Add files using upload-large-folder tool

DBCA_original_Estonian/Qwen3-14B-Base_original_features_structural_train_original_features_structural_test1/checkpoint-770/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen3-14B-Base
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen3-14B-Base
+- lora
+- sft
+- transformers
+- trl
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

DBCA_original_Estonian/Qwen3-14B-Base_original_features_structural_train_original_features_structural_test1/checkpoint-770/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-14B-Base",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.024002122189250963,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "down_proj",
+    "q_proj",
+    "k_proj",
+    "up_proj",
+    "gate_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

DBCA_original_Estonian/Qwen3-14B-Base_original_features_structural_train_original_features_structural_test1/checkpoint-770/chat_template.jinja

@@ -0,0 +1,85 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

DBCA_original_Estonian/Qwen3-14B-Base_original_features_structural_train_original_features_structural_test1/checkpoint-770/tokenizer_config.json

@@ -0,0 +1,29 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

DBCA_original_Estonian/Qwen3-14B-Base_original_features_structural_train_original_features_structural_test1/checkpoint-770/trainer_state.json

@@ -0,0 +1,206 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 770,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 1.7284933185577394,
+      "epoch": 0.12987012987012986,
+      "grad_norm": 0.6036743521690369,
+      "learning_rate": 1.542172096032852e-05,
+      "loss": 1.6924661254882813,
+      "mean_token_accuracy": 0.6527876782417298,
+      "num_tokens": 133878.0,
+      "step": 50
+    },
+    {
+      "entropy": 0.9392665743827819,
+      "epoch": 0.2597402597402597,
+      "grad_norm": 0.41290706396102905,
+      "learning_rate": 3.115817091984742e-05,
+      "loss": 0.8887335968017578,
+      "mean_token_accuracy": 0.7722584992647171,
+      "num_tokens": 265533.0,
+      "step": 100
+    },
+    {
+      "entropy": 0.7594830900430679,
+      "epoch": 0.38961038961038963,
+      "grad_norm": 0.5102436542510986,
+      "learning_rate": 4.689462087936632e-05,
+      "loss": 0.7127500915527344,
+      "mean_token_accuracy": 0.8033800584077835,
+      "num_tokens": 386480.0,
+      "step": 150
+    },
+    {
+      "entropy": 0.6976428633928299,
+      "epoch": 0.5194805194805194,
+      "grad_norm": 0.45923739671707153,
+      "learning_rate": 6.263107083888522e-05,
+      "loss": 0.653647689819336,
+      "mean_token_accuracy": 0.8134089750051499,
+      "num_tokens": 520135.0,
+      "step": 200
+    },
+    {
+      "entropy": 0.7009701311588288,
+      "epoch": 0.6493506493506493,
+      "grad_norm": 0.5048617720603943,
+      "learning_rate": 7.836752079840412e-05,
+      "loss": 0.6510067749023437,
+      "mean_token_accuracy": 0.8158668893575668,
+      "num_tokens": 640303.0,
+      "step": 250
+    },
+    {
+      "entropy": 0.6815944242477417,
+      "epoch": 0.7792207792207793,
+      "grad_norm": 0.3944048285484314,
+      "learning_rate": 9.410397075792302e-05,
+      "loss": 0.6347948837280274,
+      "mean_token_accuracy": 0.8184031039476395,
+      "num_tokens": 766027.0,
+      "step": 300
+    },
+    {
+      "entropy": 0.6565104895830154,
+      "epoch": 0.9090909090909091,
+      "grad_norm": 0.34919872879981995,
+      "learning_rate": 0.00010984042071744191,
+      "loss": 0.6166216278076172,
+      "mean_token_accuracy": 0.8238802373409271,
+      "num_tokens": 894871.0,
+      "step": 350
+    },
+    {
+      "epoch": 1.0,
+      "eval_entropy": 0.6728800308819001,
+      "eval_loss": 0.6640146970748901,
+      "eval_mean_token_accuracy": 0.8102126026955935,
+      "eval_num_tokens": 981550.0,
+      "eval_runtime": 105.0345,
+      "eval_samples_per_second": 15.776,
+      "eval_steps_per_second": 1.98,
+      "step": 385
+    },
+    {
+      "entropy": 0.6388555985689163,
+      "epoch": 1.0389610389610389,
+      "grad_norm": 0.4504299461841583,
+      "learning_rate": 0.00012116578400073899,
+      "loss": 0.5894550704956054,
+      "mean_token_accuracy": 0.8289365124702454,
+      "num_tokens": 1021041.0,
+      "step": 400
+    },
+    {
+      "entropy": 0.6055982691049576,
+      "epoch": 1.1688311688311688,
+      "grad_norm": 0.4405532479286194,
+      "learning_rate": 0.00012106869552579844,
+      "loss": 0.5607559585571289,
+      "mean_token_accuracy": 0.8378401637077332,
+      "num_tokens": 1149161.0,
+      "step": 450
+    },
+    {
+      "entropy": 0.5999097889661789,
+      "epoch": 1.2987012987012987,
+      "grad_norm": 0.3318058252334595,
+      "learning_rate": 0.00012084732888635301,
+      "loss": 0.5543725204467773,
+      "mean_token_accuracy": 0.8376027828454972,
+      "num_tokens": 1278079.0,
+      "step": 500
+    },
+    {
+      "entropy": 0.6045731788873673,
+      "epoch": 1.4285714285714286,
+      "grad_norm": 0.32081809639930725,
+      "learning_rate": 0.00012050213893540995,
+      "loss": 0.5592476272583008,
+      "mean_token_accuracy": 0.836240154504776,
+      "num_tokens": 1405326.0,
+      "step": 550
+    },
+    {
+      "entropy": 0.5909591257572174,
+      "epoch": 1.5584415584415585,
+      "grad_norm": 0.3941470682621002,
+      "learning_rate": 0.00012003383495187264,
+      "loss": 0.5478679275512696,
+      "mean_token_accuracy": 0.8397009265422821,
+      "num_tokens": 1532507.0,
+      "step": 600
+    },
+    {
+      "entropy": 0.5752921235561371,
+      "epoch": 1.6883116883116882,
+      "grad_norm": 0.300540953874588,
+      "learning_rate": 0.00011944337918314997,
+      "loss": 0.5345719146728516,
+      "mean_token_accuracy": 0.8423490756750107,
+      "num_tokens": 1662747.0,
+      "step": 650
+    },
+    {
+      "entropy": 0.5910927414894104,
+      "epoch": 1.8181818181818183,
+      "grad_norm": 0.24327129125595093,
+      "learning_rate": 0.000118731984867979,
+      "loss": 0.5490537643432617,
+      "mean_token_accuracy": 0.8396700692176818,
+      "num_tokens": 1783817.0,
+      "step": 700
+    },
+    {
+      "entropy": 0.5870773965120315,
+      "epoch": 1.948051948051948,
+      "grad_norm": 0.41987577080726624,
+      "learning_rate": 0.00011790111374352309,
+      "loss": 0.5455797958374023,
+      "mean_token_accuracy": 0.8395824021100998,
+      "num_tokens": 1914573.0,
+      "step": 750
+    },
+    {
+      "epoch": 2.0,
+      "eval_entropy": 0.6101865721150086,
+      "eval_loss": 0.6203432083129883,
+      "eval_mean_token_accuracy": 0.819772845850541,
+      "eval_num_tokens": 1963100.0,
+      "eval_runtime": 105.1641,
+      "eval_samples_per_second": 15.756,
+      "eval_steps_per_second": 1.978,
+      "step": 770
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 3850,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.13724721019904e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-100/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen3-14B-Base
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen3-14B-Base
+- lora
+- sft
+- transformers
+- trl
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-100/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-14B-Base",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 128,
+  "lora_bias": false,
+  "lora_dropout": 0.0539675908752296,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "gate_proj",
+    "q_proj",
+    "down_proj",
+    "o_proj",
+    "k_proj",
+    "v_proj",
+    "up_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-100/chat_template.jinja

@@ -0,0 +1,85 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-100/tokenizer_config.json

@@ -0,0 +1,29 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-100/trainer_state.json

@@ -0,0 +1,139 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.24125452352231605,
+  "eval_steps": 20,
+  "global_step": 100,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 1.8160341799259185,
+      "epoch": 0.04825090470446321,
+      "grad_norm": 1.1313843727111816,
+      "learning_rate": 1.9794859683580766e-05,
+      "loss": 1.7627403259277343,
+      "mean_token_accuracy": 0.6354069545865059,
+      "num_tokens": 50237.0,
+      "step": 20
+    },
+    {
+      "epoch": 0.04825090470446321,
+      "eval_entropy": 1.538598193881217,
+      "eval_loss": 1.4881398677825928,
+      "eval_mean_token_accuracy": 0.686459881201219,
+      "eval_num_tokens": 50237.0,
+      "eval_runtime": 100.8923,
+      "eval_samples_per_second": 14.074,
+      "eval_steps_per_second": 1.764,
+      "step": 20
+    },
+    {
+      "entropy": 1.1673978716135025,
+      "epoch": 0.09650180940892641,
+      "grad_norm": 0.6167025566101074,
+      "learning_rate": 4.0631554087349995e-05,
+      "loss": 1.0887075424194337,
+      "mean_token_accuracy": 0.737513878941536,
+      "num_tokens": 101098.0,
+      "step": 40
+    },
+    {
+      "epoch": 0.09650180940892641,
+      "eval_entropy": 0.887577452351538,
+      "eval_loss": 0.8519237041473389,
+      "eval_mean_token_accuracy": 0.7741017204322173,
+      "eval_num_tokens": 101098.0,
+      "eval_runtime": 100.4819,
+      "eval_samples_per_second": 14.132,
+      "eval_steps_per_second": 1.771,
+      "step": 40
+    },
+    {
+      "entropy": 0.8220535531640053,
+      "epoch": 0.14475271411338964,
+      "grad_norm": 0.8396430611610413,
+      "learning_rate": 6.146824849111923e-05,
+      "loss": 0.7812146186828614,
+      "mean_token_accuracy": 0.7872247830033302,
+      "num_tokens": 150385.0,
+      "step": 60
+    },
+    {
+      "epoch": 0.14475271411338964,
+      "eval_entropy": 0.770665143982748,
+      "eval_loss": 0.7478981614112854,
+      "eval_mean_token_accuracy": 0.7949167068084974,
+      "eval_num_tokens": 150385.0,
+      "eval_runtime": 100.6518,
+      "eval_samples_per_second": 14.108,
+      "eval_steps_per_second": 1.768,
+      "step": 60
+    },
+    {
+      "entropy": 0.7776695728302002,
+      "epoch": 0.19300361881785283,
+      "grad_norm": 0.5731202363967896,
+      "learning_rate": 8.230494289488845e-05,
+      "loss": 0.732721471786499,
+      "mean_token_accuracy": 0.7976997002959252,
+      "num_tokens": 197354.0,
+      "step": 80
+    },
+    {
+      "epoch": 0.19300361881785283,
+      "eval_entropy": 0.7650937618834249,
+      "eval_loss": 0.7109046578407288,
+      "eval_mean_token_accuracy": 0.8018777065732506,
+      "eval_num_tokens": 197354.0,
+      "eval_runtime": 100.5983,
+      "eval_samples_per_second": 14.116,
+      "eval_steps_per_second": 1.769,
+      "step": 80
+    },
+    {
+      "entropy": 0.7309209659695626,
+      "epoch": 0.24125452352231605,
+      "grad_norm": 0.48046866059303284,
+      "learning_rate": 0.00010314163729865769,
+      "loss": 0.680583143234253,
+      "mean_token_accuracy": 0.8073364540934562,
+      "num_tokens": 248557.0,
+      "step": 100
+    },
+    {
+      "epoch": 0.24125452352231605,
+      "eval_entropy": 0.7152130687504672,
+      "eval_loss": 0.6989707350730896,
+      "eval_mean_token_accuracy": 0.8057264320636064,
+      "eval_num_tokens": 248557.0,
+      "eval_runtime": 100.6264,
+      "eval_samples_per_second": 14.112,
+      "eval_steps_per_second": 1.769,
+      "step": 100
+    }
+  ],
+  "logging_steps": 20,
+  "max_steps": 4150,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 20,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.0561259958272e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1000/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen3-14B-Base
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen3-14B-Base
+- lora
+- sft
+- transformers
+- trl
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1000/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-14B-Base",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 128,
+  "lora_bias": false,
+  "lora_dropout": 0.0539675908752296,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "gate_proj",
+    "q_proj",
+    "down_proj",
+    "o_proj",
+    "k_proj",
+    "v_proj",
+    "up_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1000/chat_template.jinja

@@ -0,0 +1,85 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1000/tokenizer_config.json

@@ -0,0 +1,29 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1000/trainer_state.json

@@ -0,0 +1,1084 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.4101326899879374,
+  "eval_steps": 20,
+  "global_step": 1000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 1.8160341799259185,
+      "epoch": 0.04825090470446321,
+      "grad_norm": 1.1313843727111816,
+      "learning_rate": 1.9794859683580766e-05,
+      "loss": 1.7627403259277343,
+      "mean_token_accuracy": 0.6354069545865059,
+      "num_tokens": 50237.0,
+      "step": 20
+    },
+    {
+      "epoch": 0.04825090470446321,
+      "eval_entropy": 1.538598193881217,
+      "eval_loss": 1.4881398677825928,
+      "eval_mean_token_accuracy": 0.686459881201219,
+      "eval_num_tokens": 50237.0,
+      "eval_runtime": 100.8923,
+      "eval_samples_per_second": 14.074,
+      "eval_steps_per_second": 1.764,
+      "step": 20
+    },
+    {
+      "entropy": 1.1673978716135025,
+      "epoch": 0.09650180940892641,
+      "grad_norm": 0.6167025566101074,
+      "learning_rate": 4.0631554087349995e-05,
+      "loss": 1.0887075424194337,
+      "mean_token_accuracy": 0.737513878941536,
+      "num_tokens": 101098.0,
+      "step": 40
+    },
+    {
+      "epoch": 0.09650180940892641,
+      "eval_entropy": 0.887577452351538,
+      "eval_loss": 0.8519237041473389,
+      "eval_mean_token_accuracy": 0.7741017204322173,
+      "eval_num_tokens": 101098.0,
+      "eval_runtime": 100.4819,
+      "eval_samples_per_second": 14.132,
+      "eval_steps_per_second": 1.771,
+      "step": 40
+    },
+    {
+      "entropy": 0.8220535531640053,
+      "epoch": 0.14475271411338964,
+      "grad_norm": 0.8396430611610413,
+      "learning_rate": 6.146824849111923e-05,
+      "loss": 0.7812146186828614,
+      "mean_token_accuracy": 0.7872247830033302,
+      "num_tokens": 150385.0,
+      "step": 60
+    },
+    {
+      "epoch": 0.14475271411338964,
+      "eval_entropy": 0.770665143982748,
+      "eval_loss": 0.7478981614112854,
+      "eval_mean_token_accuracy": 0.7949167068084974,
+      "eval_num_tokens": 150385.0,
+      "eval_runtime": 100.6518,
+      "eval_samples_per_second": 14.108,
+      "eval_steps_per_second": 1.768,
+      "step": 60
+    },
+    {
+      "entropy": 0.7776695728302002,
+      "epoch": 0.19300361881785283,
+      "grad_norm": 0.5731202363967896,
+      "learning_rate": 8.230494289488845e-05,
+      "loss": 0.732721471786499,
+      "mean_token_accuracy": 0.7976997002959252,
+      "num_tokens": 197354.0,
+      "step": 80
+    },
+    {
+      "epoch": 0.19300361881785283,
+      "eval_entropy": 0.7650937618834249,
+      "eval_loss": 0.7109046578407288,
+      "eval_mean_token_accuracy": 0.8018777065732506,
+      "eval_num_tokens": 197354.0,
+      "eval_runtime": 100.5983,
+      "eval_samples_per_second": 14.116,
+      "eval_steps_per_second": 1.769,
+      "step": 80
+    },
+    {
+      "entropy": 0.7309209659695626,
+      "epoch": 0.24125452352231605,
+      "grad_norm": 0.48046866059303284,
+      "learning_rate": 0.00010314163729865769,
+      "loss": 0.680583143234253,
+      "mean_token_accuracy": 0.8073364540934562,
+      "num_tokens": 248557.0,
+      "step": 100
+    },
+    {
+      "epoch": 0.24125452352231605,
+      "eval_entropy": 0.7152130687504672,
+      "eval_loss": 0.6989707350730896,
+      "eval_mean_token_accuracy": 0.8057264320636064,
+      "eval_num_tokens": 248557.0,
+      "eval_runtime": 100.6264,
+      "eval_samples_per_second": 14.112,
+      "eval_steps_per_second": 1.769,
+      "step": 100
+    },
+    {
+      "entropy": 0.7105271726846695,
+      "epoch": 0.28950542822677927,
+      "grad_norm": 0.5512468814849854,
+      "learning_rate": 0.00012397833170242693,
+      "loss": 0.6674309253692627,
+      "mean_token_accuracy": 0.8098619893193245,
+      "num_tokens": 298041.0,
+      "step": 120
+    },
+    {
+      "epoch": 0.28950542822677927,
+      "eval_entropy": 0.7314498555124476,
+      "eval_loss": 0.6796280145645142,
+      "eval_mean_token_accuracy": 0.809063873264227,
+      "eval_num_tokens": 298041.0,
+      "eval_runtime": 100.6132,
+      "eval_samples_per_second": 14.113,
+      "eval_steps_per_second": 1.769,
+      "step": 120
+    },
+    {
+      "entropy": 0.7079856500029564,
+      "epoch": 0.33775633293124246,
+      "grad_norm": 0.5903022885322571,
+      "learning_rate": 0.00014481502610619613,
+      "loss": 0.6623000144958496,
+      "mean_token_accuracy": 0.8101129323244095,
+      "num_tokens": 348955.0,
+      "step": 140
+    },
+    {
+      "epoch": 0.33775633293124246,
+      "eval_entropy": 0.7005330203624254,
+      "eval_loss": 0.671779453754425,
+      "eval_mean_token_accuracy": 0.8106091374091888,
+      "eval_num_tokens": 348955.0,
+      "eval_runtime": 100.6904,
+      "eval_samples_per_second": 14.103,
+      "eval_steps_per_second": 1.768,
+      "step": 140
+    },
+    {
+      "entropy": 0.7109143257141113,
+      "epoch": 0.38600723763570566,
+      "grad_norm": 0.4516378343105316,
+      "learning_rate": 0.00016565172050996536,
+      "loss": 0.658915901184082,
+      "mean_token_accuracy": 0.8112782314419746,
+      "num_tokens": 395798.0,
+      "step": 160
+    },
+    {
+      "epoch": 0.38600723763570566,
+      "eval_entropy": 0.6891461650976974,
+      "eval_loss": 0.6598312258720398,
+      "eval_mean_token_accuracy": 0.8133360951134329,
+      "eval_num_tokens": 395798.0,
+      "eval_runtime": 100.5057,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 160
+    },
+    {
+      "entropy": 0.6896152347326279,
+      "epoch": 0.43425814234016885,
+      "grad_norm": 0.49479565024375916,
+      "learning_rate": 0.0001864884149137346,
+      "loss": 0.6466911315917969,
+      "mean_token_accuracy": 0.8141682580113411,
+      "num_tokens": 446437.0,
+      "step": 180
+    },
+    {
+      "epoch": 0.43425814234016885,
+      "eval_entropy": 0.6984815888860253,
+      "eval_loss": 0.652909517288208,
+      "eval_mean_token_accuracy": 0.8133517600847094,
+      "eval_num_tokens": 446437.0,
+      "eval_runtime": 100.5867,
+      "eval_samples_per_second": 14.117,
+      "eval_steps_per_second": 1.77,
+      "step": 180
+    },
+    {
+      "entropy": 0.6829591259360314,
+      "epoch": 0.4825090470446321,
+      "grad_norm": 0.4835337698459625,
+      "learning_rate": 0.00020732510931750382,
+      "loss": 0.6401876449584961,
+      "mean_token_accuracy": 0.8175502926111221,
+      "num_tokens": 497682.0,
+      "step": 200
+    },
+    {
+      "epoch": 0.4825090470446321,
+      "eval_entropy": 0.6963563973314306,
+      "eval_loss": 0.650087296962738,
+      "eval_mean_token_accuracy": 0.8159684401549651,
+      "eval_num_tokens": 497682.0,
+      "eval_runtime": 100.6277,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 200
+    },
+    {
+      "entropy": 0.6903425320982933,
+      "epoch": 0.5307599517490953,
+      "grad_norm": 0.44980454444885254,
+      "learning_rate": 0.00022816180372127305,
+      "loss": 0.6409444808959961,
+      "mean_token_accuracy": 0.8163257524371147,
+      "num_tokens": 544670.0,
+      "step": 220
+    },
+    {
+      "epoch": 0.5307599517490953,
+      "eval_entropy": 0.6687790182199371,
+      "eval_loss": 0.6451645493507385,
+      "eval_mean_token_accuracy": 0.8157875551266617,
+      "eval_num_tokens": 544670.0,
+      "eval_runtime": 100.5444,
+      "eval_samples_per_second": 14.123,
+      "eval_steps_per_second": 1.77,
+      "step": 220
+    },
+    {
+      "entropy": 0.6740772560238838,
+      "epoch": 0.5790108564535585,
+      "grad_norm": 0.5312877297401428,
+      "learning_rate": 0.0002489984981250423,
+      "loss": 0.6307676315307618,
+      "mean_token_accuracy": 0.8174169659614563,
+      "num_tokens": 592926.0,
+      "step": 240
+    },
+    {
+      "epoch": 0.5790108564535585,
+      "eval_entropy": 0.7063750872451268,
+      "eval_loss": 0.6494986414909363,
+      "eval_mean_token_accuracy": 0.8156777198395032,
+      "eval_num_tokens": 592926.0,
+      "eval_runtime": 100.5094,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 240
+    },
+    {
+      "entropy": 0.6876470118761062,
+      "epoch": 0.6272617611580217,
+      "grad_norm": 0.46149370074272156,
+      "learning_rate": 0.00026983519252881153,
+      "loss": 0.6392203807830811,
+      "mean_token_accuracy": 0.8172321930527687,
+      "num_tokens": 639077.0,
+      "step": 260
+    },
+    {
+      "epoch": 0.6272617611580217,
+      "eval_entropy": 0.625594850838854,
+      "eval_loss": 0.6463819742202759,
+      "eval_mean_token_accuracy": 0.8179670525400826,
+      "eval_num_tokens": 639077.0,
+      "eval_runtime": 100.5775,
+      "eval_samples_per_second": 14.118,
+      "eval_steps_per_second": 1.77,
+      "step": 260
+    },
+    {
+      "entropy": 0.6673879995942116,
+      "epoch": 0.6755126658624849,
+      "grad_norm": 0.42477288842201233,
+      "learning_rate": 0.0002906718869325807,
+      "loss": 0.6300291538238525,
+      "mean_token_accuracy": 0.8203139930963517,
+      "num_tokens": 691028.0,
+      "step": 280
+    },
+    {
+      "epoch": 0.6755126658624849,
+      "eval_entropy": 0.6863534494062488,
+      "eval_loss": 0.6392037272453308,
+      "eval_mean_token_accuracy": 0.8190520385677895,
+      "eval_num_tokens": 691028.0,
+      "eval_runtime": 100.5106,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 280
+    },
+    {
+      "entropy": 0.6573980256915093,
+      "epoch": 0.7237635705669482,
+      "grad_norm": 0.4656430780887604,
+      "learning_rate": 0.00031150858133635,
+      "loss": 0.6215964794158936,
+      "mean_token_accuracy": 0.8208559453487396,
+      "num_tokens": 740255.0,
+      "step": 300
+    },
+    {
+      "epoch": 0.7237635705669482,
+      "eval_entropy": 0.6834242812703165,
+      "eval_loss": 0.6479960083961487,
+      "eval_mean_token_accuracy": 0.8128752906001016,
+      "eval_num_tokens": 740255.0,
+      "eval_runtime": 100.6101,
+      "eval_samples_per_second": 14.114,
+      "eval_steps_per_second": 1.769,
+      "step": 300
+    },
+    {
+      "entropy": 0.6751310348510742,
+      "epoch": 0.7720144752714113,
+      "grad_norm": 0.4495522081851959,
+      "learning_rate": 0.00033234527574011916,
+      "loss": 0.6402668952941895,
+      "mean_token_accuracy": 0.8170571058988572,
+      "num_tokens": 789529.0,
+      "step": 320
+    },
+    {
+      "epoch": 0.7720144752714113,
+      "eval_entropy": 0.6920045178927733,
+      "eval_loss": 0.6402787566184998,
+      "eval_mean_token_accuracy": 0.817791668886549,
+      "eval_num_tokens": 789529.0,
+      "eval_runtime": 100.6347,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 320
+    },
+    {
+      "entropy": 0.6846290081739426,
+      "epoch": 0.8202653799758746,
+      "grad_norm": 0.5661432147026062,
+      "learning_rate": 0.00035318197014388845,
+      "loss": 0.6508871078491211,
+      "mean_token_accuracy": 0.8164169207215309,
+      "num_tokens": 841584.0,
+      "step": 340
+    },
+    {
+      "epoch": 0.8202653799758746,
+      "eval_entropy": 0.6772527078564248,
+      "eval_loss": 0.6512064337730408,
+      "eval_mean_token_accuracy": 0.8170605137776793,
+      "eval_num_tokens": 841584.0,
+      "eval_runtime": 100.5157,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 340
+    },
+    {
+      "entropy": 0.7022885024547577,
+      "epoch": 0.8685162846803377,
+      "grad_norm": 0.6357236504554749,
+      "learning_rate": 0.0003740186645476577,
+      "loss": 0.6649135112762451,
+      "mean_token_accuracy": 0.8136772960424423,
+      "num_tokens": 885769.0,
+      "step": 360
+    },
+    {
+      "epoch": 0.8685162846803377,
+      "eval_entropy": 0.7435127543599418,
+      "eval_loss": 0.6573262214660645,
+      "eval_mean_token_accuracy": 0.8147802225659403,
+      "eval_num_tokens": 885769.0,
+      "eval_runtime": 100.5271,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 360
+    },
+    {
+      "entropy": 0.6962979376316071,
+      "epoch": 0.916767189384801,
+      "grad_norm": 0.6179454326629639,
+      "learning_rate": 0.00039485535895142685,
+      "loss": 0.6509191513061523,
+      "mean_token_accuracy": 0.81364226937294,
+      "num_tokens": 930056.0,
+      "step": 380
+    },
+    {
+      "epoch": 0.916767189384801,
+      "eval_entropy": 0.6630894883295123,
+      "eval_loss": 0.6524088978767395,
+      "eval_mean_token_accuracy": 0.8165898587596551,
+      "eval_num_tokens": 930056.0,
+      "eval_runtime": 100.5343,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 380
+    },
+    {
+      "entropy": 0.694319449365139,
+      "epoch": 0.9650180940892642,
+      "grad_norm": 0.5458664894104004,
+      "learning_rate": 0.00041569205335519613,
+      "loss": 0.6624011993408203,
+      "mean_token_accuracy": 0.8136080905795098,
+      "num_tokens": 976892.0,
+      "step": 400
+    },
+    {
+      "epoch": 0.9650180940892642,
+      "eval_entropy": 0.7424925324622165,
+      "eval_loss": 0.6717661023139954,
+      "eval_mean_token_accuracy": 0.8099894781460922,
+      "eval_num_tokens": 976892.0,
+      "eval_runtime": 100.5992,
+      "eval_samples_per_second": 14.115,
+      "eval_steps_per_second": 1.769,
+      "step": 400
+    },
+    {
+      "entropy": 0.6861714736009256,
+      "epoch": 1.0120627261761157,
+      "grad_norm": 0.6108037829399109,
+      "learning_rate": 0.0004323601853188671,
+      "loss": 0.6551628112792969,
+      "mean_token_accuracy": 0.8134882618219424,
+      "num_tokens": 1024472.0,
+      "step": 420
+    },
+    {
+      "epoch": 1.0120627261761157,
+      "eval_entropy": 0.6593745705786715,
+      "eval_loss": 0.6720747947692871,
+      "eval_mean_token_accuracy": 0.812802647941568,
+      "eval_num_tokens": 1024472.0,
+      "eval_runtime": 100.4866,
+      "eval_samples_per_second": 14.131,
+      "eval_steps_per_second": 1.771,
+      "step": 420
+    },
+    {
+      "entropy": 0.639520151913166,
+      "epoch": 1.060313630880579,
+      "grad_norm": 0.6054353713989258,
+      "learning_rate": 0.00043231736219609266,
+      "loss": 0.6096667766571044,
+      "mean_token_accuracy": 0.8227144703269005,
+      "num_tokens": 1076964.0,
+      "step": 440
+    },
+    {
+      "epoch": 1.060313630880579,
+      "eval_entropy": 0.6316307876886946,
+      "eval_loss": 0.6621950268745422,
+      "eval_mean_token_accuracy": 0.8140057914712456,
+      "eval_num_tokens": 1076964.0,
+      "eval_runtime": 100.6409,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 440
+    },
+    {
+      "entropy": 0.6544070363044738,
+      "epoch": 1.1085645355850422,
+      "grad_norm": 0.5111057758331299,
+      "learning_rate": 0.0004322133749558233,
+      "loss": 0.6176822185516357,
+      "mean_token_accuracy": 0.8201584875583648,
+      "num_tokens": 1126064.0,
+      "step": 460
+    },
+    {
+      "epoch": 1.1085645355850422,
+      "eval_entropy": 0.6714627766207363,
+      "eval_loss": 0.6738516092300415,
+      "eval_mean_token_accuracy": 0.8128616829936424,
+      "eval_num_tokens": 1126064.0,
+      "eval_runtime": 100.4986,
+      "eval_samples_per_second": 14.13,
+      "eval_steps_per_second": 1.771,
+      "step": 460
+    },
+    {
+      "entropy": 0.6470697179436684,
+      "epoch": 1.1568154402895054,
+      "grad_norm": 0.5434669256210327,
+      "learning_rate": 0.0004320482530252179,
+      "loss": 0.6144762516021729,
+      "mean_token_accuracy": 0.819926056265831,
+      "num_tokens": 1176912.0,
+      "step": 480
+    },
+    {
+      "epoch": 1.1568154402895054,
+      "eval_entropy": 0.6475969080844622,
+      "eval_loss": 0.6624214053153992,
+      "eval_mean_token_accuracy": 0.8137228820431098,
+      "eval_num_tokens": 1176912.0,
+      "eval_runtime": 100.5617,
+      "eval_samples_per_second": 14.121,
+      "eval_steps_per_second": 1.77,
+      "step": 480
+    },
+    {
+      "entropy": 0.6659130036830903,
+      "epoch": 1.2050663449939687,
+      "grad_norm": 0.6470707058906555,
+      "learning_rate": 0.0004318220431318296,
+      "loss": 0.6366308212280274,
+      "mean_token_accuracy": 0.8175261527299881,
+      "num_tokens": 1223383.0,
+      "step": 500
+    },
+    {
+      "epoch": 1.2050663449939687,
+      "eval_entropy": 0.7232425132494295,
+      "eval_loss": 0.6877738833427429,
+      "eval_mean_token_accuracy": 0.8007767260744331,
+      "eval_num_tokens": 1223383.0,
+      "eval_runtime": 100.477,
+      "eval_samples_per_second": 14.133,
+      "eval_steps_per_second": 1.772,
+      "step": 500
+    },
+    {
+      "entropy": 0.6759189993143082,
+      "epoch": 1.2533172496984317,
+      "grad_norm": 0.6453561186790466,
+      "learning_rate": 0.00043153480929038245,
+      "loss": 0.637772512435913,
+      "mean_token_accuracy": 0.8162627041339874,
+      "num_tokens": 1273534.0,
+      "step": 520
+    },
+    {
+      "epoch": 1.2533172496984317,
+      "eval_entropy": 0.7082584312792575,
+      "eval_loss": 0.6695764064788818,
+      "eval_mean_token_accuracy": 0.812019992410467,
+      "eval_num_tokens": 1273534.0,
+      "eval_runtime": 100.4957,
+      "eval_samples_per_second": 14.13,
+      "eval_steps_per_second": 1.771,
+      "step": 520
+    },
+    {
+      "entropy": 0.6730152070522308,
+      "epoch": 1.301568154402895,
+      "grad_norm": 0.6234476566314697,
+      "learning_rate": 0.000431186632784656,
+      "loss": 0.640014362335205,
+      "mean_token_accuracy": 0.8168858170509339,
+      "num_tokens": 1323351.0,
+      "step": 540
+    },
+    {
+      "epoch": 1.301568154402895,
+      "eval_entropy": 0.6655137846978862,
+      "eval_loss": 0.6578248739242554,
+      "eval_mean_token_accuracy": 0.8147600979617472,
+      "eval_num_tokens": 1323351.0,
+      "eval_runtime": 100.503,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 540
+    },
+    {
+      "entropy": 0.6636043325066566,
+      "epoch": 1.3498190591073582,
+      "grad_norm": 0.5882512331008911,
+      "learning_rate": 0.0004307776121444827,
+      "loss": 0.6286775588989257,
+      "mean_token_accuracy": 0.8183174058794975,
+      "num_tokens": 1371508.0,
+      "step": 560
+    },
+    {
+      "epoch": 1.3498190591073582,
+      "eval_entropy": 0.6304813810948575,
+      "eval_loss": 0.6607436537742615,
+      "eval_mean_token_accuracy": 0.8169659268989992,
+      "eval_num_tokens": 1371508.0,
+      "eval_runtime": 100.6233,
+      "eval_samples_per_second": 14.112,
+      "eval_steps_per_second": 1.769,
+      "step": 560
+    },
+    {
+      "entropy": 0.6608738481998444,
+      "epoch": 1.3980699638118215,
+      "grad_norm": 0.6796398758888245,
+      "learning_rate": 0.00043030786311786564,
+      "loss": 0.622726583480835,
+      "mean_token_accuracy": 0.8230827242136002,
+      "num_tokens": 1415221.0,
+      "step": 580
+    },
+    {
+      "epoch": 1.3980699638118215,
+      "eval_entropy": 0.6645242689030894,
+      "eval_loss": 0.6662811636924744,
+      "eval_mean_token_accuracy": 0.8147623907314258,
+      "eval_num_tokens": 1415221.0,
+      "eval_runtime": 100.5074,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 580
+    },
+    {
+      "entropy": 0.6682895936071873,
+      "epoch": 1.4463208685162847,
+      "grad_norm": 0.5750718712806702,
+      "learning_rate": 0.0004297775186382226,
+      "loss": 0.6304276466369629,
+      "mean_token_accuracy": 0.8188625335693359,
+      "num_tokens": 1463356.0,
+      "step": 600
+    },
+    {
+      "epoch": 1.4463208685162847,
+      "eval_entropy": 0.6320532578765676,
+      "eval_loss": 0.650333046913147,
+      "eval_mean_token_accuracy": 0.8172853793321031,
+      "eval_num_tokens": 1463356.0,
+      "eval_runtime": 100.5116,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 600
+    },
+    {
+      "entropy": 0.653739221394062,
+      "epoch": 1.494571773220748,
+      "grad_norm": 0.5552534461021423,
+      "learning_rate": 0.00042918672878676804,
+      "loss": 0.6215689659118653,
+      "mean_token_accuracy": 0.8208896175026894,
+      "num_tokens": 1512340.0,
+      "step": 620
+    },
+    {
+      "epoch": 1.494571773220748,
+      "eval_entropy": 0.6285932062382109,
+      "eval_loss": 0.6392536163330078,
+      "eval_mean_token_accuracy": 0.8205167825302381,
+      "eval_num_tokens": 1512340.0,
+      "eval_runtime": 100.6392,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 620
+    },
+    {
+      "entropy": 0.6476897612214089,
+      "epoch": 1.5428226779252112,
+      "grad_norm": 0.536978542804718,
+      "learning_rate": 0.00042853566075004156,
+      "loss": 0.6131608009338378,
+      "mean_token_accuracy": 0.8208152890205384,
+      "num_tokens": 1560911.0,
+      "step": 640
+    },
+    {
+      "epoch": 1.5428226779252112,
+      "eval_entropy": 0.6626498856571283,
+      "eval_loss": 0.6479025483131409,
+      "eval_mean_token_accuracy": 0.8180894101603647,
+      "eval_num_tokens": 1560911.0,
+      "eval_runtime": 100.5035,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 640
+    },
+    {
+      "entropy": 0.6338633447885513,
+      "epoch": 1.5910735826296745,
+      "grad_norm": 0.4518965780735016,
+      "learning_rate": 0.0004278244987725961,
+      "loss": 0.6010610580444335,
+      "mean_token_accuracy": 0.8247816175222397,
+      "num_tokens": 1609409.0,
+      "step": 660
+    },
+    {
+      "epoch": 1.5910735826296745,
+      "eval_entropy": 0.64079723699709,
+      "eval_loss": 0.6417157053947449,
+      "eval_mean_token_accuracy": 0.8214994581897607,
+      "eval_num_tokens": 1609409.0,
+      "eval_runtime": 100.6393,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 660
+    },
+    {
+      "entropy": 0.6485912457108498,
+      "epoch": 1.6393244873341375,
+      "grad_norm": 0.6028368473052979,
+      "learning_rate": 0.00042705344410485905,
+      "loss": 0.6120437145233154,
+      "mean_token_accuracy": 0.8236212745308876,
+      "num_tokens": 1657430.0,
+      "step": 680
+    },
+    {
+      "epoch": 1.6393244873341375,
+      "eval_entropy": 0.6863006427046958,
+      "eval_loss": 0.6420931816101074,
+      "eval_mean_token_accuracy": 0.8191533580924688,
+      "eval_num_tokens": 1657430.0,
+      "eval_runtime": 100.4639,
+      "eval_samples_per_second": 14.134,
+      "eval_steps_per_second": 1.772,
+      "step": 680
+    },
+    {
+      "entropy": 0.6459382772445679,
+      "epoch": 1.6875753920386007,
+      "grad_norm": 0.5069310069084167,
+      "learning_rate": 0.00042622271494618073,
+      "loss": 0.6160889625549316,
+      "mean_token_accuracy": 0.8253828629851341,
+      "num_tokens": 1708058.0,
+      "step": 700
+    },
+    {
+      "epoch": 1.6875753920386007,
+      "eval_entropy": 0.6773919463157654,
+      "eval_loss": 0.6307896375656128,
+      "eval_mean_token_accuracy": 0.8217989825800563,
+      "eval_num_tokens": 1708058.0,
+      "eval_runtime": 100.6301,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 700
+    },
+    {
+      "entropy": 0.64687028080225,
+      "epoch": 1.7358262967430638,
+      "grad_norm": 0.3830573856830597,
+      "learning_rate": 0.00042533254638308645,
+      "loss": 0.6104001998901367,
+      "mean_token_accuracy": 0.8255606815218925,
+      "num_tokens": 1761262.0,
+      "step": 720
+    },
+    {
+      "epoch": 1.7358262967430638,
+      "eval_entropy": 0.6179541145147902,
+      "eval_loss": 0.6170215606689453,
+      "eval_mean_token_accuracy": 0.8259220900160543,
+      "eval_num_tokens": 1761262.0,
+      "eval_runtime": 100.61,
+      "eval_samples_per_second": 14.114,
+      "eval_steps_per_second": 1.769,
+      "step": 720
+    },
+    {
+      "entropy": 0.6306161753833294,
+      "epoch": 1.784077201447527,
+      "grad_norm": 0.5888957977294922,
+      "learning_rate": 0.0004243831903227498,
+      "loss": 0.5981546878814697,
+      "mean_token_accuracy": 0.8266776099801063,
+      "num_tokens": 1812127.0,
+      "step": 740
+    },
+    {
+      "epoch": 1.784077201447527,
+      "eval_entropy": 0.6357416808605194,
+      "eval_loss": 0.6202242374420166,
+      "eval_mean_token_accuracy": 0.8253380304641937,
+      "eval_num_tokens": 1812127.0,
+      "eval_runtime": 100.5145,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 740
+    },
+    {
+      "entropy": 0.6290688961744308,
+      "epoch": 1.8323281061519903,
+      "grad_norm": 0.6138001680374146,
+      "learning_rate": 0.0004233749154217063,
+      "loss": 0.5853167057037354,
+      "mean_token_accuracy": 0.8281400203704834,
+      "num_tokens": 1858403.0,
+      "step": 760
+    },
+    {
+      "epoch": 1.8323281061519903,
+      "eval_entropy": 0.6402715869833914,
+      "eval_loss": 0.6168262958526611,
+      "eval_mean_token_accuracy": 0.8246346306934785,
+      "eval_num_tokens": 1858403.0,
+      "eval_runtime": 100.5237,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 760
+    },
+    {
+      "entropy": 0.6218615680932998,
+      "epoch": 1.8805790108564535,
+      "grad_norm": 0.574312686920166,
+      "learning_rate": 0.00042230800700982586,
+      "loss": 0.5841915607452393,
+      "mean_token_accuracy": 0.8304707780480385,
+      "num_tokens": 1907361.0,
+      "step": 780
+    },
+    {
+      "epoch": 1.8805790108564535,
+      "eval_entropy": 0.6301163148679091,
+      "eval_loss": 0.6073815226554871,
+      "eval_mean_token_accuracy": 0.8274695993809218,
+      "eval_num_tokens": 1907361.0,
+      "eval_runtime": 100.5111,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 780
+    },
+    {
+      "entropy": 0.6101033873856068,
+      "epoch": 1.9288299155609168,
+      "grad_norm": 0.5054804682731628,
+      "learning_rate": 0.000421182767009569,
+      "loss": 0.5774452209472656,
+      "mean_token_accuracy": 0.8321943372488022,
+      "num_tokens": 1956605.0,
+      "step": 800
+    },
+    {
+      "epoch": 1.9288299155609168,
+      "eval_entropy": 0.631400960215022,
+      "eval_loss": 0.6115217804908752,
+      "eval_mean_token_accuracy": 0.8249618762664581,
+      "eval_num_tokens": 1956605.0,
+      "eval_runtime": 100.4916,
+      "eval_samples_per_second": 14.131,
+      "eval_steps_per_second": 1.771,
+      "step": 800
+    },
+    {
+      "entropy": 0.6217561915516854,
+      "epoch": 1.97708082026538,
+      "grad_norm": 0.5333649516105652,
+      "learning_rate": 0.0004199995138505454,
+      "loss": 0.5842787265777588,
+      "mean_token_accuracy": 0.8305525362491608,
+      "num_tokens": 2003308.0,
+      "step": 820
+    },
+    {
+      "epoch": 1.97708082026538,
+      "eval_entropy": 0.605928597657868,
+      "eval_loss": 0.6067516207695007,
+      "eval_mean_token_accuracy": 0.82715918408351,
+      "eval_num_tokens": 2003308.0,
+      "eval_runtime": 100.5286,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 820
+    },
+    {
+      "entropy": 0.5585076526189462,
+      "epoch": 2.0241254523522314,
+      "grad_norm": 0.44819608330726624,
+      "learning_rate": 0.00041875858237940244,
+      "loss": 0.5252367496490479,
+      "mean_token_accuracy": 0.8405626966403081,
+      "num_tokens": 2051331.0,
+      "step": 840
+    },
+    {
+      "epoch": 2.0241254523522314,
+      "eval_entropy": 0.6123150769579276,
+      "eval_loss": 0.6175946593284607,
+      "eval_mean_token_accuracy": 0.827637837509091,
+      "eval_num_tokens": 2051331.0,
+      "eval_runtime": 100.5414,
+      "eval_samples_per_second": 14.124,
+      "eval_steps_per_second": 1.77,
+      "step": 840
+    },
+    {
+      "entropy": 0.5359115816652775,
+      "epoch": 2.0723763570566947,
+      "grad_norm": 0.576781690120697,
+      "learning_rate": 0.0004174603237650678,
+      "loss": 0.5023207664489746,
+      "mean_token_accuracy": 0.8459693863987923,
+      "num_tokens": 2101528.0,
+      "step": 860
+    },
+    {
+      "epoch": 2.0723763570566947,
+      "eval_entropy": 0.5592134784111816,
+      "eval_loss": 0.620693564414978,
+      "eval_mean_token_accuracy": 0.827402507656076,
+      "eval_num_tokens": 2101528.0,
+      "eval_runtime": 100.6274,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 860
+    },
+    {
+      "entropy": 0.5214565828442573,
+      "epoch": 2.120627261761158,
+      "grad_norm": 0.45878875255584717,
+      "learning_rate": 0.00041610510539937234,
+      "loss": 0.48367843627929685,
+      "mean_token_accuracy": 0.8515541926026344,
+      "num_tokens": 2149550.0,
+      "step": 880
+    },
+    {
+      "epoch": 2.120627261761158,
+      "eval_entropy": 0.5860532344727034,
+      "eval_loss": 0.6134791970252991,
+      "eval_mean_token_accuracy": 0.8291774777214179,
+      "eval_num_tokens": 2149550.0,
+      "eval_runtime": 100.5295,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 880
+    },
+    {
+      "entropy": 0.5414379432797432,
+      "epoch": 2.168878166465621,
+      "grad_norm": 0.5976294875144958,
+      "learning_rate": 0.0004146933107930829,
+      "loss": 0.5012908935546875,
+      "mean_token_accuracy": 0.8466357633471489,
+      "num_tokens": 2202162.0,
+      "step": 900
+    },
+    {
+      "epoch": 2.168878166465621,
+      "eval_entropy": 0.5427783280611038,
+      "eval_loss": 0.612716019153595,
+      "eval_mean_token_accuracy": 0.829505049445656,
+      "eval_num_tokens": 2202162.0,
+      "eval_runtime": 100.5207,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 900
+    },
+    {
+      "entropy": 0.521108990907669,
+      "epoch": 2.2171290711700844,
+      "grad_norm": 0.5145775675773621,
+      "learning_rate": 0.000413225339467373,
+      "loss": 0.4877732276916504,
+      "mean_token_accuracy": 0.8504189744591713,
+      "num_tokens": 2256539.0,
+      "step": 920
+    },
+    {
+      "epoch": 2.2171290711700844,
+      "eval_entropy": 0.5620684869838565,
+      "eval_loss": 0.6176337599754333,
+      "eval_mean_token_accuracy": 0.8291444483767735,
+      "eval_num_tokens": 2256539.0,
+      "eval_runtime": 100.6819,
+      "eval_samples_per_second": 14.104,
+      "eval_steps_per_second": 1.768,
+      "step": 920
+    },
+    {
+      "entropy": 0.5452766507863999,
+      "epoch": 2.2653799758745476,
+      "grad_norm": 0.4820452928543091,
+      "learning_rate": 0.0004117016068407636,
+      "loss": 0.4991307735443115,
+      "mean_token_accuracy": 0.8474363312125206,
+      "num_tokens": 2299698.0,
+      "step": 940
+    },
+    {
+      "epoch": 2.2653799758745476,
+      "eval_entropy": 0.5611683193887218,
+      "eval_loss": 0.6153580546379089,
+      "eval_mean_token_accuracy": 0.8286041272490212,
+      "eval_num_tokens": 2299698.0,
+      "eval_runtime": 100.5249,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 940
+    },
+    {
+      "entropy": 0.5572576269507408,
+      "epoch": 2.313630880579011,
+      "grad_norm": 0.3947993814945221,
+      "learning_rate": 0.00041012254411156387,
+      "loss": 0.5191003799438476,
+      "mean_token_accuracy": 0.8422838419675827,
+      "num_tokens": 2347134.0,
+      "step": 960
+    },
+    {
+      "epoch": 2.313630880579011,
+      "eval_entropy": 0.5651392345682958,
+      "eval_loss": 0.6121005415916443,
+      "eval_mean_token_accuracy": 0.8286536242185014,
+      "eval_num_tokens": 2347134.0,
+      "eval_runtime": 100.7033,
+      "eval_samples_per_second": 14.101,
+      "eval_steps_per_second": 1.768,
+      "step": 960
+    },
+    {
+      "entropy": 0.5441671140491963,
+      "epoch": 2.361881785283474,
+      "grad_norm": 0.7719791531562805,
+      "learning_rate": 0.0004084885981358477,
+      "loss": 0.5072606086730957,
+      "mean_token_accuracy": 0.8478495761752128,
+      "num_tokens": 2394568.0,
+      "step": 980
+    },
+    {
+      "epoch": 2.361881785283474,
+      "eval_entropy": 0.5651004270891126,
+      "eval_loss": 0.6069501042366028,
+      "eval_mean_token_accuracy": 0.8311042156112328,
+      "eval_num_tokens": 2394568.0,
+      "eval_runtime": 100.5487,
+      "eval_samples_per_second": 14.123,
+      "eval_steps_per_second": 1.77,
+      "step": 980
+    },
+    {
+      "entropy": 0.5462074346840382,
+      "epoch": 2.4101326899879374,
+      "grad_norm": 0.5694436430931091,
+      "learning_rate": 0.00040680023130099856,
+      "loss": 0.5097150325775146,
+      "mean_token_accuracy": 0.8474388793110847,
+      "num_tokens": 2443253.0,
+      "step": 1000
+    },
+    {
+      "epoch": 2.4101326899879374,
+      "eval_entropy": 0.5895893243926295,
+      "eval_loss": 0.6078168749809265,
+      "eval_mean_token_accuracy": 0.828030709470256,
+      "eval_num_tokens": 2443253.0,
+      "eval_runtime": 100.5413,
+      "eval_samples_per_second": 14.124,
+      "eval_steps_per_second": 1.77,
+      "step": 1000
+    }
+  ],
+  "logging_steps": 20,
+  "max_steps": 4150,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 20,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.0729150094336e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1020/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen3-14B-Base
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen3-14B-Base
+- lora
+- sft
+- transformers
+- trl
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1020/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-14B-Base",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 128,
+  "lora_bias": false,
+  "lora_dropout": 0.0539675908752296,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "gate_proj",
+    "q_proj",
+    "down_proj",
+    "o_proj",
+    "k_proj",
+    "v_proj",
+    "up_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1020/chat_template.jinja

@@ -0,0 +1,85 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1020/tokenizer_config.json

@@ -0,0 +1,29 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1020/trainer_state.json

@@ -0,0 +1,1105 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.4583835946924006,
+  "eval_steps": 20,
+  "global_step": 1020,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 1.8160341799259185,
+      "epoch": 0.04825090470446321,
+      "grad_norm": 1.1313843727111816,
+      "learning_rate": 1.9794859683580766e-05,
+      "loss": 1.7627403259277343,
+      "mean_token_accuracy": 0.6354069545865059,
+      "num_tokens": 50237.0,
+      "step": 20
+    },
+    {
+      "epoch": 0.04825090470446321,
+      "eval_entropy": 1.538598193881217,
+      "eval_loss": 1.4881398677825928,
+      "eval_mean_token_accuracy": 0.686459881201219,
+      "eval_num_tokens": 50237.0,
+      "eval_runtime": 100.8923,
+      "eval_samples_per_second": 14.074,
+      "eval_steps_per_second": 1.764,
+      "step": 20
+    },
+    {
+      "entropy": 1.1673978716135025,
+      "epoch": 0.09650180940892641,
+      "grad_norm": 0.6167025566101074,
+      "learning_rate": 4.0631554087349995e-05,
+      "loss": 1.0887075424194337,
+      "mean_token_accuracy": 0.737513878941536,
+      "num_tokens": 101098.0,
+      "step": 40
+    },
+    {
+      "epoch": 0.09650180940892641,
+      "eval_entropy": 0.887577452351538,
+      "eval_loss": 0.8519237041473389,
+      "eval_mean_token_accuracy": 0.7741017204322173,
+      "eval_num_tokens": 101098.0,
+      "eval_runtime": 100.4819,
+      "eval_samples_per_second": 14.132,
+      "eval_steps_per_second": 1.771,
+      "step": 40
+    },
+    {
+      "entropy": 0.8220535531640053,
+      "epoch": 0.14475271411338964,
+      "grad_norm": 0.8396430611610413,
+      "learning_rate": 6.146824849111923e-05,
+      "loss": 0.7812146186828614,
+      "mean_token_accuracy": 0.7872247830033302,
+      "num_tokens": 150385.0,
+      "step": 60
+    },
+    {
+      "epoch": 0.14475271411338964,
+      "eval_entropy": 0.770665143982748,
+      "eval_loss": 0.7478981614112854,
+      "eval_mean_token_accuracy": 0.7949167068084974,
+      "eval_num_tokens": 150385.0,
+      "eval_runtime": 100.6518,
+      "eval_samples_per_second": 14.108,
+      "eval_steps_per_second": 1.768,
+      "step": 60
+    },
+    {
+      "entropy": 0.7776695728302002,
+      "epoch": 0.19300361881785283,
+      "grad_norm": 0.5731202363967896,
+      "learning_rate": 8.230494289488845e-05,
+      "loss": 0.732721471786499,
+      "mean_token_accuracy": 0.7976997002959252,
+      "num_tokens": 197354.0,
+      "step": 80
+    },
+    {
+      "epoch": 0.19300361881785283,
+      "eval_entropy": 0.7650937618834249,
+      "eval_loss": 0.7109046578407288,
+      "eval_mean_token_accuracy": 0.8018777065732506,
+      "eval_num_tokens": 197354.0,
+      "eval_runtime": 100.5983,
+      "eval_samples_per_second": 14.116,
+      "eval_steps_per_second": 1.769,
+      "step": 80
+    },
+    {
+      "entropy": 0.7309209659695626,
+      "epoch": 0.24125452352231605,
+      "grad_norm": 0.48046866059303284,
+      "learning_rate": 0.00010314163729865769,
+      "loss": 0.680583143234253,
+      "mean_token_accuracy": 0.8073364540934562,
+      "num_tokens": 248557.0,
+      "step": 100
+    },
+    {
+      "epoch": 0.24125452352231605,
+      "eval_entropy": 0.7152130687504672,
+      "eval_loss": 0.6989707350730896,
+      "eval_mean_token_accuracy": 0.8057264320636064,
+      "eval_num_tokens": 248557.0,
+      "eval_runtime": 100.6264,
+      "eval_samples_per_second": 14.112,
+      "eval_steps_per_second": 1.769,
+      "step": 100
+    },
+    {
+      "entropy": 0.7105271726846695,
+      "epoch": 0.28950542822677927,
+      "grad_norm": 0.5512468814849854,
+      "learning_rate": 0.00012397833170242693,
+      "loss": 0.6674309253692627,
+      "mean_token_accuracy": 0.8098619893193245,
+      "num_tokens": 298041.0,
+      "step": 120
+    },
+    {
+      "epoch": 0.28950542822677927,
+      "eval_entropy": 0.7314498555124476,
+      "eval_loss": 0.6796280145645142,
+      "eval_mean_token_accuracy": 0.809063873264227,
+      "eval_num_tokens": 298041.0,
+      "eval_runtime": 100.6132,
+      "eval_samples_per_second": 14.113,
+      "eval_steps_per_second": 1.769,
+      "step": 120
+    },
+    {
+      "entropy": 0.7079856500029564,
+      "epoch": 0.33775633293124246,
+      "grad_norm": 0.5903022885322571,
+      "learning_rate": 0.00014481502610619613,
+      "loss": 0.6623000144958496,
+      "mean_token_accuracy": 0.8101129323244095,
+      "num_tokens": 348955.0,
+      "step": 140
+    },
+    {
+      "epoch": 0.33775633293124246,
+      "eval_entropy": 0.7005330203624254,
+      "eval_loss": 0.671779453754425,
+      "eval_mean_token_accuracy": 0.8106091374091888,
+      "eval_num_tokens": 348955.0,
+      "eval_runtime": 100.6904,
+      "eval_samples_per_second": 14.103,
+      "eval_steps_per_second": 1.768,
+      "step": 140
+    },
+    {
+      "entropy": 0.7109143257141113,
+      "epoch": 0.38600723763570566,
+      "grad_norm": 0.4516378343105316,
+      "learning_rate": 0.00016565172050996536,
+      "loss": 0.658915901184082,
+      "mean_token_accuracy": 0.8112782314419746,
+      "num_tokens": 395798.0,
+      "step": 160
+    },
+    {
+      "epoch": 0.38600723763570566,
+      "eval_entropy": 0.6891461650976974,
+      "eval_loss": 0.6598312258720398,
+      "eval_mean_token_accuracy": 0.8133360951134329,
+      "eval_num_tokens": 395798.0,
+      "eval_runtime": 100.5057,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 160
+    },
+    {
+      "entropy": 0.6896152347326279,
+      "epoch": 0.43425814234016885,
+      "grad_norm": 0.49479565024375916,
+      "learning_rate": 0.0001864884149137346,
+      "loss": 0.6466911315917969,
+      "mean_token_accuracy": 0.8141682580113411,
+      "num_tokens": 446437.0,
+      "step": 180
+    },
+    {
+      "epoch": 0.43425814234016885,
+      "eval_entropy": 0.6984815888860253,
+      "eval_loss": 0.652909517288208,
+      "eval_mean_token_accuracy": 0.8133517600847094,
+      "eval_num_tokens": 446437.0,
+      "eval_runtime": 100.5867,
+      "eval_samples_per_second": 14.117,
+      "eval_steps_per_second": 1.77,
+      "step": 180
+    },
+    {
+      "entropy": 0.6829591259360314,
+      "epoch": 0.4825090470446321,
+      "grad_norm": 0.4835337698459625,
+      "learning_rate": 0.00020732510931750382,
+      "loss": 0.6401876449584961,
+      "mean_token_accuracy": 0.8175502926111221,
+      "num_tokens": 497682.0,
+      "step": 200
+    },
+    {
+      "epoch": 0.4825090470446321,
+      "eval_entropy": 0.6963563973314306,
+      "eval_loss": 0.650087296962738,
+      "eval_mean_token_accuracy": 0.8159684401549651,
+      "eval_num_tokens": 497682.0,
+      "eval_runtime": 100.6277,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 200
+    },
+    {
+      "entropy": 0.6903425320982933,
+      "epoch": 0.5307599517490953,
+      "grad_norm": 0.44980454444885254,
+      "learning_rate": 0.00022816180372127305,
+      "loss": 0.6409444808959961,
+      "mean_token_accuracy": 0.8163257524371147,
+      "num_tokens": 544670.0,
+      "step": 220
+    },
+    {
+      "epoch": 0.5307599517490953,
+      "eval_entropy": 0.6687790182199371,
+      "eval_loss": 0.6451645493507385,
+      "eval_mean_token_accuracy": 0.8157875551266617,
+      "eval_num_tokens": 544670.0,
+      "eval_runtime": 100.5444,
+      "eval_samples_per_second": 14.123,
+      "eval_steps_per_second": 1.77,
+      "step": 220
+    },
+    {
+      "entropy": 0.6740772560238838,
+      "epoch": 0.5790108564535585,
+      "grad_norm": 0.5312877297401428,
+      "learning_rate": 0.0002489984981250423,
+      "loss": 0.6307676315307618,
+      "mean_token_accuracy": 0.8174169659614563,
+      "num_tokens": 592926.0,
+      "step": 240
+    },
+    {
+      "epoch": 0.5790108564535585,
+      "eval_entropy": 0.7063750872451268,
+      "eval_loss": 0.6494986414909363,
+      "eval_mean_token_accuracy": 0.8156777198395032,
+      "eval_num_tokens": 592926.0,
+      "eval_runtime": 100.5094,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 240
+    },
+    {
+      "entropy": 0.6876470118761062,
+      "epoch": 0.6272617611580217,
+      "grad_norm": 0.46149370074272156,
+      "learning_rate": 0.00026983519252881153,
+      "loss": 0.6392203807830811,
+      "mean_token_accuracy": 0.8172321930527687,
+      "num_tokens": 639077.0,
+      "step": 260
+    },
+    {
+      "epoch": 0.6272617611580217,
+      "eval_entropy": 0.625594850838854,
+      "eval_loss": 0.6463819742202759,
+      "eval_mean_token_accuracy": 0.8179670525400826,
+      "eval_num_tokens": 639077.0,
+      "eval_runtime": 100.5775,
+      "eval_samples_per_second": 14.118,
+      "eval_steps_per_second": 1.77,
+      "step": 260
+    },
+    {
+      "entropy": 0.6673879995942116,
+      "epoch": 0.6755126658624849,
+      "grad_norm": 0.42477288842201233,
+      "learning_rate": 0.0002906718869325807,
+      "loss": 0.6300291538238525,
+      "mean_token_accuracy": 0.8203139930963517,
+      "num_tokens": 691028.0,
+      "step": 280
+    },
+    {
+      "epoch": 0.6755126658624849,
+      "eval_entropy": 0.6863534494062488,
+      "eval_loss": 0.6392037272453308,
+      "eval_mean_token_accuracy": 0.8190520385677895,
+      "eval_num_tokens": 691028.0,
+      "eval_runtime": 100.5106,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 280
+    },
+    {
+      "entropy": 0.6573980256915093,
+      "epoch": 0.7237635705669482,
+      "grad_norm": 0.4656430780887604,
+      "learning_rate": 0.00031150858133635,
+      "loss": 0.6215964794158936,
+      "mean_token_accuracy": 0.8208559453487396,
+      "num_tokens": 740255.0,
+      "step": 300
+    },
+    {
+      "epoch": 0.7237635705669482,
+      "eval_entropy": 0.6834242812703165,
+      "eval_loss": 0.6479960083961487,
+      "eval_mean_token_accuracy": 0.8128752906001016,
+      "eval_num_tokens": 740255.0,
+      "eval_runtime": 100.6101,
+      "eval_samples_per_second": 14.114,
+      "eval_steps_per_second": 1.769,
+      "step": 300
+    },
+    {
+      "entropy": 0.6751310348510742,
+      "epoch": 0.7720144752714113,
+      "grad_norm": 0.4495522081851959,
+      "learning_rate": 0.00033234527574011916,
+      "loss": 0.6402668952941895,
+      "mean_token_accuracy": 0.8170571058988572,
+      "num_tokens": 789529.0,
+      "step": 320
+    },
+    {
+      "epoch": 0.7720144752714113,
+      "eval_entropy": 0.6920045178927733,
+      "eval_loss": 0.6402787566184998,
+      "eval_mean_token_accuracy": 0.817791668886549,
+      "eval_num_tokens": 789529.0,
+      "eval_runtime": 100.6347,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 320
+    },
+    {
+      "entropy": 0.6846290081739426,
+      "epoch": 0.8202653799758746,
+      "grad_norm": 0.5661432147026062,
+      "learning_rate": 0.00035318197014388845,
+      "loss": 0.6508871078491211,
+      "mean_token_accuracy": 0.8164169207215309,
+      "num_tokens": 841584.0,
+      "step": 340
+    },
+    {
+      "epoch": 0.8202653799758746,
+      "eval_entropy": 0.6772527078564248,
+      "eval_loss": 0.6512064337730408,
+      "eval_mean_token_accuracy": 0.8170605137776793,
+      "eval_num_tokens": 841584.0,
+      "eval_runtime": 100.5157,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 340
+    },
+    {
+      "entropy": 0.7022885024547577,
+      "epoch": 0.8685162846803377,
+      "grad_norm": 0.6357236504554749,
+      "learning_rate": 0.0003740186645476577,
+      "loss": 0.6649135112762451,
+      "mean_token_accuracy": 0.8136772960424423,
+      "num_tokens": 885769.0,
+      "step": 360
+    },
+    {
+      "epoch": 0.8685162846803377,
+      "eval_entropy": 0.7435127543599418,
+      "eval_loss": 0.6573262214660645,
+      "eval_mean_token_accuracy": 0.8147802225659403,
+      "eval_num_tokens": 885769.0,
+      "eval_runtime": 100.5271,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 360
+    },
+    {
+      "entropy": 0.6962979376316071,
+      "epoch": 0.916767189384801,
+      "grad_norm": 0.6179454326629639,
+      "learning_rate": 0.00039485535895142685,
+      "loss": 0.6509191513061523,
+      "mean_token_accuracy": 0.81364226937294,
+      "num_tokens": 930056.0,
+      "step": 380
+    },
+    {
+      "epoch": 0.916767189384801,
+      "eval_entropy": 0.6630894883295123,
+      "eval_loss": 0.6524088978767395,
+      "eval_mean_token_accuracy": 0.8165898587596551,
+      "eval_num_tokens": 930056.0,
+      "eval_runtime": 100.5343,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 380
+    },
+    {
+      "entropy": 0.694319449365139,
+      "epoch": 0.9650180940892642,
+      "grad_norm": 0.5458664894104004,
+      "learning_rate": 0.00041569205335519613,
+      "loss": 0.6624011993408203,
+      "mean_token_accuracy": 0.8136080905795098,
+      "num_tokens": 976892.0,
+      "step": 400
+    },
+    {
+      "epoch": 0.9650180940892642,
+      "eval_entropy": 0.7424925324622165,
+      "eval_loss": 0.6717661023139954,
+      "eval_mean_token_accuracy": 0.8099894781460922,
+      "eval_num_tokens": 976892.0,
+      "eval_runtime": 100.5992,
+      "eval_samples_per_second": 14.115,
+      "eval_steps_per_second": 1.769,
+      "step": 400
+    },
+    {
+      "entropy": 0.6861714736009256,
+      "epoch": 1.0120627261761157,
+      "grad_norm": 0.6108037829399109,
+      "learning_rate": 0.0004323601853188671,
+      "loss": 0.6551628112792969,
+      "mean_token_accuracy": 0.8134882618219424,
+      "num_tokens": 1024472.0,
+      "step": 420
+    },
+    {
+      "epoch": 1.0120627261761157,
+      "eval_entropy": 0.6593745705786715,
+      "eval_loss": 0.6720747947692871,
+      "eval_mean_token_accuracy": 0.812802647941568,
+      "eval_num_tokens": 1024472.0,
+      "eval_runtime": 100.4866,
+      "eval_samples_per_second": 14.131,
+      "eval_steps_per_second": 1.771,
+      "step": 420
+    },
+    {
+      "entropy": 0.639520151913166,
+      "epoch": 1.060313630880579,
+      "grad_norm": 0.6054353713989258,
+      "learning_rate": 0.00043231736219609266,
+      "loss": 0.6096667766571044,
+      "mean_token_accuracy": 0.8227144703269005,
+      "num_tokens": 1076964.0,
+      "step": 440
+    },
+    {
+      "epoch": 1.060313630880579,
+      "eval_entropy": 0.6316307876886946,
+      "eval_loss": 0.6621950268745422,
+      "eval_mean_token_accuracy": 0.8140057914712456,
+      "eval_num_tokens": 1076964.0,
+      "eval_runtime": 100.6409,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 440
+    },
+    {
+      "entropy": 0.6544070363044738,
+      "epoch": 1.1085645355850422,
+      "grad_norm": 0.5111057758331299,
+      "learning_rate": 0.0004322133749558233,
+      "loss": 0.6176822185516357,
+      "mean_token_accuracy": 0.8201584875583648,
+      "num_tokens": 1126064.0,
+      "step": 460
+    },
+    {
+      "epoch": 1.1085645355850422,
+      "eval_entropy": 0.6714627766207363,
+      "eval_loss": 0.6738516092300415,
+      "eval_mean_token_accuracy": 0.8128616829936424,
+      "eval_num_tokens": 1126064.0,
+      "eval_runtime": 100.4986,
+      "eval_samples_per_second": 14.13,
+      "eval_steps_per_second": 1.771,
+      "step": 460
+    },
+    {
+      "entropy": 0.6470697179436684,
+      "epoch": 1.1568154402895054,
+      "grad_norm": 0.5434669256210327,
+      "learning_rate": 0.0004320482530252179,
+      "loss": 0.6144762516021729,
+      "mean_token_accuracy": 0.819926056265831,
+      "num_tokens": 1176912.0,
+      "step": 480
+    },
+    {
+      "epoch": 1.1568154402895054,
+      "eval_entropy": 0.6475969080844622,
+      "eval_loss": 0.6624214053153992,
+      "eval_mean_token_accuracy": 0.8137228820431098,
+      "eval_num_tokens": 1176912.0,
+      "eval_runtime": 100.5617,
+      "eval_samples_per_second": 14.121,
+      "eval_steps_per_second": 1.77,
+      "step": 480
+    },
+    {
+      "entropy": 0.6659130036830903,
+      "epoch": 1.2050663449939687,
+      "grad_norm": 0.6470707058906555,
+      "learning_rate": 0.0004318220431318296,
+      "loss": 0.6366308212280274,
+      "mean_token_accuracy": 0.8175261527299881,
+      "num_tokens": 1223383.0,
+      "step": 500
+    },
+    {
+      "epoch": 1.2050663449939687,
+      "eval_entropy": 0.7232425132494295,
+      "eval_loss": 0.6877738833427429,
+      "eval_mean_token_accuracy": 0.8007767260744331,
+      "eval_num_tokens": 1223383.0,
+      "eval_runtime": 100.477,
+      "eval_samples_per_second": 14.133,
+      "eval_steps_per_second": 1.772,
+      "step": 500
+    },
+    {
+      "entropy": 0.6759189993143082,
+      "epoch": 1.2533172496984317,
+      "grad_norm": 0.6453561186790466,
+      "learning_rate": 0.00043153480929038245,
+      "loss": 0.637772512435913,
+      "mean_token_accuracy": 0.8162627041339874,
+      "num_tokens": 1273534.0,
+      "step": 520
+    },
+    {
+      "epoch": 1.2533172496984317,
+      "eval_entropy": 0.7082584312792575,
+      "eval_loss": 0.6695764064788818,
+      "eval_mean_token_accuracy": 0.812019992410467,
+      "eval_num_tokens": 1273534.0,
+      "eval_runtime": 100.4957,
+      "eval_samples_per_second": 14.13,
+      "eval_steps_per_second": 1.771,
+      "step": 520
+    },
+    {
+      "entropy": 0.6730152070522308,
+      "epoch": 1.301568154402895,
+      "grad_norm": 0.6234476566314697,
+      "learning_rate": 0.000431186632784656,
+      "loss": 0.640014362335205,
+      "mean_token_accuracy": 0.8168858170509339,
+      "num_tokens": 1323351.0,
+      "step": 540
+    },
+    {
+      "epoch": 1.301568154402895,
+      "eval_entropy": 0.6655137846978862,
+      "eval_loss": 0.6578248739242554,
+      "eval_mean_token_accuracy": 0.8147600979617472,
+      "eval_num_tokens": 1323351.0,
+      "eval_runtime": 100.503,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 540
+    },
+    {
+      "entropy": 0.6636043325066566,
+      "epoch": 1.3498190591073582,
+      "grad_norm": 0.5882512331008911,
+      "learning_rate": 0.0004307776121444827,
+      "loss": 0.6286775588989257,
+      "mean_token_accuracy": 0.8183174058794975,
+      "num_tokens": 1371508.0,
+      "step": 560
+    },
+    {
+      "epoch": 1.3498190591073582,
+      "eval_entropy": 0.6304813810948575,
+      "eval_loss": 0.6607436537742615,
+      "eval_mean_token_accuracy": 0.8169659268989992,
+      "eval_num_tokens": 1371508.0,
+      "eval_runtime": 100.6233,
+      "eval_samples_per_second": 14.112,
+      "eval_steps_per_second": 1.769,
+      "step": 560
+    },
+    {
+      "entropy": 0.6608738481998444,
+      "epoch": 1.3980699638118215,
+      "grad_norm": 0.6796398758888245,
+      "learning_rate": 0.00043030786311786564,
+      "loss": 0.622726583480835,
+      "mean_token_accuracy": 0.8230827242136002,
+      "num_tokens": 1415221.0,
+      "step": 580
+    },
+    {
+      "epoch": 1.3980699638118215,
+      "eval_entropy": 0.6645242689030894,
+      "eval_loss": 0.6662811636924744,
+      "eval_mean_token_accuracy": 0.8147623907314258,
+      "eval_num_tokens": 1415221.0,
+      "eval_runtime": 100.5074,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 580
+    },
+    {
+      "entropy": 0.6682895936071873,
+      "epoch": 1.4463208685162847,
+      "grad_norm": 0.5750718712806702,
+      "learning_rate": 0.0004297775186382226,
+      "loss": 0.6304276466369629,
+      "mean_token_accuracy": 0.8188625335693359,
+      "num_tokens": 1463356.0,
+      "step": 600
+    },
+    {
+      "epoch": 1.4463208685162847,
+      "eval_entropy": 0.6320532578765676,
+      "eval_loss": 0.650333046913147,
+      "eval_mean_token_accuracy": 0.8172853793321031,
+      "eval_num_tokens": 1463356.0,
+      "eval_runtime": 100.5116,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 600
+    },
+    {
+      "entropy": 0.653739221394062,
+      "epoch": 1.494571773220748,
+      "grad_norm": 0.5552534461021423,
+      "learning_rate": 0.00042918672878676804,
+      "loss": 0.6215689659118653,
+      "mean_token_accuracy": 0.8208896175026894,
+      "num_tokens": 1512340.0,
+      "step": 620
+    },
+    {
+      "epoch": 1.494571773220748,
+      "eval_entropy": 0.6285932062382109,
+      "eval_loss": 0.6392536163330078,
+      "eval_mean_token_accuracy": 0.8205167825302381,
+      "eval_num_tokens": 1512340.0,
+      "eval_runtime": 100.6392,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 620
+    },
+    {
+      "entropy": 0.6476897612214089,
+      "epoch": 1.5428226779252112,
+      "grad_norm": 0.536978542804718,
+      "learning_rate": 0.00042853566075004156,
+      "loss": 0.6131608009338378,
+      "mean_token_accuracy": 0.8208152890205384,
+      "num_tokens": 1560911.0,
+      "step": 640
+    },
+    {
+      "epoch": 1.5428226779252112,
+      "eval_entropy": 0.6626498856571283,
+      "eval_loss": 0.6479025483131409,
+      "eval_mean_token_accuracy": 0.8180894101603647,
+      "eval_num_tokens": 1560911.0,
+      "eval_runtime": 100.5035,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 640
+    },
+    {
+      "entropy": 0.6338633447885513,
+      "epoch": 1.5910735826296745,
+      "grad_norm": 0.4518965780735016,
+      "learning_rate": 0.0004278244987725961,
+      "loss": 0.6010610580444335,
+      "mean_token_accuracy": 0.8247816175222397,
+      "num_tokens": 1609409.0,
+      "step": 660
+    },
+    {
+      "epoch": 1.5910735826296745,
+      "eval_entropy": 0.64079723699709,
+      "eval_loss": 0.6417157053947449,
+      "eval_mean_token_accuracy": 0.8214994581897607,
+      "eval_num_tokens": 1609409.0,
+      "eval_runtime": 100.6393,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 660
+    },
+    {
+      "entropy": 0.6485912457108498,
+      "epoch": 1.6393244873341375,
+      "grad_norm": 0.6028368473052979,
+      "learning_rate": 0.00042705344410485905,
+      "loss": 0.6120437145233154,
+      "mean_token_accuracy": 0.8236212745308876,
+      "num_tokens": 1657430.0,
+      "step": 680
+    },
+    {
+      "epoch": 1.6393244873341375,
+      "eval_entropy": 0.6863006427046958,
+      "eval_loss": 0.6420931816101074,
+      "eval_mean_token_accuracy": 0.8191533580924688,
+      "eval_num_tokens": 1657430.0,
+      "eval_runtime": 100.4639,
+      "eval_samples_per_second": 14.134,
+      "eval_steps_per_second": 1.772,
+      "step": 680
+    },
+    {
+      "entropy": 0.6459382772445679,
+      "epoch": 1.6875753920386007,
+      "grad_norm": 0.5069310069084167,
+      "learning_rate": 0.00042622271494618073,
+      "loss": 0.6160889625549316,
+      "mean_token_accuracy": 0.8253828629851341,
+      "num_tokens": 1708058.0,
+      "step": 700
+    },
+    {
+      "epoch": 1.6875753920386007,
+      "eval_entropy": 0.6773919463157654,
+      "eval_loss": 0.6307896375656128,
+      "eval_mean_token_accuracy": 0.8217989825800563,
+      "eval_num_tokens": 1708058.0,
+      "eval_runtime": 100.6301,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 700
+    },
+    {
+      "entropy": 0.64687028080225,
+      "epoch": 1.7358262967430638,
+      "grad_norm": 0.3830573856830597,
+      "learning_rate": 0.00042533254638308645,
+      "loss": 0.6104001998901367,
+      "mean_token_accuracy": 0.8255606815218925,
+      "num_tokens": 1761262.0,
+      "step": 720
+    },
+    {
+      "epoch": 1.7358262967430638,
+      "eval_entropy": 0.6179541145147902,
+      "eval_loss": 0.6170215606689453,
+      "eval_mean_token_accuracy": 0.8259220900160543,
+      "eval_num_tokens": 1761262.0,
+      "eval_runtime": 100.61,
+      "eval_samples_per_second": 14.114,
+      "eval_steps_per_second": 1.769,
+      "step": 720
+    },
+    {
+      "entropy": 0.6306161753833294,
+      "epoch": 1.784077201447527,
+      "grad_norm": 0.5888957977294922,
+      "learning_rate": 0.0004243831903227498,
+      "loss": 0.5981546878814697,
+      "mean_token_accuracy": 0.8266776099801063,
+      "num_tokens": 1812127.0,
+      "step": 740
+    },
+    {
+      "epoch": 1.784077201447527,
+      "eval_entropy": 0.6357416808605194,
+      "eval_loss": 0.6202242374420166,
+      "eval_mean_token_accuracy": 0.8253380304641937,
+      "eval_num_tokens": 1812127.0,
+      "eval_runtime": 100.5145,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 740
+    },
+    {
+      "entropy": 0.6290688961744308,
+      "epoch": 1.8323281061519903,
+      "grad_norm": 0.6138001680374146,
+      "learning_rate": 0.0004233749154217063,
+      "loss": 0.5853167057037354,
+      "mean_token_accuracy": 0.8281400203704834,
+      "num_tokens": 1858403.0,
+      "step": 760
+    },
+    {
+      "epoch": 1.8323281061519903,
+      "eval_entropy": 0.6402715869833914,
+      "eval_loss": 0.6168262958526611,
+      "eval_mean_token_accuracy": 0.8246346306934785,
+      "eval_num_tokens": 1858403.0,
+      "eval_runtime": 100.5237,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 760
+    },
+    {
+      "entropy": 0.6218615680932998,
+      "epoch": 1.8805790108564535,
+      "grad_norm": 0.574312686920166,
+      "learning_rate": 0.00042230800700982586,
+      "loss": 0.5841915607452393,
+      "mean_token_accuracy": 0.8304707780480385,
+      "num_tokens": 1907361.0,
+      "step": 780
+    },
+    {
+      "epoch": 1.8805790108564535,
+      "eval_entropy": 0.6301163148679091,
+      "eval_loss": 0.6073815226554871,
+      "eval_mean_token_accuracy": 0.8274695993809218,
+      "eval_num_tokens": 1907361.0,
+      "eval_runtime": 100.5111,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 780
+    },
+    {
+      "entropy": 0.6101033873856068,
+      "epoch": 1.9288299155609168,
+      "grad_norm": 0.5054804682731628,
+      "learning_rate": 0.000421182767009569,
+      "loss": 0.5774452209472656,
+      "mean_token_accuracy": 0.8321943372488022,
+      "num_tokens": 1956605.0,
+      "step": 800
+    },
+    {
+      "epoch": 1.9288299155609168,
+      "eval_entropy": 0.631400960215022,
+      "eval_loss": 0.6115217804908752,
+      "eval_mean_token_accuracy": 0.8249618762664581,
+      "eval_num_tokens": 1956605.0,
+      "eval_runtime": 100.4916,
+      "eval_samples_per_second": 14.131,
+      "eval_steps_per_second": 1.771,
+      "step": 800
+    },
+    {
+      "entropy": 0.6217561915516854,
+      "epoch": 1.97708082026538,
+      "grad_norm": 0.5333649516105652,
+      "learning_rate": 0.0004199995138505454,
+      "loss": 0.5842787265777588,
+      "mean_token_accuracy": 0.8305525362491608,
+      "num_tokens": 2003308.0,
+      "step": 820
+    },
+    {
+      "epoch": 1.97708082026538,
+      "eval_entropy": 0.605928597657868,
+      "eval_loss": 0.6067516207695007,
+      "eval_mean_token_accuracy": 0.82715918408351,
+      "eval_num_tokens": 2003308.0,
+      "eval_runtime": 100.5286,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 820
+    },
+    {
+      "entropy": 0.5585076526189462,
+      "epoch": 2.0241254523522314,
+      "grad_norm": 0.44819608330726624,
+      "learning_rate": 0.00041875858237940244,
+      "loss": 0.5252367496490479,
+      "mean_token_accuracy": 0.8405626966403081,
+      "num_tokens": 2051331.0,
+      "step": 840
+    },
+    {
+      "epoch": 2.0241254523522314,
+      "eval_entropy": 0.6123150769579276,
+      "eval_loss": 0.6175946593284607,
+      "eval_mean_token_accuracy": 0.827637837509091,
+      "eval_num_tokens": 2051331.0,
+      "eval_runtime": 100.5414,
+      "eval_samples_per_second": 14.124,
+      "eval_steps_per_second": 1.77,
+      "step": 840
+    },
+    {
+      "entropy": 0.5359115816652775,
+      "epoch": 2.0723763570566947,
+      "grad_norm": 0.576781690120697,
+      "learning_rate": 0.0004174603237650678,
+      "loss": 0.5023207664489746,
+      "mean_token_accuracy": 0.8459693863987923,
+      "num_tokens": 2101528.0,
+      "step": 860
+    },
+    {
+      "epoch": 2.0723763570566947,
+      "eval_entropy": 0.5592134784111816,
+      "eval_loss": 0.620693564414978,
+      "eval_mean_token_accuracy": 0.827402507656076,
+      "eval_num_tokens": 2101528.0,
+      "eval_runtime": 100.6274,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 860
+    },
+    {
+      "entropy": 0.5214565828442573,
+      "epoch": 2.120627261761158,
+      "grad_norm": 0.45878875255584717,
+      "learning_rate": 0.00041610510539937234,
+      "loss": 0.48367843627929685,
+      "mean_token_accuracy": 0.8515541926026344,
+      "num_tokens": 2149550.0,
+      "step": 880
+    },
+    {
+      "epoch": 2.120627261761158,
+      "eval_entropy": 0.5860532344727034,
+      "eval_loss": 0.6134791970252991,
+      "eval_mean_token_accuracy": 0.8291774777214179,
+      "eval_num_tokens": 2149550.0,
+      "eval_runtime": 100.5295,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 880
+    },
+    {
+      "entropy": 0.5414379432797432,
+      "epoch": 2.168878166465621,
+      "grad_norm": 0.5976294875144958,
+      "learning_rate": 0.0004146933107930829,
+      "loss": 0.5012908935546875,
+      "mean_token_accuracy": 0.8466357633471489,
+      "num_tokens": 2202162.0,
+      "step": 900
+    },
+    {
+      "epoch": 2.168878166465621,
+      "eval_entropy": 0.5427783280611038,
+      "eval_loss": 0.612716019153595,
+      "eval_mean_token_accuracy": 0.829505049445656,
+      "eval_num_tokens": 2202162.0,
+      "eval_runtime": 100.5207,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 900
+    },
+    {
+      "entropy": 0.521108990907669,
+      "epoch": 2.2171290711700844,
+      "grad_norm": 0.5145775675773621,
+      "learning_rate": 0.000413225339467373,
+      "loss": 0.4877732276916504,
+      "mean_token_accuracy": 0.8504189744591713,
+      "num_tokens": 2256539.0,
+      "step": 920
+    },
+    {
+      "epoch": 2.2171290711700844,
+      "eval_entropy": 0.5620684869838565,
+      "eval_loss": 0.6176337599754333,
+      "eval_mean_token_accuracy": 0.8291444483767735,
+      "eval_num_tokens": 2256539.0,
+      "eval_runtime": 100.6819,
+      "eval_samples_per_second": 14.104,
+      "eval_steps_per_second": 1.768,
+      "step": 920
+    },
+    {
+      "entropy": 0.5452766507863999,
+      "epoch": 2.2653799758745476,
+      "grad_norm": 0.4820452928543091,
+      "learning_rate": 0.0004117016068407636,
+      "loss": 0.4991307735443115,
+      "mean_token_accuracy": 0.8474363312125206,
+      "num_tokens": 2299698.0,
+      "step": 940
+    },
+    {
+      "epoch": 2.2653799758745476,
+      "eval_entropy": 0.5611683193887218,
+      "eval_loss": 0.6153580546379089,
+      "eval_mean_token_accuracy": 0.8286041272490212,
+      "eval_num_tokens": 2299698.0,
+      "eval_runtime": 100.5249,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 940
+    },
+    {
+      "entropy": 0.5572576269507408,
+      "epoch": 2.313630880579011,
+      "grad_norm": 0.3947993814945221,
+      "learning_rate": 0.00041012254411156387,
+      "loss": 0.5191003799438476,
+      "mean_token_accuracy": 0.8422838419675827,
+      "num_tokens": 2347134.0,
+      "step": 960
+    },
+    {
+      "epoch": 2.313630880579011,
+      "eval_entropy": 0.5651392345682958,
+      "eval_loss": 0.6121005415916443,
+      "eval_mean_token_accuracy": 0.8286536242185014,
+      "eval_num_tokens": 2347134.0,
+      "eval_runtime": 100.7033,
+      "eval_samples_per_second": 14.101,
+      "eval_steps_per_second": 1.768,
+      "step": 960
+    },
+    {
+      "entropy": 0.5441671140491963,
+      "epoch": 2.361881785283474,
+      "grad_norm": 0.7719791531562805,
+      "learning_rate": 0.0004084885981358477,
+      "loss": 0.5072606086730957,
+      "mean_token_accuracy": 0.8478495761752128,
+      "num_tokens": 2394568.0,
+      "step": 980
+    },
+    {
+      "epoch": 2.361881785283474,
+      "eval_entropy": 0.5651004270891126,
+      "eval_loss": 0.6069501042366028,
+      "eval_mean_token_accuracy": 0.8311042156112328,
+      "eval_num_tokens": 2394568.0,
+      "eval_runtime": 100.5487,
+      "eval_samples_per_second": 14.123,
+      "eval_steps_per_second": 1.77,
+      "step": 980
+    },
+    {
+      "entropy": 0.5462074346840382,
+      "epoch": 2.4101326899879374,
+      "grad_norm": 0.5694436430931091,
+      "learning_rate": 0.00040680023130099856,
+      "loss": 0.5097150325775146,
+      "mean_token_accuracy": 0.8474388793110847,
+      "num_tokens": 2443253.0,
+      "step": 1000
+    },
+    {
+      "epoch": 2.4101326899879374,
+      "eval_entropy": 0.5895893243926295,
+      "eval_loss": 0.6078168749809265,
+      "eval_mean_token_accuracy": 0.828030709470256,
+      "eval_num_tokens": 2443253.0,
+      "eval_runtime": 100.5413,
+      "eval_samples_per_second": 14.124,
+      "eval_steps_per_second": 1.77,
+      "step": 1000
+    },
+    {
+      "entropy": 0.5473216332495212,
+      "epoch": 2.4583835946924006,
+      "grad_norm": 0.3989660143852234,
+      "learning_rate": 0.00040505792139485913,
+      "loss": 0.5087806224822998,
+      "mean_token_accuracy": 0.8473173275589942,
+      "num_tokens": 2495178.0,
+      "step": 1020
+    },
+    {
+      "epoch": 2.4583835946924006,
+      "eval_entropy": 0.5588909702354603,
+      "eval_loss": 0.6087605953216553,
+      "eval_mean_token_accuracy": 0.8309365524334854,
+      "eval_num_tokens": 2495178.0,
+      "eval_runtime": 100.5137,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 1020
+    }
+  ],
+  "logging_steps": 20,
+  "max_steps": 4150,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 20,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.17605631420416e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1040/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen3-14B-Base
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen3-14B-Base
+- lora
+- sft
+- transformers
+- trl
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1040/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-14B-Base",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 128,
+  "lora_bias": false,
+  "lora_dropout": 0.0539675908752296,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "gate_proj",
+    "q_proj",
+    "down_proj",
+    "o_proj",
+    "k_proj",
+    "v_proj",
+    "up_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1040/chat_template.jinja

@@ -0,0 +1,85 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1040/tokenizer_config.json

@@ -0,0 +1,29 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1040/trainer_state.json

@@ -0,0 +1,1126 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.5066344993968634,
+  "eval_steps": 20,
+  "global_step": 1040,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 1.8160341799259185,
+      "epoch": 0.04825090470446321,
+      "grad_norm": 1.1313843727111816,
+      "learning_rate": 1.9794859683580766e-05,
+      "loss": 1.7627403259277343,
+      "mean_token_accuracy": 0.6354069545865059,
+      "num_tokens": 50237.0,
+      "step": 20
+    },
+    {
+      "epoch": 0.04825090470446321,
+      "eval_entropy": 1.538598193881217,
+      "eval_loss": 1.4881398677825928,
+      "eval_mean_token_accuracy": 0.686459881201219,
+      "eval_num_tokens": 50237.0,
+      "eval_runtime": 100.8923,
+      "eval_samples_per_second": 14.074,
+      "eval_steps_per_second": 1.764,
+      "step": 20
+    },
+    {
+      "entropy": 1.1673978716135025,
+      "epoch": 0.09650180940892641,
+      "grad_norm": 0.6167025566101074,
+      "learning_rate": 4.0631554087349995e-05,
+      "loss": 1.0887075424194337,
+      "mean_token_accuracy": 0.737513878941536,
+      "num_tokens": 101098.0,
+      "step": 40
+    },
+    {
+      "epoch": 0.09650180940892641,
+      "eval_entropy": 0.887577452351538,
+      "eval_loss": 0.8519237041473389,
+      "eval_mean_token_accuracy": 0.7741017204322173,
+      "eval_num_tokens": 101098.0,
+      "eval_runtime": 100.4819,
+      "eval_samples_per_second": 14.132,
+      "eval_steps_per_second": 1.771,
+      "step": 40
+    },
+    {
+      "entropy": 0.8220535531640053,
+      "epoch": 0.14475271411338964,
+      "grad_norm": 0.8396430611610413,
+      "learning_rate": 6.146824849111923e-05,
+      "loss": 0.7812146186828614,
+      "mean_token_accuracy": 0.7872247830033302,
+      "num_tokens": 150385.0,
+      "step": 60
+    },
+    {
+      "epoch": 0.14475271411338964,
+      "eval_entropy": 0.770665143982748,
+      "eval_loss": 0.7478981614112854,
+      "eval_mean_token_accuracy": 0.7949167068084974,
+      "eval_num_tokens": 150385.0,
+      "eval_runtime": 100.6518,
+      "eval_samples_per_second": 14.108,
+      "eval_steps_per_second": 1.768,
+      "step": 60
+    },
+    {
+      "entropy": 0.7776695728302002,
+      "epoch": 0.19300361881785283,
+      "grad_norm": 0.5731202363967896,
+      "learning_rate": 8.230494289488845e-05,
+      "loss": 0.732721471786499,
+      "mean_token_accuracy": 0.7976997002959252,
+      "num_tokens": 197354.0,
+      "step": 80
+    },
+    {
+      "epoch": 0.19300361881785283,
+      "eval_entropy": 0.7650937618834249,
+      "eval_loss": 0.7109046578407288,
+      "eval_mean_token_accuracy": 0.8018777065732506,
+      "eval_num_tokens": 197354.0,
+      "eval_runtime": 100.5983,
+      "eval_samples_per_second": 14.116,
+      "eval_steps_per_second": 1.769,
+      "step": 80
+    },
+    {
+      "entropy": 0.7309209659695626,
+      "epoch": 0.24125452352231605,
+      "grad_norm": 0.48046866059303284,
+      "learning_rate": 0.00010314163729865769,
+      "loss": 0.680583143234253,
+      "mean_token_accuracy": 0.8073364540934562,
+      "num_tokens": 248557.0,
+      "step": 100
+    },
+    {
+      "epoch": 0.24125452352231605,
+      "eval_entropy": 0.7152130687504672,
+      "eval_loss": 0.6989707350730896,
+      "eval_mean_token_accuracy": 0.8057264320636064,
+      "eval_num_tokens": 248557.0,
+      "eval_runtime": 100.6264,
+      "eval_samples_per_second": 14.112,
+      "eval_steps_per_second": 1.769,
+      "step": 100
+    },
+    {
+      "entropy": 0.7105271726846695,
+      "epoch": 0.28950542822677927,
+      "grad_norm": 0.5512468814849854,
+      "learning_rate": 0.00012397833170242693,
+      "loss": 0.6674309253692627,
+      "mean_token_accuracy": 0.8098619893193245,
+      "num_tokens": 298041.0,
+      "step": 120
+    },
+    {
+      "epoch": 0.28950542822677927,
+      "eval_entropy": 0.7314498555124476,
+      "eval_loss": 0.6796280145645142,
+      "eval_mean_token_accuracy": 0.809063873264227,
+      "eval_num_tokens": 298041.0,
+      "eval_runtime": 100.6132,
+      "eval_samples_per_second": 14.113,
+      "eval_steps_per_second": 1.769,
+      "step": 120
+    },
+    {
+      "entropy": 0.7079856500029564,
+      "epoch": 0.33775633293124246,
+      "grad_norm": 0.5903022885322571,
+      "learning_rate": 0.00014481502610619613,
+      "loss": 0.6623000144958496,
+      "mean_token_accuracy": 0.8101129323244095,
+      "num_tokens": 348955.0,
+      "step": 140
+    },
+    {
+      "epoch": 0.33775633293124246,
+      "eval_entropy": 0.7005330203624254,
+      "eval_loss": 0.671779453754425,
+      "eval_mean_token_accuracy": 0.8106091374091888,
+      "eval_num_tokens": 348955.0,
+      "eval_runtime": 100.6904,
+      "eval_samples_per_second": 14.103,
+      "eval_steps_per_second": 1.768,
+      "step": 140
+    },
+    {
+      "entropy": 0.7109143257141113,
+      "epoch": 0.38600723763570566,
+      "grad_norm": 0.4516378343105316,
+      "learning_rate": 0.00016565172050996536,
+      "loss": 0.658915901184082,
+      "mean_token_accuracy": 0.8112782314419746,
+      "num_tokens": 395798.0,
+      "step": 160
+    },
+    {
+      "epoch": 0.38600723763570566,
+      "eval_entropy": 0.6891461650976974,
+      "eval_loss": 0.6598312258720398,
+      "eval_mean_token_accuracy": 0.8133360951134329,
+      "eval_num_tokens": 395798.0,
+      "eval_runtime": 100.5057,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 160
+    },
+    {
+      "entropy": 0.6896152347326279,
+      "epoch": 0.43425814234016885,
+      "grad_norm": 0.49479565024375916,
+      "learning_rate": 0.0001864884149137346,
+      "loss": 0.6466911315917969,
+      "mean_token_accuracy": 0.8141682580113411,
+      "num_tokens": 446437.0,
+      "step": 180
+    },
+    {
+      "epoch": 0.43425814234016885,
+      "eval_entropy": 0.6984815888860253,
+      "eval_loss": 0.652909517288208,
+      "eval_mean_token_accuracy": 0.8133517600847094,
+      "eval_num_tokens": 446437.0,
+      "eval_runtime": 100.5867,
+      "eval_samples_per_second": 14.117,
+      "eval_steps_per_second": 1.77,
+      "step": 180
+    },
+    {
+      "entropy": 0.6829591259360314,
+      "epoch": 0.4825090470446321,
+      "grad_norm": 0.4835337698459625,
+      "learning_rate": 0.00020732510931750382,
+      "loss": 0.6401876449584961,
+      "mean_token_accuracy": 0.8175502926111221,
+      "num_tokens": 497682.0,
+      "step": 200
+    },
+    {
+      "epoch": 0.4825090470446321,
+      "eval_entropy": 0.6963563973314306,
+      "eval_loss": 0.650087296962738,
+      "eval_mean_token_accuracy": 0.8159684401549651,
+      "eval_num_tokens": 497682.0,
+      "eval_runtime": 100.6277,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 200
+    },
+    {
+      "entropy": 0.6903425320982933,
+      "epoch": 0.5307599517490953,
+      "grad_norm": 0.44980454444885254,
+      "learning_rate": 0.00022816180372127305,
+      "loss": 0.6409444808959961,
+      "mean_token_accuracy": 0.8163257524371147,
+      "num_tokens": 544670.0,
+      "step": 220
+    },
+    {
+      "epoch": 0.5307599517490953,
+      "eval_entropy": 0.6687790182199371,
+      "eval_loss": 0.6451645493507385,
+      "eval_mean_token_accuracy": 0.8157875551266617,
+      "eval_num_tokens": 544670.0,
+      "eval_runtime": 100.5444,
+      "eval_samples_per_second": 14.123,
+      "eval_steps_per_second": 1.77,
+      "step": 220
+    },
+    {
+      "entropy": 0.6740772560238838,
+      "epoch": 0.5790108564535585,
+      "grad_norm": 0.5312877297401428,
+      "learning_rate": 0.0002489984981250423,
+      "loss": 0.6307676315307618,
+      "mean_token_accuracy": 0.8174169659614563,
+      "num_tokens": 592926.0,
+      "step": 240
+    },
+    {
+      "epoch": 0.5790108564535585,
+      "eval_entropy": 0.7063750872451268,
+      "eval_loss": 0.6494986414909363,
+      "eval_mean_token_accuracy": 0.8156777198395032,
+      "eval_num_tokens": 592926.0,
+      "eval_runtime": 100.5094,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 240
+    },
+    {
+      "entropy": 0.6876470118761062,
+      "epoch": 0.6272617611580217,
+      "grad_norm": 0.46149370074272156,
+      "learning_rate": 0.00026983519252881153,
+      "loss": 0.6392203807830811,
+      "mean_token_accuracy": 0.8172321930527687,
+      "num_tokens": 639077.0,
+      "step": 260
+    },
+    {
+      "epoch": 0.6272617611580217,
+      "eval_entropy": 0.625594850838854,
+      "eval_loss": 0.6463819742202759,
+      "eval_mean_token_accuracy": 0.8179670525400826,
+      "eval_num_tokens": 639077.0,
+      "eval_runtime": 100.5775,
+      "eval_samples_per_second": 14.118,
+      "eval_steps_per_second": 1.77,
+      "step": 260
+    },
+    {
+      "entropy": 0.6673879995942116,
+      "epoch": 0.6755126658624849,
+      "grad_norm": 0.42477288842201233,
+      "learning_rate": 0.0002906718869325807,
+      "loss": 0.6300291538238525,
+      "mean_token_accuracy": 0.8203139930963517,
+      "num_tokens": 691028.0,
+      "step": 280
+    },
+    {
+      "epoch": 0.6755126658624849,
+      "eval_entropy": 0.6863534494062488,
+      "eval_loss": 0.6392037272453308,
+      "eval_mean_token_accuracy": 0.8190520385677895,
+      "eval_num_tokens": 691028.0,
+      "eval_runtime": 100.5106,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 280
+    },
+    {
+      "entropy": 0.6573980256915093,
+      "epoch": 0.7237635705669482,
+      "grad_norm": 0.4656430780887604,
+      "learning_rate": 0.00031150858133635,
+      "loss": 0.6215964794158936,
+      "mean_token_accuracy": 0.8208559453487396,
+      "num_tokens": 740255.0,
+      "step": 300
+    },
+    {
+      "epoch": 0.7237635705669482,
+      "eval_entropy": 0.6834242812703165,
+      "eval_loss": 0.6479960083961487,
+      "eval_mean_token_accuracy": 0.8128752906001016,
+      "eval_num_tokens": 740255.0,
+      "eval_runtime": 100.6101,
+      "eval_samples_per_second": 14.114,
+      "eval_steps_per_second": 1.769,
+      "step": 300
+    },
+    {
+      "entropy": 0.6751310348510742,
+      "epoch": 0.7720144752714113,
+      "grad_norm": 0.4495522081851959,
+      "learning_rate": 0.00033234527574011916,
+      "loss": 0.6402668952941895,
+      "mean_token_accuracy": 0.8170571058988572,
+      "num_tokens": 789529.0,
+      "step": 320
+    },
+    {
+      "epoch": 0.7720144752714113,
+      "eval_entropy": 0.6920045178927733,
+      "eval_loss": 0.6402787566184998,
+      "eval_mean_token_accuracy": 0.817791668886549,
+      "eval_num_tokens": 789529.0,
+      "eval_runtime": 100.6347,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 320
+    },
+    {
+      "entropy": 0.6846290081739426,
+      "epoch": 0.8202653799758746,
+      "grad_norm": 0.5661432147026062,
+      "learning_rate": 0.00035318197014388845,
+      "loss": 0.6508871078491211,
+      "mean_token_accuracy": 0.8164169207215309,
+      "num_tokens": 841584.0,
+      "step": 340
+    },
+    {
+      "epoch": 0.8202653799758746,
+      "eval_entropy": 0.6772527078564248,
+      "eval_loss": 0.6512064337730408,
+      "eval_mean_token_accuracy": 0.8170605137776793,
+      "eval_num_tokens": 841584.0,
+      "eval_runtime": 100.5157,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 340
+    },
+    {
+      "entropy": 0.7022885024547577,
+      "epoch": 0.8685162846803377,
+      "grad_norm": 0.6357236504554749,
+      "learning_rate": 0.0003740186645476577,
+      "loss": 0.6649135112762451,
+      "mean_token_accuracy": 0.8136772960424423,
+      "num_tokens": 885769.0,
+      "step": 360
+    },
+    {
+      "epoch": 0.8685162846803377,
+      "eval_entropy": 0.7435127543599418,
+      "eval_loss": 0.6573262214660645,
+      "eval_mean_token_accuracy": 0.8147802225659403,
+      "eval_num_tokens": 885769.0,
+      "eval_runtime": 100.5271,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 360
+    },
+    {
+      "entropy": 0.6962979376316071,
+      "epoch": 0.916767189384801,
+      "grad_norm": 0.6179454326629639,
+      "learning_rate": 0.00039485535895142685,
+      "loss": 0.6509191513061523,
+      "mean_token_accuracy": 0.81364226937294,
+      "num_tokens": 930056.0,
+      "step": 380
+    },
+    {
+      "epoch": 0.916767189384801,
+      "eval_entropy": 0.6630894883295123,
+      "eval_loss": 0.6524088978767395,
+      "eval_mean_token_accuracy": 0.8165898587596551,
+      "eval_num_tokens": 930056.0,
+      "eval_runtime": 100.5343,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 380
+    },
+    {
+      "entropy": 0.694319449365139,
+      "epoch": 0.9650180940892642,
+      "grad_norm": 0.5458664894104004,
+      "learning_rate": 0.00041569205335519613,
+      "loss": 0.6624011993408203,
+      "mean_token_accuracy": 0.8136080905795098,
+      "num_tokens": 976892.0,
+      "step": 400
+    },
+    {
+      "epoch": 0.9650180940892642,
+      "eval_entropy": 0.7424925324622165,
+      "eval_loss": 0.6717661023139954,
+      "eval_mean_token_accuracy": 0.8099894781460922,
+      "eval_num_tokens": 976892.0,
+      "eval_runtime": 100.5992,
+      "eval_samples_per_second": 14.115,
+      "eval_steps_per_second": 1.769,
+      "step": 400
+    },
+    {
+      "entropy": 0.6861714736009256,
+      "epoch": 1.0120627261761157,
+      "grad_norm": 0.6108037829399109,
+      "learning_rate": 0.0004323601853188671,
+      "loss": 0.6551628112792969,
+      "mean_token_accuracy": 0.8134882618219424,
+      "num_tokens": 1024472.0,
+      "step": 420
+    },
+    {
+      "epoch": 1.0120627261761157,
+      "eval_entropy": 0.6593745705786715,
+      "eval_loss": 0.6720747947692871,
+      "eval_mean_token_accuracy": 0.812802647941568,
+      "eval_num_tokens": 1024472.0,
+      "eval_runtime": 100.4866,
+      "eval_samples_per_second": 14.131,
+      "eval_steps_per_second": 1.771,
+      "step": 420
+    },
+    {
+      "entropy": 0.639520151913166,
+      "epoch": 1.060313630880579,
+      "grad_norm": 0.6054353713989258,
+      "learning_rate": 0.00043231736219609266,
+      "loss": 0.6096667766571044,
+      "mean_token_accuracy": 0.8227144703269005,
+      "num_tokens": 1076964.0,
+      "step": 440
+    },
+    {
+      "epoch": 1.060313630880579,
+      "eval_entropy": 0.6316307876886946,
+      "eval_loss": 0.6621950268745422,
+      "eval_mean_token_accuracy": 0.8140057914712456,
+      "eval_num_tokens": 1076964.0,
+      "eval_runtime": 100.6409,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 440
+    },
+    {
+      "entropy": 0.6544070363044738,
+      "epoch": 1.1085645355850422,
+      "grad_norm": 0.5111057758331299,
+      "learning_rate": 0.0004322133749558233,
+      "loss": 0.6176822185516357,
+      "mean_token_accuracy": 0.8201584875583648,
+      "num_tokens": 1126064.0,
+      "step": 460
+    },
+    {
+      "epoch": 1.1085645355850422,
+      "eval_entropy": 0.6714627766207363,
+      "eval_loss": 0.6738516092300415,
+      "eval_mean_token_accuracy": 0.8128616829936424,
+      "eval_num_tokens": 1126064.0,
+      "eval_runtime": 100.4986,
+      "eval_samples_per_second": 14.13,
+      "eval_steps_per_second": 1.771,
+      "step": 460
+    },
+    {
+      "entropy": 0.6470697179436684,
+      "epoch": 1.1568154402895054,
+      "grad_norm": 0.5434669256210327,
+      "learning_rate": 0.0004320482530252179,
+      "loss": 0.6144762516021729,
+      "mean_token_accuracy": 0.819926056265831,
+      "num_tokens": 1176912.0,
+      "step": 480
+    },
+    {
+      "epoch": 1.1568154402895054,
+      "eval_entropy": 0.6475969080844622,
+      "eval_loss": 0.6624214053153992,
+      "eval_mean_token_accuracy": 0.8137228820431098,
+      "eval_num_tokens": 1176912.0,
+      "eval_runtime": 100.5617,
+      "eval_samples_per_second": 14.121,
+      "eval_steps_per_second": 1.77,
+      "step": 480
+    },
+    {
+      "entropy": 0.6659130036830903,
+      "epoch": 1.2050663449939687,
+      "grad_norm": 0.6470707058906555,
+      "learning_rate": 0.0004318220431318296,
+      "loss": 0.6366308212280274,
+      "mean_token_accuracy": 0.8175261527299881,
+      "num_tokens": 1223383.0,
+      "step": 500
+    },
+    {
+      "epoch": 1.2050663449939687,
+      "eval_entropy": 0.7232425132494295,
+      "eval_loss": 0.6877738833427429,
+      "eval_mean_token_accuracy": 0.8007767260744331,
+      "eval_num_tokens": 1223383.0,
+      "eval_runtime": 100.477,
+      "eval_samples_per_second": 14.133,
+      "eval_steps_per_second": 1.772,
+      "step": 500
+    },
+    {
+      "entropy": 0.6759189993143082,
+      "epoch": 1.2533172496984317,
+      "grad_norm": 0.6453561186790466,
+      "learning_rate": 0.00043153480929038245,
+      "loss": 0.637772512435913,
+      "mean_token_accuracy": 0.8162627041339874,
+      "num_tokens": 1273534.0,
+      "step": 520
+    },
+    {
+      "epoch": 1.2533172496984317,
+      "eval_entropy": 0.7082584312792575,
+      "eval_loss": 0.6695764064788818,
+      "eval_mean_token_accuracy": 0.812019992410467,
+      "eval_num_tokens": 1273534.0,
+      "eval_runtime": 100.4957,
+      "eval_samples_per_second": 14.13,
+      "eval_steps_per_second": 1.771,
+      "step": 520
+    },
+    {
+      "entropy": 0.6730152070522308,
+      "epoch": 1.301568154402895,
+      "grad_norm": 0.6234476566314697,
+      "learning_rate": 0.000431186632784656,
+      "loss": 0.640014362335205,
+      "mean_token_accuracy": 0.8168858170509339,
+      "num_tokens": 1323351.0,
+      "step": 540
+    },
+    {
+      "epoch": 1.301568154402895,
+      "eval_entropy": 0.6655137846978862,
+      "eval_loss": 0.6578248739242554,
+      "eval_mean_token_accuracy": 0.8147600979617472,
+      "eval_num_tokens": 1323351.0,
+      "eval_runtime": 100.503,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 540
+    },
+    {
+      "entropy": 0.6636043325066566,
+      "epoch": 1.3498190591073582,
+      "grad_norm": 0.5882512331008911,
+      "learning_rate": 0.0004307776121444827,
+      "loss": 0.6286775588989257,
+      "mean_token_accuracy": 0.8183174058794975,
+      "num_tokens": 1371508.0,
+      "step": 560
+    },
+    {
+      "epoch": 1.3498190591073582,
+      "eval_entropy": 0.6304813810948575,
+      "eval_loss": 0.6607436537742615,
+      "eval_mean_token_accuracy": 0.8169659268989992,
+      "eval_num_tokens": 1371508.0,
+      "eval_runtime": 100.6233,
+      "eval_samples_per_second": 14.112,
+      "eval_steps_per_second": 1.769,
+      "step": 560
+    },
+    {
+      "entropy": 0.6608738481998444,
+      "epoch": 1.3980699638118215,
+      "grad_norm": 0.6796398758888245,
+      "learning_rate": 0.00043030786311786564,
+      "loss": 0.622726583480835,
+      "mean_token_accuracy": 0.8230827242136002,
+      "num_tokens": 1415221.0,
+      "step": 580
+    },
+    {
+      "epoch": 1.3980699638118215,
+      "eval_entropy": 0.6645242689030894,
+      "eval_loss": 0.6662811636924744,
+      "eval_mean_token_accuracy": 0.8147623907314258,
+      "eval_num_tokens": 1415221.0,
+      "eval_runtime": 100.5074,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 580
+    },
+    {
+      "entropy": 0.6682895936071873,
+      "epoch": 1.4463208685162847,
+      "grad_norm": 0.5750718712806702,
+      "learning_rate": 0.0004297775186382226,
+      "loss": 0.6304276466369629,
+      "mean_token_accuracy": 0.8188625335693359,
+      "num_tokens": 1463356.0,
+      "step": 600
+    },
+    {
+      "epoch": 1.4463208685162847,
+      "eval_entropy": 0.6320532578765676,
+      "eval_loss": 0.650333046913147,
+      "eval_mean_token_accuracy": 0.8172853793321031,
+      "eval_num_tokens": 1463356.0,
+      "eval_runtime": 100.5116,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 600
+    },
+    {
+      "entropy": 0.653739221394062,
+      "epoch": 1.494571773220748,
+      "grad_norm": 0.5552534461021423,
+      "learning_rate": 0.00042918672878676804,
+      "loss": 0.6215689659118653,
+      "mean_token_accuracy": 0.8208896175026894,
+      "num_tokens": 1512340.0,
+      "step": 620
+    },
+    {
+      "epoch": 1.494571773220748,
+      "eval_entropy": 0.6285932062382109,
+      "eval_loss": 0.6392536163330078,
+      "eval_mean_token_accuracy": 0.8205167825302381,
+      "eval_num_tokens": 1512340.0,
+      "eval_runtime": 100.6392,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 620
+    },
+    {
+      "entropy": 0.6476897612214089,
+      "epoch": 1.5428226779252112,
+      "grad_norm": 0.536978542804718,
+      "learning_rate": 0.00042853566075004156,
+      "loss": 0.6131608009338378,
+      "mean_token_accuracy": 0.8208152890205384,
+      "num_tokens": 1560911.0,
+      "step": 640
+    },
+    {
+      "epoch": 1.5428226779252112,
+      "eval_entropy": 0.6626498856571283,
+      "eval_loss": 0.6479025483131409,
+      "eval_mean_token_accuracy": 0.8180894101603647,
+      "eval_num_tokens": 1560911.0,
+      "eval_runtime": 100.5035,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 640
+    },
+    {
+      "entropy": 0.6338633447885513,
+      "epoch": 1.5910735826296745,
+      "grad_norm": 0.4518965780735016,
+      "learning_rate": 0.0004278244987725961,
+      "loss": 0.6010610580444335,
+      "mean_token_accuracy": 0.8247816175222397,
+      "num_tokens": 1609409.0,
+      "step": 660
+    },
+    {
+      "epoch": 1.5910735826296745,
+      "eval_entropy": 0.64079723699709,
+      "eval_loss": 0.6417157053947449,
+      "eval_mean_token_accuracy": 0.8214994581897607,
+      "eval_num_tokens": 1609409.0,
+      "eval_runtime": 100.6393,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 660
+    },
+    {
+      "entropy": 0.6485912457108498,
+      "epoch": 1.6393244873341375,
+      "grad_norm": 0.6028368473052979,
+      "learning_rate": 0.00042705344410485905,
+      "loss": 0.6120437145233154,
+      "mean_token_accuracy": 0.8236212745308876,
+      "num_tokens": 1657430.0,
+      "step": 680
+    },
+    {
+      "epoch": 1.6393244873341375,
+      "eval_entropy": 0.6863006427046958,
+      "eval_loss": 0.6420931816101074,
+      "eval_mean_token_accuracy": 0.8191533580924688,
+      "eval_num_tokens": 1657430.0,
+      "eval_runtime": 100.4639,
+      "eval_samples_per_second": 14.134,
+      "eval_steps_per_second": 1.772,
+      "step": 680
+    },
+    {
+      "entropy": 0.6459382772445679,
+      "epoch": 1.6875753920386007,
+      "grad_norm": 0.5069310069084167,
+      "learning_rate": 0.00042622271494618073,
+      "loss": 0.6160889625549316,
+      "mean_token_accuracy": 0.8253828629851341,
+      "num_tokens": 1708058.0,
+      "step": 700
+    },
+    {
+      "epoch": 1.6875753920386007,
+      "eval_entropy": 0.6773919463157654,
+      "eval_loss": 0.6307896375656128,
+      "eval_mean_token_accuracy": 0.8217989825800563,
+      "eval_num_tokens": 1708058.0,
+      "eval_runtime": 100.6301,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 700
+    },
+    {
+      "entropy": 0.64687028080225,
+      "epoch": 1.7358262967430638,
+      "grad_norm": 0.3830573856830597,
+      "learning_rate": 0.00042533254638308645,
+      "loss": 0.6104001998901367,
+      "mean_token_accuracy": 0.8255606815218925,
+      "num_tokens": 1761262.0,
+      "step": 720
+    },
+    {
+      "epoch": 1.7358262967430638,
+      "eval_entropy": 0.6179541145147902,
+      "eval_loss": 0.6170215606689453,
+      "eval_mean_token_accuracy": 0.8259220900160543,
+      "eval_num_tokens": 1761262.0,
+      "eval_runtime": 100.61,
+      "eval_samples_per_second": 14.114,
+      "eval_steps_per_second": 1.769,
+      "step": 720
+    },
+    {
+      "entropy": 0.6306161753833294,
+      "epoch": 1.784077201447527,
+      "grad_norm": 0.5888957977294922,
+      "learning_rate": 0.0004243831903227498,
+      "loss": 0.5981546878814697,
+      "mean_token_accuracy": 0.8266776099801063,
+      "num_tokens": 1812127.0,
+      "step": 740
+    },
+    {
+      "epoch": 1.784077201447527,
+      "eval_entropy": 0.6357416808605194,
+      "eval_loss": 0.6202242374420166,
+      "eval_mean_token_accuracy": 0.8253380304641937,
+      "eval_num_tokens": 1812127.0,
+      "eval_runtime": 100.5145,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 740
+    },
+    {
+      "entropy": 0.6290688961744308,
+      "epoch": 1.8323281061519903,
+      "grad_norm": 0.6138001680374146,
+      "learning_rate": 0.0004233749154217063,
+      "loss": 0.5853167057037354,
+      "mean_token_accuracy": 0.8281400203704834,
+      "num_tokens": 1858403.0,
+      "step": 760
+    },
+    {
+      "epoch": 1.8323281061519903,
+      "eval_entropy": 0.6402715869833914,
+      "eval_loss": 0.6168262958526611,
+      "eval_mean_token_accuracy": 0.8246346306934785,
+      "eval_num_tokens": 1858403.0,
+      "eval_runtime": 100.5237,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 760
+    },
+    {
+      "entropy": 0.6218615680932998,
+      "epoch": 1.8805790108564535,
+      "grad_norm": 0.574312686920166,
+      "learning_rate": 0.00042230800700982586,
+      "loss": 0.5841915607452393,
+      "mean_token_accuracy": 0.8304707780480385,
+      "num_tokens": 1907361.0,
+      "step": 780
+    },
+    {
+      "epoch": 1.8805790108564535,
+      "eval_entropy": 0.6301163148679091,
+      "eval_loss": 0.6073815226554871,
+      "eval_mean_token_accuracy": 0.8274695993809218,
+      "eval_num_tokens": 1907361.0,
+      "eval_runtime": 100.5111,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 780
+    },
+    {
+      "entropy": 0.6101033873856068,
+      "epoch": 1.9288299155609168,
+      "grad_norm": 0.5054804682731628,
+      "learning_rate": 0.000421182767009569,
+      "loss": 0.5774452209472656,
+      "mean_token_accuracy": 0.8321943372488022,
+      "num_tokens": 1956605.0,
+      "step": 800
+    },
+    {
+      "epoch": 1.9288299155609168,
+      "eval_entropy": 0.631400960215022,
+      "eval_loss": 0.6115217804908752,
+      "eval_mean_token_accuracy": 0.8249618762664581,
+      "eval_num_tokens": 1956605.0,
+      "eval_runtime": 100.4916,
+      "eval_samples_per_second": 14.131,
+      "eval_steps_per_second": 1.771,
+      "step": 800
+    },
+    {
+      "entropy": 0.6217561915516854,
+      "epoch": 1.97708082026538,
+      "grad_norm": 0.5333649516105652,
+      "learning_rate": 0.0004199995138505454,
+      "loss": 0.5842787265777588,
+      "mean_token_accuracy": 0.8305525362491608,
+      "num_tokens": 2003308.0,
+      "step": 820
+    },
+    {
+      "epoch": 1.97708082026538,
+      "eval_entropy": 0.605928597657868,
+      "eval_loss": 0.6067516207695007,
+      "eval_mean_token_accuracy": 0.82715918408351,
+      "eval_num_tokens": 2003308.0,
+      "eval_runtime": 100.5286,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 820
+    },
+    {
+      "entropy": 0.5585076526189462,
+      "epoch": 2.0241254523522314,
+      "grad_norm": 0.44819608330726624,
+      "learning_rate": 0.00041875858237940244,
+      "loss": 0.5252367496490479,
+      "mean_token_accuracy": 0.8405626966403081,
+      "num_tokens": 2051331.0,
+      "step": 840
+    },
+    {
+      "epoch": 2.0241254523522314,
+      "eval_entropy": 0.6123150769579276,
+      "eval_loss": 0.6175946593284607,
+      "eval_mean_token_accuracy": 0.827637837509091,
+      "eval_num_tokens": 2051331.0,
+      "eval_runtime": 100.5414,
+      "eval_samples_per_second": 14.124,
+      "eval_steps_per_second": 1.77,
+      "step": 840
+    },
+    {
+      "entropy": 0.5359115816652775,
+      "epoch": 2.0723763570566947,
+      "grad_norm": 0.576781690120697,
+      "learning_rate": 0.0004174603237650678,
+      "loss": 0.5023207664489746,
+      "mean_token_accuracy": 0.8459693863987923,
+      "num_tokens": 2101528.0,
+      "step": 860
+    },
+    {
+      "epoch": 2.0723763570566947,
+      "eval_entropy": 0.5592134784111816,
+      "eval_loss": 0.620693564414978,
+      "eval_mean_token_accuracy": 0.827402507656076,
+      "eval_num_tokens": 2101528.0,
+      "eval_runtime": 100.6274,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 860
+    },
+    {
+      "entropy": 0.5214565828442573,
+      "epoch": 2.120627261761158,
+      "grad_norm": 0.45878875255584717,
+      "learning_rate": 0.00041610510539937234,
+      "loss": 0.48367843627929685,
+      "mean_token_accuracy": 0.8515541926026344,
+      "num_tokens": 2149550.0,
+      "step": 880
+    },
+    {
+      "epoch": 2.120627261761158,
+      "eval_entropy": 0.5860532344727034,
+      "eval_loss": 0.6134791970252991,
+      "eval_mean_token_accuracy": 0.8291774777214179,
+      "eval_num_tokens": 2149550.0,
+      "eval_runtime": 100.5295,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 880
+    },
+    {
+      "entropy": 0.5414379432797432,
+      "epoch": 2.168878166465621,
+      "grad_norm": 0.5976294875144958,
+      "learning_rate": 0.0004146933107930829,
+      "loss": 0.5012908935546875,
+      "mean_token_accuracy": 0.8466357633471489,
+      "num_tokens": 2202162.0,
+      "step": 900
+    },
+    {
+      "epoch": 2.168878166465621,
+      "eval_entropy": 0.5427783280611038,
+      "eval_loss": 0.612716019153595,
+      "eval_mean_token_accuracy": 0.829505049445656,
+      "eval_num_tokens": 2202162.0,
+      "eval_runtime": 100.5207,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 900
+    },
+    {
+      "entropy": 0.521108990907669,
+      "epoch": 2.2171290711700844,
+      "grad_norm": 0.5145775675773621,
+      "learning_rate": 0.000413225339467373,
+      "loss": 0.4877732276916504,
+      "mean_token_accuracy": 0.8504189744591713,
+      "num_tokens": 2256539.0,
+      "step": 920
+    },
+    {
+      "epoch": 2.2171290711700844,
+      "eval_entropy": 0.5620684869838565,
+      "eval_loss": 0.6176337599754333,
+      "eval_mean_token_accuracy": 0.8291444483767735,
+      "eval_num_tokens": 2256539.0,
+      "eval_runtime": 100.6819,
+      "eval_samples_per_second": 14.104,
+      "eval_steps_per_second": 1.768,
+      "step": 920
+    },
+    {
+      "entropy": 0.5452766507863999,
+      "epoch": 2.2653799758745476,
+      "grad_norm": 0.4820452928543091,
+      "learning_rate": 0.0004117016068407636,
+      "loss": 0.4991307735443115,
+      "mean_token_accuracy": 0.8474363312125206,
+      "num_tokens": 2299698.0,
+      "step": 940
+    },
+    {
+      "epoch": 2.2653799758745476,
+      "eval_entropy": 0.5611683193887218,
+      "eval_loss": 0.6153580546379089,
+      "eval_mean_token_accuracy": 0.8286041272490212,
+      "eval_num_tokens": 2299698.0,
+      "eval_runtime": 100.5249,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 940
+    },
+    {
+      "entropy": 0.5572576269507408,
+      "epoch": 2.313630880579011,
+      "grad_norm": 0.3947993814945221,
+      "learning_rate": 0.00041012254411156387,
+      "loss": 0.5191003799438476,
+      "mean_token_accuracy": 0.8422838419675827,
+      "num_tokens": 2347134.0,
+      "step": 960
+    },
+    {
+      "epoch": 2.313630880579011,
+      "eval_entropy": 0.5651392345682958,
+      "eval_loss": 0.6121005415916443,
+      "eval_mean_token_accuracy": 0.8286536242185014,
+      "eval_num_tokens": 2347134.0,
+      "eval_runtime": 100.7033,
+      "eval_samples_per_second": 14.101,
+      "eval_steps_per_second": 1.768,
+      "step": 960
+    },
+    {
+      "entropy": 0.5441671140491963,
+      "epoch": 2.361881785283474,
+      "grad_norm": 0.7719791531562805,
+      "learning_rate": 0.0004084885981358477,
+      "loss": 0.5072606086730957,
+      "mean_token_accuracy": 0.8478495761752128,
+      "num_tokens": 2394568.0,
+      "step": 980
+    },
+    {
+      "epoch": 2.361881785283474,
+      "eval_entropy": 0.5651004270891126,
+      "eval_loss": 0.6069501042366028,
+      "eval_mean_token_accuracy": 0.8311042156112328,
+      "eval_num_tokens": 2394568.0,
+      "eval_runtime": 100.5487,
+      "eval_samples_per_second": 14.123,
+      "eval_steps_per_second": 1.77,
+      "step": 980
+    },
+    {
+      "entropy": 0.5462074346840382,
+      "epoch": 2.4101326899879374,
+      "grad_norm": 0.5694436430931091,
+      "learning_rate": 0.00040680023130099856,
+      "loss": 0.5097150325775146,
+      "mean_token_accuracy": 0.8474388793110847,
+      "num_tokens": 2443253.0,
+      "step": 1000
+    },
+    {
+      "epoch": 2.4101326899879374,
+      "eval_entropy": 0.5895893243926295,
+      "eval_loss": 0.6078168749809265,
+      "eval_mean_token_accuracy": 0.828030709470256,
+      "eval_num_tokens": 2443253.0,
+      "eval_runtime": 100.5413,
+      "eval_samples_per_second": 14.124,
+      "eval_steps_per_second": 1.77,
+      "step": 1000
+    },
+    {
+      "entropy": 0.5473216332495212,
+      "epoch": 2.4583835946924006,
+      "grad_norm": 0.3989660143852234,
+      "learning_rate": 0.00040505792139485913,
+      "loss": 0.5087806224822998,
+      "mean_token_accuracy": 0.8473173275589942,
+      "num_tokens": 2495178.0,
+      "step": 1020
+    },
+    {
+      "epoch": 2.4583835946924006,
+      "eval_entropy": 0.5588909702354603,
+      "eval_loss": 0.6087605953216553,
+      "eval_mean_token_accuracy": 0.8309365524334854,
+      "eval_num_tokens": 2495178.0,
+      "eval_runtime": 100.5137,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 1020
+    },
+    {
+      "entropy": 0.5430163390934467,
+      "epoch": 2.5066344993968634,
+      "grad_norm": 0.4916608929634094,
+      "learning_rate": 0.00040326216147052336,
+      "loss": 0.5027707576751709,
+      "mean_token_accuracy": 0.8466959804296493,
+      "num_tokens": 2541733.0,
+      "step": 1040
+    },
+    {
+      "epoch": 2.5066344993968634,
+      "eval_entropy": 0.5819033988741007,
+      "eval_loss": 0.609398365020752,
+      "eval_mean_token_accuracy": 0.8297666612635838,
+      "eval_num_tokens": 2541733.0,
+      "eval_runtime": 100.6375,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 1040
+    }
+  ],
+  "logging_steps": 20,
+  "max_steps": 4150,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 20,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.27477297872896e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1060/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen3-14B-Base
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen3-14B-Base
+- lora
+- sft
+- transformers
+- trl
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1060/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-14B-Base",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 128,
+  "lora_bias": false,
+  "lora_dropout": 0.0539675908752296,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "gate_proj",
+    "q_proj",
+    "down_proj",
+    "o_proj",
+    "k_proj",
+    "v_proj",
+    "up_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1060/chat_template.jinja

@@ -0,0 +1,85 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1060/tokenizer_config.json

@@ -0,0 +1,29 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1060/trainer_state.json

@@ -0,0 +1,1147 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.554885404101327,
+  "eval_steps": 20,
+  "global_step": 1060,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 1.8160341799259185,
+      "epoch": 0.04825090470446321,
+      "grad_norm": 1.1313843727111816,
+      "learning_rate": 1.9794859683580766e-05,
+      "loss": 1.7627403259277343,
+      "mean_token_accuracy": 0.6354069545865059,
+      "num_tokens": 50237.0,
+      "step": 20
+    },
+    {
+      "epoch": 0.04825090470446321,
+      "eval_entropy": 1.538598193881217,
+      "eval_loss": 1.4881398677825928,
+      "eval_mean_token_accuracy": 0.686459881201219,
+      "eval_num_tokens": 50237.0,
+      "eval_runtime": 100.8923,
+      "eval_samples_per_second": 14.074,
+      "eval_steps_per_second": 1.764,
+      "step": 20
+    },
+    {
+      "entropy": 1.1673978716135025,
+      "epoch": 0.09650180940892641,
+      "grad_norm": 0.6167025566101074,
+      "learning_rate": 4.0631554087349995e-05,
+      "loss": 1.0887075424194337,
+      "mean_token_accuracy": 0.737513878941536,
+      "num_tokens": 101098.0,
+      "step": 40
+    },
+    {
+      "epoch": 0.09650180940892641,
+      "eval_entropy": 0.887577452351538,
+      "eval_loss": 0.8519237041473389,
+      "eval_mean_token_accuracy": 0.7741017204322173,
+      "eval_num_tokens": 101098.0,
+      "eval_runtime": 100.4819,
+      "eval_samples_per_second": 14.132,
+      "eval_steps_per_second": 1.771,
+      "step": 40
+    },
+    {
+      "entropy": 0.8220535531640053,
+      "epoch": 0.14475271411338964,
+      "grad_norm": 0.8396430611610413,
+      "learning_rate": 6.146824849111923e-05,
+      "loss": 0.7812146186828614,
+      "mean_token_accuracy": 0.7872247830033302,
+      "num_tokens": 150385.0,
+      "step": 60
+    },
+    {
+      "epoch": 0.14475271411338964,
+      "eval_entropy": 0.770665143982748,
+      "eval_loss": 0.7478981614112854,
+      "eval_mean_token_accuracy": 0.7949167068084974,
+      "eval_num_tokens": 150385.0,
+      "eval_runtime": 100.6518,
+      "eval_samples_per_second": 14.108,
+      "eval_steps_per_second": 1.768,
+      "step": 60
+    },
+    {
+      "entropy": 0.7776695728302002,
+      "epoch": 0.19300361881785283,
+      "grad_norm": 0.5731202363967896,
+      "learning_rate": 8.230494289488845e-05,
+      "loss": 0.732721471786499,
+      "mean_token_accuracy": 0.7976997002959252,
+      "num_tokens": 197354.0,
+      "step": 80
+    },
+    {
+      "epoch": 0.19300361881785283,
+      "eval_entropy": 0.7650937618834249,
+      "eval_loss": 0.7109046578407288,
+      "eval_mean_token_accuracy": 0.8018777065732506,
+      "eval_num_tokens": 197354.0,
+      "eval_runtime": 100.5983,
+      "eval_samples_per_second": 14.116,
+      "eval_steps_per_second": 1.769,
+      "step": 80
+    },
+    {
+      "entropy": 0.7309209659695626,
+      "epoch": 0.24125452352231605,
+      "grad_norm": 0.48046866059303284,
+      "learning_rate": 0.00010314163729865769,
+      "loss": 0.680583143234253,
+      "mean_token_accuracy": 0.8073364540934562,
+      "num_tokens": 248557.0,
+      "step": 100
+    },
+    {
+      "epoch": 0.24125452352231605,
+      "eval_entropy": 0.7152130687504672,
+      "eval_loss": 0.6989707350730896,
+      "eval_mean_token_accuracy": 0.8057264320636064,
+      "eval_num_tokens": 248557.0,
+      "eval_runtime": 100.6264,
+      "eval_samples_per_second": 14.112,
+      "eval_steps_per_second": 1.769,
+      "step": 100
+    },
+    {
+      "entropy": 0.7105271726846695,
+      "epoch": 0.28950542822677927,
+      "grad_norm": 0.5512468814849854,
+      "learning_rate": 0.00012397833170242693,
+      "loss": 0.6674309253692627,
+      "mean_token_accuracy": 0.8098619893193245,
+      "num_tokens": 298041.0,
+      "step": 120
+    },
+    {
+      "epoch": 0.28950542822677927,
+      "eval_entropy": 0.7314498555124476,
+      "eval_loss": 0.6796280145645142,
+      "eval_mean_token_accuracy": 0.809063873264227,
+      "eval_num_tokens": 298041.0,
+      "eval_runtime": 100.6132,
+      "eval_samples_per_second": 14.113,
+      "eval_steps_per_second": 1.769,
+      "step": 120
+    },
+    {
+      "entropy": 0.7079856500029564,
+      "epoch": 0.33775633293124246,
+      "grad_norm": 0.5903022885322571,
+      "learning_rate": 0.00014481502610619613,
+      "loss": 0.6623000144958496,
+      "mean_token_accuracy": 0.8101129323244095,
+      "num_tokens": 348955.0,
+      "step": 140
+    },
+    {
+      "epoch": 0.33775633293124246,
+      "eval_entropy": 0.7005330203624254,
+      "eval_loss": 0.671779453754425,
+      "eval_mean_token_accuracy": 0.8106091374091888,
+      "eval_num_tokens": 348955.0,
+      "eval_runtime": 100.6904,
+      "eval_samples_per_second": 14.103,
+      "eval_steps_per_second": 1.768,
+      "step": 140
+    },
+    {
+      "entropy": 0.7109143257141113,
+      "epoch": 0.38600723763570566,
+      "grad_norm": 0.4516378343105316,
+      "learning_rate": 0.00016565172050996536,
+      "loss": 0.658915901184082,
+      "mean_token_accuracy": 0.8112782314419746,
+      "num_tokens": 395798.0,
+      "step": 160
+    },
+    {
+      "epoch": 0.38600723763570566,
+      "eval_entropy": 0.6891461650976974,
+      "eval_loss": 0.6598312258720398,
+      "eval_mean_token_accuracy": 0.8133360951134329,
+      "eval_num_tokens": 395798.0,
+      "eval_runtime": 100.5057,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 160
+    },
+    {
+      "entropy": 0.6896152347326279,
+      "epoch": 0.43425814234016885,
+      "grad_norm": 0.49479565024375916,
+      "learning_rate": 0.0001864884149137346,
+      "loss": 0.6466911315917969,
+      "mean_token_accuracy": 0.8141682580113411,
+      "num_tokens": 446437.0,
+      "step": 180
+    },
+    {
+      "epoch": 0.43425814234016885,
+      "eval_entropy": 0.6984815888860253,
+      "eval_loss": 0.652909517288208,
+      "eval_mean_token_accuracy": 0.8133517600847094,
+      "eval_num_tokens": 446437.0,
+      "eval_runtime": 100.5867,
+      "eval_samples_per_second": 14.117,
+      "eval_steps_per_second": 1.77,
+      "step": 180
+    },
+    {
+      "entropy": 0.6829591259360314,
+      "epoch": 0.4825090470446321,
+      "grad_norm": 0.4835337698459625,
+      "learning_rate": 0.00020732510931750382,
+      "loss": 0.6401876449584961,
+      "mean_token_accuracy": 0.8175502926111221,
+      "num_tokens": 497682.0,
+      "step": 200
+    },
+    {
+      "epoch": 0.4825090470446321,
+      "eval_entropy": 0.6963563973314306,
+      "eval_loss": 0.650087296962738,
+      "eval_mean_token_accuracy": 0.8159684401549651,
+      "eval_num_tokens": 497682.0,
+      "eval_runtime": 100.6277,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 200
+    },
+    {
+      "entropy": 0.6903425320982933,
+      "epoch": 0.5307599517490953,
+      "grad_norm": 0.44980454444885254,
+      "learning_rate": 0.00022816180372127305,
+      "loss": 0.6409444808959961,
+      "mean_token_accuracy": 0.8163257524371147,
+      "num_tokens": 544670.0,
+      "step": 220
+    },
+    {
+      "epoch": 0.5307599517490953,
+      "eval_entropy": 0.6687790182199371,
+      "eval_loss": 0.6451645493507385,
+      "eval_mean_token_accuracy": 0.8157875551266617,
+      "eval_num_tokens": 544670.0,
+      "eval_runtime": 100.5444,
+      "eval_samples_per_second": 14.123,
+      "eval_steps_per_second": 1.77,
+      "step": 220
+    },
+    {
+      "entropy": 0.6740772560238838,
+      "epoch": 0.5790108564535585,
+      "grad_norm": 0.5312877297401428,
+      "learning_rate": 0.0002489984981250423,
+      "loss": 0.6307676315307618,
+      "mean_token_accuracy": 0.8174169659614563,
+      "num_tokens": 592926.0,
+      "step": 240
+    },
+    {
+      "epoch": 0.5790108564535585,
+      "eval_entropy": 0.7063750872451268,
+      "eval_loss": 0.6494986414909363,
+      "eval_mean_token_accuracy": 0.8156777198395032,
+      "eval_num_tokens": 592926.0,
+      "eval_runtime": 100.5094,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 240
+    },
+    {
+      "entropy": 0.6876470118761062,
+      "epoch": 0.6272617611580217,
+      "grad_norm": 0.46149370074272156,
+      "learning_rate": 0.00026983519252881153,
+      "loss": 0.6392203807830811,
+      "mean_token_accuracy": 0.8172321930527687,
+      "num_tokens": 639077.0,
+      "step": 260
+    },
+    {
+      "epoch": 0.6272617611580217,
+      "eval_entropy": 0.625594850838854,
+      "eval_loss": 0.6463819742202759,
+      "eval_mean_token_accuracy": 0.8179670525400826,
+      "eval_num_tokens": 639077.0,
+      "eval_runtime": 100.5775,
+      "eval_samples_per_second": 14.118,
+      "eval_steps_per_second": 1.77,
+      "step": 260
+    },
+    {
+      "entropy": 0.6673879995942116,
+      "epoch": 0.6755126658624849,
+      "grad_norm": 0.42477288842201233,
+      "learning_rate": 0.0002906718869325807,
+      "loss": 0.6300291538238525,
+      "mean_token_accuracy": 0.8203139930963517,
+      "num_tokens": 691028.0,
+      "step": 280
+    },
+    {
+      "epoch": 0.6755126658624849,
+      "eval_entropy": 0.6863534494062488,
+      "eval_loss": 0.6392037272453308,
+      "eval_mean_token_accuracy": 0.8190520385677895,
+      "eval_num_tokens": 691028.0,
+      "eval_runtime": 100.5106,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 280
+    },
+    {
+      "entropy": 0.6573980256915093,
+      "epoch": 0.7237635705669482,
+      "grad_norm": 0.4656430780887604,
+      "learning_rate": 0.00031150858133635,
+      "loss": 0.6215964794158936,
+      "mean_token_accuracy": 0.8208559453487396,
+      "num_tokens": 740255.0,
+      "step": 300
+    },
+    {
+      "epoch": 0.7237635705669482,
+      "eval_entropy": 0.6834242812703165,
+      "eval_loss": 0.6479960083961487,
+      "eval_mean_token_accuracy": 0.8128752906001016,
+      "eval_num_tokens": 740255.0,
+      "eval_runtime": 100.6101,
+      "eval_samples_per_second": 14.114,
+      "eval_steps_per_second": 1.769,
+      "step": 300
+    },
+    {
+      "entropy": 0.6751310348510742,
+      "epoch": 0.7720144752714113,
+      "grad_norm": 0.4495522081851959,
+      "learning_rate": 0.00033234527574011916,
+      "loss": 0.6402668952941895,
+      "mean_token_accuracy": 0.8170571058988572,
+      "num_tokens": 789529.0,
+      "step": 320
+    },
+    {
+      "epoch": 0.7720144752714113,
+      "eval_entropy": 0.6920045178927733,
+      "eval_loss": 0.6402787566184998,
+      "eval_mean_token_accuracy": 0.817791668886549,
+      "eval_num_tokens": 789529.0,
+      "eval_runtime": 100.6347,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 320
+    },
+    {
+      "entropy": 0.6846290081739426,
+      "epoch": 0.8202653799758746,
+      "grad_norm": 0.5661432147026062,
+      "learning_rate": 0.00035318197014388845,
+      "loss": 0.6508871078491211,
+      "mean_token_accuracy": 0.8164169207215309,
+      "num_tokens": 841584.0,
+      "step": 340
+    },
+    {
+      "epoch": 0.8202653799758746,
+      "eval_entropy": 0.6772527078564248,
+      "eval_loss": 0.6512064337730408,
+      "eval_mean_token_accuracy": 0.8170605137776793,
+      "eval_num_tokens": 841584.0,
+      "eval_runtime": 100.5157,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 340
+    },
+    {
+      "entropy": 0.7022885024547577,
+      "epoch": 0.8685162846803377,
+      "grad_norm": 0.6357236504554749,
+      "learning_rate": 0.0003740186645476577,
+      "loss": 0.6649135112762451,
+      "mean_token_accuracy": 0.8136772960424423,
+      "num_tokens": 885769.0,
+      "step": 360
+    },
+    {
+      "epoch": 0.8685162846803377,
+      "eval_entropy": 0.7435127543599418,
+      "eval_loss": 0.6573262214660645,
+      "eval_mean_token_accuracy": 0.8147802225659403,
+      "eval_num_tokens": 885769.0,
+      "eval_runtime": 100.5271,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 360
+    },
+    {
+      "entropy": 0.6962979376316071,
+      "epoch": 0.916767189384801,
+      "grad_norm": 0.6179454326629639,
+      "learning_rate": 0.00039485535895142685,
+      "loss": 0.6509191513061523,
+      "mean_token_accuracy": 0.81364226937294,
+      "num_tokens": 930056.0,
+      "step": 380
+    },
+    {
+      "epoch": 0.916767189384801,
+      "eval_entropy": 0.6630894883295123,
+      "eval_loss": 0.6524088978767395,
+      "eval_mean_token_accuracy": 0.8165898587596551,
+      "eval_num_tokens": 930056.0,
+      "eval_runtime": 100.5343,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 380
+    },
+    {
+      "entropy": 0.694319449365139,
+      "epoch": 0.9650180940892642,
+      "grad_norm": 0.5458664894104004,
+      "learning_rate": 0.00041569205335519613,
+      "loss": 0.6624011993408203,
+      "mean_token_accuracy": 0.8136080905795098,
+      "num_tokens": 976892.0,
+      "step": 400
+    },
+    {
+      "epoch": 0.9650180940892642,
+      "eval_entropy": 0.7424925324622165,
+      "eval_loss": 0.6717661023139954,
+      "eval_mean_token_accuracy": 0.8099894781460922,
+      "eval_num_tokens": 976892.0,
+      "eval_runtime": 100.5992,
+      "eval_samples_per_second": 14.115,
+      "eval_steps_per_second": 1.769,
+      "step": 400
+    },
+    {
+      "entropy": 0.6861714736009256,
+      "epoch": 1.0120627261761157,
+      "grad_norm": 0.6108037829399109,
+      "learning_rate": 0.0004323601853188671,
+      "loss": 0.6551628112792969,
+      "mean_token_accuracy": 0.8134882618219424,
+      "num_tokens": 1024472.0,
+      "step": 420
+    },
+    {
+      "epoch": 1.0120627261761157,
+      "eval_entropy": 0.6593745705786715,
+      "eval_loss": 0.6720747947692871,
+      "eval_mean_token_accuracy": 0.812802647941568,
+      "eval_num_tokens": 1024472.0,
+      "eval_runtime": 100.4866,
+      "eval_samples_per_second": 14.131,
+      "eval_steps_per_second": 1.771,
+      "step": 420
+    },
+    {
+      "entropy": 0.639520151913166,
+      "epoch": 1.060313630880579,
+      "grad_norm": 0.6054353713989258,
+      "learning_rate": 0.00043231736219609266,
+      "loss": 0.6096667766571044,
+      "mean_token_accuracy": 0.8227144703269005,
+      "num_tokens": 1076964.0,
+      "step": 440
+    },
+    {
+      "epoch": 1.060313630880579,
+      "eval_entropy": 0.6316307876886946,
+      "eval_loss": 0.6621950268745422,
+      "eval_mean_token_accuracy": 0.8140057914712456,
+      "eval_num_tokens": 1076964.0,
+      "eval_runtime": 100.6409,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 440
+    },
+    {
+      "entropy": 0.6544070363044738,
+      "epoch": 1.1085645355850422,
+      "grad_norm": 0.5111057758331299,
+      "learning_rate": 0.0004322133749558233,
+      "loss": 0.6176822185516357,
+      "mean_token_accuracy": 0.8201584875583648,
+      "num_tokens": 1126064.0,
+      "step": 460
+    },
+    {
+      "epoch": 1.1085645355850422,
+      "eval_entropy": 0.6714627766207363,
+      "eval_loss": 0.6738516092300415,
+      "eval_mean_token_accuracy": 0.8128616829936424,
+      "eval_num_tokens": 1126064.0,
+      "eval_runtime": 100.4986,
+      "eval_samples_per_second": 14.13,
+      "eval_steps_per_second": 1.771,
+      "step": 460
+    },
+    {
+      "entropy": 0.6470697179436684,
+      "epoch": 1.1568154402895054,
+      "grad_norm": 0.5434669256210327,
+      "learning_rate": 0.0004320482530252179,
+      "loss": 0.6144762516021729,
+      "mean_token_accuracy": 0.819926056265831,
+      "num_tokens": 1176912.0,
+      "step": 480
+    },
+    {
+      "epoch": 1.1568154402895054,
+      "eval_entropy": 0.6475969080844622,
+      "eval_loss": 0.6624214053153992,
+      "eval_mean_token_accuracy": 0.8137228820431098,
+      "eval_num_tokens": 1176912.0,
+      "eval_runtime": 100.5617,
+      "eval_samples_per_second": 14.121,
+      "eval_steps_per_second": 1.77,
+      "step": 480
+    },
+    {
+      "entropy": 0.6659130036830903,
+      "epoch": 1.2050663449939687,
+      "grad_norm": 0.6470707058906555,
+      "learning_rate": 0.0004318220431318296,
+      "loss": 0.6366308212280274,
+      "mean_token_accuracy": 0.8175261527299881,
+      "num_tokens": 1223383.0,
+      "step": 500
+    },
+    {
+      "epoch": 1.2050663449939687,
+      "eval_entropy": 0.7232425132494295,
+      "eval_loss": 0.6877738833427429,
+      "eval_mean_token_accuracy": 0.8007767260744331,
+      "eval_num_tokens": 1223383.0,
+      "eval_runtime": 100.477,
+      "eval_samples_per_second": 14.133,
+      "eval_steps_per_second": 1.772,
+      "step": 500
+    },
+    {
+      "entropy": 0.6759189993143082,
+      "epoch": 1.2533172496984317,
+      "grad_norm": 0.6453561186790466,
+      "learning_rate": 0.00043153480929038245,
+      "loss": 0.637772512435913,
+      "mean_token_accuracy": 0.8162627041339874,
+      "num_tokens": 1273534.0,
+      "step": 520
+    },
+    {
+      "epoch": 1.2533172496984317,
+      "eval_entropy": 0.7082584312792575,
+      "eval_loss": 0.6695764064788818,
+      "eval_mean_token_accuracy": 0.812019992410467,
+      "eval_num_tokens": 1273534.0,
+      "eval_runtime": 100.4957,
+      "eval_samples_per_second": 14.13,
+      "eval_steps_per_second": 1.771,
+      "step": 520
+    },
+    {
+      "entropy": 0.6730152070522308,
+      "epoch": 1.301568154402895,
+      "grad_norm": 0.6234476566314697,
+      "learning_rate": 0.000431186632784656,
+      "loss": 0.640014362335205,
+      "mean_token_accuracy": 0.8168858170509339,
+      "num_tokens": 1323351.0,
+      "step": 540
+    },
+    {
+      "epoch": 1.301568154402895,
+      "eval_entropy": 0.6655137846978862,
+      "eval_loss": 0.6578248739242554,
+      "eval_mean_token_accuracy": 0.8147600979617472,
+      "eval_num_tokens": 1323351.0,
+      "eval_runtime": 100.503,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 540
+    },
+    {
+      "entropy": 0.6636043325066566,
+      "epoch": 1.3498190591073582,
+      "grad_norm": 0.5882512331008911,
+      "learning_rate": 0.0004307776121444827,
+      "loss": 0.6286775588989257,
+      "mean_token_accuracy": 0.8183174058794975,
+      "num_tokens": 1371508.0,
+      "step": 560
+    },
+    {
+      "epoch": 1.3498190591073582,
+      "eval_entropy": 0.6304813810948575,
+      "eval_loss": 0.6607436537742615,
+      "eval_mean_token_accuracy": 0.8169659268989992,
+      "eval_num_tokens": 1371508.0,
+      "eval_runtime": 100.6233,
+      "eval_samples_per_second": 14.112,
+      "eval_steps_per_second": 1.769,
+      "step": 560
+    },
+    {
+      "entropy": 0.6608738481998444,
+      "epoch": 1.3980699638118215,
+      "grad_norm": 0.6796398758888245,
+      "learning_rate": 0.00043030786311786564,
+      "loss": 0.622726583480835,
+      "mean_token_accuracy": 0.8230827242136002,
+      "num_tokens": 1415221.0,
+      "step": 580
+    },
+    {
+      "epoch": 1.3980699638118215,
+      "eval_entropy": 0.6645242689030894,
+      "eval_loss": 0.6662811636924744,
+      "eval_mean_token_accuracy": 0.8147623907314258,
+      "eval_num_tokens": 1415221.0,
+      "eval_runtime": 100.5074,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 580
+    },
+    {
+      "entropy": 0.6682895936071873,
+      "epoch": 1.4463208685162847,
+      "grad_norm": 0.5750718712806702,
+      "learning_rate": 0.0004297775186382226,
+      "loss": 0.6304276466369629,
+      "mean_token_accuracy": 0.8188625335693359,
+      "num_tokens": 1463356.0,
+      "step": 600
+    },
+    {
+      "epoch": 1.4463208685162847,
+      "eval_entropy": 0.6320532578765676,
+      "eval_loss": 0.650333046913147,
+      "eval_mean_token_accuracy": 0.8172853793321031,
+      "eval_num_tokens": 1463356.0,
+      "eval_runtime": 100.5116,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 600
+    },
+    {
+      "entropy": 0.653739221394062,
+      "epoch": 1.494571773220748,
+      "grad_norm": 0.5552534461021423,
+      "learning_rate": 0.00042918672878676804,
+      "loss": 0.6215689659118653,
+      "mean_token_accuracy": 0.8208896175026894,
+      "num_tokens": 1512340.0,
+      "step": 620
+    },
+    {
+      "epoch": 1.494571773220748,
+      "eval_entropy": 0.6285932062382109,
+      "eval_loss": 0.6392536163330078,
+      "eval_mean_token_accuracy": 0.8205167825302381,
+      "eval_num_tokens": 1512340.0,
+      "eval_runtime": 100.6392,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 620
+    },
+    {
+      "entropy": 0.6476897612214089,
+      "epoch": 1.5428226779252112,
+      "grad_norm": 0.536978542804718,
+      "learning_rate": 0.00042853566075004156,
+      "loss": 0.6131608009338378,
+      "mean_token_accuracy": 0.8208152890205384,
+      "num_tokens": 1560911.0,
+      "step": 640
+    },
+    {
+      "epoch": 1.5428226779252112,
+      "eval_entropy": 0.6626498856571283,
+      "eval_loss": 0.6479025483131409,
+      "eval_mean_token_accuracy": 0.8180894101603647,
+      "eval_num_tokens": 1560911.0,
+      "eval_runtime": 100.5035,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 640
+    },
+    {
+      "entropy": 0.6338633447885513,
+      "epoch": 1.5910735826296745,
+      "grad_norm": 0.4518965780735016,
+      "learning_rate": 0.0004278244987725961,
+      "loss": 0.6010610580444335,
+      "mean_token_accuracy": 0.8247816175222397,
+      "num_tokens": 1609409.0,
+      "step": 660
+    },
+    {
+      "epoch": 1.5910735826296745,
+      "eval_entropy": 0.64079723699709,
+      "eval_loss": 0.6417157053947449,
+      "eval_mean_token_accuracy": 0.8214994581897607,
+      "eval_num_tokens": 1609409.0,
+      "eval_runtime": 100.6393,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 660
+    },
+    {
+      "entropy": 0.6485912457108498,
+      "epoch": 1.6393244873341375,
+      "grad_norm": 0.6028368473052979,
+      "learning_rate": 0.00042705344410485905,
+      "loss": 0.6120437145233154,
+      "mean_token_accuracy": 0.8236212745308876,
+      "num_tokens": 1657430.0,
+      "step": 680
+    },
+    {
+      "epoch": 1.6393244873341375,
+      "eval_entropy": 0.6863006427046958,
+      "eval_loss": 0.6420931816101074,
+      "eval_mean_token_accuracy": 0.8191533580924688,
+      "eval_num_tokens": 1657430.0,
+      "eval_runtime": 100.4639,
+      "eval_samples_per_second": 14.134,
+      "eval_steps_per_second": 1.772,
+      "step": 680
+    },
+    {
+      "entropy": 0.6459382772445679,
+      "epoch": 1.6875753920386007,
+      "grad_norm": 0.5069310069084167,
+      "learning_rate": 0.00042622271494618073,
+      "loss": 0.6160889625549316,
+      "mean_token_accuracy": 0.8253828629851341,
+      "num_tokens": 1708058.0,
+      "step": 700
+    },
+    {
+      "epoch": 1.6875753920386007,
+      "eval_entropy": 0.6773919463157654,
+      "eval_loss": 0.6307896375656128,
+      "eval_mean_token_accuracy": 0.8217989825800563,
+      "eval_num_tokens": 1708058.0,
+      "eval_runtime": 100.6301,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 700
+    },
+    {
+      "entropy": 0.64687028080225,
+      "epoch": 1.7358262967430638,
+      "grad_norm": 0.3830573856830597,
+      "learning_rate": 0.00042533254638308645,
+      "loss": 0.6104001998901367,
+      "mean_token_accuracy": 0.8255606815218925,
+      "num_tokens": 1761262.0,
+      "step": 720
+    },
+    {
+      "epoch": 1.7358262967430638,
+      "eval_entropy": 0.6179541145147902,
+      "eval_loss": 0.6170215606689453,
+      "eval_mean_token_accuracy": 0.8259220900160543,
+      "eval_num_tokens": 1761262.0,
+      "eval_runtime": 100.61,
+      "eval_samples_per_second": 14.114,
+      "eval_steps_per_second": 1.769,
+      "step": 720
+    },
+    {
+      "entropy": 0.6306161753833294,
+      "epoch": 1.784077201447527,
+      "grad_norm": 0.5888957977294922,
+      "learning_rate": 0.0004243831903227498,
+      "loss": 0.5981546878814697,
+      "mean_token_accuracy": 0.8266776099801063,
+      "num_tokens": 1812127.0,
+      "step": 740
+    },
+    {
+      "epoch": 1.784077201447527,
+      "eval_entropy": 0.6357416808605194,
+      "eval_loss": 0.6202242374420166,
+      "eval_mean_token_accuracy": 0.8253380304641937,
+      "eval_num_tokens": 1812127.0,
+      "eval_runtime": 100.5145,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 740
+    },
+    {
+      "entropy": 0.6290688961744308,
+      "epoch": 1.8323281061519903,
+      "grad_norm": 0.6138001680374146,
+      "learning_rate": 0.0004233749154217063,
+      "loss": 0.5853167057037354,
+      "mean_token_accuracy": 0.8281400203704834,
+      "num_tokens": 1858403.0,
+      "step": 760
+    },
+    {
+      "epoch": 1.8323281061519903,
+      "eval_entropy": 0.6402715869833914,
+      "eval_loss": 0.6168262958526611,
+      "eval_mean_token_accuracy": 0.8246346306934785,
+      "eval_num_tokens": 1858403.0,
+      "eval_runtime": 100.5237,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 760
+    },
+    {
+      "entropy": 0.6218615680932998,
+      "epoch": 1.8805790108564535,
+      "grad_norm": 0.574312686920166,
+      "learning_rate": 0.00042230800700982586,
+      "loss": 0.5841915607452393,
+      "mean_token_accuracy": 0.8304707780480385,
+      "num_tokens": 1907361.0,
+      "step": 780
+    },
+    {
+      "epoch": 1.8805790108564535,
+      "eval_entropy": 0.6301163148679091,
+      "eval_loss": 0.6073815226554871,
+      "eval_mean_token_accuracy": 0.8274695993809218,
+      "eval_num_tokens": 1907361.0,
+      "eval_runtime": 100.5111,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 780
+    },
+    {
+      "entropy": 0.6101033873856068,
+      "epoch": 1.9288299155609168,
+      "grad_norm": 0.5054804682731628,
+      "learning_rate": 0.000421182767009569,
+      "loss": 0.5774452209472656,
+      "mean_token_accuracy": 0.8321943372488022,
+      "num_tokens": 1956605.0,
+      "step": 800
+    },
+    {
+      "epoch": 1.9288299155609168,
+      "eval_entropy": 0.631400960215022,
+      "eval_loss": 0.6115217804908752,
+      "eval_mean_token_accuracy": 0.8249618762664581,
+      "eval_num_tokens": 1956605.0,
+      "eval_runtime": 100.4916,
+      "eval_samples_per_second": 14.131,
+      "eval_steps_per_second": 1.771,
+      "step": 800
+    },
+    {
+      "entropy": 0.6217561915516854,
+      "epoch": 1.97708082026538,
+      "grad_norm": 0.5333649516105652,
+      "learning_rate": 0.0004199995138505454,
+      "loss": 0.5842787265777588,
+      "mean_token_accuracy": 0.8305525362491608,
+      "num_tokens": 2003308.0,
+      "step": 820
+    },
+    {
+      "epoch": 1.97708082026538,
+      "eval_entropy": 0.605928597657868,
+      "eval_loss": 0.6067516207695007,
+      "eval_mean_token_accuracy": 0.82715918408351,
+      "eval_num_tokens": 2003308.0,
+      "eval_runtime": 100.5286,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 820
+    },
+    {
+      "entropy": 0.5585076526189462,
+      "epoch": 2.0241254523522314,
+      "grad_norm": 0.44819608330726624,
+      "learning_rate": 0.00041875858237940244,
+      "loss": 0.5252367496490479,
+      "mean_token_accuracy": 0.8405626966403081,
+      "num_tokens": 2051331.0,
+      "step": 840
+    },
+    {
+      "epoch": 2.0241254523522314,
+      "eval_entropy": 0.6123150769579276,
+      "eval_loss": 0.6175946593284607,
+      "eval_mean_token_accuracy": 0.827637837509091,
+      "eval_num_tokens": 2051331.0,
+      "eval_runtime": 100.5414,
+      "eval_samples_per_second": 14.124,
+      "eval_steps_per_second": 1.77,
+      "step": 840
+    },
+    {
+      "entropy": 0.5359115816652775,
+      "epoch": 2.0723763570566947,
+      "grad_norm": 0.576781690120697,
+      "learning_rate": 0.0004174603237650678,
+      "loss": 0.5023207664489746,
+      "mean_token_accuracy": 0.8459693863987923,
+      "num_tokens": 2101528.0,
+      "step": 860
+    },
+    {
+      "epoch": 2.0723763570566947,
+      "eval_entropy": 0.5592134784111816,
+      "eval_loss": 0.620693564414978,
+      "eval_mean_token_accuracy": 0.827402507656076,
+      "eval_num_tokens": 2101528.0,
+      "eval_runtime": 100.6274,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 860
+    },
+    {
+      "entropy": 0.5214565828442573,
+      "epoch": 2.120627261761158,
+      "grad_norm": 0.45878875255584717,
+      "learning_rate": 0.00041610510539937234,
+      "loss": 0.48367843627929685,
+      "mean_token_accuracy": 0.8515541926026344,
+      "num_tokens": 2149550.0,
+      "step": 880
+    },
+    {
+      "epoch": 2.120627261761158,
+      "eval_entropy": 0.5860532344727034,
+      "eval_loss": 0.6134791970252991,
+      "eval_mean_token_accuracy": 0.8291774777214179,
+      "eval_num_tokens": 2149550.0,
+      "eval_runtime": 100.5295,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 880
+    },
+    {
+      "entropy": 0.5414379432797432,
+      "epoch": 2.168878166465621,
+      "grad_norm": 0.5976294875144958,
+      "learning_rate": 0.0004146933107930829,
+      "loss": 0.5012908935546875,
+      "mean_token_accuracy": 0.8466357633471489,
+      "num_tokens": 2202162.0,
+      "step": 900
+    },
+    {
+      "epoch": 2.168878166465621,
+      "eval_entropy": 0.5427783280611038,
+      "eval_loss": 0.612716019153595,
+      "eval_mean_token_accuracy": 0.829505049445656,
+      "eval_num_tokens": 2202162.0,
+      "eval_runtime": 100.5207,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 900
+    },
+    {
+      "entropy": 0.521108990907669,
+      "epoch": 2.2171290711700844,
+      "grad_norm": 0.5145775675773621,
+      "learning_rate": 0.000413225339467373,
+      "loss": 0.4877732276916504,
+      "mean_token_accuracy": 0.8504189744591713,
+      "num_tokens": 2256539.0,
+      "step": 920
+    },
+    {
+      "epoch": 2.2171290711700844,
+      "eval_entropy": 0.5620684869838565,
+      "eval_loss": 0.6176337599754333,
+      "eval_mean_token_accuracy": 0.8291444483767735,
+      "eval_num_tokens": 2256539.0,
+      "eval_runtime": 100.6819,
+      "eval_samples_per_second": 14.104,
+      "eval_steps_per_second": 1.768,
+      "step": 920
+    },
+    {
+      "entropy": 0.5452766507863999,
+      "epoch": 2.2653799758745476,
+      "grad_norm": 0.4820452928543091,
+      "learning_rate": 0.0004117016068407636,
+      "loss": 0.4991307735443115,
+      "mean_token_accuracy": 0.8474363312125206,
+      "num_tokens": 2299698.0,
+      "step": 940
+    },
+    {
+      "epoch": 2.2653799758745476,
+      "eval_entropy": 0.5611683193887218,
+      "eval_loss": 0.6153580546379089,
+      "eval_mean_token_accuracy": 0.8286041272490212,
+      "eval_num_tokens": 2299698.0,
+      "eval_runtime": 100.5249,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 940
+    },
+    {
+      "entropy": 0.5572576269507408,
+      "epoch": 2.313630880579011,
+      "grad_norm": 0.3947993814945221,
+      "learning_rate": 0.00041012254411156387,
+      "loss": 0.5191003799438476,
+      "mean_token_accuracy": 0.8422838419675827,
+      "num_tokens": 2347134.0,
+      "step": 960
+    },
+    {
+      "epoch": 2.313630880579011,
+      "eval_entropy": 0.5651392345682958,
+      "eval_loss": 0.6121005415916443,
+      "eval_mean_token_accuracy": 0.8286536242185014,
+      "eval_num_tokens": 2347134.0,
+      "eval_runtime": 100.7033,
+      "eval_samples_per_second": 14.101,
+      "eval_steps_per_second": 1.768,
+      "step": 960
+    },
+    {
+      "entropy": 0.5441671140491963,
+      "epoch": 2.361881785283474,
+      "grad_norm": 0.7719791531562805,
+      "learning_rate": 0.0004084885981358477,
+      "loss": 0.5072606086730957,
+      "mean_token_accuracy": 0.8478495761752128,
+      "num_tokens": 2394568.0,
+      "step": 980
+    },
+    {
+      "epoch": 2.361881785283474,
+      "eval_entropy": 0.5651004270891126,
+      "eval_loss": 0.6069501042366028,
+      "eval_mean_token_accuracy": 0.8311042156112328,
+      "eval_num_tokens": 2394568.0,
+      "eval_runtime": 100.5487,
+      "eval_samples_per_second": 14.123,
+      "eval_steps_per_second": 1.77,
+      "step": 980
+    },
+    {
+      "entropy": 0.5462074346840382,
+      "epoch": 2.4101326899879374,
+      "grad_norm": 0.5694436430931091,
+      "learning_rate": 0.00040680023130099856,
+      "loss": 0.5097150325775146,
+      "mean_token_accuracy": 0.8474388793110847,
+      "num_tokens": 2443253.0,
+      "step": 1000
+    },
+    {
+      "epoch": 2.4101326899879374,
+      "eval_entropy": 0.5895893243926295,
+      "eval_loss": 0.6078168749809265,
+      "eval_mean_token_accuracy": 0.828030709470256,
+      "eval_num_tokens": 2443253.0,
+      "eval_runtime": 100.5413,
+      "eval_samples_per_second": 14.124,
+      "eval_steps_per_second": 1.77,
+      "step": 1000
+    },
+    {
+      "entropy": 0.5473216332495212,
+      "epoch": 2.4583835946924006,
+      "grad_norm": 0.3989660143852234,
+      "learning_rate": 0.00040505792139485913,
+      "loss": 0.5087806224822998,
+      "mean_token_accuracy": 0.8473173275589942,
+      "num_tokens": 2495178.0,
+      "step": 1020
+    },
+    {
+      "epoch": 2.4583835946924006,
+      "eval_entropy": 0.5588909702354603,
+      "eval_loss": 0.6087605953216553,
+      "eval_mean_token_accuracy": 0.8309365524334854,
+      "eval_num_tokens": 2495178.0,
+      "eval_runtime": 100.5137,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 1020
+    },
+    {
+      "entropy": 0.5430163390934467,
+      "epoch": 2.5066344993968634,
+      "grad_norm": 0.4916608929634094,
+      "learning_rate": 0.00040326216147052336,
+      "loss": 0.5027707576751709,
+      "mean_token_accuracy": 0.8466959804296493,
+      "num_tokens": 2541733.0,
+      "step": 1040
+    },
+    {
+      "epoch": 2.5066344993968634,
+      "eval_entropy": 0.5819033988741007,
+      "eval_loss": 0.609398365020752,
+      "eval_mean_token_accuracy": 0.8297666612635838,
+      "eval_num_tokens": 2541733.0,
+      "eval_runtime": 100.6375,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 1040
+    },
+    {
+      "entropy": 0.5632683664560318,
+      "epoch": 2.554885404101327,
+      "grad_norm": 0.6062753796577454,
+      "learning_rate": 0.00040141345970680814,
+      "loss": 0.5272740840911865,
+      "mean_token_accuracy": 0.8411559104919434,
+      "num_tokens": 2589479.0,
+      "step": 1060
+    },
+    {
+      "epoch": 2.554885404101327,
+      "eval_entropy": 0.5785442567273472,
+      "eval_loss": 0.6207635998725891,
+      "eval_mean_token_accuracy": 0.8259770478425401,
+      "eval_num_tokens": 2589479.0,
+      "eval_runtime": 100.5161,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 1060
+    }
+  ],
+  "logging_steps": 20,
+  "max_steps": 4150,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 20,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.370774678528e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1080/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen3-14B-Base
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen3-14B-Base
+- lora
+- sft
+- transformers
+- trl
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1080/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-14B-Base",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 128,
+  "lora_bias": false,
+  "lora_dropout": 0.0539675908752296,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "gate_proj",
+    "q_proj",
+    "down_proj",
+    "o_proj",
+    "k_proj",
+    "v_proj",
+    "up_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1080/chat_template.jinja

@@ -0,0 +1,85 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1080/tokenizer_config.json

@@ -0,0 +1,29 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1080/trainer_state.json

@@ -0,0 +1,1168 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.60313630880579,
+  "eval_steps": 20,
+  "global_step": 1080,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 1.8160341799259185,
+      "epoch": 0.04825090470446321,
+      "grad_norm": 1.1313843727111816,
+      "learning_rate": 1.9794859683580766e-05,
+      "loss": 1.7627403259277343,
+      "mean_token_accuracy": 0.6354069545865059,
+      "num_tokens": 50237.0,
+      "step": 20
+    },
+    {
+      "epoch": 0.04825090470446321,
+      "eval_entropy": 1.538598193881217,
+      "eval_loss": 1.4881398677825928,
+      "eval_mean_token_accuracy": 0.686459881201219,
+      "eval_num_tokens": 50237.0,
+      "eval_runtime": 100.8923,
+      "eval_samples_per_second": 14.074,
+      "eval_steps_per_second": 1.764,
+      "step": 20
+    },
+    {
+      "entropy": 1.1673978716135025,
+      "epoch": 0.09650180940892641,
+      "grad_norm": 0.6167025566101074,
+      "learning_rate": 4.0631554087349995e-05,
+      "loss": 1.0887075424194337,
+      "mean_token_accuracy": 0.737513878941536,
+      "num_tokens": 101098.0,
+      "step": 40
+    },
+    {
+      "epoch": 0.09650180940892641,
+      "eval_entropy": 0.887577452351538,
+      "eval_loss": 0.8519237041473389,
+      "eval_mean_token_accuracy": 0.7741017204322173,
+      "eval_num_tokens": 101098.0,
+      "eval_runtime": 100.4819,
+      "eval_samples_per_second": 14.132,
+      "eval_steps_per_second": 1.771,
+      "step": 40
+    },
+    {
+      "entropy": 0.8220535531640053,
+      "epoch": 0.14475271411338964,
+      "grad_norm": 0.8396430611610413,
+      "learning_rate": 6.146824849111923e-05,
+      "loss": 0.7812146186828614,
+      "mean_token_accuracy": 0.7872247830033302,
+      "num_tokens": 150385.0,
+      "step": 60
+    },
+    {
+      "epoch": 0.14475271411338964,
+      "eval_entropy": 0.770665143982748,
+      "eval_loss": 0.7478981614112854,
+      "eval_mean_token_accuracy": 0.7949167068084974,
+      "eval_num_tokens": 150385.0,
+      "eval_runtime": 100.6518,
+      "eval_samples_per_second": 14.108,
+      "eval_steps_per_second": 1.768,
+      "step": 60
+    },
+    {
+      "entropy": 0.7776695728302002,
+      "epoch": 0.19300361881785283,
+      "grad_norm": 0.5731202363967896,
+      "learning_rate": 8.230494289488845e-05,
+      "loss": 0.732721471786499,
+      "mean_token_accuracy": 0.7976997002959252,
+      "num_tokens": 197354.0,
+      "step": 80
+    },
+    {
+      "epoch": 0.19300361881785283,
+      "eval_entropy": 0.7650937618834249,
+      "eval_loss": 0.7109046578407288,
+      "eval_mean_token_accuracy": 0.8018777065732506,
+      "eval_num_tokens": 197354.0,
+      "eval_runtime": 100.5983,
+      "eval_samples_per_second": 14.116,
+      "eval_steps_per_second": 1.769,
+      "step": 80
+    },
+    {
+      "entropy": 0.7309209659695626,
+      "epoch": 0.24125452352231605,
+      "grad_norm": 0.48046866059303284,
+      "learning_rate": 0.00010314163729865769,
+      "loss": 0.680583143234253,
+      "mean_token_accuracy": 0.8073364540934562,
+      "num_tokens": 248557.0,
+      "step": 100
+    },
+    {
+      "epoch": 0.24125452352231605,
+      "eval_entropy": 0.7152130687504672,
+      "eval_loss": 0.6989707350730896,
+      "eval_mean_token_accuracy": 0.8057264320636064,
+      "eval_num_tokens": 248557.0,
+      "eval_runtime": 100.6264,
+      "eval_samples_per_second": 14.112,
+      "eval_steps_per_second": 1.769,
+      "step": 100
+    },
+    {
+      "entropy": 0.7105271726846695,
+      "epoch": 0.28950542822677927,
+      "grad_norm": 0.5512468814849854,
+      "learning_rate": 0.00012397833170242693,
+      "loss": 0.6674309253692627,
+      "mean_token_accuracy": 0.8098619893193245,
+      "num_tokens": 298041.0,
+      "step": 120
+    },
+    {
+      "epoch": 0.28950542822677927,
+      "eval_entropy": 0.7314498555124476,
+      "eval_loss": 0.6796280145645142,
+      "eval_mean_token_accuracy": 0.809063873264227,
+      "eval_num_tokens": 298041.0,
+      "eval_runtime": 100.6132,
+      "eval_samples_per_second": 14.113,
+      "eval_steps_per_second": 1.769,
+      "step": 120
+    },
+    {
+      "entropy": 0.7079856500029564,
+      "epoch": 0.33775633293124246,
+      "grad_norm": 0.5903022885322571,
+      "learning_rate": 0.00014481502610619613,
+      "loss": 0.6623000144958496,
+      "mean_token_accuracy": 0.8101129323244095,
+      "num_tokens": 348955.0,
+      "step": 140
+    },
+    {
+      "epoch": 0.33775633293124246,
+      "eval_entropy": 0.7005330203624254,
+      "eval_loss": 0.671779453754425,
+      "eval_mean_token_accuracy": 0.8106091374091888,
+      "eval_num_tokens": 348955.0,
+      "eval_runtime": 100.6904,
+      "eval_samples_per_second": 14.103,
+      "eval_steps_per_second": 1.768,
+      "step": 140
+    },
+    {
+      "entropy": 0.7109143257141113,
+      "epoch": 0.38600723763570566,
+      "grad_norm": 0.4516378343105316,
+      "learning_rate": 0.00016565172050996536,
+      "loss": 0.658915901184082,
+      "mean_token_accuracy": 0.8112782314419746,
+      "num_tokens": 395798.0,
+      "step": 160
+    },
+    {
+      "epoch": 0.38600723763570566,
+      "eval_entropy": 0.6891461650976974,
+      "eval_loss": 0.6598312258720398,
+      "eval_mean_token_accuracy": 0.8133360951134329,
+      "eval_num_tokens": 395798.0,
+      "eval_runtime": 100.5057,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 160
+    },
+    {
+      "entropy": 0.6896152347326279,
+      "epoch": 0.43425814234016885,
+      "grad_norm": 0.49479565024375916,
+      "learning_rate": 0.0001864884149137346,
+      "loss": 0.6466911315917969,
+      "mean_token_accuracy": 0.8141682580113411,
+      "num_tokens": 446437.0,
+      "step": 180
+    },
+    {
+      "epoch": 0.43425814234016885,
+      "eval_entropy": 0.6984815888860253,
+      "eval_loss": 0.652909517288208,
+      "eval_mean_token_accuracy": 0.8133517600847094,
+      "eval_num_tokens": 446437.0,
+      "eval_runtime": 100.5867,
+      "eval_samples_per_second": 14.117,
+      "eval_steps_per_second": 1.77,
+      "step": 180
+    },
+    {
+      "entropy": 0.6829591259360314,
+      "epoch": 0.4825090470446321,
+      "grad_norm": 0.4835337698459625,
+      "learning_rate": 0.00020732510931750382,
+      "loss": 0.6401876449584961,
+      "mean_token_accuracy": 0.8175502926111221,
+      "num_tokens": 497682.0,
+      "step": 200
+    },
+    {
+      "epoch": 0.4825090470446321,
+      "eval_entropy": 0.6963563973314306,
+      "eval_loss": 0.650087296962738,
+      "eval_mean_token_accuracy": 0.8159684401549651,
+      "eval_num_tokens": 497682.0,
+      "eval_runtime": 100.6277,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 200
+    },
+    {
+      "entropy": 0.6903425320982933,
+      "epoch": 0.5307599517490953,
+      "grad_norm": 0.44980454444885254,
+      "learning_rate": 0.00022816180372127305,
+      "loss": 0.6409444808959961,
+      "mean_token_accuracy": 0.8163257524371147,
+      "num_tokens": 544670.0,
+      "step": 220
+    },
+    {
+      "epoch": 0.5307599517490953,
+      "eval_entropy": 0.6687790182199371,
+      "eval_loss": 0.6451645493507385,
+      "eval_mean_token_accuracy": 0.8157875551266617,
+      "eval_num_tokens": 544670.0,
+      "eval_runtime": 100.5444,
+      "eval_samples_per_second": 14.123,
+      "eval_steps_per_second": 1.77,
+      "step": 220
+    },
+    {
+      "entropy": 0.6740772560238838,
+      "epoch": 0.5790108564535585,
+      "grad_norm": 0.5312877297401428,
+      "learning_rate": 0.0002489984981250423,
+      "loss": 0.6307676315307618,
+      "mean_token_accuracy": 0.8174169659614563,
+      "num_tokens": 592926.0,
+      "step": 240
+    },
+    {
+      "epoch": 0.5790108564535585,
+      "eval_entropy": 0.7063750872451268,
+      "eval_loss": 0.6494986414909363,
+      "eval_mean_token_accuracy": 0.8156777198395032,
+      "eval_num_tokens": 592926.0,
+      "eval_runtime": 100.5094,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 240
+    },
+    {
+      "entropy": 0.6876470118761062,
+      "epoch": 0.6272617611580217,
+      "grad_norm": 0.46149370074272156,
+      "learning_rate": 0.00026983519252881153,
+      "loss": 0.6392203807830811,
+      "mean_token_accuracy": 0.8172321930527687,
+      "num_tokens": 639077.0,
+      "step": 260
+    },
+    {
+      "epoch": 0.6272617611580217,
+      "eval_entropy": 0.625594850838854,
+      "eval_loss": 0.6463819742202759,
+      "eval_mean_token_accuracy": 0.8179670525400826,
+      "eval_num_tokens": 639077.0,
+      "eval_runtime": 100.5775,
+      "eval_samples_per_second": 14.118,
+      "eval_steps_per_second": 1.77,
+      "step": 260
+    },
+    {
+      "entropy": 0.6673879995942116,
+      "epoch": 0.6755126658624849,
+      "grad_norm": 0.42477288842201233,
+      "learning_rate": 0.0002906718869325807,
+      "loss": 0.6300291538238525,
+      "mean_token_accuracy": 0.8203139930963517,
+      "num_tokens": 691028.0,
+      "step": 280
+    },
+    {
+      "epoch": 0.6755126658624849,
+      "eval_entropy": 0.6863534494062488,
+      "eval_loss": 0.6392037272453308,
+      "eval_mean_token_accuracy": 0.8190520385677895,
+      "eval_num_tokens": 691028.0,
+      "eval_runtime": 100.5106,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 280
+    },
+    {
+      "entropy": 0.6573980256915093,
+      "epoch": 0.7237635705669482,
+      "grad_norm": 0.4656430780887604,
+      "learning_rate": 0.00031150858133635,
+      "loss": 0.6215964794158936,
+      "mean_token_accuracy": 0.8208559453487396,
+      "num_tokens": 740255.0,
+      "step": 300
+    },
+    {
+      "epoch": 0.7237635705669482,
+      "eval_entropy": 0.6834242812703165,
+      "eval_loss": 0.6479960083961487,
+      "eval_mean_token_accuracy": 0.8128752906001016,
+      "eval_num_tokens": 740255.0,
+      "eval_runtime": 100.6101,
+      "eval_samples_per_second": 14.114,
+      "eval_steps_per_second": 1.769,
+      "step": 300
+    },
+    {
+      "entropy": 0.6751310348510742,
+      "epoch": 0.7720144752714113,
+      "grad_norm": 0.4495522081851959,
+      "learning_rate": 0.00033234527574011916,
+      "loss": 0.6402668952941895,
+      "mean_token_accuracy": 0.8170571058988572,
+      "num_tokens": 789529.0,
+      "step": 320
+    },
+    {
+      "epoch": 0.7720144752714113,
+      "eval_entropy": 0.6920045178927733,
+      "eval_loss": 0.6402787566184998,
+      "eval_mean_token_accuracy": 0.817791668886549,
+      "eval_num_tokens": 789529.0,
+      "eval_runtime": 100.6347,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 320
+    },
+    {
+      "entropy": 0.6846290081739426,
+      "epoch": 0.8202653799758746,
+      "grad_norm": 0.5661432147026062,
+      "learning_rate": 0.00035318197014388845,
+      "loss": 0.6508871078491211,
+      "mean_token_accuracy": 0.8164169207215309,
+      "num_tokens": 841584.0,
+      "step": 340
+    },
+    {
+      "epoch": 0.8202653799758746,
+      "eval_entropy": 0.6772527078564248,
+      "eval_loss": 0.6512064337730408,
+      "eval_mean_token_accuracy": 0.8170605137776793,
+      "eval_num_tokens": 841584.0,
+      "eval_runtime": 100.5157,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 340
+    },
+    {
+      "entropy": 0.7022885024547577,
+      "epoch": 0.8685162846803377,
+      "grad_norm": 0.6357236504554749,
+      "learning_rate": 0.0003740186645476577,
+      "loss": 0.6649135112762451,
+      "mean_token_accuracy": 0.8136772960424423,
+      "num_tokens": 885769.0,
+      "step": 360
+    },
+    {
+      "epoch": 0.8685162846803377,
+      "eval_entropy": 0.7435127543599418,
+      "eval_loss": 0.6573262214660645,
+      "eval_mean_token_accuracy": 0.8147802225659403,
+      "eval_num_tokens": 885769.0,
+      "eval_runtime": 100.5271,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 360
+    },
+    {
+      "entropy": 0.6962979376316071,
+      "epoch": 0.916767189384801,
+      "grad_norm": 0.6179454326629639,
+      "learning_rate": 0.00039485535895142685,
+      "loss": 0.6509191513061523,
+      "mean_token_accuracy": 0.81364226937294,
+      "num_tokens": 930056.0,
+      "step": 380
+    },
+    {
+      "epoch": 0.916767189384801,
+      "eval_entropy": 0.6630894883295123,
+      "eval_loss": 0.6524088978767395,
+      "eval_mean_token_accuracy": 0.8165898587596551,
+      "eval_num_tokens": 930056.0,
+      "eval_runtime": 100.5343,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 380
+    },
+    {
+      "entropy": 0.694319449365139,
+      "epoch": 0.9650180940892642,
+      "grad_norm": 0.5458664894104004,
+      "learning_rate": 0.00041569205335519613,
+      "loss": 0.6624011993408203,
+      "mean_token_accuracy": 0.8136080905795098,
+      "num_tokens": 976892.0,
+      "step": 400
+    },
+    {
+      "epoch": 0.9650180940892642,
+      "eval_entropy": 0.7424925324622165,
+      "eval_loss": 0.6717661023139954,
+      "eval_mean_token_accuracy": 0.8099894781460922,
+      "eval_num_tokens": 976892.0,
+      "eval_runtime": 100.5992,
+      "eval_samples_per_second": 14.115,
+      "eval_steps_per_second": 1.769,
+      "step": 400
+    },
+    {
+      "entropy": 0.6861714736009256,
+      "epoch": 1.0120627261761157,
+      "grad_norm": 0.6108037829399109,
+      "learning_rate": 0.0004323601853188671,
+      "loss": 0.6551628112792969,
+      "mean_token_accuracy": 0.8134882618219424,
+      "num_tokens": 1024472.0,
+      "step": 420
+    },
+    {
+      "epoch": 1.0120627261761157,
+      "eval_entropy": 0.6593745705786715,
+      "eval_loss": 0.6720747947692871,
+      "eval_mean_token_accuracy": 0.812802647941568,
+      "eval_num_tokens": 1024472.0,
+      "eval_runtime": 100.4866,
+      "eval_samples_per_second": 14.131,
+      "eval_steps_per_second": 1.771,
+      "step": 420
+    },
+    {
+      "entropy": 0.639520151913166,
+      "epoch": 1.060313630880579,
+      "grad_norm": 0.6054353713989258,
+      "learning_rate": 0.00043231736219609266,
+      "loss": 0.6096667766571044,
+      "mean_token_accuracy": 0.8227144703269005,
+      "num_tokens": 1076964.0,
+      "step": 440
+    },
+    {
+      "epoch": 1.060313630880579,
+      "eval_entropy": 0.6316307876886946,
+      "eval_loss": 0.6621950268745422,
+      "eval_mean_token_accuracy": 0.8140057914712456,
+      "eval_num_tokens": 1076964.0,
+      "eval_runtime": 100.6409,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 440
+    },
+    {
+      "entropy": 0.6544070363044738,
+      "epoch": 1.1085645355850422,
+      "grad_norm": 0.5111057758331299,
+      "learning_rate": 0.0004322133749558233,
+      "loss": 0.6176822185516357,
+      "mean_token_accuracy": 0.8201584875583648,
+      "num_tokens": 1126064.0,
+      "step": 460
+    },
+    {
+      "epoch": 1.1085645355850422,
+      "eval_entropy": 0.6714627766207363,
+      "eval_loss": 0.6738516092300415,
+      "eval_mean_token_accuracy": 0.8128616829936424,
+      "eval_num_tokens": 1126064.0,
+      "eval_runtime": 100.4986,
+      "eval_samples_per_second": 14.13,
+      "eval_steps_per_second": 1.771,
+      "step": 460
+    },
+    {
+      "entropy": 0.6470697179436684,
+      "epoch": 1.1568154402895054,
+      "grad_norm": 0.5434669256210327,
+      "learning_rate": 0.0004320482530252179,
+      "loss": 0.6144762516021729,
+      "mean_token_accuracy": 0.819926056265831,
+      "num_tokens": 1176912.0,
+      "step": 480
+    },
+    {
+      "epoch": 1.1568154402895054,
+      "eval_entropy": 0.6475969080844622,
+      "eval_loss": 0.6624214053153992,
+      "eval_mean_token_accuracy": 0.8137228820431098,
+      "eval_num_tokens": 1176912.0,
+      "eval_runtime": 100.5617,
+      "eval_samples_per_second": 14.121,
+      "eval_steps_per_second": 1.77,
+      "step": 480
+    },
+    {
+      "entropy": 0.6659130036830903,
+      "epoch": 1.2050663449939687,
+      "grad_norm": 0.6470707058906555,
+      "learning_rate": 0.0004318220431318296,
+      "loss": 0.6366308212280274,
+      "mean_token_accuracy": 0.8175261527299881,
+      "num_tokens": 1223383.0,
+      "step": 500
+    },
+    {
+      "epoch": 1.2050663449939687,
+      "eval_entropy": 0.7232425132494295,
+      "eval_loss": 0.6877738833427429,
+      "eval_mean_token_accuracy": 0.8007767260744331,
+      "eval_num_tokens": 1223383.0,
+      "eval_runtime": 100.477,
+      "eval_samples_per_second": 14.133,
+      "eval_steps_per_second": 1.772,
+      "step": 500
+    },
+    {
+      "entropy": 0.6759189993143082,
+      "epoch": 1.2533172496984317,
+      "grad_norm": 0.6453561186790466,
+      "learning_rate": 0.00043153480929038245,
+      "loss": 0.637772512435913,
+      "mean_token_accuracy": 0.8162627041339874,
+      "num_tokens": 1273534.0,
+      "step": 520
+    },
+    {
+      "epoch": 1.2533172496984317,
+      "eval_entropy": 0.7082584312792575,
+      "eval_loss": 0.6695764064788818,
+      "eval_mean_token_accuracy": 0.812019992410467,
+      "eval_num_tokens": 1273534.0,
+      "eval_runtime": 100.4957,
+      "eval_samples_per_second": 14.13,
+      "eval_steps_per_second": 1.771,
+      "step": 520
+    },
+    {
+      "entropy": 0.6730152070522308,
+      "epoch": 1.301568154402895,
+      "grad_norm": 0.6234476566314697,
+      "learning_rate": 0.000431186632784656,
+      "loss": 0.640014362335205,
+      "mean_token_accuracy": 0.8168858170509339,
+      "num_tokens": 1323351.0,
+      "step": 540
+    },
+    {
+      "epoch": 1.301568154402895,
+      "eval_entropy": 0.6655137846978862,
+      "eval_loss": 0.6578248739242554,
+      "eval_mean_token_accuracy": 0.8147600979617472,
+      "eval_num_tokens": 1323351.0,
+      "eval_runtime": 100.503,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 540
+    },
+    {
+      "entropy": 0.6636043325066566,
+      "epoch": 1.3498190591073582,
+      "grad_norm": 0.5882512331008911,
+      "learning_rate": 0.0004307776121444827,
+      "loss": 0.6286775588989257,
+      "mean_token_accuracy": 0.8183174058794975,
+      "num_tokens": 1371508.0,
+      "step": 560
+    },
+    {
+      "epoch": 1.3498190591073582,
+      "eval_entropy": 0.6304813810948575,
+      "eval_loss": 0.6607436537742615,
+      "eval_mean_token_accuracy": 0.8169659268989992,
+      "eval_num_tokens": 1371508.0,
+      "eval_runtime": 100.6233,
+      "eval_samples_per_second": 14.112,
+      "eval_steps_per_second": 1.769,
+      "step": 560
+    },
+    {
+      "entropy": 0.6608738481998444,
+      "epoch": 1.3980699638118215,
+      "grad_norm": 0.6796398758888245,
+      "learning_rate": 0.00043030786311786564,
+      "loss": 0.622726583480835,
+      "mean_token_accuracy": 0.8230827242136002,
+      "num_tokens": 1415221.0,
+      "step": 580
+    },
+    {
+      "epoch": 1.3980699638118215,
+      "eval_entropy": 0.6645242689030894,
+      "eval_loss": 0.6662811636924744,
+      "eval_mean_token_accuracy": 0.8147623907314258,
+      "eval_num_tokens": 1415221.0,
+      "eval_runtime": 100.5074,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 580
+    },
+    {
+      "entropy": 0.6682895936071873,
+      "epoch": 1.4463208685162847,
+      "grad_norm": 0.5750718712806702,
+      "learning_rate": 0.0004297775186382226,
+      "loss": 0.6304276466369629,
+      "mean_token_accuracy": 0.8188625335693359,
+      "num_tokens": 1463356.0,
+      "step": 600
+    },
+    {
+      "epoch": 1.4463208685162847,
+      "eval_entropy": 0.6320532578765676,
+      "eval_loss": 0.650333046913147,
+      "eval_mean_token_accuracy": 0.8172853793321031,
+      "eval_num_tokens": 1463356.0,
+      "eval_runtime": 100.5116,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 600
+    },
+    {
+      "entropy": 0.653739221394062,
+      "epoch": 1.494571773220748,
+      "grad_norm": 0.5552534461021423,
+      "learning_rate": 0.00042918672878676804,
+      "loss": 0.6215689659118653,
+      "mean_token_accuracy": 0.8208896175026894,
+      "num_tokens": 1512340.0,
+      "step": 620
+    },
+    {
+      "epoch": 1.494571773220748,
+      "eval_entropy": 0.6285932062382109,
+      "eval_loss": 0.6392536163330078,
+      "eval_mean_token_accuracy": 0.8205167825302381,
+      "eval_num_tokens": 1512340.0,
+      "eval_runtime": 100.6392,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 620
+    },
+    {
+      "entropy": 0.6476897612214089,
+      "epoch": 1.5428226779252112,
+      "grad_norm": 0.536978542804718,
+      "learning_rate": 0.00042853566075004156,
+      "loss": 0.6131608009338378,
+      "mean_token_accuracy": 0.8208152890205384,
+      "num_tokens": 1560911.0,
+      "step": 640
+    },
+    {
+      "epoch": 1.5428226779252112,
+      "eval_entropy": 0.6626498856571283,
+      "eval_loss": 0.6479025483131409,
+      "eval_mean_token_accuracy": 0.8180894101603647,
+      "eval_num_tokens": 1560911.0,
+      "eval_runtime": 100.5035,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 640
+    },
+    {
+      "entropy": 0.6338633447885513,
+      "epoch": 1.5910735826296745,
+      "grad_norm": 0.4518965780735016,
+      "learning_rate": 0.0004278244987725961,
+      "loss": 0.6010610580444335,
+      "mean_token_accuracy": 0.8247816175222397,
+      "num_tokens": 1609409.0,
+      "step": 660
+    },
+    {
+      "epoch": 1.5910735826296745,
+      "eval_entropy": 0.64079723699709,
+      "eval_loss": 0.6417157053947449,
+      "eval_mean_token_accuracy": 0.8214994581897607,
+      "eval_num_tokens": 1609409.0,
+      "eval_runtime": 100.6393,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 660
+    },
+    {
+      "entropy": 0.6485912457108498,
+      "epoch": 1.6393244873341375,
+      "grad_norm": 0.6028368473052979,
+      "learning_rate": 0.00042705344410485905,
+      "loss": 0.6120437145233154,
+      "mean_token_accuracy": 0.8236212745308876,
+      "num_tokens": 1657430.0,
+      "step": 680
+    },
+    {
+      "epoch": 1.6393244873341375,
+      "eval_entropy": 0.6863006427046958,
+      "eval_loss": 0.6420931816101074,
+      "eval_mean_token_accuracy": 0.8191533580924688,
+      "eval_num_tokens": 1657430.0,
+      "eval_runtime": 100.4639,
+      "eval_samples_per_second": 14.134,
+      "eval_steps_per_second": 1.772,
+      "step": 680
+    },
+    {
+      "entropy": 0.6459382772445679,
+      "epoch": 1.6875753920386007,
+      "grad_norm": 0.5069310069084167,
+      "learning_rate": 0.00042622271494618073,
+      "loss": 0.6160889625549316,
+      "mean_token_accuracy": 0.8253828629851341,
+      "num_tokens": 1708058.0,
+      "step": 700
+    },
+    {
+      "epoch": 1.6875753920386007,
+      "eval_entropy": 0.6773919463157654,
+      "eval_loss": 0.6307896375656128,
+      "eval_mean_token_accuracy": 0.8217989825800563,
+      "eval_num_tokens": 1708058.0,
+      "eval_runtime": 100.6301,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 700
+    },
+    {
+      "entropy": 0.64687028080225,
+      "epoch": 1.7358262967430638,
+      "grad_norm": 0.3830573856830597,
+      "learning_rate": 0.00042533254638308645,
+      "loss": 0.6104001998901367,
+      "mean_token_accuracy": 0.8255606815218925,
+      "num_tokens": 1761262.0,
+      "step": 720
+    },
+    {
+      "epoch": 1.7358262967430638,
+      "eval_entropy": 0.6179541145147902,
+      "eval_loss": 0.6170215606689453,
+      "eval_mean_token_accuracy": 0.8259220900160543,
+      "eval_num_tokens": 1761262.0,
+      "eval_runtime": 100.61,
+      "eval_samples_per_second": 14.114,
+      "eval_steps_per_second": 1.769,
+      "step": 720
+    },
+    {
+      "entropy": 0.6306161753833294,
+      "epoch": 1.784077201447527,
+      "grad_norm": 0.5888957977294922,
+      "learning_rate": 0.0004243831903227498,
+      "loss": 0.5981546878814697,
+      "mean_token_accuracy": 0.8266776099801063,
+      "num_tokens": 1812127.0,
+      "step": 740
+    },
+    {
+      "epoch": 1.784077201447527,
+      "eval_entropy": 0.6357416808605194,
+      "eval_loss": 0.6202242374420166,
+      "eval_mean_token_accuracy": 0.8253380304641937,
+      "eval_num_tokens": 1812127.0,
+      "eval_runtime": 100.5145,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 740
+    },
+    {
+      "entropy": 0.6290688961744308,
+      "epoch": 1.8323281061519903,
+      "grad_norm": 0.6138001680374146,
+      "learning_rate": 0.0004233749154217063,
+      "loss": 0.5853167057037354,
+      "mean_token_accuracy": 0.8281400203704834,
+      "num_tokens": 1858403.0,
+      "step": 760
+    },
+    {
+      "epoch": 1.8323281061519903,
+      "eval_entropy": 0.6402715869833914,
+      "eval_loss": 0.6168262958526611,
+      "eval_mean_token_accuracy": 0.8246346306934785,
+      "eval_num_tokens": 1858403.0,
+      "eval_runtime": 100.5237,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 760
+    },
+    {
+      "entropy": 0.6218615680932998,
+      "epoch": 1.8805790108564535,
+      "grad_norm": 0.574312686920166,
+      "learning_rate": 0.00042230800700982586,
+      "loss": 0.5841915607452393,
+      "mean_token_accuracy": 0.8304707780480385,
+      "num_tokens": 1907361.0,
+      "step": 780
+    },
+    {
+      "epoch": 1.8805790108564535,
+      "eval_entropy": 0.6301163148679091,
+      "eval_loss": 0.6073815226554871,
+      "eval_mean_token_accuracy": 0.8274695993809218,
+      "eval_num_tokens": 1907361.0,
+      "eval_runtime": 100.5111,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 780
+    },
+    {
+      "entropy": 0.6101033873856068,
+      "epoch": 1.9288299155609168,
+      "grad_norm": 0.5054804682731628,
+      "learning_rate": 0.000421182767009569,
+      "loss": 0.5774452209472656,
+      "mean_token_accuracy": 0.8321943372488022,
+      "num_tokens": 1956605.0,
+      "step": 800
+    },
+    {
+      "epoch": 1.9288299155609168,
+      "eval_entropy": 0.631400960215022,
+      "eval_loss": 0.6115217804908752,
+      "eval_mean_token_accuracy": 0.8249618762664581,
+      "eval_num_tokens": 1956605.0,
+      "eval_runtime": 100.4916,
+      "eval_samples_per_second": 14.131,
+      "eval_steps_per_second": 1.771,
+      "step": 800
+    },
+    {
+      "entropy": 0.6217561915516854,
+      "epoch": 1.97708082026538,
+      "grad_norm": 0.5333649516105652,
+      "learning_rate": 0.0004199995138505454,
+      "loss": 0.5842787265777588,
+      "mean_token_accuracy": 0.8305525362491608,
+      "num_tokens": 2003308.0,
+      "step": 820
+    },
+    {
+      "epoch": 1.97708082026538,
+      "eval_entropy": 0.605928597657868,
+      "eval_loss": 0.6067516207695007,
+      "eval_mean_token_accuracy": 0.82715918408351,
+      "eval_num_tokens": 2003308.0,
+      "eval_runtime": 100.5286,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 820
+    },
+    {
+      "entropy": 0.5585076526189462,
+      "epoch": 2.0241254523522314,
+      "grad_norm": 0.44819608330726624,
+      "learning_rate": 0.00041875858237940244,
+      "loss": 0.5252367496490479,
+      "mean_token_accuracy": 0.8405626966403081,
+      "num_tokens": 2051331.0,
+      "step": 840
+    },
+    {
+      "epoch": 2.0241254523522314,
+      "eval_entropy": 0.6123150769579276,
+      "eval_loss": 0.6175946593284607,
+      "eval_mean_token_accuracy": 0.827637837509091,
+      "eval_num_tokens": 2051331.0,
+      "eval_runtime": 100.5414,
+      "eval_samples_per_second": 14.124,
+      "eval_steps_per_second": 1.77,
+      "step": 840
+    },
+    {
+      "entropy": 0.5359115816652775,
+      "epoch": 2.0723763570566947,
+      "grad_norm": 0.576781690120697,
+      "learning_rate": 0.0004174603237650678,
+      "loss": 0.5023207664489746,
+      "mean_token_accuracy": 0.8459693863987923,
+      "num_tokens": 2101528.0,
+      "step": 860
+    },
+    {
+      "epoch": 2.0723763570566947,
+      "eval_entropy": 0.5592134784111816,
+      "eval_loss": 0.620693564414978,
+      "eval_mean_token_accuracy": 0.827402507656076,
+      "eval_num_tokens": 2101528.0,
+      "eval_runtime": 100.6274,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 860
+    },
+    {
+      "entropy": 0.5214565828442573,
+      "epoch": 2.120627261761158,
+      "grad_norm": 0.45878875255584717,
+      "learning_rate": 0.00041610510539937234,
+      "loss": 0.48367843627929685,
+      "mean_token_accuracy": 0.8515541926026344,
+      "num_tokens": 2149550.0,
+      "step": 880
+    },
+    {
+      "epoch": 2.120627261761158,
+      "eval_entropy": 0.5860532344727034,
+      "eval_loss": 0.6134791970252991,
+      "eval_mean_token_accuracy": 0.8291774777214179,
+      "eval_num_tokens": 2149550.0,
+      "eval_runtime": 100.5295,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 880
+    },
+    {
+      "entropy": 0.5414379432797432,
+      "epoch": 2.168878166465621,
+      "grad_norm": 0.5976294875144958,
+      "learning_rate": 0.0004146933107930829,
+      "loss": 0.5012908935546875,
+      "mean_token_accuracy": 0.8466357633471489,
+      "num_tokens": 2202162.0,
+      "step": 900
+    },
+    {
+      "epoch": 2.168878166465621,
+      "eval_entropy": 0.5427783280611038,
+      "eval_loss": 0.612716019153595,
+      "eval_mean_token_accuracy": 0.829505049445656,
+      "eval_num_tokens": 2202162.0,
+      "eval_runtime": 100.5207,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 900
+    },
+    {
+      "entropy": 0.521108990907669,
+      "epoch": 2.2171290711700844,
+      "grad_norm": 0.5145775675773621,
+      "learning_rate": 0.000413225339467373,
+      "loss": 0.4877732276916504,
+      "mean_token_accuracy": 0.8504189744591713,
+      "num_tokens": 2256539.0,
+      "step": 920
+    },
+    {
+      "epoch": 2.2171290711700844,
+      "eval_entropy": 0.5620684869838565,
+      "eval_loss": 0.6176337599754333,
+      "eval_mean_token_accuracy": 0.8291444483767735,
+      "eval_num_tokens": 2256539.0,
+      "eval_runtime": 100.6819,
+      "eval_samples_per_second": 14.104,
+      "eval_steps_per_second": 1.768,
+      "step": 920
+    },
+    {
+      "entropy": 0.5452766507863999,
+      "epoch": 2.2653799758745476,
+      "grad_norm": 0.4820452928543091,
+      "learning_rate": 0.0004117016068407636,
+      "loss": 0.4991307735443115,
+      "mean_token_accuracy": 0.8474363312125206,
+      "num_tokens": 2299698.0,
+      "step": 940
+    },
+    {
+      "epoch": 2.2653799758745476,
+      "eval_entropy": 0.5611683193887218,
+      "eval_loss": 0.6153580546379089,
+      "eval_mean_token_accuracy": 0.8286041272490212,
+      "eval_num_tokens": 2299698.0,
+      "eval_runtime": 100.5249,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 940
+    },
+    {
+      "entropy": 0.5572576269507408,
+      "epoch": 2.313630880579011,
+      "grad_norm": 0.3947993814945221,
+      "learning_rate": 0.00041012254411156387,
+      "loss": 0.5191003799438476,
+      "mean_token_accuracy": 0.8422838419675827,
+      "num_tokens": 2347134.0,
+      "step": 960
+    },
+    {
+      "epoch": 2.313630880579011,
+      "eval_entropy": 0.5651392345682958,
+      "eval_loss": 0.6121005415916443,
+      "eval_mean_token_accuracy": 0.8286536242185014,
+      "eval_num_tokens": 2347134.0,
+      "eval_runtime": 100.7033,
+      "eval_samples_per_second": 14.101,
+      "eval_steps_per_second": 1.768,
+      "step": 960
+    },
+    {
+      "entropy": 0.5441671140491963,
+      "epoch": 2.361881785283474,
+      "grad_norm": 0.7719791531562805,
+      "learning_rate": 0.0004084885981358477,
+      "loss": 0.5072606086730957,
+      "mean_token_accuracy": 0.8478495761752128,
+      "num_tokens": 2394568.0,
+      "step": 980
+    },
+    {
+      "epoch": 2.361881785283474,
+      "eval_entropy": 0.5651004270891126,
+      "eval_loss": 0.6069501042366028,
+      "eval_mean_token_accuracy": 0.8311042156112328,
+      "eval_num_tokens": 2394568.0,
+      "eval_runtime": 100.5487,
+      "eval_samples_per_second": 14.123,
+      "eval_steps_per_second": 1.77,
+      "step": 980
+    },
+    {
+      "entropy": 0.5462074346840382,
+      "epoch": 2.4101326899879374,
+      "grad_norm": 0.5694436430931091,
+      "learning_rate": 0.00040680023130099856,
+      "loss": 0.5097150325775146,
+      "mean_token_accuracy": 0.8474388793110847,
+      "num_tokens": 2443253.0,
+      "step": 1000
+    },
+    {
+      "epoch": 2.4101326899879374,
+      "eval_entropy": 0.5895893243926295,
+      "eval_loss": 0.6078168749809265,
+      "eval_mean_token_accuracy": 0.828030709470256,
+      "eval_num_tokens": 2443253.0,
+      "eval_runtime": 100.5413,
+      "eval_samples_per_second": 14.124,
+      "eval_steps_per_second": 1.77,
+      "step": 1000
+    },
+    {
+      "entropy": 0.5473216332495212,
+      "epoch": 2.4583835946924006,
+      "grad_norm": 0.3989660143852234,
+      "learning_rate": 0.00040505792139485913,
+      "loss": 0.5087806224822998,
+      "mean_token_accuracy": 0.8473173275589942,
+      "num_tokens": 2495178.0,
+      "step": 1020
+    },
+    {
+      "epoch": 2.4583835946924006,
+      "eval_entropy": 0.5588909702354603,
+      "eval_loss": 0.6087605953216553,
+      "eval_mean_token_accuracy": 0.8309365524334854,
+      "eval_num_tokens": 2495178.0,
+      "eval_runtime": 100.5137,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 1020
+    },
+    {
+      "entropy": 0.5430163390934467,
+      "epoch": 2.5066344993968634,
+      "grad_norm": 0.4916608929634094,
+      "learning_rate": 0.00040326216147052336,
+      "loss": 0.5027707576751709,
+      "mean_token_accuracy": 0.8466959804296493,
+      "num_tokens": 2541733.0,
+      "step": 1040
+    },
+    {
+      "epoch": 2.5066344993968634,
+      "eval_entropy": 0.5819033988741007,
+      "eval_loss": 0.609398365020752,
+      "eval_mean_token_accuracy": 0.8297666612635838,
+      "eval_num_tokens": 2541733.0,
+      "eval_runtime": 100.6375,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 1040
+    },
+    {
+      "entropy": 0.5632683664560318,
+      "epoch": 2.554885404101327,
+      "grad_norm": 0.6062753796577454,
+      "learning_rate": 0.00040141345970680814,
+      "loss": 0.5272740840911865,
+      "mean_token_accuracy": 0.8411559104919434,
+      "num_tokens": 2589479.0,
+      "step": 1060
+    },
+    {
+      "epoch": 2.554885404101327,
+      "eval_entropy": 0.5785442567273472,
+      "eval_loss": 0.6207635998725891,
+      "eval_mean_token_accuracy": 0.8259770478425401,
+      "eval_num_tokens": 2589479.0,
+      "eval_runtime": 100.5161,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 1060
+    },
+    {
+      "entropy": 0.5591374494135379,
+      "epoch": 2.60313630880579,
+      "grad_norm": 0.5771989226341248,
+      "learning_rate": 0.00039951233926444437,
+      "loss": 0.5237057685852051,
+      "mean_token_accuracy": 0.8413331478834152,
+      "num_tokens": 2639134.0,
+      "step": 1080
+    },
+    {
+      "epoch": 2.60313630880579,
+      "eval_entropy": 0.5765141543377651,
+      "eval_loss": 0.6119762063026428,
+      "eval_mean_token_accuracy": 0.8289416998289945,
+      "eval_num_tokens": 2639134.0,
+      "eval_runtime": 100.5288,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 1080
+    }
+  ],
+  "logging_steps": 20,
+  "max_steps": 4150,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 20,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.47206953373696e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1100/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen3-14B-Base
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen3-14B-Base
+- lora
+- sft
+- transformers
+- trl
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1100/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-14B-Base",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 128,
+  "lora_bias": false,
+  "lora_dropout": 0.0539675908752296,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "gate_proj",
+    "q_proj",
+    "down_proj",
+    "o_proj",
+    "k_proj",
+    "v_proj",
+    "up_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1100/chat_template.jinja

@@ -0,0 +1,85 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1100/tokenizer_config.json

@@ -0,0 +1,29 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1100/trainer_state.json

@@ -0,0 +1,1189 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.651387213510253,
+  "eval_steps": 20,
+  "global_step": 1100,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 1.8160341799259185,
+      "epoch": 0.04825090470446321,
+      "grad_norm": 1.1313843727111816,
+      "learning_rate": 1.9794859683580766e-05,
+      "loss": 1.7627403259277343,
+      "mean_token_accuracy": 0.6354069545865059,
+      "num_tokens": 50237.0,
+      "step": 20
+    },
+    {
+      "epoch": 0.04825090470446321,
+      "eval_entropy": 1.538598193881217,
+      "eval_loss": 1.4881398677825928,
+      "eval_mean_token_accuracy": 0.686459881201219,
+      "eval_num_tokens": 50237.0,
+      "eval_runtime": 100.8923,
+      "eval_samples_per_second": 14.074,
+      "eval_steps_per_second": 1.764,
+      "step": 20
+    },
+    {
+      "entropy": 1.1673978716135025,
+      "epoch": 0.09650180940892641,
+      "grad_norm": 0.6167025566101074,
+      "learning_rate": 4.0631554087349995e-05,
+      "loss": 1.0887075424194337,
+      "mean_token_accuracy": 0.737513878941536,
+      "num_tokens": 101098.0,
+      "step": 40
+    },
+    {
+      "epoch": 0.09650180940892641,
+      "eval_entropy": 0.887577452351538,
+      "eval_loss": 0.8519237041473389,
+      "eval_mean_token_accuracy": 0.7741017204322173,
+      "eval_num_tokens": 101098.0,
+      "eval_runtime": 100.4819,
+      "eval_samples_per_second": 14.132,
+      "eval_steps_per_second": 1.771,
+      "step": 40
+    },
+    {
+      "entropy": 0.8220535531640053,
+      "epoch": 0.14475271411338964,
+      "grad_norm": 0.8396430611610413,
+      "learning_rate": 6.146824849111923e-05,
+      "loss": 0.7812146186828614,
+      "mean_token_accuracy": 0.7872247830033302,
+      "num_tokens": 150385.0,
+      "step": 60
+    },
+    {
+      "epoch": 0.14475271411338964,
+      "eval_entropy": 0.770665143982748,
+      "eval_loss": 0.7478981614112854,
+      "eval_mean_token_accuracy": 0.7949167068084974,
+      "eval_num_tokens": 150385.0,
+      "eval_runtime": 100.6518,
+      "eval_samples_per_second": 14.108,
+      "eval_steps_per_second": 1.768,
+      "step": 60
+    },
+    {
+      "entropy": 0.7776695728302002,
+      "epoch": 0.19300361881785283,
+      "grad_norm": 0.5731202363967896,
+      "learning_rate": 8.230494289488845e-05,
+      "loss": 0.732721471786499,
+      "mean_token_accuracy": 0.7976997002959252,
+      "num_tokens": 197354.0,
+      "step": 80
+    },
+    {
+      "epoch": 0.19300361881785283,
+      "eval_entropy": 0.7650937618834249,
+      "eval_loss": 0.7109046578407288,
+      "eval_mean_token_accuracy": 0.8018777065732506,
+      "eval_num_tokens": 197354.0,
+      "eval_runtime": 100.5983,
+      "eval_samples_per_second": 14.116,
+      "eval_steps_per_second": 1.769,
+      "step": 80
+    },
+    {
+      "entropy": 0.7309209659695626,
+      "epoch": 0.24125452352231605,
+      "grad_norm": 0.48046866059303284,
+      "learning_rate": 0.00010314163729865769,
+      "loss": 0.680583143234253,
+      "mean_token_accuracy": 0.8073364540934562,
+      "num_tokens": 248557.0,
+      "step": 100
+    },
+    {
+      "epoch": 0.24125452352231605,
+      "eval_entropy": 0.7152130687504672,
+      "eval_loss": 0.6989707350730896,
+      "eval_mean_token_accuracy": 0.8057264320636064,
+      "eval_num_tokens": 248557.0,
+      "eval_runtime": 100.6264,
+      "eval_samples_per_second": 14.112,
+      "eval_steps_per_second": 1.769,
+      "step": 100
+    },
+    {
+      "entropy": 0.7105271726846695,
+      "epoch": 0.28950542822677927,
+      "grad_norm": 0.5512468814849854,
+      "learning_rate": 0.00012397833170242693,
+      "loss": 0.6674309253692627,
+      "mean_token_accuracy": 0.8098619893193245,
+      "num_tokens": 298041.0,
+      "step": 120
+    },
+    {
+      "epoch": 0.28950542822677927,
+      "eval_entropy": 0.7314498555124476,
+      "eval_loss": 0.6796280145645142,
+      "eval_mean_token_accuracy": 0.809063873264227,
+      "eval_num_tokens": 298041.0,
+      "eval_runtime": 100.6132,
+      "eval_samples_per_second": 14.113,
+      "eval_steps_per_second": 1.769,
+      "step": 120
+    },
+    {
+      "entropy": 0.7079856500029564,
+      "epoch": 0.33775633293124246,
+      "grad_norm": 0.5903022885322571,
+      "learning_rate": 0.00014481502610619613,
+      "loss": 0.6623000144958496,
+      "mean_token_accuracy": 0.8101129323244095,
+      "num_tokens": 348955.0,
+      "step": 140
+    },
+    {
+      "epoch": 0.33775633293124246,
+      "eval_entropy": 0.7005330203624254,
+      "eval_loss": 0.671779453754425,
+      "eval_mean_token_accuracy": 0.8106091374091888,
+      "eval_num_tokens": 348955.0,
+      "eval_runtime": 100.6904,
+      "eval_samples_per_second": 14.103,
+      "eval_steps_per_second": 1.768,
+      "step": 140
+    },
+    {
+      "entropy": 0.7109143257141113,
+      "epoch": 0.38600723763570566,
+      "grad_norm": 0.4516378343105316,
+      "learning_rate": 0.00016565172050996536,
+      "loss": 0.658915901184082,
+      "mean_token_accuracy": 0.8112782314419746,
+      "num_tokens": 395798.0,
+      "step": 160
+    },
+    {
+      "epoch": 0.38600723763570566,
+      "eval_entropy": 0.6891461650976974,
+      "eval_loss": 0.6598312258720398,
+      "eval_mean_token_accuracy": 0.8133360951134329,
+      "eval_num_tokens": 395798.0,
+      "eval_runtime": 100.5057,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 160
+    },
+    {
+      "entropy": 0.6896152347326279,
+      "epoch": 0.43425814234016885,
+      "grad_norm": 0.49479565024375916,
+      "learning_rate": 0.0001864884149137346,
+      "loss": 0.6466911315917969,
+      "mean_token_accuracy": 0.8141682580113411,
+      "num_tokens": 446437.0,
+      "step": 180
+    },
+    {
+      "epoch": 0.43425814234016885,
+      "eval_entropy": 0.6984815888860253,
+      "eval_loss": 0.652909517288208,
+      "eval_mean_token_accuracy": 0.8133517600847094,
+      "eval_num_tokens": 446437.0,
+      "eval_runtime": 100.5867,
+      "eval_samples_per_second": 14.117,
+      "eval_steps_per_second": 1.77,
+      "step": 180
+    },
+    {
+      "entropy": 0.6829591259360314,
+      "epoch": 0.4825090470446321,
+      "grad_norm": 0.4835337698459625,
+      "learning_rate": 0.00020732510931750382,
+      "loss": 0.6401876449584961,
+      "mean_token_accuracy": 0.8175502926111221,
+      "num_tokens": 497682.0,
+      "step": 200
+    },
+    {
+      "epoch": 0.4825090470446321,
+      "eval_entropy": 0.6963563973314306,
+      "eval_loss": 0.650087296962738,
+      "eval_mean_token_accuracy": 0.8159684401549651,
+      "eval_num_tokens": 497682.0,
+      "eval_runtime": 100.6277,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 200
+    },
+    {
+      "entropy": 0.6903425320982933,
+      "epoch": 0.5307599517490953,
+      "grad_norm": 0.44980454444885254,
+      "learning_rate": 0.00022816180372127305,
+      "loss": 0.6409444808959961,
+      "mean_token_accuracy": 0.8163257524371147,
+      "num_tokens": 544670.0,
+      "step": 220
+    },
+    {
+      "epoch": 0.5307599517490953,
+      "eval_entropy": 0.6687790182199371,
+      "eval_loss": 0.6451645493507385,
+      "eval_mean_token_accuracy": 0.8157875551266617,
+      "eval_num_tokens": 544670.0,
+      "eval_runtime": 100.5444,
+      "eval_samples_per_second": 14.123,
+      "eval_steps_per_second": 1.77,
+      "step": 220
+    },
+    {
+      "entropy": 0.6740772560238838,
+      "epoch": 0.5790108564535585,
+      "grad_norm": 0.5312877297401428,
+      "learning_rate": 0.0002489984981250423,
+      "loss": 0.6307676315307618,
+      "mean_token_accuracy": 0.8174169659614563,
+      "num_tokens": 592926.0,
+      "step": 240
+    },
+    {
+      "epoch": 0.5790108564535585,
+      "eval_entropy": 0.7063750872451268,
+      "eval_loss": 0.6494986414909363,
+      "eval_mean_token_accuracy": 0.8156777198395032,
+      "eval_num_tokens": 592926.0,
+      "eval_runtime": 100.5094,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 240
+    },
+    {
+      "entropy": 0.6876470118761062,
+      "epoch": 0.6272617611580217,
+      "grad_norm": 0.46149370074272156,
+      "learning_rate": 0.00026983519252881153,
+      "loss": 0.6392203807830811,
+      "mean_token_accuracy": 0.8172321930527687,
+      "num_tokens": 639077.0,
+      "step": 260
+    },
+    {
+      "epoch": 0.6272617611580217,
+      "eval_entropy": 0.625594850838854,
+      "eval_loss": 0.6463819742202759,
+      "eval_mean_token_accuracy": 0.8179670525400826,
+      "eval_num_tokens": 639077.0,
+      "eval_runtime": 100.5775,
+      "eval_samples_per_second": 14.118,
+      "eval_steps_per_second": 1.77,
+      "step": 260
+    },
+    {
+      "entropy": 0.6673879995942116,
+      "epoch": 0.6755126658624849,
+      "grad_norm": 0.42477288842201233,
+      "learning_rate": 0.0002906718869325807,
+      "loss": 0.6300291538238525,
+      "mean_token_accuracy": 0.8203139930963517,
+      "num_tokens": 691028.0,
+      "step": 280
+    },
+    {
+      "epoch": 0.6755126658624849,
+      "eval_entropy": 0.6863534494062488,
+      "eval_loss": 0.6392037272453308,
+      "eval_mean_token_accuracy": 0.8190520385677895,
+      "eval_num_tokens": 691028.0,
+      "eval_runtime": 100.5106,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 280
+    },
+    {
+      "entropy": 0.6573980256915093,
+      "epoch": 0.7237635705669482,
+      "grad_norm": 0.4656430780887604,
+      "learning_rate": 0.00031150858133635,
+      "loss": 0.6215964794158936,
+      "mean_token_accuracy": 0.8208559453487396,
+      "num_tokens": 740255.0,
+      "step": 300
+    },
+    {
+      "epoch": 0.7237635705669482,
+      "eval_entropy": 0.6834242812703165,
+      "eval_loss": 0.6479960083961487,
+      "eval_mean_token_accuracy": 0.8128752906001016,
+      "eval_num_tokens": 740255.0,
+      "eval_runtime": 100.6101,
+      "eval_samples_per_second": 14.114,
+      "eval_steps_per_second": 1.769,
+      "step": 300
+    },
+    {
+      "entropy": 0.6751310348510742,
+      "epoch": 0.7720144752714113,
+      "grad_norm": 0.4495522081851959,
+      "learning_rate": 0.00033234527574011916,
+      "loss": 0.6402668952941895,
+      "mean_token_accuracy": 0.8170571058988572,
+      "num_tokens": 789529.0,
+      "step": 320
+    },
+    {
+      "epoch": 0.7720144752714113,
+      "eval_entropy": 0.6920045178927733,
+      "eval_loss": 0.6402787566184998,
+      "eval_mean_token_accuracy": 0.817791668886549,
+      "eval_num_tokens": 789529.0,
+      "eval_runtime": 100.6347,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 320
+    },
+    {
+      "entropy": 0.6846290081739426,
+      "epoch": 0.8202653799758746,
+      "grad_norm": 0.5661432147026062,
+      "learning_rate": 0.00035318197014388845,
+      "loss": 0.6508871078491211,
+      "mean_token_accuracy": 0.8164169207215309,
+      "num_tokens": 841584.0,
+      "step": 340
+    },
+    {
+      "epoch": 0.8202653799758746,
+      "eval_entropy": 0.6772527078564248,
+      "eval_loss": 0.6512064337730408,
+      "eval_mean_token_accuracy": 0.8170605137776793,
+      "eval_num_tokens": 841584.0,
+      "eval_runtime": 100.5157,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 340
+    },
+    {
+      "entropy": 0.7022885024547577,
+      "epoch": 0.8685162846803377,
+      "grad_norm": 0.6357236504554749,
+      "learning_rate": 0.0003740186645476577,
+      "loss": 0.6649135112762451,
+      "mean_token_accuracy": 0.8136772960424423,
+      "num_tokens": 885769.0,
+      "step": 360
+    },
+    {
+      "epoch": 0.8685162846803377,
+      "eval_entropy": 0.7435127543599418,
+      "eval_loss": 0.6573262214660645,
+      "eval_mean_token_accuracy": 0.8147802225659403,
+      "eval_num_tokens": 885769.0,
+      "eval_runtime": 100.5271,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 360
+    },
+    {
+      "entropy": 0.6962979376316071,
+      "epoch": 0.916767189384801,
+      "grad_norm": 0.6179454326629639,
+      "learning_rate": 0.00039485535895142685,
+      "loss": 0.6509191513061523,
+      "mean_token_accuracy": 0.81364226937294,
+      "num_tokens": 930056.0,
+      "step": 380
+    },
+    {
+      "epoch": 0.916767189384801,
+      "eval_entropy": 0.6630894883295123,
+      "eval_loss": 0.6524088978767395,
+      "eval_mean_token_accuracy": 0.8165898587596551,
+      "eval_num_tokens": 930056.0,
+      "eval_runtime": 100.5343,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 380
+    },
+    {
+      "entropy": 0.694319449365139,
+      "epoch": 0.9650180940892642,
+      "grad_norm": 0.5458664894104004,
+      "learning_rate": 0.00041569205335519613,
+      "loss": 0.6624011993408203,
+      "mean_token_accuracy": 0.8136080905795098,
+      "num_tokens": 976892.0,
+      "step": 400
+    },
+    {
+      "epoch": 0.9650180940892642,
+      "eval_entropy": 0.7424925324622165,
+      "eval_loss": 0.6717661023139954,
+      "eval_mean_token_accuracy": 0.8099894781460922,
+      "eval_num_tokens": 976892.0,
+      "eval_runtime": 100.5992,
+      "eval_samples_per_second": 14.115,
+      "eval_steps_per_second": 1.769,
+      "step": 400
+    },
+    {
+      "entropy": 0.6861714736009256,
+      "epoch": 1.0120627261761157,
+      "grad_norm": 0.6108037829399109,
+      "learning_rate": 0.0004323601853188671,
+      "loss": 0.6551628112792969,
+      "mean_token_accuracy": 0.8134882618219424,
+      "num_tokens": 1024472.0,
+      "step": 420
+    },
+    {
+      "epoch": 1.0120627261761157,
+      "eval_entropy": 0.6593745705786715,
+      "eval_loss": 0.6720747947692871,
+      "eval_mean_token_accuracy": 0.812802647941568,
+      "eval_num_tokens": 1024472.0,
+      "eval_runtime": 100.4866,
+      "eval_samples_per_second": 14.131,
+      "eval_steps_per_second": 1.771,
+      "step": 420
+    },
+    {
+      "entropy": 0.639520151913166,
+      "epoch": 1.060313630880579,
+      "grad_norm": 0.6054353713989258,
+      "learning_rate": 0.00043231736219609266,
+      "loss": 0.6096667766571044,
+      "mean_token_accuracy": 0.8227144703269005,
+      "num_tokens": 1076964.0,
+      "step": 440
+    },
+    {
+      "epoch": 1.060313630880579,
+      "eval_entropy": 0.6316307876886946,
+      "eval_loss": 0.6621950268745422,
+      "eval_mean_token_accuracy": 0.8140057914712456,
+      "eval_num_tokens": 1076964.0,
+      "eval_runtime": 100.6409,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 440
+    },
+    {
+      "entropy": 0.6544070363044738,
+      "epoch": 1.1085645355850422,
+      "grad_norm": 0.5111057758331299,
+      "learning_rate": 0.0004322133749558233,
+      "loss": 0.6176822185516357,
+      "mean_token_accuracy": 0.8201584875583648,
+      "num_tokens": 1126064.0,
+      "step": 460
+    },
+    {
+      "epoch": 1.1085645355850422,
+      "eval_entropy": 0.6714627766207363,
+      "eval_loss": 0.6738516092300415,
+      "eval_mean_token_accuracy": 0.8128616829936424,
+      "eval_num_tokens": 1126064.0,
+      "eval_runtime": 100.4986,
+      "eval_samples_per_second": 14.13,
+      "eval_steps_per_second": 1.771,
+      "step": 460
+    },
+    {
+      "entropy": 0.6470697179436684,
+      "epoch": 1.1568154402895054,
+      "grad_norm": 0.5434669256210327,
+      "learning_rate": 0.0004320482530252179,
+      "loss": 0.6144762516021729,
+      "mean_token_accuracy": 0.819926056265831,
+      "num_tokens": 1176912.0,
+      "step": 480
+    },
+    {
+      "epoch": 1.1568154402895054,
+      "eval_entropy": 0.6475969080844622,
+      "eval_loss": 0.6624214053153992,
+      "eval_mean_token_accuracy": 0.8137228820431098,
+      "eval_num_tokens": 1176912.0,
+      "eval_runtime": 100.5617,
+      "eval_samples_per_second": 14.121,
+      "eval_steps_per_second": 1.77,
+      "step": 480
+    },
+    {
+      "entropy": 0.6659130036830903,
+      "epoch": 1.2050663449939687,
+      "grad_norm": 0.6470707058906555,
+      "learning_rate": 0.0004318220431318296,
+      "loss": 0.6366308212280274,
+      "mean_token_accuracy": 0.8175261527299881,
+      "num_tokens": 1223383.0,
+      "step": 500
+    },
+    {
+      "epoch": 1.2050663449939687,
+      "eval_entropy": 0.7232425132494295,
+      "eval_loss": 0.6877738833427429,
+      "eval_mean_token_accuracy": 0.8007767260744331,
+      "eval_num_tokens": 1223383.0,
+      "eval_runtime": 100.477,
+      "eval_samples_per_second": 14.133,
+      "eval_steps_per_second": 1.772,
+      "step": 500
+    },
+    {
+      "entropy": 0.6759189993143082,
+      "epoch": 1.2533172496984317,
+      "grad_norm": 0.6453561186790466,
+      "learning_rate": 0.00043153480929038245,
+      "loss": 0.637772512435913,
+      "mean_token_accuracy": 0.8162627041339874,
+      "num_tokens": 1273534.0,
+      "step": 520
+    },
+    {
+      "epoch": 1.2533172496984317,
+      "eval_entropy": 0.7082584312792575,
+      "eval_loss": 0.6695764064788818,
+      "eval_mean_token_accuracy": 0.812019992410467,
+      "eval_num_tokens": 1273534.0,
+      "eval_runtime": 100.4957,
+      "eval_samples_per_second": 14.13,
+      "eval_steps_per_second": 1.771,
+      "step": 520
+    },
+    {
+      "entropy": 0.6730152070522308,
+      "epoch": 1.301568154402895,
+      "grad_norm": 0.6234476566314697,
+      "learning_rate": 0.000431186632784656,
+      "loss": 0.640014362335205,
+      "mean_token_accuracy": 0.8168858170509339,
+      "num_tokens": 1323351.0,
+      "step": 540
+    },
+    {
+      "epoch": 1.301568154402895,
+      "eval_entropy": 0.6655137846978862,
+      "eval_loss": 0.6578248739242554,
+      "eval_mean_token_accuracy": 0.8147600979617472,
+      "eval_num_tokens": 1323351.0,
+      "eval_runtime": 100.503,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 540
+    },
+    {
+      "entropy": 0.6636043325066566,
+      "epoch": 1.3498190591073582,
+      "grad_norm": 0.5882512331008911,
+      "learning_rate": 0.0004307776121444827,
+      "loss": 0.6286775588989257,
+      "mean_token_accuracy": 0.8183174058794975,
+      "num_tokens": 1371508.0,
+      "step": 560
+    },
+    {
+      "epoch": 1.3498190591073582,
+      "eval_entropy": 0.6304813810948575,
+      "eval_loss": 0.6607436537742615,
+      "eval_mean_token_accuracy": 0.8169659268989992,
+      "eval_num_tokens": 1371508.0,
+      "eval_runtime": 100.6233,
+      "eval_samples_per_second": 14.112,
+      "eval_steps_per_second": 1.769,
+      "step": 560
+    },
+    {
+      "entropy": 0.6608738481998444,
+      "epoch": 1.3980699638118215,
+      "grad_norm": 0.6796398758888245,
+      "learning_rate": 0.00043030786311786564,
+      "loss": 0.622726583480835,
+      "mean_token_accuracy": 0.8230827242136002,
+      "num_tokens": 1415221.0,
+      "step": 580
+    },
+    {
+      "epoch": 1.3980699638118215,
+      "eval_entropy": 0.6645242689030894,
+      "eval_loss": 0.6662811636924744,
+      "eval_mean_token_accuracy": 0.8147623907314258,
+      "eval_num_tokens": 1415221.0,
+      "eval_runtime": 100.5074,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 580
+    },
+    {
+      "entropy": 0.6682895936071873,
+      "epoch": 1.4463208685162847,
+      "grad_norm": 0.5750718712806702,
+      "learning_rate": 0.0004297775186382226,
+      "loss": 0.6304276466369629,
+      "mean_token_accuracy": 0.8188625335693359,
+      "num_tokens": 1463356.0,
+      "step": 600
+    },
+    {
+      "epoch": 1.4463208685162847,
+      "eval_entropy": 0.6320532578765676,
+      "eval_loss": 0.650333046913147,
+      "eval_mean_token_accuracy": 0.8172853793321031,
+      "eval_num_tokens": 1463356.0,
+      "eval_runtime": 100.5116,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 600
+    },
+    {
+      "entropy": 0.653739221394062,
+      "epoch": 1.494571773220748,
+      "grad_norm": 0.5552534461021423,
+      "learning_rate": 0.00042918672878676804,
+      "loss": 0.6215689659118653,
+      "mean_token_accuracy": 0.8208896175026894,
+      "num_tokens": 1512340.0,
+      "step": 620
+    },
+    {
+      "epoch": 1.494571773220748,
+      "eval_entropy": 0.6285932062382109,
+      "eval_loss": 0.6392536163330078,
+      "eval_mean_token_accuracy": 0.8205167825302381,
+      "eval_num_tokens": 1512340.0,
+      "eval_runtime": 100.6392,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 620
+    },
+    {
+      "entropy": 0.6476897612214089,
+      "epoch": 1.5428226779252112,
+      "grad_norm": 0.536978542804718,
+      "learning_rate": 0.00042853566075004156,
+      "loss": 0.6131608009338378,
+      "mean_token_accuracy": 0.8208152890205384,
+      "num_tokens": 1560911.0,
+      "step": 640
+    },
+    {
+      "epoch": 1.5428226779252112,
+      "eval_entropy": 0.6626498856571283,
+      "eval_loss": 0.6479025483131409,
+      "eval_mean_token_accuracy": 0.8180894101603647,
+      "eval_num_tokens": 1560911.0,
+      "eval_runtime": 100.5035,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 640
+    },
+    {
+      "entropy": 0.6338633447885513,
+      "epoch": 1.5910735826296745,
+      "grad_norm": 0.4518965780735016,
+      "learning_rate": 0.0004278244987725961,
+      "loss": 0.6010610580444335,
+      "mean_token_accuracy": 0.8247816175222397,
+      "num_tokens": 1609409.0,
+      "step": 660
+    },
+    {
+      "epoch": 1.5910735826296745,
+      "eval_entropy": 0.64079723699709,
+      "eval_loss": 0.6417157053947449,
+      "eval_mean_token_accuracy": 0.8214994581897607,
+      "eval_num_tokens": 1609409.0,
+      "eval_runtime": 100.6393,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 660
+    },
+    {
+      "entropy": 0.6485912457108498,
+      "epoch": 1.6393244873341375,
+      "grad_norm": 0.6028368473052979,
+      "learning_rate": 0.00042705344410485905,
+      "loss": 0.6120437145233154,
+      "mean_token_accuracy": 0.8236212745308876,
+      "num_tokens": 1657430.0,
+      "step": 680
+    },
+    {
+      "epoch": 1.6393244873341375,
+      "eval_entropy": 0.6863006427046958,
+      "eval_loss": 0.6420931816101074,
+      "eval_mean_token_accuracy": 0.8191533580924688,
+      "eval_num_tokens": 1657430.0,
+      "eval_runtime": 100.4639,
+      "eval_samples_per_second": 14.134,
+      "eval_steps_per_second": 1.772,
+      "step": 680
+    },
+    {
+      "entropy": 0.6459382772445679,
+      "epoch": 1.6875753920386007,
+      "grad_norm": 0.5069310069084167,
+      "learning_rate": 0.00042622271494618073,
+      "loss": 0.6160889625549316,
+      "mean_token_accuracy": 0.8253828629851341,
+      "num_tokens": 1708058.0,
+      "step": 700
+    },
+    {
+      "epoch": 1.6875753920386007,
+      "eval_entropy": 0.6773919463157654,
+      "eval_loss": 0.6307896375656128,
+      "eval_mean_token_accuracy": 0.8217989825800563,
+      "eval_num_tokens": 1708058.0,
+      "eval_runtime": 100.6301,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 700
+    },
+    {
+      "entropy": 0.64687028080225,
+      "epoch": 1.7358262967430638,
+      "grad_norm": 0.3830573856830597,
+      "learning_rate": 0.00042533254638308645,
+      "loss": 0.6104001998901367,
+      "mean_token_accuracy": 0.8255606815218925,
+      "num_tokens": 1761262.0,
+      "step": 720
+    },
+    {
+      "epoch": 1.7358262967430638,
+      "eval_entropy": 0.6179541145147902,
+      "eval_loss": 0.6170215606689453,
+      "eval_mean_token_accuracy": 0.8259220900160543,
+      "eval_num_tokens": 1761262.0,
+      "eval_runtime": 100.61,
+      "eval_samples_per_second": 14.114,
+      "eval_steps_per_second": 1.769,
+      "step": 720
+    },
+    {
+      "entropy": 0.6306161753833294,
+      "epoch": 1.784077201447527,
+      "grad_norm": 0.5888957977294922,
+      "learning_rate": 0.0004243831903227498,
+      "loss": 0.5981546878814697,
+      "mean_token_accuracy": 0.8266776099801063,
+      "num_tokens": 1812127.0,
+      "step": 740
+    },
+    {
+      "epoch": 1.784077201447527,
+      "eval_entropy": 0.6357416808605194,
+      "eval_loss": 0.6202242374420166,
+      "eval_mean_token_accuracy": 0.8253380304641937,
+      "eval_num_tokens": 1812127.0,
+      "eval_runtime": 100.5145,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 740
+    },
+    {
+      "entropy": 0.6290688961744308,
+      "epoch": 1.8323281061519903,
+      "grad_norm": 0.6138001680374146,
+      "learning_rate": 0.0004233749154217063,
+      "loss": 0.5853167057037354,
+      "mean_token_accuracy": 0.8281400203704834,
+      "num_tokens": 1858403.0,
+      "step": 760
+    },
+    {
+      "epoch": 1.8323281061519903,
+      "eval_entropy": 0.6402715869833914,
+      "eval_loss": 0.6168262958526611,
+      "eval_mean_token_accuracy": 0.8246346306934785,
+      "eval_num_tokens": 1858403.0,
+      "eval_runtime": 100.5237,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 760
+    },
+    {
+      "entropy": 0.6218615680932998,
+      "epoch": 1.8805790108564535,
+      "grad_norm": 0.574312686920166,
+      "learning_rate": 0.00042230800700982586,
+      "loss": 0.5841915607452393,
+      "mean_token_accuracy": 0.8304707780480385,
+      "num_tokens": 1907361.0,
+      "step": 780
+    },
+    {
+      "epoch": 1.8805790108564535,
+      "eval_entropy": 0.6301163148679091,
+      "eval_loss": 0.6073815226554871,
+      "eval_mean_token_accuracy": 0.8274695993809218,
+      "eval_num_tokens": 1907361.0,
+      "eval_runtime": 100.5111,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 780
+    },
+    {
+      "entropy": 0.6101033873856068,
+      "epoch": 1.9288299155609168,
+      "grad_norm": 0.5054804682731628,
+      "learning_rate": 0.000421182767009569,
+      "loss": 0.5774452209472656,
+      "mean_token_accuracy": 0.8321943372488022,
+      "num_tokens": 1956605.0,
+      "step": 800
+    },
+    {
+      "epoch": 1.9288299155609168,
+      "eval_entropy": 0.631400960215022,
+      "eval_loss": 0.6115217804908752,
+      "eval_mean_token_accuracy": 0.8249618762664581,
+      "eval_num_tokens": 1956605.0,
+      "eval_runtime": 100.4916,
+      "eval_samples_per_second": 14.131,
+      "eval_steps_per_second": 1.771,
+      "step": 800
+    },
+    {
+      "entropy": 0.6217561915516854,
+      "epoch": 1.97708082026538,
+      "grad_norm": 0.5333649516105652,
+      "learning_rate": 0.0004199995138505454,
+      "loss": 0.5842787265777588,
+      "mean_token_accuracy": 0.8305525362491608,
+      "num_tokens": 2003308.0,
+      "step": 820
+    },
+    {
+      "epoch": 1.97708082026538,
+      "eval_entropy": 0.605928597657868,
+      "eval_loss": 0.6067516207695007,
+      "eval_mean_token_accuracy": 0.82715918408351,
+      "eval_num_tokens": 2003308.0,
+      "eval_runtime": 100.5286,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 820
+    },
+    {
+      "entropy": 0.5585076526189462,
+      "epoch": 2.0241254523522314,
+      "grad_norm": 0.44819608330726624,
+      "learning_rate": 0.00041875858237940244,
+      "loss": 0.5252367496490479,
+      "mean_token_accuracy": 0.8405626966403081,
+      "num_tokens": 2051331.0,
+      "step": 840
+    },
+    {
+      "epoch": 2.0241254523522314,
+      "eval_entropy": 0.6123150769579276,
+      "eval_loss": 0.6175946593284607,
+      "eval_mean_token_accuracy": 0.827637837509091,
+      "eval_num_tokens": 2051331.0,
+      "eval_runtime": 100.5414,
+      "eval_samples_per_second": 14.124,
+      "eval_steps_per_second": 1.77,
+      "step": 840
+    },
+    {
+      "entropy": 0.5359115816652775,
+      "epoch": 2.0723763570566947,
+      "grad_norm": 0.576781690120697,
+      "learning_rate": 0.0004174603237650678,
+      "loss": 0.5023207664489746,
+      "mean_token_accuracy": 0.8459693863987923,
+      "num_tokens": 2101528.0,
+      "step": 860
+    },
+    {
+      "epoch": 2.0723763570566947,
+      "eval_entropy": 0.5592134784111816,
+      "eval_loss": 0.620693564414978,
+      "eval_mean_token_accuracy": 0.827402507656076,
+      "eval_num_tokens": 2101528.0,
+      "eval_runtime": 100.6274,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 860
+    },
+    {
+      "entropy": 0.5214565828442573,
+      "epoch": 2.120627261761158,
+      "grad_norm": 0.45878875255584717,
+      "learning_rate": 0.00041610510539937234,
+      "loss": 0.48367843627929685,
+      "mean_token_accuracy": 0.8515541926026344,
+      "num_tokens": 2149550.0,
+      "step": 880
+    },
+    {
+      "epoch": 2.120627261761158,
+      "eval_entropy": 0.5860532344727034,
+      "eval_loss": 0.6134791970252991,
+      "eval_mean_token_accuracy": 0.8291774777214179,
+      "eval_num_tokens": 2149550.0,
+      "eval_runtime": 100.5295,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 880
+    },
+    {
+      "entropy": 0.5414379432797432,
+      "epoch": 2.168878166465621,
+      "grad_norm": 0.5976294875144958,
+      "learning_rate": 0.0004146933107930829,
+      "loss": 0.5012908935546875,
+      "mean_token_accuracy": 0.8466357633471489,
+      "num_tokens": 2202162.0,
+      "step": 900
+    },
+    {
+      "epoch": 2.168878166465621,
+      "eval_entropy": 0.5427783280611038,
+      "eval_loss": 0.612716019153595,
+      "eval_mean_token_accuracy": 0.829505049445656,
+      "eval_num_tokens": 2202162.0,
+      "eval_runtime": 100.5207,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 900
+    },
+    {
+      "entropy": 0.521108990907669,
+      "epoch": 2.2171290711700844,
+      "grad_norm": 0.5145775675773621,
+      "learning_rate": 0.000413225339467373,
+      "loss": 0.4877732276916504,
+      "mean_token_accuracy": 0.8504189744591713,
+      "num_tokens": 2256539.0,
+      "step": 920
+    },
+    {
+      "epoch": 2.2171290711700844,
+      "eval_entropy": 0.5620684869838565,
+      "eval_loss": 0.6176337599754333,
+      "eval_mean_token_accuracy": 0.8291444483767735,
+      "eval_num_tokens": 2256539.0,
+      "eval_runtime": 100.6819,
+      "eval_samples_per_second": 14.104,
+      "eval_steps_per_second": 1.768,
+      "step": 920
+    },
+    {
+      "entropy": 0.5452766507863999,
+      "epoch": 2.2653799758745476,
+      "grad_norm": 0.4820452928543091,
+      "learning_rate": 0.0004117016068407636,
+      "loss": 0.4991307735443115,
+      "mean_token_accuracy": 0.8474363312125206,
+      "num_tokens": 2299698.0,
+      "step": 940
+    },
+    {
+      "epoch": 2.2653799758745476,
+      "eval_entropy": 0.5611683193887218,
+      "eval_loss": 0.6153580546379089,
+      "eval_mean_token_accuracy": 0.8286041272490212,
+      "eval_num_tokens": 2299698.0,
+      "eval_runtime": 100.5249,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 940
+    },
+    {
+      "entropy": 0.5572576269507408,
+      "epoch": 2.313630880579011,
+      "grad_norm": 0.3947993814945221,
+      "learning_rate": 0.00041012254411156387,
+      "loss": 0.5191003799438476,
+      "mean_token_accuracy": 0.8422838419675827,
+      "num_tokens": 2347134.0,
+      "step": 960
+    },
+    {
+      "epoch": 2.313630880579011,
+      "eval_entropy": 0.5651392345682958,
+      "eval_loss": 0.6121005415916443,
+      "eval_mean_token_accuracy": 0.8286536242185014,
+      "eval_num_tokens": 2347134.0,
+      "eval_runtime": 100.7033,
+      "eval_samples_per_second": 14.101,
+      "eval_steps_per_second": 1.768,
+      "step": 960
+    },
+    {
+      "entropy": 0.5441671140491963,
+      "epoch": 2.361881785283474,
+      "grad_norm": 0.7719791531562805,
+      "learning_rate": 0.0004084885981358477,
+      "loss": 0.5072606086730957,
+      "mean_token_accuracy": 0.8478495761752128,
+      "num_tokens": 2394568.0,
+      "step": 980
+    },
+    {
+      "epoch": 2.361881785283474,
+      "eval_entropy": 0.5651004270891126,
+      "eval_loss": 0.6069501042366028,
+      "eval_mean_token_accuracy": 0.8311042156112328,
+      "eval_num_tokens": 2394568.0,
+      "eval_runtime": 100.5487,
+      "eval_samples_per_second": 14.123,
+      "eval_steps_per_second": 1.77,
+      "step": 980
+    },
+    {
+      "entropy": 0.5462074346840382,
+      "epoch": 2.4101326899879374,
+      "grad_norm": 0.5694436430931091,
+      "learning_rate": 0.00040680023130099856,
+      "loss": 0.5097150325775146,
+      "mean_token_accuracy": 0.8474388793110847,
+      "num_tokens": 2443253.0,
+      "step": 1000
+    },
+    {
+      "epoch": 2.4101326899879374,
+      "eval_entropy": 0.5895893243926295,
+      "eval_loss": 0.6078168749809265,
+      "eval_mean_token_accuracy": 0.828030709470256,
+      "eval_num_tokens": 2443253.0,
+      "eval_runtime": 100.5413,
+      "eval_samples_per_second": 14.124,
+      "eval_steps_per_second": 1.77,
+      "step": 1000
+    },
+    {
+      "entropy": 0.5473216332495212,
+      "epoch": 2.4583835946924006,
+      "grad_norm": 0.3989660143852234,
+      "learning_rate": 0.00040505792139485913,
+      "loss": 0.5087806224822998,
+      "mean_token_accuracy": 0.8473173275589942,
+      "num_tokens": 2495178.0,
+      "step": 1020
+    },
+    {
+      "epoch": 2.4583835946924006,
+      "eval_entropy": 0.5588909702354603,
+      "eval_loss": 0.6087605953216553,
+      "eval_mean_token_accuracy": 0.8309365524334854,
+      "eval_num_tokens": 2495178.0,
+      "eval_runtime": 100.5137,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 1020
+    },
+    {
+      "entropy": 0.5430163390934467,
+      "epoch": 2.5066344993968634,
+      "grad_norm": 0.4916608929634094,
+      "learning_rate": 0.00040326216147052336,
+      "loss": 0.5027707576751709,
+      "mean_token_accuracy": 0.8466959804296493,
+      "num_tokens": 2541733.0,
+      "step": 1040
+    },
+    {
+      "epoch": 2.5066344993968634,
+      "eval_entropy": 0.5819033988741007,
+      "eval_loss": 0.609398365020752,
+      "eval_mean_token_accuracy": 0.8297666612635838,
+      "eval_num_tokens": 2541733.0,
+      "eval_runtime": 100.6375,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 1040
+    },
+    {
+      "entropy": 0.5632683664560318,
+      "epoch": 2.554885404101327,
+      "grad_norm": 0.6062753796577454,
+      "learning_rate": 0.00040141345970680814,
+      "loss": 0.5272740840911865,
+      "mean_token_accuracy": 0.8411559104919434,
+      "num_tokens": 2589479.0,
+      "step": 1060
+    },
+    {
+      "epoch": 2.554885404101327,
+      "eval_entropy": 0.5785442567273472,
+      "eval_loss": 0.6207635998725891,
+      "eval_mean_token_accuracy": 0.8259770478425401,
+      "eval_num_tokens": 2589479.0,
+      "eval_runtime": 100.5161,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 1060
+    },
+    {
+      "entropy": 0.5591374494135379,
+      "epoch": 2.60313630880579,
+      "grad_norm": 0.5771989226341248,
+      "learning_rate": 0.00039951233926444437,
+      "loss": 0.5237057685852051,
+      "mean_token_accuracy": 0.8413331478834152,
+      "num_tokens": 2639134.0,
+      "step": 1080
+    },
+    {
+      "epoch": 2.60313630880579,
+      "eval_entropy": 0.5765141543377651,
+      "eval_loss": 0.6119762063026428,
+      "eval_mean_token_accuracy": 0.8289416998289945,
+      "eval_num_tokens": 2639134.0,
+      "eval_runtime": 100.5288,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 1080
+    },
+    {
+      "entropy": 0.5470930114388466,
+      "epoch": 2.651387213510253,
+      "grad_norm": 0.4211709499359131,
+      "learning_rate": 0.00039755933813802937,
+      "loss": 0.5157970428466797,
+      "mean_token_accuracy": 0.8436284482479095,
+      "num_tokens": 2689324.0,
+      "step": 1100
+    },
+    {
+      "epoch": 2.651387213510253,
+      "eval_entropy": 0.5593592817528864,
+      "eval_loss": 0.6080028414726257,
+      "eval_mean_token_accuracy": 0.8277805718813049,
+      "eval_num_tokens": 2689324.0,
+      "eval_runtime": 100.495,
+      "eval_samples_per_second": 14.13,
+      "eval_steps_per_second": 1.771,
+      "step": 1100
+    }
+  ],
+  "logging_steps": 20,
+  "max_steps": 4150,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 20,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.57585367650304e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1120/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen3-14B-Base
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen3-14B-Base
+- lora
+- sft
+- transformers
+- trl
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1120/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-14B-Base",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 128,
+  "lora_bias": false,
+  "lora_dropout": 0.0539675908752296,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "gate_proj",
+    "q_proj",
+    "down_proj",
+    "o_proj",
+    "k_proj",
+    "v_proj",
+    "up_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1120/chat_template.jinja

@@ -0,0 +1,85 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1120/tokenizer_config.json

@@ -0,0 +1,29 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1120/trainer_state.json

@@ -0,0 +1,1210 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.6996381182147164,
+  "eval_steps": 20,
+  "global_step": 1120,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 1.8160341799259185,
+      "epoch": 0.04825090470446321,
+      "grad_norm": 1.1313843727111816,
+      "learning_rate": 1.9794859683580766e-05,
+      "loss": 1.7627403259277343,
+      "mean_token_accuracy": 0.6354069545865059,
+      "num_tokens": 50237.0,
+      "step": 20
+    },
+    {
+      "epoch": 0.04825090470446321,
+      "eval_entropy": 1.538598193881217,
+      "eval_loss": 1.4881398677825928,
+      "eval_mean_token_accuracy": 0.686459881201219,
+      "eval_num_tokens": 50237.0,
+      "eval_runtime": 100.8923,
+      "eval_samples_per_second": 14.074,
+      "eval_steps_per_second": 1.764,
+      "step": 20
+    },
+    {
+      "entropy": 1.1673978716135025,
+      "epoch": 0.09650180940892641,
+      "grad_norm": 0.6167025566101074,
+      "learning_rate": 4.0631554087349995e-05,
+      "loss": 1.0887075424194337,
+      "mean_token_accuracy": 0.737513878941536,
+      "num_tokens": 101098.0,
+      "step": 40
+    },
+    {
+      "epoch": 0.09650180940892641,
+      "eval_entropy": 0.887577452351538,
+      "eval_loss": 0.8519237041473389,
+      "eval_mean_token_accuracy": 0.7741017204322173,
+      "eval_num_tokens": 101098.0,
+      "eval_runtime": 100.4819,
+      "eval_samples_per_second": 14.132,
+      "eval_steps_per_second": 1.771,
+      "step": 40
+    },
+    {
+      "entropy": 0.8220535531640053,
+      "epoch": 0.14475271411338964,
+      "grad_norm": 0.8396430611610413,
+      "learning_rate": 6.146824849111923e-05,
+      "loss": 0.7812146186828614,
+      "mean_token_accuracy": 0.7872247830033302,
+      "num_tokens": 150385.0,
+      "step": 60
+    },
+    {
+      "epoch": 0.14475271411338964,
+      "eval_entropy": 0.770665143982748,
+      "eval_loss": 0.7478981614112854,
+      "eval_mean_token_accuracy": 0.7949167068084974,
+      "eval_num_tokens": 150385.0,
+      "eval_runtime": 100.6518,
+      "eval_samples_per_second": 14.108,
+      "eval_steps_per_second": 1.768,
+      "step": 60
+    },
+    {
+      "entropy": 0.7776695728302002,
+      "epoch": 0.19300361881785283,
+      "grad_norm": 0.5731202363967896,
+      "learning_rate": 8.230494289488845e-05,
+      "loss": 0.732721471786499,
+      "mean_token_accuracy": 0.7976997002959252,
+      "num_tokens": 197354.0,
+      "step": 80
+    },
+    {
+      "epoch": 0.19300361881785283,
+      "eval_entropy": 0.7650937618834249,
+      "eval_loss": 0.7109046578407288,
+      "eval_mean_token_accuracy": 0.8018777065732506,
+      "eval_num_tokens": 197354.0,
+      "eval_runtime": 100.5983,
+      "eval_samples_per_second": 14.116,
+      "eval_steps_per_second": 1.769,
+      "step": 80
+    },
+    {
+      "entropy": 0.7309209659695626,
+      "epoch": 0.24125452352231605,
+      "grad_norm": 0.48046866059303284,
+      "learning_rate": 0.00010314163729865769,
+      "loss": 0.680583143234253,
+      "mean_token_accuracy": 0.8073364540934562,
+      "num_tokens": 248557.0,
+      "step": 100
+    },
+    {
+      "epoch": 0.24125452352231605,
+      "eval_entropy": 0.7152130687504672,
+      "eval_loss": 0.6989707350730896,
+      "eval_mean_token_accuracy": 0.8057264320636064,
+      "eval_num_tokens": 248557.0,
+      "eval_runtime": 100.6264,
+      "eval_samples_per_second": 14.112,
+      "eval_steps_per_second": 1.769,
+      "step": 100
+    },
+    {
+      "entropy": 0.7105271726846695,
+      "epoch": 0.28950542822677927,
+      "grad_norm": 0.5512468814849854,
+      "learning_rate": 0.00012397833170242693,
+      "loss": 0.6674309253692627,
+      "mean_token_accuracy": 0.8098619893193245,
+      "num_tokens": 298041.0,
+      "step": 120
+    },
+    {
+      "epoch": 0.28950542822677927,
+      "eval_entropy": 0.7314498555124476,
+      "eval_loss": 0.6796280145645142,
+      "eval_mean_token_accuracy": 0.809063873264227,
+      "eval_num_tokens": 298041.0,
+      "eval_runtime": 100.6132,
+      "eval_samples_per_second": 14.113,
+      "eval_steps_per_second": 1.769,
+      "step": 120
+    },
+    {
+      "entropy": 0.7079856500029564,
+      "epoch": 0.33775633293124246,
+      "grad_norm": 0.5903022885322571,
+      "learning_rate": 0.00014481502610619613,
+      "loss": 0.6623000144958496,
+      "mean_token_accuracy": 0.8101129323244095,
+      "num_tokens": 348955.0,
+      "step": 140
+    },
+    {
+      "epoch": 0.33775633293124246,
+      "eval_entropy": 0.7005330203624254,
+      "eval_loss": 0.671779453754425,
+      "eval_mean_token_accuracy": 0.8106091374091888,
+      "eval_num_tokens": 348955.0,
+      "eval_runtime": 100.6904,
+      "eval_samples_per_second": 14.103,
+      "eval_steps_per_second": 1.768,
+      "step": 140
+    },
+    {
+      "entropy": 0.7109143257141113,
+      "epoch": 0.38600723763570566,
+      "grad_norm": 0.4516378343105316,
+      "learning_rate": 0.00016565172050996536,
+      "loss": 0.658915901184082,
+      "mean_token_accuracy": 0.8112782314419746,
+      "num_tokens": 395798.0,
+      "step": 160
+    },
+    {
+      "epoch": 0.38600723763570566,
+      "eval_entropy": 0.6891461650976974,
+      "eval_loss": 0.6598312258720398,
+      "eval_mean_token_accuracy": 0.8133360951134329,
+      "eval_num_tokens": 395798.0,
+      "eval_runtime": 100.5057,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 160
+    },
+    {
+      "entropy": 0.6896152347326279,
+      "epoch": 0.43425814234016885,
+      "grad_norm": 0.49479565024375916,
+      "learning_rate": 0.0001864884149137346,
+      "loss": 0.6466911315917969,
+      "mean_token_accuracy": 0.8141682580113411,
+      "num_tokens": 446437.0,
+      "step": 180
+    },
+    {
+      "epoch": 0.43425814234016885,
+      "eval_entropy": 0.6984815888860253,
+      "eval_loss": 0.652909517288208,
+      "eval_mean_token_accuracy": 0.8133517600847094,
+      "eval_num_tokens": 446437.0,
+      "eval_runtime": 100.5867,
+      "eval_samples_per_second": 14.117,
+      "eval_steps_per_second": 1.77,
+      "step": 180
+    },
+    {
+      "entropy": 0.6829591259360314,
+      "epoch": 0.4825090470446321,
+      "grad_norm": 0.4835337698459625,
+      "learning_rate": 0.00020732510931750382,
+      "loss": 0.6401876449584961,
+      "mean_token_accuracy": 0.8175502926111221,
+      "num_tokens": 497682.0,
+      "step": 200
+    },
+    {
+      "epoch": 0.4825090470446321,
+      "eval_entropy": 0.6963563973314306,
+      "eval_loss": 0.650087296962738,
+      "eval_mean_token_accuracy": 0.8159684401549651,
+      "eval_num_tokens": 497682.0,
+      "eval_runtime": 100.6277,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 200
+    },
+    {
+      "entropy": 0.6903425320982933,
+      "epoch": 0.5307599517490953,
+      "grad_norm": 0.44980454444885254,
+      "learning_rate": 0.00022816180372127305,
+      "loss": 0.6409444808959961,
+      "mean_token_accuracy": 0.8163257524371147,
+      "num_tokens": 544670.0,
+      "step": 220
+    },
+    {
+      "epoch": 0.5307599517490953,
+      "eval_entropy": 0.6687790182199371,
+      "eval_loss": 0.6451645493507385,
+      "eval_mean_token_accuracy": 0.8157875551266617,
+      "eval_num_tokens": 544670.0,
+      "eval_runtime": 100.5444,
+      "eval_samples_per_second": 14.123,
+      "eval_steps_per_second": 1.77,
+      "step": 220
+    },
+    {
+      "entropy": 0.6740772560238838,
+      "epoch": 0.5790108564535585,
+      "grad_norm": 0.5312877297401428,
+      "learning_rate": 0.0002489984981250423,
+      "loss": 0.6307676315307618,
+      "mean_token_accuracy": 0.8174169659614563,
+      "num_tokens": 592926.0,
+      "step": 240
+    },
+    {
+      "epoch": 0.5790108564535585,
+      "eval_entropy": 0.7063750872451268,
+      "eval_loss": 0.6494986414909363,
+      "eval_mean_token_accuracy": 0.8156777198395032,
+      "eval_num_tokens": 592926.0,
+      "eval_runtime": 100.5094,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 240
+    },
+    {
+      "entropy": 0.6876470118761062,
+      "epoch": 0.6272617611580217,
+      "grad_norm": 0.46149370074272156,
+      "learning_rate": 0.00026983519252881153,
+      "loss": 0.6392203807830811,
+      "mean_token_accuracy": 0.8172321930527687,
+      "num_tokens": 639077.0,
+      "step": 260
+    },
+    {
+      "epoch": 0.6272617611580217,
+      "eval_entropy": 0.625594850838854,
+      "eval_loss": 0.6463819742202759,
+      "eval_mean_token_accuracy": 0.8179670525400826,
+      "eval_num_tokens": 639077.0,
+      "eval_runtime": 100.5775,
+      "eval_samples_per_second": 14.118,
+      "eval_steps_per_second": 1.77,
+      "step": 260
+    },
+    {
+      "entropy": 0.6673879995942116,
+      "epoch": 0.6755126658624849,
+      "grad_norm": 0.42477288842201233,
+      "learning_rate": 0.0002906718869325807,
+      "loss": 0.6300291538238525,
+      "mean_token_accuracy": 0.8203139930963517,
+      "num_tokens": 691028.0,
+      "step": 280
+    },
+    {
+      "epoch": 0.6755126658624849,
+      "eval_entropy": 0.6863534494062488,
+      "eval_loss": 0.6392037272453308,
+      "eval_mean_token_accuracy": 0.8190520385677895,
+      "eval_num_tokens": 691028.0,
+      "eval_runtime": 100.5106,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 280
+    },
+    {
+      "entropy": 0.6573980256915093,
+      "epoch": 0.7237635705669482,
+      "grad_norm": 0.4656430780887604,
+      "learning_rate": 0.00031150858133635,
+      "loss": 0.6215964794158936,
+      "mean_token_accuracy": 0.8208559453487396,
+      "num_tokens": 740255.0,
+      "step": 300
+    },
+    {
+      "epoch": 0.7237635705669482,
+      "eval_entropy": 0.6834242812703165,
+      "eval_loss": 0.6479960083961487,
+      "eval_mean_token_accuracy": 0.8128752906001016,
+      "eval_num_tokens": 740255.0,
+      "eval_runtime": 100.6101,
+      "eval_samples_per_second": 14.114,
+      "eval_steps_per_second": 1.769,
+      "step": 300
+    },
+    {
+      "entropy": 0.6751310348510742,
+      "epoch": 0.7720144752714113,
+      "grad_norm": 0.4495522081851959,
+      "learning_rate": 0.00033234527574011916,
+      "loss": 0.6402668952941895,
+      "mean_token_accuracy": 0.8170571058988572,
+      "num_tokens": 789529.0,
+      "step": 320
+    },
+    {
+      "epoch": 0.7720144752714113,
+      "eval_entropy": 0.6920045178927733,
+      "eval_loss": 0.6402787566184998,
+      "eval_mean_token_accuracy": 0.817791668886549,
+      "eval_num_tokens": 789529.0,
+      "eval_runtime": 100.6347,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 320
+    },
+    {
+      "entropy": 0.6846290081739426,
+      "epoch": 0.8202653799758746,
+      "grad_norm": 0.5661432147026062,
+      "learning_rate": 0.00035318197014388845,
+      "loss": 0.6508871078491211,
+      "mean_token_accuracy": 0.8164169207215309,
+      "num_tokens": 841584.0,
+      "step": 340
+    },
+    {
+      "epoch": 0.8202653799758746,
+      "eval_entropy": 0.6772527078564248,
+      "eval_loss": 0.6512064337730408,
+      "eval_mean_token_accuracy": 0.8170605137776793,
+      "eval_num_tokens": 841584.0,
+      "eval_runtime": 100.5157,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 340
+    },
+    {
+      "entropy": 0.7022885024547577,
+      "epoch": 0.8685162846803377,
+      "grad_norm": 0.6357236504554749,
+      "learning_rate": 0.0003740186645476577,
+      "loss": 0.6649135112762451,
+      "mean_token_accuracy": 0.8136772960424423,
+      "num_tokens": 885769.0,
+      "step": 360
+    },
+    {
+      "epoch": 0.8685162846803377,
+      "eval_entropy": 0.7435127543599418,
+      "eval_loss": 0.6573262214660645,
+      "eval_mean_token_accuracy": 0.8147802225659403,
+      "eval_num_tokens": 885769.0,
+      "eval_runtime": 100.5271,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 360
+    },
+    {
+      "entropy": 0.6962979376316071,
+      "epoch": 0.916767189384801,
+      "grad_norm": 0.6179454326629639,
+      "learning_rate": 0.00039485535895142685,
+      "loss": 0.6509191513061523,
+      "mean_token_accuracy": 0.81364226937294,
+      "num_tokens": 930056.0,
+      "step": 380
+    },
+    {
+      "epoch": 0.916767189384801,
+      "eval_entropy": 0.6630894883295123,
+      "eval_loss": 0.6524088978767395,
+      "eval_mean_token_accuracy": 0.8165898587596551,
+      "eval_num_tokens": 930056.0,
+      "eval_runtime": 100.5343,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 380
+    },
+    {
+      "entropy": 0.694319449365139,
+      "epoch": 0.9650180940892642,
+      "grad_norm": 0.5458664894104004,
+      "learning_rate": 0.00041569205335519613,
+      "loss": 0.6624011993408203,
+      "mean_token_accuracy": 0.8136080905795098,
+      "num_tokens": 976892.0,
+      "step": 400
+    },
+    {
+      "epoch": 0.9650180940892642,
+      "eval_entropy": 0.7424925324622165,
+      "eval_loss": 0.6717661023139954,
+      "eval_mean_token_accuracy": 0.8099894781460922,
+      "eval_num_tokens": 976892.0,
+      "eval_runtime": 100.5992,
+      "eval_samples_per_second": 14.115,
+      "eval_steps_per_second": 1.769,
+      "step": 400
+    },
+    {
+      "entropy": 0.6861714736009256,
+      "epoch": 1.0120627261761157,
+      "grad_norm": 0.6108037829399109,
+      "learning_rate": 0.0004323601853188671,
+      "loss": 0.6551628112792969,
+      "mean_token_accuracy": 0.8134882618219424,
+      "num_tokens": 1024472.0,
+      "step": 420
+    },
+    {
+      "epoch": 1.0120627261761157,
+      "eval_entropy": 0.6593745705786715,
+      "eval_loss": 0.6720747947692871,
+      "eval_mean_token_accuracy": 0.812802647941568,
+      "eval_num_tokens": 1024472.0,
+      "eval_runtime": 100.4866,
+      "eval_samples_per_second": 14.131,
+      "eval_steps_per_second": 1.771,
+      "step": 420
+    },
+    {
+      "entropy": 0.639520151913166,
+      "epoch": 1.060313630880579,
+      "grad_norm": 0.6054353713989258,
+      "learning_rate": 0.00043231736219609266,
+      "loss": 0.6096667766571044,
+      "mean_token_accuracy": 0.8227144703269005,
+      "num_tokens": 1076964.0,
+      "step": 440
+    },
+    {
+      "epoch": 1.060313630880579,
+      "eval_entropy": 0.6316307876886946,
+      "eval_loss": 0.6621950268745422,
+      "eval_mean_token_accuracy": 0.8140057914712456,
+      "eval_num_tokens": 1076964.0,
+      "eval_runtime": 100.6409,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 440
+    },
+    {
+      "entropy": 0.6544070363044738,
+      "epoch": 1.1085645355850422,
+      "grad_norm": 0.5111057758331299,
+      "learning_rate": 0.0004322133749558233,
+      "loss": 0.6176822185516357,
+      "mean_token_accuracy": 0.8201584875583648,
+      "num_tokens": 1126064.0,
+      "step": 460
+    },
+    {
+      "epoch": 1.1085645355850422,
+      "eval_entropy": 0.6714627766207363,
+      "eval_loss": 0.6738516092300415,
+      "eval_mean_token_accuracy": 0.8128616829936424,
+      "eval_num_tokens": 1126064.0,
+      "eval_runtime": 100.4986,
+      "eval_samples_per_second": 14.13,
+      "eval_steps_per_second": 1.771,
+      "step": 460
+    },
+    {
+      "entropy": 0.6470697179436684,
+      "epoch": 1.1568154402895054,
+      "grad_norm": 0.5434669256210327,
+      "learning_rate": 0.0004320482530252179,
+      "loss": 0.6144762516021729,
+      "mean_token_accuracy": 0.819926056265831,
+      "num_tokens": 1176912.0,
+      "step": 480
+    },
+    {
+      "epoch": 1.1568154402895054,
+      "eval_entropy": 0.6475969080844622,
+      "eval_loss": 0.6624214053153992,
+      "eval_mean_token_accuracy": 0.8137228820431098,
+      "eval_num_tokens": 1176912.0,
+      "eval_runtime": 100.5617,
+      "eval_samples_per_second": 14.121,
+      "eval_steps_per_second": 1.77,
+      "step": 480
+    },
+    {
+      "entropy": 0.6659130036830903,
+      "epoch": 1.2050663449939687,
+      "grad_norm": 0.6470707058906555,
+      "learning_rate": 0.0004318220431318296,
+      "loss": 0.6366308212280274,
+      "mean_token_accuracy": 0.8175261527299881,
+      "num_tokens": 1223383.0,
+      "step": 500
+    },
+    {
+      "epoch": 1.2050663449939687,
+      "eval_entropy": 0.7232425132494295,
+      "eval_loss": 0.6877738833427429,
+      "eval_mean_token_accuracy": 0.8007767260744331,
+      "eval_num_tokens": 1223383.0,
+      "eval_runtime": 100.477,
+      "eval_samples_per_second": 14.133,
+      "eval_steps_per_second": 1.772,
+      "step": 500
+    },
+    {
+      "entropy": 0.6759189993143082,
+      "epoch": 1.2533172496984317,
+      "grad_norm": 0.6453561186790466,
+      "learning_rate": 0.00043153480929038245,
+      "loss": 0.637772512435913,
+      "mean_token_accuracy": 0.8162627041339874,
+      "num_tokens": 1273534.0,
+      "step": 520
+    },
+    {
+      "epoch": 1.2533172496984317,
+      "eval_entropy": 0.7082584312792575,
+      "eval_loss": 0.6695764064788818,
+      "eval_mean_token_accuracy": 0.812019992410467,
+      "eval_num_tokens": 1273534.0,
+      "eval_runtime": 100.4957,
+      "eval_samples_per_second": 14.13,
+      "eval_steps_per_second": 1.771,
+      "step": 520
+    },
+    {
+      "entropy": 0.6730152070522308,
+      "epoch": 1.301568154402895,
+      "grad_norm": 0.6234476566314697,
+      "learning_rate": 0.000431186632784656,
+      "loss": 0.640014362335205,
+      "mean_token_accuracy": 0.8168858170509339,
+      "num_tokens": 1323351.0,
+      "step": 540
+    },
+    {
+      "epoch": 1.301568154402895,
+      "eval_entropy": 0.6655137846978862,
+      "eval_loss": 0.6578248739242554,
+      "eval_mean_token_accuracy": 0.8147600979617472,
+      "eval_num_tokens": 1323351.0,
+      "eval_runtime": 100.503,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 540
+    },
+    {
+      "entropy": 0.6636043325066566,
+      "epoch": 1.3498190591073582,
+      "grad_norm": 0.5882512331008911,
+      "learning_rate": 0.0004307776121444827,
+      "loss": 0.6286775588989257,
+      "mean_token_accuracy": 0.8183174058794975,
+      "num_tokens": 1371508.0,
+      "step": 560
+    },
+    {
+      "epoch": 1.3498190591073582,
+      "eval_entropy": 0.6304813810948575,
+      "eval_loss": 0.6607436537742615,
+      "eval_mean_token_accuracy": 0.8169659268989992,
+      "eval_num_tokens": 1371508.0,
+      "eval_runtime": 100.6233,
+      "eval_samples_per_second": 14.112,
+      "eval_steps_per_second": 1.769,
+      "step": 560
+    },
+    {
+      "entropy": 0.6608738481998444,
+      "epoch": 1.3980699638118215,
+      "grad_norm": 0.6796398758888245,
+      "learning_rate": 0.00043030786311786564,
+      "loss": 0.622726583480835,
+      "mean_token_accuracy": 0.8230827242136002,
+      "num_tokens": 1415221.0,
+      "step": 580
+    },
+    {
+      "epoch": 1.3980699638118215,
+      "eval_entropy": 0.6645242689030894,
+      "eval_loss": 0.6662811636924744,
+      "eval_mean_token_accuracy": 0.8147623907314258,
+      "eval_num_tokens": 1415221.0,
+      "eval_runtime": 100.5074,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 580
+    },
+    {
+      "entropy": 0.6682895936071873,
+      "epoch": 1.4463208685162847,
+      "grad_norm": 0.5750718712806702,
+      "learning_rate": 0.0004297775186382226,
+      "loss": 0.6304276466369629,
+      "mean_token_accuracy": 0.8188625335693359,
+      "num_tokens": 1463356.0,
+      "step": 600
+    },
+    {
+      "epoch": 1.4463208685162847,
+      "eval_entropy": 0.6320532578765676,
+      "eval_loss": 0.650333046913147,
+      "eval_mean_token_accuracy": 0.8172853793321031,
+      "eval_num_tokens": 1463356.0,
+      "eval_runtime": 100.5116,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 600
+    },
+    {
+      "entropy": 0.653739221394062,
+      "epoch": 1.494571773220748,
+      "grad_norm": 0.5552534461021423,
+      "learning_rate": 0.00042918672878676804,
+      "loss": 0.6215689659118653,
+      "mean_token_accuracy": 0.8208896175026894,
+      "num_tokens": 1512340.0,
+      "step": 620
+    },
+    {
+      "epoch": 1.494571773220748,
+      "eval_entropy": 0.6285932062382109,
+      "eval_loss": 0.6392536163330078,
+      "eval_mean_token_accuracy": 0.8205167825302381,
+      "eval_num_tokens": 1512340.0,
+      "eval_runtime": 100.6392,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 620
+    },
+    {
+      "entropy": 0.6476897612214089,
+      "epoch": 1.5428226779252112,
+      "grad_norm": 0.536978542804718,
+      "learning_rate": 0.00042853566075004156,
+      "loss": 0.6131608009338378,
+      "mean_token_accuracy": 0.8208152890205384,
+      "num_tokens": 1560911.0,
+      "step": 640
+    },
+    {
+      "epoch": 1.5428226779252112,
+      "eval_entropy": 0.6626498856571283,
+      "eval_loss": 0.6479025483131409,
+      "eval_mean_token_accuracy": 0.8180894101603647,
+      "eval_num_tokens": 1560911.0,
+      "eval_runtime": 100.5035,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 640
+    },
+    {
+      "entropy": 0.6338633447885513,
+      "epoch": 1.5910735826296745,
+      "grad_norm": 0.4518965780735016,
+      "learning_rate": 0.0004278244987725961,
+      "loss": 0.6010610580444335,
+      "mean_token_accuracy": 0.8247816175222397,
+      "num_tokens": 1609409.0,
+      "step": 660
+    },
+    {
+      "epoch": 1.5910735826296745,
+      "eval_entropy": 0.64079723699709,
+      "eval_loss": 0.6417157053947449,
+      "eval_mean_token_accuracy": 0.8214994581897607,
+      "eval_num_tokens": 1609409.0,
+      "eval_runtime": 100.6393,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 660
+    },
+    {
+      "entropy": 0.6485912457108498,
+      "epoch": 1.6393244873341375,
+      "grad_norm": 0.6028368473052979,
+      "learning_rate": 0.00042705344410485905,
+      "loss": 0.6120437145233154,
+      "mean_token_accuracy": 0.8236212745308876,
+      "num_tokens": 1657430.0,
+      "step": 680
+    },
+    {
+      "epoch": 1.6393244873341375,
+      "eval_entropy": 0.6863006427046958,
+      "eval_loss": 0.6420931816101074,
+      "eval_mean_token_accuracy": 0.8191533580924688,
+      "eval_num_tokens": 1657430.0,
+      "eval_runtime": 100.4639,
+      "eval_samples_per_second": 14.134,
+      "eval_steps_per_second": 1.772,
+      "step": 680
+    },
+    {
+      "entropy": 0.6459382772445679,
+      "epoch": 1.6875753920386007,
+      "grad_norm": 0.5069310069084167,
+      "learning_rate": 0.00042622271494618073,
+      "loss": 0.6160889625549316,
+      "mean_token_accuracy": 0.8253828629851341,
+      "num_tokens": 1708058.0,
+      "step": 700
+    },
+    {
+      "epoch": 1.6875753920386007,
+      "eval_entropy": 0.6773919463157654,
+      "eval_loss": 0.6307896375656128,
+      "eval_mean_token_accuracy": 0.8217989825800563,
+      "eval_num_tokens": 1708058.0,
+      "eval_runtime": 100.6301,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 700
+    },
+    {
+      "entropy": 0.64687028080225,
+      "epoch": 1.7358262967430638,
+      "grad_norm": 0.3830573856830597,
+      "learning_rate": 0.00042533254638308645,
+      "loss": 0.6104001998901367,
+      "mean_token_accuracy": 0.8255606815218925,
+      "num_tokens": 1761262.0,
+      "step": 720
+    },
+    {
+      "epoch": 1.7358262967430638,
+      "eval_entropy": 0.6179541145147902,
+      "eval_loss": 0.6170215606689453,
+      "eval_mean_token_accuracy": 0.8259220900160543,
+      "eval_num_tokens": 1761262.0,
+      "eval_runtime": 100.61,
+      "eval_samples_per_second": 14.114,
+      "eval_steps_per_second": 1.769,
+      "step": 720
+    },
+    {
+      "entropy": 0.6306161753833294,
+      "epoch": 1.784077201447527,
+      "grad_norm": 0.5888957977294922,
+      "learning_rate": 0.0004243831903227498,
+      "loss": 0.5981546878814697,
+      "mean_token_accuracy": 0.8266776099801063,
+      "num_tokens": 1812127.0,
+      "step": 740
+    },
+    {
+      "epoch": 1.784077201447527,
+      "eval_entropy": 0.6357416808605194,
+      "eval_loss": 0.6202242374420166,
+      "eval_mean_token_accuracy": 0.8253380304641937,
+      "eval_num_tokens": 1812127.0,
+      "eval_runtime": 100.5145,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 740
+    },
+    {
+      "entropy": 0.6290688961744308,
+      "epoch": 1.8323281061519903,
+      "grad_norm": 0.6138001680374146,
+      "learning_rate": 0.0004233749154217063,
+      "loss": 0.5853167057037354,
+      "mean_token_accuracy": 0.8281400203704834,
+      "num_tokens": 1858403.0,
+      "step": 760
+    },
+    {
+      "epoch": 1.8323281061519903,
+      "eval_entropy": 0.6402715869833914,
+      "eval_loss": 0.6168262958526611,
+      "eval_mean_token_accuracy": 0.8246346306934785,
+      "eval_num_tokens": 1858403.0,
+      "eval_runtime": 100.5237,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 760
+    },
+    {
+      "entropy": 0.6218615680932998,
+      "epoch": 1.8805790108564535,
+      "grad_norm": 0.574312686920166,
+      "learning_rate": 0.00042230800700982586,
+      "loss": 0.5841915607452393,
+      "mean_token_accuracy": 0.8304707780480385,
+      "num_tokens": 1907361.0,
+      "step": 780
+    },
+    {
+      "epoch": 1.8805790108564535,
+      "eval_entropy": 0.6301163148679091,
+      "eval_loss": 0.6073815226554871,
+      "eval_mean_token_accuracy": 0.8274695993809218,
+      "eval_num_tokens": 1907361.0,
+      "eval_runtime": 100.5111,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 780
+    },
+    {
+      "entropy": 0.6101033873856068,
+      "epoch": 1.9288299155609168,
+      "grad_norm": 0.5054804682731628,
+      "learning_rate": 0.000421182767009569,
+      "loss": 0.5774452209472656,
+      "mean_token_accuracy": 0.8321943372488022,
+      "num_tokens": 1956605.0,
+      "step": 800
+    },
+    {
+      "epoch": 1.9288299155609168,
+      "eval_entropy": 0.631400960215022,
+      "eval_loss": 0.6115217804908752,
+      "eval_mean_token_accuracy": 0.8249618762664581,
+      "eval_num_tokens": 1956605.0,
+      "eval_runtime": 100.4916,
+      "eval_samples_per_second": 14.131,
+      "eval_steps_per_second": 1.771,
+      "step": 800
+    },
+    {
+      "entropy": 0.6217561915516854,
+      "epoch": 1.97708082026538,
+      "grad_norm": 0.5333649516105652,
+      "learning_rate": 0.0004199995138505454,
+      "loss": 0.5842787265777588,
+      "mean_token_accuracy": 0.8305525362491608,
+      "num_tokens": 2003308.0,
+      "step": 820
+    },
+    {
+      "epoch": 1.97708082026538,
+      "eval_entropy": 0.605928597657868,
+      "eval_loss": 0.6067516207695007,
+      "eval_mean_token_accuracy": 0.82715918408351,
+      "eval_num_tokens": 2003308.0,
+      "eval_runtime": 100.5286,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 820
+    },
+    {
+      "entropy": 0.5585076526189462,
+      "epoch": 2.0241254523522314,
+      "grad_norm": 0.44819608330726624,
+      "learning_rate": 0.00041875858237940244,
+      "loss": 0.5252367496490479,
+      "mean_token_accuracy": 0.8405626966403081,
+      "num_tokens": 2051331.0,
+      "step": 840
+    },
+    {
+      "epoch": 2.0241254523522314,
+      "eval_entropy": 0.6123150769579276,
+      "eval_loss": 0.6175946593284607,
+      "eval_mean_token_accuracy": 0.827637837509091,
+      "eval_num_tokens": 2051331.0,
+      "eval_runtime": 100.5414,
+      "eval_samples_per_second": 14.124,
+      "eval_steps_per_second": 1.77,
+      "step": 840
+    },
+    {
+      "entropy": 0.5359115816652775,
+      "epoch": 2.0723763570566947,
+      "grad_norm": 0.576781690120697,
+      "learning_rate": 0.0004174603237650678,
+      "loss": 0.5023207664489746,
+      "mean_token_accuracy": 0.8459693863987923,
+      "num_tokens": 2101528.0,
+      "step": 860
+    },
+    {
+      "epoch": 2.0723763570566947,
+      "eval_entropy": 0.5592134784111816,
+      "eval_loss": 0.620693564414978,
+      "eval_mean_token_accuracy": 0.827402507656076,
+      "eval_num_tokens": 2101528.0,
+      "eval_runtime": 100.6274,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 860
+    },
+    {
+      "entropy": 0.5214565828442573,
+      "epoch": 2.120627261761158,
+      "grad_norm": 0.45878875255584717,
+      "learning_rate": 0.00041610510539937234,
+      "loss": 0.48367843627929685,
+      "mean_token_accuracy": 0.8515541926026344,
+      "num_tokens": 2149550.0,
+      "step": 880
+    },
+    {
+      "epoch": 2.120627261761158,
+      "eval_entropy": 0.5860532344727034,
+      "eval_loss": 0.6134791970252991,
+      "eval_mean_token_accuracy": 0.8291774777214179,
+      "eval_num_tokens": 2149550.0,
+      "eval_runtime": 100.5295,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 880
+    },
+    {
+      "entropy": 0.5414379432797432,
+      "epoch": 2.168878166465621,
+      "grad_norm": 0.5976294875144958,
+      "learning_rate": 0.0004146933107930829,
+      "loss": 0.5012908935546875,
+      "mean_token_accuracy": 0.8466357633471489,
+      "num_tokens": 2202162.0,
+      "step": 900
+    },
+    {
+      "epoch": 2.168878166465621,
+      "eval_entropy": 0.5427783280611038,
+      "eval_loss": 0.612716019153595,
+      "eval_mean_token_accuracy": 0.829505049445656,
+      "eval_num_tokens": 2202162.0,
+      "eval_runtime": 100.5207,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 900
+    },
+    {
+      "entropy": 0.521108990907669,
+      "epoch": 2.2171290711700844,
+      "grad_norm": 0.5145775675773621,
+      "learning_rate": 0.000413225339467373,
+      "loss": 0.4877732276916504,
+      "mean_token_accuracy": 0.8504189744591713,
+      "num_tokens": 2256539.0,
+      "step": 920
+    },
+    {
+      "epoch": 2.2171290711700844,
+      "eval_entropy": 0.5620684869838565,
+      "eval_loss": 0.6176337599754333,
+      "eval_mean_token_accuracy": 0.8291444483767735,
+      "eval_num_tokens": 2256539.0,
+      "eval_runtime": 100.6819,
+      "eval_samples_per_second": 14.104,
+      "eval_steps_per_second": 1.768,
+      "step": 920
+    },
+    {
+      "entropy": 0.5452766507863999,
+      "epoch": 2.2653799758745476,
+      "grad_norm": 0.4820452928543091,
+      "learning_rate": 0.0004117016068407636,
+      "loss": 0.4991307735443115,
+      "mean_token_accuracy": 0.8474363312125206,
+      "num_tokens": 2299698.0,
+      "step": 940
+    },
+    {
+      "epoch": 2.2653799758745476,
+      "eval_entropy": 0.5611683193887218,
+      "eval_loss": 0.6153580546379089,
+      "eval_mean_token_accuracy": 0.8286041272490212,
+      "eval_num_tokens": 2299698.0,
+      "eval_runtime": 100.5249,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 940
+    },
+    {
+      "entropy": 0.5572576269507408,
+      "epoch": 2.313630880579011,
+      "grad_norm": 0.3947993814945221,
+      "learning_rate": 0.00041012254411156387,
+      "loss": 0.5191003799438476,
+      "mean_token_accuracy": 0.8422838419675827,
+      "num_tokens": 2347134.0,
+      "step": 960
+    },
+    {
+      "epoch": 2.313630880579011,
+      "eval_entropy": 0.5651392345682958,
+      "eval_loss": 0.6121005415916443,
+      "eval_mean_token_accuracy": 0.8286536242185014,
+      "eval_num_tokens": 2347134.0,
+      "eval_runtime": 100.7033,
+      "eval_samples_per_second": 14.101,
+      "eval_steps_per_second": 1.768,
+      "step": 960
+    },
+    {
+      "entropy": 0.5441671140491963,
+      "epoch": 2.361881785283474,
+      "grad_norm": 0.7719791531562805,
+      "learning_rate": 0.0004084885981358477,
+      "loss": 0.5072606086730957,
+      "mean_token_accuracy": 0.8478495761752128,
+      "num_tokens": 2394568.0,
+      "step": 980
+    },
+    {
+      "epoch": 2.361881785283474,
+      "eval_entropy": 0.5651004270891126,
+      "eval_loss": 0.6069501042366028,
+      "eval_mean_token_accuracy": 0.8311042156112328,
+      "eval_num_tokens": 2394568.0,
+      "eval_runtime": 100.5487,
+      "eval_samples_per_second": 14.123,
+      "eval_steps_per_second": 1.77,
+      "step": 980
+    },
+    {
+      "entropy": 0.5462074346840382,
+      "epoch": 2.4101326899879374,
+      "grad_norm": 0.5694436430931091,
+      "learning_rate": 0.00040680023130099856,
+      "loss": 0.5097150325775146,
+      "mean_token_accuracy": 0.8474388793110847,
+      "num_tokens": 2443253.0,
+      "step": 1000
+    },
+    {
+      "epoch": 2.4101326899879374,
+      "eval_entropy": 0.5895893243926295,
+      "eval_loss": 0.6078168749809265,
+      "eval_mean_token_accuracy": 0.828030709470256,
+      "eval_num_tokens": 2443253.0,
+      "eval_runtime": 100.5413,
+      "eval_samples_per_second": 14.124,
+      "eval_steps_per_second": 1.77,
+      "step": 1000
+    },
+    {
+      "entropy": 0.5473216332495212,
+      "epoch": 2.4583835946924006,
+      "grad_norm": 0.3989660143852234,
+      "learning_rate": 0.00040505792139485913,
+      "loss": 0.5087806224822998,
+      "mean_token_accuracy": 0.8473173275589942,
+      "num_tokens": 2495178.0,
+      "step": 1020
+    },
+    {
+      "epoch": 2.4583835946924006,
+      "eval_entropy": 0.5588909702354603,
+      "eval_loss": 0.6087605953216553,
+      "eval_mean_token_accuracy": 0.8309365524334854,
+      "eval_num_tokens": 2495178.0,
+      "eval_runtime": 100.5137,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 1020
+    },
+    {
+      "entropy": 0.5430163390934467,
+      "epoch": 2.5066344993968634,
+      "grad_norm": 0.4916608929634094,
+      "learning_rate": 0.00040326216147052336,
+      "loss": 0.5027707576751709,
+      "mean_token_accuracy": 0.8466959804296493,
+      "num_tokens": 2541733.0,
+      "step": 1040
+    },
+    {
+      "epoch": 2.5066344993968634,
+      "eval_entropy": 0.5819033988741007,
+      "eval_loss": 0.609398365020752,
+      "eval_mean_token_accuracy": 0.8297666612635838,
+      "eval_num_tokens": 2541733.0,
+      "eval_runtime": 100.6375,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 1040
+    },
+    {
+      "entropy": 0.5632683664560318,
+      "epoch": 2.554885404101327,
+      "grad_norm": 0.6062753796577454,
+      "learning_rate": 0.00040141345970680814,
+      "loss": 0.5272740840911865,
+      "mean_token_accuracy": 0.8411559104919434,
+      "num_tokens": 2589479.0,
+      "step": 1060
+    },
+    {
+      "epoch": 2.554885404101327,
+      "eval_entropy": 0.5785442567273472,
+      "eval_loss": 0.6207635998725891,
+      "eval_mean_token_accuracy": 0.8259770478425401,
+      "eval_num_tokens": 2589479.0,
+      "eval_runtime": 100.5161,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 1060
+    },
+    {
+      "entropy": 0.5591374494135379,
+      "epoch": 2.60313630880579,
+      "grad_norm": 0.5771989226341248,
+      "learning_rate": 0.00039951233926444437,
+      "loss": 0.5237057685852051,
+      "mean_token_accuracy": 0.8413331478834152,
+      "num_tokens": 2639134.0,
+      "step": 1080
+    },
+    {
+      "epoch": 2.60313630880579,
+      "eval_entropy": 0.5765141543377651,
+      "eval_loss": 0.6119762063026428,
+      "eval_mean_token_accuracy": 0.8289416998289945,
+      "eval_num_tokens": 2639134.0,
+      "eval_runtime": 100.5288,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 1080
+    },
+    {
+      "entropy": 0.5470930114388466,
+      "epoch": 2.651387213510253,
+      "grad_norm": 0.4211709499359131,
+      "learning_rate": 0.00039755933813802937,
+      "loss": 0.5157970428466797,
+      "mean_token_accuracy": 0.8436284482479095,
+      "num_tokens": 2689324.0,
+      "step": 1100
+    },
+    {
+      "epoch": 2.651387213510253,
+      "eval_entropy": 0.5593592817528864,
+      "eval_loss": 0.6080028414726257,
+      "eval_mean_token_accuracy": 0.8277805718813049,
+      "eval_num_tokens": 2689324.0,
+      "eval_runtime": 100.495,
+      "eval_samples_per_second": 14.13,
+      "eval_steps_per_second": 1.771,
+      "step": 1100
+    },
+    {
+      "entropy": 0.5636695981025696,
+      "epoch": 2.6996381182147164,
+      "grad_norm": 0.6020221710205078,
+      "learning_rate": 0.00039555500900377994,
+      "loss": 0.5277415752410889,
+      "mean_token_accuracy": 0.840576496720314,
+      "num_tokens": 2736182.0,
+      "step": 1120
+    },
+    {
+      "epoch": 2.6996381182147164,
+      "eval_entropy": 0.5837533748551701,
+      "eval_loss": 0.6048349142074585,
+      "eval_mean_token_accuracy": 0.8288552148288555,
+      "eval_num_tokens": 2736182.0,
+      "eval_runtime": 100.5342,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 1120
+    }
+  ],
+  "logging_steps": 20,
+  "max_steps": 4150,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 20,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.68034904428544e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1140/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen3-14B-Base
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen3-14B-Base
+- lora
+- sft
+- transformers
+- trl
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1140/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-14B-Base",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 128,
+  "lora_bias": false,
+  "lora_dropout": 0.0539675908752296,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "gate_proj",
+    "q_proj",
+    "down_proj",
+    "o_proj",
+    "k_proj",
+    "v_proj",
+    "up_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1140/chat_template.jinja

@@ -0,0 +1,85 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1140/tokenizer_config.json

@@ -0,0 +1,29 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

overgeneralisation_original_Estonian/Qwen3-14B-Base_overgeneralisation_splits_original_features_train_overgeneralisation_splits_original_features_test1/checkpoint-1140/trainer_state.json

@@ -0,0 +1,1231 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.7478890229191797,
+  "eval_steps": 20,
+  "global_step": 1140,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 1.8160341799259185,
+      "epoch": 0.04825090470446321,
+      "grad_norm": 1.1313843727111816,
+      "learning_rate": 1.9794859683580766e-05,
+      "loss": 1.7627403259277343,
+      "mean_token_accuracy": 0.6354069545865059,
+      "num_tokens": 50237.0,
+      "step": 20
+    },
+    {
+      "epoch": 0.04825090470446321,
+      "eval_entropy": 1.538598193881217,
+      "eval_loss": 1.4881398677825928,
+      "eval_mean_token_accuracy": 0.686459881201219,
+      "eval_num_tokens": 50237.0,
+      "eval_runtime": 100.8923,
+      "eval_samples_per_second": 14.074,
+      "eval_steps_per_second": 1.764,
+      "step": 20
+    },
+    {
+      "entropy": 1.1673978716135025,
+      "epoch": 0.09650180940892641,
+      "grad_norm": 0.6167025566101074,
+      "learning_rate": 4.0631554087349995e-05,
+      "loss": 1.0887075424194337,
+      "mean_token_accuracy": 0.737513878941536,
+      "num_tokens": 101098.0,
+      "step": 40
+    },
+    {
+      "epoch": 0.09650180940892641,
+      "eval_entropy": 0.887577452351538,
+      "eval_loss": 0.8519237041473389,
+      "eval_mean_token_accuracy": 0.7741017204322173,
+      "eval_num_tokens": 101098.0,
+      "eval_runtime": 100.4819,
+      "eval_samples_per_second": 14.132,
+      "eval_steps_per_second": 1.771,
+      "step": 40
+    },
+    {
+      "entropy": 0.8220535531640053,
+      "epoch": 0.14475271411338964,
+      "grad_norm": 0.8396430611610413,
+      "learning_rate": 6.146824849111923e-05,
+      "loss": 0.7812146186828614,
+      "mean_token_accuracy": 0.7872247830033302,
+      "num_tokens": 150385.0,
+      "step": 60
+    },
+    {
+      "epoch": 0.14475271411338964,
+      "eval_entropy": 0.770665143982748,
+      "eval_loss": 0.7478981614112854,
+      "eval_mean_token_accuracy": 0.7949167068084974,
+      "eval_num_tokens": 150385.0,
+      "eval_runtime": 100.6518,
+      "eval_samples_per_second": 14.108,
+      "eval_steps_per_second": 1.768,
+      "step": 60
+    },
+    {
+      "entropy": 0.7776695728302002,
+      "epoch": 0.19300361881785283,
+      "grad_norm": 0.5731202363967896,
+      "learning_rate": 8.230494289488845e-05,
+      "loss": 0.732721471786499,
+      "mean_token_accuracy": 0.7976997002959252,
+      "num_tokens": 197354.0,
+      "step": 80
+    },
+    {
+      "epoch": 0.19300361881785283,
+      "eval_entropy": 0.7650937618834249,
+      "eval_loss": 0.7109046578407288,
+      "eval_mean_token_accuracy": 0.8018777065732506,
+      "eval_num_tokens": 197354.0,
+      "eval_runtime": 100.5983,
+      "eval_samples_per_second": 14.116,
+      "eval_steps_per_second": 1.769,
+      "step": 80
+    },
+    {
+      "entropy": 0.7309209659695626,
+      "epoch": 0.24125452352231605,
+      "grad_norm": 0.48046866059303284,
+      "learning_rate": 0.00010314163729865769,
+      "loss": 0.680583143234253,
+      "mean_token_accuracy": 0.8073364540934562,
+      "num_tokens": 248557.0,
+      "step": 100
+    },
+    {
+      "epoch": 0.24125452352231605,
+      "eval_entropy": 0.7152130687504672,
+      "eval_loss": 0.6989707350730896,
+      "eval_mean_token_accuracy": 0.8057264320636064,
+      "eval_num_tokens": 248557.0,
+      "eval_runtime": 100.6264,
+      "eval_samples_per_second": 14.112,
+      "eval_steps_per_second": 1.769,
+      "step": 100
+    },
+    {
+      "entropy": 0.7105271726846695,
+      "epoch": 0.28950542822677927,
+      "grad_norm": 0.5512468814849854,
+      "learning_rate": 0.00012397833170242693,
+      "loss": 0.6674309253692627,
+      "mean_token_accuracy": 0.8098619893193245,
+      "num_tokens": 298041.0,
+      "step": 120
+    },
+    {
+      "epoch": 0.28950542822677927,
+      "eval_entropy": 0.7314498555124476,
+      "eval_loss": 0.6796280145645142,
+      "eval_mean_token_accuracy": 0.809063873264227,
+      "eval_num_tokens": 298041.0,
+      "eval_runtime": 100.6132,
+      "eval_samples_per_second": 14.113,
+      "eval_steps_per_second": 1.769,
+      "step": 120
+    },
+    {
+      "entropy": 0.7079856500029564,
+      "epoch": 0.33775633293124246,
+      "grad_norm": 0.5903022885322571,
+      "learning_rate": 0.00014481502610619613,
+      "loss": 0.6623000144958496,
+      "mean_token_accuracy": 0.8101129323244095,
+      "num_tokens": 348955.0,
+      "step": 140
+    },
+    {
+      "epoch": 0.33775633293124246,
+      "eval_entropy": 0.7005330203624254,
+      "eval_loss": 0.671779453754425,
+      "eval_mean_token_accuracy": 0.8106091374091888,
+      "eval_num_tokens": 348955.0,
+      "eval_runtime": 100.6904,
+      "eval_samples_per_second": 14.103,
+      "eval_steps_per_second": 1.768,
+      "step": 140
+    },
+    {
+      "entropy": 0.7109143257141113,
+      "epoch": 0.38600723763570566,
+      "grad_norm": 0.4516378343105316,
+      "learning_rate": 0.00016565172050996536,
+      "loss": 0.658915901184082,
+      "mean_token_accuracy": 0.8112782314419746,
+      "num_tokens": 395798.0,
+      "step": 160
+    },
+    {
+      "epoch": 0.38600723763570566,
+      "eval_entropy": 0.6891461650976974,
+      "eval_loss": 0.6598312258720398,
+      "eval_mean_token_accuracy": 0.8133360951134329,
+      "eval_num_tokens": 395798.0,
+      "eval_runtime": 100.5057,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 160
+    },
+    {
+      "entropy": 0.6896152347326279,
+      "epoch": 0.43425814234016885,
+      "grad_norm": 0.49479565024375916,
+      "learning_rate": 0.0001864884149137346,
+      "loss": 0.6466911315917969,
+      "mean_token_accuracy": 0.8141682580113411,
+      "num_tokens": 446437.0,
+      "step": 180
+    },
+    {
+      "epoch": 0.43425814234016885,
+      "eval_entropy": 0.6984815888860253,
+      "eval_loss": 0.652909517288208,
+      "eval_mean_token_accuracy": 0.8133517600847094,
+      "eval_num_tokens": 446437.0,
+      "eval_runtime": 100.5867,
+      "eval_samples_per_second": 14.117,
+      "eval_steps_per_second": 1.77,
+      "step": 180
+    },
+    {
+      "entropy": 0.6829591259360314,
+      "epoch": 0.4825090470446321,
+      "grad_norm": 0.4835337698459625,
+      "learning_rate": 0.00020732510931750382,
+      "loss": 0.6401876449584961,
+      "mean_token_accuracy": 0.8175502926111221,
+      "num_tokens": 497682.0,
+      "step": 200
+    },
+    {
+      "epoch": 0.4825090470446321,
+      "eval_entropy": 0.6963563973314306,
+      "eval_loss": 0.650087296962738,
+      "eval_mean_token_accuracy": 0.8159684401549651,
+      "eval_num_tokens": 497682.0,
+      "eval_runtime": 100.6277,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 200
+    },
+    {
+      "entropy": 0.6903425320982933,
+      "epoch": 0.5307599517490953,
+      "grad_norm": 0.44980454444885254,
+      "learning_rate": 0.00022816180372127305,
+      "loss": 0.6409444808959961,
+      "mean_token_accuracy": 0.8163257524371147,
+      "num_tokens": 544670.0,
+      "step": 220
+    },
+    {
+      "epoch": 0.5307599517490953,
+      "eval_entropy": 0.6687790182199371,
+      "eval_loss": 0.6451645493507385,
+      "eval_mean_token_accuracy": 0.8157875551266617,
+      "eval_num_tokens": 544670.0,
+      "eval_runtime": 100.5444,
+      "eval_samples_per_second": 14.123,
+      "eval_steps_per_second": 1.77,
+      "step": 220
+    },
+    {
+      "entropy": 0.6740772560238838,
+      "epoch": 0.5790108564535585,
+      "grad_norm": 0.5312877297401428,
+      "learning_rate": 0.0002489984981250423,
+      "loss": 0.6307676315307618,
+      "mean_token_accuracy": 0.8174169659614563,
+      "num_tokens": 592926.0,
+      "step": 240
+    },
+    {
+      "epoch": 0.5790108564535585,
+      "eval_entropy": 0.7063750872451268,
+      "eval_loss": 0.6494986414909363,
+      "eval_mean_token_accuracy": 0.8156777198395032,
+      "eval_num_tokens": 592926.0,
+      "eval_runtime": 100.5094,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 240
+    },
+    {
+      "entropy": 0.6876470118761062,
+      "epoch": 0.6272617611580217,
+      "grad_norm": 0.46149370074272156,
+      "learning_rate": 0.00026983519252881153,
+      "loss": 0.6392203807830811,
+      "mean_token_accuracy": 0.8172321930527687,
+      "num_tokens": 639077.0,
+      "step": 260
+    },
+    {
+      "epoch": 0.6272617611580217,
+      "eval_entropy": 0.625594850838854,
+      "eval_loss": 0.6463819742202759,
+      "eval_mean_token_accuracy": 0.8179670525400826,
+      "eval_num_tokens": 639077.0,
+      "eval_runtime": 100.5775,
+      "eval_samples_per_second": 14.118,
+      "eval_steps_per_second": 1.77,
+      "step": 260
+    },
+    {
+      "entropy": 0.6673879995942116,
+      "epoch": 0.6755126658624849,
+      "grad_norm": 0.42477288842201233,
+      "learning_rate": 0.0002906718869325807,
+      "loss": 0.6300291538238525,
+      "mean_token_accuracy": 0.8203139930963517,
+      "num_tokens": 691028.0,
+      "step": 280
+    },
+    {
+      "epoch": 0.6755126658624849,
+      "eval_entropy": 0.6863534494062488,
+      "eval_loss": 0.6392037272453308,
+      "eval_mean_token_accuracy": 0.8190520385677895,
+      "eval_num_tokens": 691028.0,
+      "eval_runtime": 100.5106,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 280
+    },
+    {
+      "entropy": 0.6573980256915093,
+      "epoch": 0.7237635705669482,
+      "grad_norm": 0.4656430780887604,
+      "learning_rate": 0.00031150858133635,
+      "loss": 0.6215964794158936,
+      "mean_token_accuracy": 0.8208559453487396,
+      "num_tokens": 740255.0,
+      "step": 300
+    },
+    {
+      "epoch": 0.7237635705669482,
+      "eval_entropy": 0.6834242812703165,
+      "eval_loss": 0.6479960083961487,
+      "eval_mean_token_accuracy": 0.8128752906001016,
+      "eval_num_tokens": 740255.0,
+      "eval_runtime": 100.6101,
+      "eval_samples_per_second": 14.114,
+      "eval_steps_per_second": 1.769,
+      "step": 300
+    },
+    {
+      "entropy": 0.6751310348510742,
+      "epoch": 0.7720144752714113,
+      "grad_norm": 0.4495522081851959,
+      "learning_rate": 0.00033234527574011916,
+      "loss": 0.6402668952941895,
+      "mean_token_accuracy": 0.8170571058988572,
+      "num_tokens": 789529.0,
+      "step": 320
+    },
+    {
+      "epoch": 0.7720144752714113,
+      "eval_entropy": 0.6920045178927733,
+      "eval_loss": 0.6402787566184998,
+      "eval_mean_token_accuracy": 0.817791668886549,
+      "eval_num_tokens": 789529.0,
+      "eval_runtime": 100.6347,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 320
+    },
+    {
+      "entropy": 0.6846290081739426,
+      "epoch": 0.8202653799758746,
+      "grad_norm": 0.5661432147026062,
+      "learning_rate": 0.00035318197014388845,
+      "loss": 0.6508871078491211,
+      "mean_token_accuracy": 0.8164169207215309,
+      "num_tokens": 841584.0,
+      "step": 340
+    },
+    {
+      "epoch": 0.8202653799758746,
+      "eval_entropy": 0.6772527078564248,
+      "eval_loss": 0.6512064337730408,
+      "eval_mean_token_accuracy": 0.8170605137776793,
+      "eval_num_tokens": 841584.0,
+      "eval_runtime": 100.5157,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 340
+    },
+    {
+      "entropy": 0.7022885024547577,
+      "epoch": 0.8685162846803377,
+      "grad_norm": 0.6357236504554749,
+      "learning_rate": 0.0003740186645476577,
+      "loss": 0.6649135112762451,
+      "mean_token_accuracy": 0.8136772960424423,
+      "num_tokens": 885769.0,
+      "step": 360
+    },
+    {
+      "epoch": 0.8685162846803377,
+      "eval_entropy": 0.7435127543599418,
+      "eval_loss": 0.6573262214660645,
+      "eval_mean_token_accuracy": 0.8147802225659403,
+      "eval_num_tokens": 885769.0,
+      "eval_runtime": 100.5271,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 360
+    },
+    {
+      "entropy": 0.6962979376316071,
+      "epoch": 0.916767189384801,
+      "grad_norm": 0.6179454326629639,
+      "learning_rate": 0.00039485535895142685,
+      "loss": 0.6509191513061523,
+      "mean_token_accuracy": 0.81364226937294,
+      "num_tokens": 930056.0,
+      "step": 380
+    },
+    {
+      "epoch": 0.916767189384801,
+      "eval_entropy": 0.6630894883295123,
+      "eval_loss": 0.6524088978767395,
+      "eval_mean_token_accuracy": 0.8165898587596551,
+      "eval_num_tokens": 930056.0,
+      "eval_runtime": 100.5343,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 380
+    },
+    {
+      "entropy": 0.694319449365139,
+      "epoch": 0.9650180940892642,
+      "grad_norm": 0.5458664894104004,
+      "learning_rate": 0.00041569205335519613,
+      "loss": 0.6624011993408203,
+      "mean_token_accuracy": 0.8136080905795098,
+      "num_tokens": 976892.0,
+      "step": 400
+    },
+    {
+      "epoch": 0.9650180940892642,
+      "eval_entropy": 0.7424925324622165,
+      "eval_loss": 0.6717661023139954,
+      "eval_mean_token_accuracy": 0.8099894781460922,
+      "eval_num_tokens": 976892.0,
+      "eval_runtime": 100.5992,
+      "eval_samples_per_second": 14.115,
+      "eval_steps_per_second": 1.769,
+      "step": 400
+    },
+    {
+      "entropy": 0.6861714736009256,
+      "epoch": 1.0120627261761157,
+      "grad_norm": 0.6108037829399109,
+      "learning_rate": 0.0004323601853188671,
+      "loss": 0.6551628112792969,
+      "mean_token_accuracy": 0.8134882618219424,
+      "num_tokens": 1024472.0,
+      "step": 420
+    },
+    {
+      "epoch": 1.0120627261761157,
+      "eval_entropy": 0.6593745705786715,
+      "eval_loss": 0.6720747947692871,
+      "eval_mean_token_accuracy": 0.812802647941568,
+      "eval_num_tokens": 1024472.0,
+      "eval_runtime": 100.4866,
+      "eval_samples_per_second": 14.131,
+      "eval_steps_per_second": 1.771,
+      "step": 420
+    },
+    {
+      "entropy": 0.639520151913166,
+      "epoch": 1.060313630880579,
+      "grad_norm": 0.6054353713989258,
+      "learning_rate": 0.00043231736219609266,
+      "loss": 0.6096667766571044,
+      "mean_token_accuracy": 0.8227144703269005,
+      "num_tokens": 1076964.0,
+      "step": 440
+    },
+    {
+      "epoch": 1.060313630880579,
+      "eval_entropy": 0.6316307876886946,
+      "eval_loss": 0.6621950268745422,
+      "eval_mean_token_accuracy": 0.8140057914712456,
+      "eval_num_tokens": 1076964.0,
+      "eval_runtime": 100.6409,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 440
+    },
+    {
+      "entropy": 0.6544070363044738,
+      "epoch": 1.1085645355850422,
+      "grad_norm": 0.5111057758331299,
+      "learning_rate": 0.0004322133749558233,
+      "loss": 0.6176822185516357,
+      "mean_token_accuracy": 0.8201584875583648,
+      "num_tokens": 1126064.0,
+      "step": 460
+    },
+    {
+      "epoch": 1.1085645355850422,
+      "eval_entropy": 0.6714627766207363,
+      "eval_loss": 0.6738516092300415,
+      "eval_mean_token_accuracy": 0.8128616829936424,
+      "eval_num_tokens": 1126064.0,
+      "eval_runtime": 100.4986,
+      "eval_samples_per_second": 14.13,
+      "eval_steps_per_second": 1.771,
+      "step": 460
+    },
+    {
+      "entropy": 0.6470697179436684,
+      "epoch": 1.1568154402895054,
+      "grad_norm": 0.5434669256210327,
+      "learning_rate": 0.0004320482530252179,
+      "loss": 0.6144762516021729,
+      "mean_token_accuracy": 0.819926056265831,
+      "num_tokens": 1176912.0,
+      "step": 480
+    },
+    {
+      "epoch": 1.1568154402895054,
+      "eval_entropy": 0.6475969080844622,
+      "eval_loss": 0.6624214053153992,
+      "eval_mean_token_accuracy": 0.8137228820431098,
+      "eval_num_tokens": 1176912.0,
+      "eval_runtime": 100.5617,
+      "eval_samples_per_second": 14.121,
+      "eval_steps_per_second": 1.77,
+      "step": 480
+    },
+    {
+      "entropy": 0.6659130036830903,
+      "epoch": 1.2050663449939687,
+      "grad_norm": 0.6470707058906555,
+      "learning_rate": 0.0004318220431318296,
+      "loss": 0.6366308212280274,
+      "mean_token_accuracy": 0.8175261527299881,
+      "num_tokens": 1223383.0,
+      "step": 500
+    },
+    {
+      "epoch": 1.2050663449939687,
+      "eval_entropy": 0.7232425132494295,
+      "eval_loss": 0.6877738833427429,
+      "eval_mean_token_accuracy": 0.8007767260744331,
+      "eval_num_tokens": 1223383.0,
+      "eval_runtime": 100.477,
+      "eval_samples_per_second": 14.133,
+      "eval_steps_per_second": 1.772,
+      "step": 500
+    },
+    {
+      "entropy": 0.6759189993143082,
+      "epoch": 1.2533172496984317,
+      "grad_norm": 0.6453561186790466,
+      "learning_rate": 0.00043153480929038245,
+      "loss": 0.637772512435913,
+      "mean_token_accuracy": 0.8162627041339874,
+      "num_tokens": 1273534.0,
+      "step": 520
+    },
+    {
+      "epoch": 1.2533172496984317,
+      "eval_entropy": 0.7082584312792575,
+      "eval_loss": 0.6695764064788818,
+      "eval_mean_token_accuracy": 0.812019992410467,
+      "eval_num_tokens": 1273534.0,
+      "eval_runtime": 100.4957,
+      "eval_samples_per_second": 14.13,
+      "eval_steps_per_second": 1.771,
+      "step": 520
+    },
+    {
+      "entropy": 0.6730152070522308,
+      "epoch": 1.301568154402895,
+      "grad_norm": 0.6234476566314697,
+      "learning_rate": 0.000431186632784656,
+      "loss": 0.640014362335205,
+      "mean_token_accuracy": 0.8168858170509339,
+      "num_tokens": 1323351.0,
+      "step": 540
+    },
+    {
+      "epoch": 1.301568154402895,
+      "eval_entropy": 0.6655137846978862,
+      "eval_loss": 0.6578248739242554,
+      "eval_mean_token_accuracy": 0.8147600979617472,
+      "eval_num_tokens": 1323351.0,
+      "eval_runtime": 100.503,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 540
+    },
+    {
+      "entropy": 0.6636043325066566,
+      "epoch": 1.3498190591073582,
+      "grad_norm": 0.5882512331008911,
+      "learning_rate": 0.0004307776121444827,
+      "loss": 0.6286775588989257,
+      "mean_token_accuracy": 0.8183174058794975,
+      "num_tokens": 1371508.0,
+      "step": 560
+    },
+    {
+      "epoch": 1.3498190591073582,
+      "eval_entropy": 0.6304813810948575,
+      "eval_loss": 0.6607436537742615,
+      "eval_mean_token_accuracy": 0.8169659268989992,
+      "eval_num_tokens": 1371508.0,
+      "eval_runtime": 100.6233,
+      "eval_samples_per_second": 14.112,
+      "eval_steps_per_second": 1.769,
+      "step": 560
+    },
+    {
+      "entropy": 0.6608738481998444,
+      "epoch": 1.3980699638118215,
+      "grad_norm": 0.6796398758888245,
+      "learning_rate": 0.00043030786311786564,
+      "loss": 0.622726583480835,
+      "mean_token_accuracy": 0.8230827242136002,
+      "num_tokens": 1415221.0,
+      "step": 580
+    },
+    {
+      "epoch": 1.3980699638118215,
+      "eval_entropy": 0.6645242689030894,
+      "eval_loss": 0.6662811636924744,
+      "eval_mean_token_accuracy": 0.8147623907314258,
+      "eval_num_tokens": 1415221.0,
+      "eval_runtime": 100.5074,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 580
+    },
+    {
+      "entropy": 0.6682895936071873,
+      "epoch": 1.4463208685162847,
+      "grad_norm": 0.5750718712806702,
+      "learning_rate": 0.0004297775186382226,
+      "loss": 0.6304276466369629,
+      "mean_token_accuracy": 0.8188625335693359,
+      "num_tokens": 1463356.0,
+      "step": 600
+    },
+    {
+      "epoch": 1.4463208685162847,
+      "eval_entropy": 0.6320532578765676,
+      "eval_loss": 0.650333046913147,
+      "eval_mean_token_accuracy": 0.8172853793321031,
+      "eval_num_tokens": 1463356.0,
+      "eval_runtime": 100.5116,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 600
+    },
+    {
+      "entropy": 0.653739221394062,
+      "epoch": 1.494571773220748,
+      "grad_norm": 0.5552534461021423,
+      "learning_rate": 0.00042918672878676804,
+      "loss": 0.6215689659118653,
+      "mean_token_accuracy": 0.8208896175026894,
+      "num_tokens": 1512340.0,
+      "step": 620
+    },
+    {
+      "epoch": 1.494571773220748,
+      "eval_entropy": 0.6285932062382109,
+      "eval_loss": 0.6392536163330078,
+      "eval_mean_token_accuracy": 0.8205167825302381,
+      "eval_num_tokens": 1512340.0,
+      "eval_runtime": 100.6392,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 620
+    },
+    {
+      "entropy": 0.6476897612214089,
+      "epoch": 1.5428226779252112,
+      "grad_norm": 0.536978542804718,
+      "learning_rate": 0.00042853566075004156,
+      "loss": 0.6131608009338378,
+      "mean_token_accuracy": 0.8208152890205384,
+      "num_tokens": 1560911.0,
+      "step": 640
+    },
+    {
+      "epoch": 1.5428226779252112,
+      "eval_entropy": 0.6626498856571283,
+      "eval_loss": 0.6479025483131409,
+      "eval_mean_token_accuracy": 0.8180894101603647,
+      "eval_num_tokens": 1560911.0,
+      "eval_runtime": 100.5035,
+      "eval_samples_per_second": 14.129,
+      "eval_steps_per_second": 1.771,
+      "step": 640
+    },
+    {
+      "entropy": 0.6338633447885513,
+      "epoch": 1.5910735826296745,
+      "grad_norm": 0.4518965780735016,
+      "learning_rate": 0.0004278244987725961,
+      "loss": 0.6010610580444335,
+      "mean_token_accuracy": 0.8247816175222397,
+      "num_tokens": 1609409.0,
+      "step": 660
+    },
+    {
+      "epoch": 1.5910735826296745,
+      "eval_entropy": 0.64079723699709,
+      "eval_loss": 0.6417157053947449,
+      "eval_mean_token_accuracy": 0.8214994581897607,
+      "eval_num_tokens": 1609409.0,
+      "eval_runtime": 100.6393,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 660
+    },
+    {
+      "entropy": 0.6485912457108498,
+      "epoch": 1.6393244873341375,
+      "grad_norm": 0.6028368473052979,
+      "learning_rate": 0.00042705344410485905,
+      "loss": 0.6120437145233154,
+      "mean_token_accuracy": 0.8236212745308876,
+      "num_tokens": 1657430.0,
+      "step": 680
+    },
+    {
+      "epoch": 1.6393244873341375,
+      "eval_entropy": 0.6863006427046958,
+      "eval_loss": 0.6420931816101074,
+      "eval_mean_token_accuracy": 0.8191533580924688,
+      "eval_num_tokens": 1657430.0,
+      "eval_runtime": 100.4639,
+      "eval_samples_per_second": 14.134,
+      "eval_steps_per_second": 1.772,
+      "step": 680
+    },
+    {
+      "entropy": 0.6459382772445679,
+      "epoch": 1.6875753920386007,
+      "grad_norm": 0.5069310069084167,
+      "learning_rate": 0.00042622271494618073,
+      "loss": 0.6160889625549316,
+      "mean_token_accuracy": 0.8253828629851341,
+      "num_tokens": 1708058.0,
+      "step": 700
+    },
+    {
+      "epoch": 1.6875753920386007,
+      "eval_entropy": 0.6773919463157654,
+      "eval_loss": 0.6307896375656128,
+      "eval_mean_token_accuracy": 0.8217989825800563,
+      "eval_num_tokens": 1708058.0,
+      "eval_runtime": 100.6301,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 700
+    },
+    {
+      "entropy": 0.64687028080225,
+      "epoch": 1.7358262967430638,
+      "grad_norm": 0.3830573856830597,
+      "learning_rate": 0.00042533254638308645,
+      "loss": 0.6104001998901367,
+      "mean_token_accuracy": 0.8255606815218925,
+      "num_tokens": 1761262.0,
+      "step": 720
+    },
+    {
+      "epoch": 1.7358262967430638,
+      "eval_entropy": 0.6179541145147902,
+      "eval_loss": 0.6170215606689453,
+      "eval_mean_token_accuracy": 0.8259220900160543,
+      "eval_num_tokens": 1761262.0,
+      "eval_runtime": 100.61,
+      "eval_samples_per_second": 14.114,
+      "eval_steps_per_second": 1.769,
+      "step": 720
+    },
+    {
+      "entropy": 0.6306161753833294,
+      "epoch": 1.784077201447527,
+      "grad_norm": 0.5888957977294922,
+      "learning_rate": 0.0004243831903227498,
+      "loss": 0.5981546878814697,
+      "mean_token_accuracy": 0.8266776099801063,
+      "num_tokens": 1812127.0,
+      "step": 740
+    },
+    {
+      "epoch": 1.784077201447527,
+      "eval_entropy": 0.6357416808605194,
+      "eval_loss": 0.6202242374420166,
+      "eval_mean_token_accuracy": 0.8253380304641937,
+      "eval_num_tokens": 1812127.0,
+      "eval_runtime": 100.5145,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 740
+    },
+    {
+      "entropy": 0.6290688961744308,
+      "epoch": 1.8323281061519903,
+      "grad_norm": 0.6138001680374146,
+      "learning_rate": 0.0004233749154217063,
+      "loss": 0.5853167057037354,
+      "mean_token_accuracy": 0.8281400203704834,
+      "num_tokens": 1858403.0,
+      "step": 760
+    },
+    {
+      "epoch": 1.8323281061519903,
+      "eval_entropy": 0.6402715869833914,
+      "eval_loss": 0.6168262958526611,
+      "eval_mean_token_accuracy": 0.8246346306934785,
+      "eval_num_tokens": 1858403.0,
+      "eval_runtime": 100.5237,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 760
+    },
+    {
+      "entropy": 0.6218615680932998,
+      "epoch": 1.8805790108564535,
+      "grad_norm": 0.574312686920166,
+      "learning_rate": 0.00042230800700982586,
+      "loss": 0.5841915607452393,
+      "mean_token_accuracy": 0.8304707780480385,
+      "num_tokens": 1907361.0,
+      "step": 780
+    },
+    {
+      "epoch": 1.8805790108564535,
+      "eval_entropy": 0.6301163148679091,
+      "eval_loss": 0.6073815226554871,
+      "eval_mean_token_accuracy": 0.8274695993809218,
+      "eval_num_tokens": 1907361.0,
+      "eval_runtime": 100.5111,
+      "eval_samples_per_second": 14.128,
+      "eval_steps_per_second": 1.771,
+      "step": 780
+    },
+    {
+      "entropy": 0.6101033873856068,
+      "epoch": 1.9288299155609168,
+      "grad_norm": 0.5054804682731628,
+      "learning_rate": 0.000421182767009569,
+      "loss": 0.5774452209472656,
+      "mean_token_accuracy": 0.8321943372488022,
+      "num_tokens": 1956605.0,
+      "step": 800
+    },
+    {
+      "epoch": 1.9288299155609168,
+      "eval_entropy": 0.631400960215022,
+      "eval_loss": 0.6115217804908752,
+      "eval_mean_token_accuracy": 0.8249618762664581,
+      "eval_num_tokens": 1956605.0,
+      "eval_runtime": 100.4916,
+      "eval_samples_per_second": 14.131,
+      "eval_steps_per_second": 1.771,
+      "step": 800
+    },
+    {
+      "entropy": 0.6217561915516854,
+      "epoch": 1.97708082026538,
+      "grad_norm": 0.5333649516105652,
+      "learning_rate": 0.0004199995138505454,
+      "loss": 0.5842787265777588,
+      "mean_token_accuracy": 0.8305525362491608,
+      "num_tokens": 2003308.0,
+      "step": 820
+    },
+    {
+      "epoch": 1.97708082026538,
+      "eval_entropy": 0.605928597657868,
+      "eval_loss": 0.6067516207695007,
+      "eval_mean_token_accuracy": 0.82715918408351,
+      "eval_num_tokens": 2003308.0,
+      "eval_runtime": 100.5286,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 820
+    },
+    {
+      "entropy": 0.5585076526189462,
+      "epoch": 2.0241254523522314,
+      "grad_norm": 0.44819608330726624,
+      "learning_rate": 0.00041875858237940244,
+      "loss": 0.5252367496490479,
+      "mean_token_accuracy": 0.8405626966403081,
+      "num_tokens": 2051331.0,
+      "step": 840
+    },
+    {
+      "epoch": 2.0241254523522314,
+      "eval_entropy": 0.6123150769579276,
+      "eval_loss": 0.6175946593284607,
+      "eval_mean_token_accuracy": 0.827637837509091,
+      "eval_num_tokens": 2051331.0,
+      "eval_runtime": 100.5414,
+      "eval_samples_per_second": 14.124,
+      "eval_steps_per_second": 1.77,
+      "step": 840
+    },
+    {
+      "entropy": 0.5359115816652775,
+      "epoch": 2.0723763570566947,
+      "grad_norm": 0.576781690120697,
+      "learning_rate": 0.0004174603237650678,
+      "loss": 0.5023207664489746,
+      "mean_token_accuracy": 0.8459693863987923,
+      "num_tokens": 2101528.0,
+      "step": 860
+    },
+    {
+      "epoch": 2.0723763570566947,
+      "eval_entropy": 0.5592134784111816,
+      "eval_loss": 0.620693564414978,
+      "eval_mean_token_accuracy": 0.827402507656076,
+      "eval_num_tokens": 2101528.0,
+      "eval_runtime": 100.6274,
+      "eval_samples_per_second": 14.111,
+      "eval_steps_per_second": 1.769,
+      "step": 860
+    },
+    {
+      "entropy": 0.5214565828442573,
+      "epoch": 2.120627261761158,
+      "grad_norm": 0.45878875255584717,
+      "learning_rate": 0.00041610510539937234,
+      "loss": 0.48367843627929685,
+      "mean_token_accuracy": 0.8515541926026344,
+      "num_tokens": 2149550.0,
+      "step": 880
+    },
+    {
+      "epoch": 2.120627261761158,
+      "eval_entropy": 0.5860532344727034,
+      "eval_loss": 0.6134791970252991,
+      "eval_mean_token_accuracy": 0.8291774777214179,
+      "eval_num_tokens": 2149550.0,
+      "eval_runtime": 100.5295,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 880
+    },
+    {
+      "entropy": 0.5414379432797432,
+      "epoch": 2.168878166465621,
+      "grad_norm": 0.5976294875144958,
+      "learning_rate": 0.0004146933107930829,
+      "loss": 0.5012908935546875,
+      "mean_token_accuracy": 0.8466357633471489,
+      "num_tokens": 2202162.0,
+      "step": 900
+    },
+    {
+      "epoch": 2.168878166465621,
+      "eval_entropy": 0.5427783280611038,
+      "eval_loss": 0.612716019153595,
+      "eval_mean_token_accuracy": 0.829505049445656,
+      "eval_num_tokens": 2202162.0,
+      "eval_runtime": 100.5207,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 900
+    },
+    {
+      "entropy": 0.521108990907669,
+      "epoch": 2.2171290711700844,
+      "grad_norm": 0.5145775675773621,
+      "learning_rate": 0.000413225339467373,
+      "loss": 0.4877732276916504,
+      "mean_token_accuracy": 0.8504189744591713,
+      "num_tokens": 2256539.0,
+      "step": 920
+    },
+    {
+      "epoch": 2.2171290711700844,
+      "eval_entropy": 0.5620684869838565,
+      "eval_loss": 0.6176337599754333,
+      "eval_mean_token_accuracy": 0.8291444483767735,
+      "eval_num_tokens": 2256539.0,
+      "eval_runtime": 100.6819,
+      "eval_samples_per_second": 14.104,
+      "eval_steps_per_second": 1.768,
+      "step": 920
+    },
+    {
+      "entropy": 0.5452766507863999,
+      "epoch": 2.2653799758745476,
+      "grad_norm": 0.4820452928543091,
+      "learning_rate": 0.0004117016068407636,
+      "loss": 0.4991307735443115,
+      "mean_token_accuracy": 0.8474363312125206,
+      "num_tokens": 2299698.0,
+      "step": 940
+    },
+    {
+      "epoch": 2.2653799758745476,
+      "eval_entropy": 0.5611683193887218,
+      "eval_loss": 0.6153580546379089,
+      "eval_mean_token_accuracy": 0.8286041272490212,
+      "eval_num_tokens": 2299698.0,
+      "eval_runtime": 100.5249,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 1.771,
+      "step": 940
+    },
+    {
+      "entropy": 0.5572576269507408,
+      "epoch": 2.313630880579011,
+      "grad_norm": 0.3947993814945221,
+      "learning_rate": 0.00041012254411156387,
+      "loss": 0.5191003799438476,
+      "mean_token_accuracy": 0.8422838419675827,
+      "num_tokens": 2347134.0,
+      "step": 960
+    },
+    {
+      "epoch": 2.313630880579011,
+      "eval_entropy": 0.5651392345682958,
+      "eval_loss": 0.6121005415916443,
+      "eval_mean_token_accuracy": 0.8286536242185014,
+      "eval_num_tokens": 2347134.0,
+      "eval_runtime": 100.7033,
+      "eval_samples_per_second": 14.101,
+      "eval_steps_per_second": 1.768,
+      "step": 960
+    },
+    {
+      "entropy": 0.5441671140491963,
+      "epoch": 2.361881785283474,
+      "grad_norm": 0.7719791531562805,
+      "learning_rate": 0.0004084885981358477,
+      "loss": 0.5072606086730957,
+      "mean_token_accuracy": 0.8478495761752128,
+      "num_tokens": 2394568.0,
+      "step": 980
+    },
+    {
+      "epoch": 2.361881785283474,
+      "eval_entropy": 0.5651004270891126,
+      "eval_loss": 0.6069501042366028,
+      "eval_mean_token_accuracy": 0.8311042156112328,
+      "eval_num_tokens": 2394568.0,
+      "eval_runtime": 100.5487,
+      "eval_samples_per_second": 14.123,
+      "eval_steps_per_second": 1.77,
+      "step": 980
+    },
+    {
+      "entropy": 0.5462074346840382,
+      "epoch": 2.4101326899879374,
+      "grad_norm": 0.5694436430931091,
+      "learning_rate": 0.00040680023130099856,
+      "loss": 0.5097150325775146,
+      "mean_token_accuracy": 0.8474388793110847,
+      "num_tokens": 2443253.0,
+      "step": 1000
+    },
+    {
+      "epoch": 2.4101326899879374,
+      "eval_entropy": 0.5895893243926295,
+      "eval_loss": 0.6078168749809265,
+      "eval_mean_token_accuracy": 0.828030709470256,
+      "eval_num_tokens": 2443253.0,
+      "eval_runtime": 100.5413,
+      "eval_samples_per_second": 14.124,
+      "eval_steps_per_second": 1.77,
+      "step": 1000
+    },
+    {
+      "entropy": 0.5473216332495212,
+      "epoch": 2.4583835946924006,
+      "grad_norm": 0.3989660143852234,
+      "learning_rate": 0.00040505792139485913,
+      "loss": 0.5087806224822998,
+      "mean_token_accuracy": 0.8473173275589942,
+      "num_tokens": 2495178.0,
+      "step": 1020
+    },
+    {
+      "epoch": 2.4583835946924006,
+      "eval_entropy": 0.5588909702354603,
+      "eval_loss": 0.6087605953216553,
+      "eval_mean_token_accuracy": 0.8309365524334854,
+      "eval_num_tokens": 2495178.0,
+      "eval_runtime": 100.5137,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 1020
+    },
+    {
+      "entropy": 0.5430163390934467,
+      "epoch": 2.5066344993968634,
+      "grad_norm": 0.4916608929634094,
+      "learning_rate": 0.00040326216147052336,
+      "loss": 0.5027707576751709,
+      "mean_token_accuracy": 0.8466959804296493,
+      "num_tokens": 2541733.0,
+      "step": 1040
+    },
+    {
+      "epoch": 2.5066344993968634,
+      "eval_entropy": 0.5819033988741007,
+      "eval_loss": 0.609398365020752,
+      "eval_mean_token_accuracy": 0.8297666612635838,
+      "eval_num_tokens": 2541733.0,
+      "eval_runtime": 100.6375,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 1.769,
+      "step": 1040
+    },
+    {
+      "entropy": 0.5632683664560318,
+      "epoch": 2.554885404101327,
+      "grad_norm": 0.6062753796577454,
+      "learning_rate": 0.00040141345970680814,
+      "loss": 0.5272740840911865,
+      "mean_token_accuracy": 0.8411559104919434,
+      "num_tokens": 2589479.0,
+      "step": 1060
+    },
+    {
+      "epoch": 2.554885404101327,
+      "eval_entropy": 0.5785442567273472,
+      "eval_loss": 0.6207635998725891,
+      "eval_mean_token_accuracy": 0.8259770478425401,
+      "eval_num_tokens": 2589479.0,
+      "eval_runtime": 100.5161,
+      "eval_samples_per_second": 14.127,
+      "eval_steps_per_second": 1.771,
+      "step": 1060
+    },
+    {
+      "entropy": 0.5591374494135379,
+      "epoch": 2.60313630880579,
+      "grad_norm": 0.5771989226341248,
+      "learning_rate": 0.00039951233926444437,
+      "loss": 0.5237057685852051,
+      "mean_token_accuracy": 0.8413331478834152,
+      "num_tokens": 2639134.0,
+      "step": 1080
+    },
+    {
+      "epoch": 2.60313630880579,
+      "eval_entropy": 0.5765141543377651,
+      "eval_loss": 0.6119762063026428,
+      "eval_mean_token_accuracy": 0.8289416998289945,
+      "eval_num_tokens": 2639134.0,
+      "eval_runtime": 100.5288,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 1080
+    },
+    {
+      "entropy": 0.5470930114388466,
+      "epoch": 2.651387213510253,
+      "grad_norm": 0.4211709499359131,
+      "learning_rate": 0.00039755933813802937,
+      "loss": 0.5157970428466797,
+      "mean_token_accuracy": 0.8436284482479095,
+      "num_tokens": 2689324.0,
+      "step": 1100
+    },
+    {
+      "epoch": 2.651387213510253,
+      "eval_entropy": 0.5593592817528864,
+      "eval_loss": 0.6080028414726257,
+      "eval_mean_token_accuracy": 0.8277805718813049,
+      "eval_num_tokens": 2689324.0,
+      "eval_runtime": 100.495,
+      "eval_samples_per_second": 14.13,
+      "eval_steps_per_second": 1.771,
+      "step": 1100
+    },
+    {
+      "entropy": 0.5636695981025696,
+      "epoch": 2.6996381182147164,
+      "grad_norm": 0.6020221710205078,
+      "learning_rate": 0.00039555500900377994,
+      "loss": 0.5277415752410889,
+      "mean_token_accuracy": 0.840576496720314,
+      "num_tokens": 2736182.0,
+      "step": 1120
+    },
+    {
+      "epoch": 2.6996381182147164,
+      "eval_entropy": 0.5837533748551701,
+      "eval_loss": 0.6048349142074585,
+      "eval_mean_token_accuracy": 0.8288552148288555,
+      "eval_num_tokens": 2736182.0,
+      "eval_runtime": 100.5342,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 1120
+    },
+    {
+      "entropy": 0.5649972304701805,
+      "epoch": 2.7478890229191797,
+      "grad_norm": 0.42798763513565063,
+      "learning_rate": 0.00039349991906313236,
+      "loss": 0.5282054901123047,
+      "mean_token_accuracy": 0.8409290999174118,
+      "num_tokens": 2781503.0,
+      "step": 1140
+    },
+    {
+      "epoch": 2.7478890229191797,
+      "eval_entropy": 0.5473378684078709,
+      "eval_loss": 0.6050286889076233,
+      "eval_mean_token_accuracy": 0.8313557287950194,
+      "eval_num_tokens": 2781503.0,
+      "eval_runtime": 100.5342,
+      "eval_samples_per_second": 14.125,
+      "eval_steps_per_second": 1.771,
+      "step": 1140
+    }
+  ],
+  "logging_steps": 20,
+  "max_steps": 4150,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 20,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.77882635423744e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}