Add files using upload-large-folder tool

ab4/inverter/ao.ab4.mean.frombase.e1cend/step1000/lora/README.md

@@ -0,0 +1,206 @@
+---
+base_model: Qwen/Qwen3.6-27B
+library_name: peft
+tags:
+- base_model:adapter:Qwen/Qwen3.6-27B
+- lora
+- transformers
+---
+
+# 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 0.19.1

ab4/inverter/ao.ab4.mean.frombase.e1cend/step1000/lora/adapter_config.json

@@ -0,0 +1,56 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForCausalLM",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5"
+  },
+  "base_model_name_or_path": "Qwen/Qwen3.6-27B",
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.19.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "o_proj",
+    "in_proj_qkv",
+    "q_proj",
+    "v_proj",
+    "out_proj",
+    "in_proj_z",
+    "k_proj",
+    "down_proj",
+    "in_proj_b",
+    "in_proj_a",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend/step1000/meta.json

@@ -0,0 +1,104 @@
+{
+  "micro_steps": 1000,
+  "tokens_span": 18284544.0,
+  "tokens_sup": 19308544.0,
+  "examples_seen": 1024000,
+  "val_ce": 1.3160197734832764,
+  "val_ce_L20": 1.3070255517959595,
+  "val_ce_L24": 1.3439029455184937,
+  "val_ce_L28": 1.2906959056854248,
+  "val_ce_L32": 1.3591547012329102,
+  "val_ce_L36": 1.2168859243392944,
+  "val_ce_L40": 1.2491480112075806,
+  "val_ce_L44": 1.2386751174926758,
+  "val_ce_L48": 1.2798957824707031,
+  "val_ce_L52": 1.2783353328704834,
+  "val_ce_L56": 1.4324098825454712,
+  "val_ce_L60": 1.501736044883728,
+  "train_ce": 1.2460187441706658,
+  "train_val_gap": 0.07000102931261054,
+  "config": {
+    "run_name": "ao.ab4.mean.frombase.e1cend",
+    "transform": "scaled",
+    "layers": [
+      20,
+      24,
+      28,
+      32,
+      36,
+      40,
+      44,
+      48,
+      52,
+      56,
+      60
+    ],
+    "alpha": 16000.0,
+    "clip_mult": 4.0,
+    "scales": {
+      "20": 115.07945174569515,
+      "24": 115.07945174569515,
+      "28": 115.07945174569515,
+      "32": 115.07945174569515,
+      "36": 115.07945174569515,
+      "40": 115.07945174569515,
+      "44": 115.07945174569515,
+      "48": 115.07945174569515,
+      "52": 115.07945174569515,
+      "56": 115.07945174569515,
+      "60": 115.07945174569515
+    },
+    "freeze_base": false,
+    "xform_hidden": 0,
+    "xform_arch": "resmlp",
+    "logits_on_targets": false,
+    "prefix_cache": false,
+    "n_examples": 3333472,
+    "min_len": 1,
+    "max_len": 64,
+    "lora_r": 16,
+    "lora_alpha": 32,
+    "lora_dropout": 0.05,
+    "lr": 0.0003,
+    "lr_sched": "constant",
+    "micro_batch": 128,
+    "val_micro_batch": 16,
+    "grad_accum": 8,
+    "epochs": 1,
+    "eval_every_steps": 500,
+    "max_eval_rows": 1024,
+    "warmup_steps": 100,
+    "seed": 0,
+    "init_from": "",
+    "skip_examples": 0,
+    "skip_batches": 0,
+    "exclude_idx_file": "",
+    "num_workers": 2,
+    "extra": {
+      "kind": "ao_ladder",
+      "ar_run": "ar.ab4.mean.e1c/end",
+      "target_prefix": "",
+      "global_scale": "115.07945174569515",
+      "pool": "ao_pool/pool_iolens_u64w2.safetensors",
+      "scale_mode": "global_frozen",
+      "grad_ckpt": "on",
+      "compile_blocks": "off",
+      "prompt": "continuation_raw",
+      "layers_per_crop": "4",
+      "val_source": "pool",
+      "max_len_filter": "32",
+      "adopt_arout_picks": "False",
+      "split_seed": "1234",
+      "arout_dir": "/scratch/ola/ao_arout_u64w2/ar.ab4.mean.e1c/end",
+      "val_pool": "ao_pool/pool_iolens_u64.safetensors",
+      "val_arout_dir": "ao_arout_u64/ar.ab4.mean.e1c/end"
+    },
+    "exact_resume": false,
+    "decay_frac": 0.1,
+    "warmup_examples": 0,
+    "expected_effective_batch": 0,
+    "eval_every_examples": 0,
+    "train_examples_cap": 0,
+    "total_opt_steps_override": null
+  }
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend/step1500/lora/README.md

@@ -0,0 +1,206 @@
+---
+base_model: Qwen/Qwen3.6-27B
+library_name: peft
+tags:
+- base_model:adapter:Qwen/Qwen3.6-27B
+- lora
+- transformers
+---
+
+# 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 0.19.1

ab4/inverter/ao.ab4.mean.frombase.e1cend/step1500/lora/adapter_config.json

@@ -0,0 +1,56 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForCausalLM",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5"
+  },
+  "base_model_name_or_path": "Qwen/Qwen3.6-27B",
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.19.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "o_proj",
+    "in_proj_qkv",
+    "q_proj",
+    "v_proj",
+    "out_proj",
+    "in_proj_z",
+    "k_proj",
+    "down_proj",
+    "in_proj_b",
+    "in_proj_a",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend/step1500/meta.json

@@ -0,0 +1,104 @@
+{
+  "micro_steps": 1500,
+  "tokens_span": 27467776.0,
+  "tokens_sup": 29003776.0,
+  "examples_seen": 1536000,
+  "val_ce": 1.2897164821624756,
+  "val_ce_L20": 1.2548513412475586,
+  "val_ce_L24": 1.3059653043746948,
+  "val_ce_L28": 1.2445114850997925,
+  "val_ce_L32": 1.3372445106506348,
+  "val_ce_L36": 1.1796061992645264,
+  "val_ce_L40": 1.2273386716842651,
+  "val_ce_L44": 1.210042119026184,
+  "val_ce_L48": 1.2801361083984375,
+  "val_ce_L52": 1.2656112909317017,
+  "val_ce_L56": 1.4198745489120483,
+  "val_ce_L60": 1.488851547241211,
+  "train_ce": 1.1425579063892364,
+  "train_val_gap": 0.1471585757732392,
+  "config": {
+    "run_name": "ao.ab4.mean.frombase.e1cend",
+    "transform": "scaled",
+    "layers": [
+      20,
+      24,
+      28,
+      32,
+      36,
+      40,
+      44,
+      48,
+      52,
+      56,
+      60
+    ],
+    "alpha": 16000.0,
+    "clip_mult": 4.0,
+    "scales": {
+      "20": 115.07945174569515,
+      "24": 115.07945174569515,
+      "28": 115.07945174569515,
+      "32": 115.07945174569515,
+      "36": 115.07945174569515,
+      "40": 115.07945174569515,
+      "44": 115.07945174569515,
+      "48": 115.07945174569515,
+      "52": 115.07945174569515,
+      "56": 115.07945174569515,
+      "60": 115.07945174569515
+    },
+    "freeze_base": false,
+    "xform_hidden": 0,
+    "xform_arch": "resmlp",
+    "logits_on_targets": false,
+    "prefix_cache": false,
+    "n_examples": 3333472,
+    "min_len": 1,
+    "max_len": 64,
+    "lora_r": 16,
+    "lora_alpha": 32,
+    "lora_dropout": 0.05,
+    "lr": 0.0003,
+    "lr_sched": "constant",
+    "micro_batch": 128,
+    "val_micro_batch": 16,
+    "grad_accum": 8,
+    "epochs": 1,
+    "eval_every_steps": 500,
+    "max_eval_rows": 1024,
+    "warmup_steps": 100,
+    "seed": 0,
+    "init_from": "",
+    "skip_examples": 0,
+    "skip_batches": 0,
+    "exclude_idx_file": "",
+    "num_workers": 2,
+    "extra": {
+      "kind": "ao_ladder",
+      "ar_run": "ar.ab4.mean.e1c/end",
+      "target_prefix": "",
+      "global_scale": "115.07945174569515",
+      "pool": "ao_pool/pool_iolens_u64w2.safetensors",
+      "scale_mode": "global_frozen",
+      "grad_ckpt": "on",
+      "compile_blocks": "off",
+      "prompt": "continuation_raw",
+      "layers_per_crop": "4",
+      "val_source": "pool",
+      "max_len_filter": "32",
+      "adopt_arout_picks": "False",
+      "split_seed": "1234",
+      "arout_dir": "/scratch/ola/ao_arout_u64w2/ar.ab4.mean.e1c/end",
+      "val_pool": "ao_pool/pool_iolens_u64.safetensors",
+      "val_arout_dir": "ao_arout_u64/ar.ab4.mean.e1c/end"
+    },
+    "exact_resume": false,
+    "decay_frac": 0.1,
+    "warmup_examples": 0,
+    "expected_effective_batch": 0,
+    "eval_every_examples": 0,
+    "train_examples_cap": 0,
+    "total_opt_steps_override": null
+  }
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend/step2000/lora/README.md

@@ -0,0 +1,206 @@
+---
+base_model: Qwen/Qwen3.6-27B
+library_name: peft
+tags:
+- base_model:adapter:Qwen/Qwen3.6-27B
+- lora
+- transformers
+---
+
+# 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 0.19.1

ab4/inverter/ao.ab4.mean.frombase.e1cend/step2000/lora/adapter_config.json

@@ -0,0 +1,56 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForCausalLM",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5"
+  },
+  "base_model_name_or_path": "Qwen/Qwen3.6-27B",
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.19.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "o_proj",
+    "in_proj_qkv",
+    "q_proj",
+    "v_proj",
+    "out_proj",
+    "in_proj_z",
+    "k_proj",
+    "down_proj",
+    "in_proj_b",
+    "in_proj_a",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend/step2000/meta.json

@@ -0,0 +1,104 @@
+{
+  "micro_steps": 2000,
+  "tokens_span": 36853760.0,
+  "tokens_sup": 38901760.0,
+  "examples_seen": 2048000,
+  "val_ce": 1.2604827880859375,
+  "val_ce_L20": 1.210369348526001,
+  "val_ce_L24": 1.2374399900436401,
+  "val_ce_L28": 1.1942862272262573,
+  "val_ce_L32": 1.3073383569717407,
+  "val_ce_L36": 1.1571012735366821,
+  "val_ce_L40": 1.2083390951156616,
+  "val_ce_L44": 1.188383936882019,
+  "val_ce_L48": 1.2531511783599854,
+  "val_ce_L52": 1.248670220375061,
+  "val_ce_L56": 1.4093008041381836,
+  "val_ce_L60": 1.4756282567977905,
+  "train_ce": 1.0634016906917094,
+  "train_val_gap": 0.19708109739422808,
+  "config": {
+    "run_name": "ao.ab4.mean.frombase.e1cend",
+    "transform": "scaled",
+    "layers": [
+      20,
+      24,
+      28,
+      32,
+      36,
+      40,
+      44,
+      48,
+      52,
+      56,
+      60
+    ],
+    "alpha": 16000.0,
+    "clip_mult": 4.0,
+    "scales": {
+      "20": 115.07945174569515,
+      "24": 115.07945174569515,
+      "28": 115.07945174569515,
+      "32": 115.07945174569515,
+      "36": 115.07945174569515,
+      "40": 115.07945174569515,
+      "44": 115.07945174569515,
+      "48": 115.07945174569515,
+      "52": 115.07945174569515,
+      "56": 115.07945174569515,
+      "60": 115.07945174569515
+    },
+    "freeze_base": false,
+    "xform_hidden": 0,
+    "xform_arch": "resmlp",
+    "logits_on_targets": false,
+    "prefix_cache": false,
+    "n_examples": 3333472,
+    "min_len": 1,
+    "max_len": 64,
+    "lora_r": 16,
+    "lora_alpha": 32,
+    "lora_dropout": 0.05,
+    "lr": 0.0003,
+    "lr_sched": "constant",
+    "micro_batch": 128,
+    "val_micro_batch": 16,
+    "grad_accum": 8,
+    "epochs": 1,
+    "eval_every_steps": 500,
+    "max_eval_rows": 1024,
+    "warmup_steps": 100,
+    "seed": 0,
+    "init_from": "",
+    "skip_examples": 0,
+    "skip_batches": 0,
+    "exclude_idx_file": "",
+    "num_workers": 2,
+    "extra": {
+      "kind": "ao_ladder",
+      "ar_run": "ar.ab4.mean.e1c/end",
+      "target_prefix": "",
+      "global_scale": "115.07945174569515",
+      "pool": "ao_pool/pool_iolens_u64w2.safetensors",
+      "scale_mode": "global_frozen",
+      "grad_ckpt": "on",
+      "compile_blocks": "off",
+      "prompt": "continuation_raw",
+      "layers_per_crop": "4",
+      "val_source": "pool",
+      "max_len_filter": "32",
+      "adopt_arout_picks": "False",
+      "split_seed": "1234",
+      "arout_dir": "/scratch/ola/ao_arout_u64w2/ar.ab4.mean.e1c/end",
+      "val_pool": "ao_pool/pool_iolens_u64.safetensors",
+      "val_arout_dir": "ao_arout_u64/ar.ab4.mean.e1c/end"
+    },
+    "exact_resume": false,
+    "decay_frac": 0.1,
+    "warmup_examples": 0,
+    "expected_effective_batch": 0,
+    "eval_every_examples": 0,
+    "train_examples_cap": 0,
+    "total_opt_steps_override": null
+  }
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend/step2500/lora/README.md

@@ -0,0 +1,206 @@
+---
+base_model: Qwen/Qwen3.6-27B
+library_name: peft
+tags:
+- base_model:adapter:Qwen/Qwen3.6-27B
+- lora
+- transformers
+---
+
+# 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 0.19.1

ab4/inverter/ao.ab4.mean.frombase.e1cend/step2500/lora/adapter_config.json

@@ -0,0 +1,56 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForCausalLM",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5"
+  },
+  "base_model_name_or_path": "Qwen/Qwen3.6-27B",
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.19.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "o_proj",
+    "in_proj_qkv",
+    "q_proj",
+    "v_proj",
+    "out_proj",
+    "in_proj_z",
+    "k_proj",
+    "down_proj",
+    "in_proj_b",
+    "in_proj_a",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend/step2500/meta.json

@@ -0,0 +1,104 @@
+{
+  "micro_steps": 2500,
+  "tokens_span": 45751296.0,
+  "tokens_sup": 48311296.0,
+  "examples_seen": 2560000,
+  "val_ce": 1.287048101425171,
+  "val_ce_L20": 1.2432713508605957,
+  "val_ce_L24": 1.2799700498580933,
+  "val_ce_L28": 1.196609616279602,
+  "val_ce_L32": 1.3380208015441895,
+  "val_ce_L36": 1.1756632328033447,
+  "val_ce_L40": 1.259678840637207,
+  "val_ce_L44": 1.2235233783721924,
+  "val_ce_L48": 1.2836198806762695,
+  "val_ce_L52": 1.2515300512313843,
+  "val_ce_L56": 1.4206113815307617,
+  "val_ce_L60": 1.507584571838379,
+  "train_ce": 0.9606544596254826,
+  "train_val_gap": 0.3263936417996883,
+  "config": {
+    "run_name": "ao.ab4.mean.frombase.e1cend",
+    "transform": "scaled",
+    "layers": [
+      20,
+      24,
+      28,
+      32,
+      36,
+      40,
+      44,
+      48,
+      52,
+      56,
+      60
+    ],
+    "alpha": 16000.0,
+    "clip_mult": 4.0,
+    "scales": {
+      "20": 115.07945174569515,
+      "24": 115.07945174569515,
+      "28": 115.07945174569515,
+      "32": 115.07945174569515,
+      "36": 115.07945174569515,
+      "40": 115.07945174569515,
+      "44": 115.07945174569515,
+      "48": 115.07945174569515,
+      "52": 115.07945174569515,
+      "56": 115.07945174569515,
+      "60": 115.07945174569515
+    },
+    "freeze_base": false,
+    "xform_hidden": 0,
+    "xform_arch": "resmlp",
+    "logits_on_targets": false,
+    "prefix_cache": false,
+    "n_examples": 3333472,
+    "min_len": 1,
+    "max_len": 64,
+    "lora_r": 16,
+    "lora_alpha": 32,
+    "lora_dropout": 0.05,
+    "lr": 0.0003,
+    "lr_sched": "constant",
+    "micro_batch": 128,
+    "val_micro_batch": 16,
+    "grad_accum": 8,
+    "epochs": 1,
+    "eval_every_steps": 500,
+    "max_eval_rows": 1024,
+    "warmup_steps": 100,
+    "seed": 0,
+    "init_from": "",
+    "skip_examples": 0,
+    "skip_batches": 0,
+    "exclude_idx_file": "",
+    "num_workers": 2,
+    "extra": {
+      "kind": "ao_ladder",
+      "ar_run": "ar.ab4.mean.e1c/end",
+      "target_prefix": "",
+      "global_scale": "115.07945174569515",
+      "pool": "ao_pool/pool_iolens_u64w2.safetensors",
+      "scale_mode": "global_frozen",
+      "grad_ckpt": "on",
+      "compile_blocks": "off",
+      "prompt": "continuation_raw",
+      "layers_per_crop": "4",
+      "val_source": "pool",
+      "max_len_filter": "32",
+      "adopt_arout_picks": "False",
+      "split_seed": "1234",
+      "arout_dir": "/scratch/ola/ao_arout_u64w2/ar.ab4.mean.e1c/end",
+      "val_pool": "ao_pool/pool_iolens_u64.safetensors",
+      "val_arout_dir": "ao_arout_u64/ar.ab4.mean.e1c/end"
+    },
+    "exact_resume": false,
+    "decay_frac": 0.1,
+    "warmup_examples": 0,
+    "expected_effective_batch": 0,
+    "eval_every_examples": 0,
+    "train_examples_cap": 0,
+    "total_opt_steps_override": null
+  }
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend/step3000/lora/README.md

@@ -0,0 +1,206 @@
+---
+base_model: Qwen/Qwen3.6-27B
+library_name: peft
+tags:
+- base_model:adapter:Qwen/Qwen3.6-27B
+- lora
+- transformers
+---
+
+# 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 0.19.1

ab4/inverter/ao.ab4.mean.frombase.e1cend/step3000/lora/adapter_config.json

@@ -0,0 +1,56 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForCausalLM",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5"
+  },
+  "base_model_name_or_path": "Qwen/Qwen3.6-27B",
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.19.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "o_proj",
+    "in_proj_qkv",
+    "q_proj",
+    "v_proj",
+    "out_proj",
+    "in_proj_z",
+    "k_proj",
+    "down_proj",
+    "in_proj_b",
+    "in_proj_a",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend/step3000/meta.json

@@ -0,0 +1,104 @@
+{
+  "micro_steps": 3000,
+  "tokens_span": 55110656.0,
+  "tokens_sup": 58182656.0,
+  "examples_seen": 3072000,
+  "val_ce": 1.2937355041503906,
+  "val_ce_L20": 1.2337371110916138,
+  "val_ce_L24": 1.2826884984970093,
+  "val_ce_L28": 1.2169009447097778,
+  "val_ce_L32": 1.3512821197509766,
+  "val_ce_L36": 1.2057968378067017,
+  "val_ce_L40": 1.2689497470855713,
+  "val_ce_L44": 1.1988202333450317,
+  "val_ce_L48": 1.3063496351242065,
+  "val_ce_L52": 1.268314003944397,
+  "val_ce_L56": 1.413094162940979,
+  "val_ce_L60": 1.5060462951660156,
+  "train_ce": 0.9122910050153732,
+  "train_val_gap": 0.38144449913501743,
+  "config": {
+    "run_name": "ao.ab4.mean.frombase.e1cend",
+    "transform": "scaled",
+    "layers": [
+      20,
+      24,
+      28,
+      32,
+      36,
+      40,
+      44,
+      48,
+      52,
+      56,
+      60
+    ],
+    "alpha": 16000.0,
+    "clip_mult": 4.0,
+    "scales": {
+      "20": 115.07945174569515,
+      "24": 115.07945174569515,
+      "28": 115.07945174569515,
+      "32": 115.07945174569515,
+      "36": 115.07945174569515,
+      "40": 115.07945174569515,
+      "44": 115.07945174569515,
+      "48": 115.07945174569515,
+      "52": 115.07945174569515,
+      "56": 115.07945174569515,
+      "60": 115.07945174569515
+    },
+    "freeze_base": false,
+    "xform_hidden": 0,
+    "xform_arch": "resmlp",
+    "logits_on_targets": false,
+    "prefix_cache": false,
+    "n_examples": 3333472,
+    "min_len": 1,
+    "max_len": 64,
+    "lora_r": 16,
+    "lora_alpha": 32,
+    "lora_dropout": 0.05,
+    "lr": 0.0003,
+    "lr_sched": "constant",
+    "micro_batch": 128,
+    "val_micro_batch": 16,
+    "grad_accum": 8,
+    "epochs": 1,
+    "eval_every_steps": 500,
+    "max_eval_rows": 1024,
+    "warmup_steps": 100,
+    "seed": 0,
+    "init_from": "",
+    "skip_examples": 0,
+    "skip_batches": 0,
+    "exclude_idx_file": "",
+    "num_workers": 2,
+    "extra": {
+      "kind": "ao_ladder",
+      "ar_run": "ar.ab4.mean.e1c/end",
+      "target_prefix": "",
+      "global_scale": "115.07945174569515",
+      "pool": "ao_pool/pool_iolens_u64w2.safetensors",
+      "scale_mode": "global_frozen",
+      "grad_ckpt": "on",
+      "compile_blocks": "off",
+      "prompt": "continuation_raw",
+      "layers_per_crop": "4",
+      "val_source": "pool",
+      "max_len_filter": "32",
+      "adopt_arout_picks": "False",
+      "split_seed": "1234",
+      "arout_dir": "/scratch/ola/ao_arout_u64w2/ar.ab4.mean.e1c/end",
+      "val_pool": "ao_pool/pool_iolens_u64.safetensors",
+      "val_arout_dir": "ao_arout_u64/ar.ab4.mean.e1c/end"
+    },
+    "exact_resume": false,
+    "decay_frac": 0.1,
+    "warmup_examples": 0,
+    "expected_effective_batch": 0,
+    "eval_every_examples": 0,
+    "train_examples_cap": 0,
+    "total_opt_steps_override": null
+  }
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend/step3221/lora/README.md

@@ -0,0 +1,206 @@
+---
+base_model: Qwen/Qwen3.6-27B
+library_name: peft
+tags:
+- base_model:adapter:Qwen/Qwen3.6-27B
+- lora
+- transformers
+---
+
+# 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 0.19.1

ab4/inverter/ao.ab4.mean.frombase.e1cend/step3221/lora/adapter_config.json

@@ -0,0 +1,56 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForCausalLM",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5"
+  },
+  "base_model_name_or_path": "Qwen/Qwen3.6-27B",
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.19.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "o_proj",
+    "in_proj_qkv",
+    "q_proj",
+    "v_proj",
+    "out_proj",
+    "in_proj_z",
+    "k_proj",
+    "down_proj",
+    "in_proj_b",
+    "in_proj_a",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend/step3221/meta.json

@@ -0,0 +1,104 @@
+{
+  "micro_steps": 3221,
+  "tokens_span": 58909696.0,
+  "tokens_sup": 62208000.0,
+  "examples_seen": 3298304,
+  "val_ce": 1.3266960382461548,
+  "val_ce_L20": 1.250138282775879,
+  "val_ce_L24": 1.3243069648742676,
+  "val_ce_L28": 1.2651358842849731,
+  "val_ce_L32": 1.379777431488037,
+  "val_ce_L36": 1.210906744003296,
+  "val_ce_L40": 1.2793513536453247,
+  "val_ce_L44": 1.2760872840881348,
+  "val_ce_L48": 1.3282365798950195,
+  "val_ce_L52": 1.2896722555160522,
+  "val_ce_L56": 1.4525333642959595,
+  "val_ce_L60": 1.564955711364746,
+  "train_ce": 0.8271289839329223,
+  "train_val_gap": 0.49956705431323245,
+  "config": {
+    "run_name": "ao.ab4.mean.frombase.e1cend",
+    "transform": "scaled",
+    "layers": [
+      20,
+      24,
+      28,
+      32,
+      36,
+      40,
+      44,
+      48,
+      52,
+      56,
+      60
+    ],
+    "alpha": 16000.0,
+    "clip_mult": 4.0,
+    "scales": {
+      "20": 115.07945174569515,
+      "24": 115.07945174569515,
+      "28": 115.07945174569515,
+      "32": 115.07945174569515,
+      "36": 115.07945174569515,
+      "40": 115.07945174569515,
+      "44": 115.07945174569515,
+      "48": 115.07945174569515,
+      "52": 115.07945174569515,
+      "56": 115.07945174569515,
+      "60": 115.07945174569515
+    },
+    "freeze_base": false,
+    "xform_hidden": 0,
+    "xform_arch": "resmlp",
+    "logits_on_targets": false,
+    "prefix_cache": false,
+    "n_examples": 3333472,
+    "min_len": 1,
+    "max_len": 64,
+    "lora_r": 16,
+    "lora_alpha": 32,
+    "lora_dropout": 0.05,
+    "lr": 0.0003,
+    "lr_sched": "constant",
+    "micro_batch": 128,
+    "val_micro_batch": 16,
+    "grad_accum": 8,
+    "epochs": 1,
+    "eval_every_steps": 500,
+    "max_eval_rows": 1024,
+    "warmup_steps": 100,
+    "seed": 0,
+    "init_from": "",
+    "skip_examples": 0,
+    "skip_batches": 0,
+    "exclude_idx_file": "",
+    "num_workers": 2,
+    "extra": {
+      "kind": "ao_ladder",
+      "ar_run": "ar.ab4.mean.e1c/end",
+      "target_prefix": "",
+      "global_scale": "115.07945174569515",
+      "pool": "ao_pool/pool_iolens_u64w2.safetensors",
+      "scale_mode": "global_frozen",
+      "grad_ckpt": "on",
+      "compile_blocks": "off",
+      "prompt": "continuation_raw",
+      "layers_per_crop": "4",
+      "val_source": "pool",
+      "max_len_filter": "32",
+      "adopt_arout_picks": "False",
+      "split_seed": "1234",
+      "arout_dir": "/scratch/ola/ao_arout_u64w2/ar.ab4.mean.e1c/end",
+      "val_pool": "ao_pool/pool_iolens_u64.safetensors",
+      "val_arout_dir": "ao_arout_u64/ar.ab4.mean.e1c/end"
+    },
+    "exact_resume": false,
+    "decay_frac": 0.1,
+    "warmup_examples": 0,
+    "expected_effective_batch": 0,
+    "eval_every_examples": 0,
+    "train_examples_cap": 0,
+    "total_opt_steps_override": null
+  }
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend/step500/lora/README.md

@@ -0,0 +1,206 @@
+---
+base_model: Qwen/Qwen3.6-27B
+library_name: peft
+tags:
+- base_model:adapter:Qwen/Qwen3.6-27B
+- lora
+- transformers
+---
+
+# 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 0.19.1

ab4/inverter/ao.ab4.mean.frombase.e1cend/step500/lora/adapter_config.json

@@ -0,0 +1,56 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForCausalLM",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5"
+  },
+  "base_model_name_or_path": "Qwen/Qwen3.6-27B",
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.19.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "o_proj",
+    "in_proj_qkv",
+    "q_proj",
+    "v_proj",
+    "out_proj",
+    "in_proj_z",
+    "k_proj",
+    "down_proj",
+    "in_proj_b",
+    "in_proj_a",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend/step500/meta.json

@@ -0,0 +1,104 @@
+{
+  "micro_steps": 500,
+  "tokens_span": 9432064.0,
+  "tokens_sup": 9944064.0,
+  "examples_seen": 512000,
+  "val_ce": 1.4270954132080078,
+  "val_ce_L20": 1.4415392875671387,
+  "val_ce_L24": 1.4663161039352417,
+  "val_ce_L28": 1.4298101663589478,
+  "val_ce_L32": 1.4932904243469238,
+  "val_ce_L36": 1.347434639930725,
+  "val_ce_L40": 1.340230107307434,
+  "val_ce_L44": 1.3578640222549438,
+  "val_ce_L48": 1.36846923828125,
+  "val_ce_L52": 1.3750728368759155,
+  "val_ce_L56": 1.5103765726089478,
+  "val_ce_L60": 1.585629940032959,
+  "train_ce": 1.6530334638357163,
+  "train_val_gap": -0.22593805062770844,
+  "config": {
+    "run_name": "ao.ab4.mean.frombase.e1cend",
+    "transform": "scaled",
+    "layers": [
+      20,
+      24,
+      28,
+      32,
+      36,
+      40,
+      44,
+      48,
+      52,
+      56,
+      60
+    ],
+    "alpha": 16000.0,
+    "clip_mult": 4.0,
+    "scales": {
+      "20": 115.07945174569515,
+      "24": 115.07945174569515,
+      "28": 115.07945174569515,
+      "32": 115.07945174569515,
+      "36": 115.07945174569515,
+      "40": 115.07945174569515,
+      "44": 115.07945174569515,
+      "48": 115.07945174569515,
+      "52": 115.07945174569515,
+      "56": 115.07945174569515,
+      "60": 115.07945174569515
+    },
+    "freeze_base": false,
+    "xform_hidden": 0,
+    "xform_arch": "resmlp",
+    "logits_on_targets": false,
+    "prefix_cache": false,
+    "n_examples": 3333472,
+    "min_len": 1,
+    "max_len": 64,
+    "lora_r": 16,
+    "lora_alpha": 32,
+    "lora_dropout": 0.05,
+    "lr": 0.0003,
+    "lr_sched": "constant",
+    "micro_batch": 128,
+    "val_micro_batch": 16,
+    "grad_accum": 8,
+    "epochs": 1,
+    "eval_every_steps": 500,
+    "max_eval_rows": 1024,
+    "warmup_steps": 100,
+    "seed": 0,
+    "init_from": "",
+    "skip_examples": 0,
+    "skip_batches": 0,
+    "exclude_idx_file": "",
+    "num_workers": 2,
+    "extra": {
+      "kind": "ao_ladder",
+      "ar_run": "ar.ab4.mean.e1c/end",
+      "target_prefix": "",
+      "global_scale": "115.07945174569515",
+      "pool": "ao_pool/pool_iolens_u64w2.safetensors",
+      "scale_mode": "global_frozen",
+      "grad_ckpt": "on",
+      "compile_blocks": "off",
+      "prompt": "continuation_raw",
+      "layers_per_crop": "4",
+      "val_source": "pool",
+      "max_len_filter": "32",
+      "adopt_arout_picks": "False",
+      "split_seed": "1234",
+      "arout_dir": "/scratch/ola/ao_arout_u64w2/ar.ab4.mean.e1c/end",
+      "val_pool": "ao_pool/pool_iolens_u64.safetensors",
+      "val_arout_dir": "ao_arout_u64/ar.ab4.mean.e1c/end"
+    },
+    "exact_resume": false,
+    "decay_frac": 0.1,
+    "warmup_examples": 0,
+    "expected_effective_batch": 0,
+    "eval_every_examples": 0,
+    "train_examples_cap": 0,
+    "total_opt_steps_override": null
+  }
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step1000/lora/README.md

@@ -0,0 +1,206 @@
+---
+base_model: Qwen/Qwen3.6-27B
+library_name: peft
+tags:
+- base_model:adapter:Qwen/Qwen3.6-27B
+- lora
+- transformers
+---
+
+# 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 0.19.1

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step1000/lora/adapter_config.json

@@ -0,0 +1,56 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForCausalLM",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5"
+  },
+  "base_model_name_or_path": "Qwen/Qwen3.6-27B",
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.19.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "o_proj",
+    "in_proj_qkv",
+    "q_proj",
+    "v_proj",
+    "out_proj",
+    "in_proj_z",
+    "k_proj",
+    "down_proj",
+    "in_proj_b",
+    "in_proj_a",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step1000/meta.json

@@ -0,0 +1,104 @@
+{
+  "micro_steps": 1000,
+  "tokens_span": 18199552.0,
+  "tokens_sup": 19223552.0,
+  "examples_seen": 1024000,
+  "val_ce": 1.2794396877288818,
+  "val_ce_L20": 1.2609106302261353,
+  "val_ce_L24": 1.2861063480377197,
+  "val_ce_L28": 1.2240310907363892,
+  "val_ce_L32": 1.347455620765686,
+  "val_ce_L36": 1.1587861776351929,
+  "val_ce_L40": 1.2171308994293213,
+  "val_ce_L44": 1.2146910429000854,
+  "val_ce_L48": 1.2505067586898804,
+  "val_ce_L52": 1.26174795627594,
+  "val_ce_L56": 1.4096708297729492,
+  "val_ce_L60": 1.4684720039367676,
+  "train_ce": 1.2844150429964065,
+  "train_val_gap": -0.004975355267524684,
+  "config": {
+    "run_name": "ao.ab4.mean.frombase.e1cend_full",
+    "transform": "scaled",
+    "layers": [
+      20,
+      24,
+      28,
+      32,
+      36,
+      40,
+      44,
+      48,
+      52,
+      56,
+      60
+    ],
+    "alpha": 16000.0,
+    "clip_mult": 4.0,
+    "scales": {
+      "20": 115.07945174569515,
+      "24": 115.07945174569515,
+      "28": 115.07945174569515,
+      "32": 115.07945174569515,
+      "36": 115.07945174569515,
+      "40": 115.07945174569515,
+      "44": 115.07945174569515,
+      "48": 115.07945174569515,
+      "52": 115.07945174569515,
+      "56": 115.07945174569515,
+      "60": 115.07945174569515
+    },
+    "freeze_base": false,
+    "xform_hidden": 0,
+    "xform_arch": "resmlp",
+    "logits_on_targets": false,
+    "prefix_cache": false,
+    "n_examples": 10891600,
+    "min_len": 1,
+    "max_len": 64,
+    "lora_r": 16,
+    "lora_alpha": 32,
+    "lora_dropout": 0.05,
+    "lr": 0.0003,
+    "lr_sched": "constant",
+    "micro_batch": 128,
+    "val_micro_batch": 16,
+    "grad_accum": 8,
+    "epochs": 1,
+    "eval_every_steps": 500,
+    "max_eval_rows": 1024,
+    "warmup_steps": 100,
+    "seed": 0,
+    "init_from": "",
+    "skip_examples": 0,
+    "skip_batches": 0,
+    "exclude_idx_file": "",
+    "num_workers": 2,
+    "extra": {
+      "kind": "ao_ladder",
+      "ar_run": "ar.ab4.mean.e1c/end",
+      "target_prefix": "",
+      "global_scale": "115.07945174569515",
+      "pool": "ao_pool/pool_iolens_u64w2.safetensors",
+      "scale_mode": "global_frozen",
+      "grad_ckpt": "on",
+      "compile_blocks": "off",
+      "prompt": "continuation_raw",
+      "layers_per_crop": "4",
+      "val_source": "pool",
+      "max_len_filter": "32",
+      "adopt_arout_picks": "False",
+      "split_seed": "1234",
+      "arout_dir": "/scratch/ola/ao_arout_u64w2_n9579668/ar.ab4.mean.e1c/end",
+      "val_pool": "ao_pool/pool_iolens_u64.safetensors",
+      "val_arout_dir": "ao_arout_u64/ar.ab4.mean.e1c/end"
+    },
+    "exact_resume": false,
+    "decay_frac": 0.1,
+    "warmup_examples": 0,
+    "expected_effective_batch": 0,
+    "eval_every_examples": 0,
+    "train_examples_cap": 0,
+    "total_opt_steps_override": null
+  }
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step10000/lora/README.md

@@ -0,0 +1,206 @@
+---
+base_model: Qwen/Qwen3.6-27B
+library_name: peft
+tags:
+- base_model:adapter:Qwen/Qwen3.6-27B
+- lora
+- transformers
+---
+
+# 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 0.19.1

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step10000/lora/adapter_config.json

@@ -0,0 +1,56 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForCausalLM",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5"
+  },
+  "base_model_name_or_path": "Qwen/Qwen3.6-27B",
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.19.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "o_proj",
+    "in_proj_qkv",
+    "q_proj",
+    "v_proj",
+    "out_proj",
+    "in_proj_z",
+    "k_proj",
+    "down_proj",
+    "in_proj_b",
+    "in_proj_a",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step10000/meta.json

@@ -0,0 +1,104 @@
+{
+  "micro_steps": 10000,
+  "tokens_span": 182770688.0,
+  "tokens_sup": 193010688.0,
+  "examples_seen": 10240000,
+  "val_ce": 1.147078514099121,
+  "val_ce_L20": 1.057914137840271,
+  "val_ce_L24": 1.1253106594085693,
+  "val_ce_L28": 1.0774312019348145,
+  "val_ce_L32": 1.1754080057144165,
+  "val_ce_L36": 1.0492713451385498,
+  "val_ce_L40": 1.120781660079956,
+  "val_ce_L44": 1.0773696899414062,
+  "val_ce_L48": 1.172749638557434,
+  "val_ce_L52": 1.1298480033874512,
+  "val_ce_L56": 1.2743521928787231,
+  "val_ce_L60": 1.382391333580017,
+  "train_ce": 0.8972788824960589,
+  "train_val_gap": 0.24979963160306218,
+  "config": {
+    "run_name": "ao.ab4.mean.frombase.e1cend_full",
+    "transform": "scaled",
+    "layers": [
+      20,
+      24,
+      28,
+      32,
+      36,
+      40,
+      44,
+      48,
+      52,
+      56,
+      60
+    ],
+    "alpha": 16000.0,
+    "clip_mult": 4.0,
+    "scales": {
+      "20": 115.07945174569515,
+      "24": 115.07945174569515,
+      "28": 115.07945174569515,
+      "32": 115.07945174569515,
+      "36": 115.07945174569515,
+      "40": 115.07945174569515,
+      "44": 115.07945174569515,
+      "48": 115.07945174569515,
+      "52": 115.07945174569515,
+      "56": 115.07945174569515,
+      "60": 115.07945174569515
+    },
+    "freeze_base": false,
+    "xform_hidden": 0,
+    "xform_arch": "resmlp",
+    "logits_on_targets": false,
+    "prefix_cache": false,
+    "n_examples": 10891600,
+    "min_len": 1,
+    "max_len": 64,
+    "lora_r": 16,
+    "lora_alpha": 32,
+    "lora_dropout": 0.05,
+    "lr": 0.0003,
+    "lr_sched": "constant",
+    "micro_batch": 128,
+    "val_micro_batch": 16,
+    "grad_accum": 8,
+    "epochs": 1,
+    "eval_every_steps": 500,
+    "max_eval_rows": 1024,
+    "warmup_steps": 100,
+    "seed": 0,
+    "init_from": "",
+    "skip_examples": 0,
+    "skip_batches": 0,
+    "exclude_idx_file": "",
+    "num_workers": 2,
+    "extra": {
+      "kind": "ao_ladder",
+      "ar_run": "ar.ab4.mean.e1c/end",
+      "target_prefix": "",
+      "global_scale": "115.07945174569515",
+      "pool": "ao_pool/pool_iolens_u64w2.safetensors",
+      "scale_mode": "global_frozen",
+      "grad_ckpt": "on",
+      "compile_blocks": "off",
+      "prompt": "continuation_raw",
+      "layers_per_crop": "4",
+      "val_source": "pool",
+      "max_len_filter": "32",
+      "adopt_arout_picks": "False",
+      "split_seed": "1234",
+      "arout_dir": "/scratch/ola/ao_arout_u64w2_n9579668/ar.ab4.mean.e1c/end",
+      "val_pool": "ao_pool/pool_iolens_u64.safetensors",
+      "val_arout_dir": "ao_arout_u64/ar.ab4.mean.e1c/end"
+    },
+    "exact_resume": false,
+    "decay_frac": 0.1,
+    "warmup_examples": 0,
+    "expected_effective_batch": 0,
+    "eval_every_examples": 0,
+    "train_examples_cap": 0,
+    "total_opt_steps_override": null
+  }
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step10500/lora/README.md

@@ -0,0 +1,206 @@
+---
+base_model: Qwen/Qwen3.6-27B
+library_name: peft
+tags:
+- base_model:adapter:Qwen/Qwen3.6-27B
+- lora
+- transformers
+---
+
+# 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 0.19.1

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step10500/lora/adapter_config.json

@@ -0,0 +1,56 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForCausalLM",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5"
+  },
+  "base_model_name_or_path": "Qwen/Qwen3.6-27B",
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.19.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "o_proj",
+    "in_proj_qkv",
+    "q_proj",
+    "v_proj",
+    "out_proj",
+    "in_proj_z",
+    "k_proj",
+    "down_proj",
+    "in_proj_b",
+    "in_proj_a",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step10500/meta.json

@@ -0,0 +1,104 @@
+{
+  "micro_steps": 10500,
+  "tokens_span": 191866880.0,
+  "tokens_sup": 202618880.0,
+  "examples_seen": 10752000,
+  "val_ce": 1.1499695777893066,
+  "val_ce_L20": 1.0618640184402466,
+  "val_ce_L24": 1.1147408485412598,
+  "val_ce_L28": 1.0773769617080688,
+  "val_ce_L32": 1.180230736732483,
+  "val_ce_L36": 1.0658936500549316,
+  "val_ce_L40": 1.1348178386688232,
+  "val_ce_L44": 1.0911335945129395,
+  "val_ce_L48": 1.1644901037216187,
+  "val_ce_L52": 1.1320792436599731,
+  "val_ce_L56": 1.2735570669174194,
+  "val_ce_L60": 1.3724385499954224,
+  "train_ce": 0.8760272574573755,
+  "train_val_gap": 0.2739423203319311,
+  "config": {
+    "run_name": "ao.ab4.mean.frombase.e1cend_full",
+    "transform": "scaled",
+    "layers": [
+      20,
+      24,
+      28,
+      32,
+      36,
+      40,
+      44,
+      48,
+      52,
+      56,
+      60
+    ],
+    "alpha": 16000.0,
+    "clip_mult": 4.0,
+    "scales": {
+      "20": 115.07945174569515,
+      "24": 115.07945174569515,
+      "28": 115.07945174569515,
+      "32": 115.07945174569515,
+      "36": 115.07945174569515,
+      "40": 115.07945174569515,
+      "44": 115.07945174569515,
+      "48": 115.07945174569515,
+      "52": 115.07945174569515,
+      "56": 115.07945174569515,
+      "60": 115.07945174569515
+    },
+    "freeze_base": false,
+    "xform_hidden": 0,
+    "xform_arch": "resmlp",
+    "logits_on_targets": false,
+    "prefix_cache": false,
+    "n_examples": 10891600,
+    "min_len": 1,
+    "max_len": 64,
+    "lora_r": 16,
+    "lora_alpha": 32,
+    "lora_dropout": 0.05,
+    "lr": 0.0003,
+    "lr_sched": "constant",
+    "micro_batch": 128,
+    "val_micro_batch": 16,
+    "grad_accum": 8,
+    "epochs": 1,
+    "eval_every_steps": 500,
+    "max_eval_rows": 1024,
+    "warmup_steps": 100,
+    "seed": 0,
+    "init_from": "",
+    "skip_examples": 0,
+    "skip_batches": 0,
+    "exclude_idx_file": "",
+    "num_workers": 2,
+    "extra": {
+      "kind": "ao_ladder",
+      "ar_run": "ar.ab4.mean.e1c/end",
+      "target_prefix": "",
+      "global_scale": "115.07945174569515",
+      "pool": "ao_pool/pool_iolens_u64w2.safetensors",
+      "scale_mode": "global_frozen",
+      "grad_ckpt": "on",
+      "compile_blocks": "off",
+      "prompt": "continuation_raw",
+      "layers_per_crop": "4",
+      "val_source": "pool",
+      "max_len_filter": "32",
+      "adopt_arout_picks": "False",
+      "split_seed": "1234",
+      "arout_dir": "/scratch/ola/ao_arout_u64w2_n9579668/ar.ab4.mean.e1c/end",
+      "val_pool": "ao_pool/pool_iolens_u64.safetensors",
+      "val_arout_dir": "ao_arout_u64/ar.ab4.mean.e1c/end"
+    },
+    "exact_resume": false,
+    "decay_frac": 0.1,
+    "warmup_examples": 0,
+    "expected_effective_batch": 0,
+    "eval_every_examples": 0,
+    "train_examples_cap": 0,
+    "total_opt_steps_override": null
+  }
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step10603/lora/README.md

@@ -0,0 +1,206 @@
+---
+base_model: Qwen/Qwen3.6-27B
+library_name: peft
+tags:
+- base_model:adapter:Qwen/Qwen3.6-27B
+- lora
+- transformers
+---
+
+# 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 0.19.1

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step10603/lora/adapter_config.json

@@ -0,0 +1,56 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForCausalLM",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5"
+  },
+  "base_model_name_or_path": "Qwen/Qwen3.6-27B",
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.19.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "o_proj",
+    "in_proj_qkv",
+    "q_proj",
+    "v_proj",
+    "out_proj",
+    "in_proj_z",
+    "k_proj",
+    "down_proj",
+    "in_proj_b",
+    "in_proj_a",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step10603/meta.json

@@ -0,0 +1,104 @@
+{
+  "micro_steps": 10603,
+  "tokens_span": 193746944.0,
+  "tokens_sup": 204604416.0,
+  "examples_seen": 10857472,
+  "val_ce": 1.1377981901168823,
+  "val_ce_L20": 1.0405043363571167,
+  "val_ce_L24": 1.1026939153671265,
+  "val_ce_L28": 1.0527781248092651,
+  "val_ce_L32": 1.185759425163269,
+  "val_ce_L36": 1.0371209383010864,
+  "val_ce_L40": 1.1009219884872437,
+  "val_ce_L44": 1.0850839614868164,
+  "val_ce_L48": 1.1659104824066162,
+  "val_ce_L52": 1.1336084604263306,
+  "val_ce_L56": 1.2689943313598633,
+  "val_ce_L60": 1.3700145483016968,
+  "train_ce": 0.8688119102305579,
+  "train_val_gap": 0.2689862798863244,
+  "config": {
+    "run_name": "ao.ab4.mean.frombase.e1cend_full",
+    "transform": "scaled",
+    "layers": [
+      20,
+      24,
+      28,
+      32,
+      36,
+      40,
+      44,
+      48,
+      52,
+      56,
+      60
+    ],
+    "alpha": 16000.0,
+    "clip_mult": 4.0,
+    "scales": {
+      "20": 115.07945174569515,
+      "24": 115.07945174569515,
+      "28": 115.07945174569515,
+      "32": 115.07945174569515,
+      "36": 115.07945174569515,
+      "40": 115.07945174569515,
+      "44": 115.07945174569515,
+      "48": 115.07945174569515,
+      "52": 115.07945174569515,
+      "56": 115.07945174569515,
+      "60": 115.07945174569515
+    },
+    "freeze_base": false,
+    "xform_hidden": 0,
+    "xform_arch": "resmlp",
+    "logits_on_targets": false,
+    "prefix_cache": false,
+    "n_examples": 10891600,
+    "min_len": 1,
+    "max_len": 64,
+    "lora_r": 16,
+    "lora_alpha": 32,
+    "lora_dropout": 0.05,
+    "lr": 0.0003,
+    "lr_sched": "constant",
+    "micro_batch": 128,
+    "val_micro_batch": 16,
+    "grad_accum": 8,
+    "epochs": 1,
+    "eval_every_steps": 500,
+    "max_eval_rows": 1024,
+    "warmup_steps": 100,
+    "seed": 0,
+    "init_from": "",
+    "skip_examples": 0,
+    "skip_batches": 0,
+    "exclude_idx_file": "",
+    "num_workers": 2,
+    "extra": {
+      "kind": "ao_ladder",
+      "ar_run": "ar.ab4.mean.e1c/end",
+      "target_prefix": "",
+      "global_scale": "115.07945174569515",
+      "pool": "ao_pool/pool_iolens_u64w2.safetensors",
+      "scale_mode": "global_frozen",
+      "grad_ckpt": "on",
+      "compile_blocks": "off",
+      "prompt": "continuation_raw",
+      "layers_per_crop": "4",
+      "val_source": "pool",
+      "max_len_filter": "32",
+      "adopt_arout_picks": "False",
+      "split_seed": "1234",
+      "arout_dir": "/scratch/ola/ao_arout_u64w2_n9579668/ar.ab4.mean.e1c/end",
+      "val_pool": "ao_pool/pool_iolens_u64.safetensors",
+      "val_arout_dir": "ao_arout_u64/ar.ab4.mean.e1c/end"
+    },
+    "exact_resume": false,
+    "decay_frac": 0.1,
+    "warmup_examples": 0,
+    "expected_effective_batch": 0,
+    "eval_every_examples": 0,
+    "train_examples_cap": 0,
+    "total_opt_steps_override": null
+  }
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step1500/lora/README.md

@@ -0,0 +1,206 @@
+---
+base_model: Qwen/Qwen3.6-27B
+library_name: peft
+tags:
+- base_model:adapter:Qwen/Qwen3.6-27B
+- lora
+- transformers
+---
+
+# 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 0.19.1

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step1500/lora/adapter_config.json

@@ -0,0 +1,56 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForCausalLM",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5"
+  },
+  "base_model_name_or_path": "Qwen/Qwen3.6-27B",
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.19.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "o_proj",
+    "in_proj_qkv",
+    "q_proj",
+    "v_proj",
+    "out_proj",
+    "in_proj_z",
+    "k_proj",
+    "down_proj",
+    "in_proj_b",
+    "in_proj_a",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step1500/meta.json

@@ -0,0 +1,104 @@
+{
+  "micro_steps": 1500,
+  "tokens_span": 27310080.0,
+  "tokens_sup": 28846080.0,
+  "examples_seen": 1536000,
+  "val_ce": 1.2267848253250122,
+  "val_ce_L20": 1.1825789213180542,
+  "val_ce_L24": 1.208807349205017,
+  "val_ce_L28": 1.1780914068222046,
+  "val_ce_L32": 1.2686996459960938,
+  "val_ce_L36": 1.1071984767913818,
+  "val_ce_L40": 1.1559374332427979,
+  "val_ce_L44": 1.1627734899520874,
+  "val_ce_L48": 1.214462161064148,
+  "val_ce_L52": 1.224327802658081,
+  "val_ce_L56": 1.3841696977615356,
+  "val_ce_L60": 1.4372719526290894,
+  "train_ce": 1.2000047302246093,
+  "train_val_gap": 0.026780095100402912,
+  "config": {
+    "run_name": "ao.ab4.mean.frombase.e1cend_full",
+    "transform": "scaled",
+    "layers": [
+      20,
+      24,
+      28,
+      32,
+      36,
+      40,
+      44,
+      48,
+      52,
+      56,
+      60
+    ],
+    "alpha": 16000.0,
+    "clip_mult": 4.0,
+    "scales": {
+      "20": 115.07945174569515,
+      "24": 115.07945174569515,
+      "28": 115.07945174569515,
+      "32": 115.07945174569515,
+      "36": 115.07945174569515,
+      "40": 115.07945174569515,
+      "44": 115.07945174569515,
+      "48": 115.07945174569515,
+      "52": 115.07945174569515,
+      "56": 115.07945174569515,
+      "60": 115.07945174569515
+    },
+    "freeze_base": false,
+    "xform_hidden": 0,
+    "xform_arch": "resmlp",
+    "logits_on_targets": false,
+    "prefix_cache": false,
+    "n_examples": 10891600,
+    "min_len": 1,
+    "max_len": 64,
+    "lora_r": 16,
+    "lora_alpha": 32,
+    "lora_dropout": 0.05,
+    "lr": 0.0003,
+    "lr_sched": "constant",
+    "micro_batch": 128,
+    "val_micro_batch": 16,
+    "grad_accum": 8,
+    "epochs": 1,
+    "eval_every_steps": 500,
+    "max_eval_rows": 1024,
+    "warmup_steps": 100,
+    "seed": 0,
+    "init_from": "",
+    "skip_examples": 0,
+    "skip_batches": 0,
+    "exclude_idx_file": "",
+    "num_workers": 2,
+    "extra": {
+      "kind": "ao_ladder",
+      "ar_run": "ar.ab4.mean.e1c/end",
+      "target_prefix": "",
+      "global_scale": "115.07945174569515",
+      "pool": "ao_pool/pool_iolens_u64w2.safetensors",
+      "scale_mode": "global_frozen",
+      "grad_ckpt": "on",
+      "compile_blocks": "off",
+      "prompt": "continuation_raw",
+      "layers_per_crop": "4",
+      "val_source": "pool",
+      "max_len_filter": "32",
+      "adopt_arout_picks": "False",
+      "split_seed": "1234",
+      "arout_dir": "/scratch/ola/ao_arout_u64w2_n9579668/ar.ab4.mean.e1c/end",
+      "val_pool": "ao_pool/pool_iolens_u64.safetensors",
+      "val_arout_dir": "ao_arout_u64/ar.ab4.mean.e1c/end"
+    },
+    "exact_resume": false,
+    "decay_frac": 0.1,
+    "warmup_examples": 0,
+    "expected_effective_batch": 0,
+    "eval_every_examples": 0,
+    "train_examples_cap": 0,
+    "total_opt_steps_override": null
+  }
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step2000/lora/README.md

@@ -0,0 +1,206 @@
+---
+base_model: Qwen/Qwen3.6-27B
+library_name: peft
+tags:
+- base_model:adapter:Qwen/Qwen3.6-27B
+- lora
+- transformers
+---
+
+# 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 0.19.1

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step2000/lora/adapter_config.json

@@ -0,0 +1,56 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForCausalLM",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5"
+  },
+  "base_model_name_or_path": "Qwen/Qwen3.6-27B",
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.19.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "o_proj",
+    "in_proj_qkv",
+    "q_proj",
+    "v_proj",
+    "out_proj",
+    "in_proj_z",
+    "k_proj",
+    "down_proj",
+    "in_proj_b",
+    "in_proj_a",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step2000/meta.json

@@ -0,0 +1,104 @@
+{
+  "micro_steps": 2000,
+  "tokens_span": 36663296.0,
+  "tokens_sup": 38711296.0,
+  "examples_seen": 2048000,
+  "val_ce": 1.2051620483398438,
+  "val_ce_L20": 1.1682356595993042,
+  "val_ce_L24": 1.1859829425811768,
+  "val_ce_L28": 1.1621588468551636,
+  "val_ce_L32": 1.2459697723388672,
+  "val_ce_L36": 1.088072419166565,
+  "val_ce_L40": 1.1438509225845337,
+  "val_ce_L44": 1.1326831579208374,
+  "val_ce_L48": 1.2070726156234741,
+  "val_ce_L52": 1.1805998086929321,
+  "val_ce_L56": 1.3471125364303589,
+  "val_ce_L60": 1.4250956773757935,
+  "train_ce": 1.1669836622178555,
+  "train_val_gap": 0.038178386121988206,
+  "config": {
+    "run_name": "ao.ab4.mean.frombase.e1cend_full",
+    "transform": "scaled",
+    "layers": [
+      20,
+      24,
+      28,
+      32,
+      36,
+      40,
+      44,
+      48,
+      52,
+      56,
+      60
+    ],
+    "alpha": 16000.0,
+    "clip_mult": 4.0,
+    "scales": {
+      "20": 115.07945174569515,
+      "24": 115.07945174569515,
+      "28": 115.07945174569515,
+      "32": 115.07945174569515,
+      "36": 115.07945174569515,
+      "40": 115.07945174569515,
+      "44": 115.07945174569515,
+      "48": 115.07945174569515,
+      "52": 115.07945174569515,
+      "56": 115.07945174569515,
+      "60": 115.07945174569515
+    },
+    "freeze_base": false,
+    "xform_hidden": 0,
+    "xform_arch": "resmlp",
+    "logits_on_targets": false,
+    "prefix_cache": false,
+    "n_examples": 10891600,
+    "min_len": 1,
+    "max_len": 64,
+    "lora_r": 16,
+    "lora_alpha": 32,
+    "lora_dropout": 0.05,
+    "lr": 0.0003,
+    "lr_sched": "constant",
+    "micro_batch": 128,
+    "val_micro_batch": 16,
+    "grad_accum": 8,
+    "epochs": 1,
+    "eval_every_steps": 500,
+    "max_eval_rows": 1024,
+    "warmup_steps": 100,
+    "seed": 0,
+    "init_from": "",
+    "skip_examples": 0,
+    "skip_batches": 0,
+    "exclude_idx_file": "",
+    "num_workers": 2,
+    "extra": {
+      "kind": "ao_ladder",
+      "ar_run": "ar.ab4.mean.e1c/end",
+      "target_prefix": "",
+      "global_scale": "115.07945174569515",
+      "pool": "ao_pool/pool_iolens_u64w2.safetensors",
+      "scale_mode": "global_frozen",
+      "grad_ckpt": "on",
+      "compile_blocks": "off",
+      "prompt": "continuation_raw",
+      "layers_per_crop": "4",
+      "val_source": "pool",
+      "max_len_filter": "32",
+      "adopt_arout_picks": "False",
+      "split_seed": "1234",
+      "arout_dir": "/scratch/ola/ao_arout_u64w2_n9579668/ar.ab4.mean.e1c/end",
+      "val_pool": "ao_pool/pool_iolens_u64.safetensors",
+      "val_arout_dir": "ao_arout_u64/ar.ab4.mean.e1c/end"
+    },
+    "exact_resume": false,
+    "decay_frac": 0.1,
+    "warmup_examples": 0,
+    "expected_effective_batch": 0,
+    "eval_every_examples": 0,
+    "train_examples_cap": 0,
+    "total_opt_steps_override": null
+  }
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step2500/lora/README.md

@@ -0,0 +1,206 @@
+---
+base_model: Qwen/Qwen3.6-27B
+library_name: peft
+tags:
+- base_model:adapter:Qwen/Qwen3.6-27B
+- lora
+- transformers
+---
+
+# 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 0.19.1

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step2500/lora/adapter_config.json

@@ -0,0 +1,56 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForCausalLM",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5"
+  },
+  "base_model_name_or_path": "Qwen/Qwen3.6-27B",
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.19.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "o_proj",
+    "in_proj_qkv",
+    "q_proj",
+    "v_proj",
+    "out_proj",
+    "in_proj_z",
+    "k_proj",
+    "down_proj",
+    "in_proj_b",
+    "in_proj_a",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step2500/meta.json

@@ -0,0 +1,104 @@
+{
+  "micro_steps": 2500,
+  "tokens_span": 45873152.0,
+  "tokens_sup": 48433152.0,
+  "examples_seen": 2560000,
+  "val_ce": 1.2405095100402832,
+  "val_ce_L20": 1.214521050453186,
+  "val_ce_L24": 1.2365515232086182,
+  "val_ce_L28": 1.1566509008407593,
+  "val_ce_L32": 1.2783430814743042,
+  "val_ce_L36": 1.1128908395767212,
+  "val_ce_L40": 1.1881002187728882,
+  "val_ce_L44": 1.1614257097244263,
+  "val_ce_L48": 1.222954511642456,
+  "val_ce_L52": 1.2283871173858643,
+  "val_ce_L56": 1.3860965967178345,
+  "val_ce_L60": 1.4867610931396484,
+  "train_ce": 1.1127116147279739,
+  "train_val_gap": 0.12779789531230934,
+  "config": {
+    "run_name": "ao.ab4.mean.frombase.e1cend_full",
+    "transform": "scaled",
+    "layers": [
+      20,
+      24,
+      28,
+      32,
+      36,
+      40,
+      44,
+      48,
+      52,
+      56,
+      60
+    ],
+    "alpha": 16000.0,
+    "clip_mult": 4.0,
+    "scales": {
+      "20": 115.07945174569515,
+      "24": 115.07945174569515,
+      "28": 115.07945174569515,
+      "32": 115.07945174569515,
+      "36": 115.07945174569515,
+      "40": 115.07945174569515,
+      "44": 115.07945174569515,
+      "48": 115.07945174569515,
+      "52": 115.07945174569515,
+      "56": 115.07945174569515,
+      "60": 115.07945174569515
+    },
+    "freeze_base": false,
+    "xform_hidden": 0,
+    "xform_arch": "resmlp",
+    "logits_on_targets": false,
+    "prefix_cache": false,
+    "n_examples": 10891600,
+    "min_len": 1,
+    "max_len": 64,
+    "lora_r": 16,
+    "lora_alpha": 32,
+    "lora_dropout": 0.05,
+    "lr": 0.0003,
+    "lr_sched": "constant",
+    "micro_batch": 128,
+    "val_micro_batch": 16,
+    "grad_accum": 8,
+    "epochs": 1,
+    "eval_every_steps": 500,
+    "max_eval_rows": 1024,
+    "warmup_steps": 100,
+    "seed": 0,
+    "init_from": "",
+    "skip_examples": 0,
+    "skip_batches": 0,
+    "exclude_idx_file": "",
+    "num_workers": 2,
+    "extra": {
+      "kind": "ao_ladder",
+      "ar_run": "ar.ab4.mean.e1c/end",
+      "target_prefix": "",
+      "global_scale": "115.07945174569515",
+      "pool": "ao_pool/pool_iolens_u64w2.safetensors",
+      "scale_mode": "global_frozen",
+      "grad_ckpt": "on",
+      "compile_blocks": "off",
+      "prompt": "continuation_raw",
+      "layers_per_crop": "4",
+      "val_source": "pool",
+      "max_len_filter": "32",
+      "adopt_arout_picks": "False",
+      "split_seed": "1234",
+      "arout_dir": "/scratch/ola/ao_arout_u64w2_n9579668/ar.ab4.mean.e1c/end",
+      "val_pool": "ao_pool/pool_iolens_u64.safetensors",
+      "val_arout_dir": "ao_arout_u64/ar.ab4.mean.e1c/end"
+    },
+    "exact_resume": false,
+    "decay_frac": 0.1,
+    "warmup_examples": 0,
+    "expected_effective_batch": 0,
+    "eval_every_examples": 0,
+    "train_examples_cap": 0,
+    "total_opt_steps_override": null
+  }
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step3000/lora/README.md

@@ -0,0 +1,206 @@
+---
+base_model: Qwen/Qwen3.6-27B
+library_name: peft
+tags:
+- base_model:adapter:Qwen/Qwen3.6-27B
+- lora
+- transformers
+---
+
+# 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 0.19.1

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step3000/lora/adapter_config.json

@@ -0,0 +1,56 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForCausalLM",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5"
+  },
+  "base_model_name_or_path": "Qwen/Qwen3.6-27B",
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.19.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "o_proj",
+    "in_proj_qkv",
+    "q_proj",
+    "v_proj",
+    "out_proj",
+    "in_proj_z",
+    "k_proj",
+    "down_proj",
+    "in_proj_b",
+    "in_proj_a",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step3000/meta.json

@@ -0,0 +1,104 @@
+{
+  "micro_steps": 3000,
+  "tokens_span": 55260160.0,
+  "tokens_sup": 58332160.0,
+  "examples_seen": 3072000,
+  "val_ce": 1.1841005086898804,
+  "val_ce_L20": 1.1110998392105103,
+  "val_ce_L24": 1.1693896055221558,
+  "val_ce_L28": 1.1327086687088013,
+  "val_ce_L32": 1.2244523763656616,
+  "val_ce_L36": 1.0591293573379517,
+  "val_ce_L40": 1.1514557600021362,
+  "val_ce_L44": 1.0979892015457153,
+  "val_ce_L48": 1.2017370462417603,
+  "val_ce_L52": 1.1796592473983765,
+  "val_ce_L56": 1.326141595840454,
+  "val_ce_L60": 1.4005502462387085,
+  "train_ce": 1.088088481605053,
+  "train_val_gap": 0.09601202708482748,
+  "config": {
+    "run_name": "ao.ab4.mean.frombase.e1cend_full",
+    "transform": "scaled",
+    "layers": [
+      20,
+      24,
+      28,
+      32,
+      36,
+      40,
+      44,
+      48,
+      52,
+      56,
+      60
+    ],
+    "alpha": 16000.0,
+    "clip_mult": 4.0,
+    "scales": {
+      "20": 115.07945174569515,
+      "24": 115.07945174569515,
+      "28": 115.07945174569515,
+      "32": 115.07945174569515,
+      "36": 115.07945174569515,
+      "40": 115.07945174569515,
+      "44": 115.07945174569515,
+      "48": 115.07945174569515,
+      "52": 115.07945174569515,
+      "56": 115.07945174569515,
+      "60": 115.07945174569515
+    },
+    "freeze_base": false,
+    "xform_hidden": 0,
+    "xform_arch": "resmlp",
+    "logits_on_targets": false,
+    "prefix_cache": false,
+    "n_examples": 10891600,
+    "min_len": 1,
+    "max_len": 64,
+    "lora_r": 16,
+    "lora_alpha": 32,
+    "lora_dropout": 0.05,
+    "lr": 0.0003,
+    "lr_sched": "constant",
+    "micro_batch": 128,
+    "val_micro_batch": 16,
+    "grad_accum": 8,
+    "epochs": 1,
+    "eval_every_steps": 500,
+    "max_eval_rows": 1024,
+    "warmup_steps": 100,
+    "seed": 0,
+    "init_from": "",
+    "skip_examples": 0,
+    "skip_batches": 0,
+    "exclude_idx_file": "",
+    "num_workers": 2,
+    "extra": {
+      "kind": "ao_ladder",
+      "ar_run": "ar.ab4.mean.e1c/end",
+      "target_prefix": "",
+      "global_scale": "115.07945174569515",
+      "pool": "ao_pool/pool_iolens_u64w2.safetensors",
+      "scale_mode": "global_frozen",
+      "grad_ckpt": "on",
+      "compile_blocks": "off",
+      "prompt": "continuation_raw",
+      "layers_per_crop": "4",
+      "val_source": "pool",
+      "max_len_filter": "32",
+      "adopt_arout_picks": "False",
+      "split_seed": "1234",
+      "arout_dir": "/scratch/ola/ao_arout_u64w2_n9579668/ar.ab4.mean.e1c/end",
+      "val_pool": "ao_pool/pool_iolens_u64.safetensors",
+      "val_arout_dir": "ao_arout_u64/ar.ab4.mean.e1c/end"
+    },
+    "exact_resume": false,
+    "decay_frac": 0.1,
+    "warmup_examples": 0,
+    "expected_effective_batch": 0,
+    "eval_every_examples": 0,
+    "train_examples_cap": 0,
+    "total_opt_steps_override": null
+  }
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step3500/lora/README.md

@@ -0,0 +1,206 @@
+---
+base_model: Qwen/Qwen3.6-27B
+library_name: peft
+tags:
+- base_model:adapter:Qwen/Qwen3.6-27B
+- lora
+- transformers
+---
+
+# 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 0.19.1

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step3500/lora/adapter_config.json

@@ -0,0 +1,56 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForCausalLM",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5"
+  },
+  "base_model_name_or_path": "Qwen/Qwen3.6-27B",
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.19.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "o_proj",
+    "in_proj_qkv",
+    "q_proj",
+    "v_proj",
+    "out_proj",
+    "in_proj_z",
+    "k_proj",
+    "down_proj",
+    "in_proj_b",
+    "in_proj_a",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step3500/meta.json

@@ -0,0 +1,104 @@
+{
+  "micro_steps": 3500,
+  "tokens_span": 64128000.0,
+  "tokens_sup": 67712000.0,
+  "examples_seen": 3584000,
+  "val_ce": 1.1771403551101685,
+  "val_ce_L20": 1.1341434717178345,
+  "val_ce_L24": 1.1635295152664185,
+  "val_ce_L28": 1.1177664995193481,
+  "val_ce_L32": 1.203652024269104,
+  "val_ce_L36": 1.0604240894317627,
+  "val_ce_L40": 1.1395676136016846,
+  "val_ce_L44": 1.1139315366744995,
+  "val_ce_L48": 1.1645500659942627,
+  "val_ce_L52": 1.160385251045227,
+  "val_ce_L56": 1.3070547580718994,
+  "val_ce_L60": 1.4071545600891113,
+  "train_ce": 1.0349715248048306,
+  "train_val_gap": 0.1421688303053379,
+  "config": {
+    "run_name": "ao.ab4.mean.frombase.e1cend_full",
+    "transform": "scaled",
+    "layers": [
+      20,
+      24,
+      28,
+      32,
+      36,
+      40,
+      44,
+      48,
+      52,
+      56,
+      60
+    ],
+    "alpha": 16000.0,
+    "clip_mult": 4.0,
+    "scales": {
+      "20": 115.07945174569515,
+      "24": 115.07945174569515,
+      "28": 115.07945174569515,
+      "32": 115.07945174569515,
+      "36": 115.07945174569515,
+      "40": 115.07945174569515,
+      "44": 115.07945174569515,
+      "48": 115.07945174569515,
+      "52": 115.07945174569515,
+      "56": 115.07945174569515,
+      "60": 115.07945174569515
+    },
+    "freeze_base": false,
+    "xform_hidden": 0,
+    "xform_arch": "resmlp",
+    "logits_on_targets": false,
+    "prefix_cache": false,
+    "n_examples": 10891600,
+    "min_len": 1,
+    "max_len": 64,
+    "lora_r": 16,
+    "lora_alpha": 32,
+    "lora_dropout": 0.05,
+    "lr": 0.0003,
+    "lr_sched": "constant",
+    "micro_batch": 128,
+    "val_micro_batch": 16,
+    "grad_accum": 8,
+    "epochs": 1,
+    "eval_every_steps": 500,
+    "max_eval_rows": 1024,
+    "warmup_steps": 100,
+    "seed": 0,
+    "init_from": "",
+    "skip_examples": 0,
+    "skip_batches": 0,
+    "exclude_idx_file": "",
+    "num_workers": 2,
+    "extra": {
+      "kind": "ao_ladder",
+      "ar_run": "ar.ab4.mean.e1c/end",
+      "target_prefix": "",
+      "global_scale": "115.07945174569515",
+      "pool": "ao_pool/pool_iolens_u64w2.safetensors",
+      "scale_mode": "global_frozen",
+      "grad_ckpt": "on",
+      "compile_blocks": "off",
+      "prompt": "continuation_raw",
+      "layers_per_crop": "4",
+      "val_source": "pool",
+      "max_len_filter": "32",
+      "adopt_arout_picks": "False",
+      "split_seed": "1234",
+      "arout_dir": "/scratch/ola/ao_arout_u64w2_n9579668/ar.ab4.mean.e1c/end",
+      "val_pool": "ao_pool/pool_iolens_u64.safetensors",
+      "val_arout_dir": "ao_arout_u64/ar.ab4.mean.e1c/end"
+    },
+    "exact_resume": false,
+    "decay_frac": 0.1,
+    "warmup_examples": 0,
+    "expected_effective_batch": 0,
+    "eval_every_examples": 0,
+    "train_examples_cap": 0,
+    "total_opt_steps_override": null
+  }
+}

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step4000/lora/README.md

@@ -0,0 +1,206 @@
+---
+base_model: Qwen/Qwen3.6-27B
+library_name: peft
+tags:
+- base_model:adapter:Qwen/Qwen3.6-27B
+- lora
+- transformers
+---
+
+# 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 0.19.1

ab4/inverter/ao.ab4.mean.frombase.e1cend_full/step4000/lora/adapter_config.json

@@ -0,0 +1,56 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForCausalLM",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5"
+  },
+  "base_model_name_or_path": "Qwen/Qwen3.6-27B",
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.19.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "o_proj",
+    "in_proj_qkv",
+    "q_proj",
+    "v_proj",
+    "out_proj",
+    "in_proj_z",
+    "k_proj",
+    "down_proj",
+    "in_proj_b",
+    "in_proj_a",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}