Upload folder using huggingface_hub

qian_sg_backup/qwen3/qwen3_p109m_t1_8_suite-owt_scaling_v3_family-qwen3_geometry_id-euclid_variant-base_init_slope-std/attempt3_20260310_081106/README.md

@@ -0,0 +1,82 @@
+---
+library_name: transformers
+license: other
+base_model: /data/jrlin/HyperScale/.cache/model_specs/qwen3/qwen3/base/std/p109m
+tags:
+- llama-factory
+- full
+- generated_from_trainer
+datasets:
+- arrow
+model-index:
+- name: attempt3_20260310_081106
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# attempt3_20260310_081106
+
+This model is a fine-tuned version of [/data/jrlin/HyperScale/.cache/model_specs/qwen3/qwen3/base/std/p109m](https://huggingface.co//data/jrlin/HyperScale/.cache/model_specs/qwen3/qwen3/base/std/p109m) on the owt_local_train dataset.
+It achieves the following results on the evaluation set:
+- Loss: 4.5874
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0001
+- train_batch_size: 4
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 2
+- gradient_accumulation_steps: 16
+- total_train_batch_size: 128
+- total_eval_batch_size: 16
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.02
+- training_steps: 4003
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss |
+|:-------------:|:------:|:----:|:---------------:|
+| 7.2528        | 0.0077 | 256  | 6.4774          |
+| 5.9516        | 0.0154 | 512  | 5.8442          |
+| 5.6485        | 0.0231 | 768  | 5.5200          |
+| 5.3609        | 0.0308 | 1024 | 5.3005          |
+| 5.2063        | 0.0385 | 1280 | 5.1112          |
+| 5.0085        | 0.0462 | 1536 | 4.9671          |
+| 4.9145        | 0.0539 | 1792 | 4.8567          |
+| 4.8034        | 0.0616 | 2048 | 4.7742          |
+| 4.7276        | 0.0693 | 2304 | 4.7127          |
+| 4.6877        | 0.0770 | 2560 | 4.6659          |
+| 4.6423        | 0.0847 | 2816 | 4.6327          |
+| 4.618         | 0.0924 | 3072 | 4.6106          |
+| 4.6031        | 0.1001 | 3328 | 4.5971          |
+| 4.5861        | 0.1078 | 3584 | 4.5899          |
+| 4.591         | 0.1155 | 3840 | 4.5876          |
+
+
+### Framework versions
+
+- Transformers 4.51.3
+- Pytorch 2.5.1+cu124
+- Datasets 3.6.0
+- Tokenizers 0.21.4

qian_sg_backup/qwen3/qwen3_p109m_t1_8_suite-owt_scaling_v3_family-qwen3_geometry_id-euclid_variant-base_init_slope-std/attempt3_20260310_081106/added_tokens.json

@@ -0,0 +1,28 @@
+{
+  "</think>": 151668,
+  "</tool_call>": 151658,
+  "</tool_response>": 151666,
+  "<think>": 151667,
+  "<tool_call>": 151657,
+  "<tool_response>": 151665,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

qian_sg_backup/qwen3/qwen3_p109m_t1_8_suite-owt_scaling_v3_family-qwen3_geometry_id-euclid_variant-base_init_slope-std/attempt3_20260310_081106/all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 0.12042538469637942,
+    "eval_loss": 4.58743143081665,
+    "eval_perplexity": 98.2417650367421,
+    "eval_runtime": 11.4076,
+    "eval_samples_per_second": 371.507,
+    "eval_steps_per_second": 23.23,
+    "total_flos": 1.9812822353864294e+17,
+    "train_loss": 5.166637740610481,
+    "train_runtime": 5050.3912,
+    "train_samples_per_second": 101.454,
+    "train_steps_per_second": 0.793
+}

qian_sg_backup/qwen3/qwen3_p109m_t1_8_suite-owt_scaling_v3_family-qwen3_geometry_id-euclid_variant-base_init_slope-std/attempt3_20260310_081106/config.json

@@ -0,0 +1,41 @@
+{
+  "architectures": [
+    "Qwen3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "dtype": "bfloat16",
+  "eos_token_id": 151645,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 512,
+  "initializer_range": 0.02,
+  "intermediate_size": 1536,
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 40960,
+  "max_window_layers": 8,
+  "model_type": "qwen3",
+  "num_attention_heads": 8,
+  "num_hidden_layers": 8,
+  "num_key_value_heads": 4,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.3",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

qian_sg_backup/qwen3/qwen3_p109m_t1_8_suite-owt_scaling_v3_family-qwen3_geometry_id-euclid_variant-base_init_slope-std/attempt3_20260310_081106/eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.12042538469637942,
+    "eval_loss": 4.58743143081665,
+    "eval_perplexity": 98.2417650367421,
+    "eval_runtime": 11.4076,
+    "eval_samples_per_second": 371.507,
+    "eval_steps_per_second": 23.23
+}

qian_sg_backup/qwen3/qwen3_p109m_t1_8_suite-owt_scaling_v3_family-qwen3_geometry_id-euclid_variant-base_init_slope-std/attempt3_20260310_081106/generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "transformers_version": "4.51.3",
+  "use_cache": false
+}

qian_sg_backup/qwen3/qwen3_p109m_t1_8_suite-owt_scaling_v3_family-qwen3_geometry_id-euclid_variant-base_init_slope-std/attempt3_20260310_081106/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dca5e6c824608b0f4443b44701ad4f9103368c84187d7b31f8483ef04bfe1720
+size 374111136

qian_sg_backup/qwen3/qwen3_p109m_t1_8_suite-owt_scaling_v3_family-qwen3_geometry_id-euclid_variant-base_init_slope-std/attempt3_20260310_081106/special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

qian_sg_backup/qwen3/qwen3_p109m_t1_8_suite-owt_scaling_v3_family-qwen3_geometry_id-euclid_variant-base_init_slope-std/attempt3_20260310_081106/tokenizer_config.json

@@ -0,0 +1,241 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151665": {
+      "content": "<tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151666": {
+      "content": "</tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151667": {
+      "content": "<think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151668": {
+      "content": "</think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0].role == 'system' %}\n        {{- messages[0].content + '\\n\\n' }}\n    {%- endif %}\n    {{- \"# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0].role == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0].content + '<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}\n{%- for message in messages[::-1] %}\n    {%- set index = (messages|length - 1) - loop.index0 %}\n    {%- if ns.multi_step_tool and message.role == \"user\" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}\n        {%- set ns.multi_step_tool = false %}\n        {%- set ns.last_query_index = index %}\n    {%- endif %}\n{%- endfor %}\n{%- for message in messages %}\n    {%- if message.content is string %}\n        {%- set content = message.content %}\n    {%- else %}\n        {%- set content = '' %}\n    {%- endif %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) %}\n        {{- '<|im_start|>' + message.role + '\\n' + content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {%- set reasoning_content = '' %}\n        {%- if message.reasoning_content is string %}\n            {%- set reasoning_content = message.reasoning_content %}\n        {%- else %}\n            {%- if '</think>' in content %}\n                {%- set reasoning_content = content.split('</think>')[0].rstrip('\\n').split('<think>')[-1].lstrip('\\n') %}\n                {%- set content = content.split('</think>')[-1].lstrip('\\n') %}\n            {%- endif %}\n        {%- endif %}\n        {%- if loop.index0 > ns.last_query_index %}\n            {%- if loop.last or (not loop.last and reasoning_content) %}\n                {{- '<|im_start|>' + message.role + '\\n<think>\\n' + reasoning_content.strip('\\n') + '\\n</think>\\n\\n' + content.lstrip('\\n') }}\n            {%- else %}\n                {{- '<|im_start|>' + message.role + '\\n' + content }}\n            {%- endif %}\n        {%- else %}\n            {{- '<|im_start|>' + message.role + '\\n' + content }}\n        {%- endif %}\n        {%- if message.tool_calls %}\n            {%- for tool_call in message.tool_calls %}\n                {%- if (loop.first and content) or (not loop.first) %}\n                    {{- '\\n' }}\n                {%- endif %}\n                {%- if tool_call.function %}\n                    {%- set tool_call = tool_call.function %}\n                {%- endif %}\n                {{- '<tool_call>\\n{\"name\": \"' }}\n                {{- tool_call.name }}\n                {{- '\", \"arguments\": ' }}\n                {%- if tool_call.arguments is string %}\n                    {{- tool_call.arguments }}\n                {%- else %}\n                    {{- tool_call.arguments | tojson }}\n                {%- endif %}\n                {{- '}\\n</tool_call>' }}\n            {%- endfor %}\n        {%- endif %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if loop.first or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n    {%- if enable_thinking is defined and enable_thinking is false %}\n        {{- '<think>\\n\\n</think>\\n\\n' }}\n    {%- endif %}\n{%- endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

qian_sg_backup/qwen3/qwen3_p109m_t1_8_suite-owt_scaling_v3_family-qwen3_geometry_id-euclid_variant-base_init_slope-std/attempt3_20260310_081106/train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.12042538469637942,
+    "total_flos": 1.9812822353864294e+17,
+    "train_loss": 5.166637740610481,
+    "train_runtime": 5050.3912,
+    "train_samples_per_second": 101.454,
+    "train_steps_per_second": 0.793
+}

qian_sg_backup/qwen3/qwen3_p109m_t1_8_suite-owt_scaling_v3_family-qwen3_geometry_id-euclid_variant-base_init_slope-std/attempt3_20260310_081106/trainer_log.jsonl

@@ -0,0 +1,56 @@
+{"current_steps": 100, "total_steps": 4003, "loss": 10.138, "lr": 9.999480288817096e-05, "epoch": 0.0030083783336592408, "percentage": 2.5, "elapsed_time": "0:02:02", "remaining_time": "1:19:42"}
+{"current_steps": 200, "total_steps": 4003, "loss": 7.2528, "lr": 9.977681488236596e-05, "epoch": 0.0060167566673184816, "percentage": 5.0, "elapsed_time": "0:04:04", "remaining_time": "1:17:22"}
+{"current_steps": 256, "total_steps": 4003, "eval_loss": 6.477409362792969, "epoch": 0.007701448534167657, "percentage": 6.4, "elapsed_time": "0:05:24", "remaining_time": "1:19:04"}
+{"current_steps": 300, "total_steps": 4003, "loss": 6.5217, "lr": 9.92396147038165e-05, "epoch": 0.009025135000977723, "percentage": 7.49, "elapsed_time": "0:06:17", "remaining_time": "1:17:42"}
+{"current_steps": 400, "total_steps": 4003, "loss": 6.1661, "lr": 9.838664734667495e-05, "epoch": 0.012033513334636963, "percentage": 9.99, "elapsed_time": "0:08:19", "remaining_time": "1:14:58"}
+{"current_steps": 500, "total_steps": 4003, "loss": 5.9516, "lr": 9.722338277851551e-05, "epoch": 0.015041891668296205, "percentage": 12.49, "elapsed_time": "0:10:21", "remaining_time": "1:12:34"}
+{"current_steps": 512, "total_steps": 4003, "eval_loss": 5.844233989715576, "epoch": 0.015402897068335314, "percentage": 12.79, "elapsed_time": "0:10:47", "remaining_time": "1:13:35"}
+{"current_steps": 600, "total_steps": 4003, "loss": 5.7934, "lr": 9.575728086215092e-05, "epoch": 0.018050270001955446, "percentage": 14.99, "elapsed_time": "0:12:35", "remaining_time": "1:11:27"}
+{"current_steps": 700, "total_steps": 4003, "loss": 5.6485, "lr": 9.399774351651263e-05, "epoch": 0.021058648335614688, "percentage": 17.49, "elapsed_time": "0:14:37", "remaining_time": "1:08:58"}
+{"current_steps": 768, "total_steps": 4003, "eval_loss": 5.520048141479492, "epoch": 0.02310434560250297, "percentage": 19.19, "elapsed_time": "0:16:11", "remaining_time": "1:08:10"}
+{"current_steps": 800, "total_steps": 4003, "loss": 5.5455, "lr": 9.195605442338004e-05, "epoch": 0.024067026669273926, "percentage": 19.99, "elapsed_time": "0:16:49", "remaining_time": "1:07:23"}
+{"current_steps": 900, "total_steps": 4003, "loss": 5.4401, "lr": 8.964530666661213e-05, "epoch": 0.027075405002933168, "percentage": 22.48, "elapsed_time": "0:18:50", "remaining_time": "1:04:59"}
+{"current_steps": 1000, "total_steps": 4003, "loss": 5.3609, "lr": 8.708031876792118e-05, "epoch": 0.03008378333659241, "percentage": 24.98, "elapsed_time": "0:20:51", "remaining_time": "1:02:39"}
+{"current_steps": 1024, "total_steps": 4003, "eval_loss": 5.300453186035156, "epoch": 0.030805794136670627, "percentage": 25.58, "elapsed_time": "0:21:32", "remaining_time": "1:02:39"}
+{"current_steps": 1100, "total_steps": 4003, "loss": 5.2772, "lr": 8.427753965764032e-05, "epoch": 0.03309216167025165, "percentage": 27.48, "elapsed_time": "0:23:06", "remaining_time": "1:00:58"}
+{"current_steps": 1200, "total_steps": 4003, "loss": 5.2063, "lr": 8.125494318989478e-05, "epoch": 0.03610054000391089, "percentage": 29.98, "elapsed_time": "0:25:09", "remaining_time": "0:58:45"}
+{"current_steps": 1280, "total_steps": 4003, "eval_loss": 5.111225128173828, "epoch": 0.038507242670838285, "percentage": 31.98, "elapsed_time": "0:26:58", "remaining_time": "0:57:23"}
+{"current_steps": 1300, "total_steps": 4003, "loss": 5.1365, "lr": 7.803191287863736e-05, "epoch": 0.03910891833757013, "percentage": 32.48, "elapsed_time": "0:27:22", "remaining_time": "0:56:55"}
+{"current_steps": 1400, "total_steps": 4003, "loss": 5.0652, "lr": 7.462911759372063e-05, "epoch": 0.042117296671229376, "percentage": 34.97, "elapsed_time": "0:29:23", "remaining_time": "0:54:39"}
+{"current_steps": 1500, "total_steps": 4003, "loss": 5.0085, "lr": 7.106837901415058e-05, "epoch": 0.045125675004888614, "percentage": 37.47, "elapsed_time": "0:31:24", "remaining_time": "0:52:25"}
+{"current_steps": 1536, "total_steps": 4003, "eval_loss": 4.967124938964844, "epoch": 0.04620869120500594, "percentage": 38.37, "elapsed_time": "0:32:20", "remaining_time": "0:51:56"}
+{"current_steps": 1600, "total_steps": 4003, "loss": 4.9624, "lr": 6.737253168852667e-05, "epoch": 0.04813405333854785, "percentage": 39.97, "elapsed_time": "0:33:39", "remaining_time": "0:50:33"}
+{"current_steps": 1700, "total_steps": 4003, "loss": 4.9145, "lr": 6.356527660008243e-05, "epoch": 0.0511424316722071, "percentage": 42.47, "elapsed_time": "0:35:41", "remaining_time": "0:48:20"}
+{"current_steps": 1792, "total_steps": 4003, "eval_loss": 4.856684684753418, "epoch": 0.0539101397391736, "percentage": 44.77, "elapsed_time": "0:37:44", "remaining_time": "0:46:33"}
+{"current_steps": 1800, "total_steps": 4003, "loss": 4.8726, "lr": 5.9671029175394365e-05, "epoch": 0.054150810005866336, "percentage": 44.97, "elapsed_time": "0:37:54", "remaining_time": "0:46:23"}
+{"current_steps": 1900, "total_steps": 4003, "loss": 4.8265, "lr": 5.5714762711459876e-05, "epoch": 0.05715918833952558, "percentage": 47.46, "elapsed_time": "0:39:54", "remaining_time": "0:44:10"}
+{"current_steps": 2000, "total_steps": 4003, "loss": 4.8034, "lr": 5.172184822522515e-05, "epoch": 0.06016756667318482, "percentage": 49.96, "elapsed_time": "0:41:55", "remaining_time": "0:41:59"}
+{"current_steps": 2048, "total_steps": 4003, "eval_loss": 4.774194717407227, "epoch": 0.061611588273341254, "percentage": 51.16, "elapsed_time": "0:43:05", "remaining_time": "0:41:07"}
+{"current_steps": 2100, "total_steps": 4003, "loss": 4.7758, "lr": 4.7717891752587515e-05, "epoch": 0.06317594500684406, "percentage": 52.46, "elapsed_time": "0:44:10", "remaining_time": "0:40:01"}
+{"current_steps": 2200, "total_steps": 4003, "loss": 4.7438, "lr": 4.3728570140252134e-05, "epoch": 0.0661843233405033, "percentage": 54.96, "elapsed_time": "0:46:11", "remaining_time": "0:37:51"}
+{"current_steps": 2300, "total_steps": 4003, "loss": 4.7276, "lr": 3.97794663834882e-05, "epoch": 0.06919270167416254, "percentage": 57.46, "elapsed_time": "0:48:13", "remaining_time": "0:35:42"}
+{"current_steps": 2304, "total_steps": 4003, "eval_loss": 4.712698459625244, "epoch": 0.06931303680750892, "percentage": 57.56, "elapsed_time": "0:48:30", "remaining_time": "0:35:45"}
+{"current_steps": 2400, "total_steps": 4003, "loss": 4.7033, "lr": 3.5895905565742016e-05, "epoch": 0.07220108000782179, "percentage": 59.96, "elapsed_time": "0:50:26", "remaining_time": "0:33:41"}
+{"current_steps": 2500, "total_steps": 4003, "loss": 4.6877, "lr": 3.2102792452204977e-05, "epoch": 0.07520945834148103, "percentage": 62.45, "elapsed_time": "0:52:27", "remaining_time": "0:31:32"}
+{"current_steps": 2560, "total_steps": 4003, "eval_loss": 4.665860176086426, "epoch": 0.07701448534167657, "percentage": 63.95, "elapsed_time": "0:53:52", "remaining_time": "0:30:21"}
+{"current_steps": 2600, "total_steps": 4003, "loss": 4.6662, "lr": 2.842445177882692e-05, "epoch": 0.07821783667514026, "percentage": 64.95, "elapsed_time": "0:54:42", "remaining_time": "0:29:31"}
+{"current_steps": 2700, "total_steps": 4003, "loss": 4.6539, "lr": 2.488447226098098e-05, "epoch": 0.08122621500879951, "percentage": 67.45, "elapsed_time": "0:56:44", "remaining_time": "0:27:22"}
+{"current_steps": 2800, "total_steps": 4003, "loss": 4.6423, "lr": 2.150555532213176e-05, "epoch": 0.08423459334245875, "percentage": 69.95, "elapsed_time": "0:58:45", "remaining_time": "0:25:14"}
+{"current_steps": 2816, "total_steps": 4003, "eval_loss": 4.632678508758545, "epoch": 0.08471593387584422, "percentage": 70.35, "elapsed_time": "0:59:16", "remaining_time": "0:24:59"}
+{"current_steps": 2900, "total_steps": 4003, "loss": 4.6291, "lr": 1.8309369512589713e-05, "epoch": 0.08724297167611798, "percentage": 72.45, "elapsed_time": "1:00:58", "remaining_time": "0:23:11"}
+{"current_steps": 3000, "total_steps": 4003, "loss": 4.618, "lr": 1.5316411551945487e-05, "epoch": 0.09025135000977723, "percentage": 74.94, "elapsed_time": "1:02:58", "remaining_time": "0:21:03"}
+{"current_steps": 3072, "total_steps": 4003, "eval_loss": 4.610559940338135, "epoch": 0.09241738241001188, "percentage": 76.74, "elapsed_time": "1:04:37", "remaining_time": "0:19:35"}
+{"current_steps": 3100, "total_steps": 4003, "loss": 4.6174, "lr": 1.2545874886300612e-05, "epoch": 0.09325972834343647, "percentage": 77.44, "elapsed_time": "1:05:13", "remaining_time": "0:19:00"}
+{"current_steps": 3200, "total_steps": 4003, "loss": 4.6052, "lr": 1.00155266032198e-05, "epoch": 0.0962681066770957, "percentage": 79.94, "elapsed_time": "1:07:15", "remaining_time": "0:16:52"}
+{"current_steps": 3300, "total_steps": 4003, "loss": 4.6031, "lr": 7.741593493733208e-06, "epoch": 0.09927648501075495, "percentage": 82.44, "elapsed_time": "1:09:16", "remaining_time": "0:14:45"}
+{"current_steps": 3328, "total_steps": 4003, "eval_loss": 4.597111701965332, "epoch": 0.10011883094417955, "percentage": 83.14, "elapsed_time": "1:10:01", "remaining_time": "0:14:12"}
+{"current_steps": 3400, "total_steps": 4003, "loss": 4.5971, "lr": 5.738657992057856e-06, "epoch": 0.1022848633444142, "percentage": 84.94, "elapsed_time": "1:11:28", "remaining_time": "0:12:40"}
+{"current_steps": 3500, "total_steps": 4003, "loss": 4.5861, "lr": 4.019564660363262e-06, "epoch": 0.10529324167807344, "percentage": 87.43, "elapsed_time": "1:13:29", "remaining_time": "0:10:33"}
+{"current_steps": 3584, "total_steps": 4003, "eval_loss": 4.589923858642578, "epoch": 0.1078202794783472, "percentage": 89.53, "elapsed_time": "1:15:22", "remaining_time": "0:08:48"}
+{"current_steps": 3600, "total_steps": 4003, "loss": 4.589, "lr": 2.5953378182821066e-06, "epoch": 0.10830162001173267, "percentage": 89.93, "elapsed_time": "1:15:43", "remaining_time": "0:08:28"}
+{"current_steps": 3700, "total_steps": 4003, "loss": 4.5836, "lr": 1.4751108453971597e-06, "epoch": 0.11130999834539192, "percentage": 92.43, "elapsed_time": "1:17:45", "remaining_time": "0:06:22"}
+{"current_steps": 3800, "total_steps": 4003, "loss": 4.591, "lr": 6.660676100782081e-07, "epoch": 0.11431837667905116, "percentage": 94.93, "elapsed_time": "1:19:47", "remaining_time": "0:04:15"}
+{"current_steps": 3840, "total_steps": 4003, "eval_loss": 4.587612628936768, "epoch": 0.11552172801251485, "percentage": 95.93, "elapsed_time": "1:20:47", "remaining_time": "0:03:25"}
+{"current_steps": 3900, "total_steps": 4003, "loss": 4.5838, "lr": 1.7339640027818626e-07, "epoch": 0.1173267550127104, "percentage": 97.43, "elapsed_time": "1:22:00", "remaining_time": "0:02:09"}
+{"current_steps": 4000, "total_steps": 4003, "loss": 4.5879, "lr": 2.566517243729472e-10, "epoch": 0.12033513334636964, "percentage": 99.93, "elapsed_time": "1:24:02", "remaining_time": "0:00:03"}
+{"current_steps": 4003, "total_steps": 4003, "epoch": 0.12042538469637942, "percentage": 100.0, "elapsed_time": "1:24:08", "remaining_time": "0:00:00"}

qian_sg_backup/qwen3/qwen3_p109m_t1_8_suite-owt_scaling_v3_family-qwen3_geometry_id-euclid_variant-base_init_slope-std/attempt3_20260310_081106/trainer_state.json

@@ -0,0 +1,443 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.12042538469637942,
+  "eval_steps": 256,
+  "global_step": 4003,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0030083783336592408,
+      "grad_norm": 1.0657947063446045,
+      "learning_rate": 9.999480288817096e-05,
+      "loss": 10.138,
+      "step": 100
+    },
+    {
+      "epoch": 0.0060167566673184816,
+      "grad_norm": 0.5770671963691711,
+      "learning_rate": 9.977681488236596e-05,
+      "loss": 7.2528,
+      "step": 200
+    },
+    {
+      "epoch": 0.007701448534167657,
+      "eval_loss": 6.477409362792969,
+      "eval_runtime": 11.4963,
+      "eval_samples_per_second": 368.641,
+      "eval_steps_per_second": 23.051,
+      "step": 256
+    },
+    {
+      "epoch": 0.009025135000977723,
+      "grad_norm": 1.0136157274246216,
+      "learning_rate": 9.92396147038165e-05,
+      "loss": 6.5217,
+      "step": 300
+    },
+    {
+      "epoch": 0.012033513334636963,
+      "grad_norm": 1.6077970266342163,
+      "learning_rate": 9.838664734667495e-05,
+      "loss": 6.1661,
+      "step": 400
+    },
+    {
+      "epoch": 0.015041891668296205,
+      "grad_norm": 1.1440670490264893,
+      "learning_rate": 9.722338277851551e-05,
+      "loss": 5.9516,
+      "step": 500
+    },
+    {
+      "epoch": 0.015402897068335314,
+      "eval_loss": 5.844233989715576,
+      "eval_runtime": 11.4701,
+      "eval_samples_per_second": 369.482,
+      "eval_steps_per_second": 23.104,
+      "step": 512
+    },
+    {
+      "epoch": 0.018050270001955446,
+      "grad_norm": 1.7339913845062256,
+      "learning_rate": 9.575728086215092e-05,
+      "loss": 5.7934,
+      "step": 600
+    },
+    {
+      "epoch": 0.021058648335614688,
+      "grad_norm": 1.2664787769317627,
+      "learning_rate": 9.399774351651263e-05,
+      "loss": 5.6485,
+      "step": 700
+    },
+    {
+      "epoch": 0.02310434560250297,
+      "eval_loss": 5.520048141479492,
+      "eval_runtime": 11.4436,
+      "eval_samples_per_second": 370.338,
+      "eval_steps_per_second": 23.157,
+      "step": 768
+    },
+    {
+      "epoch": 0.024067026669273926,
+      "grad_norm": 1.1303988695144653,
+      "learning_rate": 9.195605442338004e-05,
+      "loss": 5.5455,
+      "step": 800
+    },
+    {
+      "epoch": 0.027075405002933168,
+      "grad_norm": 0.9063535928726196,
+      "learning_rate": 8.964530666661213e-05,
+      "loss": 5.4401,
+      "step": 900
+    },
+    {
+      "epoch": 0.03008378333659241,
+      "grad_norm": 1.0469073057174683,
+      "learning_rate": 8.708031876792118e-05,
+      "loss": 5.3609,
+      "step": 1000
+    },
+    {
+      "epoch": 0.030805794136670627,
+      "eval_loss": 5.300453186035156,
+      "eval_runtime": 11.4548,
+      "eval_samples_per_second": 369.974,
+      "eval_steps_per_second": 23.134,
+      "step": 1024
+    },
+    {
+      "epoch": 0.03309216167025165,
+      "grad_norm": 1.3508996963500977,
+      "learning_rate": 8.427753965764032e-05,
+      "loss": 5.2772,
+      "step": 1100
+    },
+    {
+      "epoch": 0.03610054000391089,
+      "grad_norm": 1.4069072008132935,
+      "learning_rate": 8.125494318989478e-05,
+      "loss": 5.2063,
+      "step": 1200
+    },
+    {
+      "epoch": 0.038507242670838285,
+      "eval_loss": 5.111225128173828,
+      "eval_runtime": 11.4462,
+      "eval_samples_per_second": 370.254,
+      "eval_steps_per_second": 23.152,
+      "step": 1280
+    },
+    {
+      "epoch": 0.03910891833757013,
+      "grad_norm": 1.2563368082046509,
+      "learning_rate": 7.803191287863736e-05,
+      "loss": 5.1365,
+      "step": 1300
+    },
+    {
+      "epoch": 0.042117296671229376,
+      "grad_norm": 0.8728069067001343,
+      "learning_rate": 7.462911759372063e-05,
+      "loss": 5.0652,
+      "step": 1400
+    },
+    {
+      "epoch": 0.045125675004888614,
+      "grad_norm": 1.136967420578003,
+      "learning_rate": 7.106837901415058e-05,
+      "loss": 5.0085,
+      "step": 1500
+    },
+    {
+      "epoch": 0.04620869120500594,
+      "eval_loss": 4.967124938964844,
+      "eval_runtime": 11.4443,
+      "eval_samples_per_second": 370.317,
+      "eval_steps_per_second": 23.156,
+      "step": 1536
+    },
+    {
+      "epoch": 0.04813405333854785,
+      "grad_norm": 1.111749529838562,
+      "learning_rate": 6.737253168852667e-05,
+      "loss": 4.9624,
+      "step": 1600
+    },
+    {
+      "epoch": 0.0511424316722071,
+      "grad_norm": 0.9046854972839355,
+      "learning_rate": 6.356527660008243e-05,
+      "loss": 4.9145,
+      "step": 1700
+    },
+    {
+      "epoch": 0.0539101397391736,
+      "eval_loss": 4.856684684753418,
+      "eval_runtime": 11.4398,
+      "eval_samples_per_second": 370.462,
+      "eval_steps_per_second": 23.165,
+      "step": 1792
+    },
+    {
+      "epoch": 0.054150810005866336,
+      "grad_norm": 1.2382330894470215,
+      "learning_rate": 5.9671029175394365e-05,
+      "loss": 4.8726,
+      "step": 1800
+    },
+    {
+      "epoch": 0.05715918833952558,
+      "grad_norm": 0.9098496437072754,
+      "learning_rate": 5.5714762711459876e-05,
+      "loss": 4.8265,
+      "step": 1900
+    },
+    {
+      "epoch": 0.06016756667318482,
+      "grad_norm": 1.033302903175354,
+      "learning_rate": 5.172184822522515e-05,
+      "loss": 4.8034,
+      "step": 2000
+    },
+    {
+      "epoch": 0.061611588273341254,
+      "eval_loss": 4.774194717407227,
+      "eval_runtime": 11.4477,
+      "eval_samples_per_second": 370.206,
+      "eval_steps_per_second": 23.149,
+      "step": 2048
+    },
+    {
+      "epoch": 0.06317594500684406,
+      "grad_norm": 1.0562632083892822,
+      "learning_rate": 4.7717891752587515e-05,
+      "loss": 4.7758,
+      "step": 2100
+    },
+    {
+      "epoch": 0.0661843233405033,
+      "grad_norm": 0.7925425171852112,
+      "learning_rate": 4.3728570140252134e-05,
+      "loss": 4.7438,
+      "step": 2200
+    },
+    {
+      "epoch": 0.06919270167416254,
+      "grad_norm": 0.8904197216033936,
+      "learning_rate": 3.97794663834882e-05,
+      "loss": 4.7276,
+      "step": 2300
+    },
+    {
+      "epoch": 0.06931303680750892,
+      "eval_loss": 4.712698459625244,
+      "eval_runtime": 11.4486,
+      "eval_samples_per_second": 370.177,
+      "eval_steps_per_second": 23.147,
+      "step": 2304
+    },
+    {
+      "epoch": 0.07220108000782179,
+      "grad_norm": 0.8380971550941467,
+      "learning_rate": 3.5895905565742016e-05,
+      "loss": 4.7033,
+      "step": 2400
+    },
+    {
+      "epoch": 0.07520945834148103,
+      "grad_norm": 0.8202688097953796,
+      "learning_rate": 3.2102792452204977e-05,
+      "loss": 4.6877,
+      "step": 2500
+    },
+    {
+      "epoch": 0.07701448534167657,
+      "eval_loss": 4.665860176086426,
+      "eval_runtime": 11.4816,
+      "eval_samples_per_second": 369.113,
+      "eval_steps_per_second": 23.08,
+      "step": 2560
+    },
+    {
+      "epoch": 0.07821783667514026,
+      "grad_norm": 0.8118715286254883,
+      "learning_rate": 2.842445177882692e-05,
+      "loss": 4.6662,
+      "step": 2600
+    },
+    {
+      "epoch": 0.08122621500879951,
+      "grad_norm": 0.9480417966842651,
+      "learning_rate": 2.488447226098098e-05,
+      "loss": 4.6539,
+      "step": 2700
+    },
+    {
+      "epoch": 0.08423459334245875,
+      "grad_norm": 0.7308496236801147,
+      "learning_rate": 2.150555532213176e-05,
+      "loss": 4.6423,
+      "step": 2800
+    },
+    {
+      "epoch": 0.08471593387584422,
+      "eval_loss": 4.632678508758545,
+      "eval_runtime": 11.4771,
+      "eval_samples_per_second": 369.258,
+      "eval_steps_per_second": 23.09,
+      "step": 2816
+    },
+    {
+      "epoch": 0.08724297167611798,
+      "grad_norm": 0.6865440011024475,
+      "learning_rate": 1.8309369512589713e-05,
+      "loss": 4.6291,
+      "step": 2900
+    },
+    {
+      "epoch": 0.09025135000977723,
+      "grad_norm": 0.7402969002723694,
+      "learning_rate": 1.5316411551945487e-05,
+      "loss": 4.618,
+      "step": 3000
+    },
+    {
+      "epoch": 0.09241738241001188,
+      "eval_loss": 4.610559940338135,
+      "eval_runtime": 11.4651,
+      "eval_samples_per_second": 369.645,
+      "eval_steps_per_second": 23.114,
+      "step": 3072
+    },
+    {
+      "epoch": 0.09325972834343647,
+      "grad_norm": 0.7089903950691223,
+      "learning_rate": 1.2545874886300612e-05,
+      "loss": 4.6174,
+      "step": 3100
+    },
+    {
+      "epoch": 0.0962681066770957,
+      "grad_norm": 0.6320310831069946,
+      "learning_rate": 1.00155266032198e-05,
+      "loss": 4.6052,
+      "step": 3200
+    },
+    {
+      "epoch": 0.09927648501075495,
+      "grad_norm": 0.720451831817627,
+      "learning_rate": 7.741593493733208e-06,
+      "loss": 4.6031,
+      "step": 3300
+    },
+    {
+      "epoch": 0.10011883094417955,
+      "eval_loss": 4.597111701965332,
+      "eval_runtime": 11.4429,
+      "eval_samples_per_second": 370.36,
+      "eval_steps_per_second": 23.158,
+      "step": 3328
+    },
+    {
+      "epoch": 0.1022848633444142,
+      "grad_norm": 0.5505564212799072,
+      "learning_rate": 5.738657992057856e-06,
+      "loss": 4.5971,
+      "step": 3400
+    },
+    {
+      "epoch": 0.10529324167807344,
+      "grad_norm": 0.5950909852981567,
+      "learning_rate": 4.019564660363262e-06,
+      "loss": 4.5861,
+      "step": 3500
+    },
+    {
+      "epoch": 0.1078202794783472,
+      "eval_loss": 4.589923858642578,
+      "eval_runtime": 11.4331,
+      "eval_samples_per_second": 370.678,
+      "eval_steps_per_second": 23.178,
+      "step": 3584
+    },
+    {
+      "epoch": 0.10830162001173267,
+      "grad_norm": 0.5519059300422668,
+      "learning_rate": 2.5953378182821066e-06,
+      "loss": 4.589,
+      "step": 3600
+    },
+    {
+      "epoch": 0.11130999834539192,
+      "grad_norm": 0.5641384720802307,
+      "learning_rate": 1.4751108453971597e-06,
+      "loss": 4.5836,
+      "step": 3700
+    },
+    {
+      "epoch": 0.11431837667905116,
+      "grad_norm": 0.4828135371208191,
+      "learning_rate": 6.660676100782081e-07,
+      "loss": 4.591,
+      "step": 3800
+    },
+    {
+      "epoch": 0.11552172801251485,
+      "eval_loss": 4.587612628936768,
+      "eval_runtime": 11.4927,
+      "eval_samples_per_second": 368.755,
+      "eval_steps_per_second": 23.058,
+      "step": 3840
+    },
+    {
+      "epoch": 0.1173267550127104,
+      "grad_norm": 0.44198083877563477,
+      "learning_rate": 1.7339640027818626e-07,
+      "loss": 4.5838,
+      "step": 3900
+    },
+    {
+      "epoch": 0.12033513334636964,
+      "grad_norm": 0.5587871074676514,
+      "learning_rate": 2.566517243729472e-10,
+      "loss": 4.5879,
+      "step": 4000
+    },
+    {
+      "epoch": 0.12042538469637942,
+      "step": 4003,
+      "total_flos": 1.9812822353864294e+17,
+      "train_loss": 5.166637740610481,
+      "train_runtime": 5050.3912,
+      "train_samples_per_second": 101.454,
+      "train_steps_per_second": 0.793
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 4003,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 512,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.9812822353864294e+17,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

qian_sg_backup/qwen3/qwen3_p109m_t1_8_suite-owt_scaling_v3_family-qwen3_geometry_id-euclid_variant-base_init_slope-std/attempt3_20260310_081106/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f78069dfb0691064a6967e30733398192b1eb83578ac1a9e894cd398f9ffa66
+size 7544