Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,7 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+checkpoint-105/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-210/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-315/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,149 @@
+---
+library_name: peft
+license: apache-2.0
+base_model: Qwen/Qwen3-4B-Instruct-2507
+tags:
+- axolotl
+- base_model:adapter:Qwen/Qwen3-4B-Instruct-2507
+- lora
+- transformers
+datasets:
+- LuxMuseAI/StreetMath
+pipeline_tag: text-generation
+model-index:
+- name: outputs/mymodel
+  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. -->
+
+[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
+<details><summary>See axolotl config</summary>
+
+axolotl version: `0.13.0.dev0`
+```yaml
+adapter: lora
+base_model: Qwen/Qwen3-4B-Instruct-2507
+bf16: auto
+dataset_processes: 32
+datasets:
+- type: alpaca
+  message_property_mappings:
+    instruction: instruction
+    input: input
+    output: output
+  path: LuxMuseAI/StreetMath
+  trust_remote_code: false
+gradient_accumulation_steps: 1
+gradient_checkpointing: false
+learning_rate: 0.0002
+lisa_layers_attribute: model.layers
+load_best_model_at_end: false
+load_in_4bit: false
+load_in_8bit: true
+lora_alpha: 16
+lora_dropout: 0.05
+lora_r: 8
+lora_target_modules:
+- q_proj
+- v_proj
+- k_proj
+- o_proj
+- gate_proj
+- down_proj
+- up_proj
+loraplus_lr_embedding: 1.0e-06
+lr_scheduler: cosine
+max_prompt_len: 512
+mean_resizing_embeddings: false
+micro_batch_size: 16
+num_epochs: 3.0
+optimizer: adamw_bnb_8bit
+output_dir: ./outputs/mymodel
+pretrain_multipack_attn: true
+pretrain_multipack_buffer_size: 10000
+qlora_sharded_model_loading: false
+ray_num_workers: 1
+resources_per_worker:
+  GPU: 1
+sample_packing_bin_size: 200
+sample_packing_group_size: 100000
+save_only_model: false
+save_safetensors: true
+sequence_len: 4096
+shuffle_merged_datasets: true
+skip_prepare_dataset: false
+strict: false
+train_on_inputs: false
+trl:
+  log_completions: false
+  ref_model_mixup_alpha: 0.9
+  ref_model_sync_steps: 64
+  sync_ref_model: false
+  use_vllm: false
+  vllm_device: auto
+  vllm_dtype: auto
+  vllm_gpu_memory_utilization: 0.9
+use_ray: false
+val_set_size: 0.0
+weight_decay: 0.0
+
+# Weights & Biases logging (optional)
+wandb_project: StreetMath
+wandb_entity: ctseng-  # Replace with your actual entity name
+wandb_watch: all
+wandb_name: StreetMath-4B-Instruct-2507
+wandb_log_model: "end"
+
+
+```
+
+</details><br>
+
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/ctseng-/StreetMath/runs/2tnvvjjg)
+# outputs/mymodel
+
+This model is a fine-tuned version of [Qwen/Qwen3-4B-Instruct-2507](https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507) on the LuxMuseAI/StreetMath dataset.
+
+## 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: 0.0002
+- train_batch_size: 16
+- eval_batch_size: 16
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 2
+- total_train_batch_size: 32
+- total_eval_batch_size: 32
+- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_steps: 9
+- training_steps: 315
+
+### Training results
+
+
+
+### Framework versions
+
+- PEFT 0.17.0
+- Transformers 4.55.2
+- Pytorch 2.6.0+cu124
+- Datasets 4.0.0
+- Tokenizers 0.21.4

adapter_config.json

@@ -0,0 +1,42 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-4B-Instruct-2507",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": null,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "qalora_group_size": 16,
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "gate_proj",
+    "up_proj",
+    "v_proj",
+    "down_proj",
+    "k_proj",
+    "o_proj"
+  ],
+  "target_parameters": [],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e45b31c137803c2351f0144a9d9cdee51846c1e3ed927ec07b20c1cc38ea3b8
+size 66126768

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
+}

chat_template.jinja

@@ -0,0 +1,86 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if message.content is string %}
+        {%- set content = message.content %}
+    {%- else %}
+        {%- set content = '' %}
+    {%- endif %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

checkpoint-105/README.md

@@ -0,0 +1,208 @@
+---
+base_model: Qwen/Qwen3-4B-Instruct-2507
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- axolotl
+- base_model:adapter:Qwen/Qwen3-4B-Instruct-2507
+- 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 0.17.0

checkpoint-105/adapter_config.json

@@ -0,0 +1,42 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-4B-Instruct-2507",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": null,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "qalora_group_size": 16,
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "gate_proj",
+    "up_proj",
+    "v_proj",
+    "down_proj",
+    "k_proj",
+    "o_proj"
+  ],
+  "target_parameters": [],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoint-105/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:54d3caf7bf83634e0bf622c9761f1cdc965624c297a13b8d0f45fa42e6a4b3a1
+size 66126768

checkpoint-105/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
+}

checkpoint-105/chat_template.jinja

@@ -0,0 +1,86 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if message.content is string %}
+        {%- set content = message.content %}
+    {%- else %}
+        {%- set content = '' %}
+    {%- endif %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

checkpoint-105/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ed27fee10f7b70fc015612c250d7a1cbad38b5bc091082ba56b6843e9687f45
+size 34135860

checkpoint-105/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8573018b24651647d06d6a5ca1df2de350b37825737c4c52dd99d5ae33013b32
+size 14512

checkpoint-105/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4940f16376ab751aa7d868bd930443d7668be27833ea35bde376bcd2d4202ec1
+size 14512

checkpoint-105/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2dfcd9279d0def3579f388da83224a4a93c3472451c0ccedd9f00c9689fd66a8
+size 1064

checkpoint-105/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
+  }
+}

checkpoint-105/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

checkpoint-105/tokenizer_config.json

@@ -0,0 +1,239 @@
+{
+  "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,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 262144,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoint-105/trainer_state.json

@@ -0,0 +1,1084 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 105,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.009523809523809525,
+      "grad_norm": 2.447430372238159,
+      "learning_rate": 0.0,
+      "loss": 1.8171,
+      "memory/device_mem_reserved(gib)": 36.42,
+      "memory/max_mem_active(gib)": 35.17,
+      "memory/max_mem_allocated(gib)": 35.17,
+      "step": 1
+    },
+    {
+      "epoch": 0.01904761904761905,
+      "grad_norm": 2.280487537384033,
+      "learning_rate": 2.2222222222222223e-05,
+      "loss": 1.7544,
+      "memory/device_mem_reserved(gib)": 36.45,
+      "memory/max_mem_active(gib)": 35.23,
+      "memory/max_mem_allocated(gib)": 35.23,
+      "step": 2
+    },
+    {
+      "epoch": 0.02857142857142857,
+      "grad_norm": 2.248966693878174,
+      "learning_rate": 4.4444444444444447e-05,
+      "loss": 1.7546,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 3
+    },
+    {
+      "epoch": 0.0380952380952381,
+      "grad_norm": 2.2559540271759033,
+      "learning_rate": 6.666666666666667e-05,
+      "loss": 1.8011,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 4
+    },
+    {
+      "epoch": 0.047619047619047616,
+      "grad_norm": 2.149872064590454,
+      "learning_rate": 8.888888888888889e-05,
+      "loss": 1.6465,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 5
+    },
+    {
+      "epoch": 0.05714285714285714,
+      "grad_norm": 1.6111228466033936,
+      "learning_rate": 0.00011111111111111112,
+      "loss": 1.5935,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 6
+    },
+    {
+      "epoch": 0.06666666666666667,
+      "grad_norm": 1.363952875137329,
+      "learning_rate": 0.00013333333333333334,
+      "loss": 1.4371,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 7
+    },
+    {
+      "epoch": 0.0761904761904762,
+      "grad_norm": 1.1439425945281982,
+      "learning_rate": 0.00015555555555555556,
+      "loss": 1.2409,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 8
+    },
+    {
+      "epoch": 0.08571428571428572,
+      "grad_norm": 0.877252995967865,
+      "learning_rate": 0.00017777777777777779,
+      "loss": 1.1083,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 9
+    },
+    {
+      "epoch": 0.09523809523809523,
+      "grad_norm": 0.8071479797363281,
+      "learning_rate": 0.0002,
+      "loss": 0.9937,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 10
+    },
+    {
+      "epoch": 0.10476190476190476,
+      "grad_norm": 0.7772606015205383,
+      "learning_rate": 0.00019999472984871732,
+      "loss": 0.9235,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 11
+    },
+    {
+      "epoch": 0.11428571428571428,
+      "grad_norm": 0.811764657497406,
+      "learning_rate": 0.00019997891995035912,
+      "loss": 0.7635,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 12
+    },
+    {
+      "epoch": 0.12380952380952381,
+      "grad_norm": 0.8423659801483154,
+      "learning_rate": 0.0001999525719713366,
+      "loss": 0.6836,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 13
+    },
+    {
+      "epoch": 0.13333333333333333,
+      "grad_norm": 0.7858280539512634,
+      "learning_rate": 0.0001999156886888064,
+      "loss": 0.6159,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 14
+    },
+    {
+      "epoch": 0.14285714285714285,
+      "grad_norm": 0.739560604095459,
+      "learning_rate": 0.00019986827399037812,
+      "loss": 0.5082,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 15
+    },
+    {
+      "epoch": 0.1523809523809524,
+      "grad_norm": 0.6787500381469727,
+      "learning_rate": 0.00019981033287370443,
+      "loss": 0.4553,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 16
+    },
+    {
+      "epoch": 0.1619047619047619,
+      "grad_norm": 0.6258607506752014,
+      "learning_rate": 0.00019974187144595432,
+      "loss": 0.3913,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 17
+    },
+    {
+      "epoch": 0.17142857142857143,
+      "grad_norm": 0.47939175367355347,
+      "learning_rate": 0.00019966289692316944,
+      "loss": 0.3048,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 18
+    },
+    {
+      "epoch": 0.18095238095238095,
+      "grad_norm": 0.49321815371513367,
+      "learning_rate": 0.00019957341762950344,
+      "loss": 0.3417,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 19
+    },
+    {
+      "epoch": 0.19047619047619047,
+      "grad_norm": 0.4252234101295471,
+      "learning_rate": 0.00019947344299634464,
+      "loss": 0.3033,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 20
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.46225014328956604,
+      "learning_rate": 0.00019936298356132176,
+      "loss": 0.2984,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 21
+    },
+    {
+      "epoch": 0.20952380952380953,
+      "grad_norm": 0.4106466472148895,
+      "learning_rate": 0.0001992420509671936,
+      "loss": 0.2764,
+      "memory/device_mem_reserved(gib)": 47.12,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 22
+    },
+    {
+      "epoch": 0.21904761904761905,
+      "grad_norm": 0.3764067590236664,
+      "learning_rate": 0.00019911065796062135,
+      "loss": 0.2352,
+      "memory/device_mem_reserved(gib)": 47.12,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 23
+    },
+    {
+      "epoch": 0.22857142857142856,
+      "grad_norm": 0.38518962264060974,
+      "learning_rate": 0.00019896881839082556,
+      "loss": 0.2239,
+      "memory/device_mem_reserved(gib)": 47.12,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 24
+    },
+    {
+      "epoch": 0.23809523809523808,
+      "grad_norm": 0.39728403091430664,
+      "learning_rate": 0.00019881654720812594,
+      "loss": 0.2192,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 25
+    },
+    {
+      "epoch": 0.24761904761904763,
+      "grad_norm": 0.3667093813419342,
+      "learning_rate": 0.00019865386046236596,
+      "loss": 0.2834,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 26
+    },
+    {
+      "epoch": 0.2571428571428571,
+      "grad_norm": 0.32502660155296326,
+      "learning_rate": 0.00019848077530122083,
+      "loss": 0.2166,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 27
+    },
+    {
+      "epoch": 0.26666666666666666,
+      "grad_norm": 0.31225672364234924,
+      "learning_rate": 0.0001982973099683902,
+      "loss": 0.2238,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 28
+    },
+    {
+      "epoch": 0.2761904761904762,
+      "grad_norm": 0.27448731660842896,
+      "learning_rate": 0.00019810348380167527,
+      "loss": 0.1983,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 29
+    },
+    {
+      "epoch": 0.2857142857142857,
+      "grad_norm": 0.32014158368110657,
+      "learning_rate": 0.00019789931723094046,
+      "loss": 0.2554,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 30
+    },
+    {
+      "epoch": 0.29523809523809524,
+      "grad_norm": 0.39480143785476685,
+      "learning_rate": 0.0001976848317759601,
+      "loss": 0.2438,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 31
+    },
+    {
+      "epoch": 0.3047619047619048,
+      "grad_norm": 0.28004321455955505,
+      "learning_rate": 0.00019746005004415005,
+      "loss": 0.2145,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 32
+    },
+    {
+      "epoch": 0.3142857142857143,
+      "grad_norm": 0.2754361033439636,
+      "learning_rate": 0.00019722499572818496,
+      "loss": 0.2284,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 33
+    },
+    {
+      "epoch": 0.3238095238095238,
+      "grad_norm": 0.3160915970802307,
+      "learning_rate": 0.00019697969360350098,
+      "loss": 0.2482,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 34
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.24699093401432037,
+      "learning_rate": 0.00019672416952568416,
+      "loss": 0.1916,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 35
+    },
+    {
+      "epoch": 0.34285714285714286,
+      "grad_norm": 0.2674485743045807,
+      "learning_rate": 0.00019645845042774553,
+      "loss": 0.2204,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 36
+    },
+    {
+      "epoch": 0.3523809523809524,
+      "grad_norm": 0.33608749508857727,
+      "learning_rate": 0.00019618256431728194,
+      "loss": 0.235,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 37
+    },
+    {
+      "epoch": 0.3619047619047619,
+      "grad_norm": 0.29581594467163086,
+      "learning_rate": 0.00019589654027352414,
+      "loss": 0.2175,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 38
+    },
+    {
+      "epoch": 0.37142857142857144,
+      "grad_norm": 0.3526640832424164,
+      "learning_rate": 0.0001956004084442718,
+      "loss": 0.1989,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 39
+    },
+    {
+      "epoch": 0.38095238095238093,
+      "grad_norm": 0.2882135808467865,
+      "learning_rate": 0.00019529420004271567,
+      "loss": 0.2144,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 40
+    },
+    {
+      "epoch": 0.3904761904761905,
+      "grad_norm": 0.2675183117389679,
+      "learning_rate": 0.0001949779473441478,
+      "loss": 0.1987,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 41
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.2908264696598053,
+      "learning_rate": 0.00019465168368255946,
+      "loss": 0.1893,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 42
+    },
+    {
+      "epoch": 0.4095238095238095,
+      "grad_norm": 0.3211234211921692,
+      "learning_rate": 0.00019431544344712776,
+      "loss": 0.2008,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 43
+    },
+    {
+      "epoch": 0.41904761904761906,
+      "grad_norm": 0.2795293927192688,
+      "learning_rate": 0.00019396926207859084,
+      "loss": 0.1935,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 44
+    },
+    {
+      "epoch": 0.42857142857142855,
+      "grad_norm": 0.28679871559143066,
+      "learning_rate": 0.00019361317606551238,
+      "loss": 0.1987,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 45
+    },
+    {
+      "epoch": 0.4380952380952381,
+      "grad_norm": 0.3165067732334137,
+      "learning_rate": 0.00019324722294043558,
+      "loss": 0.2118,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 46
+    },
+    {
+      "epoch": 0.44761904761904764,
+      "grad_norm": 0.25720369815826416,
+      "learning_rate": 0.00019287144127592704,
+      "loss": 0.1962,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 47
+    },
+    {
+      "epoch": 0.45714285714285713,
+      "grad_norm": 0.2577202022075653,
+      "learning_rate": 0.0001924858706805112,
+      "loss": 0.2022,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 48
+    },
+    {
+      "epoch": 0.4666666666666667,
+      "grad_norm": 0.2605520188808441,
+      "learning_rate": 0.0001920905517944954,
+      "loss": 0.1891,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 49
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "grad_norm": 0.2890350818634033,
+      "learning_rate": 0.00019168552628568631,
+      "loss": 0.17,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 50
+    },
+    {
+      "epoch": 0.4857142857142857,
+      "grad_norm": 0.25532475113868713,
+      "learning_rate": 0.00019127083684499806,
+      "loss": 0.1796,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 51
+    },
+    {
+      "epoch": 0.49523809523809526,
+      "grad_norm": 0.2963904142379761,
+      "learning_rate": 0.00019084652718195238,
+      "loss": 0.19,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 52
+    },
+    {
+      "epoch": 0.5047619047619047,
+      "grad_norm": 0.24455095827579498,
+      "learning_rate": 0.0001904126420200716,
+      "loss": 0.2017,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 53
+    },
+    {
+      "epoch": 0.5142857142857142,
+      "grad_norm": 0.2518145442008972,
+      "learning_rate": 0.00018996922709216455,
+      "loss": 0.1896,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 54
+    },
+    {
+      "epoch": 0.5238095238095238,
+      "grad_norm": 0.26225805282592773,
+      "learning_rate": 0.00018951632913550626,
+      "loss": 0.1893,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 55
+    },
+    {
+      "epoch": 0.5333333333333333,
+      "grad_norm": 0.25169169902801514,
+      "learning_rate": 0.00018905399588691163,
+      "loss": 0.1939,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 56
+    },
+    {
+      "epoch": 0.5428571428571428,
+      "grad_norm": 0.23847182095050812,
+      "learning_rate": 0.00018858227607770398,
+      "loss": 0.1686,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 57
+    },
+    {
+      "epoch": 0.5523809523809524,
+      "grad_norm": 0.30888280272483826,
+      "learning_rate": 0.00018810121942857845,
+      "loss": 0.1768,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 58
+    },
+    {
+      "epoch": 0.5619047619047619,
+      "grad_norm": 0.27275580167770386,
+      "learning_rate": 0.00018761087664436138,
+      "loss": 0.1941,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 59
+    },
+    {
+      "epoch": 0.5714285714285714,
+      "grad_norm": 0.22851359844207764,
+      "learning_rate": 0.00018711129940866575,
+      "loss": 0.1658,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 60
+    },
+    {
+      "epoch": 0.580952380952381,
+      "grad_norm": 0.25515425205230713,
+      "learning_rate": 0.00018660254037844388,
+      "loss": 0.1858,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 61
+    },
+    {
+      "epoch": 0.5904761904761905,
+      "grad_norm": 0.24301765859127045,
+      "learning_rate": 0.00018608465317843678,
+      "loss": 0.1892,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 62
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.24329794943332672,
+      "learning_rate": 0.00018555769239552233,
+      "loss": 0.1719,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 63
+    },
+    {
+      "epoch": 0.6095238095238096,
+      "grad_norm": 0.3206610381603241,
+      "learning_rate": 0.00018502171357296144,
+      "loss": 0.2075,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 64
+    },
+    {
+      "epoch": 0.6190476190476191,
+      "grad_norm": 0.3089618980884552,
+      "learning_rate": 0.00018447677320454367,
+      "loss": 0.185,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 65
+    },
+    {
+      "epoch": 0.6285714285714286,
+      "grad_norm": 0.291916161775589,
+      "learning_rate": 0.00018392292872863267,
+      "loss": 0.1739,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 66
+    },
+    {
+      "epoch": 0.638095238095238,
+      "grad_norm": 0.25783228874206543,
+      "learning_rate": 0.00018336023852211195,
+      "loss": 0.2061,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 67
+    },
+    {
+      "epoch": 0.6476190476190476,
+      "grad_norm": 0.310614675283432,
+      "learning_rate": 0.00018278876189423179,
+      "loss": 0.1984,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 68
+    },
+    {
+      "epoch": 0.6571428571428571,
+      "grad_norm": 0.23836325109004974,
+      "learning_rate": 0.00018220855908035783,
+      "loss": 0.1742,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 69
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.30577945709228516,
+      "learning_rate": 0.0001816196912356222,
+      "loss": 0.1954,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 70
+    },
+    {
+      "epoch": 0.6761904761904762,
+      "grad_norm": 0.3177741467952728,
+      "learning_rate": 0.00018102222042847737,
+      "loss": 0.193,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 71
+    },
+    {
+      "epoch": 0.6857142857142857,
+      "grad_norm": 0.3444816470146179,
+      "learning_rate": 0.00018041620963415417,
+      "loss": 0.1837,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 72
+    },
+    {
+      "epoch": 0.6952380952380952,
+      "grad_norm": 0.3139638900756836,
+      "learning_rate": 0.000179801722728024,
+      "loss": 0.1902,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 73
+    },
+    {
+      "epoch": 0.7047619047619048,
+      "grad_norm": 0.24754184484481812,
+      "learning_rate": 0.00017917882447886582,
+      "loss": 0.1612,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 74
+    },
+    {
+      "epoch": 0.7142857142857143,
+      "grad_norm": 0.23204036056995392,
+      "learning_rate": 0.00017854758054203988,
+      "loss": 0.1694,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 75
+    },
+    {
+      "epoch": 0.7238095238095238,
+      "grad_norm": 0.2823260426521301,
+      "learning_rate": 0.00017790805745256704,
+      "loss": 0.1739,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 76
+    },
+    {
+      "epoch": 0.7333333333333333,
+      "grad_norm": 0.3193572163581848,
+      "learning_rate": 0.0001772603226181159,
+      "loss": 0.1924,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 77
+    },
+    {
+      "epoch": 0.7428571428571429,
+      "grad_norm": 0.30530717968940735,
+      "learning_rate": 0.0001766044443118978,
+      "loss": 0.1375,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 78
+    },
+    {
+      "epoch": 0.7523809523809524,
+      "grad_norm": 0.2974378764629364,
+      "learning_rate": 0.00017594049166547073,
+      "loss": 0.1866,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 79
+    },
+    {
+      "epoch": 0.7619047619047619,
+      "grad_norm": 0.26553574204444885,
+      "learning_rate": 0.00017526853466145244,
+      "loss": 0.193,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 80
+    },
+    {
+      "epoch": 0.7714285714285715,
+      "grad_norm": 0.32059594988822937,
+      "learning_rate": 0.00017458864412614434,
+      "loss": 0.1822,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 81
+    },
+    {
+      "epoch": 0.780952380952381,
+      "grad_norm": 0.3260822594165802,
+      "learning_rate": 0.00017390089172206592,
+      "loss": 0.1734,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 82
+    },
+    {
+      "epoch": 0.7904761904761904,
+      "grad_norm": 0.2754499912261963,
+      "learning_rate": 0.00017320534994040148,
+      "loss": 0.1814,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 83
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.2639872431755066,
+      "learning_rate": 0.00017250209209335927,
+      "loss": 0.1764,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 84
+    },
+    {
+      "epoch": 0.8095238095238095,
+      "grad_norm": 0.22905386984348297,
+      "learning_rate": 0.0001717911923064442,
+      "loss": 0.1365,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 85
+    },
+    {
+      "epoch": 0.819047619047619,
+      "grad_norm": 0.26005449891090393,
+      "learning_rate": 0.00017107272551064473,
+      "loss": 0.1756,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 86
+    },
+    {
+      "epoch": 0.8285714285714286,
+      "grad_norm": 0.25042417645454407,
+      "learning_rate": 0.00017034676743453499,
+      "loss": 0.15,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 87
+    },
+    {
+      "epoch": 0.8380952380952381,
+      "grad_norm": 0.34358277916908264,
+      "learning_rate": 0.0001696133945962927,
+      "loss": 0.1763,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 88
+    },
+    {
+      "epoch": 0.8476190476190476,
+      "grad_norm": 0.23855257034301758,
+      "learning_rate": 0.0001688726842956339,
+      "loss": 0.1547,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 89
+    },
+    {
+      "epoch": 0.8571428571428571,
+      "grad_norm": 0.2579350173473358,
+      "learning_rate": 0.0001681247146056654,
+      "loss": 0.1872,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 90
+    },
+    {
+      "epoch": 0.8666666666666667,
+      "grad_norm": 0.30034390091896057,
+      "learning_rate": 0.00016736956436465573,
+      "loss": 0.1558,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 91
+    },
+    {
+      "epoch": 0.8761904761904762,
+      "grad_norm": 0.28185054659843445,
+      "learning_rate": 0.00016660731316772505,
+      "loss": 0.1878,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 92
+    },
+    {
+      "epoch": 0.8857142857142857,
+      "grad_norm": 0.32514333724975586,
+      "learning_rate": 0.0001658380413584558,
+      "loss": 0.1705,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 93
+    },
+    {
+      "epoch": 0.8952380952380953,
+      "grad_norm": 0.30903860926628113,
+      "learning_rate": 0.0001650618300204242,
+      "loss": 0.1708,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 94
+    },
+    {
+      "epoch": 0.9047619047619048,
+      "grad_norm": 0.24723806977272034,
+      "learning_rate": 0.00016427876096865394,
+      "loss": 0.1571,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 95
+    },
+    {
+      "epoch": 0.9142857142857143,
+      "grad_norm": 0.2727813422679901,
+      "learning_rate": 0.0001634889167409923,
+      "loss": 0.1779,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 96
+    },
+    {
+      "epoch": 0.9238095238095239,
+      "grad_norm": 0.2521083354949951,
+      "learning_rate": 0.0001626923805894107,
+      "loss": 0.1451,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 97
+    },
+    {
+      "epoch": 0.9333333333333333,
+      "grad_norm": 0.25517475605010986,
+      "learning_rate": 0.00016188923647122947,
+      "loss": 0.1789,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 98
+    },
+    {
+      "epoch": 0.9428571428571428,
+      "grad_norm": 0.26679396629333496,
+      "learning_rate": 0.0001610795690402688,
+      "loss": 0.1835,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 99
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.22741852700710297,
+      "learning_rate": 0.00016026346363792567,
+      "loss": 0.1599,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 100
+    },
+    {
+      "epoch": 0.9619047619047619,
+      "grad_norm": 0.28691986203193665,
+      "learning_rate": 0.00015944100628417868,
+      "loss": 0.1654,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 101
+    },
+    {
+      "epoch": 0.9714285714285714,
+      "grad_norm": 0.24840302765369415,
+      "learning_rate": 0.00015861228366852148,
+      "loss": 0.1584,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 102
+    },
+    {
+      "epoch": 0.9809523809523809,
+      "grad_norm": 0.24804599583148956,
+      "learning_rate": 0.00015777738314082514,
+      "loss": 0.1391,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 103
+    },
+    {
+      "epoch": 0.9904761904761905,
+      "grad_norm": 0.2463735193014145,
+      "learning_rate": 0.00015693639270213136,
+      "loss": 0.1408,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 104
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.29896217584609985,
+      "learning_rate": 0.000156089400995377,
+      "loss": 0.1666,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 105
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 315,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.6079273462857728e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-105/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:232e77a355a770b00d79212785888b9356c3ed6653e1b633e2e2153e01eae951
+size 6776

checkpoint-210/README.md

@@ -0,0 +1,208 @@
+---
+base_model: Qwen/Qwen3-4B-Instruct-2507
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- axolotl
+- base_model:adapter:Qwen/Qwen3-4B-Instruct-2507
+- 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 0.17.0

checkpoint-210/adapter_config.json

@@ -0,0 +1,42 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-4B-Instruct-2507",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": null,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "qalora_group_size": 16,
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "gate_proj",
+    "up_proj",
+    "v_proj",
+    "down_proj",
+    "k_proj",
+    "o_proj"
+  ],
+  "target_parameters": [],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoint-210/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a5ec294b0528c1151fe3a0a022417d82f5cf536ff45577bcf337eb630a64554c
+size 66126768

checkpoint-210/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
+}

checkpoint-210/chat_template.jinja

@@ -0,0 +1,86 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if message.content is string %}
+        {%- set content = message.content %}
+    {%- else %}
+        {%- set content = '' %}
+    {%- endif %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

checkpoint-210/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bcad621f4ef1b1fae68830a27a7f62e353fb5835c7d2867f34698e414c59f356
+size 34135860

checkpoint-210/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:476e7fc8020b17595f3ef4dad3e9bb8f476f39100a9c32966926557b9d34a5e3
+size 14512

checkpoint-210/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2be463597d503ef315c7bcad8c17a62b22c0357b52eab8a909a62ea745c2e290
+size 14512

checkpoint-210/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b3262c00bc04620eb4ac6df142b7b0caf7dc06645441af56eb661f2b3afbc3f
+size 1064

checkpoint-210/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
+  }
+}

checkpoint-210/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

checkpoint-210/tokenizer_config.json

@@ -0,0 +1,239 @@
+{
+  "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,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 262144,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoint-210/trainer_state.json

@@ -0,0 +1,2134 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 210,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.009523809523809525,
+      "grad_norm": 2.447430372238159,
+      "learning_rate": 0.0,
+      "loss": 1.8171,
+      "memory/device_mem_reserved(gib)": 36.42,
+      "memory/max_mem_active(gib)": 35.17,
+      "memory/max_mem_allocated(gib)": 35.17,
+      "step": 1
+    },
+    {
+      "epoch": 0.01904761904761905,
+      "grad_norm": 2.280487537384033,
+      "learning_rate": 2.2222222222222223e-05,
+      "loss": 1.7544,
+      "memory/device_mem_reserved(gib)": 36.45,
+      "memory/max_mem_active(gib)": 35.23,
+      "memory/max_mem_allocated(gib)": 35.23,
+      "step": 2
+    },
+    {
+      "epoch": 0.02857142857142857,
+      "grad_norm": 2.248966693878174,
+      "learning_rate": 4.4444444444444447e-05,
+      "loss": 1.7546,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 3
+    },
+    {
+      "epoch": 0.0380952380952381,
+      "grad_norm": 2.2559540271759033,
+      "learning_rate": 6.666666666666667e-05,
+      "loss": 1.8011,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 4
+    },
+    {
+      "epoch": 0.047619047619047616,
+      "grad_norm": 2.149872064590454,
+      "learning_rate": 8.888888888888889e-05,
+      "loss": 1.6465,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 5
+    },
+    {
+      "epoch": 0.05714285714285714,
+      "grad_norm": 1.6111228466033936,
+      "learning_rate": 0.00011111111111111112,
+      "loss": 1.5935,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 6
+    },
+    {
+      "epoch": 0.06666666666666667,
+      "grad_norm": 1.363952875137329,
+      "learning_rate": 0.00013333333333333334,
+      "loss": 1.4371,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 7
+    },
+    {
+      "epoch": 0.0761904761904762,
+      "grad_norm": 1.1439425945281982,
+      "learning_rate": 0.00015555555555555556,
+      "loss": 1.2409,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 8
+    },
+    {
+      "epoch": 0.08571428571428572,
+      "grad_norm": 0.877252995967865,
+      "learning_rate": 0.00017777777777777779,
+      "loss": 1.1083,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 9
+    },
+    {
+      "epoch": 0.09523809523809523,
+      "grad_norm": 0.8071479797363281,
+      "learning_rate": 0.0002,
+      "loss": 0.9937,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 10
+    },
+    {
+      "epoch": 0.10476190476190476,
+      "grad_norm": 0.7772606015205383,
+      "learning_rate": 0.00019999472984871732,
+      "loss": 0.9235,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 11
+    },
+    {
+      "epoch": 0.11428571428571428,
+      "grad_norm": 0.811764657497406,
+      "learning_rate": 0.00019997891995035912,
+      "loss": 0.7635,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 12
+    },
+    {
+      "epoch": 0.12380952380952381,
+      "grad_norm": 0.8423659801483154,
+      "learning_rate": 0.0001999525719713366,
+      "loss": 0.6836,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 13
+    },
+    {
+      "epoch": 0.13333333333333333,
+      "grad_norm": 0.7858280539512634,
+      "learning_rate": 0.0001999156886888064,
+      "loss": 0.6159,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 14
+    },
+    {
+      "epoch": 0.14285714285714285,
+      "grad_norm": 0.739560604095459,
+      "learning_rate": 0.00019986827399037812,
+      "loss": 0.5082,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 15
+    },
+    {
+      "epoch": 0.1523809523809524,
+      "grad_norm": 0.6787500381469727,
+      "learning_rate": 0.00019981033287370443,
+      "loss": 0.4553,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 16
+    },
+    {
+      "epoch": 0.1619047619047619,
+      "grad_norm": 0.6258607506752014,
+      "learning_rate": 0.00019974187144595432,
+      "loss": 0.3913,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 17
+    },
+    {
+      "epoch": 0.17142857142857143,
+      "grad_norm": 0.47939175367355347,
+      "learning_rate": 0.00019966289692316944,
+      "loss": 0.3048,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 18
+    },
+    {
+      "epoch": 0.18095238095238095,
+      "grad_norm": 0.49321815371513367,
+      "learning_rate": 0.00019957341762950344,
+      "loss": 0.3417,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 19
+    },
+    {
+      "epoch": 0.19047619047619047,
+      "grad_norm": 0.4252234101295471,
+      "learning_rate": 0.00019947344299634464,
+      "loss": 0.3033,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 20
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.46225014328956604,
+      "learning_rate": 0.00019936298356132176,
+      "loss": 0.2984,
+      "memory/device_mem_reserved(gib)": 47.1,
+      "memory/max_mem_active(gib)": 45.45,
+      "memory/max_mem_allocated(gib)": 45.45,
+      "step": 21
+    },
+    {
+      "epoch": 0.20952380952380953,
+      "grad_norm": 0.4106466472148895,
+      "learning_rate": 0.0001992420509671936,
+      "loss": 0.2764,
+      "memory/device_mem_reserved(gib)": 47.12,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 22
+    },
+    {
+      "epoch": 0.21904761904761905,
+      "grad_norm": 0.3764067590236664,
+      "learning_rate": 0.00019911065796062135,
+      "loss": 0.2352,
+      "memory/device_mem_reserved(gib)": 47.12,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 23
+    },
+    {
+      "epoch": 0.22857142857142856,
+      "grad_norm": 0.38518962264060974,
+      "learning_rate": 0.00019896881839082556,
+      "loss": 0.2239,
+      "memory/device_mem_reserved(gib)": 47.12,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 24
+    },
+    {
+      "epoch": 0.23809523809523808,
+      "grad_norm": 0.39728403091430664,
+      "learning_rate": 0.00019881654720812594,
+      "loss": 0.2192,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 25
+    },
+    {
+      "epoch": 0.24761904761904763,
+      "grad_norm": 0.3667093813419342,
+      "learning_rate": 0.00019865386046236596,
+      "loss": 0.2834,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 26
+    },
+    {
+      "epoch": 0.2571428571428571,
+      "grad_norm": 0.32502660155296326,
+      "learning_rate": 0.00019848077530122083,
+      "loss": 0.2166,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 27
+    },
+    {
+      "epoch": 0.26666666666666666,
+      "grad_norm": 0.31225672364234924,
+      "learning_rate": 0.0001982973099683902,
+      "loss": 0.2238,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 28
+    },
+    {
+      "epoch": 0.2761904761904762,
+      "grad_norm": 0.27448731660842896,
+      "learning_rate": 0.00019810348380167527,
+      "loss": 0.1983,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 29
+    },
+    {
+      "epoch": 0.2857142857142857,
+      "grad_norm": 0.32014158368110657,
+      "learning_rate": 0.00019789931723094046,
+      "loss": 0.2554,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 30
+    },
+    {
+      "epoch": 0.29523809523809524,
+      "grad_norm": 0.39480143785476685,
+      "learning_rate": 0.0001976848317759601,
+      "loss": 0.2438,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 31
+    },
+    {
+      "epoch": 0.3047619047619048,
+      "grad_norm": 0.28004321455955505,
+      "learning_rate": 0.00019746005004415005,
+      "loss": 0.2145,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 32
+    },
+    {
+      "epoch": 0.3142857142857143,
+      "grad_norm": 0.2754361033439636,
+      "learning_rate": 0.00019722499572818496,
+      "loss": 0.2284,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.46,
+      "memory/max_mem_allocated(gib)": 45.46,
+      "step": 33
+    },
+    {
+      "epoch": 0.3238095238095238,
+      "grad_norm": 0.3160915970802307,
+      "learning_rate": 0.00019697969360350098,
+      "loss": 0.2482,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 34
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.24699093401432037,
+      "learning_rate": 0.00019672416952568416,
+      "loss": 0.1916,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 35
+    },
+    {
+      "epoch": 0.34285714285714286,
+      "grad_norm": 0.2674485743045807,
+      "learning_rate": 0.00019645845042774553,
+      "loss": 0.2204,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 36
+    },
+    {
+      "epoch": 0.3523809523809524,
+      "grad_norm": 0.33608749508857727,
+      "learning_rate": 0.00019618256431728194,
+      "loss": 0.235,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 37
+    },
+    {
+      "epoch": 0.3619047619047619,
+      "grad_norm": 0.29581594467163086,
+      "learning_rate": 0.00019589654027352414,
+      "loss": 0.2175,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 38
+    },
+    {
+      "epoch": 0.37142857142857144,
+      "grad_norm": 0.3526640832424164,
+      "learning_rate": 0.0001956004084442718,
+      "loss": 0.1989,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 39
+    },
+    {
+      "epoch": 0.38095238095238093,
+      "grad_norm": 0.2882135808467865,
+      "learning_rate": 0.00019529420004271567,
+      "loss": 0.2144,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 40
+    },
+    {
+      "epoch": 0.3904761904761905,
+      "grad_norm": 0.2675183117389679,
+      "learning_rate": 0.0001949779473441478,
+      "loss": 0.1987,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 41
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.2908264696598053,
+      "learning_rate": 0.00019465168368255946,
+      "loss": 0.1893,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 42
+    },
+    {
+      "epoch": 0.4095238095238095,
+      "grad_norm": 0.3211234211921692,
+      "learning_rate": 0.00019431544344712776,
+      "loss": 0.2008,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 43
+    },
+    {
+      "epoch": 0.41904761904761906,
+      "grad_norm": 0.2795293927192688,
+      "learning_rate": 0.00019396926207859084,
+      "loss": 0.1935,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 44
+    },
+    {
+      "epoch": 0.42857142857142855,
+      "grad_norm": 0.28679871559143066,
+      "learning_rate": 0.00019361317606551238,
+      "loss": 0.1987,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 45
+    },
+    {
+      "epoch": 0.4380952380952381,
+      "grad_norm": 0.3165067732334137,
+      "learning_rate": 0.00019324722294043558,
+      "loss": 0.2118,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 46
+    },
+    {
+      "epoch": 0.44761904761904764,
+      "grad_norm": 0.25720369815826416,
+      "learning_rate": 0.00019287144127592704,
+      "loss": 0.1962,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 47
+    },
+    {
+      "epoch": 0.45714285714285713,
+      "grad_norm": 0.2577202022075653,
+      "learning_rate": 0.0001924858706805112,
+      "loss": 0.2022,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 48
+    },
+    {
+      "epoch": 0.4666666666666667,
+      "grad_norm": 0.2605520188808441,
+      "learning_rate": 0.0001920905517944954,
+      "loss": 0.1891,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 49
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "grad_norm": 0.2890350818634033,
+      "learning_rate": 0.00019168552628568631,
+      "loss": 0.17,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 50
+    },
+    {
+      "epoch": 0.4857142857142857,
+      "grad_norm": 0.25532475113868713,
+      "learning_rate": 0.00019127083684499806,
+      "loss": 0.1796,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 51
+    },
+    {
+      "epoch": 0.49523809523809526,
+      "grad_norm": 0.2963904142379761,
+      "learning_rate": 0.00019084652718195238,
+      "loss": 0.19,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 52
+    },
+    {
+      "epoch": 0.5047619047619047,
+      "grad_norm": 0.24455095827579498,
+      "learning_rate": 0.0001904126420200716,
+      "loss": 0.2017,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 53
+    },
+    {
+      "epoch": 0.5142857142857142,
+      "grad_norm": 0.2518145442008972,
+      "learning_rate": 0.00018996922709216455,
+      "loss": 0.1896,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 54
+    },
+    {
+      "epoch": 0.5238095238095238,
+      "grad_norm": 0.26225805282592773,
+      "learning_rate": 0.00018951632913550626,
+      "loss": 0.1893,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 55
+    },
+    {
+      "epoch": 0.5333333333333333,
+      "grad_norm": 0.25169169902801514,
+      "learning_rate": 0.00018905399588691163,
+      "loss": 0.1939,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 56
+    },
+    {
+      "epoch": 0.5428571428571428,
+      "grad_norm": 0.23847182095050812,
+      "learning_rate": 0.00018858227607770398,
+      "loss": 0.1686,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 57
+    },
+    {
+      "epoch": 0.5523809523809524,
+      "grad_norm": 0.30888280272483826,
+      "learning_rate": 0.00018810121942857845,
+      "loss": 0.1768,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 58
+    },
+    {
+      "epoch": 0.5619047619047619,
+      "grad_norm": 0.27275580167770386,
+      "learning_rate": 0.00018761087664436138,
+      "loss": 0.1941,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 59
+    },
+    {
+      "epoch": 0.5714285714285714,
+      "grad_norm": 0.22851359844207764,
+      "learning_rate": 0.00018711129940866575,
+      "loss": 0.1658,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 60
+    },
+    {
+      "epoch": 0.580952380952381,
+      "grad_norm": 0.25515425205230713,
+      "learning_rate": 0.00018660254037844388,
+      "loss": 0.1858,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 61
+    },
+    {
+      "epoch": 0.5904761904761905,
+      "grad_norm": 0.24301765859127045,
+      "learning_rate": 0.00018608465317843678,
+      "loss": 0.1892,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 62
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.24329794943332672,
+      "learning_rate": 0.00018555769239552233,
+      "loss": 0.1719,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 63
+    },
+    {
+      "epoch": 0.6095238095238096,
+      "grad_norm": 0.3206610381603241,
+      "learning_rate": 0.00018502171357296144,
+      "loss": 0.2075,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 64
+    },
+    {
+      "epoch": 0.6190476190476191,
+      "grad_norm": 0.3089618980884552,
+      "learning_rate": 0.00018447677320454367,
+      "loss": 0.185,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 65
+    },
+    {
+      "epoch": 0.6285714285714286,
+      "grad_norm": 0.291916161775589,
+      "learning_rate": 0.00018392292872863267,
+      "loss": 0.1739,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 66
+    },
+    {
+      "epoch": 0.638095238095238,
+      "grad_norm": 0.25783228874206543,
+      "learning_rate": 0.00018336023852211195,
+      "loss": 0.2061,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 67
+    },
+    {
+      "epoch": 0.6476190476190476,
+      "grad_norm": 0.310614675283432,
+      "learning_rate": 0.00018278876189423179,
+      "loss": 0.1984,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 68
+    },
+    {
+      "epoch": 0.6571428571428571,
+      "grad_norm": 0.23836325109004974,
+      "learning_rate": 0.00018220855908035783,
+      "loss": 0.1742,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 69
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.30577945709228516,
+      "learning_rate": 0.0001816196912356222,
+      "loss": 0.1954,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 70
+    },
+    {
+      "epoch": 0.6761904761904762,
+      "grad_norm": 0.3177741467952728,
+      "learning_rate": 0.00018102222042847737,
+      "loss": 0.193,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 71
+    },
+    {
+      "epoch": 0.6857142857142857,
+      "grad_norm": 0.3444816470146179,
+      "learning_rate": 0.00018041620963415417,
+      "loss": 0.1837,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 72
+    },
+    {
+      "epoch": 0.6952380952380952,
+      "grad_norm": 0.3139638900756836,
+      "learning_rate": 0.000179801722728024,
+      "loss": 0.1902,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 73
+    },
+    {
+      "epoch": 0.7047619047619048,
+      "grad_norm": 0.24754184484481812,
+      "learning_rate": 0.00017917882447886582,
+      "loss": 0.1612,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 74
+    },
+    {
+      "epoch": 0.7142857142857143,
+      "grad_norm": 0.23204036056995392,
+      "learning_rate": 0.00017854758054203988,
+      "loss": 0.1694,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 75
+    },
+    {
+      "epoch": 0.7238095238095238,
+      "grad_norm": 0.2823260426521301,
+      "learning_rate": 0.00017790805745256704,
+      "loss": 0.1739,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 76
+    },
+    {
+      "epoch": 0.7333333333333333,
+      "grad_norm": 0.3193572163581848,
+      "learning_rate": 0.0001772603226181159,
+      "loss": 0.1924,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 77
+    },
+    {
+      "epoch": 0.7428571428571429,
+      "grad_norm": 0.30530717968940735,
+      "learning_rate": 0.0001766044443118978,
+      "loss": 0.1375,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 78
+    },
+    {
+      "epoch": 0.7523809523809524,
+      "grad_norm": 0.2974378764629364,
+      "learning_rate": 0.00017594049166547073,
+      "loss": 0.1866,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 79
+    },
+    {
+      "epoch": 0.7619047619047619,
+      "grad_norm": 0.26553574204444885,
+      "learning_rate": 0.00017526853466145244,
+      "loss": 0.193,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 80
+    },
+    {
+      "epoch": 0.7714285714285715,
+      "grad_norm": 0.32059594988822937,
+      "learning_rate": 0.00017458864412614434,
+      "loss": 0.1822,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 81
+    },
+    {
+      "epoch": 0.780952380952381,
+      "grad_norm": 0.3260822594165802,
+      "learning_rate": 0.00017390089172206592,
+      "loss": 0.1734,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 82
+    },
+    {
+      "epoch": 0.7904761904761904,
+      "grad_norm": 0.2754499912261963,
+      "learning_rate": 0.00017320534994040148,
+      "loss": 0.1814,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 83
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.2639872431755066,
+      "learning_rate": 0.00017250209209335927,
+      "loss": 0.1764,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 84
+    },
+    {
+      "epoch": 0.8095238095238095,
+      "grad_norm": 0.22905386984348297,
+      "learning_rate": 0.0001717911923064442,
+      "loss": 0.1365,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 85
+    },
+    {
+      "epoch": 0.819047619047619,
+      "grad_norm": 0.26005449891090393,
+      "learning_rate": 0.00017107272551064473,
+      "loss": 0.1756,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 86
+    },
+    {
+      "epoch": 0.8285714285714286,
+      "grad_norm": 0.25042417645454407,
+      "learning_rate": 0.00017034676743453499,
+      "loss": 0.15,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 87
+    },
+    {
+      "epoch": 0.8380952380952381,
+      "grad_norm": 0.34358277916908264,
+      "learning_rate": 0.0001696133945962927,
+      "loss": 0.1763,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 88
+    },
+    {
+      "epoch": 0.8476190476190476,
+      "grad_norm": 0.23855257034301758,
+      "learning_rate": 0.0001688726842956339,
+      "loss": 0.1547,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 89
+    },
+    {
+      "epoch": 0.8571428571428571,
+      "grad_norm": 0.2579350173473358,
+      "learning_rate": 0.0001681247146056654,
+      "loss": 0.1872,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 90
+    },
+    {
+      "epoch": 0.8666666666666667,
+      "grad_norm": 0.30034390091896057,
+      "learning_rate": 0.00016736956436465573,
+      "loss": 0.1558,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 91
+    },
+    {
+      "epoch": 0.8761904761904762,
+      "grad_norm": 0.28185054659843445,
+      "learning_rate": 0.00016660731316772505,
+      "loss": 0.1878,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 92
+    },
+    {
+      "epoch": 0.8857142857142857,
+      "grad_norm": 0.32514333724975586,
+      "learning_rate": 0.0001658380413584558,
+      "loss": 0.1705,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 93
+    },
+    {
+      "epoch": 0.8952380952380953,
+      "grad_norm": 0.30903860926628113,
+      "learning_rate": 0.0001650618300204242,
+      "loss": 0.1708,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 94
+    },
+    {
+      "epoch": 0.9047619047619048,
+      "grad_norm": 0.24723806977272034,
+      "learning_rate": 0.00016427876096865394,
+      "loss": 0.1571,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 95
+    },
+    {
+      "epoch": 0.9142857142857143,
+      "grad_norm": 0.2727813422679901,
+      "learning_rate": 0.0001634889167409923,
+      "loss": 0.1779,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 96
+    },
+    {
+      "epoch": 0.9238095238095239,
+      "grad_norm": 0.2521083354949951,
+      "learning_rate": 0.0001626923805894107,
+      "loss": 0.1451,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 97
+    },
+    {
+      "epoch": 0.9333333333333333,
+      "grad_norm": 0.25517475605010986,
+      "learning_rate": 0.00016188923647122947,
+      "loss": 0.1789,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 98
+    },
+    {
+      "epoch": 0.9428571428571428,
+      "grad_norm": 0.26679396629333496,
+      "learning_rate": 0.0001610795690402688,
+      "loss": 0.1835,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 99
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.22741852700710297,
+      "learning_rate": 0.00016026346363792567,
+      "loss": 0.1599,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 100
+    },
+    {
+      "epoch": 0.9619047619047619,
+      "grad_norm": 0.28691986203193665,
+      "learning_rate": 0.00015944100628417868,
+      "loss": 0.1654,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 101
+    },
+    {
+      "epoch": 0.9714285714285714,
+      "grad_norm": 0.24840302765369415,
+      "learning_rate": 0.00015861228366852148,
+      "loss": 0.1584,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 102
+    },
+    {
+      "epoch": 0.9809523809523809,
+      "grad_norm": 0.24804599583148956,
+      "learning_rate": 0.00015777738314082514,
+      "loss": 0.1391,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 103
+    },
+    {
+      "epoch": 0.9904761904761905,
+      "grad_norm": 0.2463735193014145,
+      "learning_rate": 0.00015693639270213136,
+      "loss": 0.1408,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 104
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.29896217584609985,
+      "learning_rate": 0.000156089400995377,
+      "loss": 0.1666,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 105
+    },
+    {
+      "epoch": 1.0095238095238095,
+      "grad_norm": 0.2636161744594574,
+      "learning_rate": 0.0001552364972960506,
+      "loss": 0.1714,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 106
+    },
+    {
+      "epoch": 1.019047619047619,
+      "grad_norm": 0.2503117322921753,
+      "learning_rate": 0.00015437777150278267,
+      "loss": 0.1522,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 107
+    },
+    {
+      "epoch": 1.0285714285714285,
+      "grad_norm": 0.3032307028770447,
+      "learning_rate": 0.00015351331412787004,
+      "loss": 0.163,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 108
+    },
+    {
+      "epoch": 1.0380952380952382,
+      "grad_norm": 0.2652963399887085,
+      "learning_rate": 0.0001526432162877356,
+      "loss": 0.1589,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 109
+    },
+    {
+      "epoch": 1.0476190476190477,
+      "grad_norm": 0.2623184621334076,
+      "learning_rate": 0.00015176756969332425,
+      "loss": 0.1429,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 110
+    },
+    {
+      "epoch": 1.0571428571428572,
+      "grad_norm": 0.28222620487213135,
+      "learning_rate": 0.0001508864666404365,
+      "loss": 0.1469,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 111
+    },
+    {
+      "epoch": 1.0666666666666667,
+      "grad_norm": 0.28255829215049744,
+      "learning_rate": 0.00015000000000000001,
+      "loss": 0.156,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 112
+    },
+    {
+      "epoch": 1.0761904761904761,
+      "grad_norm": 0.22465772926807404,
+      "learning_rate": 0.00014910826320828084,
+      "loss": 0.131,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 113
+    },
+    {
+      "epoch": 1.0857142857142856,
+      "grad_norm": 0.30199363827705383,
+      "learning_rate": 0.0001482113502570349,
+      "loss": 0.1599,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 114
+    },
+    {
+      "epoch": 1.0952380952380953,
+      "grad_norm": 0.255288302898407,
+      "learning_rate": 0.00014730935568360102,
+      "loss": 0.1636,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 115
+    },
+    {
+      "epoch": 1.1047619047619048,
+      "grad_norm": 0.26142561435699463,
+      "learning_rate": 0.00014640237456093634,
+      "loss": 0.1518,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 116
+    },
+    {
+      "epoch": 1.1142857142857143,
+      "grad_norm": 0.30472439527511597,
+      "learning_rate": 0.00014549050248759547,
+      "loss": 0.1594,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 117
+    },
+    {
+      "epoch": 1.1238095238095238,
+      "grad_norm": 0.26955774426460266,
+      "learning_rate": 0.00014457383557765386,
+      "loss": 0.1676,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 118
+    },
+    {
+      "epoch": 1.1333333333333333,
+      "grad_norm": 0.34260597825050354,
+      "learning_rate": 0.00014365247045057734,
+      "loss": 0.1583,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 119
+    },
+    {
+      "epoch": 1.1428571428571428,
+      "grad_norm": 0.3136855661869049,
+      "learning_rate": 0.0001427265042210381,
+      "loss": 0.1558,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 120
+    },
+    {
+      "epoch": 1.1523809523809523,
+      "grad_norm": 0.27992814779281616,
+      "learning_rate": 0.00014179603448867835,
+      "loss": 0.1509,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 121
+    },
+    {
+      "epoch": 1.161904761904762,
+      "grad_norm": 0.34893789887428284,
+      "learning_rate": 0.00014086115932782314,
+      "loss": 0.1395,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 122
+    },
+    {
+      "epoch": 1.1714285714285715,
+      "grad_norm": 0.3308560252189636,
+      "learning_rate": 0.0001399219772771431,
+      "loss": 0.1591,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 123
+    },
+    {
+      "epoch": 1.180952380952381,
+      "grad_norm": 0.2736824154853821,
+      "learning_rate": 0.00013897858732926793,
+      "loss": 0.1604,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 124
+    },
+    {
+      "epoch": 1.1904761904761905,
+      "grad_norm": 0.28148430585861206,
+      "learning_rate": 0.0001380310889203526,
+      "loss": 0.1368,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 125
+    },
+    {
+      "epoch": 1.2,
+      "grad_norm": 0.25899434089660645,
+      "learning_rate": 0.00013707958191959608,
+      "loss": 0.15,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 126
+    },
+    {
+      "epoch": 1.2095238095238094,
+      "grad_norm": 0.32127872109413147,
+      "learning_rate": 0.00013612416661871533,
+      "loss": 0.1522,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 127
+    },
+    {
+      "epoch": 1.2190476190476192,
+      "grad_norm": 0.350427508354187,
+      "learning_rate": 0.00013516494372137368,
+      "loss": 0.1361,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 128
+    },
+    {
+      "epoch": 1.2285714285714286,
+      "grad_norm": 0.36603009700775146,
+      "learning_rate": 0.00013420201433256689,
+      "loss": 0.1527,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 129
+    },
+    {
+      "epoch": 1.2380952380952381,
+      "grad_norm": 0.2870017886161804,
+      "learning_rate": 0.00013323547994796597,
+      "loss": 0.152,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 130
+    },
+    {
+      "epoch": 1.2476190476190476,
+      "grad_norm": 0.29138877987861633,
+      "learning_rate": 0.0001322654424432195,
+      "loss": 0.1599,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 131
+    },
+    {
+      "epoch": 1.2571428571428571,
+      "grad_norm": 0.3358840048313141,
+      "learning_rate": 0.00013129200406321545,
+      "loss": 0.1668,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 132
+    },
+    {
+      "epoch": 1.2666666666666666,
+      "grad_norm": 0.3152446746826172,
+      "learning_rate": 0.00013031526741130435,
+      "loss": 0.1382,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 133
+    },
+    {
+      "epoch": 1.276190476190476,
+      "grad_norm": 0.2602708041667938,
+      "learning_rate": 0.00012933533543848461,
+      "loss": 0.1441,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 134
+    },
+    {
+      "epoch": 1.2857142857142856,
+      "grad_norm": 0.27083972096443176,
+      "learning_rate": 0.0001283523114325511,
+      "loss": 0.1517,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 135
+    },
+    {
+      "epoch": 1.2952380952380953,
+      "grad_norm": 0.27242034673690796,
+      "learning_rate": 0.0001273662990072083,
+      "loss": 0.1511,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 136
+    },
+    {
+      "epoch": 1.3047619047619048,
+      "grad_norm": 0.28473100066185,
+      "learning_rate": 0.0001263774020911492,
+      "loss": 0.1339,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 137
+    },
+    {
+      "epoch": 1.3142857142857143,
+      "grad_norm": 0.27939581871032715,
+      "learning_rate": 0.0001253857249171008,
+      "loss": 0.1414,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 138
+    },
+    {
+      "epoch": 1.3238095238095238,
+      "grad_norm": 0.2833685874938965,
+      "learning_rate": 0.00012439137201083773,
+      "loss": 0.139,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 139
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.3193693161010742,
+      "learning_rate": 0.0001233944481801649,
+      "loss": 0.159,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 140
+    },
+    {
+      "epoch": 1.342857142857143,
+      "grad_norm": 0.28556281328201294,
+      "learning_rate": 0.0001223950585038703,
+      "loss": 0.1549,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 141
+    },
+    {
+      "epoch": 1.3523809523809525,
+      "grad_norm": 0.2670241594314575,
+      "learning_rate": 0.00012139330832064974,
+      "loss": 0.1595,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 142
+    },
+    {
+      "epoch": 1.361904761904762,
+      "grad_norm": 0.20607773959636688,
+      "learning_rate": 0.00012038930321800346,
+      "loss": 0.1299,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 143
+    },
+    {
+      "epoch": 1.3714285714285714,
+      "grad_norm": 0.2973734736442566,
+      "learning_rate": 0.00011938314902110701,
+      "loss": 0.1552,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 144
+    },
+    {
+      "epoch": 1.380952380952381,
+      "grad_norm": 0.3032817244529724,
+      "learning_rate": 0.00011837495178165706,
+      "loss": 0.1554,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 145
+    },
+    {
+      "epoch": 1.3904761904761904,
+      "grad_norm": 0.2765556871891022,
+      "learning_rate": 0.00011736481776669306,
+      "loss": 0.1346,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 146
+    },
+    {
+      "epoch": 1.4,
+      "grad_norm": 0.24869301915168762,
+      "learning_rate": 0.00011635285344739651,
+      "loss": 0.1573,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 147
+    },
+    {
+      "epoch": 1.4095238095238094,
+      "grad_norm": 0.28329774737358093,
+      "learning_rate": 0.00011533916548786857,
+      "loss": 0.166,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 148
+    },
+    {
+      "epoch": 1.4190476190476191,
+      "grad_norm": 0.31719744205474854,
+      "learning_rate": 0.00011432386073388717,
+      "loss": 0.1556,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 149
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 0.27932024002075195,
+      "learning_rate": 0.00011330704620164538,
+      "loss": 0.1722,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 150
+    },
+    {
+      "epoch": 1.438095238095238,
+      "grad_norm": 0.2447134554386139,
+      "learning_rate": 0.00011228882906647142,
+      "loss": 0.1318,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 151
+    },
+    {
+      "epoch": 1.4476190476190476,
+      "grad_norm": 0.2684175670146942,
+      "learning_rate": 0.00011126931665153212,
+      "loss": 0.1465,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 152
+    },
+    {
+      "epoch": 1.457142857142857,
+      "grad_norm": 0.24706321954727173,
+      "learning_rate": 0.00011024861641652071,
+      "loss": 0.1226,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 153
+    },
+    {
+      "epoch": 1.4666666666666668,
+      "grad_norm": 0.2958153486251831,
+      "learning_rate": 0.00010922683594633021,
+      "loss": 0.1447,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 154
+    },
+    {
+      "epoch": 1.4761904761904763,
+      "grad_norm": 0.305876761674881,
+      "learning_rate": 0.00010820408293971378,
+      "loss": 0.1566,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 155
+    },
+    {
+      "epoch": 1.4857142857142858,
+      "grad_norm": 0.2755098342895508,
+      "learning_rate": 0.00010718046519793276,
+      "loss": 0.155,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 156
+    },
+    {
+      "epoch": 1.4952380952380953,
+      "grad_norm": 0.2550102174282074,
+      "learning_rate": 0.00010615609061339432,
+      "loss": 0.1227,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 157
+    },
+    {
+      "epoch": 1.5047619047619047,
+      "grad_norm": 0.26932892203330994,
+      "learning_rate": 0.00010513106715827896,
+      "loss": 0.1533,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 158
+    },
+    {
+      "epoch": 1.5142857142857142,
+      "grad_norm": 0.24724788963794708,
+      "learning_rate": 0.00010410550287316034,
+      "loss": 0.1346,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 159
+    },
+    {
+      "epoch": 1.5238095238095237,
+      "grad_norm": 0.2605031132698059,
+      "learning_rate": 0.00010307950585561706,
+      "loss": 0.1319,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 160
+    },
+    {
+      "epoch": 1.5333333333333332,
+      "grad_norm": 0.28353723883628845,
+      "learning_rate": 0.00010205318424883905,
+      "loss": 0.1386,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 161
+    },
+    {
+      "epoch": 1.5428571428571427,
+      "grad_norm": 0.31240761280059814,
+      "learning_rate": 0.00010102664623022899,
+      "loss": 0.1468,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 162
+    },
+    {
+      "epoch": 1.5523809523809524,
+      "grad_norm": 0.2614494562149048,
+      "learning_rate": 0.0001,
+      "loss": 0.146,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.47,
+      "memory/max_mem_allocated(gib)": 45.47,
+      "step": 163
+    },
+    {
+      "epoch": 1.561904761904762,
+      "grad_norm": 0.28343483805656433,
+      "learning_rate": 9.897335376977102e-05,
+      "loss": 0.1531,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 164
+    },
+    {
+      "epoch": 1.5714285714285714,
+      "grad_norm": 0.2741709053516388,
+      "learning_rate": 9.794681575116097e-05,
+      "loss": 0.1304,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 165
+    },
+    {
+      "epoch": 1.580952380952381,
+      "grad_norm": 0.326215922832489,
+      "learning_rate": 9.692049414438299e-05,
+      "loss": 0.1551,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 166
+    },
+    {
+      "epoch": 1.5904761904761906,
+      "grad_norm": 0.3120001256465912,
+      "learning_rate": 9.58944971268397e-05,
+      "loss": 0.1387,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 167
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 0.2673357129096985,
+      "learning_rate": 9.486893284172102e-05,
+      "loss": 0.1389,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 168
+    },
+    {
+      "epoch": 1.6095238095238096,
+      "grad_norm": 0.3184024691581726,
+      "learning_rate": 9.384390938660572e-05,
+      "loss": 0.1388,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 169
+    },
+    {
+      "epoch": 1.619047619047619,
+      "grad_norm": 0.28744614124298096,
+      "learning_rate": 9.281953480206725e-05,
+      "loss": 0.1246,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 170
+    },
+    {
+      "epoch": 1.6285714285714286,
+      "grad_norm": 0.28229352831840515,
+      "learning_rate": 9.179591706028626e-05,
+      "loss": 0.1637,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 171
+    },
+    {
+      "epoch": 1.638095238095238,
+      "grad_norm": 0.2750794589519501,
+      "learning_rate": 9.077316405366981e-05,
+      "loss": 0.1176,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 172
+    },
+    {
+      "epoch": 1.6476190476190475,
+      "grad_norm": 0.28753307461738586,
+      "learning_rate": 8.975138358347931e-05,
+      "loss": 0.1517,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 173
+    },
+    {
+      "epoch": 1.657142857142857,
+      "grad_norm": 0.2737794518470764,
+      "learning_rate": 8.87306833484679e-05,
+      "loss": 0.1604,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 174
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 0.2456161379814148,
+      "learning_rate": 8.77111709335286e-05,
+      "loss": 0.1441,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 175
+    },
+    {
+      "epoch": 1.6761904761904762,
+      "grad_norm": 0.2895718812942505,
+      "learning_rate": 8.669295379835467e-05,
+      "loss": 0.152,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 176
+    },
+    {
+      "epoch": 1.6857142857142857,
+      "grad_norm": 0.3128141462802887,
+      "learning_rate": 8.567613926611288e-05,
+      "loss": 0.1692,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 177
+    },
+    {
+      "epoch": 1.6952380952380952,
+      "grad_norm": 0.29811859130859375,
+      "learning_rate": 8.466083451213144e-05,
+      "loss": 0.1351,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 178
+    },
+    {
+      "epoch": 1.704761904761905,
+      "grad_norm": 0.2686918377876282,
+      "learning_rate": 8.364714655260349e-05,
+      "loss": 0.1318,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 179
+    },
+    {
+      "epoch": 1.7142857142857144,
+      "grad_norm": 0.31242185831069946,
+      "learning_rate": 8.263518223330697e-05,
+      "loss": 0.1512,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 180
+    },
+    {
+      "epoch": 1.723809523809524,
+      "grad_norm": 0.3040373623371124,
+      "learning_rate": 8.162504821834295e-05,
+      "loss": 0.1464,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 181
+    },
+    {
+      "epoch": 1.7333333333333334,
+      "grad_norm": 0.2847082018852234,
+      "learning_rate": 8.0616850978893e-05,
+      "loss": 0.1586,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 182
+    },
+    {
+      "epoch": 1.7428571428571429,
+      "grad_norm": 0.27451273798942566,
+      "learning_rate": 7.961069678199658e-05,
+      "loss": 0.1553,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 183
+    },
+    {
+      "epoch": 1.7523809523809524,
+      "grad_norm": 0.3282047212123871,
+      "learning_rate": 7.860669167935028e-05,
+      "loss": 0.1559,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 184
+    },
+    {
+      "epoch": 1.7619047619047619,
+      "grad_norm": 0.2895129919052124,
+      "learning_rate": 7.760494149612971e-05,
+      "loss": 0.1431,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 185
+    },
+    {
+      "epoch": 1.7714285714285714,
+      "grad_norm": 0.2585572898387909,
+      "learning_rate": 7.660555181983518e-05,
+      "loss": 0.1274,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 186
+    },
+    {
+      "epoch": 1.7809523809523808,
+      "grad_norm": 0.2550938129425049,
+      "learning_rate": 7.560862798916228e-05,
+      "loss": 0.1482,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 187
+    },
+    {
+      "epoch": 1.7904761904761903,
+      "grad_norm": 0.267459511756897,
+      "learning_rate": 7.461427508289922e-05,
+      "loss": 0.1425,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 188
+    },
+    {
+      "epoch": 1.8,
+      "grad_norm": 0.273129940032959,
+      "learning_rate": 7.36225979088508e-05,
+      "loss": 0.118,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 189
+    },
+    {
+      "epoch": 1.8095238095238095,
+      "grad_norm": 0.33791351318359375,
+      "learning_rate": 7.263370099279172e-05,
+      "loss": 0.146,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 190
+    },
+    {
+      "epoch": 1.819047619047619,
+      "grad_norm": 0.3014010488986969,
+      "learning_rate": 7.164768856744892e-05,
+      "loss": 0.1461,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 191
+    },
+    {
+      "epoch": 1.8285714285714287,
+      "grad_norm": 0.32012397050857544,
+      "learning_rate": 7.066466456151541e-05,
+      "loss": 0.1782,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 192
+    },
+    {
+      "epoch": 1.8380952380952382,
+      "grad_norm": 0.25507038831710815,
+      "learning_rate": 6.968473258869566e-05,
+      "loss": 0.1319,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 193
+    },
+    {
+      "epoch": 1.8476190476190477,
+      "grad_norm": 0.28470760583877563,
+      "learning_rate": 6.870799593678459e-05,
+      "loss": 0.159,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 194
+    },
+    {
+      "epoch": 1.8571428571428572,
+      "grad_norm": 0.25228601694107056,
+      "learning_rate": 6.773455755678054e-05,
+      "loss": 0.1284,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 195
+    },
+    {
+      "epoch": 1.8666666666666667,
+      "grad_norm": 0.28710344433784485,
+      "learning_rate": 6.676452005203406e-05,
+      "loss": 0.1338,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 196
+    },
+    {
+      "epoch": 1.8761904761904762,
+      "grad_norm": 0.29423221945762634,
+      "learning_rate": 6.579798566743314e-05,
+      "loss": 0.1335,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 197
+    },
+    {
+      "epoch": 1.8857142857142857,
+      "grad_norm": 0.31776273250579834,
+      "learning_rate": 6.483505627862632e-05,
+      "loss": 0.1509,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 198
+    },
+    {
+      "epoch": 1.8952380952380952,
+      "grad_norm": 0.2739746868610382,
+      "learning_rate": 6.387583338128471e-05,
+      "loss": 0.1195,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 199
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 0.28000062704086304,
+      "learning_rate": 6.292041808040393e-05,
+      "loss": 0.1444,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 200
+    },
+    {
+      "epoch": 1.9142857142857141,
+      "grad_norm": 0.2637692987918854,
+      "learning_rate": 6.196891107964744e-05,
+      "loss": 0.1412,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 201
+    },
+    {
+      "epoch": 1.9238095238095239,
+      "grad_norm": 0.2977365255355835,
+      "learning_rate": 6.102141267073207e-05,
+      "loss": 0.1224,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 202
+    },
+    {
+      "epoch": 1.9333333333333333,
+      "grad_norm": 0.2656545042991638,
+      "learning_rate": 6.007802272285693e-05,
+      "loss": 0.1235,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 203
+    },
+    {
+      "epoch": 1.9428571428571428,
+      "grad_norm": 0.3028334975242615,
+      "learning_rate": 5.913884067217685e-05,
+      "loss": 0.1544,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 204
+    },
+    {
+      "epoch": 1.9523809523809523,
+      "grad_norm": 0.32587432861328125,
+      "learning_rate": 5.82039655113217e-05,
+      "loss": 0.1484,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 205
+    },
+    {
+      "epoch": 1.961904761904762,
+      "grad_norm": 0.2999066710472107,
+      "learning_rate": 5.727349577896194e-05,
+      "loss": 0.1392,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 206
+    },
+    {
+      "epoch": 1.9714285714285715,
+      "grad_norm": 0.2822760045528412,
+      "learning_rate": 5.634752954942264e-05,
+      "loss": 0.1334,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 207
+    },
+    {
+      "epoch": 1.980952380952381,
+      "grad_norm": 0.4100743234157562,
+      "learning_rate": 5.542616442234618e-05,
+      "loss": 0.1295,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 208
+    },
+    {
+      "epoch": 1.9904761904761905,
+      "grad_norm": 0.22527122497558594,
+      "learning_rate": 5.450949751240456e-05,
+      "loss": 0.1377,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 209
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.33393946290016174,
+      "learning_rate": 5.359762543906368e-05,
+      "loss": 0.1251,
+      "memory/device_mem_reserved(gib)": 47.13,
+      "memory/max_mem_active(gib)": 45.48,
+      "memory/max_mem_allocated(gib)": 45.48,
+      "step": 210
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 315,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.236037886423859e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-210/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:232e77a355a770b00d79212785888b9356c3ed6653e1b633e2e2153e01eae951
+size 6776

checkpoint-315/README.md

@@ -0,0 +1,208 @@
+---
+base_model: Qwen/Qwen3-4B-Instruct-2507
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- axolotl
+- base_model:adapter:Qwen/Qwen3-4B-Instruct-2507
+- 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 0.17.0

checkpoint-315/adapter_config.json

@@ -0,0 +1,42 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-4B-Instruct-2507",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": null,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "qalora_group_size": 16,
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "gate_proj",
+    "up_proj",
+    "v_proj",
+    "down_proj",
+    "k_proj",
+    "o_proj"
+  ],
+  "target_parameters": [],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoint-315/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e45b31c137803c2351f0144a9d9cdee51846c1e3ed927ec07b20c1cc38ea3b8
+size 66126768

checkpoint-315/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
+}

checkpoint-315/chat_template.jinja

@@ -0,0 +1,86 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if message.content is string %}
+        {%- set content = message.content %}
+    {%- else %}
+        {%- set content = '' %}
+    {%- endif %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

checkpoint-315/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d44680f2fdfc4ac7c7dd0e7295df76182dc0e40fb6ac329f5cbc408263404a20
+size 34136372

checkpoint-315/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c2d7d2a5d0cd064a913774d809a2a7705ac965a3966e312d27f543b5be7db2a2
+size 14512

checkpoint-315/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee6d6b8378064132b3fefd1a59db9b8c5cbbd76ac6045dd4c4cc3f89819a94f9
+size 14512

checkpoint-315/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ae1d9962c79cee29f1198d64dd25fad600fe6cac4cd38a92b31a92040718648
+size 1064

checkpoint-315/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
+  }
+}

checkpoint-315/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654