Upload folder using huggingface_hub

README.md

@@ -0,0 +1,151 @@
+---
+license: other
+library_name: peft
+tags:
+- generated_from_trainer
+base_model: Qwen/Qwen1.5-MoE-A2.7B
+model-index:
+- name: out
+  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/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
+<details><summary>See axolotl config</summary>
+
+axolotl version: `0.4.0`
+```yaml
+base_model: Qwen/Qwen1.5-MoE-A2.7B
+trust_remote_code: true
+
+load_in_8bit: false
+load_in_4bit: true
+strict: false
+
+datasets:
+  - path: mhenrichsen/alpaca_2k_test
+    type: alpaca
+dataset_prepared_path:
+val_set_size: 0.05
+output_dir: ./out
+
+sequence_len: 1024  # supports up to 32k
+sample_packing: false
+pad_to_sequence_len: false
+
+adapter: qlora
+lora_model_dir:
+lora_r: 32
+lora_alpha: 16
+lora_dropout: 0.05
+lora_target_linear: true
+lora_fan_in_fan_out:
+
+wandb_project:
+wandb_entity:
+wandb_watch:
+wandb_name:
+wandb_log_model:
+
+gradient_accumulation_steps: 4
+micro_batch_size: 1
+num_epochs: 4
+optimizer: paged_adamw_8bit
+lr_scheduler: cosine
+learning_rate: 0.0002
+
+train_on_inputs: false
+group_by_length: false
+bf16: auto
+fp16:
+tf32: true
+
+gradient_checkpointing: true
+gradient_checkpointing_kwargs:
+  use_reentrant: false
+early_stopping_patience:
+resume_from_checkpoint:
+local_rank:
+logging_steps: 1
+xformers_attention:
+flash_attention: true
+
+warmup_steps: 10
+evals_per_epoch: 4
+saves_per_epoch: 1
+debug:
+deepspeed:
+weight_decay: 0.0
+fsdp:
+fsdp_config:
+special_tokens:
+
+```
+
+</details><br>
+
+# out
+
+This model is a fine-tuned version of [Qwen/Qwen1.5-MoE-A2.7B](https://huggingface.co/Qwen/Qwen1.5-MoE-A2.7B) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.2553
+
+## 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: 1
+- eval_batch_size: 1
+- seed: 42
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 4
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_steps: 10
+- num_epochs: 4
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 0.8629        | 0.0   | 1    | 0.9370          |
+| 0.6917        | 0.25  | 119  | 0.8805          |
+| 0.9783        | 0.5   | 238  | 0.8783          |
+| 0.9578        | 0.75  | 357  | 0.8827          |
+| 0.4772        | 1.0   | 476  | 0.8900          |
+| 0.4653        | 1.25  | 595  | 0.9620          |
+| 0.5907        | 1.5   | 714  | 0.9532          |
+| 0.7364        | 1.75  | 833  | 0.9360          |
+| 0.2611        | 2.0   | 952  | 0.9570          |
+| 0.1999        | 2.25  | 1071 | 1.0415          |
+| 0.1532        | 2.51  | 1190 | 1.0776          |
+| 0.0455        | 2.76  | 1309 | 1.0920          |
+| 0.087         | 3.01  | 1428 | 1.1094          |
+| 0.0183        | 3.26  | 1547 | 1.2266          |
+| 0.0135        | 3.51  | 1666 | 1.2604          |
+| 0.1929        | 3.76  | 1785 | 1.2553          |
+
+
+### Framework versions
+
+- PEFT 0.10.0
+- Transformers 4.40.0.dev0
+- Pytorch 2.1.2+cu118
+- Datasets 2.18.0
+- Tokenizers 0.15.0

adapter_config.json

@@ -0,0 +1,36 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen1.5-MoE-A2.7B",
+  "bias": "none",
+  "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_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "gate",
+    "k_proj",
+    "up_proj",
+    "v_proj",
+    "o_proj",
+    "gate_proj",
+    "shared_expert_gate",
+    "q_proj",
+    "down_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b4e646acfb9fe47cc84526e72d8ebad6e93d56a2fd915b9f046a2b64decdc067
+size 2048275154

added_tokens.json

@@ -0,0 +1,5 @@
+{
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644
+}

checkpoint-1425/README.md

@@ -0,0 +1,202 @@
+---
+library_name: peft
+base_model: Qwen/Qwen1.5-MoE-A2.7B
+---
+
+# 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 0.10.0

checkpoint-1425/adapter_config.json

@@ -0,0 +1,36 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen1.5-MoE-A2.7B",
+  "bias": "none",
+  "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_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "gate",
+    "k_proj",
+    "up_proj",
+    "v_proj",
+    "o_proj",
+    "gate_proj",
+    "shared_expert_gate",
+    "q_proj",
+    "down_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-1425/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:39397b262953cee50ba6700a289ec36963b2dbf8f14fd531aa59b7f2a75d7e18
+size 2046221176

checkpoint-1425/added_tokens.json

@@ -0,0 +1,5 @@
+{
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644
+}

checkpoint-1425/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d4de42a4cdcf27f5dbd07f1afcd899a98646a7dd44c7d4cf9305d7da2d36a87
+size 1035165446

checkpoint-1425/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0151af1c8026eeaf57a9210a2159908878492943f90054e4ec8a87ca920f5a15
+size 14244

checkpoint-1425/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e6ba97f0db9a24831fceacc8f55345fec933ba2886e40df551479fd07daac7a
+size 1064

checkpoint-1425/special_tokens_map.json

@@ -0,0 +1,20 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-1425/tokenizer_config.json

@@ -0,0 +1,43 @@
+{
+  "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
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoint-1425/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a86f7a599d796e7b48de991f534e6b4a66f2f12ac3aac1d99bcffb3fefbdb76d
+size 5816

checkpoint-1900/README.md

@@ -0,0 +1,202 @@
+---
+library_name: peft
+base_model: Qwen/Qwen1.5-MoE-A2.7B
+---
+
+# 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 0.10.0

checkpoint-1900/adapter_config.json

@@ -0,0 +1,36 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen1.5-MoE-A2.7B",
+  "bias": "none",
+  "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_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "gate",
+    "k_proj",
+    "up_proj",
+    "v_proj",
+    "o_proj",
+    "gate_proj",
+    "shared_expert_gate",
+    "q_proj",
+    "down_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-1900/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4699694b8bcbffbaaf3317fb5e33bb899438d759504571c2b778241823bfc64d
+size 2046221176

checkpoint-1900/added_tokens.json

@@ -0,0 +1,5 @@
+{
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644
+}

checkpoint-1900/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d722de2017d50000f2c35fa44be3c6d1084dd81ee4c129572520a1822ed98bcd
+size 1035165446

checkpoint-1900/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:39cd718f410a0001352be67900cc95f8033ab65a1e56f98e06ce65e07f9a9ee4
+size 14244

checkpoint-1900/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ad8e63835341e965b25000c1b2abf3138a503da06dff17129ac435726dddc32
+size 1064

checkpoint-1900/special_tokens_map.json

@@ -0,0 +1,20 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-1900/tokenizer_config.json

@@ -0,0 +1,43 @@
+{
+  "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
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoint-1900/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a86f7a599d796e7b48de991f534e6b4a66f2f12ac3aac1d99bcffb3fefbdb76d
+size 5816

checkpoint-475/README.md

@@ -0,0 +1,202 @@
+---
+library_name: peft
+base_model: Qwen/Qwen1.5-MoE-A2.7B
+---
+
+# 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 0.10.0

checkpoint-475/adapter_config.json

@@ -0,0 +1,36 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen1.5-MoE-A2.7B",
+  "bias": "none",
+  "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_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "gate",
+    "k_proj",
+    "up_proj",
+    "v_proj",
+    "o_proj",
+    "gate_proj",
+    "shared_expert_gate",
+    "q_proj",
+    "down_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-475/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4627e913181f4c277829ffa2bf25f00aa67b9ca66e1d8e289c4104823fbe8a9e
+size 2046221176

checkpoint-475/added_tokens.json

@@ -0,0 +1,5 @@
+{
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644
+}

checkpoint-475/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3409eea3a658828d7c53c1c009dff40a8d879cd2cbeb27a172a88c3dd178dbc9
+size 1035165446

checkpoint-475/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:785d92de068d83d0334d184d93e86bdaaccda6531e4fd20d4f5fc74046802330
+size 14244

checkpoint-475/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f0fcd3293e4a09b8bf1933acfa5b9fdbd1d070eaf82e1bcd8a23bec4f4aab5fd
+size 1064

checkpoint-475/special_tokens_map.json

@@ -0,0 +1,20 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-475/tokenizer_config.json

@@ -0,0 +1,43 @@
+{
+  "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
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoint-475/trainer_state.json

@@ -0,0 +1,3378 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 119,
+  "global_step": 475,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.434104859828949,
+      "learning_rate": 2e-05,
+      "loss": 0.8629,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "eval_loss": 0.9369699358940125,
+      "eval_runtime": 61.2717,
+      "eval_samples_per_second": 1.632,
+      "eval_steps_per_second": 1.632,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.517804741859436,
+      "learning_rate": 4e-05,
+      "loss": 0.9794,
+      "step": 2
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 1.3174158334732056,
+      "learning_rate": 6e-05,
+      "loss": 1.5473,
+      "step": 3
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.6236504912376404,
+      "learning_rate": 8e-05,
+      "loss": 0.6592,
+      "step": 4
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.49080607295036316,
+      "learning_rate": 0.0001,
+      "loss": 1.1104,
+      "step": 5
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.43938764929771423,
+      "learning_rate": 0.00012,
+      "loss": 0.9446,
+      "step": 6
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.8905380368232727,
+      "learning_rate": 0.00014,
+      "loss": 0.6719,
+      "step": 7
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 0.9044575095176697,
+      "learning_rate": 0.00016,
+      "loss": 0.6471,
+      "step": 8
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 0.541214108467102,
+      "learning_rate": 0.00018,
+      "loss": 0.8339,
+      "step": 9
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 1.075484275817871,
+      "learning_rate": 0.0002,
+      "loss": 0.9616,
+      "step": 10
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 0.3413633406162262,
+      "learning_rate": 0.0001999998618515421,
+      "loss": 0.8251,
+      "step": 11
+    },
+    {
+      "epoch": 0.03,
+      "grad_norm": 0.7642049193382263,
+      "learning_rate": 0.00019999944740655014,
+      "loss": 1.3883,
+      "step": 12
+    },
+    {
+      "epoch": 0.03,
+      "grad_norm": 0.5759286284446716,
+      "learning_rate": 0.00019999875666616918,
+      "loss": 0.9162,
+      "step": 13
+    },
+    {
+      "epoch": 0.03,
+      "grad_norm": 1.16646409034729,
+      "learning_rate": 0.00019999778963230775,
+      "loss": 1.1735,
+      "step": 14
+    },
+    {
+      "epoch": 0.03,
+      "grad_norm": 0.5447037220001221,
+      "learning_rate": 0.0001999965463076377,
+      "loss": 1.2769,
+      "step": 15
+    },
+    {
+      "epoch": 0.03,
+      "grad_norm": 0.510498046875,
+      "learning_rate": 0.00019999502669559432,
+      "loss": 1.1439,
+      "step": 16
+    },
+    {
+      "epoch": 0.04,
+      "grad_norm": 0.514502227306366,
+      "learning_rate": 0.00019999323080037624,
+      "loss": 0.9929,
+      "step": 17
+    },
+    {
+      "epoch": 0.04,
+      "grad_norm": 0.2014903873205185,
+      "learning_rate": 0.00019999115862694546,
+      "loss": 0.6706,
+      "step": 18
+    },
+    {
+      "epoch": 0.04,
+      "grad_norm": 0.6941856741905212,
+      "learning_rate": 0.00019998881018102737,
+      "loss": 0.7576,
+      "step": 19
+    },
+    {
+      "epoch": 0.04,
+      "grad_norm": 0.29424500465393066,
+      "learning_rate": 0.00019998618546911056,
+      "loss": 1.5048,
+      "step": 20
+    },
+    {
+      "epoch": 0.04,
+      "grad_norm": 1.431168556213379,
+      "learning_rate": 0.00019998328449844714,
+      "loss": 0.761,
+      "step": 21
+    },
+    {
+      "epoch": 0.05,
+      "grad_norm": 2.672117233276367,
+      "learning_rate": 0.00019998010727705236,
+      "loss": 0.9434,
+      "step": 22
+    },
+    {
+      "epoch": 0.05,
+      "grad_norm": 0.40351876616477966,
+      "learning_rate": 0.00019997665381370477,
+      "loss": 1.2708,
+      "step": 23
+    },
+    {
+      "epoch": 0.05,
+      "grad_norm": 0.36313024163246155,
+      "learning_rate": 0.00019997292411794618,
+      "loss": 0.6931,
+      "step": 24
+    },
+    {
+      "epoch": 0.05,
+      "grad_norm": 0.7412884831428528,
+      "learning_rate": 0.00019996891820008164,
+      "loss": 0.827,
+      "step": 25
+    },
+    {
+      "epoch": 0.05,
+      "grad_norm": 0.7299063205718994,
+      "learning_rate": 0.00019996463607117935,
+      "loss": 0.7135,
+      "step": 26
+    },
+    {
+      "epoch": 0.06,
+      "grad_norm": 0.5188469290733337,
+      "learning_rate": 0.00019996007774307075,
+      "loss": 0.574,
+      "step": 27
+    },
+    {
+      "epoch": 0.06,
+      "grad_norm": 0.23084047436714172,
+      "learning_rate": 0.00019995524322835034,
+      "loss": 1.0736,
+      "step": 28
+    },
+    {
+      "epoch": 0.06,
+      "grad_norm": 0.7209138870239258,
+      "learning_rate": 0.00019995013254037574,
+      "loss": 0.9087,
+      "step": 29
+    },
+    {
+      "epoch": 0.06,
+      "grad_norm": 1.2266876697540283,
+      "learning_rate": 0.00019994474569326757,
+      "loss": 1.2464,
+      "step": 30
+    },
+    {
+      "epoch": 0.07,
+      "grad_norm": 0.6613232493400574,
+      "learning_rate": 0.0001999390827019096,
+      "loss": 0.9568,
+      "step": 31
+    },
+    {
+      "epoch": 0.07,
+      "grad_norm": 1.0048911571502686,
+      "learning_rate": 0.00019993314358194843,
+      "loss": 1.1935,
+      "step": 32
+    },
+    {
+      "epoch": 0.07,
+      "grad_norm": 0.9797032475471497,
+      "learning_rate": 0.00019992692834979372,
+      "loss": 1.1969,
+      "step": 33
+    },
+    {
+      "epoch": 0.07,
+      "grad_norm": 0.4784688353538513,
+      "learning_rate": 0.00019992043702261793,
+      "loss": 0.8674,
+      "step": 34
+    },
+    {
+      "epoch": 0.07,
+      "grad_norm": 2.863600015640259,
+      "learning_rate": 0.00019991366961835642,
+      "loss": 1.3276,
+      "step": 35
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 0.4809297025203705,
+      "learning_rate": 0.0001999066261557073,
+      "loss": 1.0116,
+      "step": 36
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 0.6229730844497681,
+      "learning_rate": 0.00019989930665413147,
+      "loss": 1.2794,
+      "step": 37
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 0.5807106494903564,
+      "learning_rate": 0.0001998917111338525,
+      "loss": 1.0074,
+      "step": 38
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 1.5811257362365723,
+      "learning_rate": 0.00019988383961585645,
+      "loss": 1.2491,
+      "step": 39
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 0.46693727374076843,
+      "learning_rate": 0.00019987569212189224,
+      "loss": 0.8432,
+      "step": 40
+    },
+    {
+      "epoch": 0.09,
+      "grad_norm": 0.9792713522911072,
+      "learning_rate": 0.00019986726867447107,
+      "loss": 0.8716,
+      "step": 41
+    },
+    {
+      "epoch": 0.09,
+      "grad_norm": 0.3240562081336975,
+      "learning_rate": 0.00019985856929686667,
+      "loss": 0.9934,
+      "step": 42
+    },
+    {
+      "epoch": 0.09,
+      "grad_norm": 0.7358172535896301,
+      "learning_rate": 0.0001998495940131152,
+      "loss": 0.7247,
+      "step": 43
+    },
+    {
+      "epoch": 0.09,
+      "grad_norm": 0.40497535467147827,
+      "learning_rate": 0.00019984034284801502,
+      "loss": 0.8262,
+      "step": 44
+    },
+    {
+      "epoch": 0.09,
+      "grad_norm": 0.36835265159606934,
+      "learning_rate": 0.00019983081582712685,
+      "loss": 1.0898,
+      "step": 45
+    },
+    {
+      "epoch": 0.1,
+      "grad_norm": 0.7027626633644104,
+      "learning_rate": 0.0001998210129767735,
+      "loss": 1.2949,
+      "step": 46
+    },
+    {
+      "epoch": 0.1,
+      "grad_norm": 0.35624387860298157,
+      "learning_rate": 0.00019981093432404006,
+      "loss": 0.9734,
+      "step": 47
+    },
+    {
+      "epoch": 0.1,
+      "grad_norm": 0.4896808862686157,
+      "learning_rate": 0.00019980057989677345,
+      "loss": 1.2297,
+      "step": 48
+    },
+    {
+      "epoch": 0.1,
+      "grad_norm": 0.9656649827957153,
+      "learning_rate": 0.00019978994972358265,
+      "loss": 0.8358,
+      "step": 49
+    },
+    {
+      "epoch": 0.11,
+      "grad_norm": 0.644644021987915,
+      "learning_rate": 0.0001997790438338385,
+      "loss": 0.7373,
+      "step": 50
+    },
+    {
+      "epoch": 0.11,
+      "grad_norm": 1.0634011030197144,
+      "learning_rate": 0.00019976786225767365,
+      "loss": 1.3792,
+      "step": 51
+    },
+    {
+      "epoch": 0.11,
+      "grad_norm": 0.9041422605514526,
+      "learning_rate": 0.00019975640502598244,
+      "loss": 0.856,
+      "step": 52
+    },
+    {
+      "epoch": 0.11,
+      "grad_norm": 0.5682844519615173,
+      "learning_rate": 0.00019974467217042085,
+      "loss": 0.6686,
+      "step": 53
+    },
+    {
+      "epoch": 0.11,
+      "grad_norm": 0.7197520136833191,
+      "learning_rate": 0.00019973266372340639,
+      "loss": 0.6806,
+      "step": 54
+    },
+    {
+      "epoch": 0.12,
+      "grad_norm": 1.2542258501052856,
+      "learning_rate": 0.00019972037971811802,
+      "loss": 1.353,
+      "step": 55
+    },
+    {
+      "epoch": 0.12,
+      "grad_norm": 0.6802650094032288,
+      "learning_rate": 0.0001997078201884961,
+      "loss": 1.0014,
+      "step": 56
+    },
+    {
+      "epoch": 0.12,
+      "grad_norm": 0.4886693060398102,
+      "learning_rate": 0.0001996949851692422,
+      "loss": 0.8613,
+      "step": 57
+    },
+    {
+      "epoch": 0.12,
+      "grad_norm": 1.0226856470108032,
+      "learning_rate": 0.0001996818746958191,
+      "loss": 1.0443,
+      "step": 58
+    },
+    {
+      "epoch": 0.12,
+      "grad_norm": 0.5514834523200989,
+      "learning_rate": 0.00019966848880445062,
+      "loss": 0.8816,
+      "step": 59
+    },
+    {
+      "epoch": 0.13,
+      "grad_norm": 0.4890052378177643,
+      "learning_rate": 0.00019965482753212156,
+      "loss": 0.8541,
+      "step": 60
+    },
+    {
+      "epoch": 0.13,
+      "grad_norm": 0.9011398553848267,
+      "learning_rate": 0.0001996408909165776,
+      "loss": 1.1158,
+      "step": 61
+    },
+    {
+      "epoch": 0.13,
+      "grad_norm": 0.7809276580810547,
+      "learning_rate": 0.00019962667899632518,
+      "loss": 0.6702,
+      "step": 62
+    },
+    {
+      "epoch": 0.13,
+      "grad_norm": 0.604097843170166,
+      "learning_rate": 0.00019961219181063142,
+      "loss": 1.1875,
+      "step": 63
+    },
+    {
+      "epoch": 0.13,
+      "grad_norm": 0.9003333449363708,
+      "learning_rate": 0.00019959742939952392,
+      "loss": 1.2126,
+      "step": 64
+    },
+    {
+      "epoch": 0.14,
+      "grad_norm": 0.59239661693573,
+      "learning_rate": 0.0001995823918037908,
+      "loss": 1.1083,
+      "step": 65
+    },
+    {
+      "epoch": 0.14,
+      "grad_norm": 0.6981655955314636,
+      "learning_rate": 0.00019956707906498044,
+      "loss": 0.9634,
+      "step": 66
+    },
+    {
+      "epoch": 0.14,
+      "grad_norm": 0.4635857045650482,
+      "learning_rate": 0.00019955149122540152,
+      "loss": 1.0345,
+      "step": 67
+    },
+    {
+      "epoch": 0.14,
+      "grad_norm": 0.35098958015441895,
+      "learning_rate": 0.00019953562832812272,
+      "loss": 0.8871,
+      "step": 68
+    },
+    {
+      "epoch": 0.15,
+      "grad_norm": 0.4510989189147949,
+      "learning_rate": 0.00019951949041697274,
+      "loss": 0.6967,
+      "step": 69
+    },
+    {
+      "epoch": 0.15,
+      "grad_norm": 0.39661434292793274,
+      "learning_rate": 0.00019950307753654017,
+      "loss": 1.0786,
+      "step": 70
+    },
+    {
+      "epoch": 0.15,
+      "grad_norm": 0.5836047530174255,
+      "learning_rate": 0.00019948638973217323,
+      "loss": 0.9265,
+      "step": 71
+    },
+    {
+      "epoch": 0.15,
+      "grad_norm": 0.4589115381240845,
+      "learning_rate": 0.00019946942704997982,
+      "loss": 0.6476,
+      "step": 72
+    },
+    {
+      "epoch": 0.15,
+      "grad_norm": 0.4495834410190582,
+      "learning_rate": 0.00019945218953682734,
+      "loss": 0.8693,
+      "step": 73
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 0.35905730724334717,
+      "learning_rate": 0.00019943467724034252,
+      "loss": 1.0325,
+      "step": 74
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 2.235016345977783,
+      "learning_rate": 0.0001994168902089112,
+      "loss": 1.3103,
+      "step": 75
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 0.36725524067878723,
+      "learning_rate": 0.00019939882849167852,
+      "loss": 0.908,
+      "step": 76
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 0.66635662317276,
+      "learning_rate": 0.0001993804921385484,
+      "loss": 0.682,
+      "step": 77
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 2.121004819869995,
+      "learning_rate": 0.0001993618812001836,
+      "loss": 0.8462,
+      "step": 78
+    },
+    {
+      "epoch": 0.17,
+      "grad_norm": 0.44895172119140625,
+      "learning_rate": 0.00019934299572800556,
+      "loss": 0.9625,
+      "step": 79
+    },
+    {
+      "epoch": 0.17,
+      "grad_norm": 0.5769445300102234,
+      "learning_rate": 0.00019932383577419432,
+      "loss": 0.7278,
+      "step": 80
+    },
+    {
+      "epoch": 0.17,
+      "grad_norm": 0.3807710111141205,
+      "learning_rate": 0.00019930440139168817,
+      "loss": 0.657,
+      "step": 81
+    },
+    {
+      "epoch": 0.17,
+      "grad_norm": 0.212838813662529,
+      "learning_rate": 0.00019928469263418374,
+      "loss": 0.3094,
+      "step": 82
+    },
+    {
+      "epoch": 0.17,
+      "grad_norm": 0.8039274215698242,
+      "learning_rate": 0.0001992647095561357,
+      "loss": 0.8898,
+      "step": 83
+    },
+    {
+      "epoch": 0.18,
+      "grad_norm": 0.7184070348739624,
+      "learning_rate": 0.00019924445221275675,
+      "loss": 0.8613,
+      "step": 84
+    },
+    {
+      "epoch": 0.18,
+      "grad_norm": 0.4697589874267578,
+      "learning_rate": 0.00019922392066001722,
+      "loss": 0.9533,
+      "step": 85
+    },
+    {
+      "epoch": 0.18,
+      "grad_norm": 0.7024903297424316,
+      "learning_rate": 0.00019920311495464518,
+      "loss": 0.7188,
+      "step": 86
+    },
+    {
+      "epoch": 0.18,
+      "grad_norm": 0.5008754730224609,
+      "learning_rate": 0.00019918203515412617,
+      "loss": 0.8329,
+      "step": 87
+    },
+    {
+      "epoch": 0.19,
+      "grad_norm": 0.6406499743461609,
+      "learning_rate": 0.00019916068131670302,
+      "loss": 1.4259,
+      "step": 88
+    },
+    {
+      "epoch": 0.19,
+      "grad_norm": 0.39489614963531494,
+      "learning_rate": 0.00019913905350137573,
+      "loss": 0.5831,
+      "step": 89
+    },
+    {
+      "epoch": 0.19,
+      "grad_norm": 0.7018424272537231,
+      "learning_rate": 0.0001991171517679013,
+      "loss": 0.9808,
+      "step": 90
+    },
+    {
+      "epoch": 0.19,
+      "grad_norm": 0.5011338591575623,
+      "learning_rate": 0.00019909497617679348,
+      "loss": 0.6806,
+      "step": 91
+    },
+    {
+      "epoch": 0.19,
+      "grad_norm": 0.2412053793668747,
+      "learning_rate": 0.0001990725267893228,
+      "loss": 1.0299,
+      "step": 92
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.4145759046077728,
+      "learning_rate": 0.00019904980366751624,
+      "loss": 1.4344,
+      "step": 93
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.5789082050323486,
+      "learning_rate": 0.00019902680687415705,
+      "loss": 0.4662,
+      "step": 94
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.28986847400665283,
+      "learning_rate": 0.00019900353647278466,
+      "loss": 1.296,
+      "step": 95
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.33722007274627686,
+      "learning_rate": 0.00019897999252769448,
+      "loss": 0.8011,
+      "step": 96
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.6796722412109375,
+      "learning_rate": 0.00019895617510393772,
+      "loss": 0.972,
+      "step": 97
+    },
+    {
+      "epoch": 0.21,
+      "grad_norm": 0.5858548879623413,
+      "learning_rate": 0.00019893208426732115,
+      "loss": 1.0073,
+      "step": 98
+    },
+    {
+      "epoch": 0.21,
+      "grad_norm": 0.5766484141349792,
+      "learning_rate": 0.00019890772008440704,
+      "loss": 0.7884,
+      "step": 99
+    },
+    {
+      "epoch": 0.21,
+      "grad_norm": 0.37647876143455505,
+      "learning_rate": 0.00019888308262251285,
+      "loss": 0.6407,
+      "step": 100
+    },
+    {
+      "epoch": 0.21,
+      "grad_norm": 1.5475112199783325,
+      "learning_rate": 0.00019885817194971117,
+      "loss": 1.1401,
+      "step": 101
+    },
+    {
+      "epoch": 0.21,
+      "grad_norm": 1.1082801818847656,
+      "learning_rate": 0.00019883298813482938,
+      "loss": 1.392,
+      "step": 102
+    },
+    {
+      "epoch": 0.22,
+      "grad_norm": 0.3952051103115082,
+      "learning_rate": 0.00019880753124744963,
+      "loss": 1.0498,
+      "step": 103
+    },
+    {
+      "epoch": 0.22,
+      "grad_norm": 0.19289755821228027,
+      "learning_rate": 0.00019878180135790845,
+      "loss": 0.4145,
+      "step": 104
+    },
+    {
+      "epoch": 0.22,
+      "grad_norm": 0.5658400654792786,
+      "learning_rate": 0.00019875579853729676,
+      "loss": 1.0984,
+      "step": 105
+    },
+    {
+      "epoch": 0.22,
+      "grad_norm": 0.8976437449455261,
+      "learning_rate": 0.00019872952285745959,
+      "loss": 0.6919,
+      "step": 106
+    },
+    {
+      "epoch": 0.23,
+      "grad_norm": 0.5265024900436401,
+      "learning_rate": 0.00019870297439099577,
+      "loss": 1.2932,
+      "step": 107
+    },
+    {
+      "epoch": 0.23,
+      "grad_norm": 0.8367021083831787,
+      "learning_rate": 0.00019867615321125795,
+      "loss": 1.4497,
+      "step": 108
+    },
+    {
+      "epoch": 0.23,
+      "grad_norm": 0.5223955512046814,
+      "learning_rate": 0.00019864905939235214,
+      "loss": 1.0325,
+      "step": 109
+    },
+    {
+      "epoch": 0.23,
+      "grad_norm": 1.408779501914978,
+      "learning_rate": 0.00019862169300913785,
+      "loss": 0.9026,
+      "step": 110
+    },
+    {
+      "epoch": 0.23,
+      "grad_norm": 0.24817530810832977,
+      "learning_rate": 0.00019859405413722746,
+      "loss": 0.826,
+      "step": 111
+    },
+    {
+      "epoch": 0.24,
+      "grad_norm": 1.450430154800415,
+      "learning_rate": 0.0001985661428529863,
+      "loss": 0.9791,
+      "step": 112
+    },
+    {
+      "epoch": 0.24,
+      "grad_norm": 0.4882315993309021,
+      "learning_rate": 0.0001985379592335325,
+      "loss": 0.7889,
+      "step": 113
+    },
+    {
+      "epoch": 0.24,
+      "grad_norm": 0.42783451080322266,
+      "learning_rate": 0.00019850950335673643,
+      "loss": 1.1608,
+      "step": 114
+    },
+    {
+      "epoch": 0.24,
+      "grad_norm": 1.0337790250778198,
+      "learning_rate": 0.00019848077530122083,
+      "loss": 1.045,
+      "step": 115
+    },
+    {
+      "epoch": 0.24,
+      "grad_norm": 0.3064819872379303,
+      "learning_rate": 0.00019845177514636042,
+      "loss": 0.6474,
+      "step": 116
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 0.29662173986434937,
+      "learning_rate": 0.00019842250297228176,
+      "loss": 0.9493,
+      "step": 117
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 0.515562891960144,
+      "learning_rate": 0.00019839295885986296,
+      "loss": 1.0605,
+      "step": 118
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 1.4514832496643066,
+      "learning_rate": 0.0001983631428907335,
+      "loss": 0.6917,
+      "step": 119
+    },
+    {
+      "epoch": 0.25,
+      "eval_loss": 0.8804921507835388,
+      "eval_runtime": 61.5233,
+      "eval_samples_per_second": 1.625,
+      "eval_steps_per_second": 1.625,
+      "step": 119
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 0.30757004022598267,
+      "learning_rate": 0.00019833305514727395,
+      "loss": 0.9722,
+      "step": 120
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 0.5162855386734009,
+      "learning_rate": 0.00019830269571261583,
+      "loss": 1.2197,
+      "step": 121
+    },
+    {
+      "epoch": 0.26,
+      "grad_norm": 0.5095639824867249,
+      "learning_rate": 0.00019827206467064133,
+      "loss": 0.8676,
+      "step": 122
+    },
+    {
+      "epoch": 0.26,
+      "grad_norm": 0.4804045557975769,
+      "learning_rate": 0.00019824116210598306,
+      "loss": 0.8565,
+      "step": 123
+    },
+    {
+      "epoch": 0.26,
+      "grad_norm": 0.28008362650871277,
+      "learning_rate": 0.0001982099881040239,
+      "loss": 0.9001,
+      "step": 124
+    },
+    {
+      "epoch": 0.26,
+      "grad_norm": 0.6209085583686829,
+      "learning_rate": 0.0001981785427508966,
+      "loss": 0.7188,
+      "step": 125
+    },
+    {
+      "epoch": 0.27,
+      "grad_norm": 0.32877278327941895,
+      "learning_rate": 0.0001981468261334837,
+      "loss": 0.6749,
+      "step": 126
+    },
+    {
+      "epoch": 0.27,
+      "grad_norm": 0.4256601631641388,
+      "learning_rate": 0.00019811483833941728,
+      "loss": 0.8086,
+      "step": 127
+    },
+    {
+      "epoch": 0.27,
+      "grad_norm": 1.1572288274765015,
+      "learning_rate": 0.0001980825794570786,
+      "loss": 0.8554,
+      "step": 128
+    },
+    {
+      "epoch": 0.27,
+      "grad_norm": 0.4987819194793701,
+      "learning_rate": 0.00019805004957559793,
+      "loss": 0.6999,
+      "step": 129
+    },
+    {
+      "epoch": 0.27,
+      "grad_norm": 0.6852537393569946,
+      "learning_rate": 0.00019801724878485438,
+      "loss": 0.8759,
+      "step": 130
+    },
+    {
+      "epoch": 0.28,
+      "grad_norm": 0.7970736622810364,
+      "learning_rate": 0.00019798417717547552,
+      "loss": 0.7471,
+      "step": 131
+    },
+    {
+      "epoch": 0.28,
+      "grad_norm": 0.5638220310211182,
+      "learning_rate": 0.00019795083483883715,
+      "loss": 1.0391,
+      "step": 132
+    },
+    {
+      "epoch": 0.28,
+      "grad_norm": 0.5482009649276733,
+      "learning_rate": 0.00019791722186706317,
+      "loss": 0.8363,
+      "step": 133
+    },
+    {
+      "epoch": 0.28,
+      "grad_norm": 0.23791633546352386,
+      "learning_rate": 0.0001978833383530251,
+      "loss": 0.725,
+      "step": 134
+    },
+    {
+      "epoch": 0.28,
+      "grad_norm": 0.5339345335960388,
+      "learning_rate": 0.00019784918439034216,
+      "loss": 0.9828,
+      "step": 135
+    },
+    {
+      "epoch": 0.29,
+      "grad_norm": 0.24769064784049988,
+      "learning_rate": 0.00019781476007338058,
+      "loss": 0.9496,
+      "step": 136
+    },
+    {
+      "epoch": 0.29,
+      "grad_norm": 0.46634215116500854,
+      "learning_rate": 0.00019778006549725375,
+      "loss": 1.0973,
+      "step": 137
+    },
+    {
+      "epoch": 0.29,
+      "grad_norm": 0.8007522821426392,
+      "learning_rate": 0.00019774510075782172,
+      "loss": 0.6847,
+      "step": 138
+    },
+    {
+      "epoch": 0.29,
+      "grad_norm": 0.5393804907798767,
+      "learning_rate": 0.00019770986595169096,
+      "loss": 0.6461,
+      "step": 139
+    },
+    {
+      "epoch": 0.29,
+      "grad_norm": 0.2891620695590973,
+      "learning_rate": 0.00019767436117621413,
+      "loss": 0.2937,
+      "step": 140
+    },
+    {
+      "epoch": 0.3,
+      "grad_norm": 0.9463545680046082,
+      "learning_rate": 0.0001976385865294899,
+      "loss": 0.4934,
+      "step": 141
+    },
+    {
+      "epoch": 0.3,
+      "grad_norm": 0.25647807121276855,
+      "learning_rate": 0.00019760254211036244,
+      "loss": 0.7446,
+      "step": 142
+    },
+    {
+      "epoch": 0.3,
+      "grad_norm": 0.49435535073280334,
+      "learning_rate": 0.00019756622801842143,
+      "loss": 0.3544,
+      "step": 143
+    },
+    {
+      "epoch": 0.3,
+      "grad_norm": 0.7826042175292969,
+      "learning_rate": 0.00019752964435400155,
+      "loss": 0.6972,
+      "step": 144
+    },
+    {
+      "epoch": 0.31,
+      "grad_norm": 0.7160178422927856,
+      "learning_rate": 0.00019749279121818235,
+      "loss": 0.9655,
+      "step": 145
+    },
+    {
+      "epoch": 0.31,
+      "grad_norm": 0.3925221264362335,
+      "learning_rate": 0.00019745566871278794,
+      "loss": 0.9041,
+      "step": 146
+    },
+    {
+      "epoch": 0.31,
+      "grad_norm": 0.5669321417808533,
+      "learning_rate": 0.0001974182769403866,
+      "loss": 0.9093,
+      "step": 147
+    },
+    {
+      "epoch": 0.31,
+      "grad_norm": 0.5025343298912048,
+      "learning_rate": 0.00019738061600429064,
+      "loss": 0.6226,
+      "step": 148
+    },
+    {
+      "epoch": 0.31,
+      "grad_norm": 1.1127972602844238,
+      "learning_rate": 0.0001973426860085561,
+      "loss": 0.7431,
+      "step": 149
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.4064362049102783,
+      "learning_rate": 0.00019730448705798239,
+      "loss": 0.8444,
+      "step": 150
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.9272475242614746,
+      "learning_rate": 0.00019726601925811204,
+      "loss": 0.836,
+      "step": 151
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.6594715118408203,
+      "learning_rate": 0.00019722728271523034,
+      "loss": 0.9031,
+      "step": 152
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.9399688839912415,
+      "learning_rate": 0.00019718827753636522,
+      "loss": 0.7959,
+      "step": 153
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.4452652633190155,
+      "learning_rate": 0.00019714900382928675,
+      "loss": 0.5638,
+      "step": 154
+    },
+    {
+      "epoch": 0.33,
+      "grad_norm": 0.47481146454811096,
+      "learning_rate": 0.000197109461702507,
+      "loss": 0.8291,
+      "step": 155
+    },
+    {
+      "epoch": 0.33,
+      "grad_norm": 0.1962246149778366,
+      "learning_rate": 0.00019706965126527963,
+      "loss": 0.7894,
+      "step": 156
+    },
+    {
+      "epoch": 0.33,
+      "grad_norm": 1.366571307182312,
+      "learning_rate": 0.00019702957262759965,
+      "loss": 1.1808,
+      "step": 157
+    },
+    {
+      "epoch": 0.33,
+      "grad_norm": 1.3261445760726929,
+      "learning_rate": 0.00019698922590020312,
+      "loss": 0.8769,
+      "step": 158
+    },
+    {
+      "epoch": 0.33,
+      "grad_norm": 0.5637160539627075,
+      "learning_rate": 0.00019694861119456679,
+      "loss": 0.882,
+      "step": 159
+    },
+    {
+      "epoch": 0.34,
+      "grad_norm": 0.4508800208568573,
+      "learning_rate": 0.0001969077286229078,
+      "loss": 1.2723,
+      "step": 160
+    },
+    {
+      "epoch": 0.34,
+      "grad_norm": 0.41292956471443176,
+      "learning_rate": 0.0001968665782981835,
+      "loss": 0.7919,
+      "step": 161
+    },
+    {
+      "epoch": 0.34,
+      "grad_norm": 0.6105634570121765,
+      "learning_rate": 0.00019682516033409092,
+      "loss": 1.0901,
+      "step": 162
+    },
+    {
+      "epoch": 0.34,
+      "grad_norm": 0.6460319757461548,
+      "learning_rate": 0.00019678347484506669,
+      "loss": 1.0425,
+      "step": 163
+    },
+    {
+      "epoch": 0.35,
+      "grad_norm": 0.8627430200576782,
+      "learning_rate": 0.00019674152194628638,
+      "loss": 0.8019,
+      "step": 164
+    },
+    {
+      "epoch": 0.35,
+      "grad_norm": 0.3218872547149658,
+      "learning_rate": 0.00019669930175366472,
+      "loss": 0.8345,
+      "step": 165
+    },
+    {
+      "epoch": 0.35,
+      "grad_norm": 0.6773053407669067,
+      "learning_rate": 0.00019665681438385473,
+      "loss": 1.3567,
+      "step": 166
+    },
+    {
+      "epoch": 0.35,
+      "grad_norm": 0.3802971839904785,
+      "learning_rate": 0.0001966140599542477,
+      "loss": 0.7315,
+      "step": 167
+    },
+    {
+      "epoch": 0.35,
+      "grad_norm": 0.9038891196250916,
+      "learning_rate": 0.0001965710385829728,
+      "loss": 0.6807,
+      "step": 168
+    },
+    {
+      "epoch": 0.36,
+      "grad_norm": 0.7831525802612305,
+      "learning_rate": 0.00019652775038889674,
+      "loss": 1.2796,
+      "step": 169
+    },
+    {
+      "epoch": 0.36,
+      "grad_norm": 0.3705346882343292,
+      "learning_rate": 0.00019648419549162348,
+      "loss": 0.8275,
+      "step": 170
+    },
+    {
+      "epoch": 0.36,
+      "grad_norm": 0.7794845104217529,
+      "learning_rate": 0.0001964403740114939,
+      "loss": 0.7539,
+      "step": 171
+    },
+    {
+      "epoch": 0.36,
+      "grad_norm": 0.2621815800666809,
+      "learning_rate": 0.00019639628606958533,
+      "loss": 0.976,
+      "step": 172
+    },
+    {
+      "epoch": 0.36,
+      "grad_norm": 0.6929745674133301,
+      "learning_rate": 0.00019635193178771143,
+      "loss": 0.6198,
+      "step": 173
+    },
+    {
+      "epoch": 0.37,
+      "grad_norm": 0.543230414390564,
+      "learning_rate": 0.0001963073112884217,
+      "loss": 0.9319,
+      "step": 174
+    },
+    {
+      "epoch": 0.37,
+      "grad_norm": 0.6732174158096313,
+      "learning_rate": 0.0001962624246950012,
+      "loss": 0.804,
+      "step": 175
+    },
+    {
+      "epoch": 0.37,
+      "grad_norm": 0.25452062487602234,
+      "learning_rate": 0.00019621727213147027,
+      "loss": 0.7632,
+      "step": 176
+    },
+    {
+      "epoch": 0.37,
+      "grad_norm": 0.6591973304748535,
+      "learning_rate": 0.00019617185372258392,
+      "loss": 0.9745,
+      "step": 177
+    },
+    {
+      "epoch": 0.37,
+      "grad_norm": 0.6275454163551331,
+      "learning_rate": 0.0001961261695938319,
+      "loss": 0.3411,
+      "step": 178
+    },
+    {
+      "epoch": 0.38,
+      "grad_norm": 0.6691128611564636,
+      "learning_rate": 0.00019608021987143804,
+      "loss": 0.9564,
+      "step": 179
+    },
+    {
+      "epoch": 0.38,
+      "grad_norm": 0.3190310299396515,
+      "learning_rate": 0.00019603400468235998,
+      "loss": 1.3002,
+      "step": 180
+    },
+    {
+      "epoch": 0.38,
+      "grad_norm": 0.4648153781890869,
+      "learning_rate": 0.0001959875241542889,
+      "loss": 0.9507,
+      "step": 181
+    },
+    {
+      "epoch": 0.38,
+      "grad_norm": 0.5921639800071716,
+      "learning_rate": 0.00019594077841564907,
+      "loss": 0.9397,
+      "step": 182
+    },
+    {
+      "epoch": 0.39,
+      "grad_norm": 0.5769446492195129,
+      "learning_rate": 0.00019589376759559745,
+      "loss": 0.9958,
+      "step": 183
+    },
+    {
+      "epoch": 0.39,
+      "grad_norm": 0.8454503417015076,
+      "learning_rate": 0.00019584649182402357,
+      "loss": 1.189,
+      "step": 184
+    },
+    {
+      "epoch": 0.39,
+      "grad_norm": 0.2865101099014282,
+      "learning_rate": 0.0001957989512315489,
+      "loss": 0.6747,
+      "step": 185
+    },
+    {
+      "epoch": 0.39,
+      "grad_norm": 0.3642055094242096,
+      "learning_rate": 0.0001957511459495266,
+      "loss": 0.5196,
+      "step": 186
+    },
+    {
+      "epoch": 0.39,
+      "grad_norm": 0.4965610206127167,
+      "learning_rate": 0.00019570307611004124,
+      "loss": 0.9448,
+      "step": 187
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.5694214105606079,
+      "learning_rate": 0.00019565474184590826,
+      "loss": 0.868,
+      "step": 188
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.6402484774589539,
+      "learning_rate": 0.00019560614329067378,
+      "loss": 0.8872,
+      "step": 189
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.37722048163414,
+      "learning_rate": 0.0001955572805786141,
+      "loss": 0.9253,
+      "step": 190
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 1.9157966375350952,
+      "learning_rate": 0.00019550815384473534,
+      "loss": 1.6508,
+      "step": 191
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.33376675844192505,
+      "learning_rate": 0.0001954587632247732,
+      "loss": 0.9109,
+      "step": 192
+    },
+    {
+      "epoch": 0.41,
+      "grad_norm": 0.2680880129337311,
+      "learning_rate": 0.00019540910885519242,
+      "loss": 1.0693,
+      "step": 193
+    },
+    {
+      "epoch": 0.41,
+      "grad_norm": 0.7726811766624451,
+      "learning_rate": 0.00019535919087318652,
+      "loss": 0.9574,
+      "step": 194
+    },
+    {
+      "epoch": 0.41,
+      "grad_norm": 0.8604207634925842,
+      "learning_rate": 0.0001953090094166773,
+      "loss": 0.9475,
+      "step": 195
+    },
+    {
+      "epoch": 0.41,
+      "grad_norm": 1.3954675197601318,
+      "learning_rate": 0.0001952585646243146,
+      "loss": 1.5094,
+      "step": 196
+    },
+    {
+      "epoch": 0.41,
+      "grad_norm": 0.39931145310401917,
+      "learning_rate": 0.00019520785663547586,
+      "loss": 0.9915,
+      "step": 197
+    },
+    {
+      "epoch": 0.42,
+      "grad_norm": 0.772156298160553,
+      "learning_rate": 0.00019515688559026563,
+      "loss": 1.4155,
+      "step": 198
+    },
+    {
+      "epoch": 0.42,
+      "grad_norm": 0.48633861541748047,
+      "learning_rate": 0.00019510565162951537,
+      "loss": 0.9607,
+      "step": 199
+    },
+    {
+      "epoch": 0.42,
+      "grad_norm": 0.4661516845226288,
+      "learning_rate": 0.0001950541548947829,
+      "loss": 1.0283,
+      "step": 200
+    },
+    {
+      "epoch": 0.42,
+      "grad_norm": 0.8846752047538757,
+      "learning_rate": 0.00019500239552835215,
+      "loss": 0.756,
+      "step": 201
+    },
+    {
+      "epoch": 0.43,
+      "grad_norm": 0.9870714545249939,
+      "learning_rate": 0.00019495037367323262,
+      "loss": 0.7688,
+      "step": 202
+    },
+    {
+      "epoch": 0.43,
+      "grad_norm": 0.7435501217842102,
+      "learning_rate": 0.00019489808947315915,
+      "loss": 0.4752,
+      "step": 203
+    },
+    {
+      "epoch": 0.43,
+      "grad_norm": 0.6509325504302979,
+      "learning_rate": 0.0001948455430725913,
+      "loss": 0.9053,
+      "step": 204
+    },
+    {
+      "epoch": 0.43,
+      "grad_norm": 0.30190637707710266,
+      "learning_rate": 0.0001947927346167132,
+      "loss": 0.9323,
+      "step": 205
+    },
+    {
+      "epoch": 0.43,
+      "grad_norm": 0.420055627822876,
+      "learning_rate": 0.00019473966425143292,
+      "loss": 0.6446,
+      "step": 206
+    },
+    {
+      "epoch": 0.44,
+      "grad_norm": 0.49513018131256104,
+      "learning_rate": 0.00019468633212338233,
+      "loss": 0.9022,
+      "step": 207
+    },
+    {
+      "epoch": 0.44,
+      "grad_norm": 0.4812709391117096,
+      "learning_rate": 0.00019463273837991643,
+      "loss": 0.6835,
+      "step": 208
+    },
+    {
+      "epoch": 0.44,
+      "grad_norm": 0.2101246416568756,
+      "learning_rate": 0.00019457888316911306,
+      "loss": 0.5991,
+      "step": 209
+    },
+    {
+      "epoch": 0.44,
+      "grad_norm": 0.3539298176765442,
+      "learning_rate": 0.00019452476663977248,
+      "loss": 0.7323,
+      "step": 210
+    },
+    {
+      "epoch": 0.44,
+      "grad_norm": 0.29954612255096436,
+      "learning_rate": 0.00019447038894141705,
+      "loss": 0.6868,
+      "step": 211
+    },
+    {
+      "epoch": 0.45,
+      "grad_norm": 0.4053567349910736,
+      "learning_rate": 0.00019441575022429065,
+      "loss": 1.0805,
+      "step": 212
+    },
+    {
+      "epoch": 0.45,
+      "grad_norm": 0.7733739614486694,
+      "learning_rate": 0.00019436085063935835,
+      "loss": 1.3524,
+      "step": 213
+    },
+    {
+      "epoch": 0.45,
+      "grad_norm": 0.6096423864364624,
+      "learning_rate": 0.00019430569033830605,
+      "loss": 1.0183,
+      "step": 214
+    },
+    {
+      "epoch": 0.45,
+      "grad_norm": 1.1940584182739258,
+      "learning_rate": 0.00019425026947353992,
+      "loss": 1.0919,
+      "step": 215
+    },
+    {
+      "epoch": 0.45,
+      "grad_norm": 0.4030895233154297,
+      "learning_rate": 0.00019419458819818614,
+      "loss": 0.7642,
+      "step": 216
+    },
+    {
+      "epoch": 0.46,
+      "grad_norm": 1.4116997718811035,
+      "learning_rate": 0.00019413864666609034,
+      "loss": 0.6112,
+      "step": 217
+    },
+    {
+      "epoch": 0.46,
+      "grad_norm": 0.4545953869819641,
+      "learning_rate": 0.00019408244503181724,
+      "loss": 0.7328,
+      "step": 218
+    },
+    {
+      "epoch": 0.46,
+      "grad_norm": 0.9334838390350342,
+      "learning_rate": 0.0001940259834506502,
+      "loss": 1.0518,
+      "step": 219
+    },
+    {
+      "epoch": 0.46,
+      "grad_norm": 0.2695348858833313,
+      "learning_rate": 0.00019396926207859084,
+      "loss": 0.987,
+      "step": 220
+    },
+    {
+      "epoch": 0.47,
+      "grad_norm": 1.3967281579971313,
+      "learning_rate": 0.00019391228107235858,
+      "loss": 1.0819,
+      "step": 221
+    },
+    {
+      "epoch": 0.47,
+      "grad_norm": 1.0220236778259277,
+      "learning_rate": 0.00019385504058939024,
+      "loss": 0.9621,
+      "step": 222
+    },
+    {
+      "epoch": 0.47,
+      "grad_norm": 2.5694682598114014,
+      "learning_rate": 0.00019379754078783937,
+      "loss": 1.0647,
+      "step": 223
+    },
+    {
+      "epoch": 0.47,
+      "grad_norm": 0.6181725263595581,
+      "learning_rate": 0.00019373978182657625,
+      "loss": 1.0991,
+      "step": 224
+    },
+    {
+      "epoch": 0.47,
+      "grad_norm": 0.508532702922821,
+      "learning_rate": 0.0001936817638651871,
+      "loss": 1.0276,
+      "step": 225
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.3763074278831482,
+      "learning_rate": 0.00019362348706397373,
+      "loss": 0.7447,
+      "step": 226
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.9533042311668396,
+      "learning_rate": 0.00019356495158395315,
+      "loss": 1.1979,
+      "step": 227
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.43593689799308777,
+      "learning_rate": 0.00019350615758685708,
+      "loss": 1.0028,
+      "step": 228
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.7646205425262451,
+      "learning_rate": 0.00019344710523513156,
+      "loss": 1.463,
+      "step": 229
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.29402196407318115,
+      "learning_rate": 0.00019338779469193639,
+      "loss": 1.2726,
+      "step": 230
+    },
+    {
+      "epoch": 0.49,
+      "grad_norm": 0.5773300528526306,
+      "learning_rate": 0.00019332822612114475,
+      "loss": 0.4847,
+      "step": 231
+    },
+    {
+      "epoch": 0.49,
+      "grad_norm": 1.0580178499221802,
+      "learning_rate": 0.00019326839968734279,
+      "loss": 1.0639,
+      "step": 232
+    },
+    {
+      "epoch": 0.49,
+      "grad_norm": 0.6212771534919739,
+      "learning_rate": 0.00019320831555582908,
+      "loss": 0.7302,
+      "step": 233
+    },
+    {
+      "epoch": 0.49,
+      "grad_norm": 1.1953450441360474,
+      "learning_rate": 0.00019314797389261424,
+      "loss": 0.9873,
+      "step": 234
+    },
+    {
+      "epoch": 0.49,
+      "grad_norm": 1.856995940208435,
+      "learning_rate": 0.00019308737486442045,
+      "loss": 0.9573,
+      "step": 235
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.36539939045906067,
+      "learning_rate": 0.00019302651863868092,
+      "loss": 0.6884,
+      "step": 236
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.3269266188144684,
+      "learning_rate": 0.0001929654053835395,
+      "loss": 0.9445,
+      "step": 237
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.46403074264526367,
+      "learning_rate": 0.00019290403526785025,
+      "loss": 0.9783,
+      "step": 238
+    },
+    {
+      "epoch": 0.5,
+      "eval_loss": 0.8782849311828613,
+      "eval_runtime": 61.3598,
+      "eval_samples_per_second": 1.63,
+      "eval_steps_per_second": 1.63,
+      "step": 238
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.6627680659294128,
+      "learning_rate": 0.00019284240846117697,
+      "loss": 0.9527,
+      "step": 239
+    },
+    {
+      "epoch": 0.51,
+      "grad_norm": 0.866802990436554,
+      "learning_rate": 0.00019278052513379255,
+      "loss": 0.6096,
+      "step": 240
+    },
+    {
+      "epoch": 0.51,
+      "grad_norm": 0.5304962396621704,
+      "learning_rate": 0.00019271838545667876,
+      "loss": 0.8335,
+      "step": 241
+    },
+    {
+      "epoch": 0.51,
+      "grad_norm": 1.076063632965088,
+      "learning_rate": 0.00019265598960152555,
+      "loss": 1.3308,
+      "step": 242
+    },
+    {
+      "epoch": 0.51,
+      "grad_norm": 2.491516351699829,
+      "learning_rate": 0.00019259333774073083,
+      "loss": 1.4458,
+      "step": 243
+    },
+    {
+      "epoch": 0.51,
+      "grad_norm": 1.3771064281463623,
+      "learning_rate": 0.00019253043004739968,
+      "loss": 1.4581,
+      "step": 244
+    },
+    {
+      "epoch": 0.52,
+      "grad_norm": 0.24413131177425385,
+      "learning_rate": 0.00019246726669534415,
+      "loss": 0.7537,
+      "step": 245
+    },
+    {
+      "epoch": 0.52,
+      "grad_norm": 1.02517831325531,
+      "learning_rate": 0.00019240384785908265,
+      "loss": 1.0646,
+      "step": 246
+    },
+    {
+      "epoch": 0.52,
+      "grad_norm": 0.4848421514034271,
+      "learning_rate": 0.00019234017371383945,
+      "loss": 0.6972,
+      "step": 247
+    },
+    {
+      "epoch": 0.52,
+      "grad_norm": 0.8870792388916016,
+      "learning_rate": 0.00019227624443554425,
+      "loss": 1.2114,
+      "step": 248
+    },
+    {
+      "epoch": 0.52,
+      "grad_norm": 0.5171313285827637,
+      "learning_rate": 0.00019221206020083166,
+      "loss": 0.7243,
+      "step": 249
+    },
+    {
+      "epoch": 0.53,
+      "grad_norm": 0.5975112915039062,
+      "learning_rate": 0.00019214762118704076,
+      "loss": 0.964,
+      "step": 250
+    },
+    {
+      "epoch": 0.53,
+      "grad_norm": 1.0921701192855835,
+      "learning_rate": 0.0001920829275722146,
+      "loss": 1.1413,
+      "step": 251
+    },
+    {
+      "epoch": 0.53,
+      "grad_norm": 0.6540035009384155,
+      "learning_rate": 0.00019201797953509955,
+      "loss": 0.9732,
+      "step": 252
+    },
+    {
+      "epoch": 0.53,
+      "grad_norm": 1.137863278388977,
+      "learning_rate": 0.0001919527772551451,
+      "loss": 1.3374,
+      "step": 253
+    },
+    {
+      "epoch": 0.53,
+      "grad_norm": 1.4139158725738525,
+      "learning_rate": 0.00019188732091250307,
+      "loss": 1.1147,
+      "step": 254
+    },
+    {
+      "epoch": 0.54,
+      "grad_norm": 0.5039550065994263,
+      "learning_rate": 0.00019182161068802741,
+      "loss": 0.7832,
+      "step": 255
+    },
+    {
+      "epoch": 0.54,
+      "grad_norm": 1.567670464515686,
+      "learning_rate": 0.00019175564676327339,
+      "loss": 0.6684,
+      "step": 256
+    },
+    {
+      "epoch": 0.54,
+      "grad_norm": 0.4372114837169647,
+      "learning_rate": 0.0001916894293204973,
+      "loss": 0.7285,
+      "step": 257
+    },
+    {
+      "epoch": 0.54,
+      "grad_norm": 0.4466225206851959,
+      "learning_rate": 0.00019162295854265594,
+      "loss": 0.5705,
+      "step": 258
+    },
+    {
+      "epoch": 0.55,
+      "grad_norm": 1.7975250482559204,
+      "learning_rate": 0.00019155623461340594,
+      "loss": 1.4155,
+      "step": 259
+    },
+    {
+      "epoch": 0.55,
+      "grad_norm": 0.6310514211654663,
+      "learning_rate": 0.00019148925771710347,
+      "loss": 0.7388,
+      "step": 260
+    },
+    {
+      "epoch": 0.55,
+      "grad_norm": 0.5273220539093018,
+      "learning_rate": 0.0001914220280388037,
+      "loss": 0.9241,
+      "step": 261
+    },
+    {
+      "epoch": 0.55,
+      "grad_norm": 1.8354101181030273,
+      "learning_rate": 0.0001913545457642601,
+      "loss": 0.8085,
+      "step": 262
+    },
+    {
+      "epoch": 0.55,
+      "grad_norm": 0.7362698316574097,
+      "learning_rate": 0.00019128681107992415,
+      "loss": 0.953,
+      "step": 263
+    },
+    {
+      "epoch": 0.56,
+      "grad_norm": 0.5334580540657043,
+      "learning_rate": 0.00019121882417294462,
+      "loss": 0.4416,
+      "step": 264
+    },
+    {
+      "epoch": 0.56,
+      "grad_norm": 0.6351854205131531,
+      "learning_rate": 0.00019115058523116733,
+      "loss": 0.6414,
+      "step": 265
+    },
+    {
+      "epoch": 0.56,
+      "grad_norm": 0.28386977314949036,
+      "learning_rate": 0.00019108209444313433,
+      "loss": 1.0273,
+      "step": 266
+    },
+    {
+      "epoch": 0.56,
+      "grad_norm": 0.5504246354103088,
+      "learning_rate": 0.00019101335199808354,
+      "loss": 1.1191,
+      "step": 267
+    },
+    {
+      "epoch": 0.56,
+      "grad_norm": 0.7449864149093628,
+      "learning_rate": 0.00019094435808594823,
+      "loss": 1.1073,
+      "step": 268
+    },
+    {
+      "epoch": 0.57,
+      "grad_norm": 0.6302490830421448,
+      "learning_rate": 0.00019087511289735644,
+      "loss": 1.2092,
+      "step": 269
+    },
+    {
+      "epoch": 0.57,
+      "grad_norm": 0.5618910789489746,
+      "learning_rate": 0.0001908056166236305,
+      "loss": 1.1966,
+      "step": 270
+    },
+    {
+      "epoch": 0.57,
+      "grad_norm": 0.46393775939941406,
+      "learning_rate": 0.0001907358694567865,
+      "loss": 0.7148,
+      "step": 271
+    },
+    {
+      "epoch": 0.57,
+      "grad_norm": 0.34640607237815857,
+      "learning_rate": 0.00019066587158953366,
+      "loss": 1.1297,
+      "step": 272
+    },
+    {
+      "epoch": 0.57,
+      "grad_norm": 1.3277580738067627,
+      "learning_rate": 0.00019059562321527396,
+      "loss": 1.0978,
+      "step": 273
+    },
+    {
+      "epoch": 0.58,
+      "grad_norm": 0.8730579018592834,
+      "learning_rate": 0.0001905251245281015,
+      "loss": 0.9732,
+      "step": 274
+    },
+    {
+      "epoch": 0.58,
+      "grad_norm": 0.32950034737586975,
+      "learning_rate": 0.00019045437572280194,
+      "loss": 1.0795,
+      "step": 275
+    },
+    {
+      "epoch": 0.58,
+      "grad_norm": 0.48170116543769836,
+      "learning_rate": 0.00019038337699485208,
+      "loss": 0.8124,
+      "step": 276
+    },
+    {
+      "epoch": 0.58,
+      "grad_norm": 0.858323335647583,
+      "learning_rate": 0.00019031212854041918,
+      "loss": 0.813,
+      "step": 277
+    },
+    {
+      "epoch": 0.59,
+      "grad_norm": 0.9366027116775513,
+      "learning_rate": 0.00019024063055636057,
+      "loss": 1.5074,
+      "step": 278
+    },
+    {
+      "epoch": 0.59,
+      "grad_norm": 0.4378308653831482,
+      "learning_rate": 0.00019016888324022296,
+      "loss": 0.8387,
+      "step": 279
+    },
+    {
+      "epoch": 0.59,
+      "grad_norm": 0.5781106948852539,
+      "learning_rate": 0.0001900968867902419,
+      "loss": 1.0496,
+      "step": 280
+    },
+    {
+      "epoch": 0.59,
+      "grad_norm": 0.834186851978302,
+      "learning_rate": 0.00019002464140534147,
+      "loss": 1.2684,
+      "step": 281
+    },
+    {
+      "epoch": 0.59,
+      "grad_norm": 0.752008855342865,
+      "learning_rate": 0.00018995214728513343,
+      "loss": 1.069,
+      "step": 282
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.3941871225833893,
+      "learning_rate": 0.0001898794046299167,
+      "loss": 0.942,
+      "step": 283
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.4069131314754486,
+      "learning_rate": 0.0001898064136406771,
+      "loss": 0.7116,
+      "step": 284
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.6478765606880188,
+      "learning_rate": 0.00018973317451908642,
+      "loss": 0.9494,
+      "step": 285
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 1.8658535480499268,
+      "learning_rate": 0.0001896596874675021,
+      "loss": 0.7592,
+      "step": 286
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.8622011542320251,
+      "learning_rate": 0.0001895859526889666,
+      "loss": 0.9392,
+      "step": 287
+    },
+    {
+      "epoch": 0.61,
+      "grad_norm": 0.8127020001411438,
+      "learning_rate": 0.00018951197038720688,
+      "loss": 1.3309,
+      "step": 288
+    },
+    {
+      "epoch": 0.61,
+      "grad_norm": 0.5042945146560669,
+      "learning_rate": 0.0001894377407666337,
+      "loss": 0.7607,
+      "step": 289
+    },
+    {
+      "epoch": 0.61,
+      "grad_norm": 0.7252426743507385,
+      "learning_rate": 0.00018936326403234125,
+      "loss": 1.1221,
+      "step": 290
+    },
+    {
+      "epoch": 0.61,
+      "grad_norm": 0.7334456443786621,
+      "learning_rate": 0.0001892885403901064,
+      "loss": 0.4738,
+      "step": 291
+    },
+    {
+      "epoch": 0.61,
+      "grad_norm": 2.6204662322998047,
+      "learning_rate": 0.00018921357004638835,
+      "loss": 1.2511,
+      "step": 292
+    },
+    {
+      "epoch": 0.62,
+      "grad_norm": 0.5708286762237549,
+      "learning_rate": 0.00018913835320832778,
+      "loss": 1.0887,
+      "step": 293
+    },
+    {
+      "epoch": 0.62,
+      "grad_norm": 1.0324314832687378,
+      "learning_rate": 0.00018906289008374655,
+      "loss": 1.1019,
+      "step": 294
+    },
+    {
+      "epoch": 0.62,
+      "grad_norm": 0.3663407862186432,
+      "learning_rate": 0.0001889871808811469,
+      "loss": 1.0333,
+      "step": 295
+    },
+    {
+      "epoch": 0.62,
+      "grad_norm": 0.7219849824905396,
+      "learning_rate": 0.00018891122580971098,
+      "loss": 0.858,
+      "step": 296
+    },
+    {
+      "epoch": 0.63,
+      "grad_norm": 0.7850363850593567,
+      "learning_rate": 0.00018883502507930042,
+      "loss": 0.9503,
+      "step": 297
+    },
+    {
+      "epoch": 0.63,
+      "grad_norm": 0.28012195229530334,
+      "learning_rate": 0.00018875857890045543,
+      "loss": 0.8068,
+      "step": 298
+    },
+    {
+      "epoch": 0.63,
+      "grad_norm": 0.7574068307876587,
+      "learning_rate": 0.00018868188748439444,
+      "loss": 0.7557,
+      "step": 299
+    },
+    {
+      "epoch": 0.63,
+      "grad_norm": 0.9131019711494446,
+      "learning_rate": 0.00018860495104301345,
+      "loss": 1.1462,
+      "step": 300
+    },
+    {
+      "epoch": 0.63,
+      "grad_norm": 0.24085545539855957,
+      "learning_rate": 0.00018852776978888551,
+      "loss": 0.8286,
+      "step": 301
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.4502617418766022,
+      "learning_rate": 0.00018845034393526005,
+      "loss": 1.0052,
+      "step": 302
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.7258254289627075,
+      "learning_rate": 0.00018837267369606228,
+      "loss": 0.9703,
+      "step": 303
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.6078888773918152,
+      "learning_rate": 0.00018829475928589271,
+      "loss": 0.8479,
+      "step": 304
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.5912296772003174,
+      "learning_rate": 0.00018821660092002641,
+      "loss": 1.0336,
+      "step": 305
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.3440995216369629,
+      "learning_rate": 0.0001881381988144126,
+      "loss": 0.7629,
+      "step": 306
+    },
+    {
+      "epoch": 0.65,
+      "grad_norm": 0.5613306164741516,
+      "learning_rate": 0.0001880595531856738,
+      "loss": 1.0355,
+      "step": 307
+    },
+    {
+      "epoch": 0.65,
+      "grad_norm": 0.5265874862670898,
+      "learning_rate": 0.0001879806642511055,
+      "loss": 0.9046,
+      "step": 308
+    },
+    {
+      "epoch": 0.65,
+      "grad_norm": 0.37300053238868713,
+      "learning_rate": 0.0001879015322286754,
+      "loss": 0.578,
+      "step": 309
+    },
+    {
+      "epoch": 0.65,
+      "grad_norm": 0.7948945164680481,
+      "learning_rate": 0.00018782215733702286,
+      "loss": 0.5693,
+      "step": 310
+    },
+    {
+      "epoch": 0.65,
+      "grad_norm": 0.5222792625427246,
+      "learning_rate": 0.0001877425397954582,
+      "loss": 0.812,
+      "step": 311
+    },
+    {
+      "epoch": 0.66,
+      "grad_norm": 0.6407319903373718,
+      "learning_rate": 0.00018766267982396224,
+      "loss": 0.7317,
+      "step": 312
+    },
+    {
+      "epoch": 0.66,
+      "grad_norm": 0.36041396856307983,
+      "learning_rate": 0.00018758257764318567,
+      "loss": 0.3617,
+      "step": 313
+    },
+    {
+      "epoch": 0.66,
+      "grad_norm": 0.6465966105461121,
+      "learning_rate": 0.00018750223347444828,
+      "loss": 0.6037,
+      "step": 314
+    },
+    {
+      "epoch": 0.66,
+      "grad_norm": 0.4281207025051117,
+      "learning_rate": 0.00018742164753973855,
+      "loss": 0.5269,
+      "step": 315
+    },
+    {
+      "epoch": 0.67,
+      "grad_norm": 0.3671799898147583,
+      "learning_rate": 0.00018734082006171299,
+      "loss": 0.66,
+      "step": 316
+    },
+    {
+      "epoch": 0.67,
+      "grad_norm": 0.4369129240512848,
+      "learning_rate": 0.00018725975126369535,
+      "loss": 1.1395,
+      "step": 317
+    },
+    {
+      "epoch": 0.67,
+      "grad_norm": 0.4631548523902893,
+      "learning_rate": 0.00018717844136967624,
+      "loss": 0.7871,
+      "step": 318
+    },
+    {
+      "epoch": 0.67,
+      "grad_norm": 0.4736942946910858,
+      "learning_rate": 0.00018709689060431242,
+      "loss": 1.2983,
+      "step": 319
+    },
+    {
+      "epoch": 0.67,
+      "grad_norm": 0.7346480488777161,
+      "learning_rate": 0.00018701509919292613,
+      "loss": 0.9507,
+      "step": 320
+    },
+    {
+      "epoch": 0.68,
+      "grad_norm": 0.5298660397529602,
+      "learning_rate": 0.00018693306736150444,
+      "loss": 0.6621,
+      "step": 321
+    },
+    {
+      "epoch": 0.68,
+      "grad_norm": 0.5501769781112671,
+      "learning_rate": 0.0001868507953366989,
+      "loss": 0.6954,
+      "step": 322
+    },
+    {
+      "epoch": 0.68,
+      "grad_norm": 1.565510630607605,
+      "learning_rate": 0.0001867682833458245,
+      "loss": 1.2279,
+      "step": 323
+    },
+    {
+      "epoch": 0.68,
+      "grad_norm": 0.2679019570350647,
+      "learning_rate": 0.00018668553161685933,
+      "loss": 0.6207,
+      "step": 324
+    },
+    {
+      "epoch": 0.68,
+      "grad_norm": 1.0185893774032593,
+      "learning_rate": 0.00018660254037844388,
+      "loss": 1.1179,
+      "step": 325
+    },
+    {
+      "epoch": 0.69,
+      "grad_norm": 0.400493323802948,
+      "learning_rate": 0.00018651930985988036,
+      "loss": 0.5947,
+      "step": 326
+    },
+    {
+      "epoch": 0.69,
+      "grad_norm": 0.7746186256408691,
+      "learning_rate": 0.00018643584029113215,
+      "loss": 1.0365,
+      "step": 327
+    },
+    {
+      "epoch": 0.69,
+      "grad_norm": 0.5792235136032104,
+      "learning_rate": 0.0001863521319028231,
+      "loss": 0.7102,
+      "step": 328
+    },
+    {
+      "epoch": 0.69,
+      "grad_norm": 0.35895833373069763,
+      "learning_rate": 0.00018626818492623688,
+      "loss": 0.5571,
+      "step": 329
+    },
+    {
+      "epoch": 0.69,
+      "grad_norm": 0.41158926486968994,
+      "learning_rate": 0.0001861839995933164,
+      "loss": 0.9009,
+      "step": 330
+    },
+    {
+      "epoch": 0.7,
+      "grad_norm": 0.5845640301704407,
+      "learning_rate": 0.00018609957613666315,
+      "loss": 0.3317,
+      "step": 331
+    },
+    {
+      "epoch": 0.7,
+      "grad_norm": 0.4458400309085846,
+      "learning_rate": 0.00018601491478953657,
+      "loss": 1.0094,
+      "step": 332
+    },
+    {
+      "epoch": 0.7,
+      "grad_norm": 0.6415822505950928,
+      "learning_rate": 0.00018593001578585326,
+      "loss": 0.9772,
+      "step": 333
+    },
+    {
+      "epoch": 0.7,
+      "grad_norm": 1.616220474243164,
+      "learning_rate": 0.00018584487936018661,
+      "loss": 0.6879,
+      "step": 334
+    },
+    {
+      "epoch": 0.71,
+      "grad_norm": 1.4885902404785156,
+      "learning_rate": 0.00018575950574776595,
+      "loss": 0.9627,
+      "step": 335
+    },
+    {
+      "epoch": 0.71,
+      "grad_norm": 0.2818461060523987,
+      "learning_rate": 0.0001856738951844759,
+      "loss": 0.9156,
+      "step": 336
+    },
+    {
+      "epoch": 0.71,
+      "grad_norm": 1.2286068201065063,
+      "learning_rate": 0.00018558804790685588,
+      "loss": 2.6577,
+      "step": 337
+    },
+    {
+      "epoch": 0.71,
+      "grad_norm": 0.7086435556411743,
+      "learning_rate": 0.00018550196415209914,
+      "loss": 0.8172,
+      "step": 338
+    },
+    {
+      "epoch": 0.71,
+      "grad_norm": 1.0317937135696411,
+      "learning_rate": 0.00018541564415805258,
+      "loss": 1.3381,
+      "step": 339
+    },
+    {
+      "epoch": 0.72,
+      "grad_norm": 0.693418562412262,
+      "learning_rate": 0.00018532908816321558,
+      "loss": 1.1259,
+      "step": 340
+    },
+    {
+      "epoch": 0.72,
+      "grad_norm": 1.25714910030365,
+      "learning_rate": 0.00018524229640673974,
+      "loss": 0.7892,
+      "step": 341
+    },
+    {
+      "epoch": 0.72,
+      "grad_norm": 0.6042699813842773,
+      "learning_rate": 0.00018515526912842796,
+      "loss": 0.8982,
+      "step": 342
+    },
+    {
+      "epoch": 0.72,
+      "grad_norm": 0.3453720211982727,
+      "learning_rate": 0.00018506800656873398,
+      "loss": 0.9424,
+      "step": 343
+    },
+    {
+      "epoch": 0.72,
+      "grad_norm": 0.7436335682868958,
+      "learning_rate": 0.0001849805089687615,
+      "loss": 0.7121,
+      "step": 344
+    },
+    {
+      "epoch": 0.73,
+      "grad_norm": 0.8308970928192139,
+      "learning_rate": 0.00018489277657026375,
+      "loss": 1.1099,
+      "step": 345
+    },
+    {
+      "epoch": 0.73,
+      "grad_norm": 0.6892271637916565,
+      "learning_rate": 0.0001848048096156426,
+      "loss": 0.6814,
+      "step": 346
+    },
+    {
+      "epoch": 0.73,
+      "grad_norm": 0.30851200222969055,
+      "learning_rate": 0.00018471660834794805,
+      "loss": 0.283,
+      "step": 347
+    },
+    {
+      "epoch": 0.73,
+      "grad_norm": 0.2706887722015381,
+      "learning_rate": 0.00018462817301087748,
+      "loss": 0.6258,
+      "step": 348
+    },
+    {
+      "epoch": 0.73,
+      "grad_norm": 0.9876924157142639,
+      "learning_rate": 0.00018453950384877504,
+      "loss": 0.6983,
+      "step": 349
+    },
+    {
+      "epoch": 0.74,
+      "grad_norm": 0.3037252128124237,
+      "learning_rate": 0.0001844506011066308,
+      "loss": 0.9854,
+      "step": 350
+    },
+    {
+      "epoch": 0.74,
+      "grad_norm": 1.0091379880905151,
+      "learning_rate": 0.00018436146503008035,
+      "loss": 0.9871,
+      "step": 351
+    },
+    {
+      "epoch": 0.74,
+      "grad_norm": 0.5219744443893433,
+      "learning_rate": 0.0001842720958654039,
+      "loss": 0.3771,
+      "step": 352
+    },
+    {
+      "epoch": 0.74,
+      "grad_norm": 0.49409008026123047,
+      "learning_rate": 0.00018418249385952575,
+      "loss": 1.0838,
+      "step": 353
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 0.29014095664024353,
+      "learning_rate": 0.00018409265926001343,
+      "loss": 0.9922,
+      "step": 354
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 0.3307441771030426,
+      "learning_rate": 0.00018400259231507717,
+      "loss": 1.0458,
+      "step": 355
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 0.3356322646141052,
+      "learning_rate": 0.00018391229327356916,
+      "loss": 0.9891,
+      "step": 356
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 0.3707556426525116,
+      "learning_rate": 0.00018382176238498286,
+      "loss": 0.9578,
+      "step": 357
+    },
+    {
+      "epoch": 0.75,
+      "eval_loss": 0.8826732635498047,
+      "eval_runtime": 61.8974,
+      "eval_samples_per_second": 1.616,
+      "eval_steps_per_second": 1.616,
+      "step": 357
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 0.7507327198982239,
+      "learning_rate": 0.00018373099989945236,
+      "loss": 0.8916,
+      "step": 358
+    },
+    {
+      "epoch": 0.76,
+      "grad_norm": 0.3686985373497009,
+      "learning_rate": 0.00018364000606775155,
+      "loss": 0.9855,
+      "step": 359
+    },
+    {
+      "epoch": 0.76,
+      "grad_norm": 0.34240958094596863,
+      "learning_rate": 0.00018354878114129367,
+      "loss": 1.0874,
+      "step": 360
+    },
+    {
+      "epoch": 0.76,
+      "grad_norm": 0.2911188304424286,
+      "learning_rate": 0.00018345732537213027,
+      "loss": 1.2217,
+      "step": 361
+    },
+    {
+      "epoch": 0.76,
+      "grad_norm": 0.5415646433830261,
+      "learning_rate": 0.0001833656390129509,
+      "loss": 0.6675,
+      "step": 362
+    },
+    {
+      "epoch": 0.76,
+      "grad_norm": 0.36682239174842834,
+      "learning_rate": 0.00018327372231708212,
+      "loss": 0.8702,
+      "step": 363
+    },
+    {
+      "epoch": 0.77,
+      "grad_norm": 0.5462591648101807,
+      "learning_rate": 0.0001831815755384869,
+      "loss": 0.9005,
+      "step": 364
+    },
+    {
+      "epoch": 0.77,
+      "grad_norm": 0.5059930682182312,
+      "learning_rate": 0.00018308919893176396,
+      "loss": 0.8994,
+      "step": 365
+    },
+    {
+      "epoch": 0.77,
+      "grad_norm": 0.6344266533851624,
+      "learning_rate": 0.00018299659275214706,
+      "loss": 1.1571,
+      "step": 366
+    },
+    {
+      "epoch": 0.77,
+      "grad_norm": 0.2552272081375122,
+      "learning_rate": 0.00018290375725550417,
+      "loss": 1.2492,
+      "step": 367
+    },
+    {
+      "epoch": 0.77,
+      "grad_norm": 0.5543289184570312,
+      "learning_rate": 0.00018281069269833692,
+      "loss": 1.0141,
+      "step": 368
+    },
+    {
+      "epoch": 0.78,
+      "grad_norm": 1.3686586618423462,
+      "learning_rate": 0.0001827173993377798,
+      "loss": 0.8264,
+      "step": 369
+    },
+    {
+      "epoch": 0.78,
+      "grad_norm": 0.5549390912055969,
+      "learning_rate": 0.0001826238774315995,
+      "loss": 1.0753,
+      "step": 370
+    },
+    {
+      "epoch": 0.78,
+      "grad_norm": 0.8563418388366699,
+      "learning_rate": 0.00018253012723819416,
+      "loss": 0.4458,
+      "step": 371
+    },
+    {
+      "epoch": 0.78,
+      "grad_norm": 0.4292491376399994,
+      "learning_rate": 0.00018243614901659264,
+      "loss": 1.1994,
+      "step": 372
+    },
+    {
+      "epoch": 0.79,
+      "grad_norm": 0.37186571955680847,
+      "learning_rate": 0.00018234194302645394,
+      "loss": 0.9811,
+      "step": 373
+    },
+    {
+      "epoch": 0.79,
+      "grad_norm": 0.6655788421630859,
+      "learning_rate": 0.00018224750952806624,
+      "loss": 0.5048,
+      "step": 374
+    },
+    {
+      "epoch": 0.79,
+      "grad_norm": 0.7731723785400391,
+      "learning_rate": 0.00018215284878234642,
+      "loss": 0.9481,
+      "step": 375
+    },
+    {
+      "epoch": 0.79,
+      "grad_norm": 0.36243554949760437,
+      "learning_rate": 0.00018205796105083915,
+      "loss": 1.0048,
+      "step": 376
+    },
+    {
+      "epoch": 0.79,
+      "grad_norm": 1.08484947681427,
+      "learning_rate": 0.00018196284659571639,
+      "loss": 1.0245,
+      "step": 377
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 1.128653883934021,
+      "learning_rate": 0.00018186750567977637,
+      "loss": 0.9997,
+      "step": 378
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.6685619950294495,
+      "learning_rate": 0.00018177193856644316,
+      "loss": 1.3555,
+      "step": 379
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.30426543951034546,
+      "learning_rate": 0.00018167614551976567,
+      "loss": 1.1209,
+      "step": 380
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.6189528107643127,
+      "learning_rate": 0.00018158012680441723,
+      "loss": 1.0321,
+      "step": 381
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.6775807738304138,
+      "learning_rate": 0.00018148388268569453,
+      "loss": 0.7826,
+      "step": 382
+    },
+    {
+      "epoch": 0.81,
+      "grad_norm": 0.4594517946243286,
+      "learning_rate": 0.00018138741342951705,
+      "loss": 0.6422,
+      "step": 383
+    },
+    {
+      "epoch": 0.81,
+      "grad_norm": 0.537011444568634,
+      "learning_rate": 0.00018129071930242648,
+      "loss": 0.9219,
+      "step": 384
+    },
+    {
+      "epoch": 0.81,
+      "grad_norm": 0.43772855401039124,
+      "learning_rate": 0.00018119380057158568,
+      "loss": 1.1737,
+      "step": 385
+    },
+    {
+      "epoch": 0.81,
+      "grad_norm": 0.7221130132675171,
+      "learning_rate": 0.00018109665750477806,
+      "loss": 0.8636,
+      "step": 386
+    },
+    {
+      "epoch": 0.81,
+      "grad_norm": 0.3437989354133606,
+      "learning_rate": 0.00018099929037040694,
+      "loss": 0.9238,
+      "step": 387
+    },
+    {
+      "epoch": 0.82,
+      "grad_norm": 0.47244492173194885,
+      "learning_rate": 0.00018090169943749476,
+      "loss": 0.7715,
+      "step": 388
+    },
+    {
+      "epoch": 0.82,
+      "grad_norm": 0.7109631299972534,
+      "learning_rate": 0.0001808038849756822,
+      "loss": 0.4109,
+      "step": 389
+    },
+    {
+      "epoch": 0.82,
+      "grad_norm": 0.27005669474601746,
+      "learning_rate": 0.00018070584725522762,
+      "loss": 0.7158,
+      "step": 390
+    },
+    {
+      "epoch": 0.82,
+      "grad_norm": 0.4006590247154236,
+      "learning_rate": 0.00018060758654700622,
+      "loss": 1.0167,
+      "step": 391
+    },
+    {
+      "epoch": 0.83,
+      "grad_norm": 0.5627204179763794,
+      "learning_rate": 0.00018050910312250931,
+      "loss": 0.8679,
+      "step": 392
+    },
+    {
+      "epoch": 0.83,
+      "grad_norm": 0.5019241571426392,
+      "learning_rate": 0.00018041039725384352,
+      "loss": 0.9163,
+      "step": 393
+    },
+    {
+      "epoch": 0.83,
+      "grad_norm": 1.00431227684021,
+      "learning_rate": 0.00018031146921373018,
+      "loss": 0.676,
+      "step": 394
+    },
+    {
+      "epoch": 0.83,
+      "grad_norm": 0.7062071561813354,
+      "learning_rate": 0.0001802123192755044,
+      "loss": 1.2407,
+      "step": 395
+    },
+    {
+      "epoch": 0.83,
+      "grad_norm": 1.6554285287857056,
+      "learning_rate": 0.00018011294771311435,
+      "loss": 1.1187,
+      "step": 396
+    },
+    {
+      "epoch": 0.84,
+      "grad_norm": 1.08072829246521,
+      "learning_rate": 0.00018001335480112064,
+      "loss": 0.4878,
+      "step": 397
+    },
+    {
+      "epoch": 0.84,
+      "grad_norm": 0.3923906981945038,
+      "learning_rate": 0.00017991354081469538,
+      "loss": 0.7836,
+      "step": 398
+    },
+    {
+      "epoch": 0.84,
+      "grad_norm": 0.20446747541427612,
+      "learning_rate": 0.0001798135060296216,
+      "loss": 0.4597,
+      "step": 399
+    },
+    {
+      "epoch": 0.84,
+      "grad_norm": 0.5178759098052979,
+      "learning_rate": 0.00017971325072229226,
+      "loss": 1.7021,
+      "step": 400
+    },
+    {
+      "epoch": 0.84,
+      "grad_norm": 0.5159180164337158,
+      "learning_rate": 0.0001796127751697097,
+      "loss": 0.6037,
+      "step": 401
+    },
+    {
+      "epoch": 0.85,
+      "grad_norm": 0.9448319673538208,
+      "learning_rate": 0.0001795120796494848,
+      "loss": 0.8965,
+      "step": 402
+    },
+    {
+      "epoch": 0.85,
+      "grad_norm": 1.0035223960876465,
+      "learning_rate": 0.00017941116443983613,
+      "loss": 0.9786,
+      "step": 403
+    },
+    {
+      "epoch": 0.85,
+      "grad_norm": 0.26040011644363403,
+      "learning_rate": 0.00017931002981958933,
+      "loss": 0.8624,
+      "step": 404
+    },
+    {
+      "epoch": 0.85,
+      "grad_norm": 0.518144965171814,
+      "learning_rate": 0.00017920867606817625,
+      "loss": 1.0095,
+      "step": 405
+    },
+    {
+      "epoch": 0.85,
+      "grad_norm": 0.5256940722465515,
+      "learning_rate": 0.00017910710346563416,
+      "loss": 0.7392,
+      "step": 406
+    },
+    {
+      "epoch": 0.86,
+      "grad_norm": 0.8347258567810059,
+      "learning_rate": 0.000179005312292605,
+      "loss": 0.8081,
+      "step": 407
+    },
+    {
+      "epoch": 0.86,
+      "grad_norm": 0.8221095204353333,
+      "learning_rate": 0.00017890330283033468,
+      "loss": 1.1406,
+      "step": 408
+    },
+    {
+      "epoch": 0.86,
+      "grad_norm": 0.8048923015594482,
+      "learning_rate": 0.00017880107536067218,
+      "loss": 1.4362,
+      "step": 409
+    },
+    {
+      "epoch": 0.86,
+      "grad_norm": 1.9037342071533203,
+      "learning_rate": 0.0001786986301660689,
+      "loss": 1.2935,
+      "step": 410
+    },
+    {
+      "epoch": 0.87,
+      "grad_norm": 0.5521582961082458,
+      "learning_rate": 0.00017859596752957768,
+      "loss": 1.0742,
+      "step": 411
+    },
+    {
+      "epoch": 0.87,
+      "grad_norm": 1.052284598350525,
+      "learning_rate": 0.00017849308773485226,
+      "loss": 0.7661,
+      "step": 412
+    },
+    {
+      "epoch": 0.87,
+      "grad_norm": 0.43000859022140503,
+      "learning_rate": 0.00017838999106614632,
+      "loss": 0.812,
+      "step": 413
+    },
+    {
+      "epoch": 0.87,
+      "grad_norm": 0.7804751396179199,
+      "learning_rate": 0.00017828667780831278,
+      "loss": 0.7995,
+      "step": 414
+    },
+    {
+      "epoch": 0.87,
+      "grad_norm": 0.5827552080154419,
+      "learning_rate": 0.000178183148246803,
+      "loss": 0.6489,
+      "step": 415
+    },
+    {
+      "epoch": 0.88,
+      "grad_norm": 1.3453142642974854,
+      "learning_rate": 0.00017807940266766593,
+      "loss": 0.7154,
+      "step": 416
+    },
+    {
+      "epoch": 0.88,
+      "grad_norm": 0.24924832582473755,
+      "learning_rate": 0.00017797544135754744,
+      "loss": 0.8061,
+      "step": 417
+    },
+    {
+      "epoch": 0.88,
+      "grad_norm": 0.4459979236125946,
+      "learning_rate": 0.0001778712646036894,
+      "loss": 1.1167,
+      "step": 418
+    },
+    {
+      "epoch": 0.88,
+      "grad_norm": 0.6095878481864929,
+      "learning_rate": 0.000177766872693929,
+      "loss": 0.8344,
+      "step": 419
+    },
+    {
+      "epoch": 0.88,
+      "grad_norm": 0.43662723898887634,
+      "learning_rate": 0.00017766226591669785,
+      "loss": 1.0373,
+      "step": 420
+    },
+    {
+      "epoch": 0.89,
+      "grad_norm": 0.8759774565696716,
+      "learning_rate": 0.00017755744456102122,
+      "loss": 1.0988,
+      "step": 421
+    },
+    {
+      "epoch": 0.89,
+      "grad_norm": 1.1800742149353027,
+      "learning_rate": 0.00017745240891651735,
+      "loss": 0.7385,
+      "step": 422
+    },
+    {
+      "epoch": 0.89,
+      "grad_norm": 0.5820197463035583,
+      "learning_rate": 0.0001773471592733964,
+      "loss": 0.9363,
+      "step": 423
+    },
+    {
+      "epoch": 0.89,
+      "grad_norm": 0.6128491759300232,
+      "learning_rate": 0.00017724169592245995,
+      "loss": 0.7762,
+      "step": 424
+    },
+    {
+      "epoch": 0.89,
+      "grad_norm": 0.5693449378013611,
+      "learning_rate": 0.0001771360191551,
+      "loss": 0.7526,
+      "step": 425
+    },
+    {
+      "epoch": 0.9,
+      "grad_norm": 0.7725418210029602,
+      "learning_rate": 0.00017703012926329815,
+      "loss": 0.7019,
+      "step": 426
+    },
+    {
+      "epoch": 0.9,
+      "grad_norm": 0.5068923234939575,
+      "learning_rate": 0.0001769240265396249,
+      "loss": 0.8308,
+      "step": 427
+    },
+    {
+      "epoch": 0.9,
+      "grad_norm": 0.34859699010849,
+      "learning_rate": 0.0001768177112772388,
+      "loss": 0.9593,
+      "step": 428
+    },
+    {
+      "epoch": 0.9,
+      "grad_norm": 0.34673023223876953,
+      "learning_rate": 0.00017671118376988573,
+      "loss": 1.0334,
+      "step": 429
+    },
+    {
+      "epoch": 0.91,
+      "grad_norm": 0.5354735851287842,
+      "learning_rate": 0.0001766044443118978,
+      "loss": 1.2355,
+      "step": 430
+    },
+    {
+      "epoch": 0.91,
+      "grad_norm": 1.2567592859268188,
+      "learning_rate": 0.0001764974931981929,
+      "loss": 0.8935,
+      "step": 431
+    },
+    {
+      "epoch": 0.91,
+      "grad_norm": 0.4151657521724701,
+      "learning_rate": 0.00017639033072427366,
+      "loss": 1.1042,
+      "step": 432
+    },
+    {
+      "epoch": 0.91,
+      "grad_norm": 0.4307219386100769,
+      "learning_rate": 0.00017628295718622665,
+      "loss": 1.2273,
+      "step": 433
+    },
+    {
+      "epoch": 0.91,
+      "grad_norm": 0.6330164074897766,
+      "learning_rate": 0.0001761753728807217,
+      "loss": 1.2027,
+      "step": 434
+    },
+    {
+      "epoch": 0.92,
+      "grad_norm": 0.47434625029563904,
+      "learning_rate": 0.00017606757810501088,
+      "loss": 0.9242,
+      "step": 435
+    },
+    {
+      "epoch": 0.92,
+      "grad_norm": 1.0463887453079224,
+      "learning_rate": 0.00017595957315692782,
+      "loss": 1.151,
+      "step": 436
+    },
+    {
+      "epoch": 0.92,
+      "grad_norm": 0.7210713028907776,
+      "learning_rate": 0.00017585135833488692,
+      "loss": 0.8223,
+      "step": 437
+    },
+    {
+      "epoch": 0.92,
+      "grad_norm": 0.5121049284934998,
+      "learning_rate": 0.00017574293393788235,
+      "loss": 0.6994,
+      "step": 438
+    },
+    {
+      "epoch": 0.92,
+      "grad_norm": 0.8933761119842529,
+      "learning_rate": 0.00017563430026548734,
+      "loss": 0.846,
+      "step": 439
+    },
+    {
+      "epoch": 0.93,
+      "grad_norm": 0.5270050764083862,
+      "learning_rate": 0.0001755254576178535,
+      "loss": 0.7119,
+      "step": 440
+    },
+    {
+      "epoch": 0.93,
+      "grad_norm": 0.37028369307518005,
+      "learning_rate": 0.0001754164062957096,
+      "loss": 0.8623,
+      "step": 441
+    },
+    {
+      "epoch": 0.93,
+      "grad_norm": 0.6245588660240173,
+      "learning_rate": 0.00017530714660036112,
+      "loss": 1.0409,
+      "step": 442
+    },
+    {
+      "epoch": 0.93,
+      "grad_norm": 0.878105640411377,
+      "learning_rate": 0.0001751976788336892,
+      "loss": 0.6867,
+      "step": 443
+    },
+    {
+      "epoch": 0.93,
+      "grad_norm": 0.3765283226966858,
+      "learning_rate": 0.00017508800329814995,
+      "loss": 1.2251,
+      "step": 444
+    },
+    {
+      "epoch": 0.94,
+      "grad_norm": 0.4110933840274811,
+      "learning_rate": 0.00017497812029677344,
+      "loss": 0.8676,
+      "step": 445
+    },
+    {
+      "epoch": 0.94,
+      "grad_norm": 1.3986817598342896,
+      "learning_rate": 0.000174868030133163,
+      "loss": 1.6298,
+      "step": 446
+    },
+    {
+      "epoch": 0.94,
+      "grad_norm": 0.4310443103313446,
+      "learning_rate": 0.0001747577331114945,
+      "loss": 0.7328,
+      "step": 447
+    },
+    {
+      "epoch": 0.94,
+      "grad_norm": 1.5922423601150513,
+      "learning_rate": 0.00017464722953651504,
+      "loss": 0.6629,
+      "step": 448
+    },
+    {
+      "epoch": 0.95,
+      "grad_norm": 2.27004075050354,
+      "learning_rate": 0.00017453651971354264,
+      "loss": 1.4748,
+      "step": 449
+    },
+    {
+      "epoch": 0.95,
+      "grad_norm": 0.31181880831718445,
+      "learning_rate": 0.00017442560394846516,
+      "loss": 1.0477,
+      "step": 450
+    },
+    {
+      "epoch": 0.95,
+      "grad_norm": 1.1180263757705688,
+      "learning_rate": 0.00017431448254773944,
+      "loss": 0.9225,
+      "step": 451
+    },
+    {
+      "epoch": 0.95,
+      "grad_norm": 0.7490403056144714,
+      "learning_rate": 0.00017420315581839044,
+      "loss": 0.7847,
+      "step": 452
+    },
+    {
+      "epoch": 0.95,
+      "grad_norm": 1.138551115989685,
+      "learning_rate": 0.0001740916240680105,
+      "loss": 1.2782,
+      "step": 453
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.9375423789024353,
+      "learning_rate": 0.0001739798876047584,
+      "loss": 0.8316,
+      "step": 454
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 1.0941681861877441,
+      "learning_rate": 0.0001738679467373586,
+      "loss": 0.9702,
+      "step": 455
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.2845444083213806,
+      "learning_rate": 0.00017375580177510016,
+      "loss": 0.8563,
+      "step": 456
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.7341310381889343,
+      "learning_rate": 0.0001736434530278362,
+      "loss": 0.7102,
+      "step": 457
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.886854350566864,
+      "learning_rate": 0.0001735309008059829,
+      "loss": 1.4872,
+      "step": 458
+    },
+    {
+      "epoch": 0.97,
+      "grad_norm": 0.44623202085494995,
+      "learning_rate": 0.00017341814542051845,
+      "loss": 0.8142,
+      "step": 459
+    },
+    {
+      "epoch": 0.97,
+      "grad_norm": 0.8813387155532837,
+      "learning_rate": 0.00017330518718298264,
+      "loss": 1.0869,
+      "step": 460
+    },
+    {
+      "epoch": 0.97,
+      "grad_norm": 0.8468006253242493,
+      "learning_rate": 0.0001731920264054755,
+      "loss": 1.2859,
+      "step": 461
+    },
+    {
+      "epoch": 0.97,
+      "grad_norm": 0.6797575354576111,
+      "learning_rate": 0.00017307866340065685,
+      "loss": 1.0288,
+      "step": 462
+    },
+    {
+      "epoch": 0.97,
+      "grad_norm": 0.25991517305374146,
+      "learning_rate": 0.00017296509848174508,
+      "loss": 0.7996,
+      "step": 463
+    },
+    {
+      "epoch": 0.98,
+      "grad_norm": 0.557244598865509,
+      "learning_rate": 0.00017285133196251663,
+      "loss": 0.6877,
+      "step": 464
+    },
+    {
+      "epoch": 0.98,
+      "grad_norm": 0.49986258149147034,
+      "learning_rate": 0.00017273736415730488,
+      "loss": 0.8285,
+      "step": 465
+    },
+    {
+      "epoch": 0.98,
+      "grad_norm": 0.5839115977287292,
+      "learning_rate": 0.0001726231953809993,
+      "loss": 0.8617,
+      "step": 466
+    },
+    {
+      "epoch": 0.98,
+      "grad_norm": 1.1967806816101074,
+      "learning_rate": 0.0001725088259490448,
+      "loss": 0.719,
+      "step": 467
+    },
+    {
+      "epoch": 0.99,
+      "grad_norm": 0.7830839157104492,
+      "learning_rate": 0.00017239425617744048,
+      "loss": 0.623,
+      "step": 468
+    },
+    {
+      "epoch": 0.99,
+      "grad_norm": 2.0930089950561523,
+      "learning_rate": 0.00017227948638273916,
+      "loss": 1.1768,
+      "step": 469
+    },
+    {
+      "epoch": 0.99,
+      "grad_norm": 0.562641441822052,
+      "learning_rate": 0.0001721645168820462,
+      "loss": 0.5235,
+      "step": 470
+    },
+    {
+      "epoch": 0.99,
+      "grad_norm": 0.5656068325042725,
+      "learning_rate": 0.00017204934799301883,
+      "loss": 0.9211,
+      "step": 471
+    },
+    {
+      "epoch": 0.99,
+      "grad_norm": 0.934866726398468,
+      "learning_rate": 0.0001719339800338651,
+      "loss": 0.7897,
+      "step": 472
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.30100950598716736,
+      "learning_rate": 0.00017181841332334318,
+      "loss": 1.0436,
+      "step": 473
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.5494408011436462,
+      "learning_rate": 0.00017170264818076026,
+      "loss": 0.6412,
+      "step": 474
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.6607212424278259,
+      "learning_rate": 0.00017158668492597186,
+      "loss": 1.3501,
+      "step": 475
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 1900,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 475,
+  "total_flos": 3.489377344094208e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-475/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a86f7a599d796e7b48de991f534e6b4a66f2f12ac3aac1d99bcffb3fefbdb76d
+size 5816

checkpoint-950/README.md

@@ -0,0 +1,202 @@
+---
+library_name: peft
+base_model: Qwen/Qwen1.5-MoE-A2.7B
+---
+
+# 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 0.10.0

checkpoint-950/adapter_config.json

@@ -0,0 +1,36 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen1.5-MoE-A2.7B",
+  "bias": "none",
+  "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_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "gate",
+    "k_proj",
+    "up_proj",
+    "v_proj",
+    "o_proj",
+    "gate_proj",
+    "shared_expert_gate",
+    "q_proj",
+    "down_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-950/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f6a3d8e63ba16b25387f5e6697a0b2be8dd8c783f48fa02f444b3b2e75c8066f
+size 2046221176

checkpoint-950/added_tokens.json

@@ -0,0 +1,5 @@
+{
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644
+}