diff --git a/.gitattributes b/.gitattributes
index b2bff60e0d2c0da2f619d183798c831b63c58b1c..75a226c5971c42e3dedcb951b36b7ba3f8bc30a4 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -54,3 +54,4 @@ wandb/run-20251226_155650-wbzoafvt/run-wbzoafvt.wandb filter=lfs diff=lfs merge=
 sft_devstral_24B_v2/wandb/run-20251226_180613-i1cmzyri/run-i1cmzyri.wandb filter=lfs diff=lfs merge=lfs -text
 sft_devstral_24B_v2/wandb/run-20251226_180702-oordmylf/run-oordmylf.wandb filter=lfs diff=lfs merge=lfs -text
 sft_devstral_24B_v2/wandb/run-20251226_180808-ny9q48hd/run-ny9q48hd.wandb filter=lfs diff=lfs merge=lfs -text
+sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/run-upub1jan.wandb filter=lfs diff=lfs merge=lfs -text
diff --git a/sft_qwen_14B_v2/best_adapter/README.md b/sft_qwen_14B_v2/best_adapter/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..134a3a8531b68adda3fed2a98db824f7c1b7dcfd
--- /dev/null
+++ b/sft_qwen_14B_v2/best_adapter/README.md
@@ -0,0 +1,207 @@
+---
+base_model: Models/Qwen2.5-Coder-14B-CPT
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Models/Qwen2.5-Coder-14B-CPT
+- lora
+- transformers
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.0
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/best_adapter/adapter_config.json b/sft_qwen_14B_v2/best_adapter/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..d6316742bc71d30704e2ccdb9080ed0b8062e454
--- /dev/null
+++ b/sft_qwen_14B_v2/best_adapter/adapter_config.json
@@ -0,0 +1,43 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Models/Qwen2.5-Coder-14B-CPT",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.0",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "v_proj",
+    "k_proj",
+    "q_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/best_adapter/adapter_model.safetensors b/sft_qwen_14B_v2/best_adapter/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..fd323e50fdfb9ae59802526dced950bad901fdf6
--- /dev/null
+++ b/sft_qwen_14B_v2/best_adapter/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c19a29b40a83cc48759a1d5af0d41a375dcc4f15306fcdc65ea10ce6044a6f47
+size 201378736
diff --git a/sft_qwen_14B_v2/best_adapter/training_args.bin b/sft_qwen_14B_v2/best_adapter/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..0bd5f45757bdc6c7fc2d67718b18675b51983e0e
--- /dev/null
+++ b/sft_qwen_14B_v2/best_adapter/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:afae4a8ce1391f149eb00b14eed8f891c715e892ea546bf754d22db2c2bc5969
+size 4792
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-1000/README.md b/sft_qwen_14B_v2/checkpoints/checkpoint-1000/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..134a3a8531b68adda3fed2a98db824f7c1b7dcfd
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-1000/README.md
@@ -0,0 +1,207 @@
+---
+base_model: Models/Qwen2.5-Coder-14B-CPT
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Models/Qwen2.5-Coder-14B-CPT
+- lora
+- transformers
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.0
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-1000/adapter_config.json b/sft_qwen_14B_v2/checkpoints/checkpoint-1000/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..d6316742bc71d30704e2ccdb9080ed0b8062e454
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-1000/adapter_config.json
@@ -0,0 +1,43 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Models/Qwen2.5-Coder-14B-CPT",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.0",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "v_proj",
+    "k_proj",
+    "q_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-1000/adapter_model.safetensors b/sft_qwen_14B_v2/checkpoints/checkpoint-1000/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..006cc4cfb4bb2b9161a7783dced86040fc6dff06
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-1000/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38ac379064d40f8514d7d5e379578e8d466f2ce3582db82555b1f1c5e8c2db6d
+size 201378736
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-1000/optimizer.pt b/sft_qwen_14B_v2/checkpoints/checkpoint-1000/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..8936e79827984e7d18ef50a241f3be27a1878a18
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-1000/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7467ef9a63aa9e5adc3a95a80cee071ebf3381f166d2e2b850ba12d849fac9c0
+size 402976786
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-1000/rng_state.pth b/sft_qwen_14B_v2/checkpoints/checkpoint-1000/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..f741a6d943b58bfb1ec930b7ca27dc857b8643f0
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-1000/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f37d067203ec01fd39a90eb184de470feca9355b97ee387b61a8f71801819194
+size 14244
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-1000/scheduler.pt b/sft_qwen_14B_v2/checkpoints/checkpoint-1000/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..951769923613ff2839621b00eec5987625a1a7b0
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-1000/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0acd749c3b8c2670557a4f5e96ac98dd3da877521e179bc1412a007138308dc8
+size 1064
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-1000/trainer_state.json b/sft_qwen_14B_v2/checkpoints/checkpoint-1000/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..a4bce74332f43b1a22f50c9927b49db0196cd4ab
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-1000/trainer_state.json
@@ -0,0 +1,3623 @@
+{
+  "best_global_step": 1000,
+  "best_metric": 0.9487298727035522,
+  "best_model_checkpoint": "task2file/sft_qwen_14B_v2/checkpoints/checkpoint-1000",
+  "epoch": 0.4219409282700422,
+  "eval_steps": 100,
+  "global_step": 1000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0008438818565400844,
+      "grad_norm": 0.5386583805084229,
+      "learning_rate": 1.7574692442882248e-07,
+      "loss": 1.6941628456115723,
+      "step": 2
+    },
+    {
+      "epoch": 0.0016877637130801688,
+      "grad_norm": 0.5477277636528015,
+      "learning_rate": 5.272407732864675e-07,
+      "loss": 1.7132279872894287,
+      "step": 4
+    },
+    {
+      "epoch": 0.002531645569620253,
+      "grad_norm": 0.5390765070915222,
+      "learning_rate": 8.787346221441126e-07,
+      "loss": 1.641180396080017,
+      "step": 6
+    },
+    {
+      "epoch": 0.0033755274261603376,
+      "grad_norm": 0.5023683905601501,
+      "learning_rate": 1.2302284710017575e-06,
+      "loss": 1.5616240501403809,
+      "step": 8
+    },
+    {
+      "epoch": 0.004219409282700422,
+      "grad_norm": 0.4899154603481293,
+      "learning_rate": 1.5817223198594026e-06,
+      "loss": 1.572033405303955,
+      "step": 10
+    },
+    {
+      "epoch": 0.005063291139240506,
+      "grad_norm": 0.5239788293838501,
+      "learning_rate": 1.9332161687170474e-06,
+      "loss": 1.6242921352386475,
+      "step": 12
+    },
+    {
+      "epoch": 0.00590717299578059,
+      "grad_norm": 0.5172926783561707,
+      "learning_rate": 2.2847100175746925e-06,
+      "loss": 1.6800041198730469,
+      "step": 14
+    },
+    {
+      "epoch": 0.006751054852320675,
+      "grad_norm": 0.5539224743843079,
+      "learning_rate": 2.6362038664323376e-06,
+      "loss": 1.6450834274291992,
+      "step": 16
+    },
+    {
+      "epoch": 0.007594936708860759,
+      "grad_norm": 0.5255337953567505,
+      "learning_rate": 2.9876977152899827e-06,
+      "loss": 1.6673263311386108,
+      "step": 18
+    },
+    {
+      "epoch": 0.008438818565400843,
+      "grad_norm": 0.5074548721313477,
+      "learning_rate": 3.3391915641476277e-06,
+      "loss": 1.531802773475647,
+      "step": 20
+    },
+    {
+      "epoch": 0.009282700421940928,
+      "grad_norm": 0.4160279333591461,
+      "learning_rate": 3.6906854130052724e-06,
+      "loss": 1.599354863166809,
+      "step": 22
+    },
+    {
+      "epoch": 0.010126582278481013,
+      "grad_norm": 0.5716474652290344,
+      "learning_rate": 4.0421792618629174e-06,
+      "loss": 1.6700962781906128,
+      "step": 24
+    },
+    {
+      "epoch": 0.010970464135021098,
+      "grad_norm": 0.5148899555206299,
+      "learning_rate": 4.3936731107205625e-06,
+      "loss": 1.66217839717865,
+      "step": 26
+    },
+    {
+      "epoch": 0.01181434599156118,
+      "grad_norm": 0.575722336769104,
+      "learning_rate": 4.7451669595782076e-06,
+      "loss": 1.6692266464233398,
+      "step": 28
+    },
+    {
+      "epoch": 0.012658227848101266,
+      "grad_norm": 0.5345953106880188,
+      "learning_rate": 5.096660808435853e-06,
+      "loss": 1.5518689155578613,
+      "step": 30
+    },
+    {
+      "epoch": 0.01350210970464135,
+      "grad_norm": 0.4462043344974518,
+      "learning_rate": 5.448154657293498e-06,
+      "loss": 1.5930007696151733,
+      "step": 32
+    },
+    {
+      "epoch": 0.014345991561181435,
+      "grad_norm": 0.5119605660438538,
+      "learning_rate": 5.799648506151143e-06,
+      "loss": 1.6069684028625488,
+      "step": 34
+    },
+    {
+      "epoch": 0.015189873417721518,
+      "grad_norm": 0.5328608751296997,
+      "learning_rate": 6.151142355008788e-06,
+      "loss": 1.5838109254837036,
+      "step": 36
+    },
+    {
+      "epoch": 0.016033755274261603,
+      "grad_norm": 0.5065920352935791,
+      "learning_rate": 6.502636203866433e-06,
+      "loss": 1.608130931854248,
+      "step": 38
+    },
+    {
+      "epoch": 0.016877637130801686,
+      "grad_norm": 0.4479359984397888,
+      "learning_rate": 6.854130052724078e-06,
+      "loss": 1.5942182540893555,
+      "step": 40
+    },
+    {
+      "epoch": 0.017721518987341773,
+      "grad_norm": 0.42844903469085693,
+      "learning_rate": 7.205623901581722e-06,
+      "loss": 1.6441553831100464,
+      "step": 42
+    },
+    {
+      "epoch": 0.018565400843881856,
+      "grad_norm": 0.476630836725235,
+      "learning_rate": 7.557117750439367e-06,
+      "loss": 1.6068111658096313,
+      "step": 44
+    },
+    {
+      "epoch": 0.019409282700421943,
+      "grad_norm": 0.4532654881477356,
+      "learning_rate": 7.908611599297012e-06,
+      "loss": 1.6618021726608276,
+      "step": 46
+    },
+    {
+      "epoch": 0.020253164556962026,
+      "grad_norm": 0.3701118230819702,
+      "learning_rate": 8.260105448154657e-06,
+      "loss": 1.4730033874511719,
+      "step": 48
+    },
+    {
+      "epoch": 0.02109704641350211,
+      "grad_norm": 0.38471561670303345,
+      "learning_rate": 8.611599297012302e-06,
+      "loss": 1.4828267097473145,
+      "step": 50
+    },
+    {
+      "epoch": 0.021940928270042195,
+      "grad_norm": 0.3602336347103119,
+      "learning_rate": 8.963093145869948e-06,
+      "loss": 1.3877452611923218,
+      "step": 52
+    },
+    {
+      "epoch": 0.02278481012658228,
+      "grad_norm": 0.40318572521209717,
+      "learning_rate": 9.314586994727593e-06,
+      "loss": 1.49052894115448,
+      "step": 54
+    },
+    {
+      "epoch": 0.02362869198312236,
+      "grad_norm": 0.3223826587200165,
+      "learning_rate": 9.666080843585238e-06,
+      "loss": 1.4912524223327637,
+      "step": 56
+    },
+    {
+      "epoch": 0.024472573839662448,
+      "grad_norm": 0.3873065114021301,
+      "learning_rate": 1.0017574692442883e-05,
+      "loss": 1.526674509048462,
+      "step": 58
+    },
+    {
+      "epoch": 0.02531645569620253,
+      "grad_norm": 0.410159707069397,
+      "learning_rate": 1.0369068541300528e-05,
+      "loss": 1.4480271339416504,
+      "step": 60
+    },
+    {
+      "epoch": 0.026160337552742614,
+      "grad_norm": 0.3632003962993622,
+      "learning_rate": 1.0720562390158173e-05,
+      "loss": 1.4222990274429321,
+      "step": 62
+    },
+    {
+      "epoch": 0.0270042194092827,
+      "grad_norm": 0.33118435740470886,
+      "learning_rate": 1.1072056239015818e-05,
+      "loss": 1.387171745300293,
+      "step": 64
+    },
+    {
+      "epoch": 0.027848101265822784,
+      "grad_norm": 0.3301764726638794,
+      "learning_rate": 1.1423550087873463e-05,
+      "loss": 1.3523777723312378,
+      "step": 66
+    },
+    {
+      "epoch": 0.02869198312236287,
+      "grad_norm": 0.34342435002326965,
+      "learning_rate": 1.1775043936731108e-05,
+      "loss": 1.4515162706375122,
+      "step": 68
+    },
+    {
+      "epoch": 0.029535864978902954,
+      "grad_norm": 0.3243122100830078,
+      "learning_rate": 1.2126537785588753e-05,
+      "loss": 1.3509243726730347,
+      "step": 70
+    },
+    {
+      "epoch": 0.030379746835443037,
+      "grad_norm": 0.3450150787830353,
+      "learning_rate": 1.2478031634446398e-05,
+      "loss": 1.4936245679855347,
+      "step": 72
+    },
+    {
+      "epoch": 0.031223628691983123,
+      "grad_norm": 0.38912028074264526,
+      "learning_rate": 1.2829525483304042e-05,
+      "loss": 1.3419109582901,
+      "step": 74
+    },
+    {
+      "epoch": 0.032067510548523206,
+      "grad_norm": 0.3019310235977173,
+      "learning_rate": 1.3181019332161687e-05,
+      "loss": 1.4284154176712036,
+      "step": 76
+    },
+    {
+      "epoch": 0.03291139240506329,
+      "grad_norm": 0.37803682684898376,
+      "learning_rate": 1.3532513181019332e-05,
+      "loss": 1.4256561994552612,
+      "step": 78
+    },
+    {
+      "epoch": 0.03375527426160337,
+      "grad_norm": 0.34191736578941345,
+      "learning_rate": 1.3884007029876977e-05,
+      "loss": 1.3256909847259521,
+      "step": 80
+    },
+    {
+      "epoch": 0.03459915611814346,
+      "grad_norm": 0.35242700576782227,
+      "learning_rate": 1.4235500878734624e-05,
+      "loss": 1.2710685729980469,
+      "step": 82
+    },
+    {
+      "epoch": 0.035443037974683546,
+      "grad_norm": 0.38094228506088257,
+      "learning_rate": 1.4586994727592269e-05,
+      "loss": 1.253411889076233,
+      "step": 84
+    },
+    {
+      "epoch": 0.036286919831223625,
+      "grad_norm": 0.36837366223335266,
+      "learning_rate": 1.4938488576449914e-05,
+      "loss": 1.3064342737197876,
+      "step": 86
+    },
+    {
+      "epoch": 0.03713080168776371,
+      "grad_norm": 0.3443569242954254,
+      "learning_rate": 1.5289982425307557e-05,
+      "loss": 1.293562412261963,
+      "step": 88
+    },
+    {
+      "epoch": 0.0379746835443038,
+      "grad_norm": 0.3799338936805725,
+      "learning_rate": 1.5641476274165202e-05,
+      "loss": 1.3382648229599,
+      "step": 90
+    },
+    {
+      "epoch": 0.038818565400843885,
+      "grad_norm": 0.40501922369003296,
+      "learning_rate": 1.599297012302285e-05,
+      "loss": 1.3925724029541016,
+      "step": 92
+    },
+    {
+      "epoch": 0.039662447257383965,
+      "grad_norm": 0.4419630467891693,
+      "learning_rate": 1.6344463971880492e-05,
+      "loss": 1.357171893119812,
+      "step": 94
+    },
+    {
+      "epoch": 0.04050632911392405,
+      "grad_norm": 0.3619817793369293,
+      "learning_rate": 1.6695957820738137e-05,
+      "loss": 1.3029985427856445,
+      "step": 96
+    },
+    {
+      "epoch": 0.04135021097046414,
+      "grad_norm": 0.4851357340812683,
+      "learning_rate": 1.7047451669595782e-05,
+      "loss": 1.3498191833496094,
+      "step": 98
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "grad_norm": 0.418658584356308,
+      "learning_rate": 1.7398945518453427e-05,
+      "loss": 1.185287356376648,
+      "step": 100
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "eval_loss": 1.2979938983917236,
+      "eval_runtime": 682.1979,
+      "eval_samples_per_second": 3.089,
+      "eval_steps_per_second": 3.089,
+      "step": 100
+    },
+    {
+      "epoch": 0.043037974683544304,
+      "grad_norm": 0.4464418888092041,
+      "learning_rate": 1.7750439367311073e-05,
+      "loss": 1.2217272520065308,
+      "step": 102
+    },
+    {
+      "epoch": 0.04388185654008439,
+      "grad_norm": 0.4706237316131592,
+      "learning_rate": 1.8101933216168718e-05,
+      "loss": 1.2052050828933716,
+      "step": 104
+    },
+    {
+      "epoch": 0.04472573839662447,
+      "grad_norm": 0.46394404768943787,
+      "learning_rate": 1.8453427065026363e-05,
+      "loss": 1.221343994140625,
+      "step": 106
+    },
+    {
+      "epoch": 0.04556962025316456,
+      "grad_norm": 0.4726889431476593,
+      "learning_rate": 1.8804920913884008e-05,
+      "loss": 1.2387475967407227,
+      "step": 108
+    },
+    {
+      "epoch": 0.046413502109704644,
+      "grad_norm": 0.42130985856056213,
+      "learning_rate": 1.9156414762741653e-05,
+      "loss": 1.2851309776306152,
+      "step": 110
+    },
+    {
+      "epoch": 0.04725738396624472,
+      "grad_norm": 0.4504576623439789,
+      "learning_rate": 1.9507908611599298e-05,
+      "loss": 1.2753145694732666,
+      "step": 112
+    },
+    {
+      "epoch": 0.04810126582278481,
+      "grad_norm": 0.396085262298584,
+      "learning_rate": 1.9859402460456943e-05,
+      "loss": 1.2427717447280884,
+      "step": 114
+    },
+    {
+      "epoch": 0.048945147679324896,
+      "grad_norm": 0.5106491446495056,
+      "learning_rate": 2.0210896309314588e-05,
+      "loss": 1.2943825721740723,
+      "step": 116
+    },
+    {
+      "epoch": 0.049789029535864976,
+      "grad_norm": 0.42351317405700684,
+      "learning_rate": 2.0562390158172233e-05,
+      "loss": 1.263301134109497,
+      "step": 118
+    },
+    {
+      "epoch": 0.05063291139240506,
+      "grad_norm": 0.4403539299964905,
+      "learning_rate": 2.0913884007029878e-05,
+      "loss": 1.2647849321365356,
+      "step": 120
+    },
+    {
+      "epoch": 0.05147679324894515,
+      "grad_norm": 0.5260752439498901,
+      "learning_rate": 2.1265377855887523e-05,
+      "loss": 1.2351393699645996,
+      "step": 122
+    },
+    {
+      "epoch": 0.05232067510548523,
+      "grad_norm": 0.44978851079940796,
+      "learning_rate": 2.1616871704745168e-05,
+      "loss": 1.0384471416473389,
+      "step": 124
+    },
+    {
+      "epoch": 0.053164556962025315,
+      "grad_norm": 0.47732362151145935,
+      "learning_rate": 2.1968365553602813e-05,
+      "loss": 1.1518068313598633,
+      "step": 126
+    },
+    {
+      "epoch": 0.0540084388185654,
+      "grad_norm": 0.5473551750183105,
+      "learning_rate": 2.231985940246046e-05,
+      "loss": 1.2264912128448486,
+      "step": 128
+    },
+    {
+      "epoch": 0.05485232067510549,
+      "grad_norm": 0.4473855197429657,
+      "learning_rate": 2.2671353251318103e-05,
+      "loss": 1.1615246534347534,
+      "step": 130
+    },
+    {
+      "epoch": 0.05569620253164557,
+      "grad_norm": 0.5980377197265625,
+      "learning_rate": 2.302284710017575e-05,
+      "loss": 1.1334880590438843,
+      "step": 132
+    },
+    {
+      "epoch": 0.056540084388185655,
+      "grad_norm": 0.5987792015075684,
+      "learning_rate": 2.3374340949033394e-05,
+      "loss": 1.1546804904937744,
+      "step": 134
+    },
+    {
+      "epoch": 0.05738396624472574,
+      "grad_norm": 0.45355498790740967,
+      "learning_rate": 2.372583479789104e-05,
+      "loss": 1.194953441619873,
+      "step": 136
+    },
+    {
+      "epoch": 0.05822784810126582,
+      "grad_norm": 0.5373698472976685,
+      "learning_rate": 2.4077328646748684e-05,
+      "loss": 1.1067466735839844,
+      "step": 138
+    },
+    {
+      "epoch": 0.05907172995780591,
+      "grad_norm": 0.48734328150749207,
+      "learning_rate": 2.442882249560633e-05,
+      "loss": 1.188468098640442,
+      "step": 140
+    },
+    {
+      "epoch": 0.059915611814345994,
+      "grad_norm": 0.4692173898220062,
+      "learning_rate": 2.478031634446397e-05,
+      "loss": 1.1624362468719482,
+      "step": 142
+    },
+    {
+      "epoch": 0.060759493670886074,
+      "grad_norm": 0.532554030418396,
+      "learning_rate": 2.513181019332162e-05,
+      "loss": 1.0978907346725464,
+      "step": 144
+    },
+    {
+      "epoch": 0.06160337552742616,
+      "grad_norm": 0.5853802561759949,
+      "learning_rate": 2.5483304042179264e-05,
+      "loss": 1.2030781507492065,
+      "step": 146
+    },
+    {
+      "epoch": 0.06244725738396625,
+      "grad_norm": 0.5061611533164978,
+      "learning_rate": 2.583479789103691e-05,
+      "loss": 1.082366943359375,
+      "step": 148
+    },
+    {
+      "epoch": 0.06329113924050633,
+      "grad_norm": 0.49426141381263733,
+      "learning_rate": 2.6186291739894554e-05,
+      "loss": 1.10564386844635,
+      "step": 150
+    },
+    {
+      "epoch": 0.06413502109704641,
+      "grad_norm": 0.5846618413925171,
+      "learning_rate": 2.6537785588752196e-05,
+      "loss": 1.1992807388305664,
+      "step": 152
+    },
+    {
+      "epoch": 0.06497890295358649,
+      "grad_norm": 0.5517552495002747,
+      "learning_rate": 2.6889279437609844e-05,
+      "loss": 1.1757566928863525,
+      "step": 154
+    },
+    {
+      "epoch": 0.06582278481012659,
+      "grad_norm": 0.5667305588722229,
+      "learning_rate": 2.7240773286467486e-05,
+      "loss": 1.0548783540725708,
+      "step": 156
+    },
+    {
+      "epoch": 0.06666666666666667,
+      "grad_norm": 0.6760414242744446,
+      "learning_rate": 2.7592267135325134e-05,
+      "loss": 1.184364914894104,
+      "step": 158
+    },
+    {
+      "epoch": 0.06751054852320675,
+      "grad_norm": 0.5261430740356445,
+      "learning_rate": 2.7943760984182776e-05,
+      "loss": 1.1945042610168457,
+      "step": 160
+    },
+    {
+      "epoch": 0.06835443037974684,
+      "grad_norm": 0.6155015230178833,
+      "learning_rate": 2.8295254833040425e-05,
+      "loss": 1.2021973133087158,
+      "step": 162
+    },
+    {
+      "epoch": 0.06919831223628692,
+      "grad_norm": 0.6131619215011597,
+      "learning_rate": 2.8646748681898066e-05,
+      "loss": 1.144123911857605,
+      "step": 164
+    },
+    {
+      "epoch": 0.070042194092827,
+      "grad_norm": 0.5749185681343079,
+      "learning_rate": 2.8998242530755715e-05,
+      "loss": 1.1329256296157837,
+      "step": 166
+    },
+    {
+      "epoch": 0.07088607594936709,
+      "grad_norm": 0.5243118405342102,
+      "learning_rate": 2.9349736379613356e-05,
+      "loss": 1.0892387628555298,
+      "step": 168
+    },
+    {
+      "epoch": 0.07172995780590717,
+      "grad_norm": 0.7190104722976685,
+      "learning_rate": 2.9701230228471005e-05,
+      "loss": 1.163260817527771,
+      "step": 170
+    },
+    {
+      "epoch": 0.07257383966244725,
+      "grad_norm": 0.5486982464790344,
+      "learning_rate": 3.0052724077328647e-05,
+      "loss": 1.0880777835845947,
+      "step": 172
+    },
+    {
+      "epoch": 0.07341772151898734,
+      "grad_norm": 0.5020889043807983,
+      "learning_rate": 3.0404217926186295e-05,
+      "loss": 1.0433368682861328,
+      "step": 174
+    },
+    {
+      "epoch": 0.07426160337552742,
+      "grad_norm": 0.47329774498939514,
+      "learning_rate": 3.075571177504394e-05,
+      "loss": 1.0528991222381592,
+      "step": 176
+    },
+    {
+      "epoch": 0.0751054852320675,
+      "grad_norm": 0.6635547876358032,
+      "learning_rate": 3.110720562390158e-05,
+      "loss": 1.1627811193466187,
+      "step": 178
+    },
+    {
+      "epoch": 0.0759493670886076,
+      "grad_norm": 0.5624618530273438,
+      "learning_rate": 3.145869947275923e-05,
+      "loss": 1.084869384765625,
+      "step": 180
+    },
+    {
+      "epoch": 0.07679324894514768,
+      "grad_norm": 0.6029536724090576,
+      "learning_rate": 3.181019332161687e-05,
+      "loss": 1.2227671146392822,
+      "step": 182
+    },
+    {
+      "epoch": 0.07763713080168777,
+      "grad_norm": 0.930959939956665,
+      "learning_rate": 3.216168717047452e-05,
+      "loss": 1.0955452919006348,
+      "step": 184
+    },
+    {
+      "epoch": 0.07848101265822785,
+      "grad_norm": 0.5326952338218689,
+      "learning_rate": 3.251318101933216e-05,
+      "loss": 1.0640798807144165,
+      "step": 186
+    },
+    {
+      "epoch": 0.07932489451476793,
+      "grad_norm": 0.5484727621078491,
+      "learning_rate": 3.286467486818981e-05,
+      "loss": 1.0700589418411255,
+      "step": 188
+    },
+    {
+      "epoch": 0.08016877637130802,
+      "grad_norm": 0.605273425579071,
+      "learning_rate": 3.3216168717047456e-05,
+      "loss": 1.1593081951141357,
+      "step": 190
+    },
+    {
+      "epoch": 0.0810126582278481,
+      "grad_norm": 0.5704394578933716,
+      "learning_rate": 3.35676625659051e-05,
+      "loss": 1.1617076396942139,
+      "step": 192
+    },
+    {
+      "epoch": 0.08185654008438818,
+      "grad_norm": 0.5929452180862427,
+      "learning_rate": 3.3919156414762746e-05,
+      "loss": 1.1346839666366577,
+      "step": 194
+    },
+    {
+      "epoch": 0.08270042194092828,
+      "grad_norm": 0.5624077916145325,
+      "learning_rate": 3.427065026362039e-05,
+      "loss": 1.0934710502624512,
+      "step": 196
+    },
+    {
+      "epoch": 0.08354430379746836,
+      "grad_norm": 0.6717425584793091,
+      "learning_rate": 3.4622144112478036e-05,
+      "loss": 1.1810534000396729,
+      "step": 198
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "grad_norm": 0.5120199918746948,
+      "learning_rate": 3.4973637961335674e-05,
+      "loss": 1.1525514125823975,
+      "step": 200
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "eval_loss": 1.142486810684204,
+      "eval_runtime": 668.2356,
+      "eval_samples_per_second": 3.153,
+      "eval_steps_per_second": 3.153,
+      "step": 200
+    },
+    {
+      "epoch": 0.08523206751054853,
+      "grad_norm": 0.5144487023353577,
+      "learning_rate": 3.5325131810193326e-05,
+      "loss": 1.0243735313415527,
+      "step": 202
+    },
+    {
+      "epoch": 0.08607594936708861,
+      "grad_norm": 0.6325069069862366,
+      "learning_rate": 3.5676625659050964e-05,
+      "loss": 1.118743896484375,
+      "step": 204
+    },
+    {
+      "epoch": 0.08691983122362869,
+      "grad_norm": 0.5501633882522583,
+      "learning_rate": 3.6028119507908616e-05,
+      "loss": 1.0380504131317139,
+      "step": 206
+    },
+    {
+      "epoch": 0.08776371308016878,
+      "grad_norm": 0.6133899688720703,
+      "learning_rate": 3.6379613356766254e-05,
+      "loss": 0.9837555885314941,
+      "step": 208
+    },
+    {
+      "epoch": 0.08860759493670886,
+      "grad_norm": 0.5799810886383057,
+      "learning_rate": 3.6731107205623906e-05,
+      "loss": 1.090720295906067,
+      "step": 210
+    },
+    {
+      "epoch": 0.08945147679324894,
+      "grad_norm": 0.6039511561393738,
+      "learning_rate": 3.7082601054481544e-05,
+      "loss": 1.120232343673706,
+      "step": 212
+    },
+    {
+      "epoch": 0.09029535864978903,
+      "grad_norm": 0.5983024835586548,
+      "learning_rate": 3.7434094903339196e-05,
+      "loss": 1.096949815750122,
+      "step": 214
+    },
+    {
+      "epoch": 0.09113924050632911,
+      "grad_norm": 0.5641079545021057,
+      "learning_rate": 3.7785588752196835e-05,
+      "loss": 1.1226298809051514,
+      "step": 216
+    },
+    {
+      "epoch": 0.0919831223628692,
+      "grad_norm": 0.655717134475708,
+      "learning_rate": 3.8137082601054486e-05,
+      "loss": 1.1260643005371094,
+      "step": 218
+    },
+    {
+      "epoch": 0.09282700421940929,
+      "grad_norm": 0.6111898422241211,
+      "learning_rate": 3.848857644991213e-05,
+      "loss": 1.0777709484100342,
+      "step": 220
+    },
+    {
+      "epoch": 0.09367088607594937,
+      "grad_norm": 0.6821302771568298,
+      "learning_rate": 3.884007029876977e-05,
+      "loss": 1.10588800907135,
+      "step": 222
+    },
+    {
+      "epoch": 0.09451476793248945,
+      "grad_norm": 0.693175733089447,
+      "learning_rate": 3.919156414762742e-05,
+      "loss": 1.1498671770095825,
+      "step": 224
+    },
+    {
+      "epoch": 0.09535864978902954,
+      "grad_norm": 0.5288166403770447,
+      "learning_rate": 3.954305799648506e-05,
+      "loss": 1.0587562322616577,
+      "step": 226
+    },
+    {
+      "epoch": 0.09620253164556962,
+      "grad_norm": 0.6882867813110352,
+      "learning_rate": 3.989455184534271e-05,
+      "loss": 1.1107512712478638,
+      "step": 228
+    },
+    {
+      "epoch": 0.0970464135021097,
+      "grad_norm": 0.5834154486656189,
+      "learning_rate": 4.024604569420035e-05,
+      "loss": 1.020510196685791,
+      "step": 230
+    },
+    {
+      "epoch": 0.09789029535864979,
+      "grad_norm": 0.7157064080238342,
+      "learning_rate": 4.0597539543058e-05,
+      "loss": 1.0642449855804443,
+      "step": 232
+    },
+    {
+      "epoch": 0.09873417721518987,
+      "grad_norm": 0.6530708074569702,
+      "learning_rate": 4.094903339191564e-05,
+      "loss": 1.0359872579574585,
+      "step": 234
+    },
+    {
+      "epoch": 0.09957805907172995,
+      "grad_norm": 0.6329686045646667,
+      "learning_rate": 4.130052724077329e-05,
+      "loss": 1.050504446029663,
+      "step": 236
+    },
+    {
+      "epoch": 0.10042194092827005,
+      "grad_norm": 0.6597026586532593,
+      "learning_rate": 4.165202108963093e-05,
+      "loss": 1.2621175050735474,
+      "step": 238
+    },
+    {
+      "epoch": 0.10126582278481013,
+      "grad_norm": 0.6195225119590759,
+      "learning_rate": 4.200351493848858e-05,
+      "loss": 1.1218310594558716,
+      "step": 240
+    },
+    {
+      "epoch": 0.1021097046413502,
+      "grad_norm": 0.6764137744903564,
+      "learning_rate": 4.235500878734622e-05,
+      "loss": 1.1250728368759155,
+      "step": 242
+    },
+    {
+      "epoch": 0.1029535864978903,
+      "grad_norm": 0.552363395690918,
+      "learning_rate": 4.270650263620387e-05,
+      "loss": 1.028212308883667,
+      "step": 244
+    },
+    {
+      "epoch": 0.10379746835443038,
+      "grad_norm": 0.5620495676994324,
+      "learning_rate": 4.305799648506151e-05,
+      "loss": 1.0425450801849365,
+      "step": 246
+    },
+    {
+      "epoch": 0.10464135021097046,
+      "grad_norm": 0.6860032081604004,
+      "learning_rate": 4.3409490333919156e-05,
+      "loss": 1.144278883934021,
+      "step": 248
+    },
+    {
+      "epoch": 0.10548523206751055,
+      "grad_norm": 0.6033259034156799,
+      "learning_rate": 4.37609841827768e-05,
+      "loss": 1.1223982572555542,
+      "step": 250
+    },
+    {
+      "epoch": 0.10632911392405063,
+      "grad_norm": 0.6292146444320679,
+      "learning_rate": 4.4112478031634446e-05,
+      "loss": 1.1609960794448853,
+      "step": 252
+    },
+    {
+      "epoch": 0.10717299578059072,
+      "grad_norm": 0.7982883453369141,
+      "learning_rate": 4.44639718804921e-05,
+      "loss": 1.063547968864441,
+      "step": 254
+    },
+    {
+      "epoch": 0.1080168776371308,
+      "grad_norm": 0.7719110250473022,
+      "learning_rate": 4.4815465729349736e-05,
+      "loss": 1.0719804763793945,
+      "step": 256
+    },
+    {
+      "epoch": 0.10886075949367088,
+      "grad_norm": 0.6101011633872986,
+      "learning_rate": 4.516695957820739e-05,
+      "loss": 1.0778400897979736,
+      "step": 258
+    },
+    {
+      "epoch": 0.10970464135021098,
+      "grad_norm": 0.7300994396209717,
+      "learning_rate": 4.5518453427065026e-05,
+      "loss": 1.2129558324813843,
+      "step": 260
+    },
+    {
+      "epoch": 0.11054852320675106,
+      "grad_norm": 0.8348747491836548,
+      "learning_rate": 4.586994727592268e-05,
+      "loss": 1.221714735031128,
+      "step": 262
+    },
+    {
+      "epoch": 0.11139240506329114,
+      "grad_norm": 0.5445612072944641,
+      "learning_rate": 4.6221441124780316e-05,
+      "loss": 1.0187978744506836,
+      "step": 264
+    },
+    {
+      "epoch": 0.11223628691983123,
+      "grad_norm": 0.6230319738388062,
+      "learning_rate": 4.657293497363797e-05,
+      "loss": 1.096561312675476,
+      "step": 266
+    },
+    {
+      "epoch": 0.11308016877637131,
+      "grad_norm": 0.6231237649917603,
+      "learning_rate": 4.6924428822495606e-05,
+      "loss": 1.089842438697815,
+      "step": 268
+    },
+    {
+      "epoch": 0.11392405063291139,
+      "grad_norm": 0.7178627252578735,
+      "learning_rate": 4.727592267135325e-05,
+      "loss": 1.0696645975112915,
+      "step": 270
+    },
+    {
+      "epoch": 0.11476793248945148,
+      "grad_norm": 0.6895854473114014,
+      "learning_rate": 4.7627416520210896e-05,
+      "loss": 1.0511361360549927,
+      "step": 272
+    },
+    {
+      "epoch": 0.11561181434599156,
+      "grad_norm": 0.6046878695487976,
+      "learning_rate": 4.797891036906854e-05,
+      "loss": 1.1373958587646484,
+      "step": 274
+    },
+    {
+      "epoch": 0.11645569620253164,
+      "grad_norm": 0.6524552702903748,
+      "learning_rate": 4.833040421792619e-05,
+      "loss": 1.0734186172485352,
+      "step": 276
+    },
+    {
+      "epoch": 0.11729957805907174,
+      "grad_norm": 0.6331019997596741,
+      "learning_rate": 4.868189806678383e-05,
+      "loss": 1.123913049697876,
+      "step": 278
+    },
+    {
+      "epoch": 0.11814345991561181,
+      "grad_norm": 0.5919018983840942,
+      "learning_rate": 4.903339191564148e-05,
+      "loss": 1.0635710954666138,
+      "step": 280
+    },
+    {
+      "epoch": 0.1189873417721519,
+      "grad_norm": 0.6067633032798767,
+      "learning_rate": 4.938488576449912e-05,
+      "loss": 1.0429247617721558,
+      "step": 282
+    },
+    {
+      "epoch": 0.11983122362869199,
+      "grad_norm": 0.6583750247955322,
+      "learning_rate": 4.9736379613356774e-05,
+      "loss": 1.1397464275360107,
+      "step": 284
+    },
+    {
+      "epoch": 0.12067510548523207,
+      "grad_norm": 0.6200069785118103,
+      "learning_rate": 5.008787346221442e-05,
+      "loss": 1.0590803623199463,
+      "step": 286
+    },
+    {
+      "epoch": 0.12151898734177215,
+      "grad_norm": 0.6798665523529053,
+      "learning_rate": 5.0439367311072064e-05,
+      "loss": 1.1318789720535278,
+      "step": 288
+    },
+    {
+      "epoch": 0.12236286919831224,
+      "grad_norm": 0.7508794069290161,
+      "learning_rate": 5.07908611599297e-05,
+      "loss": 1.0934956073760986,
+      "step": 290
+    },
+    {
+      "epoch": 0.12320675105485232,
+      "grad_norm": 0.6901452541351318,
+      "learning_rate": 5.114235500878735e-05,
+      "loss": 1.163407802581787,
+      "step": 292
+    },
+    {
+      "epoch": 0.1240506329113924,
+      "grad_norm": 0.6423285603523254,
+      "learning_rate": 5.1493848857644985e-05,
+      "loss": 1.09059476852417,
+      "step": 294
+    },
+    {
+      "epoch": 0.1248945147679325,
+      "grad_norm": 0.6839275360107422,
+      "learning_rate": 5.1845342706502644e-05,
+      "loss": 1.0690211057662964,
+      "step": 296
+    },
+    {
+      "epoch": 0.1257383966244726,
+      "grad_norm": 0.6350128054618835,
+      "learning_rate": 5.219683655536028e-05,
+      "loss": 0.982322096824646,
+      "step": 298
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "grad_norm": 0.7136530876159668,
+      "learning_rate": 5.254833040421793e-05,
+      "loss": 1.1132930517196655,
+      "step": 300
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "eval_loss": 1.0952109098434448,
+      "eval_runtime": 677.0652,
+      "eval_samples_per_second": 3.112,
+      "eval_steps_per_second": 3.112,
+      "step": 300
+    },
+    {
+      "epoch": 0.12742616033755275,
+      "grad_norm": 0.7339721322059631,
+      "learning_rate": 5.289982425307557e-05,
+      "loss": 0.973595917224884,
+      "step": 302
+    },
+    {
+      "epoch": 0.12827004219409283,
+      "grad_norm": 0.5941481590270996,
+      "learning_rate": 5.3251318101933224e-05,
+      "loss": 0.9819849729537964,
+      "step": 304
+    },
+    {
+      "epoch": 0.1291139240506329,
+      "grad_norm": 0.7153938412666321,
+      "learning_rate": 5.360281195079086e-05,
+      "loss": 1.0315470695495605,
+      "step": 306
+    },
+    {
+      "epoch": 0.12995780590717299,
+      "grad_norm": 0.5167180299758911,
+      "learning_rate": 5.395430579964851e-05,
+      "loss": 0.9492001533508301,
+      "step": 308
+    },
+    {
+      "epoch": 0.1308016877637131,
+      "grad_norm": 0.6055944561958313,
+      "learning_rate": 5.430579964850615e-05,
+      "loss": 1.0156209468841553,
+      "step": 310
+    },
+    {
+      "epoch": 0.13164556962025317,
+      "grad_norm": 0.7662386298179626,
+      "learning_rate": 5.4657293497363805e-05,
+      "loss": 1.1791651248931885,
+      "step": 312
+    },
+    {
+      "epoch": 0.13248945147679325,
+      "grad_norm": 0.6065546274185181,
+      "learning_rate": 5.500878734622145e-05,
+      "loss": 1.0009297132492065,
+      "step": 314
+    },
+    {
+      "epoch": 0.13333333333333333,
+      "grad_norm": 0.604225754737854,
+      "learning_rate": 5.536028119507909e-05,
+      "loss": 1.0208244323730469,
+      "step": 316
+    },
+    {
+      "epoch": 0.1341772151898734,
+      "grad_norm": 0.6186763048171997,
+      "learning_rate": 5.571177504393673e-05,
+      "loss": 0.9968416690826416,
+      "step": 318
+    },
+    {
+      "epoch": 0.1350210970464135,
+      "grad_norm": 0.7100363969802856,
+      "learning_rate": 5.606326889279437e-05,
+      "loss": 0.9540256857872009,
+      "step": 320
+    },
+    {
+      "epoch": 0.1358649789029536,
+      "grad_norm": 0.6979711055755615,
+      "learning_rate": 5.641476274165203e-05,
+      "loss": 1.0631953477859497,
+      "step": 322
+    },
+    {
+      "epoch": 0.13670886075949368,
+      "grad_norm": 0.6237109303474426,
+      "learning_rate": 5.676625659050967e-05,
+      "loss": 1.0170501470565796,
+      "step": 324
+    },
+    {
+      "epoch": 0.13755274261603376,
+      "grad_norm": 0.6525548696517944,
+      "learning_rate": 5.711775043936731e-05,
+      "loss": 1.0715603828430176,
+      "step": 326
+    },
+    {
+      "epoch": 0.13839662447257384,
+      "grad_norm": 0.6869221329689026,
+      "learning_rate": 5.746924428822495e-05,
+      "loss": 1.0111541748046875,
+      "step": 328
+    },
+    {
+      "epoch": 0.13924050632911392,
+      "grad_norm": 0.553188145160675,
+      "learning_rate": 5.782073813708261e-05,
+      "loss": 1.0311682224273682,
+      "step": 330
+    },
+    {
+      "epoch": 0.140084388185654,
+      "grad_norm": 0.6760852932929993,
+      "learning_rate": 5.817223198594025e-05,
+      "loss": 1.0213634967803955,
+      "step": 332
+    },
+    {
+      "epoch": 0.1409282700421941,
+      "grad_norm": 0.5907419919967651,
+      "learning_rate": 5.8523725834797894e-05,
+      "loss": 0.9748594164848328,
+      "step": 334
+    },
+    {
+      "epoch": 0.14177215189873418,
+      "grad_norm": 0.7044920921325684,
+      "learning_rate": 5.887521968365554e-05,
+      "loss": 1.05863356590271,
+      "step": 336
+    },
+    {
+      "epoch": 0.14261603375527426,
+      "grad_norm": 0.679073691368103,
+      "learning_rate": 5.922671353251318e-05,
+      "loss": 1.1341127157211304,
+      "step": 338
+    },
+    {
+      "epoch": 0.14345991561181434,
+      "grad_norm": 0.7676237225532532,
+      "learning_rate": 5.957820738137083e-05,
+      "loss": 0.9540836215019226,
+      "step": 340
+    },
+    {
+      "epoch": 0.14430379746835442,
+      "grad_norm": 0.6313899755477905,
+      "learning_rate": 5.9929701230228474e-05,
+      "loss": 1.0585911273956299,
+      "step": 342
+    },
+    {
+      "epoch": 0.1451476793248945,
+      "grad_norm": 0.7123099565505981,
+      "learning_rate": 6.028119507908612e-05,
+      "loss": 1.0760118961334229,
+      "step": 344
+    },
+    {
+      "epoch": 0.1459915611814346,
+      "grad_norm": 0.585935652256012,
+      "learning_rate": 6.063268892794376e-05,
+      "loss": 1.036866307258606,
+      "step": 346
+    },
+    {
+      "epoch": 0.1468354430379747,
+      "grad_norm": 0.5643263459205627,
+      "learning_rate": 6.0984182776801416e-05,
+      "loss": 1.0242938995361328,
+      "step": 348
+    },
+    {
+      "epoch": 0.14767932489451477,
+      "grad_norm": 0.626761794090271,
+      "learning_rate": 6.133567662565906e-05,
+      "loss": 1.0497376918792725,
+      "step": 350
+    },
+    {
+      "epoch": 0.14852320675105485,
+      "grad_norm": 0.5106956958770752,
+      "learning_rate": 6.16871704745167e-05,
+      "loss": 0.9811885356903076,
+      "step": 352
+    },
+    {
+      "epoch": 0.14936708860759493,
+      "grad_norm": 0.6948089003562927,
+      "learning_rate": 6.203866432337434e-05,
+      "loss": 1.0715330839157104,
+      "step": 354
+    },
+    {
+      "epoch": 0.150210970464135,
+      "grad_norm": 0.699713945388794,
+      "learning_rate": 6.239015817223199e-05,
+      "loss": 1.0405226945877075,
+      "step": 356
+    },
+    {
+      "epoch": 0.15105485232067511,
+      "grad_norm": 0.6437667012214661,
+      "learning_rate": 6.274165202108964e-05,
+      "loss": 1.0490930080413818,
+      "step": 358
+    },
+    {
+      "epoch": 0.1518987341772152,
+      "grad_norm": 0.6952699422836304,
+      "learning_rate": 6.309314586994728e-05,
+      "loss": 0.9267548322677612,
+      "step": 360
+    },
+    {
+      "epoch": 0.15274261603375527,
+      "grad_norm": 0.6713186502456665,
+      "learning_rate": 6.344463971880492e-05,
+      "loss": 1.0427420139312744,
+      "step": 362
+    },
+    {
+      "epoch": 0.15358649789029535,
+      "grad_norm": 0.6750379800796509,
+      "learning_rate": 6.379613356766257e-05,
+      "loss": 1.048950433731079,
+      "step": 364
+    },
+    {
+      "epoch": 0.15443037974683543,
+      "grad_norm": 0.6053379774093628,
+      "learning_rate": 6.414762741652022e-05,
+      "loss": 1.0156004428863525,
+      "step": 366
+    },
+    {
+      "epoch": 0.15527426160337554,
+      "grad_norm": 0.8063633441925049,
+      "learning_rate": 6.449912126537786e-05,
+      "loss": 1.0020819902420044,
+      "step": 368
+    },
+    {
+      "epoch": 0.15611814345991562,
+      "grad_norm": 0.8027494549751282,
+      "learning_rate": 6.48506151142355e-05,
+      "loss": 1.055633783340454,
+      "step": 370
+    },
+    {
+      "epoch": 0.1569620253164557,
+      "grad_norm": 0.6580121517181396,
+      "learning_rate": 6.520210896309315e-05,
+      "loss": 1.0149940252304077,
+      "step": 372
+    },
+    {
+      "epoch": 0.15780590717299578,
+      "grad_norm": 0.6561233997344971,
+      "learning_rate": 6.55536028119508e-05,
+      "loss": 0.9769611954689026,
+      "step": 374
+    },
+    {
+      "epoch": 0.15864978902953586,
+      "grad_norm": 0.6444346308708191,
+      "learning_rate": 6.590509666080844e-05,
+      "loss": 0.9099349975585938,
+      "step": 376
+    },
+    {
+      "epoch": 0.15949367088607594,
+      "grad_norm": 0.5879359245300293,
+      "learning_rate": 6.625659050966608e-05,
+      "loss": 1.0797548294067383,
+      "step": 378
+    },
+    {
+      "epoch": 0.16033755274261605,
+      "grad_norm": 0.6994144916534424,
+      "learning_rate": 6.660808435852373e-05,
+      "loss": 1.0336791276931763,
+      "step": 380
+    },
+    {
+      "epoch": 0.16118143459915613,
+      "grad_norm": 0.6128669381141663,
+      "learning_rate": 6.695957820738138e-05,
+      "loss": 1.018118143081665,
+      "step": 382
+    },
+    {
+      "epoch": 0.1620253164556962,
+      "grad_norm": 1.0237540006637573,
+      "learning_rate": 6.731107205623902e-05,
+      "loss": 1.1405497789382935,
+      "step": 384
+    },
+    {
+      "epoch": 0.16286919831223629,
+      "grad_norm": 0.6091578006744385,
+      "learning_rate": 6.766256590509666e-05,
+      "loss": 1.0314189195632935,
+      "step": 386
+    },
+    {
+      "epoch": 0.16371308016877636,
+      "grad_norm": 0.5916037559509277,
+      "learning_rate": 6.801405975395431e-05,
+      "loss": 0.9564052820205688,
+      "step": 388
+    },
+    {
+      "epoch": 0.16455696202531644,
+      "grad_norm": 0.771653950214386,
+      "learning_rate": 6.836555360281195e-05,
+      "loss": 1.0023859739303589,
+      "step": 390
+    },
+    {
+      "epoch": 0.16540084388185655,
+      "grad_norm": 0.654658317565918,
+      "learning_rate": 6.87170474516696e-05,
+      "loss": 1.07024085521698,
+      "step": 392
+    },
+    {
+      "epoch": 0.16624472573839663,
+      "grad_norm": 0.6611968874931335,
+      "learning_rate": 6.906854130052724e-05,
+      "loss": 1.0552500486373901,
+      "step": 394
+    },
+    {
+      "epoch": 0.1670886075949367,
+      "grad_norm": 0.6955893039703369,
+      "learning_rate": 6.942003514938489e-05,
+      "loss": 1.0562875270843506,
+      "step": 396
+    },
+    {
+      "epoch": 0.1679324894514768,
+      "grad_norm": 0.6666058301925659,
+      "learning_rate": 6.977152899824253e-05,
+      "loss": 0.9850592017173767,
+      "step": 398
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "grad_norm": 0.6131711006164551,
+      "learning_rate": 7.012302284710018e-05,
+      "loss": 1.0077755451202393,
+      "step": 400
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "eval_loss": 1.0625108480453491,
+      "eval_runtime": 691.0068,
+      "eval_samples_per_second": 3.049,
+      "eval_steps_per_second": 3.049,
+      "step": 400
+    },
+    {
+      "epoch": 0.16962025316455695,
+      "grad_norm": 0.6286499500274658,
+      "learning_rate": 7.047451669595783e-05,
+      "loss": 1.1012427806854248,
+      "step": 402
+    },
+    {
+      "epoch": 0.17046413502109706,
+      "grad_norm": 0.6639351844787598,
+      "learning_rate": 7.082601054481547e-05,
+      "loss": 1.0379719734191895,
+      "step": 404
+    },
+    {
+      "epoch": 0.17130801687763714,
+      "grad_norm": 0.750401496887207,
+      "learning_rate": 7.117750439367311e-05,
+      "loss": 1.031856656074524,
+      "step": 406
+    },
+    {
+      "epoch": 0.17215189873417722,
+      "grad_norm": 0.8084847331047058,
+      "learning_rate": 7.152899824253075e-05,
+      "loss": 1.0493193864822388,
+      "step": 408
+    },
+    {
+      "epoch": 0.1729957805907173,
+      "grad_norm": 0.7448462247848511,
+      "learning_rate": 7.188049209138841e-05,
+      "loss": 1.1012418270111084,
+      "step": 410
+    },
+    {
+      "epoch": 0.17383966244725738,
+      "grad_norm": 0.5841867923736572,
+      "learning_rate": 7.223198594024605e-05,
+      "loss": 0.9926692247390747,
+      "step": 412
+    },
+    {
+      "epoch": 0.17468354430379746,
+      "grad_norm": 0.7125606536865234,
+      "learning_rate": 7.258347978910369e-05,
+      "loss": 1.0588877201080322,
+      "step": 414
+    },
+    {
+      "epoch": 0.17552742616033756,
+      "grad_norm": 0.5750942230224609,
+      "learning_rate": 7.293497363796134e-05,
+      "loss": 1.038270354270935,
+      "step": 416
+    },
+    {
+      "epoch": 0.17637130801687764,
+      "grad_norm": 0.565444827079773,
+      "learning_rate": 7.328646748681899e-05,
+      "loss": 0.9843021035194397,
+      "step": 418
+    },
+    {
+      "epoch": 0.17721518987341772,
+      "grad_norm": 0.5825693011283875,
+      "learning_rate": 7.363796133567663e-05,
+      "loss": 1.0731632709503174,
+      "step": 420
+    },
+    {
+      "epoch": 0.1780590717299578,
+      "grad_norm": 0.6267391443252563,
+      "learning_rate": 7.398945518453427e-05,
+      "loss": 1.0061273574829102,
+      "step": 422
+    },
+    {
+      "epoch": 0.17890295358649788,
+      "grad_norm": 0.6621372103691101,
+      "learning_rate": 7.434094903339192e-05,
+      "loss": 1.0461612939834595,
+      "step": 424
+    },
+    {
+      "epoch": 0.17974683544303796,
+      "grad_norm": 0.6635435223579407,
+      "learning_rate": 7.469244288224957e-05,
+      "loss": 0.9789207577705383,
+      "step": 426
+    },
+    {
+      "epoch": 0.18059071729957807,
+      "grad_norm": 0.6342346668243408,
+      "learning_rate": 7.504393673110721e-05,
+      "loss": 1.0527069568634033,
+      "step": 428
+    },
+    {
+      "epoch": 0.18143459915611815,
+      "grad_norm": 0.6762149930000305,
+      "learning_rate": 7.539543057996485e-05,
+      "loss": 0.9708702564239502,
+      "step": 430
+    },
+    {
+      "epoch": 0.18227848101265823,
+      "grad_norm": 0.7073282599449158,
+      "learning_rate": 7.57469244288225e-05,
+      "loss": 1.0509834289550781,
+      "step": 432
+    },
+    {
+      "epoch": 0.1831223628691983,
+      "grad_norm": 0.6917856931686401,
+      "learning_rate": 7.609841827768014e-05,
+      "loss": 1.0128819942474365,
+      "step": 434
+    },
+    {
+      "epoch": 0.1839662447257384,
+      "grad_norm": 0.5574942231178284,
+      "learning_rate": 7.644991212653779e-05,
+      "loss": 0.989395797252655,
+      "step": 436
+    },
+    {
+      "epoch": 0.1848101265822785,
+      "grad_norm": 0.640765905380249,
+      "learning_rate": 7.680140597539543e-05,
+      "loss": 0.9846042990684509,
+      "step": 438
+    },
+    {
+      "epoch": 0.18565400843881857,
+      "grad_norm": 0.6699127554893494,
+      "learning_rate": 7.715289982425308e-05,
+      "loss": 1.0344442129135132,
+      "step": 440
+    },
+    {
+      "epoch": 0.18649789029535865,
+      "grad_norm": 0.6164930462837219,
+      "learning_rate": 7.750439367311072e-05,
+      "loss": 1.0179373025894165,
+      "step": 442
+    },
+    {
+      "epoch": 0.18734177215189873,
+      "grad_norm": 0.6880720853805542,
+      "learning_rate": 7.785588752196837e-05,
+      "loss": 1.0518895387649536,
+      "step": 444
+    },
+    {
+      "epoch": 0.1881856540084388,
+      "grad_norm": 0.6501413583755493,
+      "learning_rate": 7.820738137082601e-05,
+      "loss": 1.0442606210708618,
+      "step": 446
+    },
+    {
+      "epoch": 0.1890295358649789,
+      "grad_norm": 0.6076085567474365,
+      "learning_rate": 7.855887521968366e-05,
+      "loss": 0.9828442335128784,
+      "step": 448
+    },
+    {
+      "epoch": 0.189873417721519,
+      "grad_norm": 0.6418202519416809,
+      "learning_rate": 7.89103690685413e-05,
+      "loss": 1.0573710203170776,
+      "step": 450
+    },
+    {
+      "epoch": 0.19071729957805908,
+      "grad_norm": 0.7055076360702515,
+      "learning_rate": 7.926186291739895e-05,
+      "loss": 1.0216103792190552,
+      "step": 452
+    },
+    {
+      "epoch": 0.19156118143459916,
+      "grad_norm": 0.5668330192565918,
+      "learning_rate": 7.961335676625659e-05,
+      "loss": 0.9837722778320312,
+      "step": 454
+    },
+    {
+      "epoch": 0.19240506329113924,
+      "grad_norm": 0.6419380307197571,
+      "learning_rate": 7.996485061511424e-05,
+      "loss": 1.0003894567489624,
+      "step": 456
+    },
+    {
+      "epoch": 0.19324894514767932,
+      "grad_norm": 0.5949198007583618,
+      "learning_rate": 8.031634446397188e-05,
+      "loss": 1.0609031915664673,
+      "step": 458
+    },
+    {
+      "epoch": 0.1940928270042194,
+      "grad_norm": 0.7032039761543274,
+      "learning_rate": 8.066783831282952e-05,
+      "loss": 1.0543403625488281,
+      "step": 460
+    },
+    {
+      "epoch": 0.1949367088607595,
+      "grad_norm": 0.5775868892669678,
+      "learning_rate": 8.101933216168718e-05,
+      "loss": 0.9819303154945374,
+      "step": 462
+    },
+    {
+      "epoch": 0.19578059071729959,
+      "grad_norm": 0.9301062226295471,
+      "learning_rate": 8.137082601054482e-05,
+      "loss": 1.0542067289352417,
+      "step": 464
+    },
+    {
+      "epoch": 0.19662447257383966,
+      "grad_norm": 0.6193217039108276,
+      "learning_rate": 8.172231985940246e-05,
+      "loss": 0.9966341257095337,
+      "step": 466
+    },
+    {
+      "epoch": 0.19746835443037974,
+      "grad_norm": 0.6286146640777588,
+      "learning_rate": 8.20738137082601e-05,
+      "loss": 1.0474121570587158,
+      "step": 468
+    },
+    {
+      "epoch": 0.19831223628691982,
+      "grad_norm": 0.7418972253799438,
+      "learning_rate": 8.242530755711776e-05,
+      "loss": 0.9549239277839661,
+      "step": 470
+    },
+    {
+      "epoch": 0.1991561181434599,
+      "grad_norm": 0.6122808456420898,
+      "learning_rate": 8.27768014059754e-05,
+      "loss": 1.0191338062286377,
+      "step": 472
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.6375362277030945,
+      "learning_rate": 8.312829525483304e-05,
+      "loss": 1.0987539291381836,
+      "step": 474
+    },
+    {
+      "epoch": 0.2008438818565401,
+      "grad_norm": 0.6459513306617737,
+      "learning_rate": 8.347978910369068e-05,
+      "loss": 1.0369136333465576,
+      "step": 476
+    },
+    {
+      "epoch": 0.20168776371308017,
+      "grad_norm": 0.7029640674591064,
+      "learning_rate": 8.383128295254833e-05,
+      "loss": 1.0582096576690674,
+      "step": 478
+    },
+    {
+      "epoch": 0.20253164556962025,
+      "grad_norm": 0.6345387697219849,
+      "learning_rate": 8.418277680140598e-05,
+      "loss": 1.022916316986084,
+      "step": 480
+    },
+    {
+      "epoch": 0.20337552742616033,
+      "grad_norm": 0.5764590501785278,
+      "learning_rate": 8.453427065026362e-05,
+      "loss": 0.973024308681488,
+      "step": 482
+    },
+    {
+      "epoch": 0.2042194092827004,
+      "grad_norm": 0.5884482860565186,
+      "learning_rate": 8.488576449912127e-05,
+      "loss": 1.0292812585830688,
+      "step": 484
+    },
+    {
+      "epoch": 0.20506329113924052,
+      "grad_norm": 0.616357147693634,
+      "learning_rate": 8.523725834797891e-05,
+      "loss": 1.0083447694778442,
+      "step": 486
+    },
+    {
+      "epoch": 0.2059071729957806,
+      "grad_norm": 0.7671196460723877,
+      "learning_rate": 8.558875219683656e-05,
+      "loss": 0.9936985373497009,
+      "step": 488
+    },
+    {
+      "epoch": 0.20675105485232068,
+      "grad_norm": 0.6197299957275391,
+      "learning_rate": 8.59402460456942e-05,
+      "loss": 1.051513910293579,
+      "step": 490
+    },
+    {
+      "epoch": 0.20759493670886076,
+      "grad_norm": 0.6912890672683716,
+      "learning_rate": 8.629173989455185e-05,
+      "loss": 0.9474978446960449,
+      "step": 492
+    },
+    {
+      "epoch": 0.20843881856540084,
+      "grad_norm": 0.6941592693328857,
+      "learning_rate": 8.664323374340949e-05,
+      "loss": 1.0671660900115967,
+      "step": 494
+    },
+    {
+      "epoch": 0.20928270042194091,
+      "grad_norm": 0.5889528393745422,
+      "learning_rate": 8.699472759226714e-05,
+      "loss": 1.0020159482955933,
+      "step": 496
+    },
+    {
+      "epoch": 0.21012658227848102,
+      "grad_norm": 0.6478549838066101,
+      "learning_rate": 8.734622144112478e-05,
+      "loss": 1.0165860652923584,
+      "step": 498
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "grad_norm": 0.6018255949020386,
+      "learning_rate": 8.769771528998243e-05,
+      "loss": 0.8798263072967529,
+      "step": 500
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "eval_loss": 1.042096495628357,
+      "eval_runtime": 692.4361,
+      "eval_samples_per_second": 3.043,
+      "eval_steps_per_second": 3.043,
+      "step": 500
+    },
+    {
+      "epoch": 0.21181434599156118,
+      "grad_norm": 0.578990638256073,
+      "learning_rate": 8.804920913884007e-05,
+      "loss": 1.092096209526062,
+      "step": 502
+    },
+    {
+      "epoch": 0.21265822784810126,
+      "grad_norm": 0.6597883701324463,
+      "learning_rate": 8.840070298769771e-05,
+      "loss": 1.0413451194763184,
+      "step": 504
+    },
+    {
+      "epoch": 0.21350210970464134,
+      "grad_norm": 0.6660305261611938,
+      "learning_rate": 8.875219683655536e-05,
+      "loss": 1.0073142051696777,
+      "step": 506
+    },
+    {
+      "epoch": 0.21434599156118145,
+      "grad_norm": 0.6283115148544312,
+      "learning_rate": 8.910369068541301e-05,
+      "loss": 1.0319768190383911,
+      "step": 508
+    },
+    {
+      "epoch": 0.21518987341772153,
+      "grad_norm": 0.6257343292236328,
+      "learning_rate": 8.945518453427065e-05,
+      "loss": 1.0046353340148926,
+      "step": 510
+    },
+    {
+      "epoch": 0.2160337552742616,
+      "grad_norm": 0.5530875325202942,
+      "learning_rate": 8.980667838312829e-05,
+      "loss": 0.9169099926948547,
+      "step": 512
+    },
+    {
+      "epoch": 0.2168776371308017,
+      "grad_norm": 0.5369633436203003,
+      "learning_rate": 9.015817223198594e-05,
+      "loss": 0.9081505537033081,
+      "step": 514
+    },
+    {
+      "epoch": 0.21772151898734177,
+      "grad_norm": 0.6618232131004333,
+      "learning_rate": 9.05096660808436e-05,
+      "loss": 0.9165045022964478,
+      "step": 516
+    },
+    {
+      "epoch": 0.21856540084388185,
+      "grad_norm": 0.600666344165802,
+      "learning_rate": 9.086115992970123e-05,
+      "loss": 0.91348797082901,
+      "step": 518
+    },
+    {
+      "epoch": 0.21940928270042195,
+      "grad_norm": 0.5919831991195679,
+      "learning_rate": 9.121265377855887e-05,
+      "loss": 1.006508469581604,
+      "step": 520
+    },
+    {
+      "epoch": 0.22025316455696203,
+      "grad_norm": 0.688058614730835,
+      "learning_rate": 9.156414762741654e-05,
+      "loss": 1.0013236999511719,
+      "step": 522
+    },
+    {
+      "epoch": 0.2210970464135021,
+      "grad_norm": 0.6721227765083313,
+      "learning_rate": 9.191564147627418e-05,
+      "loss": 1.0909923315048218,
+      "step": 524
+    },
+    {
+      "epoch": 0.2219409282700422,
+      "grad_norm": 0.5987313389778137,
+      "learning_rate": 9.226713532513181e-05,
+      "loss": 1.0117096900939941,
+      "step": 526
+    },
+    {
+      "epoch": 0.22278481012658227,
+      "grad_norm": 0.6191489696502686,
+      "learning_rate": 9.261862917398945e-05,
+      "loss": 1.0153647661209106,
+      "step": 528
+    },
+    {
+      "epoch": 0.22362869198312235,
+      "grad_norm": 0.6821563243865967,
+      "learning_rate": 9.29701230228471e-05,
+      "loss": 0.9649755954742432,
+      "step": 530
+    },
+    {
+      "epoch": 0.22447257383966246,
+      "grad_norm": 1.760398268699646,
+      "learning_rate": 9.332161687170476e-05,
+      "loss": 0.8673232197761536,
+      "step": 532
+    },
+    {
+      "epoch": 0.22531645569620254,
+      "grad_norm": 0.6670058369636536,
+      "learning_rate": 9.36731107205624e-05,
+      "loss": 0.9942440986633301,
+      "step": 534
+    },
+    {
+      "epoch": 0.22616033755274262,
+      "grad_norm": 0.7345916032791138,
+      "learning_rate": 9.402460456942003e-05,
+      "loss": 1.0364389419555664,
+      "step": 536
+    },
+    {
+      "epoch": 0.2270042194092827,
+      "grad_norm": 0.5946128964424133,
+      "learning_rate": 9.437609841827768e-05,
+      "loss": 0.9314924478530884,
+      "step": 538
+    },
+    {
+      "epoch": 0.22784810126582278,
+      "grad_norm": 0.5800848603248596,
+      "learning_rate": 9.472759226713534e-05,
+      "loss": 1.0694862604141235,
+      "step": 540
+    },
+    {
+      "epoch": 0.22869198312236286,
+      "grad_norm": 0.6712192893028259,
+      "learning_rate": 9.507908611599297e-05,
+      "loss": 1.03531014919281,
+      "step": 542
+    },
+    {
+      "epoch": 0.22953586497890296,
+      "grad_norm": 0.5641416311264038,
+      "learning_rate": 9.543057996485063e-05,
+      "loss": 0.9795235991477966,
+      "step": 544
+    },
+    {
+      "epoch": 0.23037974683544304,
+      "grad_norm": 0.50412517786026,
+      "learning_rate": 9.578207381370826e-05,
+      "loss": 0.9641494750976562,
+      "step": 546
+    },
+    {
+      "epoch": 0.23122362869198312,
+      "grad_norm": 0.579118549823761,
+      "learning_rate": 9.61335676625659e-05,
+      "loss": 0.9375281929969788,
+      "step": 548
+    },
+    {
+      "epoch": 0.2320675105485232,
+      "grad_norm": 0.5888341665267944,
+      "learning_rate": 9.648506151142355e-05,
+      "loss": 0.9414046406745911,
+      "step": 550
+    },
+    {
+      "epoch": 0.23291139240506328,
+      "grad_norm": 0.5595056414604187,
+      "learning_rate": 9.68365553602812e-05,
+      "loss": 0.9005617499351501,
+      "step": 552
+    },
+    {
+      "epoch": 0.23375527426160336,
+      "grad_norm": 0.6605326533317566,
+      "learning_rate": 9.718804920913884e-05,
+      "loss": 1.0283968448638916,
+      "step": 554
+    },
+    {
+      "epoch": 0.23459915611814347,
+      "grad_norm": 0.5657313466072083,
+      "learning_rate": 9.753954305799648e-05,
+      "loss": 1.0058249235153198,
+      "step": 556
+    },
+    {
+      "epoch": 0.23544303797468355,
+      "grad_norm": 0.5433364510536194,
+      "learning_rate": 9.789103690685413e-05,
+      "loss": 0.9835494756698608,
+      "step": 558
+    },
+    {
+      "epoch": 0.23628691983122363,
+      "grad_norm": 0.6129802465438843,
+      "learning_rate": 9.824253075571179e-05,
+      "loss": 1.054532527923584,
+      "step": 560
+    },
+    {
+      "epoch": 0.2371308016877637,
+      "grad_norm": 0.6496239304542542,
+      "learning_rate": 9.859402460456942e-05,
+      "loss": 1.0240973234176636,
+      "step": 562
+    },
+    {
+      "epoch": 0.2379746835443038,
+      "grad_norm": 0.6380873918533325,
+      "learning_rate": 9.894551845342706e-05,
+      "loss": 1.0229179859161377,
+      "step": 564
+    },
+    {
+      "epoch": 0.23881856540084387,
+      "grad_norm": 0.6151993870735168,
+      "learning_rate": 9.929701230228471e-05,
+      "loss": 1.0111570358276367,
+      "step": 566
+    },
+    {
+      "epoch": 0.23966244725738398,
+      "grad_norm": 0.5727584958076477,
+      "learning_rate": 9.964850615114237e-05,
+      "loss": 0.9450829029083252,
+      "step": 568
+    },
+    {
+      "epoch": 0.24050632911392406,
+      "grad_norm": 0.6620725989341736,
+      "learning_rate": 0.0001,
+      "loss": 0.9800319075584412,
+      "step": 570
+    },
+    {
+      "epoch": 0.24135021097046414,
+      "grad_norm": 0.6151163578033447,
+      "learning_rate": 0.00010035149384885764,
+      "loss": 0.9757438898086548,
+      "step": 572
+    },
+    {
+      "epoch": 0.24219409282700421,
+      "grad_norm": 0.5672140717506409,
+      "learning_rate": 0.0001007029876977153,
+      "loss": 0.9104921817779541,
+      "step": 574
+    },
+    {
+      "epoch": 0.2430379746835443,
+      "grad_norm": 0.5697256326675415,
+      "learning_rate": 0.00010105448154657293,
+      "loss": 1.1027376651763916,
+      "step": 576
+    },
+    {
+      "epoch": 0.2438818565400844,
+      "grad_norm": 0.5590381622314453,
+      "learning_rate": 0.00010140597539543057,
+      "loss": 1.0055404901504517,
+      "step": 578
+    },
+    {
+      "epoch": 0.24472573839662448,
+      "grad_norm": 0.5518567562103271,
+      "learning_rate": 0.00010175746924428824,
+      "loss": 1.020835518836975,
+      "step": 580
+    },
+    {
+      "epoch": 0.24556962025316456,
+      "grad_norm": 0.6338496208190918,
+      "learning_rate": 0.00010210896309314588,
+      "loss": 0.9528344869613647,
+      "step": 582
+    },
+    {
+      "epoch": 0.24641350210970464,
+      "grad_norm": 0.6497329473495483,
+      "learning_rate": 0.00010246045694200353,
+      "loss": 1.0088670253753662,
+      "step": 584
+    },
+    {
+      "epoch": 0.24725738396624472,
+      "grad_norm": 0.49888095259666443,
+      "learning_rate": 0.00010281195079086117,
+      "loss": 0.9961200952529907,
+      "step": 586
+    },
+    {
+      "epoch": 0.2481012658227848,
+      "grad_norm": 0.5680158734321594,
+      "learning_rate": 0.0001031634446397188,
+      "loss": 0.9635610580444336,
+      "step": 588
+    },
+    {
+      "epoch": 0.2489451476793249,
+      "grad_norm": 0.658168375492096,
+      "learning_rate": 0.00010351493848857646,
+      "loss": 0.9392287135124207,
+      "step": 590
+    },
+    {
+      "epoch": 0.249789029535865,
+      "grad_norm": 0.618262767791748,
+      "learning_rate": 0.0001038664323374341,
+      "loss": 0.9600516557693481,
+      "step": 592
+    },
+    {
+      "epoch": 0.25063291139240507,
+      "grad_norm": 0.6003909111022949,
+      "learning_rate": 0.00010421792618629173,
+      "loss": 1.005476713180542,
+      "step": 594
+    },
+    {
+      "epoch": 0.2514767932489452,
+      "grad_norm": 0.5437078475952148,
+      "learning_rate": 0.00010456942003514938,
+      "loss": 0.9523017406463623,
+      "step": 596
+    },
+    {
+      "epoch": 0.2523206751054852,
+      "grad_norm": 0.5524541735649109,
+      "learning_rate": 0.00010492091388400705,
+      "loss": 0.9526668787002563,
+      "step": 598
+    },
+    {
+      "epoch": 0.25316455696202533,
+      "grad_norm": 0.679504930973053,
+      "learning_rate": 0.00010527240773286469,
+      "loss": 1.019660472869873,
+      "step": 600
+    },
+    {
+      "epoch": 0.25316455696202533,
+      "eval_loss": 1.0193854570388794,
+      "eval_runtime": 677.9523,
+      "eval_samples_per_second": 3.108,
+      "eval_steps_per_second": 3.108,
+      "step": 600
+    },
+    {
+      "epoch": 0.2540084388185654,
+      "grad_norm": 0.5646136999130249,
+      "learning_rate": 0.00010562390158172233,
+      "loss": 0.9910882711410522,
+      "step": 602
+    },
+    {
+      "epoch": 0.2548523206751055,
+      "grad_norm": 0.5238093137741089,
+      "learning_rate": 0.00010597539543057998,
+      "loss": 0.9616432785987854,
+      "step": 604
+    },
+    {
+      "epoch": 0.25569620253164554,
+      "grad_norm": 0.7483857274055481,
+      "learning_rate": 0.00010632688927943762,
+      "loss": 1.0078275203704834,
+      "step": 606
+    },
+    {
+      "epoch": 0.25654008438818565,
+      "grad_norm": 0.578948974609375,
+      "learning_rate": 0.00010667838312829525,
+      "loss": 0.9827103018760681,
+      "step": 608
+    },
+    {
+      "epoch": 0.25738396624472576,
+      "grad_norm": 0.5525906085968018,
+      "learning_rate": 0.00010702987697715289,
+      "loss": 1.0423277616500854,
+      "step": 610
+    },
+    {
+      "epoch": 0.2582278481012658,
+      "grad_norm": 0.6721326112747192,
+      "learning_rate": 0.00010738137082601054,
+      "loss": 0.9561693072319031,
+      "step": 612
+    },
+    {
+      "epoch": 0.2590717299578059,
+      "grad_norm": 0.5701051354408264,
+      "learning_rate": 0.00010773286467486821,
+      "loss": 0.9602992534637451,
+      "step": 614
+    },
+    {
+      "epoch": 0.25991561181434597,
+      "grad_norm": 0.6349860429763794,
+      "learning_rate": 0.00010808435852372585,
+      "loss": 1.1422650814056396,
+      "step": 616
+    },
+    {
+      "epoch": 0.2607594936708861,
+      "grad_norm": 0.5496085286140442,
+      "learning_rate": 0.00010843585237258349,
+      "loss": 0.9762773513793945,
+      "step": 618
+    },
+    {
+      "epoch": 0.2616033755274262,
+      "grad_norm": 0.6080722808837891,
+      "learning_rate": 0.00010878734622144114,
+      "loss": 1.0133616924285889,
+      "step": 620
+    },
+    {
+      "epoch": 0.26244725738396624,
+      "grad_norm": 0.5450218915939331,
+      "learning_rate": 0.00010913884007029878,
+      "loss": 0.9385587573051453,
+      "step": 622
+    },
+    {
+      "epoch": 0.26329113924050634,
+      "grad_norm": 0.592106819152832,
+      "learning_rate": 0.00010949033391915641,
+      "loss": 0.9359989762306213,
+      "step": 624
+    },
+    {
+      "epoch": 0.2641350210970464,
+      "grad_norm": 0.6449427604675293,
+      "learning_rate": 0.00010984182776801407,
+      "loss": 1.0266027450561523,
+      "step": 626
+    },
+    {
+      "epoch": 0.2649789029535865,
+      "grad_norm": 0.538299560546875,
+      "learning_rate": 0.0001101933216168717,
+      "loss": 0.9303187131881714,
+      "step": 628
+    },
+    {
+      "epoch": 0.26582278481012656,
+      "grad_norm": 0.546316921710968,
+      "learning_rate": 0.00011054481546572934,
+      "loss": 0.9368857145309448,
+      "step": 630
+    },
+    {
+      "epoch": 0.26666666666666666,
+      "grad_norm": 0.5818730592727661,
+      "learning_rate": 0.00011089630931458701,
+      "loss": 0.9573145508766174,
+      "step": 632
+    },
+    {
+      "epoch": 0.26751054852320677,
+      "grad_norm": 0.5958262085914612,
+      "learning_rate": 0.00011124780316344465,
+      "loss": 0.9345449805259705,
+      "step": 634
+    },
+    {
+      "epoch": 0.2683544303797468,
+      "grad_norm": 0.6259077787399292,
+      "learning_rate": 0.0001115992970123023,
+      "loss": 1.0906590223312378,
+      "step": 636
+    },
+    {
+      "epoch": 0.26919831223628693,
+      "grad_norm": 0.589672863483429,
+      "learning_rate": 0.00011195079086115994,
+      "loss": 1.0757447481155396,
+      "step": 638
+    },
+    {
+      "epoch": 0.270042194092827,
+      "grad_norm": 0.5714080333709717,
+      "learning_rate": 0.00011230228471001758,
+      "loss": 0.9310855269432068,
+      "step": 640
+    },
+    {
+      "epoch": 0.2708860759493671,
+      "grad_norm": 0.45342639088630676,
+      "learning_rate": 0.00011265377855887523,
+      "loss": 0.9276360273361206,
+      "step": 642
+    },
+    {
+      "epoch": 0.2717299578059072,
+      "grad_norm": 0.6386750340461731,
+      "learning_rate": 0.00011300527240773287,
+      "loss": 1.084719181060791,
+      "step": 644
+    },
+    {
+      "epoch": 0.27257383966244725,
+      "grad_norm": 0.6446163654327393,
+      "learning_rate": 0.0001133567662565905,
+      "loss": 0.9763918519020081,
+      "step": 646
+    },
+    {
+      "epoch": 0.27341772151898736,
+      "grad_norm": 0.5925686359405518,
+      "learning_rate": 0.00011370826010544816,
+      "loss": 0.9517921805381775,
+      "step": 648
+    },
+    {
+      "epoch": 0.2742616033755274,
+      "grad_norm": 0.5399773716926575,
+      "learning_rate": 0.00011405975395430582,
+      "loss": 1.0587927103042603,
+      "step": 650
+    },
+    {
+      "epoch": 0.2751054852320675,
+      "grad_norm": 0.5872456431388855,
+      "learning_rate": 0.00011441124780316346,
+      "loss": 0.883341908454895,
+      "step": 652
+    },
+    {
+      "epoch": 0.2759493670886076,
+      "grad_norm": 0.5574564337730408,
+      "learning_rate": 0.0001147627416520211,
+      "loss": 1.0306891202926636,
+      "step": 654
+    },
+    {
+      "epoch": 0.2767932489451477,
+      "grad_norm": 0.47789013385772705,
+      "learning_rate": 0.00011511423550087874,
+      "loss": 0.8814032077789307,
+      "step": 656
+    },
+    {
+      "epoch": 0.2776371308016878,
+      "grad_norm": 0.5565530061721802,
+      "learning_rate": 0.00011546572934973639,
+      "loss": 0.9460552334785461,
+      "step": 658
+    },
+    {
+      "epoch": 0.27848101265822783,
+      "grad_norm": 0.5299761295318604,
+      "learning_rate": 0.00011581722319859403,
+      "loss": 0.9475110769271851,
+      "step": 660
+    },
+    {
+      "epoch": 0.27932489451476794,
+      "grad_norm": 0.6503344178199768,
+      "learning_rate": 0.00011616871704745166,
+      "loss": 1.0630913972854614,
+      "step": 662
+    },
+    {
+      "epoch": 0.280168776371308,
+      "grad_norm": 0.5794585943222046,
+      "learning_rate": 0.00011652021089630932,
+      "loss": 0.9389138221740723,
+      "step": 664
+    },
+    {
+      "epoch": 0.2810126582278481,
+      "grad_norm": 0.5762867331504822,
+      "learning_rate": 0.00011687170474516695,
+      "loss": 0.8934136033058167,
+      "step": 666
+    },
+    {
+      "epoch": 0.2818565400843882,
+      "grad_norm": 0.6565435528755188,
+      "learning_rate": 0.00011722319859402462,
+      "loss": 1.1072614192962646,
+      "step": 668
+    },
+    {
+      "epoch": 0.28270042194092826,
+      "grad_norm": 0.5819830298423767,
+      "learning_rate": 0.00011757469244288226,
+      "loss": 1.0501434803009033,
+      "step": 670
+    },
+    {
+      "epoch": 0.28354430379746837,
+      "grad_norm": 0.6071487069129944,
+      "learning_rate": 0.00011792618629173991,
+      "loss": 0.9880793690681458,
+      "step": 672
+    },
+    {
+      "epoch": 0.2843881856540084,
+      "grad_norm": 0.5765058398246765,
+      "learning_rate": 0.00011827768014059755,
+      "loss": 0.9670693874359131,
+      "step": 674
+    },
+    {
+      "epoch": 0.2852320675105485,
+      "grad_norm": 0.5245351791381836,
+      "learning_rate": 0.00011862917398945519,
+      "loss": 0.9602360725402832,
+      "step": 676
+    },
+    {
+      "epoch": 0.28607594936708863,
+      "grad_norm": 0.6189922094345093,
+      "learning_rate": 0.00011898066783831282,
+      "loss": 0.9684560894966125,
+      "step": 678
+    },
+    {
+      "epoch": 0.2869198312236287,
+      "grad_norm": 0.6138690710067749,
+      "learning_rate": 0.00011933216168717048,
+      "loss": 0.9465792775154114,
+      "step": 680
+    },
+    {
+      "epoch": 0.2877637130801688,
+      "grad_norm": 0.5371595621109009,
+      "learning_rate": 0.00011968365553602812,
+      "loss": 0.8495944738388062,
+      "step": 682
+    },
+    {
+      "epoch": 0.28860759493670884,
+      "grad_norm": 0.5549944639205933,
+      "learning_rate": 0.00012003514938488578,
+      "loss": 0.9663267135620117,
+      "step": 684
+    },
+    {
+      "epoch": 0.28945147679324895,
+      "grad_norm": 0.6484189033508301,
+      "learning_rate": 0.00012038664323374342,
+      "loss": 0.9736058712005615,
+      "step": 686
+    },
+    {
+      "epoch": 0.290295358649789,
+      "grad_norm": 0.540351390838623,
+      "learning_rate": 0.00012073813708260107,
+      "loss": 1.0591845512390137,
+      "step": 688
+    },
+    {
+      "epoch": 0.2911392405063291,
+      "grad_norm": 0.5657922029495239,
+      "learning_rate": 0.00012108963093145871,
+      "loss": 0.944908618927002,
+      "step": 690
+    },
+    {
+      "epoch": 0.2919831223628692,
+      "grad_norm": 0.6040505170822144,
+      "learning_rate": 0.00012144112478031635,
+      "loss": 1.0018219947814941,
+      "step": 692
+    },
+    {
+      "epoch": 0.29282700421940927,
+      "grad_norm": 0.5435477495193481,
+      "learning_rate": 0.000121792618629174,
+      "loss": 1.0351502895355225,
+      "step": 694
+    },
+    {
+      "epoch": 0.2936708860759494,
+      "grad_norm": 0.5712518692016602,
+      "learning_rate": 0.00012214411247803164,
+      "loss": 0.9935672283172607,
+      "step": 696
+    },
+    {
+      "epoch": 0.29451476793248943,
+      "grad_norm": 0.6138222813606262,
+      "learning_rate": 0.00012249560632688928,
+      "loss": 1.0165108442306519,
+      "step": 698
+    },
+    {
+      "epoch": 0.29535864978902954,
+      "grad_norm": 0.4495212435722351,
+      "learning_rate": 0.00012284710017574691,
+      "loss": 0.9334425926208496,
+      "step": 700
+    },
+    {
+      "epoch": 0.29535864978902954,
+      "eval_loss": 0.996929407119751,
+      "eval_runtime": 668.6398,
+      "eval_samples_per_second": 3.151,
+      "eval_steps_per_second": 3.151,
+      "step": 700
+    },
+    {
+      "epoch": 0.29620253164556964,
+      "grad_norm": 0.5321539044380188,
+      "learning_rate": 0.00012319859402460458,
+      "loss": 1.0516537427902222,
+      "step": 702
+    },
+    {
+      "epoch": 0.2970464135021097,
+      "grad_norm": 0.5716516971588135,
+      "learning_rate": 0.00012355008787346222,
+      "loss": 0.9387198686599731,
+      "step": 704
+    },
+    {
+      "epoch": 0.2978902953586498,
+      "grad_norm": 0.5617920160293579,
+      "learning_rate": 0.00012390158172231988,
+      "loss": 0.99737948179245,
+      "step": 706
+    },
+    {
+      "epoch": 0.29873417721518986,
+      "grad_norm": 0.4922899007797241,
+      "learning_rate": 0.00012425307557117752,
+      "loss": 0.9955025911331177,
+      "step": 708
+    },
+    {
+      "epoch": 0.29957805907172996,
+      "grad_norm": 0.543501615524292,
+      "learning_rate": 0.00012460456942003516,
+      "loss": 0.9124280214309692,
+      "step": 710
+    },
+    {
+      "epoch": 0.30042194092827,
+      "grad_norm": 0.49590054154396057,
+      "learning_rate": 0.0001249560632688928,
+      "loss": 0.9820216298103333,
+      "step": 712
+    },
+    {
+      "epoch": 0.3012658227848101,
+      "grad_norm": 0.5984305739402771,
+      "learning_rate": 0.00012530755711775044,
+      "loss": 1.0152074098587036,
+      "step": 714
+    },
+    {
+      "epoch": 0.30210970464135023,
+      "grad_norm": 0.9343504905700684,
+      "learning_rate": 0.00012565905096660807,
+      "loss": 1.0577725172042847,
+      "step": 716
+    },
+    {
+      "epoch": 0.3029535864978903,
+      "grad_norm": 0.5118702054023743,
+      "learning_rate": 0.0001260105448154657,
+      "loss": 0.9830358028411865,
+      "step": 718
+    },
+    {
+      "epoch": 0.3037974683544304,
+      "grad_norm": 0.4940392076969147,
+      "learning_rate": 0.00012636203866432338,
+      "loss": 0.9466043710708618,
+      "step": 720
+    },
+    {
+      "epoch": 0.30464135021097044,
+      "grad_norm": 0.5965693593025208,
+      "learning_rate": 0.00012671353251318104,
+      "loss": 1.015270709991455,
+      "step": 722
+    },
+    {
+      "epoch": 0.30548523206751055,
+      "grad_norm": 0.5020529627799988,
+      "learning_rate": 0.00012706502636203868,
+      "loss": 0.9703927636146545,
+      "step": 724
+    },
+    {
+      "epoch": 0.30632911392405066,
+      "grad_norm": 0.6067010164260864,
+      "learning_rate": 0.00012741652021089632,
+      "loss": 1.0255526304244995,
+      "step": 726
+    },
+    {
+      "epoch": 0.3071729957805907,
+      "grad_norm": 0.5931884050369263,
+      "learning_rate": 0.00012776801405975396,
+      "loss": 0.9335633516311646,
+      "step": 728
+    },
+    {
+      "epoch": 0.3080168776371308,
+      "grad_norm": 0.5938752293586731,
+      "learning_rate": 0.0001281195079086116,
+      "loss": 1.0921578407287598,
+      "step": 730
+    },
+    {
+      "epoch": 0.30886075949367087,
+      "grad_norm": 0.49728086590766907,
+      "learning_rate": 0.00012847100175746923,
+      "loss": 0.963066041469574,
+      "step": 732
+    },
+    {
+      "epoch": 0.309704641350211,
+      "grad_norm": 0.5452080965042114,
+      "learning_rate": 0.0001288224956063269,
+      "loss": 0.9513075351715088,
+      "step": 734
+    },
+    {
+      "epoch": 0.3105485232067511,
+      "grad_norm": 0.5497731566429138,
+      "learning_rate": 0.00012917398945518454,
+      "loss": 0.8576077222824097,
+      "step": 736
+    },
+    {
+      "epoch": 0.31139240506329113,
+      "grad_norm": 0.5580397248268127,
+      "learning_rate": 0.0001295254833040422,
+      "loss": 0.9542577862739563,
+      "step": 738
+    },
+    {
+      "epoch": 0.31223628691983124,
+      "grad_norm": 0.5890427827835083,
+      "learning_rate": 0.00012987697715289984,
+      "loss": 0.8992732167243958,
+      "step": 740
+    },
+    {
+      "epoch": 0.3130801687763713,
+      "grad_norm": 0.5942965745925903,
+      "learning_rate": 0.00013022847100175748,
+      "loss": 1.0322896242141724,
+      "step": 742
+    },
+    {
+      "epoch": 0.3139240506329114,
+      "grad_norm": 0.6341713070869446,
+      "learning_rate": 0.00013057996485061512,
+      "loss": 0.9217103719711304,
+      "step": 744
+    },
+    {
+      "epoch": 0.31476793248945145,
+      "grad_norm": 0.5294105410575867,
+      "learning_rate": 0.00013093145869947276,
+      "loss": 0.951789915561676,
+      "step": 746
+    },
+    {
+      "epoch": 0.31561181434599156,
+      "grad_norm": 0.6372058391571045,
+      "learning_rate": 0.0001312829525483304,
+      "loss": 0.9459875226020813,
+      "step": 748
+    },
+    {
+      "epoch": 0.31645569620253167,
+      "grad_norm": 0.5979796648025513,
+      "learning_rate": 0.00013163444639718806,
+      "loss": 0.9626097679138184,
+      "step": 750
+    },
+    {
+      "epoch": 0.3172995780590717,
+      "grad_norm": 0.5682399868965149,
+      "learning_rate": 0.0001319859402460457,
+      "loss": 1.0261781215667725,
+      "step": 752
+    },
+    {
+      "epoch": 0.3181434599156118,
+      "grad_norm": 0.5349125266075134,
+      "learning_rate": 0.00013233743409490336,
+      "loss": 0.9319828152656555,
+      "step": 754
+    },
+    {
+      "epoch": 0.3189873417721519,
+      "grad_norm": 0.6093934178352356,
+      "learning_rate": 0.000132688927943761,
+      "loss": 0.9216550588607788,
+      "step": 756
+    },
+    {
+      "epoch": 0.319831223628692,
+      "grad_norm": 0.5188612341880798,
+      "learning_rate": 0.00013304042179261864,
+      "loss": 0.901739776134491,
+      "step": 758
+    },
+    {
+      "epoch": 0.3206751054852321,
+      "grad_norm": 0.5877130627632141,
+      "learning_rate": 0.00013339191564147628,
+      "loss": 1.0362589359283447,
+      "step": 760
+    },
+    {
+      "epoch": 0.32151898734177214,
+      "grad_norm": 0.5542771816253662,
+      "learning_rate": 0.00013374340949033392,
+      "loss": 0.8787116408348083,
+      "step": 762
+    },
+    {
+      "epoch": 0.32236286919831225,
+      "grad_norm": 0.5084902048110962,
+      "learning_rate": 0.00013409490333919156,
+      "loss": 0.9237037301063538,
+      "step": 764
+    },
+    {
+      "epoch": 0.3232067510548523,
+      "grad_norm": 0.5461528301239014,
+      "learning_rate": 0.00013444639718804922,
+      "loss": 1.0150731801986694,
+      "step": 766
+    },
+    {
+      "epoch": 0.3240506329113924,
+      "grad_norm": 0.53483647108078,
+      "learning_rate": 0.00013479789103690686,
+      "loss": 0.8985214829444885,
+      "step": 768
+    },
+    {
+      "epoch": 0.32489451476793246,
+      "grad_norm": 0.5580531358718872,
+      "learning_rate": 0.0001351493848857645,
+      "loss": 1.0225775241851807,
+      "step": 770
+    },
+    {
+      "epoch": 0.32573839662447257,
+      "grad_norm": 0.5203377604484558,
+      "learning_rate": 0.00013550087873462216,
+      "loss": 0.9571293592453003,
+      "step": 772
+    },
+    {
+      "epoch": 0.3265822784810127,
+      "grad_norm": 0.5049671530723572,
+      "learning_rate": 0.0001358523725834798,
+      "loss": 1.0468909740447998,
+      "step": 774
+    },
+    {
+      "epoch": 0.32742616033755273,
+      "grad_norm": 0.4723063111305237,
+      "learning_rate": 0.00013620386643233744,
+      "loss": 0.7743215560913086,
+      "step": 776
+    },
+    {
+      "epoch": 0.32827004219409284,
+      "grad_norm": 0.6310980916023254,
+      "learning_rate": 0.00013655536028119508,
+      "loss": 1.021510362625122,
+      "step": 778
+    },
+    {
+      "epoch": 0.3291139240506329,
+      "grad_norm": 0.47066664695739746,
+      "learning_rate": 0.00013690685413005274,
+      "loss": 0.9134382605552673,
+      "step": 780
+    },
+    {
+      "epoch": 0.329957805907173,
+      "grad_norm": 0.5725092887878418,
+      "learning_rate": 0.00013725834797891038,
+      "loss": 0.9797834753990173,
+      "step": 782
+    },
+    {
+      "epoch": 0.3308016877637131,
+      "grad_norm": 0.5139563083648682,
+      "learning_rate": 0.00013760984182776802,
+      "loss": 0.9372621178627014,
+      "step": 784
+    },
+    {
+      "epoch": 0.33164556962025316,
+      "grad_norm": 0.5275821685791016,
+      "learning_rate": 0.00013796133567662566,
+      "loss": 0.9528245329856873,
+      "step": 786
+    },
+    {
+      "epoch": 0.33248945147679326,
+      "grad_norm": 0.5702582001686096,
+      "learning_rate": 0.0001383128295254833,
+      "loss": 0.9750176072120667,
+      "step": 788
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.5281293392181396,
+      "learning_rate": 0.00013866432337434096,
+      "loss": 0.9412306547164917,
+      "step": 790
+    },
+    {
+      "epoch": 0.3341772151898734,
+      "grad_norm": 0.5578986406326294,
+      "learning_rate": 0.0001390158172231986,
+      "loss": 0.997580885887146,
+      "step": 792
+    },
+    {
+      "epoch": 0.33502109704641353,
+      "grad_norm": 0.50461345911026,
+      "learning_rate": 0.00013936731107205624,
+      "loss": 0.9082320928573608,
+      "step": 794
+    },
+    {
+      "epoch": 0.3358649789029536,
+      "grad_norm": 0.5258530378341675,
+      "learning_rate": 0.0001397188049209139,
+      "loss": 1.0082844495773315,
+      "step": 796
+    },
+    {
+      "epoch": 0.3367088607594937,
+      "grad_norm": 0.5548169016838074,
+      "learning_rate": 0.00014007029876977154,
+      "loss": 0.9729003310203552,
+      "step": 798
+    },
+    {
+      "epoch": 0.33755274261603374,
+      "grad_norm": 0.48601076006889343,
+      "learning_rate": 0.00014042179261862918,
+      "loss": 0.9099526405334473,
+      "step": 800
+    },
+    {
+      "epoch": 0.33755274261603374,
+      "eval_loss": 0.9800403714179993,
+      "eval_runtime": 678.8306,
+      "eval_samples_per_second": 3.104,
+      "eval_steps_per_second": 3.104,
+      "step": 800
+    },
+    {
+      "epoch": 0.33839662447257385,
+      "grad_norm": 0.5413158535957336,
+      "learning_rate": 0.00014077328646748682,
+      "loss": 0.8610644936561584,
+      "step": 802
+    },
+    {
+      "epoch": 0.3392405063291139,
+      "grad_norm": 0.5147035717964172,
+      "learning_rate": 0.00014112478031634446,
+      "loss": 0.9584825038909912,
+      "step": 804
+    },
+    {
+      "epoch": 0.340084388185654,
+      "grad_norm": 0.5931771397590637,
+      "learning_rate": 0.0001414762741652021,
+      "loss": 1.0142558813095093,
+      "step": 806
+    },
+    {
+      "epoch": 0.3409282700421941,
+      "grad_norm": 0.5178377032279968,
+      "learning_rate": 0.00014182776801405976,
+      "loss": 1.0078763961791992,
+      "step": 808
+    },
+    {
+      "epoch": 0.34177215189873417,
+      "grad_norm": 0.5453237295150757,
+      "learning_rate": 0.0001421792618629174,
+      "loss": 0.9107215404510498,
+      "step": 810
+    },
+    {
+      "epoch": 0.3426160337552743,
+      "grad_norm": 0.5886152982711792,
+      "learning_rate": 0.00014253075571177506,
+      "loss": 0.9981362819671631,
+      "step": 812
+    },
+    {
+      "epoch": 0.3434599156118143,
+      "grad_norm": 0.48040178418159485,
+      "learning_rate": 0.0001428822495606327,
+      "loss": 0.9636131525039673,
+      "step": 814
+    },
+    {
+      "epoch": 0.34430379746835443,
+      "grad_norm": 0.5011753439903259,
+      "learning_rate": 0.00014323374340949034,
+      "loss": 0.9590586423873901,
+      "step": 816
+    },
+    {
+      "epoch": 0.34514767932489454,
+      "grad_norm": 0.57858806848526,
+      "learning_rate": 0.00014358523725834798,
+      "loss": 0.978246808052063,
+      "step": 818
+    },
+    {
+      "epoch": 0.3459915611814346,
+      "grad_norm": 0.46092939376831055,
+      "learning_rate": 0.00014393673110720562,
+      "loss": 0.8549934029579163,
+      "step": 820
+    },
+    {
+      "epoch": 0.3468354430379747,
+      "grad_norm": 0.5756489038467407,
+      "learning_rate": 0.00014428822495606326,
+      "loss": 0.9771265387535095,
+      "step": 822
+    },
+    {
+      "epoch": 0.34767932489451475,
+      "grad_norm": 0.5501731634140015,
+      "learning_rate": 0.00014463971880492092,
+      "loss": 0.8739748001098633,
+      "step": 824
+    },
+    {
+      "epoch": 0.34852320675105486,
+      "grad_norm": 0.5451868176460266,
+      "learning_rate": 0.0001449912126537786,
+      "loss": 0.9129468202590942,
+      "step": 826
+    },
+    {
+      "epoch": 0.3493670886075949,
+      "grad_norm": 0.4624619781970978,
+      "learning_rate": 0.00014534270650263623,
+      "loss": 0.9196704030036926,
+      "step": 828
+    },
+    {
+      "epoch": 0.350210970464135,
+      "grad_norm": 0.520878791809082,
+      "learning_rate": 0.00014569420035149386,
+      "loss": 0.9976527690887451,
+      "step": 830
+    },
+    {
+      "epoch": 0.3510548523206751,
+      "grad_norm": 0.4469171464443207,
+      "learning_rate": 0.0001460456942003515,
+      "loss": 0.7753443717956543,
+      "step": 832
+    },
+    {
+      "epoch": 0.3518987341772152,
+      "grad_norm": 0.5105249881744385,
+      "learning_rate": 0.00014639718804920914,
+      "loss": 0.9584846496582031,
+      "step": 834
+    },
+    {
+      "epoch": 0.3527426160337553,
+      "grad_norm": 0.5043913125991821,
+      "learning_rate": 0.00014674868189806678,
+      "loss": 0.9013500213623047,
+      "step": 836
+    },
+    {
+      "epoch": 0.35358649789029534,
+      "grad_norm": 0.575850784778595,
+      "learning_rate": 0.00014710017574692442,
+      "loss": 0.9775562286376953,
+      "step": 838
+    },
+    {
+      "epoch": 0.35443037974683544,
+      "grad_norm": 0.5128876566886902,
+      "learning_rate": 0.00014745166959578208,
+      "loss": 0.9278940558433533,
+      "step": 840
+    },
+    {
+      "epoch": 0.35527426160337555,
+      "grad_norm": 0.5757885575294495,
+      "learning_rate": 0.00014780316344463975,
+      "loss": 1.0091488361358643,
+      "step": 842
+    },
+    {
+      "epoch": 0.3561181434599156,
+      "grad_norm": 0.500934898853302,
+      "learning_rate": 0.00014815465729349739,
+      "loss": 0.9286836981773376,
+      "step": 844
+    },
+    {
+      "epoch": 0.3569620253164557,
+      "grad_norm": 0.5220686197280884,
+      "learning_rate": 0.00014850615114235502,
+      "loss": 0.9484171867370605,
+      "step": 846
+    },
+    {
+      "epoch": 0.35780590717299576,
+      "grad_norm": 0.5494697690010071,
+      "learning_rate": 0.00014885764499121266,
+      "loss": 1.0556397438049316,
+      "step": 848
+    },
+    {
+      "epoch": 0.35864978902953587,
+      "grad_norm": 0.46633943915367126,
+      "learning_rate": 0.0001492091388400703,
+      "loss": 0.933089017868042,
+      "step": 850
+    },
+    {
+      "epoch": 0.3594936708860759,
+      "grad_norm": 0.4864962697029114,
+      "learning_rate": 0.00014956063268892794,
+      "loss": 1.0016963481903076,
+      "step": 852
+    },
+    {
+      "epoch": 0.36033755274261603,
+      "grad_norm": 0.5032764673233032,
+      "learning_rate": 0.0001499121265377856,
+      "loss": 0.9085348844528198,
+      "step": 854
+    },
+    {
+      "epoch": 0.36118143459915614,
+      "grad_norm": 0.5424998998641968,
+      "learning_rate": 0.00015026362038664324,
+      "loss": 0.923502504825592,
+      "step": 856
+    },
+    {
+      "epoch": 0.3620253164556962,
+      "grad_norm": 0.5181655287742615,
+      "learning_rate": 0.00015061511423550088,
+      "loss": 0.919174313545227,
+      "step": 858
+    },
+    {
+      "epoch": 0.3628691983122363,
+      "grad_norm": 0.5129443407058716,
+      "learning_rate": 0.00015096660808435855,
+      "loss": 0.851981520652771,
+      "step": 860
+    },
+    {
+      "epoch": 0.36371308016877635,
+      "grad_norm": 0.49540698528289795,
+      "learning_rate": 0.00015131810193321618,
+      "loss": 0.8633858561515808,
+      "step": 862
+    },
+    {
+      "epoch": 0.36455696202531646,
+      "grad_norm": 0.4706701934337616,
+      "learning_rate": 0.00015166959578207382,
+      "loss": 0.9473677277565002,
+      "step": 864
+    },
+    {
+      "epoch": 0.36540084388185656,
+      "grad_norm": 0.587704598903656,
+      "learning_rate": 0.00015202108963093146,
+      "loss": 0.942383885383606,
+      "step": 866
+    },
+    {
+      "epoch": 0.3662447257383966,
+      "grad_norm": 0.5851273536682129,
+      "learning_rate": 0.0001523725834797891,
+      "loss": 0.9811251163482666,
+      "step": 868
+    },
+    {
+      "epoch": 0.3670886075949367,
+      "grad_norm": 0.46357613801956177,
+      "learning_rate": 0.00015272407732864676,
+      "loss": 0.8151084184646606,
+      "step": 870
+    },
+    {
+      "epoch": 0.3679324894514768,
+      "grad_norm": 0.47250670194625854,
+      "learning_rate": 0.0001530755711775044,
+      "loss": 0.9023333191871643,
+      "step": 872
+    },
+    {
+      "epoch": 0.3687763713080169,
+      "grad_norm": 0.47510042786598206,
+      "learning_rate": 0.00015342706502636204,
+      "loss": 0.8758499622344971,
+      "step": 874
+    },
+    {
+      "epoch": 0.369620253164557,
+      "grad_norm": 0.5687124729156494,
+      "learning_rate": 0.00015377855887521968,
+      "loss": 0.9837421774864197,
+      "step": 876
+    },
+    {
+      "epoch": 0.37046413502109704,
+      "grad_norm": 0.49064236879348755,
+      "learning_rate": 0.00015413005272407735,
+      "loss": 0.9084216356277466,
+      "step": 878
+    },
+    {
+      "epoch": 0.37130801687763715,
+      "grad_norm": 0.5288164615631104,
+      "learning_rate": 0.00015448154657293498,
+      "loss": 0.8425542712211609,
+      "step": 880
+    },
+    {
+      "epoch": 0.3721518987341772,
+      "grad_norm": 0.5446951985359192,
+      "learning_rate": 0.00015483304042179262,
+      "loss": 1.006197452545166,
+      "step": 882
+    },
+    {
+      "epoch": 0.3729957805907173,
+      "grad_norm": 0.46872642636299133,
+      "learning_rate": 0.00015518453427065026,
+      "loss": 0.9779444932937622,
+      "step": 884
+    },
+    {
+      "epoch": 0.37383966244725736,
+      "grad_norm": 0.4852714240550995,
+      "learning_rate": 0.00015553602811950793,
+      "loss": 0.9099963307380676,
+      "step": 886
+    },
+    {
+      "epoch": 0.37468354430379747,
+      "grad_norm": 0.5219841003417969,
+      "learning_rate": 0.00015588752196836556,
+      "loss": 0.9730570316314697,
+      "step": 888
+    },
+    {
+      "epoch": 0.3755274261603376,
+      "grad_norm": 0.5258626341819763,
+      "learning_rate": 0.0001562390158172232,
+      "loss": 0.9304586052894592,
+      "step": 890
+    },
+    {
+      "epoch": 0.3763713080168776,
+      "grad_norm": 0.4266716241836548,
+      "learning_rate": 0.00015659050966608084,
+      "loss": 0.8680716156959534,
+      "step": 892
+    },
+    {
+      "epoch": 0.37721518987341773,
+      "grad_norm": 0.46361327171325684,
+      "learning_rate": 0.0001569420035149385,
+      "loss": 0.8428018093109131,
+      "step": 894
+    },
+    {
+      "epoch": 0.3780590717299578,
+      "grad_norm": 0.5313687920570374,
+      "learning_rate": 0.00015729349736379614,
+      "loss": 0.8465750217437744,
+      "step": 896
+    },
+    {
+      "epoch": 0.3789029535864979,
+      "grad_norm": 0.47166210412979126,
+      "learning_rate": 0.00015764499121265378,
+      "loss": 0.9120327234268188,
+      "step": 898
+    },
+    {
+      "epoch": 0.379746835443038,
+      "grad_norm": 0.4647318720817566,
+      "learning_rate": 0.00015799648506151145,
+      "loss": 0.8950425982475281,
+      "step": 900
+    },
+    {
+      "epoch": 0.379746835443038,
+      "eval_loss": 0.9643027186393738,
+      "eval_runtime": 691.7929,
+      "eval_samples_per_second": 3.046,
+      "eval_steps_per_second": 3.046,
+      "step": 900
+    },
+    {
+      "epoch": 0.38059071729957805,
+      "grad_norm": 0.5445119738578796,
+      "learning_rate": 0.00015834797891036909,
+      "loss": 0.908163845539093,
+      "step": 902
+    },
+    {
+      "epoch": 0.38143459915611816,
+      "grad_norm": 0.4311858117580414,
+      "learning_rate": 0.00015869947275922672,
+      "loss": 0.8945821523666382,
+      "step": 904
+    },
+    {
+      "epoch": 0.3822784810126582,
+      "grad_norm": 0.5590984225273132,
+      "learning_rate": 0.00015905096660808436,
+      "loss": 0.9478458762168884,
+      "step": 906
+    },
+    {
+      "epoch": 0.3831223628691983,
+      "grad_norm": 0.5470241904258728,
+      "learning_rate": 0.000159402460456942,
+      "loss": 0.9259957671165466,
+      "step": 908
+    },
+    {
+      "epoch": 0.38396624472573837,
+      "grad_norm": 0.5498791337013245,
+      "learning_rate": 0.00015975395430579964,
+      "loss": 0.8824930787086487,
+      "step": 910
+    },
+    {
+      "epoch": 0.3848101265822785,
+      "grad_norm": 0.4779198467731476,
+      "learning_rate": 0.0001601054481546573,
+      "loss": 0.8842340707778931,
+      "step": 912
+    },
+    {
+      "epoch": 0.3856540084388186,
+      "grad_norm": 0.5390620827674866,
+      "learning_rate": 0.00016045694200351494,
+      "loss": 0.92950040102005,
+      "step": 914
+    },
+    {
+      "epoch": 0.38649789029535864,
+      "grad_norm": 0.505519688129425,
+      "learning_rate": 0.0001608084358523726,
+      "loss": 0.8420897126197815,
+      "step": 916
+    },
+    {
+      "epoch": 0.38734177215189874,
+      "grad_norm": 0.4463907778263092,
+      "learning_rate": 0.00016115992970123025,
+      "loss": 0.8256624341011047,
+      "step": 918
+    },
+    {
+      "epoch": 0.3881856540084388,
+      "grad_norm": 0.5321422219276428,
+      "learning_rate": 0.00016151142355008788,
+      "loss": 0.8701168298721313,
+      "step": 920
+    },
+    {
+      "epoch": 0.3890295358649789,
+      "grad_norm": 0.4343073070049286,
+      "learning_rate": 0.00016186291739894552,
+      "loss": 0.9019309878349304,
+      "step": 922
+    },
+    {
+      "epoch": 0.389873417721519,
+      "grad_norm": 0.5311984419822693,
+      "learning_rate": 0.00016221441124780316,
+      "loss": 0.8560551404953003,
+      "step": 924
+    },
+    {
+      "epoch": 0.39071729957805906,
+      "grad_norm": 0.588691771030426,
+      "learning_rate": 0.0001625659050966608,
+      "loss": 0.9700050354003906,
+      "step": 926
+    },
+    {
+      "epoch": 0.39156118143459917,
+      "grad_norm": 0.5425586104393005,
+      "learning_rate": 0.00016291739894551844,
+      "loss": 1.043768048286438,
+      "step": 928
+    },
+    {
+      "epoch": 0.3924050632911392,
+      "grad_norm": 0.5228736996650696,
+      "learning_rate": 0.0001632688927943761,
+      "loss": 0.9501712918281555,
+      "step": 930
+    },
+    {
+      "epoch": 0.39324894514767933,
+      "grad_norm": 0.48960360884666443,
+      "learning_rate": 0.00016362038664323377,
+      "loss": 0.9223058223724365,
+      "step": 932
+    },
+    {
+      "epoch": 0.39409282700421944,
+      "grad_norm": 0.45204755663871765,
+      "learning_rate": 0.0001639718804920914,
+      "loss": 0.9692960977554321,
+      "step": 934
+    },
+    {
+      "epoch": 0.3949367088607595,
+      "grad_norm": 0.5299274921417236,
+      "learning_rate": 0.00016432337434094905,
+      "loss": 0.9467466473579407,
+      "step": 936
+    },
+    {
+      "epoch": 0.3957805907172996,
+      "grad_norm": 0.5607715249061584,
+      "learning_rate": 0.00016467486818980668,
+      "loss": 0.9118053317070007,
+      "step": 938
+    },
+    {
+      "epoch": 0.39662447257383965,
+      "grad_norm": 0.5271831154823303,
+      "learning_rate": 0.00016502636203866432,
+      "loss": 0.9131460189819336,
+      "step": 940
+    },
+    {
+      "epoch": 0.39746835443037976,
+      "grad_norm": 0.5075286030769348,
+      "learning_rate": 0.00016537785588752196,
+      "loss": 0.9358300566673279,
+      "step": 942
+    },
+    {
+      "epoch": 0.3983122362869198,
+      "grad_norm": 0.515731155872345,
+      "learning_rate": 0.00016572934973637963,
+      "loss": 0.8908210396766663,
+      "step": 944
+    },
+    {
+      "epoch": 0.3991561181434599,
+      "grad_norm": 0.4856977164745331,
+      "learning_rate": 0.00016608084358523726,
+      "loss": 0.9775290489196777,
+      "step": 946
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.48846355080604553,
+      "learning_rate": 0.00016643233743409493,
+      "loss": 0.8957490921020508,
+      "step": 948
+    },
+    {
+      "epoch": 0.4008438818565401,
+      "grad_norm": 0.42990800738334656,
+      "learning_rate": 0.00016678383128295257,
+      "loss": 0.9036174416542053,
+      "step": 950
+    },
+    {
+      "epoch": 0.4016877637130802,
+      "grad_norm": 0.49552062153816223,
+      "learning_rate": 0.0001671353251318102,
+      "loss": 0.991032600402832,
+      "step": 952
+    },
+    {
+      "epoch": 0.40253164556962023,
+      "grad_norm": 0.4565040171146393,
+      "learning_rate": 0.00016748681898066784,
+      "loss": 0.823063313961029,
+      "step": 954
+    },
+    {
+      "epoch": 0.40337552742616034,
+      "grad_norm": 0.4290153682231903,
+      "learning_rate": 0.00016783831282952548,
+      "loss": 0.8785063624382019,
+      "step": 956
+    },
+    {
+      "epoch": 0.40421940928270045,
+      "grad_norm": 0.5419702529907227,
+      "learning_rate": 0.00016818980667838312,
+      "loss": 0.8763971924781799,
+      "step": 958
+    },
+    {
+      "epoch": 0.4050632911392405,
+      "grad_norm": 0.5177501440048218,
+      "learning_rate": 0.00016854130052724079,
+      "loss": 0.9470553398132324,
+      "step": 960
+    },
+    {
+      "epoch": 0.4059071729957806,
+      "grad_norm": 0.539725661277771,
+      "learning_rate": 0.00016889279437609842,
+      "loss": 0.9235025644302368,
+      "step": 962
+    },
+    {
+      "epoch": 0.40675105485232066,
+      "grad_norm": 0.5324983596801758,
+      "learning_rate": 0.0001692442882249561,
+      "loss": 1.0248996019363403,
+      "step": 964
+    },
+    {
+      "epoch": 0.40759493670886077,
+      "grad_norm": 0.4936407506465912,
+      "learning_rate": 0.00016959578207381373,
+      "loss": 0.9076873660087585,
+      "step": 966
+    },
+    {
+      "epoch": 0.4084388185654008,
+      "grad_norm": 0.4960501194000244,
+      "learning_rate": 0.00016994727592267137,
+      "loss": 0.9162673950195312,
+      "step": 968
+    },
+    {
+      "epoch": 0.4092827004219409,
+      "grad_norm": 0.45093682408332825,
+      "learning_rate": 0.000170298769771529,
+      "loss": 0.904100775718689,
+      "step": 970
+    },
+    {
+      "epoch": 0.41012658227848103,
+      "grad_norm": 0.4560275077819824,
+      "learning_rate": 0.00017065026362038664,
+      "loss": 0.8633337020874023,
+      "step": 972
+    },
+    {
+      "epoch": 0.4109704641350211,
+      "grad_norm": 0.44885000586509705,
+      "learning_rate": 0.00017100175746924428,
+      "loss": 0.8454209566116333,
+      "step": 974
+    },
+    {
+      "epoch": 0.4118143459915612,
+      "grad_norm": 0.47251659631729126,
+      "learning_rate": 0.00017135325131810195,
+      "loss": 0.824730396270752,
+      "step": 976
+    },
+    {
+      "epoch": 0.41265822784810124,
+      "grad_norm": 0.6597666144371033,
+      "learning_rate": 0.00017170474516695959,
+      "loss": 0.9496501684188843,
+      "step": 978
+    },
+    {
+      "epoch": 0.41350210970464135,
+      "grad_norm": 0.39806297421455383,
+      "learning_rate": 0.00017205623901581722,
+      "loss": 0.9419087171554565,
+      "step": 980
+    },
+    {
+      "epoch": 0.41434599156118146,
+      "grad_norm": 0.48231109976768494,
+      "learning_rate": 0.0001724077328646749,
+      "loss": 0.9182976484298706,
+      "step": 982
+    },
+    {
+      "epoch": 0.4151898734177215,
+      "grad_norm": 0.5438776612281799,
+      "learning_rate": 0.00017275922671353253,
+      "loss": 0.9386967420578003,
+      "step": 984
+    },
+    {
+      "epoch": 0.4160337552742616,
+      "grad_norm": 0.4959667921066284,
+      "learning_rate": 0.00017311072056239017,
+      "loss": 0.897849440574646,
+      "step": 986
+    },
+    {
+      "epoch": 0.41687763713080167,
+      "grad_norm": 0.43533357977867126,
+      "learning_rate": 0.0001734622144112478,
+      "loss": 0.8776953816413879,
+      "step": 988
+    },
+    {
+      "epoch": 0.4177215189873418,
+      "grad_norm": 0.47513946890830994,
+      "learning_rate": 0.00017381370826010547,
+      "loss": 0.9162989854812622,
+      "step": 990
+    },
+    {
+      "epoch": 0.41856540084388183,
+      "grad_norm": 0.4907188415527344,
+      "learning_rate": 0.0001741652021089631,
+      "loss": 0.9482660889625549,
+      "step": 992
+    },
+    {
+      "epoch": 0.41940928270042194,
+      "grad_norm": 0.44499966502189636,
+      "learning_rate": 0.00017451669595782075,
+      "loss": 0.8812930583953857,
+      "step": 994
+    },
+    {
+      "epoch": 0.42025316455696204,
+      "grad_norm": 0.4535730481147766,
+      "learning_rate": 0.00017486818980667838,
+      "loss": 0.9439874887466431,
+      "step": 996
+    },
+    {
+      "epoch": 0.4210970464135021,
+      "grad_norm": 0.5240745544433594,
+      "learning_rate": 0.00017521968365553602,
+      "loss": 0.8818395137786865,
+      "step": 998
+    },
+    {
+      "epoch": 0.4219409282700422,
+      "grad_norm": 0.5301211476325989,
+      "learning_rate": 0.0001755711775043937,
+      "loss": 0.886186957359314,
+      "step": 1000
+    },
+    {
+      "epoch": 0.4219409282700422,
+      "eval_loss": 0.9487298727035522,
+      "eval_runtime": 689.4288,
+      "eval_samples_per_second": 3.056,
+      "eval_steps_per_second": 3.056,
+      "step": 1000
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 14220,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 6,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 5,
+        "early_stopping_threshold": 0.001
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 0
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.979725639330796e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-1000/training_args.bin b/sft_qwen_14B_v2/checkpoints/checkpoint-1000/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..0bd5f45757bdc6c7fc2d67718b18675b51983e0e
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-1000/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:afae4a8ce1391f149eb00b14eed8f891c715e892ea546bf754d22db2c2bc5969
+size 4792
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-1500/README.md b/sft_qwen_14B_v2/checkpoints/checkpoint-1500/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..134a3a8531b68adda3fed2a98db824f7c1b7dcfd
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-1500/README.md
@@ -0,0 +1,207 @@
+---
+base_model: Models/Qwen2.5-Coder-14B-CPT
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Models/Qwen2.5-Coder-14B-CPT
+- lora
+- transformers
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.0
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-1500/adapter_config.json b/sft_qwen_14B_v2/checkpoints/checkpoint-1500/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..d6316742bc71d30704e2ccdb9080ed0b8062e454
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-1500/adapter_config.json
@@ -0,0 +1,43 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Models/Qwen2.5-Coder-14B-CPT",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.0",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "v_proj",
+    "k_proj",
+    "q_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-1500/adapter_model.safetensors b/sft_qwen_14B_v2/checkpoints/checkpoint-1500/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..8c653b06301944b0d82ab84d629bd69b43c63873
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-1500/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72b8f86cccb14317675ea126399dc4e8a7f18c71b7076f438c3d7e9c62fda470
+size 201378736
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-1500/optimizer.pt b/sft_qwen_14B_v2/checkpoints/checkpoint-1500/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..65eba1c4e34f589f02297bbae9e1f85c1cadb0e4
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-1500/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:90432c235116acced0b5039484a54c5317932455079b5db44009c97433b3a03b
+size 402976786
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-1500/rng_state.pth b/sft_qwen_14B_v2/checkpoints/checkpoint-1500/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..18e94c37ee1aee33e36f09b2af0a47f8e41e73a0
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-1500/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a22e5aeea526845f66de80330bef50d5eab9ae3204b490a24a0489bbbf8f66f2
+size 14244
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-1500/scheduler.pt b/sft_qwen_14B_v2/checkpoints/checkpoint-1500/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..27a29dde19d6c86196c4274b73439b47a37417a0
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-1500/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c6d90f58358a00be3824aa6b916df4e44be43484ede05aad7a85da63345c506
+size 1064
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-1500/trainer_state.json b/sft_qwen_14B_v2/checkpoints/checkpoint-1500/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..c2bb20ee604382a25ebb93a4973441ae393f807b
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-1500/trainer_state.json
@@ -0,0 +1,5413 @@
+{
+  "best_global_step": 1500,
+  "best_metric": 0.887488842010498,
+  "best_model_checkpoint": "task2file/sft_qwen_14B_v2/checkpoints/checkpoint-1500",
+  "epoch": 0.6329113924050633,
+  "eval_steps": 100,
+  "global_step": 1500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0008438818565400844,
+      "grad_norm": 0.5386583805084229,
+      "learning_rate": 1.7574692442882248e-07,
+      "loss": 1.6941628456115723,
+      "step": 2
+    },
+    {
+      "epoch": 0.0016877637130801688,
+      "grad_norm": 0.5477277636528015,
+      "learning_rate": 5.272407732864675e-07,
+      "loss": 1.7132279872894287,
+      "step": 4
+    },
+    {
+      "epoch": 0.002531645569620253,
+      "grad_norm": 0.5390765070915222,
+      "learning_rate": 8.787346221441126e-07,
+      "loss": 1.641180396080017,
+      "step": 6
+    },
+    {
+      "epoch": 0.0033755274261603376,
+      "grad_norm": 0.5023683905601501,
+      "learning_rate": 1.2302284710017575e-06,
+      "loss": 1.5616240501403809,
+      "step": 8
+    },
+    {
+      "epoch": 0.004219409282700422,
+      "grad_norm": 0.4899154603481293,
+      "learning_rate": 1.5817223198594026e-06,
+      "loss": 1.572033405303955,
+      "step": 10
+    },
+    {
+      "epoch": 0.005063291139240506,
+      "grad_norm": 0.5239788293838501,
+      "learning_rate": 1.9332161687170474e-06,
+      "loss": 1.6242921352386475,
+      "step": 12
+    },
+    {
+      "epoch": 0.00590717299578059,
+      "grad_norm": 0.5172926783561707,
+      "learning_rate": 2.2847100175746925e-06,
+      "loss": 1.6800041198730469,
+      "step": 14
+    },
+    {
+      "epoch": 0.006751054852320675,
+      "grad_norm": 0.5539224743843079,
+      "learning_rate": 2.6362038664323376e-06,
+      "loss": 1.6450834274291992,
+      "step": 16
+    },
+    {
+      "epoch": 0.007594936708860759,
+      "grad_norm": 0.5255337953567505,
+      "learning_rate": 2.9876977152899827e-06,
+      "loss": 1.6673263311386108,
+      "step": 18
+    },
+    {
+      "epoch": 0.008438818565400843,
+      "grad_norm": 0.5074548721313477,
+      "learning_rate": 3.3391915641476277e-06,
+      "loss": 1.531802773475647,
+      "step": 20
+    },
+    {
+      "epoch": 0.009282700421940928,
+      "grad_norm": 0.4160279333591461,
+      "learning_rate": 3.6906854130052724e-06,
+      "loss": 1.599354863166809,
+      "step": 22
+    },
+    {
+      "epoch": 0.010126582278481013,
+      "grad_norm": 0.5716474652290344,
+      "learning_rate": 4.0421792618629174e-06,
+      "loss": 1.6700962781906128,
+      "step": 24
+    },
+    {
+      "epoch": 0.010970464135021098,
+      "grad_norm": 0.5148899555206299,
+      "learning_rate": 4.3936731107205625e-06,
+      "loss": 1.66217839717865,
+      "step": 26
+    },
+    {
+      "epoch": 0.01181434599156118,
+      "grad_norm": 0.575722336769104,
+      "learning_rate": 4.7451669595782076e-06,
+      "loss": 1.6692266464233398,
+      "step": 28
+    },
+    {
+      "epoch": 0.012658227848101266,
+      "grad_norm": 0.5345953106880188,
+      "learning_rate": 5.096660808435853e-06,
+      "loss": 1.5518689155578613,
+      "step": 30
+    },
+    {
+      "epoch": 0.01350210970464135,
+      "grad_norm": 0.4462043344974518,
+      "learning_rate": 5.448154657293498e-06,
+      "loss": 1.5930007696151733,
+      "step": 32
+    },
+    {
+      "epoch": 0.014345991561181435,
+      "grad_norm": 0.5119605660438538,
+      "learning_rate": 5.799648506151143e-06,
+      "loss": 1.6069684028625488,
+      "step": 34
+    },
+    {
+      "epoch": 0.015189873417721518,
+      "grad_norm": 0.5328608751296997,
+      "learning_rate": 6.151142355008788e-06,
+      "loss": 1.5838109254837036,
+      "step": 36
+    },
+    {
+      "epoch": 0.016033755274261603,
+      "grad_norm": 0.5065920352935791,
+      "learning_rate": 6.502636203866433e-06,
+      "loss": 1.608130931854248,
+      "step": 38
+    },
+    {
+      "epoch": 0.016877637130801686,
+      "grad_norm": 0.4479359984397888,
+      "learning_rate": 6.854130052724078e-06,
+      "loss": 1.5942182540893555,
+      "step": 40
+    },
+    {
+      "epoch": 0.017721518987341773,
+      "grad_norm": 0.42844903469085693,
+      "learning_rate": 7.205623901581722e-06,
+      "loss": 1.6441553831100464,
+      "step": 42
+    },
+    {
+      "epoch": 0.018565400843881856,
+      "grad_norm": 0.476630836725235,
+      "learning_rate": 7.557117750439367e-06,
+      "loss": 1.6068111658096313,
+      "step": 44
+    },
+    {
+      "epoch": 0.019409282700421943,
+      "grad_norm": 0.4532654881477356,
+      "learning_rate": 7.908611599297012e-06,
+      "loss": 1.6618021726608276,
+      "step": 46
+    },
+    {
+      "epoch": 0.020253164556962026,
+      "grad_norm": 0.3701118230819702,
+      "learning_rate": 8.260105448154657e-06,
+      "loss": 1.4730033874511719,
+      "step": 48
+    },
+    {
+      "epoch": 0.02109704641350211,
+      "grad_norm": 0.38471561670303345,
+      "learning_rate": 8.611599297012302e-06,
+      "loss": 1.4828267097473145,
+      "step": 50
+    },
+    {
+      "epoch": 0.021940928270042195,
+      "grad_norm": 0.3602336347103119,
+      "learning_rate": 8.963093145869948e-06,
+      "loss": 1.3877452611923218,
+      "step": 52
+    },
+    {
+      "epoch": 0.02278481012658228,
+      "grad_norm": 0.40318572521209717,
+      "learning_rate": 9.314586994727593e-06,
+      "loss": 1.49052894115448,
+      "step": 54
+    },
+    {
+      "epoch": 0.02362869198312236,
+      "grad_norm": 0.3223826587200165,
+      "learning_rate": 9.666080843585238e-06,
+      "loss": 1.4912524223327637,
+      "step": 56
+    },
+    {
+      "epoch": 0.024472573839662448,
+      "grad_norm": 0.3873065114021301,
+      "learning_rate": 1.0017574692442883e-05,
+      "loss": 1.526674509048462,
+      "step": 58
+    },
+    {
+      "epoch": 0.02531645569620253,
+      "grad_norm": 0.410159707069397,
+      "learning_rate": 1.0369068541300528e-05,
+      "loss": 1.4480271339416504,
+      "step": 60
+    },
+    {
+      "epoch": 0.026160337552742614,
+      "grad_norm": 0.3632003962993622,
+      "learning_rate": 1.0720562390158173e-05,
+      "loss": 1.4222990274429321,
+      "step": 62
+    },
+    {
+      "epoch": 0.0270042194092827,
+      "grad_norm": 0.33118435740470886,
+      "learning_rate": 1.1072056239015818e-05,
+      "loss": 1.387171745300293,
+      "step": 64
+    },
+    {
+      "epoch": 0.027848101265822784,
+      "grad_norm": 0.3301764726638794,
+      "learning_rate": 1.1423550087873463e-05,
+      "loss": 1.3523777723312378,
+      "step": 66
+    },
+    {
+      "epoch": 0.02869198312236287,
+      "grad_norm": 0.34342435002326965,
+      "learning_rate": 1.1775043936731108e-05,
+      "loss": 1.4515162706375122,
+      "step": 68
+    },
+    {
+      "epoch": 0.029535864978902954,
+      "grad_norm": 0.3243122100830078,
+      "learning_rate": 1.2126537785588753e-05,
+      "loss": 1.3509243726730347,
+      "step": 70
+    },
+    {
+      "epoch": 0.030379746835443037,
+      "grad_norm": 0.3450150787830353,
+      "learning_rate": 1.2478031634446398e-05,
+      "loss": 1.4936245679855347,
+      "step": 72
+    },
+    {
+      "epoch": 0.031223628691983123,
+      "grad_norm": 0.38912028074264526,
+      "learning_rate": 1.2829525483304042e-05,
+      "loss": 1.3419109582901,
+      "step": 74
+    },
+    {
+      "epoch": 0.032067510548523206,
+      "grad_norm": 0.3019310235977173,
+      "learning_rate": 1.3181019332161687e-05,
+      "loss": 1.4284154176712036,
+      "step": 76
+    },
+    {
+      "epoch": 0.03291139240506329,
+      "grad_norm": 0.37803682684898376,
+      "learning_rate": 1.3532513181019332e-05,
+      "loss": 1.4256561994552612,
+      "step": 78
+    },
+    {
+      "epoch": 0.03375527426160337,
+      "grad_norm": 0.34191736578941345,
+      "learning_rate": 1.3884007029876977e-05,
+      "loss": 1.3256909847259521,
+      "step": 80
+    },
+    {
+      "epoch": 0.03459915611814346,
+      "grad_norm": 0.35242700576782227,
+      "learning_rate": 1.4235500878734624e-05,
+      "loss": 1.2710685729980469,
+      "step": 82
+    },
+    {
+      "epoch": 0.035443037974683546,
+      "grad_norm": 0.38094228506088257,
+      "learning_rate": 1.4586994727592269e-05,
+      "loss": 1.253411889076233,
+      "step": 84
+    },
+    {
+      "epoch": 0.036286919831223625,
+      "grad_norm": 0.36837366223335266,
+      "learning_rate": 1.4938488576449914e-05,
+      "loss": 1.3064342737197876,
+      "step": 86
+    },
+    {
+      "epoch": 0.03713080168776371,
+      "grad_norm": 0.3443569242954254,
+      "learning_rate": 1.5289982425307557e-05,
+      "loss": 1.293562412261963,
+      "step": 88
+    },
+    {
+      "epoch": 0.0379746835443038,
+      "grad_norm": 0.3799338936805725,
+      "learning_rate": 1.5641476274165202e-05,
+      "loss": 1.3382648229599,
+      "step": 90
+    },
+    {
+      "epoch": 0.038818565400843885,
+      "grad_norm": 0.40501922369003296,
+      "learning_rate": 1.599297012302285e-05,
+      "loss": 1.3925724029541016,
+      "step": 92
+    },
+    {
+      "epoch": 0.039662447257383965,
+      "grad_norm": 0.4419630467891693,
+      "learning_rate": 1.6344463971880492e-05,
+      "loss": 1.357171893119812,
+      "step": 94
+    },
+    {
+      "epoch": 0.04050632911392405,
+      "grad_norm": 0.3619817793369293,
+      "learning_rate": 1.6695957820738137e-05,
+      "loss": 1.3029985427856445,
+      "step": 96
+    },
+    {
+      "epoch": 0.04135021097046414,
+      "grad_norm": 0.4851357340812683,
+      "learning_rate": 1.7047451669595782e-05,
+      "loss": 1.3498191833496094,
+      "step": 98
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "grad_norm": 0.418658584356308,
+      "learning_rate": 1.7398945518453427e-05,
+      "loss": 1.185287356376648,
+      "step": 100
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "eval_loss": 1.2979938983917236,
+      "eval_runtime": 682.1979,
+      "eval_samples_per_second": 3.089,
+      "eval_steps_per_second": 3.089,
+      "step": 100
+    },
+    {
+      "epoch": 0.043037974683544304,
+      "grad_norm": 0.4464418888092041,
+      "learning_rate": 1.7750439367311073e-05,
+      "loss": 1.2217272520065308,
+      "step": 102
+    },
+    {
+      "epoch": 0.04388185654008439,
+      "grad_norm": 0.4706237316131592,
+      "learning_rate": 1.8101933216168718e-05,
+      "loss": 1.2052050828933716,
+      "step": 104
+    },
+    {
+      "epoch": 0.04472573839662447,
+      "grad_norm": 0.46394404768943787,
+      "learning_rate": 1.8453427065026363e-05,
+      "loss": 1.221343994140625,
+      "step": 106
+    },
+    {
+      "epoch": 0.04556962025316456,
+      "grad_norm": 0.4726889431476593,
+      "learning_rate": 1.8804920913884008e-05,
+      "loss": 1.2387475967407227,
+      "step": 108
+    },
+    {
+      "epoch": 0.046413502109704644,
+      "grad_norm": 0.42130985856056213,
+      "learning_rate": 1.9156414762741653e-05,
+      "loss": 1.2851309776306152,
+      "step": 110
+    },
+    {
+      "epoch": 0.04725738396624472,
+      "grad_norm": 0.4504576623439789,
+      "learning_rate": 1.9507908611599298e-05,
+      "loss": 1.2753145694732666,
+      "step": 112
+    },
+    {
+      "epoch": 0.04810126582278481,
+      "grad_norm": 0.396085262298584,
+      "learning_rate": 1.9859402460456943e-05,
+      "loss": 1.2427717447280884,
+      "step": 114
+    },
+    {
+      "epoch": 0.048945147679324896,
+      "grad_norm": 0.5106491446495056,
+      "learning_rate": 2.0210896309314588e-05,
+      "loss": 1.2943825721740723,
+      "step": 116
+    },
+    {
+      "epoch": 0.049789029535864976,
+      "grad_norm": 0.42351317405700684,
+      "learning_rate": 2.0562390158172233e-05,
+      "loss": 1.263301134109497,
+      "step": 118
+    },
+    {
+      "epoch": 0.05063291139240506,
+      "grad_norm": 0.4403539299964905,
+      "learning_rate": 2.0913884007029878e-05,
+      "loss": 1.2647849321365356,
+      "step": 120
+    },
+    {
+      "epoch": 0.05147679324894515,
+      "grad_norm": 0.5260752439498901,
+      "learning_rate": 2.1265377855887523e-05,
+      "loss": 1.2351393699645996,
+      "step": 122
+    },
+    {
+      "epoch": 0.05232067510548523,
+      "grad_norm": 0.44978851079940796,
+      "learning_rate": 2.1616871704745168e-05,
+      "loss": 1.0384471416473389,
+      "step": 124
+    },
+    {
+      "epoch": 0.053164556962025315,
+      "grad_norm": 0.47732362151145935,
+      "learning_rate": 2.1968365553602813e-05,
+      "loss": 1.1518068313598633,
+      "step": 126
+    },
+    {
+      "epoch": 0.0540084388185654,
+      "grad_norm": 0.5473551750183105,
+      "learning_rate": 2.231985940246046e-05,
+      "loss": 1.2264912128448486,
+      "step": 128
+    },
+    {
+      "epoch": 0.05485232067510549,
+      "grad_norm": 0.4473855197429657,
+      "learning_rate": 2.2671353251318103e-05,
+      "loss": 1.1615246534347534,
+      "step": 130
+    },
+    {
+      "epoch": 0.05569620253164557,
+      "grad_norm": 0.5980377197265625,
+      "learning_rate": 2.302284710017575e-05,
+      "loss": 1.1334880590438843,
+      "step": 132
+    },
+    {
+      "epoch": 0.056540084388185655,
+      "grad_norm": 0.5987792015075684,
+      "learning_rate": 2.3374340949033394e-05,
+      "loss": 1.1546804904937744,
+      "step": 134
+    },
+    {
+      "epoch": 0.05738396624472574,
+      "grad_norm": 0.45355498790740967,
+      "learning_rate": 2.372583479789104e-05,
+      "loss": 1.194953441619873,
+      "step": 136
+    },
+    {
+      "epoch": 0.05822784810126582,
+      "grad_norm": 0.5373698472976685,
+      "learning_rate": 2.4077328646748684e-05,
+      "loss": 1.1067466735839844,
+      "step": 138
+    },
+    {
+      "epoch": 0.05907172995780591,
+      "grad_norm": 0.48734328150749207,
+      "learning_rate": 2.442882249560633e-05,
+      "loss": 1.188468098640442,
+      "step": 140
+    },
+    {
+      "epoch": 0.059915611814345994,
+      "grad_norm": 0.4692173898220062,
+      "learning_rate": 2.478031634446397e-05,
+      "loss": 1.1624362468719482,
+      "step": 142
+    },
+    {
+      "epoch": 0.060759493670886074,
+      "grad_norm": 0.532554030418396,
+      "learning_rate": 2.513181019332162e-05,
+      "loss": 1.0978907346725464,
+      "step": 144
+    },
+    {
+      "epoch": 0.06160337552742616,
+      "grad_norm": 0.5853802561759949,
+      "learning_rate": 2.5483304042179264e-05,
+      "loss": 1.2030781507492065,
+      "step": 146
+    },
+    {
+      "epoch": 0.06244725738396625,
+      "grad_norm": 0.5061611533164978,
+      "learning_rate": 2.583479789103691e-05,
+      "loss": 1.082366943359375,
+      "step": 148
+    },
+    {
+      "epoch": 0.06329113924050633,
+      "grad_norm": 0.49426141381263733,
+      "learning_rate": 2.6186291739894554e-05,
+      "loss": 1.10564386844635,
+      "step": 150
+    },
+    {
+      "epoch": 0.06413502109704641,
+      "grad_norm": 0.5846618413925171,
+      "learning_rate": 2.6537785588752196e-05,
+      "loss": 1.1992807388305664,
+      "step": 152
+    },
+    {
+      "epoch": 0.06497890295358649,
+      "grad_norm": 0.5517552495002747,
+      "learning_rate": 2.6889279437609844e-05,
+      "loss": 1.1757566928863525,
+      "step": 154
+    },
+    {
+      "epoch": 0.06582278481012659,
+      "grad_norm": 0.5667305588722229,
+      "learning_rate": 2.7240773286467486e-05,
+      "loss": 1.0548783540725708,
+      "step": 156
+    },
+    {
+      "epoch": 0.06666666666666667,
+      "grad_norm": 0.6760414242744446,
+      "learning_rate": 2.7592267135325134e-05,
+      "loss": 1.184364914894104,
+      "step": 158
+    },
+    {
+      "epoch": 0.06751054852320675,
+      "grad_norm": 0.5261430740356445,
+      "learning_rate": 2.7943760984182776e-05,
+      "loss": 1.1945042610168457,
+      "step": 160
+    },
+    {
+      "epoch": 0.06835443037974684,
+      "grad_norm": 0.6155015230178833,
+      "learning_rate": 2.8295254833040425e-05,
+      "loss": 1.2021973133087158,
+      "step": 162
+    },
+    {
+      "epoch": 0.06919831223628692,
+      "grad_norm": 0.6131619215011597,
+      "learning_rate": 2.8646748681898066e-05,
+      "loss": 1.144123911857605,
+      "step": 164
+    },
+    {
+      "epoch": 0.070042194092827,
+      "grad_norm": 0.5749185681343079,
+      "learning_rate": 2.8998242530755715e-05,
+      "loss": 1.1329256296157837,
+      "step": 166
+    },
+    {
+      "epoch": 0.07088607594936709,
+      "grad_norm": 0.5243118405342102,
+      "learning_rate": 2.9349736379613356e-05,
+      "loss": 1.0892387628555298,
+      "step": 168
+    },
+    {
+      "epoch": 0.07172995780590717,
+      "grad_norm": 0.7190104722976685,
+      "learning_rate": 2.9701230228471005e-05,
+      "loss": 1.163260817527771,
+      "step": 170
+    },
+    {
+      "epoch": 0.07257383966244725,
+      "grad_norm": 0.5486982464790344,
+      "learning_rate": 3.0052724077328647e-05,
+      "loss": 1.0880777835845947,
+      "step": 172
+    },
+    {
+      "epoch": 0.07341772151898734,
+      "grad_norm": 0.5020889043807983,
+      "learning_rate": 3.0404217926186295e-05,
+      "loss": 1.0433368682861328,
+      "step": 174
+    },
+    {
+      "epoch": 0.07426160337552742,
+      "grad_norm": 0.47329774498939514,
+      "learning_rate": 3.075571177504394e-05,
+      "loss": 1.0528991222381592,
+      "step": 176
+    },
+    {
+      "epoch": 0.0751054852320675,
+      "grad_norm": 0.6635547876358032,
+      "learning_rate": 3.110720562390158e-05,
+      "loss": 1.1627811193466187,
+      "step": 178
+    },
+    {
+      "epoch": 0.0759493670886076,
+      "grad_norm": 0.5624618530273438,
+      "learning_rate": 3.145869947275923e-05,
+      "loss": 1.084869384765625,
+      "step": 180
+    },
+    {
+      "epoch": 0.07679324894514768,
+      "grad_norm": 0.6029536724090576,
+      "learning_rate": 3.181019332161687e-05,
+      "loss": 1.2227671146392822,
+      "step": 182
+    },
+    {
+      "epoch": 0.07763713080168777,
+      "grad_norm": 0.930959939956665,
+      "learning_rate": 3.216168717047452e-05,
+      "loss": 1.0955452919006348,
+      "step": 184
+    },
+    {
+      "epoch": 0.07848101265822785,
+      "grad_norm": 0.5326952338218689,
+      "learning_rate": 3.251318101933216e-05,
+      "loss": 1.0640798807144165,
+      "step": 186
+    },
+    {
+      "epoch": 0.07932489451476793,
+      "grad_norm": 0.5484727621078491,
+      "learning_rate": 3.286467486818981e-05,
+      "loss": 1.0700589418411255,
+      "step": 188
+    },
+    {
+      "epoch": 0.08016877637130802,
+      "grad_norm": 0.605273425579071,
+      "learning_rate": 3.3216168717047456e-05,
+      "loss": 1.1593081951141357,
+      "step": 190
+    },
+    {
+      "epoch": 0.0810126582278481,
+      "grad_norm": 0.5704394578933716,
+      "learning_rate": 3.35676625659051e-05,
+      "loss": 1.1617076396942139,
+      "step": 192
+    },
+    {
+      "epoch": 0.08185654008438818,
+      "grad_norm": 0.5929452180862427,
+      "learning_rate": 3.3919156414762746e-05,
+      "loss": 1.1346839666366577,
+      "step": 194
+    },
+    {
+      "epoch": 0.08270042194092828,
+      "grad_norm": 0.5624077916145325,
+      "learning_rate": 3.427065026362039e-05,
+      "loss": 1.0934710502624512,
+      "step": 196
+    },
+    {
+      "epoch": 0.08354430379746836,
+      "grad_norm": 0.6717425584793091,
+      "learning_rate": 3.4622144112478036e-05,
+      "loss": 1.1810534000396729,
+      "step": 198
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "grad_norm": 0.5120199918746948,
+      "learning_rate": 3.4973637961335674e-05,
+      "loss": 1.1525514125823975,
+      "step": 200
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "eval_loss": 1.142486810684204,
+      "eval_runtime": 668.2356,
+      "eval_samples_per_second": 3.153,
+      "eval_steps_per_second": 3.153,
+      "step": 200
+    },
+    {
+      "epoch": 0.08523206751054853,
+      "grad_norm": 0.5144487023353577,
+      "learning_rate": 3.5325131810193326e-05,
+      "loss": 1.0243735313415527,
+      "step": 202
+    },
+    {
+      "epoch": 0.08607594936708861,
+      "grad_norm": 0.6325069069862366,
+      "learning_rate": 3.5676625659050964e-05,
+      "loss": 1.118743896484375,
+      "step": 204
+    },
+    {
+      "epoch": 0.08691983122362869,
+      "grad_norm": 0.5501633882522583,
+      "learning_rate": 3.6028119507908616e-05,
+      "loss": 1.0380504131317139,
+      "step": 206
+    },
+    {
+      "epoch": 0.08776371308016878,
+      "grad_norm": 0.6133899688720703,
+      "learning_rate": 3.6379613356766254e-05,
+      "loss": 0.9837555885314941,
+      "step": 208
+    },
+    {
+      "epoch": 0.08860759493670886,
+      "grad_norm": 0.5799810886383057,
+      "learning_rate": 3.6731107205623906e-05,
+      "loss": 1.090720295906067,
+      "step": 210
+    },
+    {
+      "epoch": 0.08945147679324894,
+      "grad_norm": 0.6039511561393738,
+      "learning_rate": 3.7082601054481544e-05,
+      "loss": 1.120232343673706,
+      "step": 212
+    },
+    {
+      "epoch": 0.09029535864978903,
+      "grad_norm": 0.5983024835586548,
+      "learning_rate": 3.7434094903339196e-05,
+      "loss": 1.096949815750122,
+      "step": 214
+    },
+    {
+      "epoch": 0.09113924050632911,
+      "grad_norm": 0.5641079545021057,
+      "learning_rate": 3.7785588752196835e-05,
+      "loss": 1.1226298809051514,
+      "step": 216
+    },
+    {
+      "epoch": 0.0919831223628692,
+      "grad_norm": 0.655717134475708,
+      "learning_rate": 3.8137082601054486e-05,
+      "loss": 1.1260643005371094,
+      "step": 218
+    },
+    {
+      "epoch": 0.09282700421940929,
+      "grad_norm": 0.6111898422241211,
+      "learning_rate": 3.848857644991213e-05,
+      "loss": 1.0777709484100342,
+      "step": 220
+    },
+    {
+      "epoch": 0.09367088607594937,
+      "grad_norm": 0.6821302771568298,
+      "learning_rate": 3.884007029876977e-05,
+      "loss": 1.10588800907135,
+      "step": 222
+    },
+    {
+      "epoch": 0.09451476793248945,
+      "grad_norm": 0.693175733089447,
+      "learning_rate": 3.919156414762742e-05,
+      "loss": 1.1498671770095825,
+      "step": 224
+    },
+    {
+      "epoch": 0.09535864978902954,
+      "grad_norm": 0.5288166403770447,
+      "learning_rate": 3.954305799648506e-05,
+      "loss": 1.0587562322616577,
+      "step": 226
+    },
+    {
+      "epoch": 0.09620253164556962,
+      "grad_norm": 0.6882867813110352,
+      "learning_rate": 3.989455184534271e-05,
+      "loss": 1.1107512712478638,
+      "step": 228
+    },
+    {
+      "epoch": 0.0970464135021097,
+      "grad_norm": 0.5834154486656189,
+      "learning_rate": 4.024604569420035e-05,
+      "loss": 1.020510196685791,
+      "step": 230
+    },
+    {
+      "epoch": 0.09789029535864979,
+      "grad_norm": 0.7157064080238342,
+      "learning_rate": 4.0597539543058e-05,
+      "loss": 1.0642449855804443,
+      "step": 232
+    },
+    {
+      "epoch": 0.09873417721518987,
+      "grad_norm": 0.6530708074569702,
+      "learning_rate": 4.094903339191564e-05,
+      "loss": 1.0359872579574585,
+      "step": 234
+    },
+    {
+      "epoch": 0.09957805907172995,
+      "grad_norm": 0.6329686045646667,
+      "learning_rate": 4.130052724077329e-05,
+      "loss": 1.050504446029663,
+      "step": 236
+    },
+    {
+      "epoch": 0.10042194092827005,
+      "grad_norm": 0.6597026586532593,
+      "learning_rate": 4.165202108963093e-05,
+      "loss": 1.2621175050735474,
+      "step": 238
+    },
+    {
+      "epoch": 0.10126582278481013,
+      "grad_norm": 0.6195225119590759,
+      "learning_rate": 4.200351493848858e-05,
+      "loss": 1.1218310594558716,
+      "step": 240
+    },
+    {
+      "epoch": 0.1021097046413502,
+      "grad_norm": 0.6764137744903564,
+      "learning_rate": 4.235500878734622e-05,
+      "loss": 1.1250728368759155,
+      "step": 242
+    },
+    {
+      "epoch": 0.1029535864978903,
+      "grad_norm": 0.552363395690918,
+      "learning_rate": 4.270650263620387e-05,
+      "loss": 1.028212308883667,
+      "step": 244
+    },
+    {
+      "epoch": 0.10379746835443038,
+      "grad_norm": 0.5620495676994324,
+      "learning_rate": 4.305799648506151e-05,
+      "loss": 1.0425450801849365,
+      "step": 246
+    },
+    {
+      "epoch": 0.10464135021097046,
+      "grad_norm": 0.6860032081604004,
+      "learning_rate": 4.3409490333919156e-05,
+      "loss": 1.144278883934021,
+      "step": 248
+    },
+    {
+      "epoch": 0.10548523206751055,
+      "grad_norm": 0.6033259034156799,
+      "learning_rate": 4.37609841827768e-05,
+      "loss": 1.1223982572555542,
+      "step": 250
+    },
+    {
+      "epoch": 0.10632911392405063,
+      "grad_norm": 0.6292146444320679,
+      "learning_rate": 4.4112478031634446e-05,
+      "loss": 1.1609960794448853,
+      "step": 252
+    },
+    {
+      "epoch": 0.10717299578059072,
+      "grad_norm": 0.7982883453369141,
+      "learning_rate": 4.44639718804921e-05,
+      "loss": 1.063547968864441,
+      "step": 254
+    },
+    {
+      "epoch": 0.1080168776371308,
+      "grad_norm": 0.7719110250473022,
+      "learning_rate": 4.4815465729349736e-05,
+      "loss": 1.0719804763793945,
+      "step": 256
+    },
+    {
+      "epoch": 0.10886075949367088,
+      "grad_norm": 0.6101011633872986,
+      "learning_rate": 4.516695957820739e-05,
+      "loss": 1.0778400897979736,
+      "step": 258
+    },
+    {
+      "epoch": 0.10970464135021098,
+      "grad_norm": 0.7300994396209717,
+      "learning_rate": 4.5518453427065026e-05,
+      "loss": 1.2129558324813843,
+      "step": 260
+    },
+    {
+      "epoch": 0.11054852320675106,
+      "grad_norm": 0.8348747491836548,
+      "learning_rate": 4.586994727592268e-05,
+      "loss": 1.221714735031128,
+      "step": 262
+    },
+    {
+      "epoch": 0.11139240506329114,
+      "grad_norm": 0.5445612072944641,
+      "learning_rate": 4.6221441124780316e-05,
+      "loss": 1.0187978744506836,
+      "step": 264
+    },
+    {
+      "epoch": 0.11223628691983123,
+      "grad_norm": 0.6230319738388062,
+      "learning_rate": 4.657293497363797e-05,
+      "loss": 1.096561312675476,
+      "step": 266
+    },
+    {
+      "epoch": 0.11308016877637131,
+      "grad_norm": 0.6231237649917603,
+      "learning_rate": 4.6924428822495606e-05,
+      "loss": 1.089842438697815,
+      "step": 268
+    },
+    {
+      "epoch": 0.11392405063291139,
+      "grad_norm": 0.7178627252578735,
+      "learning_rate": 4.727592267135325e-05,
+      "loss": 1.0696645975112915,
+      "step": 270
+    },
+    {
+      "epoch": 0.11476793248945148,
+      "grad_norm": 0.6895854473114014,
+      "learning_rate": 4.7627416520210896e-05,
+      "loss": 1.0511361360549927,
+      "step": 272
+    },
+    {
+      "epoch": 0.11561181434599156,
+      "grad_norm": 0.6046878695487976,
+      "learning_rate": 4.797891036906854e-05,
+      "loss": 1.1373958587646484,
+      "step": 274
+    },
+    {
+      "epoch": 0.11645569620253164,
+      "grad_norm": 0.6524552702903748,
+      "learning_rate": 4.833040421792619e-05,
+      "loss": 1.0734186172485352,
+      "step": 276
+    },
+    {
+      "epoch": 0.11729957805907174,
+      "grad_norm": 0.6331019997596741,
+      "learning_rate": 4.868189806678383e-05,
+      "loss": 1.123913049697876,
+      "step": 278
+    },
+    {
+      "epoch": 0.11814345991561181,
+      "grad_norm": 0.5919018983840942,
+      "learning_rate": 4.903339191564148e-05,
+      "loss": 1.0635710954666138,
+      "step": 280
+    },
+    {
+      "epoch": 0.1189873417721519,
+      "grad_norm": 0.6067633032798767,
+      "learning_rate": 4.938488576449912e-05,
+      "loss": 1.0429247617721558,
+      "step": 282
+    },
+    {
+      "epoch": 0.11983122362869199,
+      "grad_norm": 0.6583750247955322,
+      "learning_rate": 4.9736379613356774e-05,
+      "loss": 1.1397464275360107,
+      "step": 284
+    },
+    {
+      "epoch": 0.12067510548523207,
+      "grad_norm": 0.6200069785118103,
+      "learning_rate": 5.008787346221442e-05,
+      "loss": 1.0590803623199463,
+      "step": 286
+    },
+    {
+      "epoch": 0.12151898734177215,
+      "grad_norm": 0.6798665523529053,
+      "learning_rate": 5.0439367311072064e-05,
+      "loss": 1.1318789720535278,
+      "step": 288
+    },
+    {
+      "epoch": 0.12236286919831224,
+      "grad_norm": 0.7508794069290161,
+      "learning_rate": 5.07908611599297e-05,
+      "loss": 1.0934956073760986,
+      "step": 290
+    },
+    {
+      "epoch": 0.12320675105485232,
+      "grad_norm": 0.6901452541351318,
+      "learning_rate": 5.114235500878735e-05,
+      "loss": 1.163407802581787,
+      "step": 292
+    },
+    {
+      "epoch": 0.1240506329113924,
+      "grad_norm": 0.6423285603523254,
+      "learning_rate": 5.1493848857644985e-05,
+      "loss": 1.09059476852417,
+      "step": 294
+    },
+    {
+      "epoch": 0.1248945147679325,
+      "grad_norm": 0.6839275360107422,
+      "learning_rate": 5.1845342706502644e-05,
+      "loss": 1.0690211057662964,
+      "step": 296
+    },
+    {
+      "epoch": 0.1257383966244726,
+      "grad_norm": 0.6350128054618835,
+      "learning_rate": 5.219683655536028e-05,
+      "loss": 0.982322096824646,
+      "step": 298
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "grad_norm": 0.7136530876159668,
+      "learning_rate": 5.254833040421793e-05,
+      "loss": 1.1132930517196655,
+      "step": 300
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "eval_loss": 1.0952109098434448,
+      "eval_runtime": 677.0652,
+      "eval_samples_per_second": 3.112,
+      "eval_steps_per_second": 3.112,
+      "step": 300
+    },
+    {
+      "epoch": 0.12742616033755275,
+      "grad_norm": 0.7339721322059631,
+      "learning_rate": 5.289982425307557e-05,
+      "loss": 0.973595917224884,
+      "step": 302
+    },
+    {
+      "epoch": 0.12827004219409283,
+      "grad_norm": 0.5941481590270996,
+      "learning_rate": 5.3251318101933224e-05,
+      "loss": 0.9819849729537964,
+      "step": 304
+    },
+    {
+      "epoch": 0.1291139240506329,
+      "grad_norm": 0.7153938412666321,
+      "learning_rate": 5.360281195079086e-05,
+      "loss": 1.0315470695495605,
+      "step": 306
+    },
+    {
+      "epoch": 0.12995780590717299,
+      "grad_norm": 0.5167180299758911,
+      "learning_rate": 5.395430579964851e-05,
+      "loss": 0.9492001533508301,
+      "step": 308
+    },
+    {
+      "epoch": 0.1308016877637131,
+      "grad_norm": 0.6055944561958313,
+      "learning_rate": 5.430579964850615e-05,
+      "loss": 1.0156209468841553,
+      "step": 310
+    },
+    {
+      "epoch": 0.13164556962025317,
+      "grad_norm": 0.7662386298179626,
+      "learning_rate": 5.4657293497363805e-05,
+      "loss": 1.1791651248931885,
+      "step": 312
+    },
+    {
+      "epoch": 0.13248945147679325,
+      "grad_norm": 0.6065546274185181,
+      "learning_rate": 5.500878734622145e-05,
+      "loss": 1.0009297132492065,
+      "step": 314
+    },
+    {
+      "epoch": 0.13333333333333333,
+      "grad_norm": 0.604225754737854,
+      "learning_rate": 5.536028119507909e-05,
+      "loss": 1.0208244323730469,
+      "step": 316
+    },
+    {
+      "epoch": 0.1341772151898734,
+      "grad_norm": 0.6186763048171997,
+      "learning_rate": 5.571177504393673e-05,
+      "loss": 0.9968416690826416,
+      "step": 318
+    },
+    {
+      "epoch": 0.1350210970464135,
+      "grad_norm": 0.7100363969802856,
+      "learning_rate": 5.606326889279437e-05,
+      "loss": 0.9540256857872009,
+      "step": 320
+    },
+    {
+      "epoch": 0.1358649789029536,
+      "grad_norm": 0.6979711055755615,
+      "learning_rate": 5.641476274165203e-05,
+      "loss": 1.0631953477859497,
+      "step": 322
+    },
+    {
+      "epoch": 0.13670886075949368,
+      "grad_norm": 0.6237109303474426,
+      "learning_rate": 5.676625659050967e-05,
+      "loss": 1.0170501470565796,
+      "step": 324
+    },
+    {
+      "epoch": 0.13755274261603376,
+      "grad_norm": 0.6525548696517944,
+      "learning_rate": 5.711775043936731e-05,
+      "loss": 1.0715603828430176,
+      "step": 326
+    },
+    {
+      "epoch": 0.13839662447257384,
+      "grad_norm": 0.6869221329689026,
+      "learning_rate": 5.746924428822495e-05,
+      "loss": 1.0111541748046875,
+      "step": 328
+    },
+    {
+      "epoch": 0.13924050632911392,
+      "grad_norm": 0.553188145160675,
+      "learning_rate": 5.782073813708261e-05,
+      "loss": 1.0311682224273682,
+      "step": 330
+    },
+    {
+      "epoch": 0.140084388185654,
+      "grad_norm": 0.6760852932929993,
+      "learning_rate": 5.817223198594025e-05,
+      "loss": 1.0213634967803955,
+      "step": 332
+    },
+    {
+      "epoch": 0.1409282700421941,
+      "grad_norm": 0.5907419919967651,
+      "learning_rate": 5.8523725834797894e-05,
+      "loss": 0.9748594164848328,
+      "step": 334
+    },
+    {
+      "epoch": 0.14177215189873418,
+      "grad_norm": 0.7044920921325684,
+      "learning_rate": 5.887521968365554e-05,
+      "loss": 1.05863356590271,
+      "step": 336
+    },
+    {
+      "epoch": 0.14261603375527426,
+      "grad_norm": 0.679073691368103,
+      "learning_rate": 5.922671353251318e-05,
+      "loss": 1.1341127157211304,
+      "step": 338
+    },
+    {
+      "epoch": 0.14345991561181434,
+      "grad_norm": 0.7676237225532532,
+      "learning_rate": 5.957820738137083e-05,
+      "loss": 0.9540836215019226,
+      "step": 340
+    },
+    {
+      "epoch": 0.14430379746835442,
+      "grad_norm": 0.6313899755477905,
+      "learning_rate": 5.9929701230228474e-05,
+      "loss": 1.0585911273956299,
+      "step": 342
+    },
+    {
+      "epoch": 0.1451476793248945,
+      "grad_norm": 0.7123099565505981,
+      "learning_rate": 6.028119507908612e-05,
+      "loss": 1.0760118961334229,
+      "step": 344
+    },
+    {
+      "epoch": 0.1459915611814346,
+      "grad_norm": 0.585935652256012,
+      "learning_rate": 6.063268892794376e-05,
+      "loss": 1.036866307258606,
+      "step": 346
+    },
+    {
+      "epoch": 0.1468354430379747,
+      "grad_norm": 0.5643263459205627,
+      "learning_rate": 6.0984182776801416e-05,
+      "loss": 1.0242938995361328,
+      "step": 348
+    },
+    {
+      "epoch": 0.14767932489451477,
+      "grad_norm": 0.626761794090271,
+      "learning_rate": 6.133567662565906e-05,
+      "loss": 1.0497376918792725,
+      "step": 350
+    },
+    {
+      "epoch": 0.14852320675105485,
+      "grad_norm": 0.5106956958770752,
+      "learning_rate": 6.16871704745167e-05,
+      "loss": 0.9811885356903076,
+      "step": 352
+    },
+    {
+      "epoch": 0.14936708860759493,
+      "grad_norm": 0.6948089003562927,
+      "learning_rate": 6.203866432337434e-05,
+      "loss": 1.0715330839157104,
+      "step": 354
+    },
+    {
+      "epoch": 0.150210970464135,
+      "grad_norm": 0.699713945388794,
+      "learning_rate": 6.239015817223199e-05,
+      "loss": 1.0405226945877075,
+      "step": 356
+    },
+    {
+      "epoch": 0.15105485232067511,
+      "grad_norm": 0.6437667012214661,
+      "learning_rate": 6.274165202108964e-05,
+      "loss": 1.0490930080413818,
+      "step": 358
+    },
+    {
+      "epoch": 0.1518987341772152,
+      "grad_norm": 0.6952699422836304,
+      "learning_rate": 6.309314586994728e-05,
+      "loss": 0.9267548322677612,
+      "step": 360
+    },
+    {
+      "epoch": 0.15274261603375527,
+      "grad_norm": 0.6713186502456665,
+      "learning_rate": 6.344463971880492e-05,
+      "loss": 1.0427420139312744,
+      "step": 362
+    },
+    {
+      "epoch": 0.15358649789029535,
+      "grad_norm": 0.6750379800796509,
+      "learning_rate": 6.379613356766257e-05,
+      "loss": 1.048950433731079,
+      "step": 364
+    },
+    {
+      "epoch": 0.15443037974683543,
+      "grad_norm": 0.6053379774093628,
+      "learning_rate": 6.414762741652022e-05,
+      "loss": 1.0156004428863525,
+      "step": 366
+    },
+    {
+      "epoch": 0.15527426160337554,
+      "grad_norm": 0.8063633441925049,
+      "learning_rate": 6.449912126537786e-05,
+      "loss": 1.0020819902420044,
+      "step": 368
+    },
+    {
+      "epoch": 0.15611814345991562,
+      "grad_norm": 0.8027494549751282,
+      "learning_rate": 6.48506151142355e-05,
+      "loss": 1.055633783340454,
+      "step": 370
+    },
+    {
+      "epoch": 0.1569620253164557,
+      "grad_norm": 0.6580121517181396,
+      "learning_rate": 6.520210896309315e-05,
+      "loss": 1.0149940252304077,
+      "step": 372
+    },
+    {
+      "epoch": 0.15780590717299578,
+      "grad_norm": 0.6561233997344971,
+      "learning_rate": 6.55536028119508e-05,
+      "loss": 0.9769611954689026,
+      "step": 374
+    },
+    {
+      "epoch": 0.15864978902953586,
+      "grad_norm": 0.6444346308708191,
+      "learning_rate": 6.590509666080844e-05,
+      "loss": 0.9099349975585938,
+      "step": 376
+    },
+    {
+      "epoch": 0.15949367088607594,
+      "grad_norm": 0.5879359245300293,
+      "learning_rate": 6.625659050966608e-05,
+      "loss": 1.0797548294067383,
+      "step": 378
+    },
+    {
+      "epoch": 0.16033755274261605,
+      "grad_norm": 0.6994144916534424,
+      "learning_rate": 6.660808435852373e-05,
+      "loss": 1.0336791276931763,
+      "step": 380
+    },
+    {
+      "epoch": 0.16118143459915613,
+      "grad_norm": 0.6128669381141663,
+      "learning_rate": 6.695957820738138e-05,
+      "loss": 1.018118143081665,
+      "step": 382
+    },
+    {
+      "epoch": 0.1620253164556962,
+      "grad_norm": 1.0237540006637573,
+      "learning_rate": 6.731107205623902e-05,
+      "loss": 1.1405497789382935,
+      "step": 384
+    },
+    {
+      "epoch": 0.16286919831223629,
+      "grad_norm": 0.6091578006744385,
+      "learning_rate": 6.766256590509666e-05,
+      "loss": 1.0314189195632935,
+      "step": 386
+    },
+    {
+      "epoch": 0.16371308016877636,
+      "grad_norm": 0.5916037559509277,
+      "learning_rate": 6.801405975395431e-05,
+      "loss": 0.9564052820205688,
+      "step": 388
+    },
+    {
+      "epoch": 0.16455696202531644,
+      "grad_norm": 0.771653950214386,
+      "learning_rate": 6.836555360281195e-05,
+      "loss": 1.0023859739303589,
+      "step": 390
+    },
+    {
+      "epoch": 0.16540084388185655,
+      "grad_norm": 0.654658317565918,
+      "learning_rate": 6.87170474516696e-05,
+      "loss": 1.07024085521698,
+      "step": 392
+    },
+    {
+      "epoch": 0.16624472573839663,
+      "grad_norm": 0.6611968874931335,
+      "learning_rate": 6.906854130052724e-05,
+      "loss": 1.0552500486373901,
+      "step": 394
+    },
+    {
+      "epoch": 0.1670886075949367,
+      "grad_norm": 0.6955893039703369,
+      "learning_rate": 6.942003514938489e-05,
+      "loss": 1.0562875270843506,
+      "step": 396
+    },
+    {
+      "epoch": 0.1679324894514768,
+      "grad_norm": 0.6666058301925659,
+      "learning_rate": 6.977152899824253e-05,
+      "loss": 0.9850592017173767,
+      "step": 398
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "grad_norm": 0.6131711006164551,
+      "learning_rate": 7.012302284710018e-05,
+      "loss": 1.0077755451202393,
+      "step": 400
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "eval_loss": 1.0625108480453491,
+      "eval_runtime": 691.0068,
+      "eval_samples_per_second": 3.049,
+      "eval_steps_per_second": 3.049,
+      "step": 400
+    },
+    {
+      "epoch": 0.16962025316455695,
+      "grad_norm": 0.6286499500274658,
+      "learning_rate": 7.047451669595783e-05,
+      "loss": 1.1012427806854248,
+      "step": 402
+    },
+    {
+      "epoch": 0.17046413502109706,
+      "grad_norm": 0.6639351844787598,
+      "learning_rate": 7.082601054481547e-05,
+      "loss": 1.0379719734191895,
+      "step": 404
+    },
+    {
+      "epoch": 0.17130801687763714,
+      "grad_norm": 0.750401496887207,
+      "learning_rate": 7.117750439367311e-05,
+      "loss": 1.031856656074524,
+      "step": 406
+    },
+    {
+      "epoch": 0.17215189873417722,
+      "grad_norm": 0.8084847331047058,
+      "learning_rate": 7.152899824253075e-05,
+      "loss": 1.0493193864822388,
+      "step": 408
+    },
+    {
+      "epoch": 0.1729957805907173,
+      "grad_norm": 0.7448462247848511,
+      "learning_rate": 7.188049209138841e-05,
+      "loss": 1.1012418270111084,
+      "step": 410
+    },
+    {
+      "epoch": 0.17383966244725738,
+      "grad_norm": 0.5841867923736572,
+      "learning_rate": 7.223198594024605e-05,
+      "loss": 0.9926692247390747,
+      "step": 412
+    },
+    {
+      "epoch": 0.17468354430379746,
+      "grad_norm": 0.7125606536865234,
+      "learning_rate": 7.258347978910369e-05,
+      "loss": 1.0588877201080322,
+      "step": 414
+    },
+    {
+      "epoch": 0.17552742616033756,
+      "grad_norm": 0.5750942230224609,
+      "learning_rate": 7.293497363796134e-05,
+      "loss": 1.038270354270935,
+      "step": 416
+    },
+    {
+      "epoch": 0.17637130801687764,
+      "grad_norm": 0.565444827079773,
+      "learning_rate": 7.328646748681899e-05,
+      "loss": 0.9843021035194397,
+      "step": 418
+    },
+    {
+      "epoch": 0.17721518987341772,
+      "grad_norm": 0.5825693011283875,
+      "learning_rate": 7.363796133567663e-05,
+      "loss": 1.0731632709503174,
+      "step": 420
+    },
+    {
+      "epoch": 0.1780590717299578,
+      "grad_norm": 0.6267391443252563,
+      "learning_rate": 7.398945518453427e-05,
+      "loss": 1.0061273574829102,
+      "step": 422
+    },
+    {
+      "epoch": 0.17890295358649788,
+      "grad_norm": 0.6621372103691101,
+      "learning_rate": 7.434094903339192e-05,
+      "loss": 1.0461612939834595,
+      "step": 424
+    },
+    {
+      "epoch": 0.17974683544303796,
+      "grad_norm": 0.6635435223579407,
+      "learning_rate": 7.469244288224957e-05,
+      "loss": 0.9789207577705383,
+      "step": 426
+    },
+    {
+      "epoch": 0.18059071729957807,
+      "grad_norm": 0.6342346668243408,
+      "learning_rate": 7.504393673110721e-05,
+      "loss": 1.0527069568634033,
+      "step": 428
+    },
+    {
+      "epoch": 0.18143459915611815,
+      "grad_norm": 0.6762149930000305,
+      "learning_rate": 7.539543057996485e-05,
+      "loss": 0.9708702564239502,
+      "step": 430
+    },
+    {
+      "epoch": 0.18227848101265823,
+      "grad_norm": 0.7073282599449158,
+      "learning_rate": 7.57469244288225e-05,
+      "loss": 1.0509834289550781,
+      "step": 432
+    },
+    {
+      "epoch": 0.1831223628691983,
+      "grad_norm": 0.6917856931686401,
+      "learning_rate": 7.609841827768014e-05,
+      "loss": 1.0128819942474365,
+      "step": 434
+    },
+    {
+      "epoch": 0.1839662447257384,
+      "grad_norm": 0.5574942231178284,
+      "learning_rate": 7.644991212653779e-05,
+      "loss": 0.989395797252655,
+      "step": 436
+    },
+    {
+      "epoch": 0.1848101265822785,
+      "grad_norm": 0.640765905380249,
+      "learning_rate": 7.680140597539543e-05,
+      "loss": 0.9846042990684509,
+      "step": 438
+    },
+    {
+      "epoch": 0.18565400843881857,
+      "grad_norm": 0.6699127554893494,
+      "learning_rate": 7.715289982425308e-05,
+      "loss": 1.0344442129135132,
+      "step": 440
+    },
+    {
+      "epoch": 0.18649789029535865,
+      "grad_norm": 0.6164930462837219,
+      "learning_rate": 7.750439367311072e-05,
+      "loss": 1.0179373025894165,
+      "step": 442
+    },
+    {
+      "epoch": 0.18734177215189873,
+      "grad_norm": 0.6880720853805542,
+      "learning_rate": 7.785588752196837e-05,
+      "loss": 1.0518895387649536,
+      "step": 444
+    },
+    {
+      "epoch": 0.1881856540084388,
+      "grad_norm": 0.6501413583755493,
+      "learning_rate": 7.820738137082601e-05,
+      "loss": 1.0442606210708618,
+      "step": 446
+    },
+    {
+      "epoch": 0.1890295358649789,
+      "grad_norm": 0.6076085567474365,
+      "learning_rate": 7.855887521968366e-05,
+      "loss": 0.9828442335128784,
+      "step": 448
+    },
+    {
+      "epoch": 0.189873417721519,
+      "grad_norm": 0.6418202519416809,
+      "learning_rate": 7.89103690685413e-05,
+      "loss": 1.0573710203170776,
+      "step": 450
+    },
+    {
+      "epoch": 0.19071729957805908,
+      "grad_norm": 0.7055076360702515,
+      "learning_rate": 7.926186291739895e-05,
+      "loss": 1.0216103792190552,
+      "step": 452
+    },
+    {
+      "epoch": 0.19156118143459916,
+      "grad_norm": 0.5668330192565918,
+      "learning_rate": 7.961335676625659e-05,
+      "loss": 0.9837722778320312,
+      "step": 454
+    },
+    {
+      "epoch": 0.19240506329113924,
+      "grad_norm": 0.6419380307197571,
+      "learning_rate": 7.996485061511424e-05,
+      "loss": 1.0003894567489624,
+      "step": 456
+    },
+    {
+      "epoch": 0.19324894514767932,
+      "grad_norm": 0.5949198007583618,
+      "learning_rate": 8.031634446397188e-05,
+      "loss": 1.0609031915664673,
+      "step": 458
+    },
+    {
+      "epoch": 0.1940928270042194,
+      "grad_norm": 0.7032039761543274,
+      "learning_rate": 8.066783831282952e-05,
+      "loss": 1.0543403625488281,
+      "step": 460
+    },
+    {
+      "epoch": 0.1949367088607595,
+      "grad_norm": 0.5775868892669678,
+      "learning_rate": 8.101933216168718e-05,
+      "loss": 0.9819303154945374,
+      "step": 462
+    },
+    {
+      "epoch": 0.19578059071729959,
+      "grad_norm": 0.9301062226295471,
+      "learning_rate": 8.137082601054482e-05,
+      "loss": 1.0542067289352417,
+      "step": 464
+    },
+    {
+      "epoch": 0.19662447257383966,
+      "grad_norm": 0.6193217039108276,
+      "learning_rate": 8.172231985940246e-05,
+      "loss": 0.9966341257095337,
+      "step": 466
+    },
+    {
+      "epoch": 0.19746835443037974,
+      "grad_norm": 0.6286146640777588,
+      "learning_rate": 8.20738137082601e-05,
+      "loss": 1.0474121570587158,
+      "step": 468
+    },
+    {
+      "epoch": 0.19831223628691982,
+      "grad_norm": 0.7418972253799438,
+      "learning_rate": 8.242530755711776e-05,
+      "loss": 0.9549239277839661,
+      "step": 470
+    },
+    {
+      "epoch": 0.1991561181434599,
+      "grad_norm": 0.6122808456420898,
+      "learning_rate": 8.27768014059754e-05,
+      "loss": 1.0191338062286377,
+      "step": 472
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.6375362277030945,
+      "learning_rate": 8.312829525483304e-05,
+      "loss": 1.0987539291381836,
+      "step": 474
+    },
+    {
+      "epoch": 0.2008438818565401,
+      "grad_norm": 0.6459513306617737,
+      "learning_rate": 8.347978910369068e-05,
+      "loss": 1.0369136333465576,
+      "step": 476
+    },
+    {
+      "epoch": 0.20168776371308017,
+      "grad_norm": 0.7029640674591064,
+      "learning_rate": 8.383128295254833e-05,
+      "loss": 1.0582096576690674,
+      "step": 478
+    },
+    {
+      "epoch": 0.20253164556962025,
+      "grad_norm": 0.6345387697219849,
+      "learning_rate": 8.418277680140598e-05,
+      "loss": 1.022916316986084,
+      "step": 480
+    },
+    {
+      "epoch": 0.20337552742616033,
+      "grad_norm": 0.5764590501785278,
+      "learning_rate": 8.453427065026362e-05,
+      "loss": 0.973024308681488,
+      "step": 482
+    },
+    {
+      "epoch": 0.2042194092827004,
+      "grad_norm": 0.5884482860565186,
+      "learning_rate": 8.488576449912127e-05,
+      "loss": 1.0292812585830688,
+      "step": 484
+    },
+    {
+      "epoch": 0.20506329113924052,
+      "grad_norm": 0.616357147693634,
+      "learning_rate": 8.523725834797891e-05,
+      "loss": 1.0083447694778442,
+      "step": 486
+    },
+    {
+      "epoch": 0.2059071729957806,
+      "grad_norm": 0.7671196460723877,
+      "learning_rate": 8.558875219683656e-05,
+      "loss": 0.9936985373497009,
+      "step": 488
+    },
+    {
+      "epoch": 0.20675105485232068,
+      "grad_norm": 0.6197299957275391,
+      "learning_rate": 8.59402460456942e-05,
+      "loss": 1.051513910293579,
+      "step": 490
+    },
+    {
+      "epoch": 0.20759493670886076,
+      "grad_norm": 0.6912890672683716,
+      "learning_rate": 8.629173989455185e-05,
+      "loss": 0.9474978446960449,
+      "step": 492
+    },
+    {
+      "epoch": 0.20843881856540084,
+      "grad_norm": 0.6941592693328857,
+      "learning_rate": 8.664323374340949e-05,
+      "loss": 1.0671660900115967,
+      "step": 494
+    },
+    {
+      "epoch": 0.20928270042194091,
+      "grad_norm": 0.5889528393745422,
+      "learning_rate": 8.699472759226714e-05,
+      "loss": 1.0020159482955933,
+      "step": 496
+    },
+    {
+      "epoch": 0.21012658227848102,
+      "grad_norm": 0.6478549838066101,
+      "learning_rate": 8.734622144112478e-05,
+      "loss": 1.0165860652923584,
+      "step": 498
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "grad_norm": 0.6018255949020386,
+      "learning_rate": 8.769771528998243e-05,
+      "loss": 0.8798263072967529,
+      "step": 500
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "eval_loss": 1.042096495628357,
+      "eval_runtime": 692.4361,
+      "eval_samples_per_second": 3.043,
+      "eval_steps_per_second": 3.043,
+      "step": 500
+    },
+    {
+      "epoch": 0.21181434599156118,
+      "grad_norm": 0.578990638256073,
+      "learning_rate": 8.804920913884007e-05,
+      "loss": 1.092096209526062,
+      "step": 502
+    },
+    {
+      "epoch": 0.21265822784810126,
+      "grad_norm": 0.6597883701324463,
+      "learning_rate": 8.840070298769771e-05,
+      "loss": 1.0413451194763184,
+      "step": 504
+    },
+    {
+      "epoch": 0.21350210970464134,
+      "grad_norm": 0.6660305261611938,
+      "learning_rate": 8.875219683655536e-05,
+      "loss": 1.0073142051696777,
+      "step": 506
+    },
+    {
+      "epoch": 0.21434599156118145,
+      "grad_norm": 0.6283115148544312,
+      "learning_rate": 8.910369068541301e-05,
+      "loss": 1.0319768190383911,
+      "step": 508
+    },
+    {
+      "epoch": 0.21518987341772153,
+      "grad_norm": 0.6257343292236328,
+      "learning_rate": 8.945518453427065e-05,
+      "loss": 1.0046353340148926,
+      "step": 510
+    },
+    {
+      "epoch": 0.2160337552742616,
+      "grad_norm": 0.5530875325202942,
+      "learning_rate": 8.980667838312829e-05,
+      "loss": 0.9169099926948547,
+      "step": 512
+    },
+    {
+      "epoch": 0.2168776371308017,
+      "grad_norm": 0.5369633436203003,
+      "learning_rate": 9.015817223198594e-05,
+      "loss": 0.9081505537033081,
+      "step": 514
+    },
+    {
+      "epoch": 0.21772151898734177,
+      "grad_norm": 0.6618232131004333,
+      "learning_rate": 9.05096660808436e-05,
+      "loss": 0.9165045022964478,
+      "step": 516
+    },
+    {
+      "epoch": 0.21856540084388185,
+      "grad_norm": 0.600666344165802,
+      "learning_rate": 9.086115992970123e-05,
+      "loss": 0.91348797082901,
+      "step": 518
+    },
+    {
+      "epoch": 0.21940928270042195,
+      "grad_norm": 0.5919831991195679,
+      "learning_rate": 9.121265377855887e-05,
+      "loss": 1.006508469581604,
+      "step": 520
+    },
+    {
+      "epoch": 0.22025316455696203,
+      "grad_norm": 0.688058614730835,
+      "learning_rate": 9.156414762741654e-05,
+      "loss": 1.0013236999511719,
+      "step": 522
+    },
+    {
+      "epoch": 0.2210970464135021,
+      "grad_norm": 0.6721227765083313,
+      "learning_rate": 9.191564147627418e-05,
+      "loss": 1.0909923315048218,
+      "step": 524
+    },
+    {
+      "epoch": 0.2219409282700422,
+      "grad_norm": 0.5987313389778137,
+      "learning_rate": 9.226713532513181e-05,
+      "loss": 1.0117096900939941,
+      "step": 526
+    },
+    {
+      "epoch": 0.22278481012658227,
+      "grad_norm": 0.6191489696502686,
+      "learning_rate": 9.261862917398945e-05,
+      "loss": 1.0153647661209106,
+      "step": 528
+    },
+    {
+      "epoch": 0.22362869198312235,
+      "grad_norm": 0.6821563243865967,
+      "learning_rate": 9.29701230228471e-05,
+      "loss": 0.9649755954742432,
+      "step": 530
+    },
+    {
+      "epoch": 0.22447257383966246,
+      "grad_norm": 1.760398268699646,
+      "learning_rate": 9.332161687170476e-05,
+      "loss": 0.8673232197761536,
+      "step": 532
+    },
+    {
+      "epoch": 0.22531645569620254,
+      "grad_norm": 0.6670058369636536,
+      "learning_rate": 9.36731107205624e-05,
+      "loss": 0.9942440986633301,
+      "step": 534
+    },
+    {
+      "epoch": 0.22616033755274262,
+      "grad_norm": 0.7345916032791138,
+      "learning_rate": 9.402460456942003e-05,
+      "loss": 1.0364389419555664,
+      "step": 536
+    },
+    {
+      "epoch": 0.2270042194092827,
+      "grad_norm": 0.5946128964424133,
+      "learning_rate": 9.437609841827768e-05,
+      "loss": 0.9314924478530884,
+      "step": 538
+    },
+    {
+      "epoch": 0.22784810126582278,
+      "grad_norm": 0.5800848603248596,
+      "learning_rate": 9.472759226713534e-05,
+      "loss": 1.0694862604141235,
+      "step": 540
+    },
+    {
+      "epoch": 0.22869198312236286,
+      "grad_norm": 0.6712192893028259,
+      "learning_rate": 9.507908611599297e-05,
+      "loss": 1.03531014919281,
+      "step": 542
+    },
+    {
+      "epoch": 0.22953586497890296,
+      "grad_norm": 0.5641416311264038,
+      "learning_rate": 9.543057996485063e-05,
+      "loss": 0.9795235991477966,
+      "step": 544
+    },
+    {
+      "epoch": 0.23037974683544304,
+      "grad_norm": 0.50412517786026,
+      "learning_rate": 9.578207381370826e-05,
+      "loss": 0.9641494750976562,
+      "step": 546
+    },
+    {
+      "epoch": 0.23122362869198312,
+      "grad_norm": 0.579118549823761,
+      "learning_rate": 9.61335676625659e-05,
+      "loss": 0.9375281929969788,
+      "step": 548
+    },
+    {
+      "epoch": 0.2320675105485232,
+      "grad_norm": 0.5888341665267944,
+      "learning_rate": 9.648506151142355e-05,
+      "loss": 0.9414046406745911,
+      "step": 550
+    },
+    {
+      "epoch": 0.23291139240506328,
+      "grad_norm": 0.5595056414604187,
+      "learning_rate": 9.68365553602812e-05,
+      "loss": 0.9005617499351501,
+      "step": 552
+    },
+    {
+      "epoch": 0.23375527426160336,
+      "grad_norm": 0.6605326533317566,
+      "learning_rate": 9.718804920913884e-05,
+      "loss": 1.0283968448638916,
+      "step": 554
+    },
+    {
+      "epoch": 0.23459915611814347,
+      "grad_norm": 0.5657313466072083,
+      "learning_rate": 9.753954305799648e-05,
+      "loss": 1.0058249235153198,
+      "step": 556
+    },
+    {
+      "epoch": 0.23544303797468355,
+      "grad_norm": 0.5433364510536194,
+      "learning_rate": 9.789103690685413e-05,
+      "loss": 0.9835494756698608,
+      "step": 558
+    },
+    {
+      "epoch": 0.23628691983122363,
+      "grad_norm": 0.6129802465438843,
+      "learning_rate": 9.824253075571179e-05,
+      "loss": 1.054532527923584,
+      "step": 560
+    },
+    {
+      "epoch": 0.2371308016877637,
+      "grad_norm": 0.6496239304542542,
+      "learning_rate": 9.859402460456942e-05,
+      "loss": 1.0240973234176636,
+      "step": 562
+    },
+    {
+      "epoch": 0.2379746835443038,
+      "grad_norm": 0.6380873918533325,
+      "learning_rate": 9.894551845342706e-05,
+      "loss": 1.0229179859161377,
+      "step": 564
+    },
+    {
+      "epoch": 0.23881856540084387,
+      "grad_norm": 0.6151993870735168,
+      "learning_rate": 9.929701230228471e-05,
+      "loss": 1.0111570358276367,
+      "step": 566
+    },
+    {
+      "epoch": 0.23966244725738398,
+      "grad_norm": 0.5727584958076477,
+      "learning_rate": 9.964850615114237e-05,
+      "loss": 0.9450829029083252,
+      "step": 568
+    },
+    {
+      "epoch": 0.24050632911392406,
+      "grad_norm": 0.6620725989341736,
+      "learning_rate": 0.0001,
+      "loss": 0.9800319075584412,
+      "step": 570
+    },
+    {
+      "epoch": 0.24135021097046414,
+      "grad_norm": 0.6151163578033447,
+      "learning_rate": 0.00010035149384885764,
+      "loss": 0.9757438898086548,
+      "step": 572
+    },
+    {
+      "epoch": 0.24219409282700421,
+      "grad_norm": 0.5672140717506409,
+      "learning_rate": 0.0001007029876977153,
+      "loss": 0.9104921817779541,
+      "step": 574
+    },
+    {
+      "epoch": 0.2430379746835443,
+      "grad_norm": 0.5697256326675415,
+      "learning_rate": 0.00010105448154657293,
+      "loss": 1.1027376651763916,
+      "step": 576
+    },
+    {
+      "epoch": 0.2438818565400844,
+      "grad_norm": 0.5590381622314453,
+      "learning_rate": 0.00010140597539543057,
+      "loss": 1.0055404901504517,
+      "step": 578
+    },
+    {
+      "epoch": 0.24472573839662448,
+      "grad_norm": 0.5518567562103271,
+      "learning_rate": 0.00010175746924428824,
+      "loss": 1.020835518836975,
+      "step": 580
+    },
+    {
+      "epoch": 0.24556962025316456,
+      "grad_norm": 0.6338496208190918,
+      "learning_rate": 0.00010210896309314588,
+      "loss": 0.9528344869613647,
+      "step": 582
+    },
+    {
+      "epoch": 0.24641350210970464,
+      "grad_norm": 0.6497329473495483,
+      "learning_rate": 0.00010246045694200353,
+      "loss": 1.0088670253753662,
+      "step": 584
+    },
+    {
+      "epoch": 0.24725738396624472,
+      "grad_norm": 0.49888095259666443,
+      "learning_rate": 0.00010281195079086117,
+      "loss": 0.9961200952529907,
+      "step": 586
+    },
+    {
+      "epoch": 0.2481012658227848,
+      "grad_norm": 0.5680158734321594,
+      "learning_rate": 0.0001031634446397188,
+      "loss": 0.9635610580444336,
+      "step": 588
+    },
+    {
+      "epoch": 0.2489451476793249,
+      "grad_norm": 0.658168375492096,
+      "learning_rate": 0.00010351493848857646,
+      "loss": 0.9392287135124207,
+      "step": 590
+    },
+    {
+      "epoch": 0.249789029535865,
+      "grad_norm": 0.618262767791748,
+      "learning_rate": 0.0001038664323374341,
+      "loss": 0.9600516557693481,
+      "step": 592
+    },
+    {
+      "epoch": 0.25063291139240507,
+      "grad_norm": 0.6003909111022949,
+      "learning_rate": 0.00010421792618629173,
+      "loss": 1.005476713180542,
+      "step": 594
+    },
+    {
+      "epoch": 0.2514767932489452,
+      "grad_norm": 0.5437078475952148,
+      "learning_rate": 0.00010456942003514938,
+      "loss": 0.9523017406463623,
+      "step": 596
+    },
+    {
+      "epoch": 0.2523206751054852,
+      "grad_norm": 0.5524541735649109,
+      "learning_rate": 0.00010492091388400705,
+      "loss": 0.9526668787002563,
+      "step": 598
+    },
+    {
+      "epoch": 0.25316455696202533,
+      "grad_norm": 0.679504930973053,
+      "learning_rate": 0.00010527240773286469,
+      "loss": 1.019660472869873,
+      "step": 600
+    },
+    {
+      "epoch": 0.25316455696202533,
+      "eval_loss": 1.0193854570388794,
+      "eval_runtime": 677.9523,
+      "eval_samples_per_second": 3.108,
+      "eval_steps_per_second": 3.108,
+      "step": 600
+    },
+    {
+      "epoch": 0.2540084388185654,
+      "grad_norm": 0.5646136999130249,
+      "learning_rate": 0.00010562390158172233,
+      "loss": 0.9910882711410522,
+      "step": 602
+    },
+    {
+      "epoch": 0.2548523206751055,
+      "grad_norm": 0.5238093137741089,
+      "learning_rate": 0.00010597539543057998,
+      "loss": 0.9616432785987854,
+      "step": 604
+    },
+    {
+      "epoch": 0.25569620253164554,
+      "grad_norm": 0.7483857274055481,
+      "learning_rate": 0.00010632688927943762,
+      "loss": 1.0078275203704834,
+      "step": 606
+    },
+    {
+      "epoch": 0.25654008438818565,
+      "grad_norm": 0.578948974609375,
+      "learning_rate": 0.00010667838312829525,
+      "loss": 0.9827103018760681,
+      "step": 608
+    },
+    {
+      "epoch": 0.25738396624472576,
+      "grad_norm": 0.5525906085968018,
+      "learning_rate": 0.00010702987697715289,
+      "loss": 1.0423277616500854,
+      "step": 610
+    },
+    {
+      "epoch": 0.2582278481012658,
+      "grad_norm": 0.6721326112747192,
+      "learning_rate": 0.00010738137082601054,
+      "loss": 0.9561693072319031,
+      "step": 612
+    },
+    {
+      "epoch": 0.2590717299578059,
+      "grad_norm": 0.5701051354408264,
+      "learning_rate": 0.00010773286467486821,
+      "loss": 0.9602992534637451,
+      "step": 614
+    },
+    {
+      "epoch": 0.25991561181434597,
+      "grad_norm": 0.6349860429763794,
+      "learning_rate": 0.00010808435852372585,
+      "loss": 1.1422650814056396,
+      "step": 616
+    },
+    {
+      "epoch": 0.2607594936708861,
+      "grad_norm": 0.5496085286140442,
+      "learning_rate": 0.00010843585237258349,
+      "loss": 0.9762773513793945,
+      "step": 618
+    },
+    {
+      "epoch": 0.2616033755274262,
+      "grad_norm": 0.6080722808837891,
+      "learning_rate": 0.00010878734622144114,
+      "loss": 1.0133616924285889,
+      "step": 620
+    },
+    {
+      "epoch": 0.26244725738396624,
+      "grad_norm": 0.5450218915939331,
+      "learning_rate": 0.00010913884007029878,
+      "loss": 0.9385587573051453,
+      "step": 622
+    },
+    {
+      "epoch": 0.26329113924050634,
+      "grad_norm": 0.592106819152832,
+      "learning_rate": 0.00010949033391915641,
+      "loss": 0.9359989762306213,
+      "step": 624
+    },
+    {
+      "epoch": 0.2641350210970464,
+      "grad_norm": 0.6449427604675293,
+      "learning_rate": 0.00010984182776801407,
+      "loss": 1.0266027450561523,
+      "step": 626
+    },
+    {
+      "epoch": 0.2649789029535865,
+      "grad_norm": 0.538299560546875,
+      "learning_rate": 0.0001101933216168717,
+      "loss": 0.9303187131881714,
+      "step": 628
+    },
+    {
+      "epoch": 0.26582278481012656,
+      "grad_norm": 0.546316921710968,
+      "learning_rate": 0.00011054481546572934,
+      "loss": 0.9368857145309448,
+      "step": 630
+    },
+    {
+      "epoch": 0.26666666666666666,
+      "grad_norm": 0.5818730592727661,
+      "learning_rate": 0.00011089630931458701,
+      "loss": 0.9573145508766174,
+      "step": 632
+    },
+    {
+      "epoch": 0.26751054852320677,
+      "grad_norm": 0.5958262085914612,
+      "learning_rate": 0.00011124780316344465,
+      "loss": 0.9345449805259705,
+      "step": 634
+    },
+    {
+      "epoch": 0.2683544303797468,
+      "grad_norm": 0.6259077787399292,
+      "learning_rate": 0.0001115992970123023,
+      "loss": 1.0906590223312378,
+      "step": 636
+    },
+    {
+      "epoch": 0.26919831223628693,
+      "grad_norm": 0.589672863483429,
+      "learning_rate": 0.00011195079086115994,
+      "loss": 1.0757447481155396,
+      "step": 638
+    },
+    {
+      "epoch": 0.270042194092827,
+      "grad_norm": 0.5714080333709717,
+      "learning_rate": 0.00011230228471001758,
+      "loss": 0.9310855269432068,
+      "step": 640
+    },
+    {
+      "epoch": 0.2708860759493671,
+      "grad_norm": 0.45342639088630676,
+      "learning_rate": 0.00011265377855887523,
+      "loss": 0.9276360273361206,
+      "step": 642
+    },
+    {
+      "epoch": 0.2717299578059072,
+      "grad_norm": 0.6386750340461731,
+      "learning_rate": 0.00011300527240773287,
+      "loss": 1.084719181060791,
+      "step": 644
+    },
+    {
+      "epoch": 0.27257383966244725,
+      "grad_norm": 0.6446163654327393,
+      "learning_rate": 0.0001133567662565905,
+      "loss": 0.9763918519020081,
+      "step": 646
+    },
+    {
+      "epoch": 0.27341772151898736,
+      "grad_norm": 0.5925686359405518,
+      "learning_rate": 0.00011370826010544816,
+      "loss": 0.9517921805381775,
+      "step": 648
+    },
+    {
+      "epoch": 0.2742616033755274,
+      "grad_norm": 0.5399773716926575,
+      "learning_rate": 0.00011405975395430582,
+      "loss": 1.0587927103042603,
+      "step": 650
+    },
+    {
+      "epoch": 0.2751054852320675,
+      "grad_norm": 0.5872456431388855,
+      "learning_rate": 0.00011441124780316346,
+      "loss": 0.883341908454895,
+      "step": 652
+    },
+    {
+      "epoch": 0.2759493670886076,
+      "grad_norm": 0.5574564337730408,
+      "learning_rate": 0.0001147627416520211,
+      "loss": 1.0306891202926636,
+      "step": 654
+    },
+    {
+      "epoch": 0.2767932489451477,
+      "grad_norm": 0.47789013385772705,
+      "learning_rate": 0.00011511423550087874,
+      "loss": 0.8814032077789307,
+      "step": 656
+    },
+    {
+      "epoch": 0.2776371308016878,
+      "grad_norm": 0.5565530061721802,
+      "learning_rate": 0.00011546572934973639,
+      "loss": 0.9460552334785461,
+      "step": 658
+    },
+    {
+      "epoch": 0.27848101265822783,
+      "grad_norm": 0.5299761295318604,
+      "learning_rate": 0.00011581722319859403,
+      "loss": 0.9475110769271851,
+      "step": 660
+    },
+    {
+      "epoch": 0.27932489451476794,
+      "grad_norm": 0.6503344178199768,
+      "learning_rate": 0.00011616871704745166,
+      "loss": 1.0630913972854614,
+      "step": 662
+    },
+    {
+      "epoch": 0.280168776371308,
+      "grad_norm": 0.5794585943222046,
+      "learning_rate": 0.00011652021089630932,
+      "loss": 0.9389138221740723,
+      "step": 664
+    },
+    {
+      "epoch": 0.2810126582278481,
+      "grad_norm": 0.5762867331504822,
+      "learning_rate": 0.00011687170474516695,
+      "loss": 0.8934136033058167,
+      "step": 666
+    },
+    {
+      "epoch": 0.2818565400843882,
+      "grad_norm": 0.6565435528755188,
+      "learning_rate": 0.00011722319859402462,
+      "loss": 1.1072614192962646,
+      "step": 668
+    },
+    {
+      "epoch": 0.28270042194092826,
+      "grad_norm": 0.5819830298423767,
+      "learning_rate": 0.00011757469244288226,
+      "loss": 1.0501434803009033,
+      "step": 670
+    },
+    {
+      "epoch": 0.28354430379746837,
+      "grad_norm": 0.6071487069129944,
+      "learning_rate": 0.00011792618629173991,
+      "loss": 0.9880793690681458,
+      "step": 672
+    },
+    {
+      "epoch": 0.2843881856540084,
+      "grad_norm": 0.5765058398246765,
+      "learning_rate": 0.00011827768014059755,
+      "loss": 0.9670693874359131,
+      "step": 674
+    },
+    {
+      "epoch": 0.2852320675105485,
+      "grad_norm": 0.5245351791381836,
+      "learning_rate": 0.00011862917398945519,
+      "loss": 0.9602360725402832,
+      "step": 676
+    },
+    {
+      "epoch": 0.28607594936708863,
+      "grad_norm": 0.6189922094345093,
+      "learning_rate": 0.00011898066783831282,
+      "loss": 0.9684560894966125,
+      "step": 678
+    },
+    {
+      "epoch": 0.2869198312236287,
+      "grad_norm": 0.6138690710067749,
+      "learning_rate": 0.00011933216168717048,
+      "loss": 0.9465792775154114,
+      "step": 680
+    },
+    {
+      "epoch": 0.2877637130801688,
+      "grad_norm": 0.5371595621109009,
+      "learning_rate": 0.00011968365553602812,
+      "loss": 0.8495944738388062,
+      "step": 682
+    },
+    {
+      "epoch": 0.28860759493670884,
+      "grad_norm": 0.5549944639205933,
+      "learning_rate": 0.00012003514938488578,
+      "loss": 0.9663267135620117,
+      "step": 684
+    },
+    {
+      "epoch": 0.28945147679324895,
+      "grad_norm": 0.6484189033508301,
+      "learning_rate": 0.00012038664323374342,
+      "loss": 0.9736058712005615,
+      "step": 686
+    },
+    {
+      "epoch": 0.290295358649789,
+      "grad_norm": 0.540351390838623,
+      "learning_rate": 0.00012073813708260107,
+      "loss": 1.0591845512390137,
+      "step": 688
+    },
+    {
+      "epoch": 0.2911392405063291,
+      "grad_norm": 0.5657922029495239,
+      "learning_rate": 0.00012108963093145871,
+      "loss": 0.944908618927002,
+      "step": 690
+    },
+    {
+      "epoch": 0.2919831223628692,
+      "grad_norm": 0.6040505170822144,
+      "learning_rate": 0.00012144112478031635,
+      "loss": 1.0018219947814941,
+      "step": 692
+    },
+    {
+      "epoch": 0.29282700421940927,
+      "grad_norm": 0.5435477495193481,
+      "learning_rate": 0.000121792618629174,
+      "loss": 1.0351502895355225,
+      "step": 694
+    },
+    {
+      "epoch": 0.2936708860759494,
+      "grad_norm": 0.5712518692016602,
+      "learning_rate": 0.00012214411247803164,
+      "loss": 0.9935672283172607,
+      "step": 696
+    },
+    {
+      "epoch": 0.29451476793248943,
+      "grad_norm": 0.6138222813606262,
+      "learning_rate": 0.00012249560632688928,
+      "loss": 1.0165108442306519,
+      "step": 698
+    },
+    {
+      "epoch": 0.29535864978902954,
+      "grad_norm": 0.4495212435722351,
+      "learning_rate": 0.00012284710017574691,
+      "loss": 0.9334425926208496,
+      "step": 700
+    },
+    {
+      "epoch": 0.29535864978902954,
+      "eval_loss": 0.996929407119751,
+      "eval_runtime": 668.6398,
+      "eval_samples_per_second": 3.151,
+      "eval_steps_per_second": 3.151,
+      "step": 700
+    },
+    {
+      "epoch": 0.29620253164556964,
+      "grad_norm": 0.5321539044380188,
+      "learning_rate": 0.00012319859402460458,
+      "loss": 1.0516537427902222,
+      "step": 702
+    },
+    {
+      "epoch": 0.2970464135021097,
+      "grad_norm": 0.5716516971588135,
+      "learning_rate": 0.00012355008787346222,
+      "loss": 0.9387198686599731,
+      "step": 704
+    },
+    {
+      "epoch": 0.2978902953586498,
+      "grad_norm": 0.5617920160293579,
+      "learning_rate": 0.00012390158172231988,
+      "loss": 0.99737948179245,
+      "step": 706
+    },
+    {
+      "epoch": 0.29873417721518986,
+      "grad_norm": 0.4922899007797241,
+      "learning_rate": 0.00012425307557117752,
+      "loss": 0.9955025911331177,
+      "step": 708
+    },
+    {
+      "epoch": 0.29957805907172996,
+      "grad_norm": 0.543501615524292,
+      "learning_rate": 0.00012460456942003516,
+      "loss": 0.9124280214309692,
+      "step": 710
+    },
+    {
+      "epoch": 0.30042194092827,
+      "grad_norm": 0.49590054154396057,
+      "learning_rate": 0.0001249560632688928,
+      "loss": 0.9820216298103333,
+      "step": 712
+    },
+    {
+      "epoch": 0.3012658227848101,
+      "grad_norm": 0.5984305739402771,
+      "learning_rate": 0.00012530755711775044,
+      "loss": 1.0152074098587036,
+      "step": 714
+    },
+    {
+      "epoch": 0.30210970464135023,
+      "grad_norm": 0.9343504905700684,
+      "learning_rate": 0.00012565905096660807,
+      "loss": 1.0577725172042847,
+      "step": 716
+    },
+    {
+      "epoch": 0.3029535864978903,
+      "grad_norm": 0.5118702054023743,
+      "learning_rate": 0.0001260105448154657,
+      "loss": 0.9830358028411865,
+      "step": 718
+    },
+    {
+      "epoch": 0.3037974683544304,
+      "grad_norm": 0.4940392076969147,
+      "learning_rate": 0.00012636203866432338,
+      "loss": 0.9466043710708618,
+      "step": 720
+    },
+    {
+      "epoch": 0.30464135021097044,
+      "grad_norm": 0.5965693593025208,
+      "learning_rate": 0.00012671353251318104,
+      "loss": 1.015270709991455,
+      "step": 722
+    },
+    {
+      "epoch": 0.30548523206751055,
+      "grad_norm": 0.5020529627799988,
+      "learning_rate": 0.00012706502636203868,
+      "loss": 0.9703927636146545,
+      "step": 724
+    },
+    {
+      "epoch": 0.30632911392405066,
+      "grad_norm": 0.6067010164260864,
+      "learning_rate": 0.00012741652021089632,
+      "loss": 1.0255526304244995,
+      "step": 726
+    },
+    {
+      "epoch": 0.3071729957805907,
+      "grad_norm": 0.5931884050369263,
+      "learning_rate": 0.00012776801405975396,
+      "loss": 0.9335633516311646,
+      "step": 728
+    },
+    {
+      "epoch": 0.3080168776371308,
+      "grad_norm": 0.5938752293586731,
+      "learning_rate": 0.0001281195079086116,
+      "loss": 1.0921578407287598,
+      "step": 730
+    },
+    {
+      "epoch": 0.30886075949367087,
+      "grad_norm": 0.49728086590766907,
+      "learning_rate": 0.00012847100175746923,
+      "loss": 0.963066041469574,
+      "step": 732
+    },
+    {
+      "epoch": 0.309704641350211,
+      "grad_norm": 0.5452080965042114,
+      "learning_rate": 0.0001288224956063269,
+      "loss": 0.9513075351715088,
+      "step": 734
+    },
+    {
+      "epoch": 0.3105485232067511,
+      "grad_norm": 0.5497731566429138,
+      "learning_rate": 0.00012917398945518454,
+      "loss": 0.8576077222824097,
+      "step": 736
+    },
+    {
+      "epoch": 0.31139240506329113,
+      "grad_norm": 0.5580397248268127,
+      "learning_rate": 0.0001295254833040422,
+      "loss": 0.9542577862739563,
+      "step": 738
+    },
+    {
+      "epoch": 0.31223628691983124,
+      "grad_norm": 0.5890427827835083,
+      "learning_rate": 0.00012987697715289984,
+      "loss": 0.8992732167243958,
+      "step": 740
+    },
+    {
+      "epoch": 0.3130801687763713,
+      "grad_norm": 0.5942965745925903,
+      "learning_rate": 0.00013022847100175748,
+      "loss": 1.0322896242141724,
+      "step": 742
+    },
+    {
+      "epoch": 0.3139240506329114,
+      "grad_norm": 0.6341713070869446,
+      "learning_rate": 0.00013057996485061512,
+      "loss": 0.9217103719711304,
+      "step": 744
+    },
+    {
+      "epoch": 0.31476793248945145,
+      "grad_norm": 0.5294105410575867,
+      "learning_rate": 0.00013093145869947276,
+      "loss": 0.951789915561676,
+      "step": 746
+    },
+    {
+      "epoch": 0.31561181434599156,
+      "grad_norm": 0.6372058391571045,
+      "learning_rate": 0.0001312829525483304,
+      "loss": 0.9459875226020813,
+      "step": 748
+    },
+    {
+      "epoch": 0.31645569620253167,
+      "grad_norm": 0.5979796648025513,
+      "learning_rate": 0.00013163444639718806,
+      "loss": 0.9626097679138184,
+      "step": 750
+    },
+    {
+      "epoch": 0.3172995780590717,
+      "grad_norm": 0.5682399868965149,
+      "learning_rate": 0.0001319859402460457,
+      "loss": 1.0261781215667725,
+      "step": 752
+    },
+    {
+      "epoch": 0.3181434599156118,
+      "grad_norm": 0.5349125266075134,
+      "learning_rate": 0.00013233743409490336,
+      "loss": 0.9319828152656555,
+      "step": 754
+    },
+    {
+      "epoch": 0.3189873417721519,
+      "grad_norm": 0.6093934178352356,
+      "learning_rate": 0.000132688927943761,
+      "loss": 0.9216550588607788,
+      "step": 756
+    },
+    {
+      "epoch": 0.319831223628692,
+      "grad_norm": 0.5188612341880798,
+      "learning_rate": 0.00013304042179261864,
+      "loss": 0.901739776134491,
+      "step": 758
+    },
+    {
+      "epoch": 0.3206751054852321,
+      "grad_norm": 0.5877130627632141,
+      "learning_rate": 0.00013339191564147628,
+      "loss": 1.0362589359283447,
+      "step": 760
+    },
+    {
+      "epoch": 0.32151898734177214,
+      "grad_norm": 0.5542771816253662,
+      "learning_rate": 0.00013374340949033392,
+      "loss": 0.8787116408348083,
+      "step": 762
+    },
+    {
+      "epoch": 0.32236286919831225,
+      "grad_norm": 0.5084902048110962,
+      "learning_rate": 0.00013409490333919156,
+      "loss": 0.9237037301063538,
+      "step": 764
+    },
+    {
+      "epoch": 0.3232067510548523,
+      "grad_norm": 0.5461528301239014,
+      "learning_rate": 0.00013444639718804922,
+      "loss": 1.0150731801986694,
+      "step": 766
+    },
+    {
+      "epoch": 0.3240506329113924,
+      "grad_norm": 0.53483647108078,
+      "learning_rate": 0.00013479789103690686,
+      "loss": 0.8985214829444885,
+      "step": 768
+    },
+    {
+      "epoch": 0.32489451476793246,
+      "grad_norm": 0.5580531358718872,
+      "learning_rate": 0.0001351493848857645,
+      "loss": 1.0225775241851807,
+      "step": 770
+    },
+    {
+      "epoch": 0.32573839662447257,
+      "grad_norm": 0.5203377604484558,
+      "learning_rate": 0.00013550087873462216,
+      "loss": 0.9571293592453003,
+      "step": 772
+    },
+    {
+      "epoch": 0.3265822784810127,
+      "grad_norm": 0.5049671530723572,
+      "learning_rate": 0.0001358523725834798,
+      "loss": 1.0468909740447998,
+      "step": 774
+    },
+    {
+      "epoch": 0.32742616033755273,
+      "grad_norm": 0.4723063111305237,
+      "learning_rate": 0.00013620386643233744,
+      "loss": 0.7743215560913086,
+      "step": 776
+    },
+    {
+      "epoch": 0.32827004219409284,
+      "grad_norm": 0.6310980916023254,
+      "learning_rate": 0.00013655536028119508,
+      "loss": 1.021510362625122,
+      "step": 778
+    },
+    {
+      "epoch": 0.3291139240506329,
+      "grad_norm": 0.47066664695739746,
+      "learning_rate": 0.00013690685413005274,
+      "loss": 0.9134382605552673,
+      "step": 780
+    },
+    {
+      "epoch": 0.329957805907173,
+      "grad_norm": 0.5725092887878418,
+      "learning_rate": 0.00013725834797891038,
+      "loss": 0.9797834753990173,
+      "step": 782
+    },
+    {
+      "epoch": 0.3308016877637131,
+      "grad_norm": 0.5139563083648682,
+      "learning_rate": 0.00013760984182776802,
+      "loss": 0.9372621178627014,
+      "step": 784
+    },
+    {
+      "epoch": 0.33164556962025316,
+      "grad_norm": 0.5275821685791016,
+      "learning_rate": 0.00013796133567662566,
+      "loss": 0.9528245329856873,
+      "step": 786
+    },
+    {
+      "epoch": 0.33248945147679326,
+      "grad_norm": 0.5702582001686096,
+      "learning_rate": 0.0001383128295254833,
+      "loss": 0.9750176072120667,
+      "step": 788
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.5281293392181396,
+      "learning_rate": 0.00013866432337434096,
+      "loss": 0.9412306547164917,
+      "step": 790
+    },
+    {
+      "epoch": 0.3341772151898734,
+      "grad_norm": 0.5578986406326294,
+      "learning_rate": 0.0001390158172231986,
+      "loss": 0.997580885887146,
+      "step": 792
+    },
+    {
+      "epoch": 0.33502109704641353,
+      "grad_norm": 0.50461345911026,
+      "learning_rate": 0.00013936731107205624,
+      "loss": 0.9082320928573608,
+      "step": 794
+    },
+    {
+      "epoch": 0.3358649789029536,
+      "grad_norm": 0.5258530378341675,
+      "learning_rate": 0.0001397188049209139,
+      "loss": 1.0082844495773315,
+      "step": 796
+    },
+    {
+      "epoch": 0.3367088607594937,
+      "grad_norm": 0.5548169016838074,
+      "learning_rate": 0.00014007029876977154,
+      "loss": 0.9729003310203552,
+      "step": 798
+    },
+    {
+      "epoch": 0.33755274261603374,
+      "grad_norm": 0.48601076006889343,
+      "learning_rate": 0.00014042179261862918,
+      "loss": 0.9099526405334473,
+      "step": 800
+    },
+    {
+      "epoch": 0.33755274261603374,
+      "eval_loss": 0.9800403714179993,
+      "eval_runtime": 678.8306,
+      "eval_samples_per_second": 3.104,
+      "eval_steps_per_second": 3.104,
+      "step": 800
+    },
+    {
+      "epoch": 0.33839662447257385,
+      "grad_norm": 0.5413158535957336,
+      "learning_rate": 0.00014077328646748682,
+      "loss": 0.8610644936561584,
+      "step": 802
+    },
+    {
+      "epoch": 0.3392405063291139,
+      "grad_norm": 0.5147035717964172,
+      "learning_rate": 0.00014112478031634446,
+      "loss": 0.9584825038909912,
+      "step": 804
+    },
+    {
+      "epoch": 0.340084388185654,
+      "grad_norm": 0.5931771397590637,
+      "learning_rate": 0.0001414762741652021,
+      "loss": 1.0142558813095093,
+      "step": 806
+    },
+    {
+      "epoch": 0.3409282700421941,
+      "grad_norm": 0.5178377032279968,
+      "learning_rate": 0.00014182776801405976,
+      "loss": 1.0078763961791992,
+      "step": 808
+    },
+    {
+      "epoch": 0.34177215189873417,
+      "grad_norm": 0.5453237295150757,
+      "learning_rate": 0.0001421792618629174,
+      "loss": 0.9107215404510498,
+      "step": 810
+    },
+    {
+      "epoch": 0.3426160337552743,
+      "grad_norm": 0.5886152982711792,
+      "learning_rate": 0.00014253075571177506,
+      "loss": 0.9981362819671631,
+      "step": 812
+    },
+    {
+      "epoch": 0.3434599156118143,
+      "grad_norm": 0.48040178418159485,
+      "learning_rate": 0.0001428822495606327,
+      "loss": 0.9636131525039673,
+      "step": 814
+    },
+    {
+      "epoch": 0.34430379746835443,
+      "grad_norm": 0.5011753439903259,
+      "learning_rate": 0.00014323374340949034,
+      "loss": 0.9590586423873901,
+      "step": 816
+    },
+    {
+      "epoch": 0.34514767932489454,
+      "grad_norm": 0.57858806848526,
+      "learning_rate": 0.00014358523725834798,
+      "loss": 0.978246808052063,
+      "step": 818
+    },
+    {
+      "epoch": 0.3459915611814346,
+      "grad_norm": 0.46092939376831055,
+      "learning_rate": 0.00014393673110720562,
+      "loss": 0.8549934029579163,
+      "step": 820
+    },
+    {
+      "epoch": 0.3468354430379747,
+      "grad_norm": 0.5756489038467407,
+      "learning_rate": 0.00014428822495606326,
+      "loss": 0.9771265387535095,
+      "step": 822
+    },
+    {
+      "epoch": 0.34767932489451475,
+      "grad_norm": 0.5501731634140015,
+      "learning_rate": 0.00014463971880492092,
+      "loss": 0.8739748001098633,
+      "step": 824
+    },
+    {
+      "epoch": 0.34852320675105486,
+      "grad_norm": 0.5451868176460266,
+      "learning_rate": 0.0001449912126537786,
+      "loss": 0.9129468202590942,
+      "step": 826
+    },
+    {
+      "epoch": 0.3493670886075949,
+      "grad_norm": 0.4624619781970978,
+      "learning_rate": 0.00014534270650263623,
+      "loss": 0.9196704030036926,
+      "step": 828
+    },
+    {
+      "epoch": 0.350210970464135,
+      "grad_norm": 0.520878791809082,
+      "learning_rate": 0.00014569420035149386,
+      "loss": 0.9976527690887451,
+      "step": 830
+    },
+    {
+      "epoch": 0.3510548523206751,
+      "grad_norm": 0.4469171464443207,
+      "learning_rate": 0.0001460456942003515,
+      "loss": 0.7753443717956543,
+      "step": 832
+    },
+    {
+      "epoch": 0.3518987341772152,
+      "grad_norm": 0.5105249881744385,
+      "learning_rate": 0.00014639718804920914,
+      "loss": 0.9584846496582031,
+      "step": 834
+    },
+    {
+      "epoch": 0.3527426160337553,
+      "grad_norm": 0.5043913125991821,
+      "learning_rate": 0.00014674868189806678,
+      "loss": 0.9013500213623047,
+      "step": 836
+    },
+    {
+      "epoch": 0.35358649789029534,
+      "grad_norm": 0.575850784778595,
+      "learning_rate": 0.00014710017574692442,
+      "loss": 0.9775562286376953,
+      "step": 838
+    },
+    {
+      "epoch": 0.35443037974683544,
+      "grad_norm": 0.5128876566886902,
+      "learning_rate": 0.00014745166959578208,
+      "loss": 0.9278940558433533,
+      "step": 840
+    },
+    {
+      "epoch": 0.35527426160337555,
+      "grad_norm": 0.5757885575294495,
+      "learning_rate": 0.00014780316344463975,
+      "loss": 1.0091488361358643,
+      "step": 842
+    },
+    {
+      "epoch": 0.3561181434599156,
+      "grad_norm": 0.500934898853302,
+      "learning_rate": 0.00014815465729349739,
+      "loss": 0.9286836981773376,
+      "step": 844
+    },
+    {
+      "epoch": 0.3569620253164557,
+      "grad_norm": 0.5220686197280884,
+      "learning_rate": 0.00014850615114235502,
+      "loss": 0.9484171867370605,
+      "step": 846
+    },
+    {
+      "epoch": 0.35780590717299576,
+      "grad_norm": 0.5494697690010071,
+      "learning_rate": 0.00014885764499121266,
+      "loss": 1.0556397438049316,
+      "step": 848
+    },
+    {
+      "epoch": 0.35864978902953587,
+      "grad_norm": 0.46633943915367126,
+      "learning_rate": 0.0001492091388400703,
+      "loss": 0.933089017868042,
+      "step": 850
+    },
+    {
+      "epoch": 0.3594936708860759,
+      "grad_norm": 0.4864962697029114,
+      "learning_rate": 0.00014956063268892794,
+      "loss": 1.0016963481903076,
+      "step": 852
+    },
+    {
+      "epoch": 0.36033755274261603,
+      "grad_norm": 0.5032764673233032,
+      "learning_rate": 0.0001499121265377856,
+      "loss": 0.9085348844528198,
+      "step": 854
+    },
+    {
+      "epoch": 0.36118143459915614,
+      "grad_norm": 0.5424998998641968,
+      "learning_rate": 0.00015026362038664324,
+      "loss": 0.923502504825592,
+      "step": 856
+    },
+    {
+      "epoch": 0.3620253164556962,
+      "grad_norm": 0.5181655287742615,
+      "learning_rate": 0.00015061511423550088,
+      "loss": 0.919174313545227,
+      "step": 858
+    },
+    {
+      "epoch": 0.3628691983122363,
+      "grad_norm": 0.5129443407058716,
+      "learning_rate": 0.00015096660808435855,
+      "loss": 0.851981520652771,
+      "step": 860
+    },
+    {
+      "epoch": 0.36371308016877635,
+      "grad_norm": 0.49540698528289795,
+      "learning_rate": 0.00015131810193321618,
+      "loss": 0.8633858561515808,
+      "step": 862
+    },
+    {
+      "epoch": 0.36455696202531646,
+      "grad_norm": 0.4706701934337616,
+      "learning_rate": 0.00015166959578207382,
+      "loss": 0.9473677277565002,
+      "step": 864
+    },
+    {
+      "epoch": 0.36540084388185656,
+      "grad_norm": 0.587704598903656,
+      "learning_rate": 0.00015202108963093146,
+      "loss": 0.942383885383606,
+      "step": 866
+    },
+    {
+      "epoch": 0.3662447257383966,
+      "grad_norm": 0.5851273536682129,
+      "learning_rate": 0.0001523725834797891,
+      "loss": 0.9811251163482666,
+      "step": 868
+    },
+    {
+      "epoch": 0.3670886075949367,
+      "grad_norm": 0.46357613801956177,
+      "learning_rate": 0.00015272407732864676,
+      "loss": 0.8151084184646606,
+      "step": 870
+    },
+    {
+      "epoch": 0.3679324894514768,
+      "grad_norm": 0.47250670194625854,
+      "learning_rate": 0.0001530755711775044,
+      "loss": 0.9023333191871643,
+      "step": 872
+    },
+    {
+      "epoch": 0.3687763713080169,
+      "grad_norm": 0.47510042786598206,
+      "learning_rate": 0.00015342706502636204,
+      "loss": 0.8758499622344971,
+      "step": 874
+    },
+    {
+      "epoch": 0.369620253164557,
+      "grad_norm": 0.5687124729156494,
+      "learning_rate": 0.00015377855887521968,
+      "loss": 0.9837421774864197,
+      "step": 876
+    },
+    {
+      "epoch": 0.37046413502109704,
+      "grad_norm": 0.49064236879348755,
+      "learning_rate": 0.00015413005272407735,
+      "loss": 0.9084216356277466,
+      "step": 878
+    },
+    {
+      "epoch": 0.37130801687763715,
+      "grad_norm": 0.5288164615631104,
+      "learning_rate": 0.00015448154657293498,
+      "loss": 0.8425542712211609,
+      "step": 880
+    },
+    {
+      "epoch": 0.3721518987341772,
+      "grad_norm": 0.5446951985359192,
+      "learning_rate": 0.00015483304042179262,
+      "loss": 1.006197452545166,
+      "step": 882
+    },
+    {
+      "epoch": 0.3729957805907173,
+      "grad_norm": 0.46872642636299133,
+      "learning_rate": 0.00015518453427065026,
+      "loss": 0.9779444932937622,
+      "step": 884
+    },
+    {
+      "epoch": 0.37383966244725736,
+      "grad_norm": 0.4852714240550995,
+      "learning_rate": 0.00015553602811950793,
+      "loss": 0.9099963307380676,
+      "step": 886
+    },
+    {
+      "epoch": 0.37468354430379747,
+      "grad_norm": 0.5219841003417969,
+      "learning_rate": 0.00015588752196836556,
+      "loss": 0.9730570316314697,
+      "step": 888
+    },
+    {
+      "epoch": 0.3755274261603376,
+      "grad_norm": 0.5258626341819763,
+      "learning_rate": 0.0001562390158172232,
+      "loss": 0.9304586052894592,
+      "step": 890
+    },
+    {
+      "epoch": 0.3763713080168776,
+      "grad_norm": 0.4266716241836548,
+      "learning_rate": 0.00015659050966608084,
+      "loss": 0.8680716156959534,
+      "step": 892
+    },
+    {
+      "epoch": 0.37721518987341773,
+      "grad_norm": 0.46361327171325684,
+      "learning_rate": 0.0001569420035149385,
+      "loss": 0.8428018093109131,
+      "step": 894
+    },
+    {
+      "epoch": 0.3780590717299578,
+      "grad_norm": 0.5313687920570374,
+      "learning_rate": 0.00015729349736379614,
+      "loss": 0.8465750217437744,
+      "step": 896
+    },
+    {
+      "epoch": 0.3789029535864979,
+      "grad_norm": 0.47166210412979126,
+      "learning_rate": 0.00015764499121265378,
+      "loss": 0.9120327234268188,
+      "step": 898
+    },
+    {
+      "epoch": 0.379746835443038,
+      "grad_norm": 0.4647318720817566,
+      "learning_rate": 0.00015799648506151145,
+      "loss": 0.8950425982475281,
+      "step": 900
+    },
+    {
+      "epoch": 0.379746835443038,
+      "eval_loss": 0.9643027186393738,
+      "eval_runtime": 691.7929,
+      "eval_samples_per_second": 3.046,
+      "eval_steps_per_second": 3.046,
+      "step": 900
+    },
+    {
+      "epoch": 0.38059071729957805,
+      "grad_norm": 0.5445119738578796,
+      "learning_rate": 0.00015834797891036909,
+      "loss": 0.908163845539093,
+      "step": 902
+    },
+    {
+      "epoch": 0.38143459915611816,
+      "grad_norm": 0.4311858117580414,
+      "learning_rate": 0.00015869947275922672,
+      "loss": 0.8945821523666382,
+      "step": 904
+    },
+    {
+      "epoch": 0.3822784810126582,
+      "grad_norm": 0.5590984225273132,
+      "learning_rate": 0.00015905096660808436,
+      "loss": 0.9478458762168884,
+      "step": 906
+    },
+    {
+      "epoch": 0.3831223628691983,
+      "grad_norm": 0.5470241904258728,
+      "learning_rate": 0.000159402460456942,
+      "loss": 0.9259957671165466,
+      "step": 908
+    },
+    {
+      "epoch": 0.38396624472573837,
+      "grad_norm": 0.5498791337013245,
+      "learning_rate": 0.00015975395430579964,
+      "loss": 0.8824930787086487,
+      "step": 910
+    },
+    {
+      "epoch": 0.3848101265822785,
+      "grad_norm": 0.4779198467731476,
+      "learning_rate": 0.0001601054481546573,
+      "loss": 0.8842340707778931,
+      "step": 912
+    },
+    {
+      "epoch": 0.3856540084388186,
+      "grad_norm": 0.5390620827674866,
+      "learning_rate": 0.00016045694200351494,
+      "loss": 0.92950040102005,
+      "step": 914
+    },
+    {
+      "epoch": 0.38649789029535864,
+      "grad_norm": 0.505519688129425,
+      "learning_rate": 0.0001608084358523726,
+      "loss": 0.8420897126197815,
+      "step": 916
+    },
+    {
+      "epoch": 0.38734177215189874,
+      "grad_norm": 0.4463907778263092,
+      "learning_rate": 0.00016115992970123025,
+      "loss": 0.8256624341011047,
+      "step": 918
+    },
+    {
+      "epoch": 0.3881856540084388,
+      "grad_norm": 0.5321422219276428,
+      "learning_rate": 0.00016151142355008788,
+      "loss": 0.8701168298721313,
+      "step": 920
+    },
+    {
+      "epoch": 0.3890295358649789,
+      "grad_norm": 0.4343073070049286,
+      "learning_rate": 0.00016186291739894552,
+      "loss": 0.9019309878349304,
+      "step": 922
+    },
+    {
+      "epoch": 0.389873417721519,
+      "grad_norm": 0.5311984419822693,
+      "learning_rate": 0.00016221441124780316,
+      "loss": 0.8560551404953003,
+      "step": 924
+    },
+    {
+      "epoch": 0.39071729957805906,
+      "grad_norm": 0.588691771030426,
+      "learning_rate": 0.0001625659050966608,
+      "loss": 0.9700050354003906,
+      "step": 926
+    },
+    {
+      "epoch": 0.39156118143459917,
+      "grad_norm": 0.5425586104393005,
+      "learning_rate": 0.00016291739894551844,
+      "loss": 1.043768048286438,
+      "step": 928
+    },
+    {
+      "epoch": 0.3924050632911392,
+      "grad_norm": 0.5228736996650696,
+      "learning_rate": 0.0001632688927943761,
+      "loss": 0.9501712918281555,
+      "step": 930
+    },
+    {
+      "epoch": 0.39324894514767933,
+      "grad_norm": 0.48960360884666443,
+      "learning_rate": 0.00016362038664323377,
+      "loss": 0.9223058223724365,
+      "step": 932
+    },
+    {
+      "epoch": 0.39409282700421944,
+      "grad_norm": 0.45204755663871765,
+      "learning_rate": 0.0001639718804920914,
+      "loss": 0.9692960977554321,
+      "step": 934
+    },
+    {
+      "epoch": 0.3949367088607595,
+      "grad_norm": 0.5299274921417236,
+      "learning_rate": 0.00016432337434094905,
+      "loss": 0.9467466473579407,
+      "step": 936
+    },
+    {
+      "epoch": 0.3957805907172996,
+      "grad_norm": 0.5607715249061584,
+      "learning_rate": 0.00016467486818980668,
+      "loss": 0.9118053317070007,
+      "step": 938
+    },
+    {
+      "epoch": 0.39662447257383965,
+      "grad_norm": 0.5271831154823303,
+      "learning_rate": 0.00016502636203866432,
+      "loss": 0.9131460189819336,
+      "step": 940
+    },
+    {
+      "epoch": 0.39746835443037976,
+      "grad_norm": 0.5075286030769348,
+      "learning_rate": 0.00016537785588752196,
+      "loss": 0.9358300566673279,
+      "step": 942
+    },
+    {
+      "epoch": 0.3983122362869198,
+      "grad_norm": 0.515731155872345,
+      "learning_rate": 0.00016572934973637963,
+      "loss": 0.8908210396766663,
+      "step": 944
+    },
+    {
+      "epoch": 0.3991561181434599,
+      "grad_norm": 0.4856977164745331,
+      "learning_rate": 0.00016608084358523726,
+      "loss": 0.9775290489196777,
+      "step": 946
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.48846355080604553,
+      "learning_rate": 0.00016643233743409493,
+      "loss": 0.8957490921020508,
+      "step": 948
+    },
+    {
+      "epoch": 0.4008438818565401,
+      "grad_norm": 0.42990800738334656,
+      "learning_rate": 0.00016678383128295257,
+      "loss": 0.9036174416542053,
+      "step": 950
+    },
+    {
+      "epoch": 0.4016877637130802,
+      "grad_norm": 0.49552062153816223,
+      "learning_rate": 0.0001671353251318102,
+      "loss": 0.991032600402832,
+      "step": 952
+    },
+    {
+      "epoch": 0.40253164556962023,
+      "grad_norm": 0.4565040171146393,
+      "learning_rate": 0.00016748681898066784,
+      "loss": 0.823063313961029,
+      "step": 954
+    },
+    {
+      "epoch": 0.40337552742616034,
+      "grad_norm": 0.4290153682231903,
+      "learning_rate": 0.00016783831282952548,
+      "loss": 0.8785063624382019,
+      "step": 956
+    },
+    {
+      "epoch": 0.40421940928270045,
+      "grad_norm": 0.5419702529907227,
+      "learning_rate": 0.00016818980667838312,
+      "loss": 0.8763971924781799,
+      "step": 958
+    },
+    {
+      "epoch": 0.4050632911392405,
+      "grad_norm": 0.5177501440048218,
+      "learning_rate": 0.00016854130052724079,
+      "loss": 0.9470553398132324,
+      "step": 960
+    },
+    {
+      "epoch": 0.4059071729957806,
+      "grad_norm": 0.539725661277771,
+      "learning_rate": 0.00016889279437609842,
+      "loss": 0.9235025644302368,
+      "step": 962
+    },
+    {
+      "epoch": 0.40675105485232066,
+      "grad_norm": 0.5324983596801758,
+      "learning_rate": 0.0001692442882249561,
+      "loss": 1.0248996019363403,
+      "step": 964
+    },
+    {
+      "epoch": 0.40759493670886077,
+      "grad_norm": 0.4936407506465912,
+      "learning_rate": 0.00016959578207381373,
+      "loss": 0.9076873660087585,
+      "step": 966
+    },
+    {
+      "epoch": 0.4084388185654008,
+      "grad_norm": 0.4960501194000244,
+      "learning_rate": 0.00016994727592267137,
+      "loss": 0.9162673950195312,
+      "step": 968
+    },
+    {
+      "epoch": 0.4092827004219409,
+      "grad_norm": 0.45093682408332825,
+      "learning_rate": 0.000170298769771529,
+      "loss": 0.904100775718689,
+      "step": 970
+    },
+    {
+      "epoch": 0.41012658227848103,
+      "grad_norm": 0.4560275077819824,
+      "learning_rate": 0.00017065026362038664,
+      "loss": 0.8633337020874023,
+      "step": 972
+    },
+    {
+      "epoch": 0.4109704641350211,
+      "grad_norm": 0.44885000586509705,
+      "learning_rate": 0.00017100175746924428,
+      "loss": 0.8454209566116333,
+      "step": 974
+    },
+    {
+      "epoch": 0.4118143459915612,
+      "grad_norm": 0.47251659631729126,
+      "learning_rate": 0.00017135325131810195,
+      "loss": 0.824730396270752,
+      "step": 976
+    },
+    {
+      "epoch": 0.41265822784810124,
+      "grad_norm": 0.6597666144371033,
+      "learning_rate": 0.00017170474516695959,
+      "loss": 0.9496501684188843,
+      "step": 978
+    },
+    {
+      "epoch": 0.41350210970464135,
+      "grad_norm": 0.39806297421455383,
+      "learning_rate": 0.00017205623901581722,
+      "loss": 0.9419087171554565,
+      "step": 980
+    },
+    {
+      "epoch": 0.41434599156118146,
+      "grad_norm": 0.48231109976768494,
+      "learning_rate": 0.0001724077328646749,
+      "loss": 0.9182976484298706,
+      "step": 982
+    },
+    {
+      "epoch": 0.4151898734177215,
+      "grad_norm": 0.5438776612281799,
+      "learning_rate": 0.00017275922671353253,
+      "loss": 0.9386967420578003,
+      "step": 984
+    },
+    {
+      "epoch": 0.4160337552742616,
+      "grad_norm": 0.4959667921066284,
+      "learning_rate": 0.00017311072056239017,
+      "loss": 0.897849440574646,
+      "step": 986
+    },
+    {
+      "epoch": 0.41687763713080167,
+      "grad_norm": 0.43533357977867126,
+      "learning_rate": 0.0001734622144112478,
+      "loss": 0.8776953816413879,
+      "step": 988
+    },
+    {
+      "epoch": 0.4177215189873418,
+      "grad_norm": 0.47513946890830994,
+      "learning_rate": 0.00017381370826010547,
+      "loss": 0.9162989854812622,
+      "step": 990
+    },
+    {
+      "epoch": 0.41856540084388183,
+      "grad_norm": 0.4907188415527344,
+      "learning_rate": 0.0001741652021089631,
+      "loss": 0.9482660889625549,
+      "step": 992
+    },
+    {
+      "epoch": 0.41940928270042194,
+      "grad_norm": 0.44499966502189636,
+      "learning_rate": 0.00017451669595782075,
+      "loss": 0.8812930583953857,
+      "step": 994
+    },
+    {
+      "epoch": 0.42025316455696204,
+      "grad_norm": 0.4535730481147766,
+      "learning_rate": 0.00017486818980667838,
+      "loss": 0.9439874887466431,
+      "step": 996
+    },
+    {
+      "epoch": 0.4210970464135021,
+      "grad_norm": 0.5240745544433594,
+      "learning_rate": 0.00017521968365553602,
+      "loss": 0.8818395137786865,
+      "step": 998
+    },
+    {
+      "epoch": 0.4219409282700422,
+      "grad_norm": 0.5301211476325989,
+      "learning_rate": 0.0001755711775043937,
+      "loss": 0.886186957359314,
+      "step": 1000
+    },
+    {
+      "epoch": 0.4219409282700422,
+      "eval_loss": 0.9487298727035522,
+      "eval_runtime": 689.4288,
+      "eval_samples_per_second": 3.056,
+      "eval_steps_per_second": 3.056,
+      "step": 1000
+    },
+    {
+      "epoch": 0.42278481012658226,
+      "grad_norm": 0.47876957058906555,
+      "learning_rate": 0.00017592267135325133,
+      "loss": 0.8814021348953247,
+      "step": 1002
+    },
+    {
+      "epoch": 0.42362869198312236,
+      "grad_norm": 0.4929780960083008,
+      "learning_rate": 0.00017627416520210896,
+      "loss": 0.8295068740844727,
+      "step": 1004
+    },
+    {
+      "epoch": 0.42447257383966247,
+      "grad_norm": 0.41888436675071716,
+      "learning_rate": 0.00017662565905096663,
+      "loss": 0.8364827036857605,
+      "step": 1006
+    },
+    {
+      "epoch": 0.4253164556962025,
+      "grad_norm": 0.5175151824951172,
+      "learning_rate": 0.00017697715289982427,
+      "loss": 1.0167189836502075,
+      "step": 1008
+    },
+    {
+      "epoch": 0.42616033755274263,
+      "grad_norm": 0.4815356135368347,
+      "learning_rate": 0.0001773286467486819,
+      "loss": 0.8460752964019775,
+      "step": 1010
+    },
+    {
+      "epoch": 0.4270042194092827,
+      "grad_norm": 0.5210875272750854,
+      "learning_rate": 0.00017768014059753954,
+      "loss": 0.8595574498176575,
+      "step": 1012
+    },
+    {
+      "epoch": 0.4278481012658228,
+      "grad_norm": 0.4400486946105957,
+      "learning_rate": 0.00017803163444639718,
+      "loss": 0.8764723539352417,
+      "step": 1014
+    },
+    {
+      "epoch": 0.4286919831223629,
+      "grad_norm": 0.5282127857208252,
+      "learning_rate": 0.00017838312829525482,
+      "loss": 0.9706798791885376,
+      "step": 1016
+    },
+    {
+      "epoch": 0.42953586497890295,
+      "grad_norm": 0.47523441910743713,
+      "learning_rate": 0.00017873462214411249,
+      "loss": 0.8912002444267273,
+      "step": 1018
+    },
+    {
+      "epoch": 0.43037974683544306,
+      "grad_norm": 0.49640706181526184,
+      "learning_rate": 0.00017908611599297012,
+      "loss": 0.8835636973381042,
+      "step": 1020
+    },
+    {
+      "epoch": 0.4312236286919831,
+      "grad_norm": 0.5253039002418518,
+      "learning_rate": 0.0001794376098418278,
+      "loss": 0.8711735606193542,
+      "step": 1022
+    },
+    {
+      "epoch": 0.4320675105485232,
+      "grad_norm": 0.49285009503364563,
+      "learning_rate": 0.00017978910369068543,
+      "loss": 0.9064869284629822,
+      "step": 1024
+    },
+    {
+      "epoch": 0.43291139240506327,
+      "grad_norm": 0.48758041858673096,
+      "learning_rate": 0.00018014059753954307,
+      "loss": 0.8993359208106995,
+      "step": 1026
+    },
+    {
+      "epoch": 0.4337552742616034,
+      "grad_norm": 0.4815461039543152,
+      "learning_rate": 0.0001804920913884007,
+      "loss": 0.8747937679290771,
+      "step": 1028
+    },
+    {
+      "epoch": 0.4345991561181435,
+      "grad_norm": 0.4930349886417389,
+      "learning_rate": 0.00018084358523725834,
+      "loss": 0.8879084587097168,
+      "step": 1030
+    },
+    {
+      "epoch": 0.43544303797468353,
+      "grad_norm": 0.49303027987480164,
+      "learning_rate": 0.00018119507908611598,
+      "loss": 1.0023083686828613,
+      "step": 1032
+    },
+    {
+      "epoch": 0.43628691983122364,
+      "grad_norm": 0.5312249064445496,
+      "learning_rate": 0.00018154657293497365,
+      "loss": 0.938680112361908,
+      "step": 1034
+    },
+    {
+      "epoch": 0.4371308016877637,
+      "grad_norm": 0.5174582004547119,
+      "learning_rate": 0.0001818980667838313,
+      "loss": 0.854195773601532,
+      "step": 1036
+    },
+    {
+      "epoch": 0.4379746835443038,
+      "grad_norm": 0.5452545881271362,
+      "learning_rate": 0.00018224956063268895,
+      "loss": 0.9059375524520874,
+      "step": 1038
+    },
+    {
+      "epoch": 0.4388185654008439,
+      "grad_norm": 0.5480839014053345,
+      "learning_rate": 0.0001826010544815466,
+      "loss": 0.9708920121192932,
+      "step": 1040
+    },
+    {
+      "epoch": 0.43966244725738396,
+      "grad_norm": 0.44693151116371155,
+      "learning_rate": 0.00018295254833040423,
+      "loss": 0.8561046123504639,
+      "step": 1042
+    },
+    {
+      "epoch": 0.44050632911392407,
+      "grad_norm": 0.4024234116077423,
+      "learning_rate": 0.00018330404217926187,
+      "loss": 0.8811968564987183,
+      "step": 1044
+    },
+    {
+      "epoch": 0.4413502109704641,
+      "grad_norm": 0.5586408376693726,
+      "learning_rate": 0.0001836555360281195,
+      "loss": 0.978068470954895,
+      "step": 1046
+    },
+    {
+      "epoch": 0.4421940928270042,
+      "grad_norm": 0.4918624758720398,
+      "learning_rate": 0.00018400702987697714,
+      "loss": 0.8640981316566467,
+      "step": 1048
+    },
+    {
+      "epoch": 0.4430379746835443,
+      "grad_norm": 0.4335230588912964,
+      "learning_rate": 0.0001843585237258348,
+      "loss": 0.9077964425086975,
+      "step": 1050
+    },
+    {
+      "epoch": 0.4438818565400844,
+      "grad_norm": 0.5275123715400696,
+      "learning_rate": 0.00018471001757469247,
+      "loss": 0.960682213306427,
+      "step": 1052
+    },
+    {
+      "epoch": 0.4447257383966245,
+      "grad_norm": 0.5144415497779846,
+      "learning_rate": 0.0001850615114235501,
+      "loss": 0.8549577593803406,
+      "step": 1054
+    },
+    {
+      "epoch": 0.44556962025316454,
+      "grad_norm": 0.5166662931442261,
+      "learning_rate": 0.00018541300527240775,
+      "loss": 0.9320827126502991,
+      "step": 1056
+    },
+    {
+      "epoch": 0.44641350210970465,
+      "grad_norm": 0.5011980533599854,
+      "learning_rate": 0.0001857644991212654,
+      "loss": 0.9632431864738464,
+      "step": 1058
+    },
+    {
+      "epoch": 0.4472573839662447,
+      "grad_norm": 0.49312469363212585,
+      "learning_rate": 0.00018611599297012303,
+      "loss": 0.9442946910858154,
+      "step": 1060
+    },
+    {
+      "epoch": 0.4481012658227848,
+      "grad_norm": 0.49958568811416626,
+      "learning_rate": 0.00018646748681898066,
+      "loss": 0.952802300453186,
+      "step": 1062
+    },
+    {
+      "epoch": 0.4489451476793249,
+      "grad_norm": 0.49827462434768677,
+      "learning_rate": 0.00018681898066783833,
+      "loss": 0.9630650877952576,
+      "step": 1064
+    },
+    {
+      "epoch": 0.44978902953586497,
+      "grad_norm": 0.523980438709259,
+      "learning_rate": 0.00018717047451669597,
+      "loss": 0.9081395268440247,
+      "step": 1066
+    },
+    {
+      "epoch": 0.4506329113924051,
+      "grad_norm": 0.5108568668365479,
+      "learning_rate": 0.0001875219683655536,
+      "loss": 0.9398958683013916,
+      "step": 1068
+    },
+    {
+      "epoch": 0.45147679324894513,
+      "grad_norm": 0.4453965127468109,
+      "learning_rate": 0.00018787346221441127,
+      "loss": 0.93592768907547,
+      "step": 1070
+    },
+    {
+      "epoch": 0.45232067510548524,
+      "grad_norm": 0.4675683081150055,
+      "learning_rate": 0.0001882249560632689,
+      "loss": 0.8879633545875549,
+      "step": 1072
+    },
+    {
+      "epoch": 0.4531645569620253,
+      "grad_norm": 0.47398847341537476,
+      "learning_rate": 0.00018857644991212655,
+      "loss": 0.9966004490852356,
+      "step": 1074
+    },
+    {
+      "epoch": 0.4540084388185654,
+      "grad_norm": 0.46053192019462585,
+      "learning_rate": 0.0001889279437609842,
+      "loss": 0.9184179902076721,
+      "step": 1076
+    },
+    {
+      "epoch": 0.4548523206751055,
+      "grad_norm": 0.5601398348808289,
+      "learning_rate": 0.00018927943760984182,
+      "loss": 0.9434974193572998,
+      "step": 1078
+    },
+    {
+      "epoch": 0.45569620253164556,
+      "grad_norm": 0.48422637581825256,
+      "learning_rate": 0.0001896309314586995,
+      "loss": 0.9522465467453003,
+      "step": 1080
+    },
+    {
+      "epoch": 0.45654008438818566,
+      "grad_norm": 0.46280911564826965,
+      "learning_rate": 0.00018998242530755713,
+      "loss": 0.9315434694290161,
+      "step": 1082
+    },
+    {
+      "epoch": 0.4573839662447257,
+      "grad_norm": 0.4658683240413666,
+      "learning_rate": 0.00019033391915641477,
+      "loss": 0.8165783286094666,
+      "step": 1084
+    },
+    {
+      "epoch": 0.4582278481012658,
+      "grad_norm": 0.4192182123661041,
+      "learning_rate": 0.0001906854130052724,
+      "loss": 0.8024274110794067,
+      "step": 1086
+    },
+    {
+      "epoch": 0.45907172995780593,
+      "grad_norm": 0.4178735613822937,
+      "learning_rate": 0.00019103690685413007,
+      "loss": 0.8620653748512268,
+      "step": 1088
+    },
+    {
+      "epoch": 0.459915611814346,
+      "grad_norm": 0.4488574266433716,
+      "learning_rate": 0.0001913884007029877,
+      "loss": 0.9121530055999756,
+      "step": 1090
+    },
+    {
+      "epoch": 0.4607594936708861,
+      "grad_norm": 0.5164965987205505,
+      "learning_rate": 0.00019173989455184535,
+      "loss": 0.9496700763702393,
+      "step": 1092
+    },
+    {
+      "epoch": 0.46160337552742614,
+      "grad_norm": 0.41563132405281067,
+      "learning_rate": 0.00019209138840070299,
+      "loss": 0.8757708668708801,
+      "step": 1094
+    },
+    {
+      "epoch": 0.46244725738396625,
+      "grad_norm": 2.3222429752349854,
+      "learning_rate": 0.00019244288224956065,
+      "loss": 0.8406533002853394,
+      "step": 1096
+    },
+    {
+      "epoch": 0.46329113924050636,
+      "grad_norm": 0.4813845157623291,
+      "learning_rate": 0.0001927943760984183,
+      "loss": 0.9459465742111206,
+      "step": 1098
+    },
+    {
+      "epoch": 0.4641350210970464,
+      "grad_norm": 0.6233882308006287,
+      "learning_rate": 0.00019314586994727593,
+      "loss": 0.9376904368400574,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4641350210970464,
+      "eval_loss": 0.9357889294624329,
+      "eval_runtime": 676.9573,
+      "eval_samples_per_second": 3.112,
+      "eval_steps_per_second": 3.112,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4649789029535865,
+      "grad_norm": 0.5125579237937927,
+      "learning_rate": 0.00019349736379613357,
+      "loss": 0.8998825550079346,
+      "step": 1102
+    },
+    {
+      "epoch": 0.46582278481012657,
+      "grad_norm": 0.4534320831298828,
+      "learning_rate": 0.00019384885764499123,
+      "loss": 0.9213768839836121,
+      "step": 1104
+    },
+    {
+      "epoch": 0.4666666666666667,
+      "grad_norm": 0.4715143144130707,
+      "learning_rate": 0.00019420035149384887,
+      "loss": 0.9739661812782288,
+      "step": 1106
+    },
+    {
+      "epoch": 0.4675105485232067,
+      "grad_norm": 0.45529672503471375,
+      "learning_rate": 0.0001945518453427065,
+      "loss": 0.9113216400146484,
+      "step": 1108
+    },
+    {
+      "epoch": 0.46835443037974683,
+      "grad_norm": 0.48199015855789185,
+      "learning_rate": 0.00019490333919156417,
+      "loss": 0.92528235912323,
+      "step": 1110
+    },
+    {
+      "epoch": 0.46919831223628694,
+      "grad_norm": 0.4425188899040222,
+      "learning_rate": 0.0001952548330404218,
+      "loss": 0.8612716197967529,
+      "step": 1112
+    },
+    {
+      "epoch": 0.470042194092827,
+      "grad_norm": 0.4311593770980835,
+      "learning_rate": 0.00019560632688927945,
+      "loss": 0.8901699185371399,
+      "step": 1114
+    },
+    {
+      "epoch": 0.4708860759493671,
+      "grad_norm": 0.497806578874588,
+      "learning_rate": 0.0001959578207381371,
+      "loss": 0.9271994829177856,
+      "step": 1116
+    },
+    {
+      "epoch": 0.47172995780590715,
+      "grad_norm": 0.47149473428726196,
+      "learning_rate": 0.00019630931458699473,
+      "loss": 0.8740925788879395,
+      "step": 1118
+    },
+    {
+      "epoch": 0.47257383966244726,
+      "grad_norm": 0.5029966235160828,
+      "learning_rate": 0.00019666080843585236,
+      "loss": 0.9190115928649902,
+      "step": 1120
+    },
+    {
+      "epoch": 0.47341772151898737,
+      "grad_norm": 0.47972601652145386,
+      "learning_rate": 0.00019701230228471003,
+      "loss": 1.0043057203292847,
+      "step": 1122
+    },
+    {
+      "epoch": 0.4742616033755274,
+      "grad_norm": 0.45096471905708313,
+      "learning_rate": 0.00019736379613356767,
+      "loss": 0.9472925066947937,
+      "step": 1124
+    },
+    {
+      "epoch": 0.4751054852320675,
+      "grad_norm": 0.459852933883667,
+      "learning_rate": 0.00019771528998242533,
+      "loss": 0.8043124079704285,
+      "step": 1126
+    },
+    {
+      "epoch": 0.4759493670886076,
+      "grad_norm": 0.5313422679901123,
+      "learning_rate": 0.00019806678383128297,
+      "loss": 0.9662142395973206,
+      "step": 1128
+    },
+    {
+      "epoch": 0.4767932489451477,
+      "grad_norm": 0.43474531173706055,
+      "learning_rate": 0.0001984182776801406,
+      "loss": 0.8691151738166809,
+      "step": 1130
+    },
+    {
+      "epoch": 0.47763713080168774,
+      "grad_norm": 0.5232312083244324,
+      "learning_rate": 0.00019876977152899825,
+      "loss": 0.9024254083633423,
+      "step": 1132
+    },
+    {
+      "epoch": 0.47848101265822784,
+      "grad_norm": 0.47671905159950256,
+      "learning_rate": 0.0001991212653778559,
+      "loss": 0.8996873497962952,
+      "step": 1134
+    },
+    {
+      "epoch": 0.47932489451476795,
+      "grad_norm": 0.42289480566978455,
+      "learning_rate": 0.00019947275922671353,
+      "loss": 0.8336917757987976,
+      "step": 1136
+    },
+    {
+      "epoch": 0.480168776371308,
+      "grad_norm": 0.4700844883918762,
+      "learning_rate": 0.0001998242530755712,
+      "loss": 0.8943206071853638,
+      "step": 1138
+    },
+    {
+      "epoch": 0.4810126582278481,
+      "grad_norm": 0.5341399312019348,
+      "learning_rate": 0.00019999999711649004,
+      "loss": 0.9340365529060364,
+      "step": 1140
+    },
+    {
+      "epoch": 0.48185654008438816,
+      "grad_norm": 0.46169522404670715,
+      "learning_rate": 0.00019999997404841123,
+      "loss": 0.8241778016090393,
+      "step": 1142
+    },
+    {
+      "epoch": 0.48270042194092827,
+      "grad_norm": 0.5475223064422607,
+      "learning_rate": 0.00019999992791225896,
+      "loss": 1.0096158981323242,
+      "step": 1144
+    },
+    {
+      "epoch": 0.4835443037974684,
+      "grad_norm": 0.524641215801239,
+      "learning_rate": 0.00019999985870804385,
+      "loss": 0.9650378227233887,
+      "step": 1146
+    },
+    {
+      "epoch": 0.48438818565400843,
+      "grad_norm": 0.4326174855232239,
+      "learning_rate": 0.00019999976643578186,
+      "loss": 0.9003009796142578,
+      "step": 1148
+    },
+    {
+      "epoch": 0.48523206751054854,
+      "grad_norm": 0.49034059047698975,
+      "learning_rate": 0.0001999996510954943,
+      "loss": 0.9793432950973511,
+      "step": 1150
+    },
+    {
+      "epoch": 0.4860759493670886,
+      "grad_norm": 0.4760092496871948,
+      "learning_rate": 0.00019999951268720776,
+      "loss": 0.8793007731437683,
+      "step": 1152
+    },
+    {
+      "epoch": 0.4869198312236287,
+      "grad_norm": 0.44606879353523254,
+      "learning_rate": 0.00019999935121095417,
+      "loss": 0.888630747795105,
+      "step": 1154
+    },
+    {
+      "epoch": 0.4877637130801688,
+      "grad_norm": 0.5030332207679749,
+      "learning_rate": 0.0001999991666667708,
+      "loss": 0.9660000801086426,
+      "step": 1156
+    },
+    {
+      "epoch": 0.48860759493670886,
+      "grad_norm": 0.4295555055141449,
+      "learning_rate": 0.00019999895905470014,
+      "loss": 0.824654757976532,
+      "step": 1158
+    },
+    {
+      "epoch": 0.48945147679324896,
+      "grad_norm": 0.392167866230011,
+      "learning_rate": 0.0001999987283747902,
+      "loss": 0.8159562349319458,
+      "step": 1160
+    },
+    {
+      "epoch": 0.490295358649789,
+      "grad_norm": 0.484611839056015,
+      "learning_rate": 0.00019999847462709412,
+      "loss": 0.9630686044692993,
+      "step": 1162
+    },
+    {
+      "epoch": 0.4911392405063291,
+      "grad_norm": 0.45891445875167847,
+      "learning_rate": 0.00019999819781167042,
+      "loss": 0.8396129608154297,
+      "step": 1164
+    },
+    {
+      "epoch": 0.4919831223628692,
+      "grad_norm": 0.5553452968597412,
+      "learning_rate": 0.00019999789792858304,
+      "loss": 0.8194513916969299,
+      "step": 1166
+    },
+    {
+      "epoch": 0.4928270042194093,
+      "grad_norm": 0.38998672366142273,
+      "learning_rate": 0.00019999757497790106,
+      "loss": 0.9268721342086792,
+      "step": 1168
+    },
+    {
+      "epoch": 0.4936708860759494,
+      "grad_norm": 0.45445355772972107,
+      "learning_rate": 0.00019999722895969904,
+      "loss": 0.9204684495925903,
+      "step": 1170
+    },
+    {
+      "epoch": 0.49451476793248944,
+      "grad_norm": 0.484225332736969,
+      "learning_rate": 0.00019999685987405678,
+      "loss": 0.9270301461219788,
+      "step": 1172
+    },
+    {
+      "epoch": 0.49535864978902955,
+      "grad_norm": 0.45215415954589844,
+      "learning_rate": 0.00019999646772105942,
+      "loss": 0.8782645463943481,
+      "step": 1174
+    },
+    {
+      "epoch": 0.4962025316455696,
+      "grad_norm": 0.41578832268714905,
+      "learning_rate": 0.00019999605250079744,
+      "loss": 0.873112678527832,
+      "step": 1176
+    },
+    {
+      "epoch": 0.4970464135021097,
+      "grad_norm": 0.4632788300514221,
+      "learning_rate": 0.0001999956142133666,
+      "loss": 0.9593189358711243,
+      "step": 1178
+    },
+    {
+      "epoch": 0.4978902953586498,
+      "grad_norm": 0.5007622838020325,
+      "learning_rate": 0.000199995152858868,
+      "loss": 0.9215621948242188,
+      "step": 1180
+    },
+    {
+      "epoch": 0.49873417721518987,
+      "grad_norm": 0.5201935768127441,
+      "learning_rate": 0.0001999946684374081,
+      "loss": 0.8964219689369202,
+      "step": 1182
+    },
+    {
+      "epoch": 0.49957805907173,
+      "grad_norm": 0.5053967237472534,
+      "learning_rate": 0.0001999941609490986,
+      "loss": 0.8528663516044617,
+      "step": 1184
+    },
+    {
+      "epoch": 0.5004219409282701,
+      "grad_norm": 0.44397974014282227,
+      "learning_rate": 0.0001999936303940566,
+      "loss": 0.8214734792709351,
+      "step": 1186
+    },
+    {
+      "epoch": 0.5012658227848101,
+      "grad_norm": 0.4563101530075073,
+      "learning_rate": 0.0001999930767724045,
+      "loss": 0.8882166743278503,
+      "step": 1188
+    },
+    {
+      "epoch": 0.5021097046413502,
+      "grad_norm": 0.4666728675365448,
+      "learning_rate": 0.00019999250008426997,
+      "loss": 0.8882588148117065,
+      "step": 1190
+    },
+    {
+      "epoch": 0.5029535864978903,
+      "grad_norm": 0.4423629939556122,
+      "learning_rate": 0.00019999190032978607,
+      "loss": 0.8353691697120667,
+      "step": 1192
+    },
+    {
+      "epoch": 0.5037974683544304,
+      "grad_norm": 0.4514595866203308,
+      "learning_rate": 0.00019999127750909118,
+      "loss": 0.8721219301223755,
+      "step": 1194
+    },
+    {
+      "epoch": 0.5046413502109705,
+      "grad_norm": 0.48469996452331543,
+      "learning_rate": 0.00019999063162232886,
+      "loss": 0.902795135974884,
+      "step": 1196
+    },
+    {
+      "epoch": 0.5054852320675105,
+      "grad_norm": 0.4931983947753906,
+      "learning_rate": 0.00019998996266964823,
+      "loss": 0.8843175768852234,
+      "step": 1198
+    },
+    {
+      "epoch": 0.5063291139240507,
+      "grad_norm": 0.507408082485199,
+      "learning_rate": 0.00019998927065120357,
+      "loss": 0.9179208278656006,
+      "step": 1200
+    },
+    {
+      "epoch": 0.5063291139240507,
+      "eval_loss": 0.9224098324775696,
+      "eval_runtime": 669.7542,
+      "eval_samples_per_second": 3.146,
+      "eval_steps_per_second": 3.146,
+      "step": 1200
+    },
+    {
+      "epoch": 0.5071729957805907,
+      "grad_norm": 0.4024188220500946,
+      "learning_rate": 0.00019998855556715447,
+      "loss": 0.9078981876373291,
+      "step": 1202
+    },
+    {
+      "epoch": 0.5080168776371308,
+      "grad_norm": 0.6558667421340942,
+      "learning_rate": 0.00019998781741766594,
+      "loss": 0.8089252710342407,
+      "step": 1204
+    },
+    {
+      "epoch": 0.5088607594936709,
+      "grad_norm": 0.45128145813941956,
+      "learning_rate": 0.00019998705620290823,
+      "loss": 0.9316248297691345,
+      "step": 1206
+    },
+    {
+      "epoch": 0.509704641350211,
+      "grad_norm": 0.4733511507511139,
+      "learning_rate": 0.00019998627192305694,
+      "loss": 0.9047867655754089,
+      "step": 1208
+    },
+    {
+      "epoch": 0.510548523206751,
+      "grad_norm": 0.4026021659374237,
+      "learning_rate": 0.00019998546457829298,
+      "loss": 0.8507166504859924,
+      "step": 1210
+    },
+    {
+      "epoch": 0.5113924050632911,
+      "grad_norm": 0.43416184186935425,
+      "learning_rate": 0.00019998463416880262,
+      "loss": 0.8490090370178223,
+      "step": 1212
+    },
+    {
+      "epoch": 0.5122362869198313,
+      "grad_norm": 0.3678364157676697,
+      "learning_rate": 0.0001999837806947774,
+      "loss": 0.7730492949485779,
+      "step": 1214
+    },
+    {
+      "epoch": 0.5130801687763713,
+      "grad_norm": 0.49042677879333496,
+      "learning_rate": 0.00019998290415641415,
+      "loss": 0.9625269174575806,
+      "step": 1216
+    },
+    {
+      "epoch": 0.5139240506329114,
+      "grad_norm": 0.4576701819896698,
+      "learning_rate": 0.00019998200455391516,
+      "loss": 0.8961732983589172,
+      "step": 1218
+    },
+    {
+      "epoch": 0.5147679324894515,
+      "grad_norm": 0.4465518593788147,
+      "learning_rate": 0.00019998108188748793,
+      "loss": 0.8508996367454529,
+      "step": 1220
+    },
+    {
+      "epoch": 0.5156118143459916,
+      "grad_norm": 0.4267960488796234,
+      "learning_rate": 0.00019998013615734524,
+      "loss": 0.8664930462837219,
+      "step": 1222
+    },
+    {
+      "epoch": 0.5164556962025316,
+      "grad_norm": 0.39936602115631104,
+      "learning_rate": 0.0001999791673637053,
+      "loss": 0.8841317892074585,
+      "step": 1224
+    },
+    {
+      "epoch": 0.5172995780590718,
+      "grad_norm": 0.4442414343357086,
+      "learning_rate": 0.0001999781755067916,
+      "loss": 0.8601276874542236,
+      "step": 1226
+    },
+    {
+      "epoch": 0.5181434599156118,
+      "grad_norm": 0.4397028088569641,
+      "learning_rate": 0.00019997716058683292,
+      "loss": 0.8377046585083008,
+      "step": 1228
+    },
+    {
+      "epoch": 0.5189873417721519,
+      "grad_norm": 0.5327648520469666,
+      "learning_rate": 0.0001999761226040634,
+      "loss": 0.9853615164756775,
+      "step": 1230
+    },
+    {
+      "epoch": 0.5198312236286919,
+      "grad_norm": 0.40931421518325806,
+      "learning_rate": 0.00019997506155872244,
+      "loss": 0.9235715866088867,
+      "step": 1232
+    },
+    {
+      "epoch": 0.5206751054852321,
+      "grad_norm": 0.4599161148071289,
+      "learning_rate": 0.00019997397745105487,
+      "loss": 0.868396520614624,
+      "step": 1234
+    },
+    {
+      "epoch": 0.5215189873417722,
+      "grad_norm": 0.4813833236694336,
+      "learning_rate": 0.0001999728702813107,
+      "loss": 0.9946733117103577,
+      "step": 1236
+    },
+    {
+      "epoch": 0.5223628691983122,
+      "grad_norm": 0.43518269062042236,
+      "learning_rate": 0.00019997174004974543,
+      "loss": 0.8608635067939758,
+      "step": 1238
+    },
+    {
+      "epoch": 0.5232067510548524,
+      "grad_norm": 0.45646214485168457,
+      "learning_rate": 0.0001999705867566197,
+      "loss": 0.819354772567749,
+      "step": 1240
+    },
+    {
+      "epoch": 0.5240506329113924,
+      "grad_norm": 0.46834197640419006,
+      "learning_rate": 0.00019996941040219954,
+      "loss": 0.8934658765792847,
+      "step": 1242
+    },
+    {
+      "epoch": 0.5248945147679325,
+      "grad_norm": 0.46065032482147217,
+      "learning_rate": 0.0001999682109867564,
+      "loss": 0.8624778985977173,
+      "step": 1244
+    },
+    {
+      "epoch": 0.5257383966244725,
+      "grad_norm": 0.4506741166114807,
+      "learning_rate": 0.00019996698851056688,
+      "loss": 0.8540882468223572,
+      "step": 1246
+    },
+    {
+      "epoch": 0.5265822784810127,
+      "grad_norm": 0.4550519287586212,
+      "learning_rate": 0.00019996574297391302,
+      "loss": 0.9003006219863892,
+      "step": 1248
+    },
+    {
+      "epoch": 0.5274261603375527,
+      "grad_norm": 0.4097813367843628,
+      "learning_rate": 0.00019996447437708214,
+      "loss": 0.8490248918533325,
+      "step": 1250
+    },
+    {
+      "epoch": 0.5282700421940928,
+      "grad_norm": 0.4375711679458618,
+      "learning_rate": 0.00019996318272036688,
+      "loss": 0.9204569458961487,
+      "step": 1252
+    },
+    {
+      "epoch": 0.529113924050633,
+      "grad_norm": 0.4335198998451233,
+      "learning_rate": 0.00019996186800406518,
+      "loss": 0.8831789493560791,
+      "step": 1254
+    },
+    {
+      "epoch": 0.529957805907173,
+      "grad_norm": 0.4840933084487915,
+      "learning_rate": 0.00019996053022848035,
+      "loss": 0.8578064441680908,
+      "step": 1256
+    },
+    {
+      "epoch": 0.5308016877637131,
+      "grad_norm": 0.48911216855049133,
+      "learning_rate": 0.00019995916939392097,
+      "loss": 0.9235416650772095,
+      "step": 1258
+    },
+    {
+      "epoch": 0.5316455696202531,
+      "grad_norm": 0.41106730699539185,
+      "learning_rate": 0.00019995778550070098,
+      "loss": 0.7897764444351196,
+      "step": 1260
+    },
+    {
+      "epoch": 0.5324894514767933,
+      "grad_norm": 0.42233356833457947,
+      "learning_rate": 0.00019995637854913957,
+      "loss": 0.9392989873886108,
+      "step": 1262
+    },
+    {
+      "epoch": 0.5333333333333333,
+      "grad_norm": 0.39145180583000183,
+      "learning_rate": 0.0001999549485395613,
+      "loss": 0.9130091667175293,
+      "step": 1264
+    },
+    {
+      "epoch": 0.5341772151898734,
+      "grad_norm": 0.42780739068984985,
+      "learning_rate": 0.00019995349547229614,
+      "loss": 0.9690561890602112,
+      "step": 1266
+    },
+    {
+      "epoch": 0.5350210970464135,
+      "grad_norm": 0.4823327362537384,
+      "learning_rate": 0.0001999520193476792,
+      "loss": 0.9990131258964539,
+      "step": 1268
+    },
+    {
+      "epoch": 0.5358649789029536,
+      "grad_norm": 0.4669715166091919,
+      "learning_rate": 0.00019995052016605097,
+      "loss": 0.8343052864074707,
+      "step": 1270
+    },
+    {
+      "epoch": 0.5367088607594936,
+      "grad_norm": 0.4024743437767029,
+      "learning_rate": 0.0001999489979277574,
+      "loss": 0.9605970978736877,
+      "step": 1272
+    },
+    {
+      "epoch": 0.5375527426160338,
+      "grad_norm": 0.42063432931900024,
+      "learning_rate": 0.0001999474526331495,
+      "loss": 0.8164438009262085,
+      "step": 1274
+    },
+    {
+      "epoch": 0.5383966244725739,
+      "grad_norm": 0.47953495383262634,
+      "learning_rate": 0.00019994588428258383,
+      "loss": 0.879767656326294,
+      "step": 1276
+    },
+    {
+      "epoch": 0.5392405063291139,
+      "grad_norm": 0.39668557047843933,
+      "learning_rate": 0.00019994429287642214,
+      "loss": 0.7658726572990417,
+      "step": 1278
+    },
+    {
+      "epoch": 0.540084388185654,
+      "grad_norm": 0.5136121511459351,
+      "learning_rate": 0.00019994267841503155,
+      "loss": 0.9477384686470032,
+      "step": 1280
+    },
+    {
+      "epoch": 0.5409282700421941,
+      "grad_norm": 0.4039286673069,
+      "learning_rate": 0.00019994104089878452,
+      "loss": 0.7883896827697754,
+      "step": 1282
+    },
+    {
+      "epoch": 0.5417721518987342,
+      "grad_norm": 0.4425487816333771,
+      "learning_rate": 0.0001999393803280587,
+      "loss": 0.8526129126548767,
+      "step": 1284
+    },
+    {
+      "epoch": 0.5426160337552742,
+      "grad_norm": 0.4544559121131897,
+      "learning_rate": 0.00019993769670323725,
+      "loss": 0.8694683909416199,
+      "step": 1286
+    },
+    {
+      "epoch": 0.5434599156118144,
+      "grad_norm": 0.42872917652130127,
+      "learning_rate": 0.0001999359900247085,
+      "loss": 0.8211527466773987,
+      "step": 1288
+    },
+    {
+      "epoch": 0.5443037974683544,
+      "grad_norm": 0.41549673676490784,
+      "learning_rate": 0.00019993426029286617,
+      "loss": 0.7967932224273682,
+      "step": 1290
+    },
+    {
+      "epoch": 0.5451476793248945,
+      "grad_norm": 0.42888355255126953,
+      "learning_rate": 0.00019993250750810926,
+      "loss": 0.8291563987731934,
+      "step": 1292
+    },
+    {
+      "epoch": 0.5459915611814345,
+      "grad_norm": 0.44985485076904297,
+      "learning_rate": 0.0001999307316708421,
+      "loss": 0.8787803649902344,
+      "step": 1294
+    },
+    {
+      "epoch": 0.5468354430379747,
+      "grad_norm": 0.43326008319854736,
+      "learning_rate": 0.00019992893278147436,
+      "loss": 0.8375519514083862,
+      "step": 1296
+    },
+    {
+      "epoch": 0.5476793248945148,
+      "grad_norm": 0.406556099653244,
+      "learning_rate": 0.00019992711084042096,
+      "loss": 0.8538051247596741,
+      "step": 1298
+    },
+    {
+      "epoch": 0.5485232067510548,
+      "grad_norm": 0.46090877056121826,
+      "learning_rate": 0.00019992526584810226,
+      "loss": 0.9722132086753845,
+      "step": 1300
+    },
+    {
+      "epoch": 0.5485232067510548,
+      "eval_loss": 0.9068717360496521,
+      "eval_runtime": 680.7718,
+      "eval_samples_per_second": 3.095,
+      "eval_steps_per_second": 3.095,
+      "step": 1300
+    },
+    {
+      "epoch": 0.549367088607595,
+      "grad_norm": 0.44932207465171814,
+      "learning_rate": 0.00019992339780494382,
+      "loss": 0.8474152684211731,
+      "step": 1302
+    },
+    {
+      "epoch": 0.550210970464135,
+      "grad_norm": 0.5185276865959167,
+      "learning_rate": 0.0001999215067113766,
+      "loss": 0.8440775871276855,
+      "step": 1304
+    },
+    {
+      "epoch": 0.5510548523206751,
+      "grad_norm": 0.43811365962028503,
+      "learning_rate": 0.0001999195925678368,
+      "loss": 0.8848010897636414,
+      "step": 1306
+    },
+    {
+      "epoch": 0.5518987341772152,
+      "grad_norm": 0.4861660301685333,
+      "learning_rate": 0.00019991765537476596,
+      "loss": 0.8265100121498108,
+      "step": 1308
+    },
+    {
+      "epoch": 0.5527426160337553,
+      "grad_norm": 0.5068721771240234,
+      "learning_rate": 0.000199915695132611,
+      "loss": 0.9327963590621948,
+      "step": 1310
+    },
+    {
+      "epoch": 0.5535864978902953,
+      "grad_norm": 0.4902805685997009,
+      "learning_rate": 0.0001999137118418241,
+      "loss": 0.9315155744552612,
+      "step": 1312
+    },
+    {
+      "epoch": 0.5544303797468354,
+      "grad_norm": 0.4530964493751526,
+      "learning_rate": 0.00019991170550286272,
+      "loss": 0.9344852566719055,
+      "step": 1314
+    },
+    {
+      "epoch": 0.5552742616033756,
+      "grad_norm": 0.4106673300266266,
+      "learning_rate": 0.00019990967611618974,
+      "loss": 0.8326173424720764,
+      "step": 1316
+    },
+    {
+      "epoch": 0.5561181434599156,
+      "grad_norm": 0.4439711570739746,
+      "learning_rate": 0.0001999076236822733,
+      "loss": 0.8934909105300903,
+      "step": 1318
+    },
+    {
+      "epoch": 0.5569620253164557,
+      "grad_norm": 0.40245553851127625,
+      "learning_rate": 0.0001999055482015868,
+      "loss": 0.6691107153892517,
+      "step": 1320
+    },
+    {
+      "epoch": 0.5578059071729958,
+      "grad_norm": 0.43904435634613037,
+      "learning_rate": 0.0001999034496746091,
+      "loss": 0.8560307621955872,
+      "step": 1322
+    },
+    {
+      "epoch": 0.5586497890295359,
+      "grad_norm": 0.5131705403327942,
+      "learning_rate": 0.00019990132810182422,
+      "loss": 0.9024442434310913,
+      "step": 1324
+    },
+    {
+      "epoch": 0.5594936708860759,
+      "grad_norm": 0.4539531171321869,
+      "learning_rate": 0.00019989918348372156,
+      "loss": 0.8675428628921509,
+      "step": 1326
+    },
+    {
+      "epoch": 0.560337552742616,
+      "grad_norm": 0.44469985365867615,
+      "learning_rate": 0.00019989701582079591,
+      "loss": 0.8853429555892944,
+      "step": 1328
+    },
+    {
+      "epoch": 0.5611814345991561,
+      "grad_norm": 0.4485546946525574,
+      "learning_rate": 0.00019989482511354725,
+      "loss": 0.9675378799438477,
+      "step": 1330
+    },
+    {
+      "epoch": 0.5620253164556962,
+      "grad_norm": 0.43316468596458435,
+      "learning_rate": 0.00019989261136248097,
+      "loss": 0.8256655335426331,
+      "step": 1332
+    },
+    {
+      "epoch": 0.5628691983122363,
+      "grad_norm": 0.4315780699253082,
+      "learning_rate": 0.00019989037456810772,
+      "loss": 0.7727690935134888,
+      "step": 1334
+    },
+    {
+      "epoch": 0.5637130801687764,
+      "grad_norm": 0.45147502422332764,
+      "learning_rate": 0.00019988811473094348,
+      "loss": 0.9240618348121643,
+      "step": 1336
+    },
+    {
+      "epoch": 0.5645569620253165,
+      "grad_norm": 0.4606908857822418,
+      "learning_rate": 0.00019988583185150957,
+      "loss": 0.8005949258804321,
+      "step": 1338
+    },
+    {
+      "epoch": 0.5654008438818565,
+      "grad_norm": 0.48932451009750366,
+      "learning_rate": 0.00019988352593033255,
+      "loss": 0.9042718410491943,
+      "step": 1340
+    },
+    {
+      "epoch": 0.5662447257383966,
+      "grad_norm": 0.46022218465805054,
+      "learning_rate": 0.00019988119696794443,
+      "loss": 0.8724613785743713,
+      "step": 1342
+    },
+    {
+      "epoch": 0.5670886075949367,
+      "grad_norm": 0.4414621889591217,
+      "learning_rate": 0.00019987884496488242,
+      "loss": 0.8653066158294678,
+      "step": 1344
+    },
+    {
+      "epoch": 0.5679324894514768,
+      "grad_norm": 0.4490342140197754,
+      "learning_rate": 0.0001998764699216891,
+      "loss": 0.8843849301338196,
+      "step": 1346
+    },
+    {
+      "epoch": 0.5687763713080168,
+      "grad_norm": 0.47690069675445557,
+      "learning_rate": 0.00019987407183891232,
+      "loss": 0.8385393023490906,
+      "step": 1348
+    },
+    {
+      "epoch": 0.569620253164557,
+      "grad_norm": 0.41409361362457275,
+      "learning_rate": 0.00019987165071710527,
+      "loss": 0.77492755651474,
+      "step": 1350
+    },
+    {
+      "epoch": 0.570464135021097,
+      "grad_norm": 0.4151647984981537,
+      "learning_rate": 0.0001998692065568265,
+      "loss": 0.8368680477142334,
+      "step": 1352
+    },
+    {
+      "epoch": 0.5713080168776371,
+      "grad_norm": 0.4105067849159241,
+      "learning_rate": 0.0001998667393586398,
+      "loss": 0.833220362663269,
+      "step": 1354
+    },
+    {
+      "epoch": 0.5721518987341773,
+      "grad_norm": 0.44701820611953735,
+      "learning_rate": 0.0001998642491231143,
+      "loss": 0.8041552305221558,
+      "step": 1356
+    },
+    {
+      "epoch": 0.5729957805907173,
+      "grad_norm": 0.4621582329273224,
+      "learning_rate": 0.00019986173585082444,
+      "loss": 0.9052709937095642,
+      "step": 1358
+    },
+    {
+      "epoch": 0.5738396624472574,
+      "grad_norm": 0.4934713840484619,
+      "learning_rate": 0.00019985919954235005,
+      "loss": 0.8036054372787476,
+      "step": 1360
+    },
+    {
+      "epoch": 0.5746835443037974,
+      "grad_norm": 0.4557384252548218,
+      "learning_rate": 0.0001998566401982761,
+      "loss": 0.8340095281600952,
+      "step": 1362
+    },
+    {
+      "epoch": 0.5755274261603376,
+      "grad_norm": 0.5048403739929199,
+      "learning_rate": 0.00019985405781919305,
+      "loss": 0.8703633546829224,
+      "step": 1364
+    },
+    {
+      "epoch": 0.5763713080168776,
+      "grad_norm": 0.4707714319229126,
+      "learning_rate": 0.0001998514524056966,
+      "loss": 0.9115830659866333,
+      "step": 1366
+    },
+    {
+      "epoch": 0.5772151898734177,
+      "grad_norm": 0.4502186179161072,
+      "learning_rate": 0.00019984882395838777,
+      "loss": 0.8603149056434631,
+      "step": 1368
+    },
+    {
+      "epoch": 0.5780590717299579,
+      "grad_norm": 0.4229136109352112,
+      "learning_rate": 0.00019984617247787288,
+      "loss": 0.7482197284698486,
+      "step": 1370
+    },
+    {
+      "epoch": 0.5789029535864979,
+      "grad_norm": 0.49208030104637146,
+      "learning_rate": 0.0001998434979647636,
+      "loss": 0.9071239233016968,
+      "step": 1372
+    },
+    {
+      "epoch": 0.579746835443038,
+      "grad_norm": 0.47393161058425903,
+      "learning_rate": 0.00019984080041967687,
+      "loss": 0.9370644092559814,
+      "step": 1374
+    },
+    {
+      "epoch": 0.580590717299578,
+      "grad_norm": 0.4483324885368347,
+      "learning_rate": 0.00019983807984323492,
+      "loss": 0.8536180257797241,
+      "step": 1376
+    },
+    {
+      "epoch": 0.5814345991561182,
+      "grad_norm": 0.4163796007633209,
+      "learning_rate": 0.00019983533623606543,
+      "loss": 0.7360405325889587,
+      "step": 1378
+    },
+    {
+      "epoch": 0.5822784810126582,
+      "grad_norm": 0.44147446751594543,
+      "learning_rate": 0.00019983256959880122,
+      "loss": 0.8162824511528015,
+      "step": 1380
+    },
+    {
+      "epoch": 0.5831223628691983,
+      "grad_norm": 0.44443491101264954,
+      "learning_rate": 0.0001998297799320805,
+      "loss": 0.9169327616691589,
+      "step": 1382
+    },
+    {
+      "epoch": 0.5839662447257384,
+      "grad_norm": 0.4641949236392975,
+      "learning_rate": 0.00019982696723654686,
+      "loss": 0.8875693678855896,
+      "step": 1384
+    },
+    {
+      "epoch": 0.5848101265822785,
+      "grad_norm": 0.4327974319458008,
+      "learning_rate": 0.00019982413151284906,
+      "loss": 0.7747344970703125,
+      "step": 1386
+    },
+    {
+      "epoch": 0.5856540084388185,
+      "grad_norm": 0.44200772047042847,
+      "learning_rate": 0.0001998212727616413,
+      "loss": 0.8490481972694397,
+      "step": 1388
+    },
+    {
+      "epoch": 0.5864978902953587,
+      "grad_norm": 0.39084959030151367,
+      "learning_rate": 0.000199818390983583,
+      "loss": 0.8014808297157288,
+      "step": 1390
+    },
+    {
+      "epoch": 0.5873417721518988,
+      "grad_norm": 0.3691277801990509,
+      "learning_rate": 0.00019981548617933897,
+      "loss": 0.8650928139686584,
+      "step": 1392
+    },
+    {
+      "epoch": 0.5881856540084388,
+      "grad_norm": 0.480002760887146,
+      "learning_rate": 0.00019981255834957925,
+      "loss": 0.7713267803192139,
+      "step": 1394
+    },
+    {
+      "epoch": 0.5890295358649789,
+      "grad_norm": 0.4397226572036743,
+      "learning_rate": 0.00019980960749497927,
+      "loss": 0.9053656458854675,
+      "step": 1396
+    },
+    {
+      "epoch": 0.589873417721519,
+      "grad_norm": 0.4263727366924286,
+      "learning_rate": 0.00019980663361621973,
+      "loss": 0.8737669587135315,
+      "step": 1398
+    },
+    {
+      "epoch": 0.5907172995780591,
+      "grad_norm": 0.4468817710876465,
+      "learning_rate": 0.0001998036367139866,
+      "loss": 0.9396650791168213,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5907172995780591,
+      "eval_loss": 0.8971880674362183,
+      "eval_runtime": 692.8046,
+      "eval_samples_per_second": 3.041,
+      "eval_steps_per_second": 3.041,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5915611814345991,
+      "grad_norm": 0.4128144383430481,
+      "learning_rate": 0.00019980061678897127,
+      "loss": 0.926134467124939,
+      "step": 1402
+    },
+    {
+      "epoch": 0.5924050632911393,
+      "grad_norm": 0.45524629950523376,
+      "learning_rate": 0.00019979757384187035,
+      "loss": 0.826360285282135,
+      "step": 1404
+    },
+    {
+      "epoch": 0.5932489451476793,
+      "grad_norm": 0.42521336674690247,
+      "learning_rate": 0.0001997945078733858,
+      "loss": 0.8147702813148499,
+      "step": 1406
+    },
+    {
+      "epoch": 0.5940928270042194,
+      "grad_norm": 0.4172525107860565,
+      "learning_rate": 0.00019979141888422489,
+      "loss": 0.7711596488952637,
+      "step": 1408
+    },
+    {
+      "epoch": 0.5949367088607594,
+      "grad_norm": 0.46193623542785645,
+      "learning_rate": 0.00019978830687510013,
+      "loss": 0.7875980138778687,
+      "step": 1410
+    },
+    {
+      "epoch": 0.5957805907172996,
+      "grad_norm": 0.4455825686454773,
+      "learning_rate": 0.00019978517184672946,
+      "loss": 0.8590195178985596,
+      "step": 1412
+    },
+    {
+      "epoch": 0.5966244725738397,
+      "grad_norm": 0.4266505539417267,
+      "learning_rate": 0.00019978201379983608,
+      "loss": 0.9318227767944336,
+      "step": 1414
+    },
+    {
+      "epoch": 0.5974683544303797,
+      "grad_norm": 0.42627251148223877,
+      "learning_rate": 0.00019977883273514843,
+      "loss": 0.8644474744796753,
+      "step": 1416
+    },
+    {
+      "epoch": 0.5983122362869199,
+      "grad_norm": 0.393873393535614,
+      "learning_rate": 0.00019977562865340038,
+      "loss": 0.7760446071624756,
+      "step": 1418
+    },
+    {
+      "epoch": 0.5991561181434599,
+      "grad_norm": 0.44334208965301514,
+      "learning_rate": 0.000199772401555331,
+      "loss": 0.8071584701538086,
+      "step": 1420
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.39390063285827637,
+      "learning_rate": 0.0001997691514416848,
+      "loss": 0.9076889753341675,
+      "step": 1422
+    },
+    {
+      "epoch": 0.60084388185654,
+      "grad_norm": 0.46753421425819397,
+      "learning_rate": 0.00019976587831321144,
+      "loss": 0.8534318804740906,
+      "step": 1424
+    },
+    {
+      "epoch": 0.6016877637130802,
+      "grad_norm": 0.4840681254863739,
+      "learning_rate": 0.00019976258217066598,
+      "loss": 0.9049596190452576,
+      "step": 1426
+    },
+    {
+      "epoch": 0.6025316455696202,
+      "grad_norm": 0.4740557372570038,
+      "learning_rate": 0.00019975926301480878,
+      "loss": 0.8520918488502502,
+      "step": 1428
+    },
+    {
+      "epoch": 0.6033755274261603,
+      "grad_norm": 0.46498584747314453,
+      "learning_rate": 0.00019975592084640553,
+      "loss": 0.8221305012702942,
+      "step": 1430
+    },
+    {
+      "epoch": 0.6042194092827005,
+      "grad_norm": 0.40885692834854126,
+      "learning_rate": 0.0001997525556662272,
+      "loss": 0.8132198452949524,
+      "step": 1432
+    },
+    {
+      "epoch": 0.6050632911392405,
+      "grad_norm": 0.4375821053981781,
+      "learning_rate": 0.00019974916747505006,
+      "loss": 0.870534360408783,
+      "step": 1434
+    },
+    {
+      "epoch": 0.6059071729957806,
+      "grad_norm": 0.4780935049057007,
+      "learning_rate": 0.00019974575627365569,
+      "loss": 0.8825759291648865,
+      "step": 1436
+    },
+    {
+      "epoch": 0.6067510548523207,
+      "grad_norm": 0.4623970687389374,
+      "learning_rate": 0.00019974232206283098,
+      "loss": 0.8187641501426697,
+      "step": 1438
+    },
+    {
+      "epoch": 0.6075949367088608,
+      "grad_norm": 0.42172595858573914,
+      "learning_rate": 0.0001997388648433682,
+      "loss": 0.8569821119308472,
+      "step": 1440
+    },
+    {
+      "epoch": 0.6084388185654008,
+      "grad_norm": 0.49386659264564514,
+      "learning_rate": 0.0001997353846160648,
+      "loss": 0.9173614382743835,
+      "step": 1442
+    },
+    {
+      "epoch": 0.6092827004219409,
+      "grad_norm": 0.47122615575790405,
+      "learning_rate": 0.00019973188138172363,
+      "loss": 0.9444975256919861,
+      "step": 1444
+    },
+    {
+      "epoch": 0.610126582278481,
+      "grad_norm": 0.4831530451774597,
+      "learning_rate": 0.0001997283551411528,
+      "loss": 0.9537245631217957,
+      "step": 1446
+    },
+    {
+      "epoch": 0.6109704641350211,
+      "grad_norm": 0.44460317492485046,
+      "learning_rate": 0.0001997248058951658,
+      "loss": 0.8481367230415344,
+      "step": 1448
+    },
+    {
+      "epoch": 0.6118143459915611,
+      "grad_norm": 0.4315263628959656,
+      "learning_rate": 0.00019972123364458128,
+      "loss": 0.8572371006011963,
+      "step": 1450
+    },
+    {
+      "epoch": 0.6126582278481013,
+      "grad_norm": 0.38559582829475403,
+      "learning_rate": 0.00019971763839022336,
+      "loss": 0.762590765953064,
+      "step": 1452
+    },
+    {
+      "epoch": 0.6135021097046414,
+      "grad_norm": 0.3820290267467499,
+      "learning_rate": 0.00019971402013292138,
+      "loss": 0.7879500389099121,
+      "step": 1454
+    },
+    {
+      "epoch": 0.6143459915611814,
+      "grad_norm": 0.42509570717811584,
+      "learning_rate": 0.00019971037887351,
+      "loss": 0.8696863055229187,
+      "step": 1456
+    },
+    {
+      "epoch": 0.6151898734177215,
+      "grad_norm": 0.46750548481941223,
+      "learning_rate": 0.00019970671461282916,
+      "loss": 0.8425862789154053,
+      "step": 1458
+    },
+    {
+      "epoch": 0.6160337552742616,
+      "grad_norm": 0.3937121033668518,
+      "learning_rate": 0.0001997030273517242,
+      "loss": 0.8028931021690369,
+      "step": 1460
+    },
+    {
+      "epoch": 0.6168776371308017,
+      "grad_norm": 0.4361220896244049,
+      "learning_rate": 0.00019969931709104565,
+      "loss": 0.8662509918212891,
+      "step": 1462
+    },
+    {
+      "epoch": 0.6177215189873417,
+      "grad_norm": 0.4737963378429413,
+      "learning_rate": 0.00019969558383164943,
+      "loss": 0.8962596654891968,
+      "step": 1464
+    },
+    {
+      "epoch": 0.6185654008438819,
+      "grad_norm": 0.3952867388725281,
+      "learning_rate": 0.00019969182757439668,
+      "loss": 0.8630867600440979,
+      "step": 1466
+    },
+    {
+      "epoch": 0.619409282700422,
+      "grad_norm": 0.41688743233680725,
+      "learning_rate": 0.00019968804832015393,
+      "loss": 0.8759240508079529,
+      "step": 1468
+    },
+    {
+      "epoch": 0.620253164556962,
+      "grad_norm": 0.4599224627017975,
+      "learning_rate": 0.00019968424606979298,
+      "loss": 0.8459385633468628,
+      "step": 1470
+    },
+    {
+      "epoch": 0.6210970464135022,
+      "grad_norm": 0.5476765632629395,
+      "learning_rate": 0.00019968042082419094,
+      "loss": 0.8844659328460693,
+      "step": 1472
+    },
+    {
+      "epoch": 0.6219409282700422,
+      "grad_norm": 0.46202269196510315,
+      "learning_rate": 0.00019967657258423022,
+      "loss": 0.860946536064148,
+      "step": 1474
+    },
+    {
+      "epoch": 0.6227848101265823,
+      "grad_norm": 0.39446666836738586,
+      "learning_rate": 0.00019967270135079853,
+      "loss": 0.7587860822677612,
+      "step": 1476
+    },
+    {
+      "epoch": 0.6236286919831223,
+      "grad_norm": 0.4500375986099243,
+      "learning_rate": 0.0001996688071247889,
+      "loss": 0.9143000245094299,
+      "step": 1478
+    },
+    {
+      "epoch": 0.6244725738396625,
+      "grad_norm": 0.4203545153141022,
+      "learning_rate": 0.0001996648899070996,
+      "loss": 0.7865519523620605,
+      "step": 1480
+    },
+    {
+      "epoch": 0.6253164556962025,
+      "grad_norm": 0.45623889565467834,
+      "learning_rate": 0.00019966094969863432,
+      "loss": 0.9192912578582764,
+      "step": 1482
+    },
+    {
+      "epoch": 0.6261603375527426,
+      "grad_norm": 0.4139637053012848,
+      "learning_rate": 0.00019965698650030195,
+      "loss": 0.8459161520004272,
+      "step": 1484
+    },
+    {
+      "epoch": 0.6270042194092827,
+      "grad_norm": 0.4410824179649353,
+      "learning_rate": 0.00019965300031301678,
+      "loss": 0.8159077763557434,
+      "step": 1486
+    },
+    {
+      "epoch": 0.6278481012658228,
+      "grad_norm": 0.43821197748184204,
+      "learning_rate": 0.00019964899113769828,
+      "loss": 0.884467363357544,
+      "step": 1488
+    },
+    {
+      "epoch": 0.6286919831223629,
+      "grad_norm": 0.4692091643810272,
+      "learning_rate": 0.00019964495897527133,
+      "loss": 0.8693601489067078,
+      "step": 1490
+    },
+    {
+      "epoch": 0.6295358649789029,
+      "grad_norm": 0.40715447068214417,
+      "learning_rate": 0.00019964090382666608,
+      "loss": 0.8897743225097656,
+      "step": 1492
+    },
+    {
+      "epoch": 0.6303797468354431,
+      "grad_norm": 0.4123334586620331,
+      "learning_rate": 0.0001996368256928179,
+      "loss": 0.8274733424186707,
+      "step": 1494
+    },
+    {
+      "epoch": 0.6312236286919831,
+      "grad_norm": 0.44347211718559265,
+      "learning_rate": 0.00019963272457466767,
+      "loss": 0.8158749938011169,
+      "step": 1496
+    },
+    {
+      "epoch": 0.6320675105485232,
+      "grad_norm": 0.4700329601764679,
+      "learning_rate": 0.00019962860047316135,
+      "loss": 0.8439569473266602,
+      "step": 1498
+    },
+    {
+      "epoch": 0.6329113924050633,
+      "grad_norm": 0.46415844559669495,
+      "learning_rate": 0.00019962445338925027,
+      "loss": 0.8316822648048401,
+      "step": 1500
+    },
+    {
+      "epoch": 0.6329113924050633,
+      "eval_loss": 0.887488842010498,
+      "eval_runtime": 686.2804,
+      "eval_samples_per_second": 3.07,
+      "eval_steps_per_second": 3.07,
+      "step": 1500
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 14220,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 6,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 5,
+        "early_stopping_threshold": 0.001
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 0
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 8.966840619617587e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-1500/training_args.bin b/sft_qwen_14B_v2/checkpoints/checkpoint-1500/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..0bd5f45757bdc6c7fc2d67718b18675b51983e0e
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-1500/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:afae4a8ce1391f149eb00b14eed8f891c715e892ea546bf754d22db2c2bc5969
+size 4792
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-2000/README.md b/sft_qwen_14B_v2/checkpoints/checkpoint-2000/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..134a3a8531b68adda3fed2a98db824f7c1b7dcfd
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-2000/README.md
@@ -0,0 +1,207 @@
+---
+base_model: Models/Qwen2.5-Coder-14B-CPT
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Models/Qwen2.5-Coder-14B-CPT
+- lora
+- transformers
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.0
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-2000/adapter_config.json b/sft_qwen_14B_v2/checkpoints/checkpoint-2000/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..d6316742bc71d30704e2ccdb9080ed0b8062e454
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-2000/adapter_config.json
@@ -0,0 +1,43 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Models/Qwen2.5-Coder-14B-CPT",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.0",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "v_proj",
+    "k_proj",
+    "q_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-2000/adapter_model.safetensors b/sft_qwen_14B_v2/checkpoints/checkpoint-2000/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..4f2e973515f4da77d737eab50d4a5d869572482a
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-2000/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f36ab65082bb1805ada7a0d63595c09b088ed7d6feed08092d0234cd38ae0c0
+size 201378736
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-2000/optimizer.pt b/sft_qwen_14B_v2/checkpoints/checkpoint-2000/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..2aa1ec2e17631b1022d8fcd4b5fa0c30f3e9efea
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-2000/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d041d1505290b0d22cce08cd5ddc5af28c3ba276e2afbc492822b36858ef6b6e
+size 402976786
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-2000/rng_state.pth b/sft_qwen_14B_v2/checkpoints/checkpoint-2000/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..a97b131884950755ba50b1e6e06822b71e0bd0ad
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-2000/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86dda28e04ef063abc8ec80a126ae11e59bcc0222e7ffedfe8ffc3e09e5ff1d1
+size 14244
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-2000/scheduler.pt b/sft_qwen_14B_v2/checkpoints/checkpoint-2000/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..81c4d0471390fa212559ab76ddbb6046d9502ba3
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-2000/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:47ee9234e7cfe3ba106b70e03151eb5b4653fe797445e980958ee4b2951365df
+size 1064
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-2000/trainer_state.json b/sft_qwen_14B_v2/checkpoints/checkpoint-2000/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..069775a699e13cd85508bfc6e633d69e52cd96f3
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-2000/trainer_state.json
@@ -0,0 +1,7203 @@
+{
+  "best_global_step": 2000,
+  "best_metric": 0.8460908532142639,
+  "best_model_checkpoint": "task2file/sft_qwen_14B_v2/checkpoints/checkpoint-2000",
+  "epoch": 0.8438818565400844,
+  "eval_steps": 100,
+  "global_step": 2000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0008438818565400844,
+      "grad_norm": 0.5386583805084229,
+      "learning_rate": 1.7574692442882248e-07,
+      "loss": 1.6941628456115723,
+      "step": 2
+    },
+    {
+      "epoch": 0.0016877637130801688,
+      "grad_norm": 0.5477277636528015,
+      "learning_rate": 5.272407732864675e-07,
+      "loss": 1.7132279872894287,
+      "step": 4
+    },
+    {
+      "epoch": 0.002531645569620253,
+      "grad_norm": 0.5390765070915222,
+      "learning_rate": 8.787346221441126e-07,
+      "loss": 1.641180396080017,
+      "step": 6
+    },
+    {
+      "epoch": 0.0033755274261603376,
+      "grad_norm": 0.5023683905601501,
+      "learning_rate": 1.2302284710017575e-06,
+      "loss": 1.5616240501403809,
+      "step": 8
+    },
+    {
+      "epoch": 0.004219409282700422,
+      "grad_norm": 0.4899154603481293,
+      "learning_rate": 1.5817223198594026e-06,
+      "loss": 1.572033405303955,
+      "step": 10
+    },
+    {
+      "epoch": 0.005063291139240506,
+      "grad_norm": 0.5239788293838501,
+      "learning_rate": 1.9332161687170474e-06,
+      "loss": 1.6242921352386475,
+      "step": 12
+    },
+    {
+      "epoch": 0.00590717299578059,
+      "grad_norm": 0.5172926783561707,
+      "learning_rate": 2.2847100175746925e-06,
+      "loss": 1.6800041198730469,
+      "step": 14
+    },
+    {
+      "epoch": 0.006751054852320675,
+      "grad_norm": 0.5539224743843079,
+      "learning_rate": 2.6362038664323376e-06,
+      "loss": 1.6450834274291992,
+      "step": 16
+    },
+    {
+      "epoch": 0.007594936708860759,
+      "grad_norm": 0.5255337953567505,
+      "learning_rate": 2.9876977152899827e-06,
+      "loss": 1.6673263311386108,
+      "step": 18
+    },
+    {
+      "epoch": 0.008438818565400843,
+      "grad_norm": 0.5074548721313477,
+      "learning_rate": 3.3391915641476277e-06,
+      "loss": 1.531802773475647,
+      "step": 20
+    },
+    {
+      "epoch": 0.009282700421940928,
+      "grad_norm": 0.4160279333591461,
+      "learning_rate": 3.6906854130052724e-06,
+      "loss": 1.599354863166809,
+      "step": 22
+    },
+    {
+      "epoch": 0.010126582278481013,
+      "grad_norm": 0.5716474652290344,
+      "learning_rate": 4.0421792618629174e-06,
+      "loss": 1.6700962781906128,
+      "step": 24
+    },
+    {
+      "epoch": 0.010970464135021098,
+      "grad_norm": 0.5148899555206299,
+      "learning_rate": 4.3936731107205625e-06,
+      "loss": 1.66217839717865,
+      "step": 26
+    },
+    {
+      "epoch": 0.01181434599156118,
+      "grad_norm": 0.575722336769104,
+      "learning_rate": 4.7451669595782076e-06,
+      "loss": 1.6692266464233398,
+      "step": 28
+    },
+    {
+      "epoch": 0.012658227848101266,
+      "grad_norm": 0.5345953106880188,
+      "learning_rate": 5.096660808435853e-06,
+      "loss": 1.5518689155578613,
+      "step": 30
+    },
+    {
+      "epoch": 0.01350210970464135,
+      "grad_norm": 0.4462043344974518,
+      "learning_rate": 5.448154657293498e-06,
+      "loss": 1.5930007696151733,
+      "step": 32
+    },
+    {
+      "epoch": 0.014345991561181435,
+      "grad_norm": 0.5119605660438538,
+      "learning_rate": 5.799648506151143e-06,
+      "loss": 1.6069684028625488,
+      "step": 34
+    },
+    {
+      "epoch": 0.015189873417721518,
+      "grad_norm": 0.5328608751296997,
+      "learning_rate": 6.151142355008788e-06,
+      "loss": 1.5838109254837036,
+      "step": 36
+    },
+    {
+      "epoch": 0.016033755274261603,
+      "grad_norm": 0.5065920352935791,
+      "learning_rate": 6.502636203866433e-06,
+      "loss": 1.608130931854248,
+      "step": 38
+    },
+    {
+      "epoch": 0.016877637130801686,
+      "grad_norm": 0.4479359984397888,
+      "learning_rate": 6.854130052724078e-06,
+      "loss": 1.5942182540893555,
+      "step": 40
+    },
+    {
+      "epoch": 0.017721518987341773,
+      "grad_norm": 0.42844903469085693,
+      "learning_rate": 7.205623901581722e-06,
+      "loss": 1.6441553831100464,
+      "step": 42
+    },
+    {
+      "epoch": 0.018565400843881856,
+      "grad_norm": 0.476630836725235,
+      "learning_rate": 7.557117750439367e-06,
+      "loss": 1.6068111658096313,
+      "step": 44
+    },
+    {
+      "epoch": 0.019409282700421943,
+      "grad_norm": 0.4532654881477356,
+      "learning_rate": 7.908611599297012e-06,
+      "loss": 1.6618021726608276,
+      "step": 46
+    },
+    {
+      "epoch": 0.020253164556962026,
+      "grad_norm": 0.3701118230819702,
+      "learning_rate": 8.260105448154657e-06,
+      "loss": 1.4730033874511719,
+      "step": 48
+    },
+    {
+      "epoch": 0.02109704641350211,
+      "grad_norm": 0.38471561670303345,
+      "learning_rate": 8.611599297012302e-06,
+      "loss": 1.4828267097473145,
+      "step": 50
+    },
+    {
+      "epoch": 0.021940928270042195,
+      "grad_norm": 0.3602336347103119,
+      "learning_rate": 8.963093145869948e-06,
+      "loss": 1.3877452611923218,
+      "step": 52
+    },
+    {
+      "epoch": 0.02278481012658228,
+      "grad_norm": 0.40318572521209717,
+      "learning_rate": 9.314586994727593e-06,
+      "loss": 1.49052894115448,
+      "step": 54
+    },
+    {
+      "epoch": 0.02362869198312236,
+      "grad_norm": 0.3223826587200165,
+      "learning_rate": 9.666080843585238e-06,
+      "loss": 1.4912524223327637,
+      "step": 56
+    },
+    {
+      "epoch": 0.024472573839662448,
+      "grad_norm": 0.3873065114021301,
+      "learning_rate": 1.0017574692442883e-05,
+      "loss": 1.526674509048462,
+      "step": 58
+    },
+    {
+      "epoch": 0.02531645569620253,
+      "grad_norm": 0.410159707069397,
+      "learning_rate": 1.0369068541300528e-05,
+      "loss": 1.4480271339416504,
+      "step": 60
+    },
+    {
+      "epoch": 0.026160337552742614,
+      "grad_norm": 0.3632003962993622,
+      "learning_rate": 1.0720562390158173e-05,
+      "loss": 1.4222990274429321,
+      "step": 62
+    },
+    {
+      "epoch": 0.0270042194092827,
+      "grad_norm": 0.33118435740470886,
+      "learning_rate": 1.1072056239015818e-05,
+      "loss": 1.387171745300293,
+      "step": 64
+    },
+    {
+      "epoch": 0.027848101265822784,
+      "grad_norm": 0.3301764726638794,
+      "learning_rate": 1.1423550087873463e-05,
+      "loss": 1.3523777723312378,
+      "step": 66
+    },
+    {
+      "epoch": 0.02869198312236287,
+      "grad_norm": 0.34342435002326965,
+      "learning_rate": 1.1775043936731108e-05,
+      "loss": 1.4515162706375122,
+      "step": 68
+    },
+    {
+      "epoch": 0.029535864978902954,
+      "grad_norm": 0.3243122100830078,
+      "learning_rate": 1.2126537785588753e-05,
+      "loss": 1.3509243726730347,
+      "step": 70
+    },
+    {
+      "epoch": 0.030379746835443037,
+      "grad_norm": 0.3450150787830353,
+      "learning_rate": 1.2478031634446398e-05,
+      "loss": 1.4936245679855347,
+      "step": 72
+    },
+    {
+      "epoch": 0.031223628691983123,
+      "grad_norm": 0.38912028074264526,
+      "learning_rate": 1.2829525483304042e-05,
+      "loss": 1.3419109582901,
+      "step": 74
+    },
+    {
+      "epoch": 0.032067510548523206,
+      "grad_norm": 0.3019310235977173,
+      "learning_rate": 1.3181019332161687e-05,
+      "loss": 1.4284154176712036,
+      "step": 76
+    },
+    {
+      "epoch": 0.03291139240506329,
+      "grad_norm": 0.37803682684898376,
+      "learning_rate": 1.3532513181019332e-05,
+      "loss": 1.4256561994552612,
+      "step": 78
+    },
+    {
+      "epoch": 0.03375527426160337,
+      "grad_norm": 0.34191736578941345,
+      "learning_rate": 1.3884007029876977e-05,
+      "loss": 1.3256909847259521,
+      "step": 80
+    },
+    {
+      "epoch": 0.03459915611814346,
+      "grad_norm": 0.35242700576782227,
+      "learning_rate": 1.4235500878734624e-05,
+      "loss": 1.2710685729980469,
+      "step": 82
+    },
+    {
+      "epoch": 0.035443037974683546,
+      "grad_norm": 0.38094228506088257,
+      "learning_rate": 1.4586994727592269e-05,
+      "loss": 1.253411889076233,
+      "step": 84
+    },
+    {
+      "epoch": 0.036286919831223625,
+      "grad_norm": 0.36837366223335266,
+      "learning_rate": 1.4938488576449914e-05,
+      "loss": 1.3064342737197876,
+      "step": 86
+    },
+    {
+      "epoch": 0.03713080168776371,
+      "grad_norm": 0.3443569242954254,
+      "learning_rate": 1.5289982425307557e-05,
+      "loss": 1.293562412261963,
+      "step": 88
+    },
+    {
+      "epoch": 0.0379746835443038,
+      "grad_norm": 0.3799338936805725,
+      "learning_rate": 1.5641476274165202e-05,
+      "loss": 1.3382648229599,
+      "step": 90
+    },
+    {
+      "epoch": 0.038818565400843885,
+      "grad_norm": 0.40501922369003296,
+      "learning_rate": 1.599297012302285e-05,
+      "loss": 1.3925724029541016,
+      "step": 92
+    },
+    {
+      "epoch": 0.039662447257383965,
+      "grad_norm": 0.4419630467891693,
+      "learning_rate": 1.6344463971880492e-05,
+      "loss": 1.357171893119812,
+      "step": 94
+    },
+    {
+      "epoch": 0.04050632911392405,
+      "grad_norm": 0.3619817793369293,
+      "learning_rate": 1.6695957820738137e-05,
+      "loss": 1.3029985427856445,
+      "step": 96
+    },
+    {
+      "epoch": 0.04135021097046414,
+      "grad_norm": 0.4851357340812683,
+      "learning_rate": 1.7047451669595782e-05,
+      "loss": 1.3498191833496094,
+      "step": 98
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "grad_norm": 0.418658584356308,
+      "learning_rate": 1.7398945518453427e-05,
+      "loss": 1.185287356376648,
+      "step": 100
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "eval_loss": 1.2979938983917236,
+      "eval_runtime": 682.1979,
+      "eval_samples_per_second": 3.089,
+      "eval_steps_per_second": 3.089,
+      "step": 100
+    },
+    {
+      "epoch": 0.043037974683544304,
+      "grad_norm": 0.4464418888092041,
+      "learning_rate": 1.7750439367311073e-05,
+      "loss": 1.2217272520065308,
+      "step": 102
+    },
+    {
+      "epoch": 0.04388185654008439,
+      "grad_norm": 0.4706237316131592,
+      "learning_rate": 1.8101933216168718e-05,
+      "loss": 1.2052050828933716,
+      "step": 104
+    },
+    {
+      "epoch": 0.04472573839662447,
+      "grad_norm": 0.46394404768943787,
+      "learning_rate": 1.8453427065026363e-05,
+      "loss": 1.221343994140625,
+      "step": 106
+    },
+    {
+      "epoch": 0.04556962025316456,
+      "grad_norm": 0.4726889431476593,
+      "learning_rate": 1.8804920913884008e-05,
+      "loss": 1.2387475967407227,
+      "step": 108
+    },
+    {
+      "epoch": 0.046413502109704644,
+      "grad_norm": 0.42130985856056213,
+      "learning_rate": 1.9156414762741653e-05,
+      "loss": 1.2851309776306152,
+      "step": 110
+    },
+    {
+      "epoch": 0.04725738396624472,
+      "grad_norm": 0.4504576623439789,
+      "learning_rate": 1.9507908611599298e-05,
+      "loss": 1.2753145694732666,
+      "step": 112
+    },
+    {
+      "epoch": 0.04810126582278481,
+      "grad_norm": 0.396085262298584,
+      "learning_rate": 1.9859402460456943e-05,
+      "loss": 1.2427717447280884,
+      "step": 114
+    },
+    {
+      "epoch": 0.048945147679324896,
+      "grad_norm": 0.5106491446495056,
+      "learning_rate": 2.0210896309314588e-05,
+      "loss": 1.2943825721740723,
+      "step": 116
+    },
+    {
+      "epoch": 0.049789029535864976,
+      "grad_norm": 0.42351317405700684,
+      "learning_rate": 2.0562390158172233e-05,
+      "loss": 1.263301134109497,
+      "step": 118
+    },
+    {
+      "epoch": 0.05063291139240506,
+      "grad_norm": 0.4403539299964905,
+      "learning_rate": 2.0913884007029878e-05,
+      "loss": 1.2647849321365356,
+      "step": 120
+    },
+    {
+      "epoch": 0.05147679324894515,
+      "grad_norm": 0.5260752439498901,
+      "learning_rate": 2.1265377855887523e-05,
+      "loss": 1.2351393699645996,
+      "step": 122
+    },
+    {
+      "epoch": 0.05232067510548523,
+      "grad_norm": 0.44978851079940796,
+      "learning_rate": 2.1616871704745168e-05,
+      "loss": 1.0384471416473389,
+      "step": 124
+    },
+    {
+      "epoch": 0.053164556962025315,
+      "grad_norm": 0.47732362151145935,
+      "learning_rate": 2.1968365553602813e-05,
+      "loss": 1.1518068313598633,
+      "step": 126
+    },
+    {
+      "epoch": 0.0540084388185654,
+      "grad_norm": 0.5473551750183105,
+      "learning_rate": 2.231985940246046e-05,
+      "loss": 1.2264912128448486,
+      "step": 128
+    },
+    {
+      "epoch": 0.05485232067510549,
+      "grad_norm": 0.4473855197429657,
+      "learning_rate": 2.2671353251318103e-05,
+      "loss": 1.1615246534347534,
+      "step": 130
+    },
+    {
+      "epoch": 0.05569620253164557,
+      "grad_norm": 0.5980377197265625,
+      "learning_rate": 2.302284710017575e-05,
+      "loss": 1.1334880590438843,
+      "step": 132
+    },
+    {
+      "epoch": 0.056540084388185655,
+      "grad_norm": 0.5987792015075684,
+      "learning_rate": 2.3374340949033394e-05,
+      "loss": 1.1546804904937744,
+      "step": 134
+    },
+    {
+      "epoch": 0.05738396624472574,
+      "grad_norm": 0.45355498790740967,
+      "learning_rate": 2.372583479789104e-05,
+      "loss": 1.194953441619873,
+      "step": 136
+    },
+    {
+      "epoch": 0.05822784810126582,
+      "grad_norm": 0.5373698472976685,
+      "learning_rate": 2.4077328646748684e-05,
+      "loss": 1.1067466735839844,
+      "step": 138
+    },
+    {
+      "epoch": 0.05907172995780591,
+      "grad_norm": 0.48734328150749207,
+      "learning_rate": 2.442882249560633e-05,
+      "loss": 1.188468098640442,
+      "step": 140
+    },
+    {
+      "epoch": 0.059915611814345994,
+      "grad_norm": 0.4692173898220062,
+      "learning_rate": 2.478031634446397e-05,
+      "loss": 1.1624362468719482,
+      "step": 142
+    },
+    {
+      "epoch": 0.060759493670886074,
+      "grad_norm": 0.532554030418396,
+      "learning_rate": 2.513181019332162e-05,
+      "loss": 1.0978907346725464,
+      "step": 144
+    },
+    {
+      "epoch": 0.06160337552742616,
+      "grad_norm": 0.5853802561759949,
+      "learning_rate": 2.5483304042179264e-05,
+      "loss": 1.2030781507492065,
+      "step": 146
+    },
+    {
+      "epoch": 0.06244725738396625,
+      "grad_norm": 0.5061611533164978,
+      "learning_rate": 2.583479789103691e-05,
+      "loss": 1.082366943359375,
+      "step": 148
+    },
+    {
+      "epoch": 0.06329113924050633,
+      "grad_norm": 0.49426141381263733,
+      "learning_rate": 2.6186291739894554e-05,
+      "loss": 1.10564386844635,
+      "step": 150
+    },
+    {
+      "epoch": 0.06413502109704641,
+      "grad_norm": 0.5846618413925171,
+      "learning_rate": 2.6537785588752196e-05,
+      "loss": 1.1992807388305664,
+      "step": 152
+    },
+    {
+      "epoch": 0.06497890295358649,
+      "grad_norm": 0.5517552495002747,
+      "learning_rate": 2.6889279437609844e-05,
+      "loss": 1.1757566928863525,
+      "step": 154
+    },
+    {
+      "epoch": 0.06582278481012659,
+      "grad_norm": 0.5667305588722229,
+      "learning_rate": 2.7240773286467486e-05,
+      "loss": 1.0548783540725708,
+      "step": 156
+    },
+    {
+      "epoch": 0.06666666666666667,
+      "grad_norm": 0.6760414242744446,
+      "learning_rate": 2.7592267135325134e-05,
+      "loss": 1.184364914894104,
+      "step": 158
+    },
+    {
+      "epoch": 0.06751054852320675,
+      "grad_norm": 0.5261430740356445,
+      "learning_rate": 2.7943760984182776e-05,
+      "loss": 1.1945042610168457,
+      "step": 160
+    },
+    {
+      "epoch": 0.06835443037974684,
+      "grad_norm": 0.6155015230178833,
+      "learning_rate": 2.8295254833040425e-05,
+      "loss": 1.2021973133087158,
+      "step": 162
+    },
+    {
+      "epoch": 0.06919831223628692,
+      "grad_norm": 0.6131619215011597,
+      "learning_rate": 2.8646748681898066e-05,
+      "loss": 1.144123911857605,
+      "step": 164
+    },
+    {
+      "epoch": 0.070042194092827,
+      "grad_norm": 0.5749185681343079,
+      "learning_rate": 2.8998242530755715e-05,
+      "loss": 1.1329256296157837,
+      "step": 166
+    },
+    {
+      "epoch": 0.07088607594936709,
+      "grad_norm": 0.5243118405342102,
+      "learning_rate": 2.9349736379613356e-05,
+      "loss": 1.0892387628555298,
+      "step": 168
+    },
+    {
+      "epoch": 0.07172995780590717,
+      "grad_norm": 0.7190104722976685,
+      "learning_rate": 2.9701230228471005e-05,
+      "loss": 1.163260817527771,
+      "step": 170
+    },
+    {
+      "epoch": 0.07257383966244725,
+      "grad_norm": 0.5486982464790344,
+      "learning_rate": 3.0052724077328647e-05,
+      "loss": 1.0880777835845947,
+      "step": 172
+    },
+    {
+      "epoch": 0.07341772151898734,
+      "grad_norm": 0.5020889043807983,
+      "learning_rate": 3.0404217926186295e-05,
+      "loss": 1.0433368682861328,
+      "step": 174
+    },
+    {
+      "epoch": 0.07426160337552742,
+      "grad_norm": 0.47329774498939514,
+      "learning_rate": 3.075571177504394e-05,
+      "loss": 1.0528991222381592,
+      "step": 176
+    },
+    {
+      "epoch": 0.0751054852320675,
+      "grad_norm": 0.6635547876358032,
+      "learning_rate": 3.110720562390158e-05,
+      "loss": 1.1627811193466187,
+      "step": 178
+    },
+    {
+      "epoch": 0.0759493670886076,
+      "grad_norm": 0.5624618530273438,
+      "learning_rate": 3.145869947275923e-05,
+      "loss": 1.084869384765625,
+      "step": 180
+    },
+    {
+      "epoch": 0.07679324894514768,
+      "grad_norm": 0.6029536724090576,
+      "learning_rate": 3.181019332161687e-05,
+      "loss": 1.2227671146392822,
+      "step": 182
+    },
+    {
+      "epoch": 0.07763713080168777,
+      "grad_norm": 0.930959939956665,
+      "learning_rate": 3.216168717047452e-05,
+      "loss": 1.0955452919006348,
+      "step": 184
+    },
+    {
+      "epoch": 0.07848101265822785,
+      "grad_norm": 0.5326952338218689,
+      "learning_rate": 3.251318101933216e-05,
+      "loss": 1.0640798807144165,
+      "step": 186
+    },
+    {
+      "epoch": 0.07932489451476793,
+      "grad_norm": 0.5484727621078491,
+      "learning_rate": 3.286467486818981e-05,
+      "loss": 1.0700589418411255,
+      "step": 188
+    },
+    {
+      "epoch": 0.08016877637130802,
+      "grad_norm": 0.605273425579071,
+      "learning_rate": 3.3216168717047456e-05,
+      "loss": 1.1593081951141357,
+      "step": 190
+    },
+    {
+      "epoch": 0.0810126582278481,
+      "grad_norm": 0.5704394578933716,
+      "learning_rate": 3.35676625659051e-05,
+      "loss": 1.1617076396942139,
+      "step": 192
+    },
+    {
+      "epoch": 0.08185654008438818,
+      "grad_norm": 0.5929452180862427,
+      "learning_rate": 3.3919156414762746e-05,
+      "loss": 1.1346839666366577,
+      "step": 194
+    },
+    {
+      "epoch": 0.08270042194092828,
+      "grad_norm": 0.5624077916145325,
+      "learning_rate": 3.427065026362039e-05,
+      "loss": 1.0934710502624512,
+      "step": 196
+    },
+    {
+      "epoch": 0.08354430379746836,
+      "grad_norm": 0.6717425584793091,
+      "learning_rate": 3.4622144112478036e-05,
+      "loss": 1.1810534000396729,
+      "step": 198
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "grad_norm": 0.5120199918746948,
+      "learning_rate": 3.4973637961335674e-05,
+      "loss": 1.1525514125823975,
+      "step": 200
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "eval_loss": 1.142486810684204,
+      "eval_runtime": 668.2356,
+      "eval_samples_per_second": 3.153,
+      "eval_steps_per_second": 3.153,
+      "step": 200
+    },
+    {
+      "epoch": 0.08523206751054853,
+      "grad_norm": 0.5144487023353577,
+      "learning_rate": 3.5325131810193326e-05,
+      "loss": 1.0243735313415527,
+      "step": 202
+    },
+    {
+      "epoch": 0.08607594936708861,
+      "grad_norm": 0.6325069069862366,
+      "learning_rate": 3.5676625659050964e-05,
+      "loss": 1.118743896484375,
+      "step": 204
+    },
+    {
+      "epoch": 0.08691983122362869,
+      "grad_norm": 0.5501633882522583,
+      "learning_rate": 3.6028119507908616e-05,
+      "loss": 1.0380504131317139,
+      "step": 206
+    },
+    {
+      "epoch": 0.08776371308016878,
+      "grad_norm": 0.6133899688720703,
+      "learning_rate": 3.6379613356766254e-05,
+      "loss": 0.9837555885314941,
+      "step": 208
+    },
+    {
+      "epoch": 0.08860759493670886,
+      "grad_norm": 0.5799810886383057,
+      "learning_rate": 3.6731107205623906e-05,
+      "loss": 1.090720295906067,
+      "step": 210
+    },
+    {
+      "epoch": 0.08945147679324894,
+      "grad_norm": 0.6039511561393738,
+      "learning_rate": 3.7082601054481544e-05,
+      "loss": 1.120232343673706,
+      "step": 212
+    },
+    {
+      "epoch": 0.09029535864978903,
+      "grad_norm": 0.5983024835586548,
+      "learning_rate": 3.7434094903339196e-05,
+      "loss": 1.096949815750122,
+      "step": 214
+    },
+    {
+      "epoch": 0.09113924050632911,
+      "grad_norm": 0.5641079545021057,
+      "learning_rate": 3.7785588752196835e-05,
+      "loss": 1.1226298809051514,
+      "step": 216
+    },
+    {
+      "epoch": 0.0919831223628692,
+      "grad_norm": 0.655717134475708,
+      "learning_rate": 3.8137082601054486e-05,
+      "loss": 1.1260643005371094,
+      "step": 218
+    },
+    {
+      "epoch": 0.09282700421940929,
+      "grad_norm": 0.6111898422241211,
+      "learning_rate": 3.848857644991213e-05,
+      "loss": 1.0777709484100342,
+      "step": 220
+    },
+    {
+      "epoch": 0.09367088607594937,
+      "grad_norm": 0.6821302771568298,
+      "learning_rate": 3.884007029876977e-05,
+      "loss": 1.10588800907135,
+      "step": 222
+    },
+    {
+      "epoch": 0.09451476793248945,
+      "grad_norm": 0.693175733089447,
+      "learning_rate": 3.919156414762742e-05,
+      "loss": 1.1498671770095825,
+      "step": 224
+    },
+    {
+      "epoch": 0.09535864978902954,
+      "grad_norm": 0.5288166403770447,
+      "learning_rate": 3.954305799648506e-05,
+      "loss": 1.0587562322616577,
+      "step": 226
+    },
+    {
+      "epoch": 0.09620253164556962,
+      "grad_norm": 0.6882867813110352,
+      "learning_rate": 3.989455184534271e-05,
+      "loss": 1.1107512712478638,
+      "step": 228
+    },
+    {
+      "epoch": 0.0970464135021097,
+      "grad_norm": 0.5834154486656189,
+      "learning_rate": 4.024604569420035e-05,
+      "loss": 1.020510196685791,
+      "step": 230
+    },
+    {
+      "epoch": 0.09789029535864979,
+      "grad_norm": 0.7157064080238342,
+      "learning_rate": 4.0597539543058e-05,
+      "loss": 1.0642449855804443,
+      "step": 232
+    },
+    {
+      "epoch": 0.09873417721518987,
+      "grad_norm": 0.6530708074569702,
+      "learning_rate": 4.094903339191564e-05,
+      "loss": 1.0359872579574585,
+      "step": 234
+    },
+    {
+      "epoch": 0.09957805907172995,
+      "grad_norm": 0.6329686045646667,
+      "learning_rate": 4.130052724077329e-05,
+      "loss": 1.050504446029663,
+      "step": 236
+    },
+    {
+      "epoch": 0.10042194092827005,
+      "grad_norm": 0.6597026586532593,
+      "learning_rate": 4.165202108963093e-05,
+      "loss": 1.2621175050735474,
+      "step": 238
+    },
+    {
+      "epoch": 0.10126582278481013,
+      "grad_norm": 0.6195225119590759,
+      "learning_rate": 4.200351493848858e-05,
+      "loss": 1.1218310594558716,
+      "step": 240
+    },
+    {
+      "epoch": 0.1021097046413502,
+      "grad_norm": 0.6764137744903564,
+      "learning_rate": 4.235500878734622e-05,
+      "loss": 1.1250728368759155,
+      "step": 242
+    },
+    {
+      "epoch": 0.1029535864978903,
+      "grad_norm": 0.552363395690918,
+      "learning_rate": 4.270650263620387e-05,
+      "loss": 1.028212308883667,
+      "step": 244
+    },
+    {
+      "epoch": 0.10379746835443038,
+      "grad_norm": 0.5620495676994324,
+      "learning_rate": 4.305799648506151e-05,
+      "loss": 1.0425450801849365,
+      "step": 246
+    },
+    {
+      "epoch": 0.10464135021097046,
+      "grad_norm": 0.6860032081604004,
+      "learning_rate": 4.3409490333919156e-05,
+      "loss": 1.144278883934021,
+      "step": 248
+    },
+    {
+      "epoch": 0.10548523206751055,
+      "grad_norm": 0.6033259034156799,
+      "learning_rate": 4.37609841827768e-05,
+      "loss": 1.1223982572555542,
+      "step": 250
+    },
+    {
+      "epoch": 0.10632911392405063,
+      "grad_norm": 0.6292146444320679,
+      "learning_rate": 4.4112478031634446e-05,
+      "loss": 1.1609960794448853,
+      "step": 252
+    },
+    {
+      "epoch": 0.10717299578059072,
+      "grad_norm": 0.7982883453369141,
+      "learning_rate": 4.44639718804921e-05,
+      "loss": 1.063547968864441,
+      "step": 254
+    },
+    {
+      "epoch": 0.1080168776371308,
+      "grad_norm": 0.7719110250473022,
+      "learning_rate": 4.4815465729349736e-05,
+      "loss": 1.0719804763793945,
+      "step": 256
+    },
+    {
+      "epoch": 0.10886075949367088,
+      "grad_norm": 0.6101011633872986,
+      "learning_rate": 4.516695957820739e-05,
+      "loss": 1.0778400897979736,
+      "step": 258
+    },
+    {
+      "epoch": 0.10970464135021098,
+      "grad_norm": 0.7300994396209717,
+      "learning_rate": 4.5518453427065026e-05,
+      "loss": 1.2129558324813843,
+      "step": 260
+    },
+    {
+      "epoch": 0.11054852320675106,
+      "grad_norm": 0.8348747491836548,
+      "learning_rate": 4.586994727592268e-05,
+      "loss": 1.221714735031128,
+      "step": 262
+    },
+    {
+      "epoch": 0.11139240506329114,
+      "grad_norm": 0.5445612072944641,
+      "learning_rate": 4.6221441124780316e-05,
+      "loss": 1.0187978744506836,
+      "step": 264
+    },
+    {
+      "epoch": 0.11223628691983123,
+      "grad_norm": 0.6230319738388062,
+      "learning_rate": 4.657293497363797e-05,
+      "loss": 1.096561312675476,
+      "step": 266
+    },
+    {
+      "epoch": 0.11308016877637131,
+      "grad_norm": 0.6231237649917603,
+      "learning_rate": 4.6924428822495606e-05,
+      "loss": 1.089842438697815,
+      "step": 268
+    },
+    {
+      "epoch": 0.11392405063291139,
+      "grad_norm": 0.7178627252578735,
+      "learning_rate": 4.727592267135325e-05,
+      "loss": 1.0696645975112915,
+      "step": 270
+    },
+    {
+      "epoch": 0.11476793248945148,
+      "grad_norm": 0.6895854473114014,
+      "learning_rate": 4.7627416520210896e-05,
+      "loss": 1.0511361360549927,
+      "step": 272
+    },
+    {
+      "epoch": 0.11561181434599156,
+      "grad_norm": 0.6046878695487976,
+      "learning_rate": 4.797891036906854e-05,
+      "loss": 1.1373958587646484,
+      "step": 274
+    },
+    {
+      "epoch": 0.11645569620253164,
+      "grad_norm": 0.6524552702903748,
+      "learning_rate": 4.833040421792619e-05,
+      "loss": 1.0734186172485352,
+      "step": 276
+    },
+    {
+      "epoch": 0.11729957805907174,
+      "grad_norm": 0.6331019997596741,
+      "learning_rate": 4.868189806678383e-05,
+      "loss": 1.123913049697876,
+      "step": 278
+    },
+    {
+      "epoch": 0.11814345991561181,
+      "grad_norm": 0.5919018983840942,
+      "learning_rate": 4.903339191564148e-05,
+      "loss": 1.0635710954666138,
+      "step": 280
+    },
+    {
+      "epoch": 0.1189873417721519,
+      "grad_norm": 0.6067633032798767,
+      "learning_rate": 4.938488576449912e-05,
+      "loss": 1.0429247617721558,
+      "step": 282
+    },
+    {
+      "epoch": 0.11983122362869199,
+      "grad_norm": 0.6583750247955322,
+      "learning_rate": 4.9736379613356774e-05,
+      "loss": 1.1397464275360107,
+      "step": 284
+    },
+    {
+      "epoch": 0.12067510548523207,
+      "grad_norm": 0.6200069785118103,
+      "learning_rate": 5.008787346221442e-05,
+      "loss": 1.0590803623199463,
+      "step": 286
+    },
+    {
+      "epoch": 0.12151898734177215,
+      "grad_norm": 0.6798665523529053,
+      "learning_rate": 5.0439367311072064e-05,
+      "loss": 1.1318789720535278,
+      "step": 288
+    },
+    {
+      "epoch": 0.12236286919831224,
+      "grad_norm": 0.7508794069290161,
+      "learning_rate": 5.07908611599297e-05,
+      "loss": 1.0934956073760986,
+      "step": 290
+    },
+    {
+      "epoch": 0.12320675105485232,
+      "grad_norm": 0.6901452541351318,
+      "learning_rate": 5.114235500878735e-05,
+      "loss": 1.163407802581787,
+      "step": 292
+    },
+    {
+      "epoch": 0.1240506329113924,
+      "grad_norm": 0.6423285603523254,
+      "learning_rate": 5.1493848857644985e-05,
+      "loss": 1.09059476852417,
+      "step": 294
+    },
+    {
+      "epoch": 0.1248945147679325,
+      "grad_norm": 0.6839275360107422,
+      "learning_rate": 5.1845342706502644e-05,
+      "loss": 1.0690211057662964,
+      "step": 296
+    },
+    {
+      "epoch": 0.1257383966244726,
+      "grad_norm": 0.6350128054618835,
+      "learning_rate": 5.219683655536028e-05,
+      "loss": 0.982322096824646,
+      "step": 298
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "grad_norm": 0.7136530876159668,
+      "learning_rate": 5.254833040421793e-05,
+      "loss": 1.1132930517196655,
+      "step": 300
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "eval_loss": 1.0952109098434448,
+      "eval_runtime": 677.0652,
+      "eval_samples_per_second": 3.112,
+      "eval_steps_per_second": 3.112,
+      "step": 300
+    },
+    {
+      "epoch": 0.12742616033755275,
+      "grad_norm": 0.7339721322059631,
+      "learning_rate": 5.289982425307557e-05,
+      "loss": 0.973595917224884,
+      "step": 302
+    },
+    {
+      "epoch": 0.12827004219409283,
+      "grad_norm": 0.5941481590270996,
+      "learning_rate": 5.3251318101933224e-05,
+      "loss": 0.9819849729537964,
+      "step": 304
+    },
+    {
+      "epoch": 0.1291139240506329,
+      "grad_norm": 0.7153938412666321,
+      "learning_rate": 5.360281195079086e-05,
+      "loss": 1.0315470695495605,
+      "step": 306
+    },
+    {
+      "epoch": 0.12995780590717299,
+      "grad_norm": 0.5167180299758911,
+      "learning_rate": 5.395430579964851e-05,
+      "loss": 0.9492001533508301,
+      "step": 308
+    },
+    {
+      "epoch": 0.1308016877637131,
+      "grad_norm": 0.6055944561958313,
+      "learning_rate": 5.430579964850615e-05,
+      "loss": 1.0156209468841553,
+      "step": 310
+    },
+    {
+      "epoch": 0.13164556962025317,
+      "grad_norm": 0.7662386298179626,
+      "learning_rate": 5.4657293497363805e-05,
+      "loss": 1.1791651248931885,
+      "step": 312
+    },
+    {
+      "epoch": 0.13248945147679325,
+      "grad_norm": 0.6065546274185181,
+      "learning_rate": 5.500878734622145e-05,
+      "loss": 1.0009297132492065,
+      "step": 314
+    },
+    {
+      "epoch": 0.13333333333333333,
+      "grad_norm": 0.604225754737854,
+      "learning_rate": 5.536028119507909e-05,
+      "loss": 1.0208244323730469,
+      "step": 316
+    },
+    {
+      "epoch": 0.1341772151898734,
+      "grad_norm": 0.6186763048171997,
+      "learning_rate": 5.571177504393673e-05,
+      "loss": 0.9968416690826416,
+      "step": 318
+    },
+    {
+      "epoch": 0.1350210970464135,
+      "grad_norm": 0.7100363969802856,
+      "learning_rate": 5.606326889279437e-05,
+      "loss": 0.9540256857872009,
+      "step": 320
+    },
+    {
+      "epoch": 0.1358649789029536,
+      "grad_norm": 0.6979711055755615,
+      "learning_rate": 5.641476274165203e-05,
+      "loss": 1.0631953477859497,
+      "step": 322
+    },
+    {
+      "epoch": 0.13670886075949368,
+      "grad_norm": 0.6237109303474426,
+      "learning_rate": 5.676625659050967e-05,
+      "loss": 1.0170501470565796,
+      "step": 324
+    },
+    {
+      "epoch": 0.13755274261603376,
+      "grad_norm": 0.6525548696517944,
+      "learning_rate": 5.711775043936731e-05,
+      "loss": 1.0715603828430176,
+      "step": 326
+    },
+    {
+      "epoch": 0.13839662447257384,
+      "grad_norm": 0.6869221329689026,
+      "learning_rate": 5.746924428822495e-05,
+      "loss": 1.0111541748046875,
+      "step": 328
+    },
+    {
+      "epoch": 0.13924050632911392,
+      "grad_norm": 0.553188145160675,
+      "learning_rate": 5.782073813708261e-05,
+      "loss": 1.0311682224273682,
+      "step": 330
+    },
+    {
+      "epoch": 0.140084388185654,
+      "grad_norm": 0.6760852932929993,
+      "learning_rate": 5.817223198594025e-05,
+      "loss": 1.0213634967803955,
+      "step": 332
+    },
+    {
+      "epoch": 0.1409282700421941,
+      "grad_norm": 0.5907419919967651,
+      "learning_rate": 5.8523725834797894e-05,
+      "loss": 0.9748594164848328,
+      "step": 334
+    },
+    {
+      "epoch": 0.14177215189873418,
+      "grad_norm": 0.7044920921325684,
+      "learning_rate": 5.887521968365554e-05,
+      "loss": 1.05863356590271,
+      "step": 336
+    },
+    {
+      "epoch": 0.14261603375527426,
+      "grad_norm": 0.679073691368103,
+      "learning_rate": 5.922671353251318e-05,
+      "loss": 1.1341127157211304,
+      "step": 338
+    },
+    {
+      "epoch": 0.14345991561181434,
+      "grad_norm": 0.7676237225532532,
+      "learning_rate": 5.957820738137083e-05,
+      "loss": 0.9540836215019226,
+      "step": 340
+    },
+    {
+      "epoch": 0.14430379746835442,
+      "grad_norm": 0.6313899755477905,
+      "learning_rate": 5.9929701230228474e-05,
+      "loss": 1.0585911273956299,
+      "step": 342
+    },
+    {
+      "epoch": 0.1451476793248945,
+      "grad_norm": 0.7123099565505981,
+      "learning_rate": 6.028119507908612e-05,
+      "loss": 1.0760118961334229,
+      "step": 344
+    },
+    {
+      "epoch": 0.1459915611814346,
+      "grad_norm": 0.585935652256012,
+      "learning_rate": 6.063268892794376e-05,
+      "loss": 1.036866307258606,
+      "step": 346
+    },
+    {
+      "epoch": 0.1468354430379747,
+      "grad_norm": 0.5643263459205627,
+      "learning_rate": 6.0984182776801416e-05,
+      "loss": 1.0242938995361328,
+      "step": 348
+    },
+    {
+      "epoch": 0.14767932489451477,
+      "grad_norm": 0.626761794090271,
+      "learning_rate": 6.133567662565906e-05,
+      "loss": 1.0497376918792725,
+      "step": 350
+    },
+    {
+      "epoch": 0.14852320675105485,
+      "grad_norm": 0.5106956958770752,
+      "learning_rate": 6.16871704745167e-05,
+      "loss": 0.9811885356903076,
+      "step": 352
+    },
+    {
+      "epoch": 0.14936708860759493,
+      "grad_norm": 0.6948089003562927,
+      "learning_rate": 6.203866432337434e-05,
+      "loss": 1.0715330839157104,
+      "step": 354
+    },
+    {
+      "epoch": 0.150210970464135,
+      "grad_norm": 0.699713945388794,
+      "learning_rate": 6.239015817223199e-05,
+      "loss": 1.0405226945877075,
+      "step": 356
+    },
+    {
+      "epoch": 0.15105485232067511,
+      "grad_norm": 0.6437667012214661,
+      "learning_rate": 6.274165202108964e-05,
+      "loss": 1.0490930080413818,
+      "step": 358
+    },
+    {
+      "epoch": 0.1518987341772152,
+      "grad_norm": 0.6952699422836304,
+      "learning_rate": 6.309314586994728e-05,
+      "loss": 0.9267548322677612,
+      "step": 360
+    },
+    {
+      "epoch": 0.15274261603375527,
+      "grad_norm": 0.6713186502456665,
+      "learning_rate": 6.344463971880492e-05,
+      "loss": 1.0427420139312744,
+      "step": 362
+    },
+    {
+      "epoch": 0.15358649789029535,
+      "grad_norm": 0.6750379800796509,
+      "learning_rate": 6.379613356766257e-05,
+      "loss": 1.048950433731079,
+      "step": 364
+    },
+    {
+      "epoch": 0.15443037974683543,
+      "grad_norm": 0.6053379774093628,
+      "learning_rate": 6.414762741652022e-05,
+      "loss": 1.0156004428863525,
+      "step": 366
+    },
+    {
+      "epoch": 0.15527426160337554,
+      "grad_norm": 0.8063633441925049,
+      "learning_rate": 6.449912126537786e-05,
+      "loss": 1.0020819902420044,
+      "step": 368
+    },
+    {
+      "epoch": 0.15611814345991562,
+      "grad_norm": 0.8027494549751282,
+      "learning_rate": 6.48506151142355e-05,
+      "loss": 1.055633783340454,
+      "step": 370
+    },
+    {
+      "epoch": 0.1569620253164557,
+      "grad_norm": 0.6580121517181396,
+      "learning_rate": 6.520210896309315e-05,
+      "loss": 1.0149940252304077,
+      "step": 372
+    },
+    {
+      "epoch": 0.15780590717299578,
+      "grad_norm": 0.6561233997344971,
+      "learning_rate": 6.55536028119508e-05,
+      "loss": 0.9769611954689026,
+      "step": 374
+    },
+    {
+      "epoch": 0.15864978902953586,
+      "grad_norm": 0.6444346308708191,
+      "learning_rate": 6.590509666080844e-05,
+      "loss": 0.9099349975585938,
+      "step": 376
+    },
+    {
+      "epoch": 0.15949367088607594,
+      "grad_norm": 0.5879359245300293,
+      "learning_rate": 6.625659050966608e-05,
+      "loss": 1.0797548294067383,
+      "step": 378
+    },
+    {
+      "epoch": 0.16033755274261605,
+      "grad_norm": 0.6994144916534424,
+      "learning_rate": 6.660808435852373e-05,
+      "loss": 1.0336791276931763,
+      "step": 380
+    },
+    {
+      "epoch": 0.16118143459915613,
+      "grad_norm": 0.6128669381141663,
+      "learning_rate": 6.695957820738138e-05,
+      "loss": 1.018118143081665,
+      "step": 382
+    },
+    {
+      "epoch": 0.1620253164556962,
+      "grad_norm": 1.0237540006637573,
+      "learning_rate": 6.731107205623902e-05,
+      "loss": 1.1405497789382935,
+      "step": 384
+    },
+    {
+      "epoch": 0.16286919831223629,
+      "grad_norm": 0.6091578006744385,
+      "learning_rate": 6.766256590509666e-05,
+      "loss": 1.0314189195632935,
+      "step": 386
+    },
+    {
+      "epoch": 0.16371308016877636,
+      "grad_norm": 0.5916037559509277,
+      "learning_rate": 6.801405975395431e-05,
+      "loss": 0.9564052820205688,
+      "step": 388
+    },
+    {
+      "epoch": 0.16455696202531644,
+      "grad_norm": 0.771653950214386,
+      "learning_rate": 6.836555360281195e-05,
+      "loss": 1.0023859739303589,
+      "step": 390
+    },
+    {
+      "epoch": 0.16540084388185655,
+      "grad_norm": 0.654658317565918,
+      "learning_rate": 6.87170474516696e-05,
+      "loss": 1.07024085521698,
+      "step": 392
+    },
+    {
+      "epoch": 0.16624472573839663,
+      "grad_norm": 0.6611968874931335,
+      "learning_rate": 6.906854130052724e-05,
+      "loss": 1.0552500486373901,
+      "step": 394
+    },
+    {
+      "epoch": 0.1670886075949367,
+      "grad_norm": 0.6955893039703369,
+      "learning_rate": 6.942003514938489e-05,
+      "loss": 1.0562875270843506,
+      "step": 396
+    },
+    {
+      "epoch": 0.1679324894514768,
+      "grad_norm": 0.6666058301925659,
+      "learning_rate": 6.977152899824253e-05,
+      "loss": 0.9850592017173767,
+      "step": 398
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "grad_norm": 0.6131711006164551,
+      "learning_rate": 7.012302284710018e-05,
+      "loss": 1.0077755451202393,
+      "step": 400
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "eval_loss": 1.0625108480453491,
+      "eval_runtime": 691.0068,
+      "eval_samples_per_second": 3.049,
+      "eval_steps_per_second": 3.049,
+      "step": 400
+    },
+    {
+      "epoch": 0.16962025316455695,
+      "grad_norm": 0.6286499500274658,
+      "learning_rate": 7.047451669595783e-05,
+      "loss": 1.1012427806854248,
+      "step": 402
+    },
+    {
+      "epoch": 0.17046413502109706,
+      "grad_norm": 0.6639351844787598,
+      "learning_rate": 7.082601054481547e-05,
+      "loss": 1.0379719734191895,
+      "step": 404
+    },
+    {
+      "epoch": 0.17130801687763714,
+      "grad_norm": 0.750401496887207,
+      "learning_rate": 7.117750439367311e-05,
+      "loss": 1.031856656074524,
+      "step": 406
+    },
+    {
+      "epoch": 0.17215189873417722,
+      "grad_norm": 0.8084847331047058,
+      "learning_rate": 7.152899824253075e-05,
+      "loss": 1.0493193864822388,
+      "step": 408
+    },
+    {
+      "epoch": 0.1729957805907173,
+      "grad_norm": 0.7448462247848511,
+      "learning_rate": 7.188049209138841e-05,
+      "loss": 1.1012418270111084,
+      "step": 410
+    },
+    {
+      "epoch": 0.17383966244725738,
+      "grad_norm": 0.5841867923736572,
+      "learning_rate": 7.223198594024605e-05,
+      "loss": 0.9926692247390747,
+      "step": 412
+    },
+    {
+      "epoch": 0.17468354430379746,
+      "grad_norm": 0.7125606536865234,
+      "learning_rate": 7.258347978910369e-05,
+      "loss": 1.0588877201080322,
+      "step": 414
+    },
+    {
+      "epoch": 0.17552742616033756,
+      "grad_norm": 0.5750942230224609,
+      "learning_rate": 7.293497363796134e-05,
+      "loss": 1.038270354270935,
+      "step": 416
+    },
+    {
+      "epoch": 0.17637130801687764,
+      "grad_norm": 0.565444827079773,
+      "learning_rate": 7.328646748681899e-05,
+      "loss": 0.9843021035194397,
+      "step": 418
+    },
+    {
+      "epoch": 0.17721518987341772,
+      "grad_norm": 0.5825693011283875,
+      "learning_rate": 7.363796133567663e-05,
+      "loss": 1.0731632709503174,
+      "step": 420
+    },
+    {
+      "epoch": 0.1780590717299578,
+      "grad_norm": 0.6267391443252563,
+      "learning_rate": 7.398945518453427e-05,
+      "loss": 1.0061273574829102,
+      "step": 422
+    },
+    {
+      "epoch": 0.17890295358649788,
+      "grad_norm": 0.6621372103691101,
+      "learning_rate": 7.434094903339192e-05,
+      "loss": 1.0461612939834595,
+      "step": 424
+    },
+    {
+      "epoch": 0.17974683544303796,
+      "grad_norm": 0.6635435223579407,
+      "learning_rate": 7.469244288224957e-05,
+      "loss": 0.9789207577705383,
+      "step": 426
+    },
+    {
+      "epoch": 0.18059071729957807,
+      "grad_norm": 0.6342346668243408,
+      "learning_rate": 7.504393673110721e-05,
+      "loss": 1.0527069568634033,
+      "step": 428
+    },
+    {
+      "epoch": 0.18143459915611815,
+      "grad_norm": 0.6762149930000305,
+      "learning_rate": 7.539543057996485e-05,
+      "loss": 0.9708702564239502,
+      "step": 430
+    },
+    {
+      "epoch": 0.18227848101265823,
+      "grad_norm": 0.7073282599449158,
+      "learning_rate": 7.57469244288225e-05,
+      "loss": 1.0509834289550781,
+      "step": 432
+    },
+    {
+      "epoch": 0.1831223628691983,
+      "grad_norm": 0.6917856931686401,
+      "learning_rate": 7.609841827768014e-05,
+      "loss": 1.0128819942474365,
+      "step": 434
+    },
+    {
+      "epoch": 0.1839662447257384,
+      "grad_norm": 0.5574942231178284,
+      "learning_rate": 7.644991212653779e-05,
+      "loss": 0.989395797252655,
+      "step": 436
+    },
+    {
+      "epoch": 0.1848101265822785,
+      "grad_norm": 0.640765905380249,
+      "learning_rate": 7.680140597539543e-05,
+      "loss": 0.9846042990684509,
+      "step": 438
+    },
+    {
+      "epoch": 0.18565400843881857,
+      "grad_norm": 0.6699127554893494,
+      "learning_rate": 7.715289982425308e-05,
+      "loss": 1.0344442129135132,
+      "step": 440
+    },
+    {
+      "epoch": 0.18649789029535865,
+      "grad_norm": 0.6164930462837219,
+      "learning_rate": 7.750439367311072e-05,
+      "loss": 1.0179373025894165,
+      "step": 442
+    },
+    {
+      "epoch": 0.18734177215189873,
+      "grad_norm": 0.6880720853805542,
+      "learning_rate": 7.785588752196837e-05,
+      "loss": 1.0518895387649536,
+      "step": 444
+    },
+    {
+      "epoch": 0.1881856540084388,
+      "grad_norm": 0.6501413583755493,
+      "learning_rate": 7.820738137082601e-05,
+      "loss": 1.0442606210708618,
+      "step": 446
+    },
+    {
+      "epoch": 0.1890295358649789,
+      "grad_norm": 0.6076085567474365,
+      "learning_rate": 7.855887521968366e-05,
+      "loss": 0.9828442335128784,
+      "step": 448
+    },
+    {
+      "epoch": 0.189873417721519,
+      "grad_norm": 0.6418202519416809,
+      "learning_rate": 7.89103690685413e-05,
+      "loss": 1.0573710203170776,
+      "step": 450
+    },
+    {
+      "epoch": 0.19071729957805908,
+      "grad_norm": 0.7055076360702515,
+      "learning_rate": 7.926186291739895e-05,
+      "loss": 1.0216103792190552,
+      "step": 452
+    },
+    {
+      "epoch": 0.19156118143459916,
+      "grad_norm": 0.5668330192565918,
+      "learning_rate": 7.961335676625659e-05,
+      "loss": 0.9837722778320312,
+      "step": 454
+    },
+    {
+      "epoch": 0.19240506329113924,
+      "grad_norm": 0.6419380307197571,
+      "learning_rate": 7.996485061511424e-05,
+      "loss": 1.0003894567489624,
+      "step": 456
+    },
+    {
+      "epoch": 0.19324894514767932,
+      "grad_norm": 0.5949198007583618,
+      "learning_rate": 8.031634446397188e-05,
+      "loss": 1.0609031915664673,
+      "step": 458
+    },
+    {
+      "epoch": 0.1940928270042194,
+      "grad_norm": 0.7032039761543274,
+      "learning_rate": 8.066783831282952e-05,
+      "loss": 1.0543403625488281,
+      "step": 460
+    },
+    {
+      "epoch": 0.1949367088607595,
+      "grad_norm": 0.5775868892669678,
+      "learning_rate": 8.101933216168718e-05,
+      "loss": 0.9819303154945374,
+      "step": 462
+    },
+    {
+      "epoch": 0.19578059071729959,
+      "grad_norm": 0.9301062226295471,
+      "learning_rate": 8.137082601054482e-05,
+      "loss": 1.0542067289352417,
+      "step": 464
+    },
+    {
+      "epoch": 0.19662447257383966,
+      "grad_norm": 0.6193217039108276,
+      "learning_rate": 8.172231985940246e-05,
+      "loss": 0.9966341257095337,
+      "step": 466
+    },
+    {
+      "epoch": 0.19746835443037974,
+      "grad_norm": 0.6286146640777588,
+      "learning_rate": 8.20738137082601e-05,
+      "loss": 1.0474121570587158,
+      "step": 468
+    },
+    {
+      "epoch": 0.19831223628691982,
+      "grad_norm": 0.7418972253799438,
+      "learning_rate": 8.242530755711776e-05,
+      "loss": 0.9549239277839661,
+      "step": 470
+    },
+    {
+      "epoch": 0.1991561181434599,
+      "grad_norm": 0.6122808456420898,
+      "learning_rate": 8.27768014059754e-05,
+      "loss": 1.0191338062286377,
+      "step": 472
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.6375362277030945,
+      "learning_rate": 8.312829525483304e-05,
+      "loss": 1.0987539291381836,
+      "step": 474
+    },
+    {
+      "epoch": 0.2008438818565401,
+      "grad_norm": 0.6459513306617737,
+      "learning_rate": 8.347978910369068e-05,
+      "loss": 1.0369136333465576,
+      "step": 476
+    },
+    {
+      "epoch": 0.20168776371308017,
+      "grad_norm": 0.7029640674591064,
+      "learning_rate": 8.383128295254833e-05,
+      "loss": 1.0582096576690674,
+      "step": 478
+    },
+    {
+      "epoch": 0.20253164556962025,
+      "grad_norm": 0.6345387697219849,
+      "learning_rate": 8.418277680140598e-05,
+      "loss": 1.022916316986084,
+      "step": 480
+    },
+    {
+      "epoch": 0.20337552742616033,
+      "grad_norm": 0.5764590501785278,
+      "learning_rate": 8.453427065026362e-05,
+      "loss": 0.973024308681488,
+      "step": 482
+    },
+    {
+      "epoch": 0.2042194092827004,
+      "grad_norm": 0.5884482860565186,
+      "learning_rate": 8.488576449912127e-05,
+      "loss": 1.0292812585830688,
+      "step": 484
+    },
+    {
+      "epoch": 0.20506329113924052,
+      "grad_norm": 0.616357147693634,
+      "learning_rate": 8.523725834797891e-05,
+      "loss": 1.0083447694778442,
+      "step": 486
+    },
+    {
+      "epoch": 0.2059071729957806,
+      "grad_norm": 0.7671196460723877,
+      "learning_rate": 8.558875219683656e-05,
+      "loss": 0.9936985373497009,
+      "step": 488
+    },
+    {
+      "epoch": 0.20675105485232068,
+      "grad_norm": 0.6197299957275391,
+      "learning_rate": 8.59402460456942e-05,
+      "loss": 1.051513910293579,
+      "step": 490
+    },
+    {
+      "epoch": 0.20759493670886076,
+      "grad_norm": 0.6912890672683716,
+      "learning_rate": 8.629173989455185e-05,
+      "loss": 0.9474978446960449,
+      "step": 492
+    },
+    {
+      "epoch": 0.20843881856540084,
+      "grad_norm": 0.6941592693328857,
+      "learning_rate": 8.664323374340949e-05,
+      "loss": 1.0671660900115967,
+      "step": 494
+    },
+    {
+      "epoch": 0.20928270042194091,
+      "grad_norm": 0.5889528393745422,
+      "learning_rate": 8.699472759226714e-05,
+      "loss": 1.0020159482955933,
+      "step": 496
+    },
+    {
+      "epoch": 0.21012658227848102,
+      "grad_norm": 0.6478549838066101,
+      "learning_rate": 8.734622144112478e-05,
+      "loss": 1.0165860652923584,
+      "step": 498
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "grad_norm": 0.6018255949020386,
+      "learning_rate": 8.769771528998243e-05,
+      "loss": 0.8798263072967529,
+      "step": 500
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "eval_loss": 1.042096495628357,
+      "eval_runtime": 692.4361,
+      "eval_samples_per_second": 3.043,
+      "eval_steps_per_second": 3.043,
+      "step": 500
+    },
+    {
+      "epoch": 0.21181434599156118,
+      "grad_norm": 0.578990638256073,
+      "learning_rate": 8.804920913884007e-05,
+      "loss": 1.092096209526062,
+      "step": 502
+    },
+    {
+      "epoch": 0.21265822784810126,
+      "grad_norm": 0.6597883701324463,
+      "learning_rate": 8.840070298769771e-05,
+      "loss": 1.0413451194763184,
+      "step": 504
+    },
+    {
+      "epoch": 0.21350210970464134,
+      "grad_norm": 0.6660305261611938,
+      "learning_rate": 8.875219683655536e-05,
+      "loss": 1.0073142051696777,
+      "step": 506
+    },
+    {
+      "epoch": 0.21434599156118145,
+      "grad_norm": 0.6283115148544312,
+      "learning_rate": 8.910369068541301e-05,
+      "loss": 1.0319768190383911,
+      "step": 508
+    },
+    {
+      "epoch": 0.21518987341772153,
+      "grad_norm": 0.6257343292236328,
+      "learning_rate": 8.945518453427065e-05,
+      "loss": 1.0046353340148926,
+      "step": 510
+    },
+    {
+      "epoch": 0.2160337552742616,
+      "grad_norm": 0.5530875325202942,
+      "learning_rate": 8.980667838312829e-05,
+      "loss": 0.9169099926948547,
+      "step": 512
+    },
+    {
+      "epoch": 0.2168776371308017,
+      "grad_norm": 0.5369633436203003,
+      "learning_rate": 9.015817223198594e-05,
+      "loss": 0.9081505537033081,
+      "step": 514
+    },
+    {
+      "epoch": 0.21772151898734177,
+      "grad_norm": 0.6618232131004333,
+      "learning_rate": 9.05096660808436e-05,
+      "loss": 0.9165045022964478,
+      "step": 516
+    },
+    {
+      "epoch": 0.21856540084388185,
+      "grad_norm": 0.600666344165802,
+      "learning_rate": 9.086115992970123e-05,
+      "loss": 0.91348797082901,
+      "step": 518
+    },
+    {
+      "epoch": 0.21940928270042195,
+      "grad_norm": 0.5919831991195679,
+      "learning_rate": 9.121265377855887e-05,
+      "loss": 1.006508469581604,
+      "step": 520
+    },
+    {
+      "epoch": 0.22025316455696203,
+      "grad_norm": 0.688058614730835,
+      "learning_rate": 9.156414762741654e-05,
+      "loss": 1.0013236999511719,
+      "step": 522
+    },
+    {
+      "epoch": 0.2210970464135021,
+      "grad_norm": 0.6721227765083313,
+      "learning_rate": 9.191564147627418e-05,
+      "loss": 1.0909923315048218,
+      "step": 524
+    },
+    {
+      "epoch": 0.2219409282700422,
+      "grad_norm": 0.5987313389778137,
+      "learning_rate": 9.226713532513181e-05,
+      "loss": 1.0117096900939941,
+      "step": 526
+    },
+    {
+      "epoch": 0.22278481012658227,
+      "grad_norm": 0.6191489696502686,
+      "learning_rate": 9.261862917398945e-05,
+      "loss": 1.0153647661209106,
+      "step": 528
+    },
+    {
+      "epoch": 0.22362869198312235,
+      "grad_norm": 0.6821563243865967,
+      "learning_rate": 9.29701230228471e-05,
+      "loss": 0.9649755954742432,
+      "step": 530
+    },
+    {
+      "epoch": 0.22447257383966246,
+      "grad_norm": 1.760398268699646,
+      "learning_rate": 9.332161687170476e-05,
+      "loss": 0.8673232197761536,
+      "step": 532
+    },
+    {
+      "epoch": 0.22531645569620254,
+      "grad_norm": 0.6670058369636536,
+      "learning_rate": 9.36731107205624e-05,
+      "loss": 0.9942440986633301,
+      "step": 534
+    },
+    {
+      "epoch": 0.22616033755274262,
+      "grad_norm": 0.7345916032791138,
+      "learning_rate": 9.402460456942003e-05,
+      "loss": 1.0364389419555664,
+      "step": 536
+    },
+    {
+      "epoch": 0.2270042194092827,
+      "grad_norm": 0.5946128964424133,
+      "learning_rate": 9.437609841827768e-05,
+      "loss": 0.9314924478530884,
+      "step": 538
+    },
+    {
+      "epoch": 0.22784810126582278,
+      "grad_norm": 0.5800848603248596,
+      "learning_rate": 9.472759226713534e-05,
+      "loss": 1.0694862604141235,
+      "step": 540
+    },
+    {
+      "epoch": 0.22869198312236286,
+      "grad_norm": 0.6712192893028259,
+      "learning_rate": 9.507908611599297e-05,
+      "loss": 1.03531014919281,
+      "step": 542
+    },
+    {
+      "epoch": 0.22953586497890296,
+      "grad_norm": 0.5641416311264038,
+      "learning_rate": 9.543057996485063e-05,
+      "loss": 0.9795235991477966,
+      "step": 544
+    },
+    {
+      "epoch": 0.23037974683544304,
+      "grad_norm": 0.50412517786026,
+      "learning_rate": 9.578207381370826e-05,
+      "loss": 0.9641494750976562,
+      "step": 546
+    },
+    {
+      "epoch": 0.23122362869198312,
+      "grad_norm": 0.579118549823761,
+      "learning_rate": 9.61335676625659e-05,
+      "loss": 0.9375281929969788,
+      "step": 548
+    },
+    {
+      "epoch": 0.2320675105485232,
+      "grad_norm": 0.5888341665267944,
+      "learning_rate": 9.648506151142355e-05,
+      "loss": 0.9414046406745911,
+      "step": 550
+    },
+    {
+      "epoch": 0.23291139240506328,
+      "grad_norm": 0.5595056414604187,
+      "learning_rate": 9.68365553602812e-05,
+      "loss": 0.9005617499351501,
+      "step": 552
+    },
+    {
+      "epoch": 0.23375527426160336,
+      "grad_norm": 0.6605326533317566,
+      "learning_rate": 9.718804920913884e-05,
+      "loss": 1.0283968448638916,
+      "step": 554
+    },
+    {
+      "epoch": 0.23459915611814347,
+      "grad_norm": 0.5657313466072083,
+      "learning_rate": 9.753954305799648e-05,
+      "loss": 1.0058249235153198,
+      "step": 556
+    },
+    {
+      "epoch": 0.23544303797468355,
+      "grad_norm": 0.5433364510536194,
+      "learning_rate": 9.789103690685413e-05,
+      "loss": 0.9835494756698608,
+      "step": 558
+    },
+    {
+      "epoch": 0.23628691983122363,
+      "grad_norm": 0.6129802465438843,
+      "learning_rate": 9.824253075571179e-05,
+      "loss": 1.054532527923584,
+      "step": 560
+    },
+    {
+      "epoch": 0.2371308016877637,
+      "grad_norm": 0.6496239304542542,
+      "learning_rate": 9.859402460456942e-05,
+      "loss": 1.0240973234176636,
+      "step": 562
+    },
+    {
+      "epoch": 0.2379746835443038,
+      "grad_norm": 0.6380873918533325,
+      "learning_rate": 9.894551845342706e-05,
+      "loss": 1.0229179859161377,
+      "step": 564
+    },
+    {
+      "epoch": 0.23881856540084387,
+      "grad_norm": 0.6151993870735168,
+      "learning_rate": 9.929701230228471e-05,
+      "loss": 1.0111570358276367,
+      "step": 566
+    },
+    {
+      "epoch": 0.23966244725738398,
+      "grad_norm": 0.5727584958076477,
+      "learning_rate": 9.964850615114237e-05,
+      "loss": 0.9450829029083252,
+      "step": 568
+    },
+    {
+      "epoch": 0.24050632911392406,
+      "grad_norm": 0.6620725989341736,
+      "learning_rate": 0.0001,
+      "loss": 0.9800319075584412,
+      "step": 570
+    },
+    {
+      "epoch": 0.24135021097046414,
+      "grad_norm": 0.6151163578033447,
+      "learning_rate": 0.00010035149384885764,
+      "loss": 0.9757438898086548,
+      "step": 572
+    },
+    {
+      "epoch": 0.24219409282700421,
+      "grad_norm": 0.5672140717506409,
+      "learning_rate": 0.0001007029876977153,
+      "loss": 0.9104921817779541,
+      "step": 574
+    },
+    {
+      "epoch": 0.2430379746835443,
+      "grad_norm": 0.5697256326675415,
+      "learning_rate": 0.00010105448154657293,
+      "loss": 1.1027376651763916,
+      "step": 576
+    },
+    {
+      "epoch": 0.2438818565400844,
+      "grad_norm": 0.5590381622314453,
+      "learning_rate": 0.00010140597539543057,
+      "loss": 1.0055404901504517,
+      "step": 578
+    },
+    {
+      "epoch": 0.24472573839662448,
+      "grad_norm": 0.5518567562103271,
+      "learning_rate": 0.00010175746924428824,
+      "loss": 1.020835518836975,
+      "step": 580
+    },
+    {
+      "epoch": 0.24556962025316456,
+      "grad_norm": 0.6338496208190918,
+      "learning_rate": 0.00010210896309314588,
+      "loss": 0.9528344869613647,
+      "step": 582
+    },
+    {
+      "epoch": 0.24641350210970464,
+      "grad_norm": 0.6497329473495483,
+      "learning_rate": 0.00010246045694200353,
+      "loss": 1.0088670253753662,
+      "step": 584
+    },
+    {
+      "epoch": 0.24725738396624472,
+      "grad_norm": 0.49888095259666443,
+      "learning_rate": 0.00010281195079086117,
+      "loss": 0.9961200952529907,
+      "step": 586
+    },
+    {
+      "epoch": 0.2481012658227848,
+      "grad_norm": 0.5680158734321594,
+      "learning_rate": 0.0001031634446397188,
+      "loss": 0.9635610580444336,
+      "step": 588
+    },
+    {
+      "epoch": 0.2489451476793249,
+      "grad_norm": 0.658168375492096,
+      "learning_rate": 0.00010351493848857646,
+      "loss": 0.9392287135124207,
+      "step": 590
+    },
+    {
+      "epoch": 0.249789029535865,
+      "grad_norm": 0.618262767791748,
+      "learning_rate": 0.0001038664323374341,
+      "loss": 0.9600516557693481,
+      "step": 592
+    },
+    {
+      "epoch": 0.25063291139240507,
+      "grad_norm": 0.6003909111022949,
+      "learning_rate": 0.00010421792618629173,
+      "loss": 1.005476713180542,
+      "step": 594
+    },
+    {
+      "epoch": 0.2514767932489452,
+      "grad_norm": 0.5437078475952148,
+      "learning_rate": 0.00010456942003514938,
+      "loss": 0.9523017406463623,
+      "step": 596
+    },
+    {
+      "epoch": 0.2523206751054852,
+      "grad_norm": 0.5524541735649109,
+      "learning_rate": 0.00010492091388400705,
+      "loss": 0.9526668787002563,
+      "step": 598
+    },
+    {
+      "epoch": 0.25316455696202533,
+      "grad_norm": 0.679504930973053,
+      "learning_rate": 0.00010527240773286469,
+      "loss": 1.019660472869873,
+      "step": 600
+    },
+    {
+      "epoch": 0.25316455696202533,
+      "eval_loss": 1.0193854570388794,
+      "eval_runtime": 677.9523,
+      "eval_samples_per_second": 3.108,
+      "eval_steps_per_second": 3.108,
+      "step": 600
+    },
+    {
+      "epoch": 0.2540084388185654,
+      "grad_norm": 0.5646136999130249,
+      "learning_rate": 0.00010562390158172233,
+      "loss": 0.9910882711410522,
+      "step": 602
+    },
+    {
+      "epoch": 0.2548523206751055,
+      "grad_norm": 0.5238093137741089,
+      "learning_rate": 0.00010597539543057998,
+      "loss": 0.9616432785987854,
+      "step": 604
+    },
+    {
+      "epoch": 0.25569620253164554,
+      "grad_norm": 0.7483857274055481,
+      "learning_rate": 0.00010632688927943762,
+      "loss": 1.0078275203704834,
+      "step": 606
+    },
+    {
+      "epoch": 0.25654008438818565,
+      "grad_norm": 0.578948974609375,
+      "learning_rate": 0.00010667838312829525,
+      "loss": 0.9827103018760681,
+      "step": 608
+    },
+    {
+      "epoch": 0.25738396624472576,
+      "grad_norm": 0.5525906085968018,
+      "learning_rate": 0.00010702987697715289,
+      "loss": 1.0423277616500854,
+      "step": 610
+    },
+    {
+      "epoch": 0.2582278481012658,
+      "grad_norm": 0.6721326112747192,
+      "learning_rate": 0.00010738137082601054,
+      "loss": 0.9561693072319031,
+      "step": 612
+    },
+    {
+      "epoch": 0.2590717299578059,
+      "grad_norm": 0.5701051354408264,
+      "learning_rate": 0.00010773286467486821,
+      "loss": 0.9602992534637451,
+      "step": 614
+    },
+    {
+      "epoch": 0.25991561181434597,
+      "grad_norm": 0.6349860429763794,
+      "learning_rate": 0.00010808435852372585,
+      "loss": 1.1422650814056396,
+      "step": 616
+    },
+    {
+      "epoch": 0.2607594936708861,
+      "grad_norm": 0.5496085286140442,
+      "learning_rate": 0.00010843585237258349,
+      "loss": 0.9762773513793945,
+      "step": 618
+    },
+    {
+      "epoch": 0.2616033755274262,
+      "grad_norm": 0.6080722808837891,
+      "learning_rate": 0.00010878734622144114,
+      "loss": 1.0133616924285889,
+      "step": 620
+    },
+    {
+      "epoch": 0.26244725738396624,
+      "grad_norm": 0.5450218915939331,
+      "learning_rate": 0.00010913884007029878,
+      "loss": 0.9385587573051453,
+      "step": 622
+    },
+    {
+      "epoch": 0.26329113924050634,
+      "grad_norm": 0.592106819152832,
+      "learning_rate": 0.00010949033391915641,
+      "loss": 0.9359989762306213,
+      "step": 624
+    },
+    {
+      "epoch": 0.2641350210970464,
+      "grad_norm": 0.6449427604675293,
+      "learning_rate": 0.00010984182776801407,
+      "loss": 1.0266027450561523,
+      "step": 626
+    },
+    {
+      "epoch": 0.2649789029535865,
+      "grad_norm": 0.538299560546875,
+      "learning_rate": 0.0001101933216168717,
+      "loss": 0.9303187131881714,
+      "step": 628
+    },
+    {
+      "epoch": 0.26582278481012656,
+      "grad_norm": 0.546316921710968,
+      "learning_rate": 0.00011054481546572934,
+      "loss": 0.9368857145309448,
+      "step": 630
+    },
+    {
+      "epoch": 0.26666666666666666,
+      "grad_norm": 0.5818730592727661,
+      "learning_rate": 0.00011089630931458701,
+      "loss": 0.9573145508766174,
+      "step": 632
+    },
+    {
+      "epoch": 0.26751054852320677,
+      "grad_norm": 0.5958262085914612,
+      "learning_rate": 0.00011124780316344465,
+      "loss": 0.9345449805259705,
+      "step": 634
+    },
+    {
+      "epoch": 0.2683544303797468,
+      "grad_norm": 0.6259077787399292,
+      "learning_rate": 0.0001115992970123023,
+      "loss": 1.0906590223312378,
+      "step": 636
+    },
+    {
+      "epoch": 0.26919831223628693,
+      "grad_norm": 0.589672863483429,
+      "learning_rate": 0.00011195079086115994,
+      "loss": 1.0757447481155396,
+      "step": 638
+    },
+    {
+      "epoch": 0.270042194092827,
+      "grad_norm": 0.5714080333709717,
+      "learning_rate": 0.00011230228471001758,
+      "loss": 0.9310855269432068,
+      "step": 640
+    },
+    {
+      "epoch": 0.2708860759493671,
+      "grad_norm": 0.45342639088630676,
+      "learning_rate": 0.00011265377855887523,
+      "loss": 0.9276360273361206,
+      "step": 642
+    },
+    {
+      "epoch": 0.2717299578059072,
+      "grad_norm": 0.6386750340461731,
+      "learning_rate": 0.00011300527240773287,
+      "loss": 1.084719181060791,
+      "step": 644
+    },
+    {
+      "epoch": 0.27257383966244725,
+      "grad_norm": 0.6446163654327393,
+      "learning_rate": 0.0001133567662565905,
+      "loss": 0.9763918519020081,
+      "step": 646
+    },
+    {
+      "epoch": 0.27341772151898736,
+      "grad_norm": 0.5925686359405518,
+      "learning_rate": 0.00011370826010544816,
+      "loss": 0.9517921805381775,
+      "step": 648
+    },
+    {
+      "epoch": 0.2742616033755274,
+      "grad_norm": 0.5399773716926575,
+      "learning_rate": 0.00011405975395430582,
+      "loss": 1.0587927103042603,
+      "step": 650
+    },
+    {
+      "epoch": 0.2751054852320675,
+      "grad_norm": 0.5872456431388855,
+      "learning_rate": 0.00011441124780316346,
+      "loss": 0.883341908454895,
+      "step": 652
+    },
+    {
+      "epoch": 0.2759493670886076,
+      "grad_norm": 0.5574564337730408,
+      "learning_rate": 0.0001147627416520211,
+      "loss": 1.0306891202926636,
+      "step": 654
+    },
+    {
+      "epoch": 0.2767932489451477,
+      "grad_norm": 0.47789013385772705,
+      "learning_rate": 0.00011511423550087874,
+      "loss": 0.8814032077789307,
+      "step": 656
+    },
+    {
+      "epoch": 0.2776371308016878,
+      "grad_norm": 0.5565530061721802,
+      "learning_rate": 0.00011546572934973639,
+      "loss": 0.9460552334785461,
+      "step": 658
+    },
+    {
+      "epoch": 0.27848101265822783,
+      "grad_norm": 0.5299761295318604,
+      "learning_rate": 0.00011581722319859403,
+      "loss": 0.9475110769271851,
+      "step": 660
+    },
+    {
+      "epoch": 0.27932489451476794,
+      "grad_norm": 0.6503344178199768,
+      "learning_rate": 0.00011616871704745166,
+      "loss": 1.0630913972854614,
+      "step": 662
+    },
+    {
+      "epoch": 0.280168776371308,
+      "grad_norm": 0.5794585943222046,
+      "learning_rate": 0.00011652021089630932,
+      "loss": 0.9389138221740723,
+      "step": 664
+    },
+    {
+      "epoch": 0.2810126582278481,
+      "grad_norm": 0.5762867331504822,
+      "learning_rate": 0.00011687170474516695,
+      "loss": 0.8934136033058167,
+      "step": 666
+    },
+    {
+      "epoch": 0.2818565400843882,
+      "grad_norm": 0.6565435528755188,
+      "learning_rate": 0.00011722319859402462,
+      "loss": 1.1072614192962646,
+      "step": 668
+    },
+    {
+      "epoch": 0.28270042194092826,
+      "grad_norm": 0.5819830298423767,
+      "learning_rate": 0.00011757469244288226,
+      "loss": 1.0501434803009033,
+      "step": 670
+    },
+    {
+      "epoch": 0.28354430379746837,
+      "grad_norm": 0.6071487069129944,
+      "learning_rate": 0.00011792618629173991,
+      "loss": 0.9880793690681458,
+      "step": 672
+    },
+    {
+      "epoch": 0.2843881856540084,
+      "grad_norm": 0.5765058398246765,
+      "learning_rate": 0.00011827768014059755,
+      "loss": 0.9670693874359131,
+      "step": 674
+    },
+    {
+      "epoch": 0.2852320675105485,
+      "grad_norm": 0.5245351791381836,
+      "learning_rate": 0.00011862917398945519,
+      "loss": 0.9602360725402832,
+      "step": 676
+    },
+    {
+      "epoch": 0.28607594936708863,
+      "grad_norm": 0.6189922094345093,
+      "learning_rate": 0.00011898066783831282,
+      "loss": 0.9684560894966125,
+      "step": 678
+    },
+    {
+      "epoch": 0.2869198312236287,
+      "grad_norm": 0.6138690710067749,
+      "learning_rate": 0.00011933216168717048,
+      "loss": 0.9465792775154114,
+      "step": 680
+    },
+    {
+      "epoch": 0.2877637130801688,
+      "grad_norm": 0.5371595621109009,
+      "learning_rate": 0.00011968365553602812,
+      "loss": 0.8495944738388062,
+      "step": 682
+    },
+    {
+      "epoch": 0.28860759493670884,
+      "grad_norm": 0.5549944639205933,
+      "learning_rate": 0.00012003514938488578,
+      "loss": 0.9663267135620117,
+      "step": 684
+    },
+    {
+      "epoch": 0.28945147679324895,
+      "grad_norm": 0.6484189033508301,
+      "learning_rate": 0.00012038664323374342,
+      "loss": 0.9736058712005615,
+      "step": 686
+    },
+    {
+      "epoch": 0.290295358649789,
+      "grad_norm": 0.540351390838623,
+      "learning_rate": 0.00012073813708260107,
+      "loss": 1.0591845512390137,
+      "step": 688
+    },
+    {
+      "epoch": 0.2911392405063291,
+      "grad_norm": 0.5657922029495239,
+      "learning_rate": 0.00012108963093145871,
+      "loss": 0.944908618927002,
+      "step": 690
+    },
+    {
+      "epoch": 0.2919831223628692,
+      "grad_norm": 0.6040505170822144,
+      "learning_rate": 0.00012144112478031635,
+      "loss": 1.0018219947814941,
+      "step": 692
+    },
+    {
+      "epoch": 0.29282700421940927,
+      "grad_norm": 0.5435477495193481,
+      "learning_rate": 0.000121792618629174,
+      "loss": 1.0351502895355225,
+      "step": 694
+    },
+    {
+      "epoch": 0.2936708860759494,
+      "grad_norm": 0.5712518692016602,
+      "learning_rate": 0.00012214411247803164,
+      "loss": 0.9935672283172607,
+      "step": 696
+    },
+    {
+      "epoch": 0.29451476793248943,
+      "grad_norm": 0.6138222813606262,
+      "learning_rate": 0.00012249560632688928,
+      "loss": 1.0165108442306519,
+      "step": 698
+    },
+    {
+      "epoch": 0.29535864978902954,
+      "grad_norm": 0.4495212435722351,
+      "learning_rate": 0.00012284710017574691,
+      "loss": 0.9334425926208496,
+      "step": 700
+    },
+    {
+      "epoch": 0.29535864978902954,
+      "eval_loss": 0.996929407119751,
+      "eval_runtime": 668.6398,
+      "eval_samples_per_second": 3.151,
+      "eval_steps_per_second": 3.151,
+      "step": 700
+    },
+    {
+      "epoch": 0.29620253164556964,
+      "grad_norm": 0.5321539044380188,
+      "learning_rate": 0.00012319859402460458,
+      "loss": 1.0516537427902222,
+      "step": 702
+    },
+    {
+      "epoch": 0.2970464135021097,
+      "grad_norm": 0.5716516971588135,
+      "learning_rate": 0.00012355008787346222,
+      "loss": 0.9387198686599731,
+      "step": 704
+    },
+    {
+      "epoch": 0.2978902953586498,
+      "grad_norm": 0.5617920160293579,
+      "learning_rate": 0.00012390158172231988,
+      "loss": 0.99737948179245,
+      "step": 706
+    },
+    {
+      "epoch": 0.29873417721518986,
+      "grad_norm": 0.4922899007797241,
+      "learning_rate": 0.00012425307557117752,
+      "loss": 0.9955025911331177,
+      "step": 708
+    },
+    {
+      "epoch": 0.29957805907172996,
+      "grad_norm": 0.543501615524292,
+      "learning_rate": 0.00012460456942003516,
+      "loss": 0.9124280214309692,
+      "step": 710
+    },
+    {
+      "epoch": 0.30042194092827,
+      "grad_norm": 0.49590054154396057,
+      "learning_rate": 0.0001249560632688928,
+      "loss": 0.9820216298103333,
+      "step": 712
+    },
+    {
+      "epoch": 0.3012658227848101,
+      "grad_norm": 0.5984305739402771,
+      "learning_rate": 0.00012530755711775044,
+      "loss": 1.0152074098587036,
+      "step": 714
+    },
+    {
+      "epoch": 0.30210970464135023,
+      "grad_norm": 0.9343504905700684,
+      "learning_rate": 0.00012565905096660807,
+      "loss": 1.0577725172042847,
+      "step": 716
+    },
+    {
+      "epoch": 0.3029535864978903,
+      "grad_norm": 0.5118702054023743,
+      "learning_rate": 0.0001260105448154657,
+      "loss": 0.9830358028411865,
+      "step": 718
+    },
+    {
+      "epoch": 0.3037974683544304,
+      "grad_norm": 0.4940392076969147,
+      "learning_rate": 0.00012636203866432338,
+      "loss": 0.9466043710708618,
+      "step": 720
+    },
+    {
+      "epoch": 0.30464135021097044,
+      "grad_norm": 0.5965693593025208,
+      "learning_rate": 0.00012671353251318104,
+      "loss": 1.015270709991455,
+      "step": 722
+    },
+    {
+      "epoch": 0.30548523206751055,
+      "grad_norm": 0.5020529627799988,
+      "learning_rate": 0.00012706502636203868,
+      "loss": 0.9703927636146545,
+      "step": 724
+    },
+    {
+      "epoch": 0.30632911392405066,
+      "grad_norm": 0.6067010164260864,
+      "learning_rate": 0.00012741652021089632,
+      "loss": 1.0255526304244995,
+      "step": 726
+    },
+    {
+      "epoch": 0.3071729957805907,
+      "grad_norm": 0.5931884050369263,
+      "learning_rate": 0.00012776801405975396,
+      "loss": 0.9335633516311646,
+      "step": 728
+    },
+    {
+      "epoch": 0.3080168776371308,
+      "grad_norm": 0.5938752293586731,
+      "learning_rate": 0.0001281195079086116,
+      "loss": 1.0921578407287598,
+      "step": 730
+    },
+    {
+      "epoch": 0.30886075949367087,
+      "grad_norm": 0.49728086590766907,
+      "learning_rate": 0.00012847100175746923,
+      "loss": 0.963066041469574,
+      "step": 732
+    },
+    {
+      "epoch": 0.309704641350211,
+      "grad_norm": 0.5452080965042114,
+      "learning_rate": 0.0001288224956063269,
+      "loss": 0.9513075351715088,
+      "step": 734
+    },
+    {
+      "epoch": 0.3105485232067511,
+      "grad_norm": 0.5497731566429138,
+      "learning_rate": 0.00012917398945518454,
+      "loss": 0.8576077222824097,
+      "step": 736
+    },
+    {
+      "epoch": 0.31139240506329113,
+      "grad_norm": 0.5580397248268127,
+      "learning_rate": 0.0001295254833040422,
+      "loss": 0.9542577862739563,
+      "step": 738
+    },
+    {
+      "epoch": 0.31223628691983124,
+      "grad_norm": 0.5890427827835083,
+      "learning_rate": 0.00012987697715289984,
+      "loss": 0.8992732167243958,
+      "step": 740
+    },
+    {
+      "epoch": 0.3130801687763713,
+      "grad_norm": 0.5942965745925903,
+      "learning_rate": 0.00013022847100175748,
+      "loss": 1.0322896242141724,
+      "step": 742
+    },
+    {
+      "epoch": 0.3139240506329114,
+      "grad_norm": 0.6341713070869446,
+      "learning_rate": 0.00013057996485061512,
+      "loss": 0.9217103719711304,
+      "step": 744
+    },
+    {
+      "epoch": 0.31476793248945145,
+      "grad_norm": 0.5294105410575867,
+      "learning_rate": 0.00013093145869947276,
+      "loss": 0.951789915561676,
+      "step": 746
+    },
+    {
+      "epoch": 0.31561181434599156,
+      "grad_norm": 0.6372058391571045,
+      "learning_rate": 0.0001312829525483304,
+      "loss": 0.9459875226020813,
+      "step": 748
+    },
+    {
+      "epoch": 0.31645569620253167,
+      "grad_norm": 0.5979796648025513,
+      "learning_rate": 0.00013163444639718806,
+      "loss": 0.9626097679138184,
+      "step": 750
+    },
+    {
+      "epoch": 0.3172995780590717,
+      "grad_norm": 0.5682399868965149,
+      "learning_rate": 0.0001319859402460457,
+      "loss": 1.0261781215667725,
+      "step": 752
+    },
+    {
+      "epoch": 0.3181434599156118,
+      "grad_norm": 0.5349125266075134,
+      "learning_rate": 0.00013233743409490336,
+      "loss": 0.9319828152656555,
+      "step": 754
+    },
+    {
+      "epoch": 0.3189873417721519,
+      "grad_norm": 0.6093934178352356,
+      "learning_rate": 0.000132688927943761,
+      "loss": 0.9216550588607788,
+      "step": 756
+    },
+    {
+      "epoch": 0.319831223628692,
+      "grad_norm": 0.5188612341880798,
+      "learning_rate": 0.00013304042179261864,
+      "loss": 0.901739776134491,
+      "step": 758
+    },
+    {
+      "epoch": 0.3206751054852321,
+      "grad_norm": 0.5877130627632141,
+      "learning_rate": 0.00013339191564147628,
+      "loss": 1.0362589359283447,
+      "step": 760
+    },
+    {
+      "epoch": 0.32151898734177214,
+      "grad_norm": 0.5542771816253662,
+      "learning_rate": 0.00013374340949033392,
+      "loss": 0.8787116408348083,
+      "step": 762
+    },
+    {
+      "epoch": 0.32236286919831225,
+      "grad_norm": 0.5084902048110962,
+      "learning_rate": 0.00013409490333919156,
+      "loss": 0.9237037301063538,
+      "step": 764
+    },
+    {
+      "epoch": 0.3232067510548523,
+      "grad_norm": 0.5461528301239014,
+      "learning_rate": 0.00013444639718804922,
+      "loss": 1.0150731801986694,
+      "step": 766
+    },
+    {
+      "epoch": 0.3240506329113924,
+      "grad_norm": 0.53483647108078,
+      "learning_rate": 0.00013479789103690686,
+      "loss": 0.8985214829444885,
+      "step": 768
+    },
+    {
+      "epoch": 0.32489451476793246,
+      "grad_norm": 0.5580531358718872,
+      "learning_rate": 0.0001351493848857645,
+      "loss": 1.0225775241851807,
+      "step": 770
+    },
+    {
+      "epoch": 0.32573839662447257,
+      "grad_norm": 0.5203377604484558,
+      "learning_rate": 0.00013550087873462216,
+      "loss": 0.9571293592453003,
+      "step": 772
+    },
+    {
+      "epoch": 0.3265822784810127,
+      "grad_norm": 0.5049671530723572,
+      "learning_rate": 0.0001358523725834798,
+      "loss": 1.0468909740447998,
+      "step": 774
+    },
+    {
+      "epoch": 0.32742616033755273,
+      "grad_norm": 0.4723063111305237,
+      "learning_rate": 0.00013620386643233744,
+      "loss": 0.7743215560913086,
+      "step": 776
+    },
+    {
+      "epoch": 0.32827004219409284,
+      "grad_norm": 0.6310980916023254,
+      "learning_rate": 0.00013655536028119508,
+      "loss": 1.021510362625122,
+      "step": 778
+    },
+    {
+      "epoch": 0.3291139240506329,
+      "grad_norm": 0.47066664695739746,
+      "learning_rate": 0.00013690685413005274,
+      "loss": 0.9134382605552673,
+      "step": 780
+    },
+    {
+      "epoch": 0.329957805907173,
+      "grad_norm": 0.5725092887878418,
+      "learning_rate": 0.00013725834797891038,
+      "loss": 0.9797834753990173,
+      "step": 782
+    },
+    {
+      "epoch": 0.3308016877637131,
+      "grad_norm": 0.5139563083648682,
+      "learning_rate": 0.00013760984182776802,
+      "loss": 0.9372621178627014,
+      "step": 784
+    },
+    {
+      "epoch": 0.33164556962025316,
+      "grad_norm": 0.5275821685791016,
+      "learning_rate": 0.00013796133567662566,
+      "loss": 0.9528245329856873,
+      "step": 786
+    },
+    {
+      "epoch": 0.33248945147679326,
+      "grad_norm": 0.5702582001686096,
+      "learning_rate": 0.0001383128295254833,
+      "loss": 0.9750176072120667,
+      "step": 788
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.5281293392181396,
+      "learning_rate": 0.00013866432337434096,
+      "loss": 0.9412306547164917,
+      "step": 790
+    },
+    {
+      "epoch": 0.3341772151898734,
+      "grad_norm": 0.5578986406326294,
+      "learning_rate": 0.0001390158172231986,
+      "loss": 0.997580885887146,
+      "step": 792
+    },
+    {
+      "epoch": 0.33502109704641353,
+      "grad_norm": 0.50461345911026,
+      "learning_rate": 0.00013936731107205624,
+      "loss": 0.9082320928573608,
+      "step": 794
+    },
+    {
+      "epoch": 0.3358649789029536,
+      "grad_norm": 0.5258530378341675,
+      "learning_rate": 0.0001397188049209139,
+      "loss": 1.0082844495773315,
+      "step": 796
+    },
+    {
+      "epoch": 0.3367088607594937,
+      "grad_norm": 0.5548169016838074,
+      "learning_rate": 0.00014007029876977154,
+      "loss": 0.9729003310203552,
+      "step": 798
+    },
+    {
+      "epoch": 0.33755274261603374,
+      "grad_norm": 0.48601076006889343,
+      "learning_rate": 0.00014042179261862918,
+      "loss": 0.9099526405334473,
+      "step": 800
+    },
+    {
+      "epoch": 0.33755274261603374,
+      "eval_loss": 0.9800403714179993,
+      "eval_runtime": 678.8306,
+      "eval_samples_per_second": 3.104,
+      "eval_steps_per_second": 3.104,
+      "step": 800
+    },
+    {
+      "epoch": 0.33839662447257385,
+      "grad_norm": 0.5413158535957336,
+      "learning_rate": 0.00014077328646748682,
+      "loss": 0.8610644936561584,
+      "step": 802
+    },
+    {
+      "epoch": 0.3392405063291139,
+      "grad_norm": 0.5147035717964172,
+      "learning_rate": 0.00014112478031634446,
+      "loss": 0.9584825038909912,
+      "step": 804
+    },
+    {
+      "epoch": 0.340084388185654,
+      "grad_norm": 0.5931771397590637,
+      "learning_rate": 0.0001414762741652021,
+      "loss": 1.0142558813095093,
+      "step": 806
+    },
+    {
+      "epoch": 0.3409282700421941,
+      "grad_norm": 0.5178377032279968,
+      "learning_rate": 0.00014182776801405976,
+      "loss": 1.0078763961791992,
+      "step": 808
+    },
+    {
+      "epoch": 0.34177215189873417,
+      "grad_norm": 0.5453237295150757,
+      "learning_rate": 0.0001421792618629174,
+      "loss": 0.9107215404510498,
+      "step": 810
+    },
+    {
+      "epoch": 0.3426160337552743,
+      "grad_norm": 0.5886152982711792,
+      "learning_rate": 0.00014253075571177506,
+      "loss": 0.9981362819671631,
+      "step": 812
+    },
+    {
+      "epoch": 0.3434599156118143,
+      "grad_norm": 0.48040178418159485,
+      "learning_rate": 0.0001428822495606327,
+      "loss": 0.9636131525039673,
+      "step": 814
+    },
+    {
+      "epoch": 0.34430379746835443,
+      "grad_norm": 0.5011753439903259,
+      "learning_rate": 0.00014323374340949034,
+      "loss": 0.9590586423873901,
+      "step": 816
+    },
+    {
+      "epoch": 0.34514767932489454,
+      "grad_norm": 0.57858806848526,
+      "learning_rate": 0.00014358523725834798,
+      "loss": 0.978246808052063,
+      "step": 818
+    },
+    {
+      "epoch": 0.3459915611814346,
+      "grad_norm": 0.46092939376831055,
+      "learning_rate": 0.00014393673110720562,
+      "loss": 0.8549934029579163,
+      "step": 820
+    },
+    {
+      "epoch": 0.3468354430379747,
+      "grad_norm": 0.5756489038467407,
+      "learning_rate": 0.00014428822495606326,
+      "loss": 0.9771265387535095,
+      "step": 822
+    },
+    {
+      "epoch": 0.34767932489451475,
+      "grad_norm": 0.5501731634140015,
+      "learning_rate": 0.00014463971880492092,
+      "loss": 0.8739748001098633,
+      "step": 824
+    },
+    {
+      "epoch": 0.34852320675105486,
+      "grad_norm": 0.5451868176460266,
+      "learning_rate": 0.0001449912126537786,
+      "loss": 0.9129468202590942,
+      "step": 826
+    },
+    {
+      "epoch": 0.3493670886075949,
+      "grad_norm": 0.4624619781970978,
+      "learning_rate": 0.00014534270650263623,
+      "loss": 0.9196704030036926,
+      "step": 828
+    },
+    {
+      "epoch": 0.350210970464135,
+      "grad_norm": 0.520878791809082,
+      "learning_rate": 0.00014569420035149386,
+      "loss": 0.9976527690887451,
+      "step": 830
+    },
+    {
+      "epoch": 0.3510548523206751,
+      "grad_norm": 0.4469171464443207,
+      "learning_rate": 0.0001460456942003515,
+      "loss": 0.7753443717956543,
+      "step": 832
+    },
+    {
+      "epoch": 0.3518987341772152,
+      "grad_norm": 0.5105249881744385,
+      "learning_rate": 0.00014639718804920914,
+      "loss": 0.9584846496582031,
+      "step": 834
+    },
+    {
+      "epoch": 0.3527426160337553,
+      "grad_norm": 0.5043913125991821,
+      "learning_rate": 0.00014674868189806678,
+      "loss": 0.9013500213623047,
+      "step": 836
+    },
+    {
+      "epoch": 0.35358649789029534,
+      "grad_norm": 0.575850784778595,
+      "learning_rate": 0.00014710017574692442,
+      "loss": 0.9775562286376953,
+      "step": 838
+    },
+    {
+      "epoch": 0.35443037974683544,
+      "grad_norm": 0.5128876566886902,
+      "learning_rate": 0.00014745166959578208,
+      "loss": 0.9278940558433533,
+      "step": 840
+    },
+    {
+      "epoch": 0.35527426160337555,
+      "grad_norm": 0.5757885575294495,
+      "learning_rate": 0.00014780316344463975,
+      "loss": 1.0091488361358643,
+      "step": 842
+    },
+    {
+      "epoch": 0.3561181434599156,
+      "grad_norm": 0.500934898853302,
+      "learning_rate": 0.00014815465729349739,
+      "loss": 0.9286836981773376,
+      "step": 844
+    },
+    {
+      "epoch": 0.3569620253164557,
+      "grad_norm": 0.5220686197280884,
+      "learning_rate": 0.00014850615114235502,
+      "loss": 0.9484171867370605,
+      "step": 846
+    },
+    {
+      "epoch": 0.35780590717299576,
+      "grad_norm": 0.5494697690010071,
+      "learning_rate": 0.00014885764499121266,
+      "loss": 1.0556397438049316,
+      "step": 848
+    },
+    {
+      "epoch": 0.35864978902953587,
+      "grad_norm": 0.46633943915367126,
+      "learning_rate": 0.0001492091388400703,
+      "loss": 0.933089017868042,
+      "step": 850
+    },
+    {
+      "epoch": 0.3594936708860759,
+      "grad_norm": 0.4864962697029114,
+      "learning_rate": 0.00014956063268892794,
+      "loss": 1.0016963481903076,
+      "step": 852
+    },
+    {
+      "epoch": 0.36033755274261603,
+      "grad_norm": 0.5032764673233032,
+      "learning_rate": 0.0001499121265377856,
+      "loss": 0.9085348844528198,
+      "step": 854
+    },
+    {
+      "epoch": 0.36118143459915614,
+      "grad_norm": 0.5424998998641968,
+      "learning_rate": 0.00015026362038664324,
+      "loss": 0.923502504825592,
+      "step": 856
+    },
+    {
+      "epoch": 0.3620253164556962,
+      "grad_norm": 0.5181655287742615,
+      "learning_rate": 0.00015061511423550088,
+      "loss": 0.919174313545227,
+      "step": 858
+    },
+    {
+      "epoch": 0.3628691983122363,
+      "grad_norm": 0.5129443407058716,
+      "learning_rate": 0.00015096660808435855,
+      "loss": 0.851981520652771,
+      "step": 860
+    },
+    {
+      "epoch": 0.36371308016877635,
+      "grad_norm": 0.49540698528289795,
+      "learning_rate": 0.00015131810193321618,
+      "loss": 0.8633858561515808,
+      "step": 862
+    },
+    {
+      "epoch": 0.36455696202531646,
+      "grad_norm": 0.4706701934337616,
+      "learning_rate": 0.00015166959578207382,
+      "loss": 0.9473677277565002,
+      "step": 864
+    },
+    {
+      "epoch": 0.36540084388185656,
+      "grad_norm": 0.587704598903656,
+      "learning_rate": 0.00015202108963093146,
+      "loss": 0.942383885383606,
+      "step": 866
+    },
+    {
+      "epoch": 0.3662447257383966,
+      "grad_norm": 0.5851273536682129,
+      "learning_rate": 0.0001523725834797891,
+      "loss": 0.9811251163482666,
+      "step": 868
+    },
+    {
+      "epoch": 0.3670886075949367,
+      "grad_norm": 0.46357613801956177,
+      "learning_rate": 0.00015272407732864676,
+      "loss": 0.8151084184646606,
+      "step": 870
+    },
+    {
+      "epoch": 0.3679324894514768,
+      "grad_norm": 0.47250670194625854,
+      "learning_rate": 0.0001530755711775044,
+      "loss": 0.9023333191871643,
+      "step": 872
+    },
+    {
+      "epoch": 0.3687763713080169,
+      "grad_norm": 0.47510042786598206,
+      "learning_rate": 0.00015342706502636204,
+      "loss": 0.8758499622344971,
+      "step": 874
+    },
+    {
+      "epoch": 0.369620253164557,
+      "grad_norm": 0.5687124729156494,
+      "learning_rate": 0.00015377855887521968,
+      "loss": 0.9837421774864197,
+      "step": 876
+    },
+    {
+      "epoch": 0.37046413502109704,
+      "grad_norm": 0.49064236879348755,
+      "learning_rate": 0.00015413005272407735,
+      "loss": 0.9084216356277466,
+      "step": 878
+    },
+    {
+      "epoch": 0.37130801687763715,
+      "grad_norm": 0.5288164615631104,
+      "learning_rate": 0.00015448154657293498,
+      "loss": 0.8425542712211609,
+      "step": 880
+    },
+    {
+      "epoch": 0.3721518987341772,
+      "grad_norm": 0.5446951985359192,
+      "learning_rate": 0.00015483304042179262,
+      "loss": 1.006197452545166,
+      "step": 882
+    },
+    {
+      "epoch": 0.3729957805907173,
+      "grad_norm": 0.46872642636299133,
+      "learning_rate": 0.00015518453427065026,
+      "loss": 0.9779444932937622,
+      "step": 884
+    },
+    {
+      "epoch": 0.37383966244725736,
+      "grad_norm": 0.4852714240550995,
+      "learning_rate": 0.00015553602811950793,
+      "loss": 0.9099963307380676,
+      "step": 886
+    },
+    {
+      "epoch": 0.37468354430379747,
+      "grad_norm": 0.5219841003417969,
+      "learning_rate": 0.00015588752196836556,
+      "loss": 0.9730570316314697,
+      "step": 888
+    },
+    {
+      "epoch": 0.3755274261603376,
+      "grad_norm": 0.5258626341819763,
+      "learning_rate": 0.0001562390158172232,
+      "loss": 0.9304586052894592,
+      "step": 890
+    },
+    {
+      "epoch": 0.3763713080168776,
+      "grad_norm": 0.4266716241836548,
+      "learning_rate": 0.00015659050966608084,
+      "loss": 0.8680716156959534,
+      "step": 892
+    },
+    {
+      "epoch": 0.37721518987341773,
+      "grad_norm": 0.46361327171325684,
+      "learning_rate": 0.0001569420035149385,
+      "loss": 0.8428018093109131,
+      "step": 894
+    },
+    {
+      "epoch": 0.3780590717299578,
+      "grad_norm": 0.5313687920570374,
+      "learning_rate": 0.00015729349736379614,
+      "loss": 0.8465750217437744,
+      "step": 896
+    },
+    {
+      "epoch": 0.3789029535864979,
+      "grad_norm": 0.47166210412979126,
+      "learning_rate": 0.00015764499121265378,
+      "loss": 0.9120327234268188,
+      "step": 898
+    },
+    {
+      "epoch": 0.379746835443038,
+      "grad_norm": 0.4647318720817566,
+      "learning_rate": 0.00015799648506151145,
+      "loss": 0.8950425982475281,
+      "step": 900
+    },
+    {
+      "epoch": 0.379746835443038,
+      "eval_loss": 0.9643027186393738,
+      "eval_runtime": 691.7929,
+      "eval_samples_per_second": 3.046,
+      "eval_steps_per_second": 3.046,
+      "step": 900
+    },
+    {
+      "epoch": 0.38059071729957805,
+      "grad_norm": 0.5445119738578796,
+      "learning_rate": 0.00015834797891036909,
+      "loss": 0.908163845539093,
+      "step": 902
+    },
+    {
+      "epoch": 0.38143459915611816,
+      "grad_norm": 0.4311858117580414,
+      "learning_rate": 0.00015869947275922672,
+      "loss": 0.8945821523666382,
+      "step": 904
+    },
+    {
+      "epoch": 0.3822784810126582,
+      "grad_norm": 0.5590984225273132,
+      "learning_rate": 0.00015905096660808436,
+      "loss": 0.9478458762168884,
+      "step": 906
+    },
+    {
+      "epoch": 0.3831223628691983,
+      "grad_norm": 0.5470241904258728,
+      "learning_rate": 0.000159402460456942,
+      "loss": 0.9259957671165466,
+      "step": 908
+    },
+    {
+      "epoch": 0.38396624472573837,
+      "grad_norm": 0.5498791337013245,
+      "learning_rate": 0.00015975395430579964,
+      "loss": 0.8824930787086487,
+      "step": 910
+    },
+    {
+      "epoch": 0.3848101265822785,
+      "grad_norm": 0.4779198467731476,
+      "learning_rate": 0.0001601054481546573,
+      "loss": 0.8842340707778931,
+      "step": 912
+    },
+    {
+      "epoch": 0.3856540084388186,
+      "grad_norm": 0.5390620827674866,
+      "learning_rate": 0.00016045694200351494,
+      "loss": 0.92950040102005,
+      "step": 914
+    },
+    {
+      "epoch": 0.38649789029535864,
+      "grad_norm": 0.505519688129425,
+      "learning_rate": 0.0001608084358523726,
+      "loss": 0.8420897126197815,
+      "step": 916
+    },
+    {
+      "epoch": 0.38734177215189874,
+      "grad_norm": 0.4463907778263092,
+      "learning_rate": 0.00016115992970123025,
+      "loss": 0.8256624341011047,
+      "step": 918
+    },
+    {
+      "epoch": 0.3881856540084388,
+      "grad_norm": 0.5321422219276428,
+      "learning_rate": 0.00016151142355008788,
+      "loss": 0.8701168298721313,
+      "step": 920
+    },
+    {
+      "epoch": 0.3890295358649789,
+      "grad_norm": 0.4343073070049286,
+      "learning_rate": 0.00016186291739894552,
+      "loss": 0.9019309878349304,
+      "step": 922
+    },
+    {
+      "epoch": 0.389873417721519,
+      "grad_norm": 0.5311984419822693,
+      "learning_rate": 0.00016221441124780316,
+      "loss": 0.8560551404953003,
+      "step": 924
+    },
+    {
+      "epoch": 0.39071729957805906,
+      "grad_norm": 0.588691771030426,
+      "learning_rate": 0.0001625659050966608,
+      "loss": 0.9700050354003906,
+      "step": 926
+    },
+    {
+      "epoch": 0.39156118143459917,
+      "grad_norm": 0.5425586104393005,
+      "learning_rate": 0.00016291739894551844,
+      "loss": 1.043768048286438,
+      "step": 928
+    },
+    {
+      "epoch": 0.3924050632911392,
+      "grad_norm": 0.5228736996650696,
+      "learning_rate": 0.0001632688927943761,
+      "loss": 0.9501712918281555,
+      "step": 930
+    },
+    {
+      "epoch": 0.39324894514767933,
+      "grad_norm": 0.48960360884666443,
+      "learning_rate": 0.00016362038664323377,
+      "loss": 0.9223058223724365,
+      "step": 932
+    },
+    {
+      "epoch": 0.39409282700421944,
+      "grad_norm": 0.45204755663871765,
+      "learning_rate": 0.0001639718804920914,
+      "loss": 0.9692960977554321,
+      "step": 934
+    },
+    {
+      "epoch": 0.3949367088607595,
+      "grad_norm": 0.5299274921417236,
+      "learning_rate": 0.00016432337434094905,
+      "loss": 0.9467466473579407,
+      "step": 936
+    },
+    {
+      "epoch": 0.3957805907172996,
+      "grad_norm": 0.5607715249061584,
+      "learning_rate": 0.00016467486818980668,
+      "loss": 0.9118053317070007,
+      "step": 938
+    },
+    {
+      "epoch": 0.39662447257383965,
+      "grad_norm": 0.5271831154823303,
+      "learning_rate": 0.00016502636203866432,
+      "loss": 0.9131460189819336,
+      "step": 940
+    },
+    {
+      "epoch": 0.39746835443037976,
+      "grad_norm": 0.5075286030769348,
+      "learning_rate": 0.00016537785588752196,
+      "loss": 0.9358300566673279,
+      "step": 942
+    },
+    {
+      "epoch": 0.3983122362869198,
+      "grad_norm": 0.515731155872345,
+      "learning_rate": 0.00016572934973637963,
+      "loss": 0.8908210396766663,
+      "step": 944
+    },
+    {
+      "epoch": 0.3991561181434599,
+      "grad_norm": 0.4856977164745331,
+      "learning_rate": 0.00016608084358523726,
+      "loss": 0.9775290489196777,
+      "step": 946
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.48846355080604553,
+      "learning_rate": 0.00016643233743409493,
+      "loss": 0.8957490921020508,
+      "step": 948
+    },
+    {
+      "epoch": 0.4008438818565401,
+      "grad_norm": 0.42990800738334656,
+      "learning_rate": 0.00016678383128295257,
+      "loss": 0.9036174416542053,
+      "step": 950
+    },
+    {
+      "epoch": 0.4016877637130802,
+      "grad_norm": 0.49552062153816223,
+      "learning_rate": 0.0001671353251318102,
+      "loss": 0.991032600402832,
+      "step": 952
+    },
+    {
+      "epoch": 0.40253164556962023,
+      "grad_norm": 0.4565040171146393,
+      "learning_rate": 0.00016748681898066784,
+      "loss": 0.823063313961029,
+      "step": 954
+    },
+    {
+      "epoch": 0.40337552742616034,
+      "grad_norm": 0.4290153682231903,
+      "learning_rate": 0.00016783831282952548,
+      "loss": 0.8785063624382019,
+      "step": 956
+    },
+    {
+      "epoch": 0.40421940928270045,
+      "grad_norm": 0.5419702529907227,
+      "learning_rate": 0.00016818980667838312,
+      "loss": 0.8763971924781799,
+      "step": 958
+    },
+    {
+      "epoch": 0.4050632911392405,
+      "grad_norm": 0.5177501440048218,
+      "learning_rate": 0.00016854130052724079,
+      "loss": 0.9470553398132324,
+      "step": 960
+    },
+    {
+      "epoch": 0.4059071729957806,
+      "grad_norm": 0.539725661277771,
+      "learning_rate": 0.00016889279437609842,
+      "loss": 0.9235025644302368,
+      "step": 962
+    },
+    {
+      "epoch": 0.40675105485232066,
+      "grad_norm": 0.5324983596801758,
+      "learning_rate": 0.0001692442882249561,
+      "loss": 1.0248996019363403,
+      "step": 964
+    },
+    {
+      "epoch": 0.40759493670886077,
+      "grad_norm": 0.4936407506465912,
+      "learning_rate": 0.00016959578207381373,
+      "loss": 0.9076873660087585,
+      "step": 966
+    },
+    {
+      "epoch": 0.4084388185654008,
+      "grad_norm": 0.4960501194000244,
+      "learning_rate": 0.00016994727592267137,
+      "loss": 0.9162673950195312,
+      "step": 968
+    },
+    {
+      "epoch": 0.4092827004219409,
+      "grad_norm": 0.45093682408332825,
+      "learning_rate": 0.000170298769771529,
+      "loss": 0.904100775718689,
+      "step": 970
+    },
+    {
+      "epoch": 0.41012658227848103,
+      "grad_norm": 0.4560275077819824,
+      "learning_rate": 0.00017065026362038664,
+      "loss": 0.8633337020874023,
+      "step": 972
+    },
+    {
+      "epoch": 0.4109704641350211,
+      "grad_norm": 0.44885000586509705,
+      "learning_rate": 0.00017100175746924428,
+      "loss": 0.8454209566116333,
+      "step": 974
+    },
+    {
+      "epoch": 0.4118143459915612,
+      "grad_norm": 0.47251659631729126,
+      "learning_rate": 0.00017135325131810195,
+      "loss": 0.824730396270752,
+      "step": 976
+    },
+    {
+      "epoch": 0.41265822784810124,
+      "grad_norm": 0.6597666144371033,
+      "learning_rate": 0.00017170474516695959,
+      "loss": 0.9496501684188843,
+      "step": 978
+    },
+    {
+      "epoch": 0.41350210970464135,
+      "grad_norm": 0.39806297421455383,
+      "learning_rate": 0.00017205623901581722,
+      "loss": 0.9419087171554565,
+      "step": 980
+    },
+    {
+      "epoch": 0.41434599156118146,
+      "grad_norm": 0.48231109976768494,
+      "learning_rate": 0.0001724077328646749,
+      "loss": 0.9182976484298706,
+      "step": 982
+    },
+    {
+      "epoch": 0.4151898734177215,
+      "grad_norm": 0.5438776612281799,
+      "learning_rate": 0.00017275922671353253,
+      "loss": 0.9386967420578003,
+      "step": 984
+    },
+    {
+      "epoch": 0.4160337552742616,
+      "grad_norm": 0.4959667921066284,
+      "learning_rate": 0.00017311072056239017,
+      "loss": 0.897849440574646,
+      "step": 986
+    },
+    {
+      "epoch": 0.41687763713080167,
+      "grad_norm": 0.43533357977867126,
+      "learning_rate": 0.0001734622144112478,
+      "loss": 0.8776953816413879,
+      "step": 988
+    },
+    {
+      "epoch": 0.4177215189873418,
+      "grad_norm": 0.47513946890830994,
+      "learning_rate": 0.00017381370826010547,
+      "loss": 0.9162989854812622,
+      "step": 990
+    },
+    {
+      "epoch": 0.41856540084388183,
+      "grad_norm": 0.4907188415527344,
+      "learning_rate": 0.0001741652021089631,
+      "loss": 0.9482660889625549,
+      "step": 992
+    },
+    {
+      "epoch": 0.41940928270042194,
+      "grad_norm": 0.44499966502189636,
+      "learning_rate": 0.00017451669595782075,
+      "loss": 0.8812930583953857,
+      "step": 994
+    },
+    {
+      "epoch": 0.42025316455696204,
+      "grad_norm": 0.4535730481147766,
+      "learning_rate": 0.00017486818980667838,
+      "loss": 0.9439874887466431,
+      "step": 996
+    },
+    {
+      "epoch": 0.4210970464135021,
+      "grad_norm": 0.5240745544433594,
+      "learning_rate": 0.00017521968365553602,
+      "loss": 0.8818395137786865,
+      "step": 998
+    },
+    {
+      "epoch": 0.4219409282700422,
+      "grad_norm": 0.5301211476325989,
+      "learning_rate": 0.0001755711775043937,
+      "loss": 0.886186957359314,
+      "step": 1000
+    },
+    {
+      "epoch": 0.4219409282700422,
+      "eval_loss": 0.9487298727035522,
+      "eval_runtime": 689.4288,
+      "eval_samples_per_second": 3.056,
+      "eval_steps_per_second": 3.056,
+      "step": 1000
+    },
+    {
+      "epoch": 0.42278481012658226,
+      "grad_norm": 0.47876957058906555,
+      "learning_rate": 0.00017592267135325133,
+      "loss": 0.8814021348953247,
+      "step": 1002
+    },
+    {
+      "epoch": 0.42362869198312236,
+      "grad_norm": 0.4929780960083008,
+      "learning_rate": 0.00017627416520210896,
+      "loss": 0.8295068740844727,
+      "step": 1004
+    },
+    {
+      "epoch": 0.42447257383966247,
+      "grad_norm": 0.41888436675071716,
+      "learning_rate": 0.00017662565905096663,
+      "loss": 0.8364827036857605,
+      "step": 1006
+    },
+    {
+      "epoch": 0.4253164556962025,
+      "grad_norm": 0.5175151824951172,
+      "learning_rate": 0.00017697715289982427,
+      "loss": 1.0167189836502075,
+      "step": 1008
+    },
+    {
+      "epoch": 0.42616033755274263,
+      "grad_norm": 0.4815356135368347,
+      "learning_rate": 0.0001773286467486819,
+      "loss": 0.8460752964019775,
+      "step": 1010
+    },
+    {
+      "epoch": 0.4270042194092827,
+      "grad_norm": 0.5210875272750854,
+      "learning_rate": 0.00017768014059753954,
+      "loss": 0.8595574498176575,
+      "step": 1012
+    },
+    {
+      "epoch": 0.4278481012658228,
+      "grad_norm": 0.4400486946105957,
+      "learning_rate": 0.00017803163444639718,
+      "loss": 0.8764723539352417,
+      "step": 1014
+    },
+    {
+      "epoch": 0.4286919831223629,
+      "grad_norm": 0.5282127857208252,
+      "learning_rate": 0.00017838312829525482,
+      "loss": 0.9706798791885376,
+      "step": 1016
+    },
+    {
+      "epoch": 0.42953586497890295,
+      "grad_norm": 0.47523441910743713,
+      "learning_rate": 0.00017873462214411249,
+      "loss": 0.8912002444267273,
+      "step": 1018
+    },
+    {
+      "epoch": 0.43037974683544306,
+      "grad_norm": 0.49640706181526184,
+      "learning_rate": 0.00017908611599297012,
+      "loss": 0.8835636973381042,
+      "step": 1020
+    },
+    {
+      "epoch": 0.4312236286919831,
+      "grad_norm": 0.5253039002418518,
+      "learning_rate": 0.0001794376098418278,
+      "loss": 0.8711735606193542,
+      "step": 1022
+    },
+    {
+      "epoch": 0.4320675105485232,
+      "grad_norm": 0.49285009503364563,
+      "learning_rate": 0.00017978910369068543,
+      "loss": 0.9064869284629822,
+      "step": 1024
+    },
+    {
+      "epoch": 0.43291139240506327,
+      "grad_norm": 0.48758041858673096,
+      "learning_rate": 0.00018014059753954307,
+      "loss": 0.8993359208106995,
+      "step": 1026
+    },
+    {
+      "epoch": 0.4337552742616034,
+      "grad_norm": 0.4815461039543152,
+      "learning_rate": 0.0001804920913884007,
+      "loss": 0.8747937679290771,
+      "step": 1028
+    },
+    {
+      "epoch": 0.4345991561181435,
+      "grad_norm": 0.4930349886417389,
+      "learning_rate": 0.00018084358523725834,
+      "loss": 0.8879084587097168,
+      "step": 1030
+    },
+    {
+      "epoch": 0.43544303797468353,
+      "grad_norm": 0.49303027987480164,
+      "learning_rate": 0.00018119507908611598,
+      "loss": 1.0023083686828613,
+      "step": 1032
+    },
+    {
+      "epoch": 0.43628691983122364,
+      "grad_norm": 0.5312249064445496,
+      "learning_rate": 0.00018154657293497365,
+      "loss": 0.938680112361908,
+      "step": 1034
+    },
+    {
+      "epoch": 0.4371308016877637,
+      "grad_norm": 0.5174582004547119,
+      "learning_rate": 0.0001818980667838313,
+      "loss": 0.854195773601532,
+      "step": 1036
+    },
+    {
+      "epoch": 0.4379746835443038,
+      "grad_norm": 0.5452545881271362,
+      "learning_rate": 0.00018224956063268895,
+      "loss": 0.9059375524520874,
+      "step": 1038
+    },
+    {
+      "epoch": 0.4388185654008439,
+      "grad_norm": 0.5480839014053345,
+      "learning_rate": 0.0001826010544815466,
+      "loss": 0.9708920121192932,
+      "step": 1040
+    },
+    {
+      "epoch": 0.43966244725738396,
+      "grad_norm": 0.44693151116371155,
+      "learning_rate": 0.00018295254833040423,
+      "loss": 0.8561046123504639,
+      "step": 1042
+    },
+    {
+      "epoch": 0.44050632911392407,
+      "grad_norm": 0.4024234116077423,
+      "learning_rate": 0.00018330404217926187,
+      "loss": 0.8811968564987183,
+      "step": 1044
+    },
+    {
+      "epoch": 0.4413502109704641,
+      "grad_norm": 0.5586408376693726,
+      "learning_rate": 0.0001836555360281195,
+      "loss": 0.978068470954895,
+      "step": 1046
+    },
+    {
+      "epoch": 0.4421940928270042,
+      "grad_norm": 0.4918624758720398,
+      "learning_rate": 0.00018400702987697714,
+      "loss": 0.8640981316566467,
+      "step": 1048
+    },
+    {
+      "epoch": 0.4430379746835443,
+      "grad_norm": 0.4335230588912964,
+      "learning_rate": 0.0001843585237258348,
+      "loss": 0.9077964425086975,
+      "step": 1050
+    },
+    {
+      "epoch": 0.4438818565400844,
+      "grad_norm": 0.5275123715400696,
+      "learning_rate": 0.00018471001757469247,
+      "loss": 0.960682213306427,
+      "step": 1052
+    },
+    {
+      "epoch": 0.4447257383966245,
+      "grad_norm": 0.5144415497779846,
+      "learning_rate": 0.0001850615114235501,
+      "loss": 0.8549577593803406,
+      "step": 1054
+    },
+    {
+      "epoch": 0.44556962025316454,
+      "grad_norm": 0.5166662931442261,
+      "learning_rate": 0.00018541300527240775,
+      "loss": 0.9320827126502991,
+      "step": 1056
+    },
+    {
+      "epoch": 0.44641350210970465,
+      "grad_norm": 0.5011980533599854,
+      "learning_rate": 0.0001857644991212654,
+      "loss": 0.9632431864738464,
+      "step": 1058
+    },
+    {
+      "epoch": 0.4472573839662447,
+      "grad_norm": 0.49312469363212585,
+      "learning_rate": 0.00018611599297012303,
+      "loss": 0.9442946910858154,
+      "step": 1060
+    },
+    {
+      "epoch": 0.4481012658227848,
+      "grad_norm": 0.49958568811416626,
+      "learning_rate": 0.00018646748681898066,
+      "loss": 0.952802300453186,
+      "step": 1062
+    },
+    {
+      "epoch": 0.4489451476793249,
+      "grad_norm": 0.49827462434768677,
+      "learning_rate": 0.00018681898066783833,
+      "loss": 0.9630650877952576,
+      "step": 1064
+    },
+    {
+      "epoch": 0.44978902953586497,
+      "grad_norm": 0.523980438709259,
+      "learning_rate": 0.00018717047451669597,
+      "loss": 0.9081395268440247,
+      "step": 1066
+    },
+    {
+      "epoch": 0.4506329113924051,
+      "grad_norm": 0.5108568668365479,
+      "learning_rate": 0.0001875219683655536,
+      "loss": 0.9398958683013916,
+      "step": 1068
+    },
+    {
+      "epoch": 0.45147679324894513,
+      "grad_norm": 0.4453965127468109,
+      "learning_rate": 0.00018787346221441127,
+      "loss": 0.93592768907547,
+      "step": 1070
+    },
+    {
+      "epoch": 0.45232067510548524,
+      "grad_norm": 0.4675683081150055,
+      "learning_rate": 0.0001882249560632689,
+      "loss": 0.8879633545875549,
+      "step": 1072
+    },
+    {
+      "epoch": 0.4531645569620253,
+      "grad_norm": 0.47398847341537476,
+      "learning_rate": 0.00018857644991212655,
+      "loss": 0.9966004490852356,
+      "step": 1074
+    },
+    {
+      "epoch": 0.4540084388185654,
+      "grad_norm": 0.46053192019462585,
+      "learning_rate": 0.0001889279437609842,
+      "loss": 0.9184179902076721,
+      "step": 1076
+    },
+    {
+      "epoch": 0.4548523206751055,
+      "grad_norm": 0.5601398348808289,
+      "learning_rate": 0.00018927943760984182,
+      "loss": 0.9434974193572998,
+      "step": 1078
+    },
+    {
+      "epoch": 0.45569620253164556,
+      "grad_norm": 0.48422637581825256,
+      "learning_rate": 0.0001896309314586995,
+      "loss": 0.9522465467453003,
+      "step": 1080
+    },
+    {
+      "epoch": 0.45654008438818566,
+      "grad_norm": 0.46280911564826965,
+      "learning_rate": 0.00018998242530755713,
+      "loss": 0.9315434694290161,
+      "step": 1082
+    },
+    {
+      "epoch": 0.4573839662447257,
+      "grad_norm": 0.4658683240413666,
+      "learning_rate": 0.00019033391915641477,
+      "loss": 0.8165783286094666,
+      "step": 1084
+    },
+    {
+      "epoch": 0.4582278481012658,
+      "grad_norm": 0.4192182123661041,
+      "learning_rate": 0.0001906854130052724,
+      "loss": 0.8024274110794067,
+      "step": 1086
+    },
+    {
+      "epoch": 0.45907172995780593,
+      "grad_norm": 0.4178735613822937,
+      "learning_rate": 0.00019103690685413007,
+      "loss": 0.8620653748512268,
+      "step": 1088
+    },
+    {
+      "epoch": 0.459915611814346,
+      "grad_norm": 0.4488574266433716,
+      "learning_rate": 0.0001913884007029877,
+      "loss": 0.9121530055999756,
+      "step": 1090
+    },
+    {
+      "epoch": 0.4607594936708861,
+      "grad_norm": 0.5164965987205505,
+      "learning_rate": 0.00019173989455184535,
+      "loss": 0.9496700763702393,
+      "step": 1092
+    },
+    {
+      "epoch": 0.46160337552742614,
+      "grad_norm": 0.41563132405281067,
+      "learning_rate": 0.00019209138840070299,
+      "loss": 0.8757708668708801,
+      "step": 1094
+    },
+    {
+      "epoch": 0.46244725738396625,
+      "grad_norm": 2.3222429752349854,
+      "learning_rate": 0.00019244288224956065,
+      "loss": 0.8406533002853394,
+      "step": 1096
+    },
+    {
+      "epoch": 0.46329113924050636,
+      "grad_norm": 0.4813845157623291,
+      "learning_rate": 0.0001927943760984183,
+      "loss": 0.9459465742111206,
+      "step": 1098
+    },
+    {
+      "epoch": 0.4641350210970464,
+      "grad_norm": 0.6233882308006287,
+      "learning_rate": 0.00019314586994727593,
+      "loss": 0.9376904368400574,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4641350210970464,
+      "eval_loss": 0.9357889294624329,
+      "eval_runtime": 676.9573,
+      "eval_samples_per_second": 3.112,
+      "eval_steps_per_second": 3.112,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4649789029535865,
+      "grad_norm": 0.5125579237937927,
+      "learning_rate": 0.00019349736379613357,
+      "loss": 0.8998825550079346,
+      "step": 1102
+    },
+    {
+      "epoch": 0.46582278481012657,
+      "grad_norm": 0.4534320831298828,
+      "learning_rate": 0.00019384885764499123,
+      "loss": 0.9213768839836121,
+      "step": 1104
+    },
+    {
+      "epoch": 0.4666666666666667,
+      "grad_norm": 0.4715143144130707,
+      "learning_rate": 0.00019420035149384887,
+      "loss": 0.9739661812782288,
+      "step": 1106
+    },
+    {
+      "epoch": 0.4675105485232067,
+      "grad_norm": 0.45529672503471375,
+      "learning_rate": 0.0001945518453427065,
+      "loss": 0.9113216400146484,
+      "step": 1108
+    },
+    {
+      "epoch": 0.46835443037974683,
+      "grad_norm": 0.48199015855789185,
+      "learning_rate": 0.00019490333919156417,
+      "loss": 0.92528235912323,
+      "step": 1110
+    },
+    {
+      "epoch": 0.46919831223628694,
+      "grad_norm": 0.4425188899040222,
+      "learning_rate": 0.0001952548330404218,
+      "loss": 0.8612716197967529,
+      "step": 1112
+    },
+    {
+      "epoch": 0.470042194092827,
+      "grad_norm": 0.4311593770980835,
+      "learning_rate": 0.00019560632688927945,
+      "loss": 0.8901699185371399,
+      "step": 1114
+    },
+    {
+      "epoch": 0.4708860759493671,
+      "grad_norm": 0.497806578874588,
+      "learning_rate": 0.0001959578207381371,
+      "loss": 0.9271994829177856,
+      "step": 1116
+    },
+    {
+      "epoch": 0.47172995780590715,
+      "grad_norm": 0.47149473428726196,
+      "learning_rate": 0.00019630931458699473,
+      "loss": 0.8740925788879395,
+      "step": 1118
+    },
+    {
+      "epoch": 0.47257383966244726,
+      "grad_norm": 0.5029966235160828,
+      "learning_rate": 0.00019666080843585236,
+      "loss": 0.9190115928649902,
+      "step": 1120
+    },
+    {
+      "epoch": 0.47341772151898737,
+      "grad_norm": 0.47972601652145386,
+      "learning_rate": 0.00019701230228471003,
+      "loss": 1.0043057203292847,
+      "step": 1122
+    },
+    {
+      "epoch": 0.4742616033755274,
+      "grad_norm": 0.45096471905708313,
+      "learning_rate": 0.00019736379613356767,
+      "loss": 0.9472925066947937,
+      "step": 1124
+    },
+    {
+      "epoch": 0.4751054852320675,
+      "grad_norm": 0.459852933883667,
+      "learning_rate": 0.00019771528998242533,
+      "loss": 0.8043124079704285,
+      "step": 1126
+    },
+    {
+      "epoch": 0.4759493670886076,
+      "grad_norm": 0.5313422679901123,
+      "learning_rate": 0.00019806678383128297,
+      "loss": 0.9662142395973206,
+      "step": 1128
+    },
+    {
+      "epoch": 0.4767932489451477,
+      "grad_norm": 0.43474531173706055,
+      "learning_rate": 0.0001984182776801406,
+      "loss": 0.8691151738166809,
+      "step": 1130
+    },
+    {
+      "epoch": 0.47763713080168774,
+      "grad_norm": 0.5232312083244324,
+      "learning_rate": 0.00019876977152899825,
+      "loss": 0.9024254083633423,
+      "step": 1132
+    },
+    {
+      "epoch": 0.47848101265822784,
+      "grad_norm": 0.47671905159950256,
+      "learning_rate": 0.0001991212653778559,
+      "loss": 0.8996873497962952,
+      "step": 1134
+    },
+    {
+      "epoch": 0.47932489451476795,
+      "grad_norm": 0.42289480566978455,
+      "learning_rate": 0.00019947275922671353,
+      "loss": 0.8336917757987976,
+      "step": 1136
+    },
+    {
+      "epoch": 0.480168776371308,
+      "grad_norm": 0.4700844883918762,
+      "learning_rate": 0.0001998242530755712,
+      "loss": 0.8943206071853638,
+      "step": 1138
+    },
+    {
+      "epoch": 0.4810126582278481,
+      "grad_norm": 0.5341399312019348,
+      "learning_rate": 0.00019999999711649004,
+      "loss": 0.9340365529060364,
+      "step": 1140
+    },
+    {
+      "epoch": 0.48185654008438816,
+      "grad_norm": 0.46169522404670715,
+      "learning_rate": 0.00019999997404841123,
+      "loss": 0.8241778016090393,
+      "step": 1142
+    },
+    {
+      "epoch": 0.48270042194092827,
+      "grad_norm": 0.5475223064422607,
+      "learning_rate": 0.00019999992791225896,
+      "loss": 1.0096158981323242,
+      "step": 1144
+    },
+    {
+      "epoch": 0.4835443037974684,
+      "grad_norm": 0.524641215801239,
+      "learning_rate": 0.00019999985870804385,
+      "loss": 0.9650378227233887,
+      "step": 1146
+    },
+    {
+      "epoch": 0.48438818565400843,
+      "grad_norm": 0.4326174855232239,
+      "learning_rate": 0.00019999976643578186,
+      "loss": 0.9003009796142578,
+      "step": 1148
+    },
+    {
+      "epoch": 0.48523206751054854,
+      "grad_norm": 0.49034059047698975,
+      "learning_rate": 0.0001999996510954943,
+      "loss": 0.9793432950973511,
+      "step": 1150
+    },
+    {
+      "epoch": 0.4860759493670886,
+      "grad_norm": 0.4760092496871948,
+      "learning_rate": 0.00019999951268720776,
+      "loss": 0.8793007731437683,
+      "step": 1152
+    },
+    {
+      "epoch": 0.4869198312236287,
+      "grad_norm": 0.44606879353523254,
+      "learning_rate": 0.00019999935121095417,
+      "loss": 0.888630747795105,
+      "step": 1154
+    },
+    {
+      "epoch": 0.4877637130801688,
+      "grad_norm": 0.5030332207679749,
+      "learning_rate": 0.0001999991666667708,
+      "loss": 0.9660000801086426,
+      "step": 1156
+    },
+    {
+      "epoch": 0.48860759493670886,
+      "grad_norm": 0.4295555055141449,
+      "learning_rate": 0.00019999895905470014,
+      "loss": 0.824654757976532,
+      "step": 1158
+    },
+    {
+      "epoch": 0.48945147679324896,
+      "grad_norm": 0.392167866230011,
+      "learning_rate": 0.0001999987283747902,
+      "loss": 0.8159562349319458,
+      "step": 1160
+    },
+    {
+      "epoch": 0.490295358649789,
+      "grad_norm": 0.484611839056015,
+      "learning_rate": 0.00019999847462709412,
+      "loss": 0.9630686044692993,
+      "step": 1162
+    },
+    {
+      "epoch": 0.4911392405063291,
+      "grad_norm": 0.45891445875167847,
+      "learning_rate": 0.00019999819781167042,
+      "loss": 0.8396129608154297,
+      "step": 1164
+    },
+    {
+      "epoch": 0.4919831223628692,
+      "grad_norm": 0.5553452968597412,
+      "learning_rate": 0.00019999789792858304,
+      "loss": 0.8194513916969299,
+      "step": 1166
+    },
+    {
+      "epoch": 0.4928270042194093,
+      "grad_norm": 0.38998672366142273,
+      "learning_rate": 0.00019999757497790106,
+      "loss": 0.9268721342086792,
+      "step": 1168
+    },
+    {
+      "epoch": 0.4936708860759494,
+      "grad_norm": 0.45445355772972107,
+      "learning_rate": 0.00019999722895969904,
+      "loss": 0.9204684495925903,
+      "step": 1170
+    },
+    {
+      "epoch": 0.49451476793248944,
+      "grad_norm": 0.484225332736969,
+      "learning_rate": 0.00019999685987405678,
+      "loss": 0.9270301461219788,
+      "step": 1172
+    },
+    {
+      "epoch": 0.49535864978902955,
+      "grad_norm": 0.45215415954589844,
+      "learning_rate": 0.00019999646772105942,
+      "loss": 0.8782645463943481,
+      "step": 1174
+    },
+    {
+      "epoch": 0.4962025316455696,
+      "grad_norm": 0.41578832268714905,
+      "learning_rate": 0.00019999605250079744,
+      "loss": 0.873112678527832,
+      "step": 1176
+    },
+    {
+      "epoch": 0.4970464135021097,
+      "grad_norm": 0.4632788300514221,
+      "learning_rate": 0.0001999956142133666,
+      "loss": 0.9593189358711243,
+      "step": 1178
+    },
+    {
+      "epoch": 0.4978902953586498,
+      "grad_norm": 0.5007622838020325,
+      "learning_rate": 0.000199995152858868,
+      "loss": 0.9215621948242188,
+      "step": 1180
+    },
+    {
+      "epoch": 0.49873417721518987,
+      "grad_norm": 0.5201935768127441,
+      "learning_rate": 0.0001999946684374081,
+      "loss": 0.8964219689369202,
+      "step": 1182
+    },
+    {
+      "epoch": 0.49957805907173,
+      "grad_norm": 0.5053967237472534,
+      "learning_rate": 0.0001999941609490986,
+      "loss": 0.8528663516044617,
+      "step": 1184
+    },
+    {
+      "epoch": 0.5004219409282701,
+      "grad_norm": 0.44397974014282227,
+      "learning_rate": 0.0001999936303940566,
+      "loss": 0.8214734792709351,
+      "step": 1186
+    },
+    {
+      "epoch": 0.5012658227848101,
+      "grad_norm": 0.4563101530075073,
+      "learning_rate": 0.0001999930767724045,
+      "loss": 0.8882166743278503,
+      "step": 1188
+    },
+    {
+      "epoch": 0.5021097046413502,
+      "grad_norm": 0.4666728675365448,
+      "learning_rate": 0.00019999250008426997,
+      "loss": 0.8882588148117065,
+      "step": 1190
+    },
+    {
+      "epoch": 0.5029535864978903,
+      "grad_norm": 0.4423629939556122,
+      "learning_rate": 0.00019999190032978607,
+      "loss": 0.8353691697120667,
+      "step": 1192
+    },
+    {
+      "epoch": 0.5037974683544304,
+      "grad_norm": 0.4514595866203308,
+      "learning_rate": 0.00019999127750909118,
+      "loss": 0.8721219301223755,
+      "step": 1194
+    },
+    {
+      "epoch": 0.5046413502109705,
+      "grad_norm": 0.48469996452331543,
+      "learning_rate": 0.00019999063162232886,
+      "loss": 0.902795135974884,
+      "step": 1196
+    },
+    {
+      "epoch": 0.5054852320675105,
+      "grad_norm": 0.4931983947753906,
+      "learning_rate": 0.00019998996266964823,
+      "loss": 0.8843175768852234,
+      "step": 1198
+    },
+    {
+      "epoch": 0.5063291139240507,
+      "grad_norm": 0.507408082485199,
+      "learning_rate": 0.00019998927065120357,
+      "loss": 0.9179208278656006,
+      "step": 1200
+    },
+    {
+      "epoch": 0.5063291139240507,
+      "eval_loss": 0.9224098324775696,
+      "eval_runtime": 669.7542,
+      "eval_samples_per_second": 3.146,
+      "eval_steps_per_second": 3.146,
+      "step": 1200
+    },
+    {
+      "epoch": 0.5071729957805907,
+      "grad_norm": 0.4024188220500946,
+      "learning_rate": 0.00019998855556715447,
+      "loss": 0.9078981876373291,
+      "step": 1202
+    },
+    {
+      "epoch": 0.5080168776371308,
+      "grad_norm": 0.6558667421340942,
+      "learning_rate": 0.00019998781741766594,
+      "loss": 0.8089252710342407,
+      "step": 1204
+    },
+    {
+      "epoch": 0.5088607594936709,
+      "grad_norm": 0.45128145813941956,
+      "learning_rate": 0.00019998705620290823,
+      "loss": 0.9316248297691345,
+      "step": 1206
+    },
+    {
+      "epoch": 0.509704641350211,
+      "grad_norm": 0.4733511507511139,
+      "learning_rate": 0.00019998627192305694,
+      "loss": 0.9047867655754089,
+      "step": 1208
+    },
+    {
+      "epoch": 0.510548523206751,
+      "grad_norm": 0.4026021659374237,
+      "learning_rate": 0.00019998546457829298,
+      "loss": 0.8507166504859924,
+      "step": 1210
+    },
+    {
+      "epoch": 0.5113924050632911,
+      "grad_norm": 0.43416184186935425,
+      "learning_rate": 0.00019998463416880262,
+      "loss": 0.8490090370178223,
+      "step": 1212
+    },
+    {
+      "epoch": 0.5122362869198313,
+      "grad_norm": 0.3678364157676697,
+      "learning_rate": 0.0001999837806947774,
+      "loss": 0.7730492949485779,
+      "step": 1214
+    },
+    {
+      "epoch": 0.5130801687763713,
+      "grad_norm": 0.49042677879333496,
+      "learning_rate": 0.00019998290415641415,
+      "loss": 0.9625269174575806,
+      "step": 1216
+    },
+    {
+      "epoch": 0.5139240506329114,
+      "grad_norm": 0.4576701819896698,
+      "learning_rate": 0.00019998200455391516,
+      "loss": 0.8961732983589172,
+      "step": 1218
+    },
+    {
+      "epoch": 0.5147679324894515,
+      "grad_norm": 0.4465518593788147,
+      "learning_rate": 0.00019998108188748793,
+      "loss": 0.8508996367454529,
+      "step": 1220
+    },
+    {
+      "epoch": 0.5156118143459916,
+      "grad_norm": 0.4267960488796234,
+      "learning_rate": 0.00019998013615734524,
+      "loss": 0.8664930462837219,
+      "step": 1222
+    },
+    {
+      "epoch": 0.5164556962025316,
+      "grad_norm": 0.39936602115631104,
+      "learning_rate": 0.0001999791673637053,
+      "loss": 0.8841317892074585,
+      "step": 1224
+    },
+    {
+      "epoch": 0.5172995780590718,
+      "grad_norm": 0.4442414343357086,
+      "learning_rate": 0.0001999781755067916,
+      "loss": 0.8601276874542236,
+      "step": 1226
+    },
+    {
+      "epoch": 0.5181434599156118,
+      "grad_norm": 0.4397028088569641,
+      "learning_rate": 0.00019997716058683292,
+      "loss": 0.8377046585083008,
+      "step": 1228
+    },
+    {
+      "epoch": 0.5189873417721519,
+      "grad_norm": 0.5327648520469666,
+      "learning_rate": 0.0001999761226040634,
+      "loss": 0.9853615164756775,
+      "step": 1230
+    },
+    {
+      "epoch": 0.5198312236286919,
+      "grad_norm": 0.40931421518325806,
+      "learning_rate": 0.00019997506155872244,
+      "loss": 0.9235715866088867,
+      "step": 1232
+    },
+    {
+      "epoch": 0.5206751054852321,
+      "grad_norm": 0.4599161148071289,
+      "learning_rate": 0.00019997397745105487,
+      "loss": 0.868396520614624,
+      "step": 1234
+    },
+    {
+      "epoch": 0.5215189873417722,
+      "grad_norm": 0.4813833236694336,
+      "learning_rate": 0.0001999728702813107,
+      "loss": 0.9946733117103577,
+      "step": 1236
+    },
+    {
+      "epoch": 0.5223628691983122,
+      "grad_norm": 0.43518269062042236,
+      "learning_rate": 0.00019997174004974543,
+      "loss": 0.8608635067939758,
+      "step": 1238
+    },
+    {
+      "epoch": 0.5232067510548524,
+      "grad_norm": 0.45646214485168457,
+      "learning_rate": 0.0001999705867566197,
+      "loss": 0.819354772567749,
+      "step": 1240
+    },
+    {
+      "epoch": 0.5240506329113924,
+      "grad_norm": 0.46834197640419006,
+      "learning_rate": 0.00019996941040219954,
+      "loss": 0.8934658765792847,
+      "step": 1242
+    },
+    {
+      "epoch": 0.5248945147679325,
+      "grad_norm": 0.46065032482147217,
+      "learning_rate": 0.0001999682109867564,
+      "loss": 0.8624778985977173,
+      "step": 1244
+    },
+    {
+      "epoch": 0.5257383966244725,
+      "grad_norm": 0.4506741166114807,
+      "learning_rate": 0.00019996698851056688,
+      "loss": 0.8540882468223572,
+      "step": 1246
+    },
+    {
+      "epoch": 0.5265822784810127,
+      "grad_norm": 0.4550519287586212,
+      "learning_rate": 0.00019996574297391302,
+      "loss": 0.9003006219863892,
+      "step": 1248
+    },
+    {
+      "epoch": 0.5274261603375527,
+      "grad_norm": 0.4097813367843628,
+      "learning_rate": 0.00019996447437708214,
+      "loss": 0.8490248918533325,
+      "step": 1250
+    },
+    {
+      "epoch": 0.5282700421940928,
+      "grad_norm": 0.4375711679458618,
+      "learning_rate": 0.00019996318272036688,
+      "loss": 0.9204569458961487,
+      "step": 1252
+    },
+    {
+      "epoch": 0.529113924050633,
+      "grad_norm": 0.4335198998451233,
+      "learning_rate": 0.00019996186800406518,
+      "loss": 0.8831789493560791,
+      "step": 1254
+    },
+    {
+      "epoch": 0.529957805907173,
+      "grad_norm": 0.4840933084487915,
+      "learning_rate": 0.00019996053022848035,
+      "loss": 0.8578064441680908,
+      "step": 1256
+    },
+    {
+      "epoch": 0.5308016877637131,
+      "grad_norm": 0.48911216855049133,
+      "learning_rate": 0.00019995916939392097,
+      "loss": 0.9235416650772095,
+      "step": 1258
+    },
+    {
+      "epoch": 0.5316455696202531,
+      "grad_norm": 0.41106730699539185,
+      "learning_rate": 0.00019995778550070098,
+      "loss": 0.7897764444351196,
+      "step": 1260
+    },
+    {
+      "epoch": 0.5324894514767933,
+      "grad_norm": 0.42233356833457947,
+      "learning_rate": 0.00019995637854913957,
+      "loss": 0.9392989873886108,
+      "step": 1262
+    },
+    {
+      "epoch": 0.5333333333333333,
+      "grad_norm": 0.39145180583000183,
+      "learning_rate": 0.0001999549485395613,
+      "loss": 0.9130091667175293,
+      "step": 1264
+    },
+    {
+      "epoch": 0.5341772151898734,
+      "grad_norm": 0.42780739068984985,
+      "learning_rate": 0.00019995349547229614,
+      "loss": 0.9690561890602112,
+      "step": 1266
+    },
+    {
+      "epoch": 0.5350210970464135,
+      "grad_norm": 0.4823327362537384,
+      "learning_rate": 0.0001999520193476792,
+      "loss": 0.9990131258964539,
+      "step": 1268
+    },
+    {
+      "epoch": 0.5358649789029536,
+      "grad_norm": 0.4669715166091919,
+      "learning_rate": 0.00019995052016605097,
+      "loss": 0.8343052864074707,
+      "step": 1270
+    },
+    {
+      "epoch": 0.5367088607594936,
+      "grad_norm": 0.4024743437767029,
+      "learning_rate": 0.0001999489979277574,
+      "loss": 0.9605970978736877,
+      "step": 1272
+    },
+    {
+      "epoch": 0.5375527426160338,
+      "grad_norm": 0.42063432931900024,
+      "learning_rate": 0.0001999474526331495,
+      "loss": 0.8164438009262085,
+      "step": 1274
+    },
+    {
+      "epoch": 0.5383966244725739,
+      "grad_norm": 0.47953495383262634,
+      "learning_rate": 0.00019994588428258383,
+      "loss": 0.879767656326294,
+      "step": 1276
+    },
+    {
+      "epoch": 0.5392405063291139,
+      "grad_norm": 0.39668557047843933,
+      "learning_rate": 0.00019994429287642214,
+      "loss": 0.7658726572990417,
+      "step": 1278
+    },
+    {
+      "epoch": 0.540084388185654,
+      "grad_norm": 0.5136121511459351,
+      "learning_rate": 0.00019994267841503155,
+      "loss": 0.9477384686470032,
+      "step": 1280
+    },
+    {
+      "epoch": 0.5409282700421941,
+      "grad_norm": 0.4039286673069,
+      "learning_rate": 0.00019994104089878452,
+      "loss": 0.7883896827697754,
+      "step": 1282
+    },
+    {
+      "epoch": 0.5417721518987342,
+      "grad_norm": 0.4425487816333771,
+      "learning_rate": 0.0001999393803280587,
+      "loss": 0.8526129126548767,
+      "step": 1284
+    },
+    {
+      "epoch": 0.5426160337552742,
+      "grad_norm": 0.4544559121131897,
+      "learning_rate": 0.00019993769670323725,
+      "loss": 0.8694683909416199,
+      "step": 1286
+    },
+    {
+      "epoch": 0.5434599156118144,
+      "grad_norm": 0.42872917652130127,
+      "learning_rate": 0.0001999359900247085,
+      "loss": 0.8211527466773987,
+      "step": 1288
+    },
+    {
+      "epoch": 0.5443037974683544,
+      "grad_norm": 0.41549673676490784,
+      "learning_rate": 0.00019993426029286617,
+      "loss": 0.7967932224273682,
+      "step": 1290
+    },
+    {
+      "epoch": 0.5451476793248945,
+      "grad_norm": 0.42888355255126953,
+      "learning_rate": 0.00019993250750810926,
+      "loss": 0.8291563987731934,
+      "step": 1292
+    },
+    {
+      "epoch": 0.5459915611814345,
+      "grad_norm": 0.44985485076904297,
+      "learning_rate": 0.0001999307316708421,
+      "loss": 0.8787803649902344,
+      "step": 1294
+    },
+    {
+      "epoch": 0.5468354430379747,
+      "grad_norm": 0.43326008319854736,
+      "learning_rate": 0.00019992893278147436,
+      "loss": 0.8375519514083862,
+      "step": 1296
+    },
+    {
+      "epoch": 0.5476793248945148,
+      "grad_norm": 0.406556099653244,
+      "learning_rate": 0.00019992711084042096,
+      "loss": 0.8538051247596741,
+      "step": 1298
+    },
+    {
+      "epoch": 0.5485232067510548,
+      "grad_norm": 0.46090877056121826,
+      "learning_rate": 0.00019992526584810226,
+      "loss": 0.9722132086753845,
+      "step": 1300
+    },
+    {
+      "epoch": 0.5485232067510548,
+      "eval_loss": 0.9068717360496521,
+      "eval_runtime": 680.7718,
+      "eval_samples_per_second": 3.095,
+      "eval_steps_per_second": 3.095,
+      "step": 1300
+    },
+    {
+      "epoch": 0.549367088607595,
+      "grad_norm": 0.44932207465171814,
+      "learning_rate": 0.00019992339780494382,
+      "loss": 0.8474152684211731,
+      "step": 1302
+    },
+    {
+      "epoch": 0.550210970464135,
+      "grad_norm": 0.5185276865959167,
+      "learning_rate": 0.0001999215067113766,
+      "loss": 0.8440775871276855,
+      "step": 1304
+    },
+    {
+      "epoch": 0.5510548523206751,
+      "grad_norm": 0.43811365962028503,
+      "learning_rate": 0.0001999195925678368,
+      "loss": 0.8848010897636414,
+      "step": 1306
+    },
+    {
+      "epoch": 0.5518987341772152,
+      "grad_norm": 0.4861660301685333,
+      "learning_rate": 0.00019991765537476596,
+      "loss": 0.8265100121498108,
+      "step": 1308
+    },
+    {
+      "epoch": 0.5527426160337553,
+      "grad_norm": 0.5068721771240234,
+      "learning_rate": 0.000199915695132611,
+      "loss": 0.9327963590621948,
+      "step": 1310
+    },
+    {
+      "epoch": 0.5535864978902953,
+      "grad_norm": 0.4902805685997009,
+      "learning_rate": 0.0001999137118418241,
+      "loss": 0.9315155744552612,
+      "step": 1312
+    },
+    {
+      "epoch": 0.5544303797468354,
+      "grad_norm": 0.4530964493751526,
+      "learning_rate": 0.00019991170550286272,
+      "loss": 0.9344852566719055,
+      "step": 1314
+    },
+    {
+      "epoch": 0.5552742616033756,
+      "grad_norm": 0.4106673300266266,
+      "learning_rate": 0.00019990967611618974,
+      "loss": 0.8326173424720764,
+      "step": 1316
+    },
+    {
+      "epoch": 0.5561181434599156,
+      "grad_norm": 0.4439711570739746,
+      "learning_rate": 0.0001999076236822733,
+      "loss": 0.8934909105300903,
+      "step": 1318
+    },
+    {
+      "epoch": 0.5569620253164557,
+      "grad_norm": 0.40245553851127625,
+      "learning_rate": 0.0001999055482015868,
+      "loss": 0.6691107153892517,
+      "step": 1320
+    },
+    {
+      "epoch": 0.5578059071729958,
+      "grad_norm": 0.43904435634613037,
+      "learning_rate": 0.0001999034496746091,
+      "loss": 0.8560307621955872,
+      "step": 1322
+    },
+    {
+      "epoch": 0.5586497890295359,
+      "grad_norm": 0.5131705403327942,
+      "learning_rate": 0.00019990132810182422,
+      "loss": 0.9024442434310913,
+      "step": 1324
+    },
+    {
+      "epoch": 0.5594936708860759,
+      "grad_norm": 0.4539531171321869,
+      "learning_rate": 0.00019989918348372156,
+      "loss": 0.8675428628921509,
+      "step": 1326
+    },
+    {
+      "epoch": 0.560337552742616,
+      "grad_norm": 0.44469985365867615,
+      "learning_rate": 0.00019989701582079591,
+      "loss": 0.8853429555892944,
+      "step": 1328
+    },
+    {
+      "epoch": 0.5611814345991561,
+      "grad_norm": 0.4485546946525574,
+      "learning_rate": 0.00019989482511354725,
+      "loss": 0.9675378799438477,
+      "step": 1330
+    },
+    {
+      "epoch": 0.5620253164556962,
+      "grad_norm": 0.43316468596458435,
+      "learning_rate": 0.00019989261136248097,
+      "loss": 0.8256655335426331,
+      "step": 1332
+    },
+    {
+      "epoch": 0.5628691983122363,
+      "grad_norm": 0.4315780699253082,
+      "learning_rate": 0.00019989037456810772,
+      "loss": 0.7727690935134888,
+      "step": 1334
+    },
+    {
+      "epoch": 0.5637130801687764,
+      "grad_norm": 0.45147502422332764,
+      "learning_rate": 0.00019988811473094348,
+      "loss": 0.9240618348121643,
+      "step": 1336
+    },
+    {
+      "epoch": 0.5645569620253165,
+      "grad_norm": 0.4606908857822418,
+      "learning_rate": 0.00019988583185150957,
+      "loss": 0.8005949258804321,
+      "step": 1338
+    },
+    {
+      "epoch": 0.5654008438818565,
+      "grad_norm": 0.48932451009750366,
+      "learning_rate": 0.00019988352593033255,
+      "loss": 0.9042718410491943,
+      "step": 1340
+    },
+    {
+      "epoch": 0.5662447257383966,
+      "grad_norm": 0.46022218465805054,
+      "learning_rate": 0.00019988119696794443,
+      "loss": 0.8724613785743713,
+      "step": 1342
+    },
+    {
+      "epoch": 0.5670886075949367,
+      "grad_norm": 0.4414621889591217,
+      "learning_rate": 0.00019987884496488242,
+      "loss": 0.8653066158294678,
+      "step": 1344
+    },
+    {
+      "epoch": 0.5679324894514768,
+      "grad_norm": 0.4490342140197754,
+      "learning_rate": 0.0001998764699216891,
+      "loss": 0.8843849301338196,
+      "step": 1346
+    },
+    {
+      "epoch": 0.5687763713080168,
+      "grad_norm": 0.47690069675445557,
+      "learning_rate": 0.00019987407183891232,
+      "loss": 0.8385393023490906,
+      "step": 1348
+    },
+    {
+      "epoch": 0.569620253164557,
+      "grad_norm": 0.41409361362457275,
+      "learning_rate": 0.00019987165071710527,
+      "loss": 0.77492755651474,
+      "step": 1350
+    },
+    {
+      "epoch": 0.570464135021097,
+      "grad_norm": 0.4151647984981537,
+      "learning_rate": 0.0001998692065568265,
+      "loss": 0.8368680477142334,
+      "step": 1352
+    },
+    {
+      "epoch": 0.5713080168776371,
+      "grad_norm": 0.4105067849159241,
+      "learning_rate": 0.0001998667393586398,
+      "loss": 0.833220362663269,
+      "step": 1354
+    },
+    {
+      "epoch": 0.5721518987341773,
+      "grad_norm": 0.44701820611953735,
+      "learning_rate": 0.0001998642491231143,
+      "loss": 0.8041552305221558,
+      "step": 1356
+    },
+    {
+      "epoch": 0.5729957805907173,
+      "grad_norm": 0.4621582329273224,
+      "learning_rate": 0.00019986173585082444,
+      "loss": 0.9052709937095642,
+      "step": 1358
+    },
+    {
+      "epoch": 0.5738396624472574,
+      "grad_norm": 0.4934713840484619,
+      "learning_rate": 0.00019985919954235005,
+      "loss": 0.8036054372787476,
+      "step": 1360
+    },
+    {
+      "epoch": 0.5746835443037974,
+      "grad_norm": 0.4557384252548218,
+      "learning_rate": 0.0001998566401982761,
+      "loss": 0.8340095281600952,
+      "step": 1362
+    },
+    {
+      "epoch": 0.5755274261603376,
+      "grad_norm": 0.5048403739929199,
+      "learning_rate": 0.00019985405781919305,
+      "loss": 0.8703633546829224,
+      "step": 1364
+    },
+    {
+      "epoch": 0.5763713080168776,
+      "grad_norm": 0.4707714319229126,
+      "learning_rate": 0.0001998514524056966,
+      "loss": 0.9115830659866333,
+      "step": 1366
+    },
+    {
+      "epoch": 0.5772151898734177,
+      "grad_norm": 0.4502186179161072,
+      "learning_rate": 0.00019984882395838777,
+      "loss": 0.8603149056434631,
+      "step": 1368
+    },
+    {
+      "epoch": 0.5780590717299579,
+      "grad_norm": 0.4229136109352112,
+      "learning_rate": 0.00019984617247787288,
+      "loss": 0.7482197284698486,
+      "step": 1370
+    },
+    {
+      "epoch": 0.5789029535864979,
+      "grad_norm": 0.49208030104637146,
+      "learning_rate": 0.0001998434979647636,
+      "loss": 0.9071239233016968,
+      "step": 1372
+    },
+    {
+      "epoch": 0.579746835443038,
+      "grad_norm": 0.47393161058425903,
+      "learning_rate": 0.00019984080041967687,
+      "loss": 0.9370644092559814,
+      "step": 1374
+    },
+    {
+      "epoch": 0.580590717299578,
+      "grad_norm": 0.4483324885368347,
+      "learning_rate": 0.00019983807984323492,
+      "loss": 0.8536180257797241,
+      "step": 1376
+    },
+    {
+      "epoch": 0.5814345991561182,
+      "grad_norm": 0.4163796007633209,
+      "learning_rate": 0.00019983533623606543,
+      "loss": 0.7360405325889587,
+      "step": 1378
+    },
+    {
+      "epoch": 0.5822784810126582,
+      "grad_norm": 0.44147446751594543,
+      "learning_rate": 0.00019983256959880122,
+      "loss": 0.8162824511528015,
+      "step": 1380
+    },
+    {
+      "epoch": 0.5831223628691983,
+      "grad_norm": 0.44443491101264954,
+      "learning_rate": 0.0001998297799320805,
+      "loss": 0.9169327616691589,
+      "step": 1382
+    },
+    {
+      "epoch": 0.5839662447257384,
+      "grad_norm": 0.4641949236392975,
+      "learning_rate": 0.00019982696723654686,
+      "loss": 0.8875693678855896,
+      "step": 1384
+    },
+    {
+      "epoch": 0.5848101265822785,
+      "grad_norm": 0.4327974319458008,
+      "learning_rate": 0.00019982413151284906,
+      "loss": 0.7747344970703125,
+      "step": 1386
+    },
+    {
+      "epoch": 0.5856540084388185,
+      "grad_norm": 0.44200772047042847,
+      "learning_rate": 0.0001998212727616413,
+      "loss": 0.8490481972694397,
+      "step": 1388
+    },
+    {
+      "epoch": 0.5864978902953587,
+      "grad_norm": 0.39084959030151367,
+      "learning_rate": 0.000199818390983583,
+      "loss": 0.8014808297157288,
+      "step": 1390
+    },
+    {
+      "epoch": 0.5873417721518988,
+      "grad_norm": 0.3691277801990509,
+      "learning_rate": 0.00019981548617933897,
+      "loss": 0.8650928139686584,
+      "step": 1392
+    },
+    {
+      "epoch": 0.5881856540084388,
+      "grad_norm": 0.480002760887146,
+      "learning_rate": 0.00019981255834957925,
+      "loss": 0.7713267803192139,
+      "step": 1394
+    },
+    {
+      "epoch": 0.5890295358649789,
+      "grad_norm": 0.4397226572036743,
+      "learning_rate": 0.00019980960749497927,
+      "loss": 0.9053656458854675,
+      "step": 1396
+    },
+    {
+      "epoch": 0.589873417721519,
+      "grad_norm": 0.4263727366924286,
+      "learning_rate": 0.00019980663361621973,
+      "loss": 0.8737669587135315,
+      "step": 1398
+    },
+    {
+      "epoch": 0.5907172995780591,
+      "grad_norm": 0.4468817710876465,
+      "learning_rate": 0.0001998036367139866,
+      "loss": 0.9396650791168213,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5907172995780591,
+      "eval_loss": 0.8971880674362183,
+      "eval_runtime": 692.8046,
+      "eval_samples_per_second": 3.041,
+      "eval_steps_per_second": 3.041,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5915611814345991,
+      "grad_norm": 0.4128144383430481,
+      "learning_rate": 0.00019980061678897127,
+      "loss": 0.926134467124939,
+      "step": 1402
+    },
+    {
+      "epoch": 0.5924050632911393,
+      "grad_norm": 0.45524629950523376,
+      "learning_rate": 0.00019979757384187035,
+      "loss": 0.826360285282135,
+      "step": 1404
+    },
+    {
+      "epoch": 0.5932489451476793,
+      "grad_norm": 0.42521336674690247,
+      "learning_rate": 0.0001997945078733858,
+      "loss": 0.8147702813148499,
+      "step": 1406
+    },
+    {
+      "epoch": 0.5940928270042194,
+      "grad_norm": 0.4172525107860565,
+      "learning_rate": 0.00019979141888422489,
+      "loss": 0.7711596488952637,
+      "step": 1408
+    },
+    {
+      "epoch": 0.5949367088607594,
+      "grad_norm": 0.46193623542785645,
+      "learning_rate": 0.00019978830687510013,
+      "loss": 0.7875980138778687,
+      "step": 1410
+    },
+    {
+      "epoch": 0.5957805907172996,
+      "grad_norm": 0.4455825686454773,
+      "learning_rate": 0.00019978517184672946,
+      "loss": 0.8590195178985596,
+      "step": 1412
+    },
+    {
+      "epoch": 0.5966244725738397,
+      "grad_norm": 0.4266505539417267,
+      "learning_rate": 0.00019978201379983608,
+      "loss": 0.9318227767944336,
+      "step": 1414
+    },
+    {
+      "epoch": 0.5974683544303797,
+      "grad_norm": 0.42627251148223877,
+      "learning_rate": 0.00019977883273514843,
+      "loss": 0.8644474744796753,
+      "step": 1416
+    },
+    {
+      "epoch": 0.5983122362869199,
+      "grad_norm": 0.393873393535614,
+      "learning_rate": 0.00019977562865340038,
+      "loss": 0.7760446071624756,
+      "step": 1418
+    },
+    {
+      "epoch": 0.5991561181434599,
+      "grad_norm": 0.44334208965301514,
+      "learning_rate": 0.000199772401555331,
+      "loss": 0.8071584701538086,
+      "step": 1420
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.39390063285827637,
+      "learning_rate": 0.0001997691514416848,
+      "loss": 0.9076889753341675,
+      "step": 1422
+    },
+    {
+      "epoch": 0.60084388185654,
+      "grad_norm": 0.46753421425819397,
+      "learning_rate": 0.00019976587831321144,
+      "loss": 0.8534318804740906,
+      "step": 1424
+    },
+    {
+      "epoch": 0.6016877637130802,
+      "grad_norm": 0.4840681254863739,
+      "learning_rate": 0.00019976258217066598,
+      "loss": 0.9049596190452576,
+      "step": 1426
+    },
+    {
+      "epoch": 0.6025316455696202,
+      "grad_norm": 0.4740557372570038,
+      "learning_rate": 0.00019975926301480878,
+      "loss": 0.8520918488502502,
+      "step": 1428
+    },
+    {
+      "epoch": 0.6033755274261603,
+      "grad_norm": 0.46498584747314453,
+      "learning_rate": 0.00019975592084640553,
+      "loss": 0.8221305012702942,
+      "step": 1430
+    },
+    {
+      "epoch": 0.6042194092827005,
+      "grad_norm": 0.40885692834854126,
+      "learning_rate": 0.0001997525556662272,
+      "loss": 0.8132198452949524,
+      "step": 1432
+    },
+    {
+      "epoch": 0.6050632911392405,
+      "grad_norm": 0.4375821053981781,
+      "learning_rate": 0.00019974916747505006,
+      "loss": 0.870534360408783,
+      "step": 1434
+    },
+    {
+      "epoch": 0.6059071729957806,
+      "grad_norm": 0.4780935049057007,
+      "learning_rate": 0.00019974575627365569,
+      "loss": 0.8825759291648865,
+      "step": 1436
+    },
+    {
+      "epoch": 0.6067510548523207,
+      "grad_norm": 0.4623970687389374,
+      "learning_rate": 0.00019974232206283098,
+      "loss": 0.8187641501426697,
+      "step": 1438
+    },
+    {
+      "epoch": 0.6075949367088608,
+      "grad_norm": 0.42172595858573914,
+      "learning_rate": 0.0001997388648433682,
+      "loss": 0.8569821119308472,
+      "step": 1440
+    },
+    {
+      "epoch": 0.6084388185654008,
+      "grad_norm": 0.49386659264564514,
+      "learning_rate": 0.0001997353846160648,
+      "loss": 0.9173614382743835,
+      "step": 1442
+    },
+    {
+      "epoch": 0.6092827004219409,
+      "grad_norm": 0.47122615575790405,
+      "learning_rate": 0.00019973188138172363,
+      "loss": 0.9444975256919861,
+      "step": 1444
+    },
+    {
+      "epoch": 0.610126582278481,
+      "grad_norm": 0.4831530451774597,
+      "learning_rate": 0.0001997283551411528,
+      "loss": 0.9537245631217957,
+      "step": 1446
+    },
+    {
+      "epoch": 0.6109704641350211,
+      "grad_norm": 0.44460317492485046,
+      "learning_rate": 0.0001997248058951658,
+      "loss": 0.8481367230415344,
+      "step": 1448
+    },
+    {
+      "epoch": 0.6118143459915611,
+      "grad_norm": 0.4315263628959656,
+      "learning_rate": 0.00019972123364458128,
+      "loss": 0.8572371006011963,
+      "step": 1450
+    },
+    {
+      "epoch": 0.6126582278481013,
+      "grad_norm": 0.38559582829475403,
+      "learning_rate": 0.00019971763839022336,
+      "loss": 0.762590765953064,
+      "step": 1452
+    },
+    {
+      "epoch": 0.6135021097046414,
+      "grad_norm": 0.3820290267467499,
+      "learning_rate": 0.00019971402013292138,
+      "loss": 0.7879500389099121,
+      "step": 1454
+    },
+    {
+      "epoch": 0.6143459915611814,
+      "grad_norm": 0.42509570717811584,
+      "learning_rate": 0.00019971037887351,
+      "loss": 0.8696863055229187,
+      "step": 1456
+    },
+    {
+      "epoch": 0.6151898734177215,
+      "grad_norm": 0.46750548481941223,
+      "learning_rate": 0.00019970671461282916,
+      "loss": 0.8425862789154053,
+      "step": 1458
+    },
+    {
+      "epoch": 0.6160337552742616,
+      "grad_norm": 0.3937121033668518,
+      "learning_rate": 0.0001997030273517242,
+      "loss": 0.8028931021690369,
+      "step": 1460
+    },
+    {
+      "epoch": 0.6168776371308017,
+      "grad_norm": 0.4361220896244049,
+      "learning_rate": 0.00019969931709104565,
+      "loss": 0.8662509918212891,
+      "step": 1462
+    },
+    {
+      "epoch": 0.6177215189873417,
+      "grad_norm": 0.4737963378429413,
+      "learning_rate": 0.00019969558383164943,
+      "loss": 0.8962596654891968,
+      "step": 1464
+    },
+    {
+      "epoch": 0.6185654008438819,
+      "grad_norm": 0.3952867388725281,
+      "learning_rate": 0.00019969182757439668,
+      "loss": 0.8630867600440979,
+      "step": 1466
+    },
+    {
+      "epoch": 0.619409282700422,
+      "grad_norm": 0.41688743233680725,
+      "learning_rate": 0.00019968804832015393,
+      "loss": 0.8759240508079529,
+      "step": 1468
+    },
+    {
+      "epoch": 0.620253164556962,
+      "grad_norm": 0.4599224627017975,
+      "learning_rate": 0.00019968424606979298,
+      "loss": 0.8459385633468628,
+      "step": 1470
+    },
+    {
+      "epoch": 0.6210970464135022,
+      "grad_norm": 0.5476765632629395,
+      "learning_rate": 0.00019968042082419094,
+      "loss": 0.8844659328460693,
+      "step": 1472
+    },
+    {
+      "epoch": 0.6219409282700422,
+      "grad_norm": 0.46202269196510315,
+      "learning_rate": 0.00019967657258423022,
+      "loss": 0.860946536064148,
+      "step": 1474
+    },
+    {
+      "epoch": 0.6227848101265823,
+      "grad_norm": 0.39446666836738586,
+      "learning_rate": 0.00019967270135079853,
+      "loss": 0.7587860822677612,
+      "step": 1476
+    },
+    {
+      "epoch": 0.6236286919831223,
+      "grad_norm": 0.4500375986099243,
+      "learning_rate": 0.0001996688071247889,
+      "loss": 0.9143000245094299,
+      "step": 1478
+    },
+    {
+      "epoch": 0.6244725738396625,
+      "grad_norm": 0.4203545153141022,
+      "learning_rate": 0.0001996648899070996,
+      "loss": 0.7865519523620605,
+      "step": 1480
+    },
+    {
+      "epoch": 0.6253164556962025,
+      "grad_norm": 0.45623889565467834,
+      "learning_rate": 0.00019966094969863432,
+      "loss": 0.9192912578582764,
+      "step": 1482
+    },
+    {
+      "epoch": 0.6261603375527426,
+      "grad_norm": 0.4139637053012848,
+      "learning_rate": 0.00019965698650030195,
+      "loss": 0.8459161520004272,
+      "step": 1484
+    },
+    {
+      "epoch": 0.6270042194092827,
+      "grad_norm": 0.4410824179649353,
+      "learning_rate": 0.00019965300031301678,
+      "loss": 0.8159077763557434,
+      "step": 1486
+    },
+    {
+      "epoch": 0.6278481012658228,
+      "grad_norm": 0.43821197748184204,
+      "learning_rate": 0.00019964899113769828,
+      "loss": 0.884467363357544,
+      "step": 1488
+    },
+    {
+      "epoch": 0.6286919831223629,
+      "grad_norm": 0.4692091643810272,
+      "learning_rate": 0.00019964495897527133,
+      "loss": 0.8693601489067078,
+      "step": 1490
+    },
+    {
+      "epoch": 0.6295358649789029,
+      "grad_norm": 0.40715447068214417,
+      "learning_rate": 0.00019964090382666608,
+      "loss": 0.8897743225097656,
+      "step": 1492
+    },
+    {
+      "epoch": 0.6303797468354431,
+      "grad_norm": 0.4123334586620331,
+      "learning_rate": 0.0001996368256928179,
+      "loss": 0.8274733424186707,
+      "step": 1494
+    },
+    {
+      "epoch": 0.6312236286919831,
+      "grad_norm": 0.44347211718559265,
+      "learning_rate": 0.00019963272457466767,
+      "loss": 0.8158749938011169,
+      "step": 1496
+    },
+    {
+      "epoch": 0.6320675105485232,
+      "grad_norm": 0.4700329601764679,
+      "learning_rate": 0.00019962860047316135,
+      "loss": 0.8439569473266602,
+      "step": 1498
+    },
+    {
+      "epoch": 0.6329113924050633,
+      "grad_norm": 0.46415844559669495,
+      "learning_rate": 0.00019962445338925027,
+      "loss": 0.8316822648048401,
+      "step": 1500
+    },
+    {
+      "epoch": 0.6329113924050633,
+      "eval_loss": 0.887488842010498,
+      "eval_runtime": 686.2804,
+      "eval_samples_per_second": 3.07,
+      "eval_steps_per_second": 3.07,
+      "step": 1500
+    },
+    {
+      "epoch": 0.6337552742616034,
+      "grad_norm": 0.44183167815208435,
+      "learning_rate": 0.0001996202833238911,
+      "loss": 0.9121994972229004,
+      "step": 1502
+    },
+    {
+      "epoch": 0.6345991561181434,
+      "grad_norm": 0.4279208779335022,
+      "learning_rate": 0.00019961609027804587,
+      "loss": 0.9160211086273193,
+      "step": 1504
+    },
+    {
+      "epoch": 0.6354430379746835,
+      "grad_norm": 0.3967169523239136,
+      "learning_rate": 0.00019961187425268176,
+      "loss": 0.861677348613739,
+      "step": 1506
+    },
+    {
+      "epoch": 0.6362869198312237,
+      "grad_norm": 0.4796451926231384,
+      "learning_rate": 0.0001996076352487713,
+      "loss": 0.8478423357009888,
+      "step": 1508
+    },
+    {
+      "epoch": 0.6371308016877637,
+      "grad_norm": 0.3861426115036011,
+      "learning_rate": 0.00019960337326729245,
+      "loss": 0.8739159107208252,
+      "step": 1510
+    },
+    {
+      "epoch": 0.6379746835443038,
+      "grad_norm": 0.4567820429801941,
+      "learning_rate": 0.00019959908830922824,
+      "loss": 0.7693920731544495,
+      "step": 1512
+    },
+    {
+      "epoch": 0.6388185654008439,
+      "grad_norm": 0.4457108676433563,
+      "learning_rate": 0.00019959478037556724,
+      "loss": 0.8482301831245422,
+      "step": 1514
+    },
+    {
+      "epoch": 0.639662447257384,
+      "grad_norm": 0.43918928503990173,
+      "learning_rate": 0.00019959044946730314,
+      "loss": 0.8626812696456909,
+      "step": 1516
+    },
+    {
+      "epoch": 0.640506329113924,
+      "grad_norm": 0.4095900058746338,
+      "learning_rate": 0.00019958609558543504,
+      "loss": 0.8342230916023254,
+      "step": 1518
+    },
+    {
+      "epoch": 0.6413502109704642,
+      "grad_norm": 0.4181270897388458,
+      "learning_rate": 0.00019958171873096724,
+      "loss": 0.8794118165969849,
+      "step": 1520
+    },
+    {
+      "epoch": 0.6421940928270042,
+      "grad_norm": 0.4750959277153015,
+      "learning_rate": 0.00019957731890490947,
+      "loss": 0.7787677049636841,
+      "step": 1522
+    },
+    {
+      "epoch": 0.6430379746835443,
+      "grad_norm": 0.5129296779632568,
+      "learning_rate": 0.00019957289610827663,
+      "loss": 0.9084368348121643,
+      "step": 1524
+    },
+    {
+      "epoch": 0.6438818565400843,
+      "grad_norm": 0.4829029142856598,
+      "learning_rate": 0.000199568450342089,
+      "loss": 0.9066952466964722,
+      "step": 1526
+    },
+    {
+      "epoch": 0.6447257383966245,
+      "grad_norm": 0.4225057363510132,
+      "learning_rate": 0.00019956398160737214,
+      "loss": 0.8127874732017517,
+      "step": 1528
+    },
+    {
+      "epoch": 0.6455696202531646,
+      "grad_norm": 0.4231826364994049,
+      "learning_rate": 0.00019955948990515684,
+      "loss": 0.8781921863555908,
+      "step": 1530
+    },
+    {
+      "epoch": 0.6464135021097046,
+      "grad_norm": 0.4436216950416565,
+      "learning_rate": 0.00019955497523647933,
+      "loss": 0.8136996626853943,
+      "step": 1532
+    },
+    {
+      "epoch": 0.6472573839662448,
+      "grad_norm": 0.4018244743347168,
+      "learning_rate": 0.00019955043760238098,
+      "loss": 0.8003877401351929,
+      "step": 1534
+    },
+    {
+      "epoch": 0.6481012658227848,
+      "grad_norm": 0.40024611353874207,
+      "learning_rate": 0.0001995458770039086,
+      "loss": 0.8996267914772034,
+      "step": 1536
+    },
+    {
+      "epoch": 0.6489451476793249,
+      "grad_norm": 0.42845702171325684,
+      "learning_rate": 0.00019954129344211424,
+      "loss": 0.8610522150993347,
+      "step": 1538
+    },
+    {
+      "epoch": 0.6497890295358649,
+      "grad_norm": 0.4258182644844055,
+      "learning_rate": 0.00019953668691805517,
+      "loss": 0.8991837501525879,
+      "step": 1540
+    },
+    {
+      "epoch": 0.6506329113924051,
+      "grad_norm": 0.4483257830142975,
+      "learning_rate": 0.0001995320574327941,
+      "loss": 0.8150189518928528,
+      "step": 1542
+    },
+    {
+      "epoch": 0.6514767932489451,
+      "grad_norm": 0.4320627748966217,
+      "learning_rate": 0.0001995274049873989,
+      "loss": 0.8572840690612793,
+      "step": 1544
+    },
+    {
+      "epoch": 0.6523206751054852,
+      "grad_norm": 0.39950594305992126,
+      "learning_rate": 0.00019952272958294288,
+      "loss": 0.7766129374504089,
+      "step": 1546
+    },
+    {
+      "epoch": 0.6531645569620254,
+      "grad_norm": 0.4130117893218994,
+      "learning_rate": 0.00019951803122050446,
+      "loss": 0.9032199382781982,
+      "step": 1548
+    },
+    {
+      "epoch": 0.6540084388185654,
+      "grad_norm": 0.4436919391155243,
+      "learning_rate": 0.00019951330990116754,
+      "loss": 0.8361048102378845,
+      "step": 1550
+    },
+    {
+      "epoch": 0.6548523206751055,
+      "grad_norm": 0.4229227900505066,
+      "learning_rate": 0.00019950856562602121,
+      "loss": 0.8661768436431885,
+      "step": 1552
+    },
+    {
+      "epoch": 0.6556962025316456,
+      "grad_norm": 0.4839727580547333,
+      "learning_rate": 0.00019950379839615988,
+      "loss": 0.9737826585769653,
+      "step": 1554
+    },
+    {
+      "epoch": 0.6565400843881857,
+      "grad_norm": 0.3859386742115021,
+      "learning_rate": 0.00019949900821268328,
+      "loss": 0.7542453408241272,
+      "step": 1556
+    },
+    {
+      "epoch": 0.6573839662447257,
+      "grad_norm": 0.39643988013267517,
+      "learning_rate": 0.0001994941950766964,
+      "loss": 0.9226290583610535,
+      "step": 1558
+    },
+    {
+      "epoch": 0.6582278481012658,
+      "grad_norm": 0.4261355400085449,
+      "learning_rate": 0.00019948935898930954,
+      "loss": 0.8375602960586548,
+      "step": 1560
+    },
+    {
+      "epoch": 0.6590717299578059,
+      "grad_norm": 0.445931077003479,
+      "learning_rate": 0.00019948449995163826,
+      "loss": 0.7847310304641724,
+      "step": 1562
+    },
+    {
+      "epoch": 0.659915611814346,
+      "grad_norm": 0.4680459797382355,
+      "learning_rate": 0.00019947961796480353,
+      "loss": 0.851150631904602,
+      "step": 1564
+    },
+    {
+      "epoch": 0.660759493670886,
+      "grad_norm": 0.6826074719429016,
+      "learning_rate": 0.00019947471302993143,
+      "loss": 0.8259562253952026,
+      "step": 1566
+    },
+    {
+      "epoch": 0.6616033755274262,
+      "grad_norm": 0.47369205951690674,
+      "learning_rate": 0.00019946978514815352,
+      "loss": 0.879021406173706,
+      "step": 1568
+    },
+    {
+      "epoch": 0.6624472573839663,
+      "grad_norm": 0.41111937165260315,
+      "learning_rate": 0.0001994648343206065,
+      "loss": 0.9296056032180786,
+      "step": 1570
+    },
+    {
+      "epoch": 0.6632911392405063,
+      "grad_norm": 0.4297783076763153,
+      "learning_rate": 0.00019945986054843248,
+      "loss": 0.873822808265686,
+      "step": 1572
+    },
+    {
+      "epoch": 0.6641350210970464,
+      "grad_norm": 0.48124316334724426,
+      "learning_rate": 0.00019945486383277878,
+      "loss": 0.937151312828064,
+      "step": 1574
+    },
+    {
+      "epoch": 0.6649789029535865,
+      "grad_norm": 0.4048190116882324,
+      "learning_rate": 0.00019944984417479805,
+      "loss": 0.8398929834365845,
+      "step": 1576
+    },
+    {
+      "epoch": 0.6658227848101266,
+      "grad_norm": 0.4023774266242981,
+      "learning_rate": 0.00019944480157564826,
+      "loss": 0.8269520998001099,
+      "step": 1578
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.3784184157848358,
+      "learning_rate": 0.00019943973603649262,
+      "loss": 0.8237608671188354,
+      "step": 1580
+    },
+    {
+      "epoch": 0.6675105485232068,
+      "grad_norm": 0.4448573887348175,
+      "learning_rate": 0.00019943464755849965,
+      "loss": 0.8469099998474121,
+      "step": 1582
+    },
+    {
+      "epoch": 0.6683544303797468,
+      "grad_norm": 0.4690774083137512,
+      "learning_rate": 0.00019942953614284317,
+      "loss": 0.8630822896957397,
+      "step": 1584
+    },
+    {
+      "epoch": 0.6691983122362869,
+      "grad_norm": 0.43670853972435,
+      "learning_rate": 0.00019942440179070227,
+      "loss": 0.8904082179069519,
+      "step": 1586
+    },
+    {
+      "epoch": 0.6700421940928271,
+      "grad_norm": 0.4236849844455719,
+      "learning_rate": 0.00019941924450326138,
+      "loss": 0.8355640172958374,
+      "step": 1588
+    },
+    {
+      "epoch": 0.6708860759493671,
+      "grad_norm": 0.48624539375305176,
+      "learning_rate": 0.00019941406428171013,
+      "loss": 0.8898435831069946,
+      "step": 1590
+    },
+    {
+      "epoch": 0.6717299578059072,
+      "grad_norm": 0.4453780949115753,
+      "learning_rate": 0.00019940886112724355,
+      "loss": 0.9000012278556824,
+      "step": 1592
+    },
+    {
+      "epoch": 0.6725738396624472,
+      "grad_norm": 0.4417199194431305,
+      "learning_rate": 0.00019940363504106193,
+      "loss": 0.8171058893203735,
+      "step": 1594
+    },
+    {
+      "epoch": 0.6734177215189874,
+      "grad_norm": 0.45055362582206726,
+      "learning_rate": 0.00019939838602437074,
+      "loss": 0.8703798055648804,
+      "step": 1596
+    },
+    {
+      "epoch": 0.6742616033755274,
+      "grad_norm": 0.42316532135009766,
+      "learning_rate": 0.00019939311407838088,
+      "loss": 0.8677281737327576,
+      "step": 1598
+    },
+    {
+      "epoch": 0.6751054852320675,
+      "grad_norm": 0.3935781717300415,
+      "learning_rate": 0.0001993878192043085,
+      "loss": 0.7314013242721558,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6751054852320675,
+      "eval_loss": 0.8769772052764893,
+      "eval_runtime": 677.9338,
+      "eval_samples_per_second": 3.108,
+      "eval_steps_per_second": 3.108,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6759493670886076,
+      "grad_norm": 0.4197012186050415,
+      "learning_rate": 0.000199382501403375,
+      "loss": 0.8075380325317383,
+      "step": 1602
+    },
+    {
+      "epoch": 0.6767932489451477,
+      "grad_norm": 0.45227327942848206,
+      "learning_rate": 0.00019937716067680713,
+      "loss": 0.7885794639587402,
+      "step": 1604
+    },
+    {
+      "epoch": 0.6776371308016877,
+      "grad_norm": 0.45660385489463806,
+      "learning_rate": 0.00019937179702583682,
+      "loss": 0.868261992931366,
+      "step": 1606
+    },
+    {
+      "epoch": 0.6784810126582278,
+      "grad_norm": 0.3955288231372833,
+      "learning_rate": 0.00019936641045170144,
+      "loss": 0.8306655287742615,
+      "step": 1608
+    },
+    {
+      "epoch": 0.679324894514768,
+      "grad_norm": 0.3651324510574341,
+      "learning_rate": 0.00019936100095564353,
+      "loss": 0.7505315542221069,
+      "step": 1610
+    },
+    {
+      "epoch": 0.680168776371308,
+      "grad_norm": 0.4240955412387848,
+      "learning_rate": 0.00019935556853891096,
+      "loss": 0.8792756795883179,
+      "step": 1612
+    },
+    {
+      "epoch": 0.6810126582278481,
+      "grad_norm": 0.4142507314682007,
+      "learning_rate": 0.00019935011320275687,
+      "loss": 0.7328272461891174,
+      "step": 1614
+    },
+    {
+      "epoch": 0.6818565400843882,
+      "grad_norm": 0.48339322209358215,
+      "learning_rate": 0.00019934463494843975,
+      "loss": 0.8414849042892456,
+      "step": 1616
+    },
+    {
+      "epoch": 0.6827004219409283,
+      "grad_norm": 0.4312443435192108,
+      "learning_rate": 0.00019933913377722328,
+      "loss": 0.8588716983795166,
+      "step": 1618
+    },
+    {
+      "epoch": 0.6835443037974683,
+      "grad_norm": 0.43849512934684753,
+      "learning_rate": 0.0001993336096903765,
+      "loss": 0.846511960029602,
+      "step": 1620
+    },
+    {
+      "epoch": 0.6843881856540084,
+      "grad_norm": 0.45767566561698914,
+      "learning_rate": 0.0001993280626891737,
+      "loss": 0.9352323412895203,
+      "step": 1622
+    },
+    {
+      "epoch": 0.6852320675105485,
+      "grad_norm": 0.438412606716156,
+      "learning_rate": 0.00019932249277489444,
+      "loss": 0.8884757161140442,
+      "step": 1624
+    },
+    {
+      "epoch": 0.6860759493670886,
+      "grad_norm": 0.45893922448158264,
+      "learning_rate": 0.0001993168999488237,
+      "loss": 0.7739649415016174,
+      "step": 1626
+    },
+    {
+      "epoch": 0.6869198312236287,
+      "grad_norm": 0.38631772994995117,
+      "learning_rate": 0.0001993112842122515,
+      "loss": 0.7948258519172668,
+      "step": 1628
+    },
+    {
+      "epoch": 0.6877637130801688,
+      "grad_norm": 0.3824027180671692,
+      "learning_rate": 0.00019930564556647334,
+      "loss": 0.750511884689331,
+      "step": 1630
+    },
+    {
+      "epoch": 0.6886075949367089,
+      "grad_norm": 0.3980776071548462,
+      "learning_rate": 0.00019929998401278996,
+      "loss": 0.9006738662719727,
+      "step": 1632
+    },
+    {
+      "epoch": 0.6894514767932489,
+      "grad_norm": 0.41419896483421326,
+      "learning_rate": 0.00019929429955250734,
+      "loss": 0.6996869444847107,
+      "step": 1634
+    },
+    {
+      "epoch": 0.6902953586497891,
+      "grad_norm": 0.43783676624298096,
+      "learning_rate": 0.00019928859218693682,
+      "loss": 0.8523393273353577,
+      "step": 1636
+    },
+    {
+      "epoch": 0.6911392405063291,
+      "grad_norm": 0.4378126561641693,
+      "learning_rate": 0.00019928286191739497,
+      "loss": 0.8885987401008606,
+      "step": 1638
+    },
+    {
+      "epoch": 0.6919831223628692,
+      "grad_norm": 0.4276133179664612,
+      "learning_rate": 0.00019927710874520363,
+      "loss": 0.8563809990882874,
+      "step": 1640
+    },
+    {
+      "epoch": 0.6928270042194092,
+      "grad_norm": 0.4600282311439514,
+      "learning_rate": 0.00019927133267168992,
+      "loss": 0.854114294052124,
+      "step": 1642
+    },
+    {
+      "epoch": 0.6936708860759494,
+      "grad_norm": 0.5058461427688599,
+      "learning_rate": 0.00019926553369818635,
+      "loss": 0.8557889461517334,
+      "step": 1644
+    },
+    {
+      "epoch": 0.6945147679324895,
+      "grad_norm": 0.4947412312030792,
+      "learning_rate": 0.00019925971182603057,
+      "loss": 0.8622767925262451,
+      "step": 1646
+    },
+    {
+      "epoch": 0.6953586497890295,
+      "grad_norm": 0.42555496096611023,
+      "learning_rate": 0.00019925386705656558,
+      "loss": 0.7836558818817139,
+      "step": 1648
+    },
+    {
+      "epoch": 0.6962025316455697,
+      "grad_norm": 0.47560831904411316,
+      "learning_rate": 0.00019924799939113967,
+      "loss": 0.7877846956253052,
+      "step": 1650
+    },
+    {
+      "epoch": 0.6970464135021097,
+      "grad_norm": 0.391702264547348,
+      "learning_rate": 0.0001992421088311064,
+      "loss": 0.775681734085083,
+      "step": 1652
+    },
+    {
+      "epoch": 0.6978902953586498,
+      "grad_norm": 0.463679701089859,
+      "learning_rate": 0.0001992361953778246,
+      "loss": 0.8809882402420044,
+      "step": 1654
+    },
+    {
+      "epoch": 0.6987341772151898,
+      "grad_norm": 0.44096484780311584,
+      "learning_rate": 0.00019923025903265836,
+      "loss": 0.8513299226760864,
+      "step": 1656
+    },
+    {
+      "epoch": 0.69957805907173,
+      "grad_norm": 0.47818854451179504,
+      "learning_rate": 0.0001992242997969772,
+      "loss": 0.8309667706489563,
+      "step": 1658
+    },
+    {
+      "epoch": 0.70042194092827,
+      "grad_norm": 0.4192790985107422,
+      "learning_rate": 0.00019921831767215565,
+      "loss": 0.8817558884620667,
+      "step": 1660
+    },
+    {
+      "epoch": 0.7012658227848101,
+      "grad_norm": 0.45285239815711975,
+      "learning_rate": 0.00019921231265957373,
+      "loss": 0.8315755724906921,
+      "step": 1662
+    },
+    {
+      "epoch": 0.7021097046413503,
+      "grad_norm": 0.47118625044822693,
+      "learning_rate": 0.00019920628476061673,
+      "loss": 0.8698058724403381,
+      "step": 1664
+    },
+    {
+      "epoch": 0.7029535864978903,
+      "grad_norm": 0.4769238829612732,
+      "learning_rate": 0.0001992002339766751,
+      "loss": 0.953620970249176,
+      "step": 1666
+    },
+    {
+      "epoch": 0.7037974683544304,
+      "grad_norm": 0.39397355914115906,
+      "learning_rate": 0.00019919416030914468,
+      "loss": 0.8536615371704102,
+      "step": 1668
+    },
+    {
+      "epoch": 0.7046413502109705,
+      "grad_norm": 0.3659648299217224,
+      "learning_rate": 0.00019918806375942655,
+      "loss": 0.8159828782081604,
+      "step": 1670
+    },
+    {
+      "epoch": 0.7054852320675106,
+      "grad_norm": 0.4381789565086365,
+      "learning_rate": 0.00019918194432892703,
+      "loss": 0.782463550567627,
+      "step": 1672
+    },
+    {
+      "epoch": 0.7063291139240506,
+      "grad_norm": 0.44745752215385437,
+      "learning_rate": 0.00019917580201905778,
+      "loss": 0.9314272999763489,
+      "step": 1674
+    },
+    {
+      "epoch": 0.7071729957805907,
+      "grad_norm": 0.4515164792537689,
+      "learning_rate": 0.00019916963683123574,
+      "loss": 0.860816240310669,
+      "step": 1676
+    },
+    {
+      "epoch": 0.7080168776371308,
+      "grad_norm": 0.42735886573791504,
+      "learning_rate": 0.00019916344876688303,
+      "loss": 0.8008456826210022,
+      "step": 1678
+    },
+    {
+      "epoch": 0.7088607594936709,
+      "grad_norm": 0.418275386095047,
+      "learning_rate": 0.00019915723782742718,
+      "loss": 0.7796595096588135,
+      "step": 1680
+    },
+    {
+      "epoch": 0.7097046413502109,
+      "grad_norm": 0.44347989559173584,
+      "learning_rate": 0.00019915100401430089,
+      "loss": 0.9006506204605103,
+      "step": 1682
+    },
+    {
+      "epoch": 0.7105485232067511,
+      "grad_norm": 0.3932702839374542,
+      "learning_rate": 0.00019914474732894225,
+      "loss": 0.7750146985054016,
+      "step": 1684
+    },
+    {
+      "epoch": 0.7113924050632912,
+      "grad_norm": 0.4810502231121063,
+      "learning_rate": 0.00019913846777279447,
+      "loss": 0.7508097290992737,
+      "step": 1686
+    },
+    {
+      "epoch": 0.7122362869198312,
+      "grad_norm": 0.4166778028011322,
+      "learning_rate": 0.00019913216534730622,
+      "loss": 0.7926796078681946,
+      "step": 1688
+    },
+    {
+      "epoch": 0.7130801687763713,
+      "grad_norm": 0.47900474071502686,
+      "learning_rate": 0.00019912584005393123,
+      "loss": 0.8363928198814392,
+      "step": 1690
+    },
+    {
+      "epoch": 0.7139240506329114,
+      "grad_norm": 0.435143381357193,
+      "learning_rate": 0.00019911949189412872,
+      "loss": 0.8649623990058899,
+      "step": 1692
+    },
+    {
+      "epoch": 0.7147679324894515,
+      "grad_norm": 0.4070208966732025,
+      "learning_rate": 0.00019911312086936305,
+      "loss": 0.8196499943733215,
+      "step": 1694
+    },
+    {
+      "epoch": 0.7156118143459915,
+      "grad_norm": 0.4717724621295929,
+      "learning_rate": 0.0001991067269811039,
+      "loss": 0.9157413840293884,
+      "step": 1696
+    },
+    {
+      "epoch": 0.7164556962025317,
+      "grad_norm": 0.3942880928516388,
+      "learning_rate": 0.0001991003102308262,
+      "loss": 0.7880456447601318,
+      "step": 1698
+    },
+    {
+      "epoch": 0.7172995780590717,
+      "grad_norm": 0.4037776589393616,
+      "learning_rate": 0.0001990938706200102,
+      "loss": 0.8088646531105042,
+      "step": 1700
+    },
+    {
+      "epoch": 0.7172995780590717,
+      "eval_loss": 0.8708170056343079,
+      "eval_runtime": 670.3019,
+      "eval_samples_per_second": 3.143,
+      "eval_steps_per_second": 3.143,
+      "step": 1700
+    },
+    {
+      "epoch": 0.7181434599156118,
+      "grad_norm": 0.4443536400794983,
+      "learning_rate": 0.0001990874081501414,
+      "loss": 0.8740324378013611,
+      "step": 1702
+    },
+    {
+      "epoch": 0.7189873417721518,
+      "grad_norm": 0.45089733600616455,
+      "learning_rate": 0.00019908092282271053,
+      "loss": 0.8305023908615112,
+      "step": 1704
+    },
+    {
+      "epoch": 0.719831223628692,
+      "grad_norm": 0.3903036415576935,
+      "learning_rate": 0.00019907441463921366,
+      "loss": 0.7718394994735718,
+      "step": 1706
+    },
+    {
+      "epoch": 0.7206751054852321,
+      "grad_norm": 0.42990729212760925,
+      "learning_rate": 0.00019906788360115208,
+      "loss": 0.843289315700531,
+      "step": 1708
+    },
+    {
+      "epoch": 0.7215189873417721,
+      "grad_norm": 0.4287833571434021,
+      "learning_rate": 0.0001990613297100324,
+      "loss": 0.8332282900810242,
+      "step": 1710
+    },
+    {
+      "epoch": 0.7223628691983123,
+      "grad_norm": 0.4557269513607025,
+      "learning_rate": 0.00019905475296736646,
+      "loss": 0.784180223941803,
+      "step": 1712
+    },
+    {
+      "epoch": 0.7232067510548523,
+      "grad_norm": 0.4001525640487671,
+      "learning_rate": 0.0001990481533746714,
+      "loss": 0.7807716131210327,
+      "step": 1714
+    },
+    {
+      "epoch": 0.7240506329113924,
+      "grad_norm": 0.5208709836006165,
+      "learning_rate": 0.0001990415309334696,
+      "loss": 0.7936414480209351,
+      "step": 1716
+    },
+    {
+      "epoch": 0.7248945147679325,
+      "grad_norm": 0.46469953656196594,
+      "learning_rate": 0.00019903488564528875,
+      "loss": 0.7964845895767212,
+      "step": 1718
+    },
+    {
+      "epoch": 0.7257383966244726,
+      "grad_norm": 0.405773401260376,
+      "learning_rate": 0.00019902821751166178,
+      "loss": 0.8302215337753296,
+      "step": 1720
+    },
+    {
+      "epoch": 0.7265822784810126,
+      "grad_norm": 0.3874973654747009,
+      "learning_rate": 0.00019902152653412692,
+      "loss": 0.8486512899398804,
+      "step": 1722
+    },
+    {
+      "epoch": 0.7274261603375527,
+      "grad_norm": 0.46865755319595337,
+      "learning_rate": 0.00019901481271422758,
+      "loss": 0.837841272354126,
+      "step": 1724
+    },
+    {
+      "epoch": 0.7282700421940929,
+      "grad_norm": 0.4139123558998108,
+      "learning_rate": 0.0001990080760535126,
+      "loss": 0.7640942335128784,
+      "step": 1726
+    },
+    {
+      "epoch": 0.7291139240506329,
+      "grad_norm": 0.5003024935722351,
+      "learning_rate": 0.000199001316553536,
+      "loss": 0.8236244320869446,
+      "step": 1728
+    },
+    {
+      "epoch": 0.729957805907173,
+      "grad_norm": 0.38495776057243347,
+      "learning_rate": 0.00019899453421585696,
+      "loss": 0.7792612314224243,
+      "step": 1730
+    },
+    {
+      "epoch": 0.7308016877637131,
+      "grad_norm": 0.4462733268737793,
+      "learning_rate": 0.00019898772904204014,
+      "loss": 0.7733872532844543,
+      "step": 1732
+    },
+    {
+      "epoch": 0.7316455696202532,
+      "grad_norm": 0.398248553276062,
+      "learning_rate": 0.00019898090103365532,
+      "loss": 0.8933147192001343,
+      "step": 1734
+    },
+    {
+      "epoch": 0.7324894514767932,
+      "grad_norm": 0.4231175482273102,
+      "learning_rate": 0.00019897405019227756,
+      "loss": 0.8198356628417969,
+      "step": 1736
+    },
+    {
+      "epoch": 0.7333333333333333,
+      "grad_norm": 0.43639084696769714,
+      "learning_rate": 0.0001989671765194873,
+      "loss": 0.8266391158103943,
+      "step": 1738
+    },
+    {
+      "epoch": 0.7341772151898734,
+      "grad_norm": 0.4235835671424866,
+      "learning_rate": 0.00019896028001687008,
+      "loss": 0.7839997410774231,
+      "step": 1740
+    },
+    {
+      "epoch": 0.7350210970464135,
+      "grad_norm": 0.4033111035823822,
+      "learning_rate": 0.00019895336068601687,
+      "loss": 0.7153518199920654,
+      "step": 1742
+    },
+    {
+      "epoch": 0.7358649789029535,
+      "grad_norm": 0.44281265139579773,
+      "learning_rate": 0.00019894641852852377,
+      "loss": 0.9079289436340332,
+      "step": 1744
+    },
+    {
+      "epoch": 0.7367088607594937,
+      "grad_norm": 0.4882747232913971,
+      "learning_rate": 0.0001989394535459922,
+      "loss": 0.8645254373550415,
+      "step": 1746
+    },
+    {
+      "epoch": 0.7375527426160338,
+      "grad_norm": 0.34858641028404236,
+      "learning_rate": 0.00019893246574002887,
+      "loss": 0.8134214878082275,
+      "step": 1748
+    },
+    {
+      "epoch": 0.7383966244725738,
+      "grad_norm": 0.4273303747177124,
+      "learning_rate": 0.00019892545511224574,
+      "loss": 0.8142663240432739,
+      "step": 1750
+    },
+    {
+      "epoch": 0.739240506329114,
+      "grad_norm": 0.4289957284927368,
+      "learning_rate": 0.00019891842166426004,
+      "loss": 0.7957643866539001,
+      "step": 1752
+    },
+    {
+      "epoch": 0.740084388185654,
+      "grad_norm": 0.4812224805355072,
+      "learning_rate": 0.00019891136539769416,
+      "loss": 0.9000363945960999,
+      "step": 1754
+    },
+    {
+      "epoch": 0.7409282700421941,
+      "grad_norm": 0.4078022837638855,
+      "learning_rate": 0.00019890428631417595,
+      "loss": 0.7719792723655701,
+      "step": 1756
+    },
+    {
+      "epoch": 0.7417721518987341,
+      "grad_norm": 0.40169858932495117,
+      "learning_rate": 0.00019889718441533838,
+      "loss": 0.7681847214698792,
+      "step": 1758
+    },
+    {
+      "epoch": 0.7426160337552743,
+      "grad_norm": 0.3713594377040863,
+      "learning_rate": 0.00019889005970281972,
+      "loss": 0.7536827325820923,
+      "step": 1760
+    },
+    {
+      "epoch": 0.7434599156118143,
+      "grad_norm": 0.4183627665042877,
+      "learning_rate": 0.0001988829121782635,
+      "loss": 0.9000160694122314,
+      "step": 1762
+    },
+    {
+      "epoch": 0.7443037974683544,
+      "grad_norm": 0.4241594672203064,
+      "learning_rate": 0.00019887574184331851,
+      "loss": 0.8047307133674622,
+      "step": 1764
+    },
+    {
+      "epoch": 0.7451476793248946,
+      "grad_norm": 0.4249233603477478,
+      "learning_rate": 0.00019886854869963883,
+      "loss": 0.8932583928108215,
+      "step": 1766
+    },
+    {
+      "epoch": 0.7459915611814346,
+      "grad_norm": 0.35815340280532837,
+      "learning_rate": 0.00019886133274888378,
+      "loss": 0.7684977054595947,
+      "step": 1768
+    },
+    {
+      "epoch": 0.7468354430379747,
+      "grad_norm": 0.41035401821136475,
+      "learning_rate": 0.00019885409399271795,
+      "loss": 0.8644338250160217,
+      "step": 1770
+    },
+    {
+      "epoch": 0.7476793248945147,
+      "grad_norm": 0.5450286269187927,
+      "learning_rate": 0.00019884683243281116,
+      "loss": 0.8035860657691956,
+      "step": 1772
+    },
+    {
+      "epoch": 0.7485232067510549,
+      "grad_norm": 0.5280999541282654,
+      "learning_rate": 0.0001988395480708385,
+      "loss": 0.8861207962036133,
+      "step": 1774
+    },
+    {
+      "epoch": 0.7493670886075949,
+      "grad_norm": 0.42665231227874756,
+      "learning_rate": 0.00019883224090848036,
+      "loss": 0.7506847977638245,
+      "step": 1776
+    },
+    {
+      "epoch": 0.750210970464135,
+      "grad_norm": 0.39029282331466675,
+      "learning_rate": 0.00019882491094742232,
+      "loss": 0.7662046551704407,
+      "step": 1778
+    },
+    {
+      "epoch": 0.7510548523206751,
+      "grad_norm": 0.43870019912719727,
+      "learning_rate": 0.00019881755818935534,
+      "loss": 0.7590143084526062,
+      "step": 1780
+    },
+    {
+      "epoch": 0.7518987341772152,
+      "grad_norm": 0.4195050001144409,
+      "learning_rate": 0.0001988101826359755,
+      "loss": 0.9340365529060364,
+      "step": 1782
+    },
+    {
+      "epoch": 0.7527426160337553,
+      "grad_norm": 0.4236123263835907,
+      "learning_rate": 0.0001988027842889842,
+      "loss": 0.7969209551811218,
+      "step": 1784
+    },
+    {
+      "epoch": 0.7535864978902953,
+      "grad_norm": 0.4274357259273529,
+      "learning_rate": 0.00019879536315008808,
+      "loss": 0.7892382144927979,
+      "step": 1786
+    },
+    {
+      "epoch": 0.7544303797468355,
+      "grad_norm": 0.4622916579246521,
+      "learning_rate": 0.00019878791922099912,
+      "loss": 0.8752562403678894,
+      "step": 1788
+    },
+    {
+      "epoch": 0.7552742616033755,
+      "grad_norm": 0.3648734986782074,
+      "learning_rate": 0.00019878045250343445,
+      "loss": 0.813011884689331,
+      "step": 1790
+    },
+    {
+      "epoch": 0.7561181434599156,
+      "grad_norm": 0.41361239552497864,
+      "learning_rate": 0.00019877296299911648,
+      "loss": 0.7310198545455933,
+      "step": 1792
+    },
+    {
+      "epoch": 0.7569620253164557,
+      "grad_norm": 0.4113222062587738,
+      "learning_rate": 0.00019876545070977294,
+      "loss": 0.858386754989624,
+      "step": 1794
+    },
+    {
+      "epoch": 0.7578059071729958,
+      "grad_norm": 0.4295370876789093,
+      "learning_rate": 0.00019875791563713676,
+      "loss": 0.8377325534820557,
+      "step": 1796
+    },
+    {
+      "epoch": 0.7586497890295358,
+      "grad_norm": 0.4227522015571594,
+      "learning_rate": 0.0001987503577829461,
+      "loss": 0.9139418005943298,
+      "step": 1798
+    },
+    {
+      "epoch": 0.759493670886076,
+      "grad_norm": 0.3638151288032532,
+      "learning_rate": 0.00019874277714894442,
+      "loss": 0.6994872689247131,
+      "step": 1800
+    },
+    {
+      "epoch": 0.759493670886076,
+      "eval_loss": 0.8625519275665283,
+      "eval_runtime": 686.4271,
+      "eval_samples_per_second": 3.07,
+      "eval_steps_per_second": 3.07,
+      "step": 1800
+    },
+    {
+      "epoch": 0.760337552742616,
+      "grad_norm": 0.46954411268234253,
+      "learning_rate": 0.00019873517373688047,
+      "loss": 0.8681167960166931,
+      "step": 1802
+    },
+    {
+      "epoch": 0.7611814345991561,
+      "grad_norm": 0.41958674788475037,
+      "learning_rate": 0.00019872754754850819,
+      "loss": 0.7420852184295654,
+      "step": 1804
+    },
+    {
+      "epoch": 0.7620253164556962,
+      "grad_norm": 0.431226521730423,
+      "learning_rate": 0.00019871989858558678,
+      "loss": 0.8018608093261719,
+      "step": 1806
+    },
+    {
+      "epoch": 0.7628691983122363,
+      "grad_norm": 0.45129454135894775,
+      "learning_rate": 0.0001987122268498807,
+      "loss": 0.8793904781341553,
+      "step": 1808
+    },
+    {
+      "epoch": 0.7637130801687764,
+      "grad_norm": 0.44278961420059204,
+      "learning_rate": 0.00019870453234315972,
+      "loss": 0.8416730761528015,
+      "step": 1810
+    },
+    {
+      "epoch": 0.7645569620253164,
+      "grad_norm": 0.4689098000526428,
+      "learning_rate": 0.00019869681506719876,
+      "loss": 0.8225743174552917,
+      "step": 1812
+    },
+    {
+      "epoch": 0.7654008438818566,
+      "grad_norm": 0.4413386881351471,
+      "learning_rate": 0.00019868907502377806,
+      "loss": 0.7871913313865662,
+      "step": 1814
+    },
+    {
+      "epoch": 0.7662447257383966,
+      "grad_norm": 0.48395273089408875,
+      "learning_rate": 0.00019868131221468315,
+      "loss": 0.8616237044334412,
+      "step": 1816
+    },
+    {
+      "epoch": 0.7670886075949367,
+      "grad_norm": 0.370263934135437,
+      "learning_rate": 0.0001986735266417047,
+      "loss": 0.7243452668190002,
+      "step": 1818
+    },
+    {
+      "epoch": 0.7679324894514767,
+      "grad_norm": 0.4522445797920227,
+      "learning_rate": 0.00019866571830663875,
+      "loss": 0.8623812198638916,
+      "step": 1820
+    },
+    {
+      "epoch": 0.7687763713080169,
+      "grad_norm": 0.43821004033088684,
+      "learning_rate": 0.00019865788721128643,
+      "loss": 0.776618480682373,
+      "step": 1822
+    },
+    {
+      "epoch": 0.769620253164557,
+      "grad_norm": 0.3825650215148926,
+      "learning_rate": 0.00019865003335745436,
+      "loss": 0.8722774982452393,
+      "step": 1824
+    },
+    {
+      "epoch": 0.770464135021097,
+      "grad_norm": 0.43684205412864685,
+      "learning_rate": 0.00019864215674695416,
+      "loss": 0.907802939414978,
+      "step": 1826
+    },
+    {
+      "epoch": 0.7713080168776372,
+      "grad_norm": 0.4128975570201874,
+      "learning_rate": 0.00019863425738160287,
+      "loss": 0.8505539298057556,
+      "step": 1828
+    },
+    {
+      "epoch": 0.7721518987341772,
+      "grad_norm": 0.4254065454006195,
+      "learning_rate": 0.0001986263352632227,
+      "loss": 0.83274906873703,
+      "step": 1830
+    },
+    {
+      "epoch": 0.7729957805907173,
+      "grad_norm": 0.4137375056743622,
+      "learning_rate": 0.00019861839039364118,
+      "loss": 0.8172078728675842,
+      "step": 1832
+    },
+    {
+      "epoch": 0.7738396624472574,
+      "grad_norm": 0.4387703239917755,
+      "learning_rate": 0.00019861042277469096,
+      "loss": 0.8610842227935791,
+      "step": 1834
+    },
+    {
+      "epoch": 0.7746835443037975,
+      "grad_norm": 0.398952454328537,
+      "learning_rate": 0.00019860243240821005,
+      "loss": 0.7961188554763794,
+      "step": 1836
+    },
+    {
+      "epoch": 0.7755274261603375,
+      "grad_norm": 0.4079921543598175,
+      "learning_rate": 0.0001985944192960417,
+      "loss": 0.8078321218490601,
+      "step": 1838
+    },
+    {
+      "epoch": 0.7763713080168776,
+      "grad_norm": 0.4701489508152008,
+      "learning_rate": 0.00019858638344003433,
+      "loss": 0.8743909001350403,
+      "step": 1840
+    },
+    {
+      "epoch": 0.7772151898734178,
+      "grad_norm": 0.5391269326210022,
+      "learning_rate": 0.00019857832484204173,
+      "loss": 0.7681707143783569,
+      "step": 1842
+    },
+    {
+      "epoch": 0.7780590717299578,
+      "grad_norm": 0.36250734329223633,
+      "learning_rate": 0.00019857024350392277,
+      "loss": 0.7817525863647461,
+      "step": 1844
+    },
+    {
+      "epoch": 0.7789029535864979,
+      "grad_norm": 0.41237136721611023,
+      "learning_rate": 0.00019856213942754175,
+      "loss": 0.7614551782608032,
+      "step": 1846
+    },
+    {
+      "epoch": 0.779746835443038,
+      "grad_norm": 0.3968869745731354,
+      "learning_rate": 0.00019855401261476807,
+      "loss": 0.7492313981056213,
+      "step": 1848
+    },
+    {
+      "epoch": 0.7805907172995781,
+      "grad_norm": 0.441476434469223,
+      "learning_rate": 0.00019854586306747646,
+      "loss": 0.8615695238113403,
+      "step": 1850
+    },
+    {
+      "epoch": 0.7814345991561181,
+      "grad_norm": 0.4486154317855835,
+      "learning_rate": 0.00019853769078754686,
+      "loss": 0.8416724801063538,
+      "step": 1852
+    },
+    {
+      "epoch": 0.7822784810126582,
+      "grad_norm": 0.45148879289627075,
+      "learning_rate": 0.0001985294957768644,
+      "loss": 0.8987806439399719,
+      "step": 1854
+    },
+    {
+      "epoch": 0.7831223628691983,
+      "grad_norm": 0.39330095052719116,
+      "learning_rate": 0.0001985212780373196,
+      "loss": 0.7597590088844299,
+      "step": 1856
+    },
+    {
+      "epoch": 0.7839662447257384,
+      "grad_norm": 0.4260408580303192,
+      "learning_rate": 0.00019851303757080808,
+      "loss": 0.8429927229881287,
+      "step": 1858
+    },
+    {
+      "epoch": 0.7848101265822784,
+      "grad_norm": 0.37519171833992004,
+      "learning_rate": 0.00019850477437923075,
+      "loss": 0.7443564534187317,
+      "step": 1860
+    },
+    {
+      "epoch": 0.7856540084388186,
+      "grad_norm": 0.4991084337234497,
+      "learning_rate": 0.00019849648846449383,
+      "loss": 0.934238612651825,
+      "step": 1862
+    },
+    {
+      "epoch": 0.7864978902953587,
+      "grad_norm": 0.4711335003376007,
+      "learning_rate": 0.00019848817982850867,
+      "loss": 0.8938905000686646,
+      "step": 1864
+    },
+    {
+      "epoch": 0.7873417721518987,
+      "grad_norm": 0.4145370423793793,
+      "learning_rate": 0.00019847984847319193,
+      "loss": 0.8745643496513367,
+      "step": 1866
+    },
+    {
+      "epoch": 0.7881856540084389,
+      "grad_norm": 0.3444044888019562,
+      "learning_rate": 0.00019847149440046548,
+      "loss": 0.7178786396980286,
+      "step": 1868
+    },
+    {
+      "epoch": 0.7890295358649789,
+      "grad_norm": 0.34632962942123413,
+      "learning_rate": 0.00019846311761225646,
+      "loss": 0.9114303588867188,
+      "step": 1870
+    },
+    {
+      "epoch": 0.789873417721519,
+      "grad_norm": 0.38630741834640503,
+      "learning_rate": 0.0001984547181104972,
+      "loss": 0.836281955242157,
+      "step": 1872
+    },
+    {
+      "epoch": 0.790717299578059,
+      "grad_norm": 0.40947434306144714,
+      "learning_rate": 0.00019844629589712534,
+      "loss": 0.8247858881950378,
+      "step": 1874
+    },
+    {
+      "epoch": 0.7915611814345992,
+      "grad_norm": 0.45878684520721436,
+      "learning_rate": 0.00019843785097408372,
+      "loss": 0.8446351289749146,
+      "step": 1876
+    },
+    {
+      "epoch": 0.7924050632911392,
+      "grad_norm": 0.37202852964401245,
+      "learning_rate": 0.0001984293833433204,
+      "loss": 0.8567686676979065,
+      "step": 1878
+    },
+    {
+      "epoch": 0.7932489451476793,
+      "grad_norm": 0.3839069902896881,
+      "learning_rate": 0.00019842089300678876,
+      "loss": 0.8134047985076904,
+      "step": 1880
+    },
+    {
+      "epoch": 0.7940928270042195,
+      "grad_norm": 0.4018687605857849,
+      "learning_rate": 0.00019841237996644727,
+      "loss": 0.8670110106468201,
+      "step": 1882
+    },
+    {
+      "epoch": 0.7949367088607595,
+      "grad_norm": 0.372089684009552,
+      "learning_rate": 0.0001984038442242598,
+      "loss": 0.7975355386734009,
+      "step": 1884
+    },
+    {
+      "epoch": 0.7957805907172996,
+      "grad_norm": 0.4492705166339874,
+      "learning_rate": 0.0001983952857821953,
+      "loss": 0.8755611777305603,
+      "step": 1886
+    },
+    {
+      "epoch": 0.7966244725738396,
+      "grad_norm": 0.451054185628891,
+      "learning_rate": 0.00019838670464222813,
+      "loss": 0.8680241107940674,
+      "step": 1888
+    },
+    {
+      "epoch": 0.7974683544303798,
+      "grad_norm": 0.4036900997161865,
+      "learning_rate": 0.00019837810080633773,
+      "loss": 0.7990990877151489,
+      "step": 1890
+    },
+    {
+      "epoch": 0.7983122362869198,
+      "grad_norm": 0.46125802397727966,
+      "learning_rate": 0.0001983694742765089,
+      "loss": 0.8598953485488892,
+      "step": 1892
+    },
+    {
+      "epoch": 0.7991561181434599,
+      "grad_norm": 0.4458785951137543,
+      "learning_rate": 0.00019836082505473153,
+      "loss": 0.7707474827766418,
+      "step": 1894
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.44501590728759766,
+      "learning_rate": 0.0001983521531430009,
+      "loss": 0.8069534301757812,
+      "step": 1896
+    },
+    {
+      "epoch": 0.8008438818565401,
+      "grad_norm": 0.400771826505661,
+      "learning_rate": 0.0001983434585433174,
+      "loss": 0.862334668636322,
+      "step": 1898
+    },
+    {
+      "epoch": 0.8016877637130801,
+      "grad_norm": 0.39193347096443176,
+      "learning_rate": 0.00019833474125768676,
+      "loss": 0.8125433325767517,
+      "step": 1900
+    },
+    {
+      "epoch": 0.8016877637130801,
+      "eval_loss": 0.8546335697174072,
+      "eval_runtime": 688.5301,
+      "eval_samples_per_second": 3.06,
+      "eval_steps_per_second": 3.06,
+      "step": 1900
+    },
+    {
+      "epoch": 0.8025316455696202,
+      "grad_norm": 0.441129207611084,
+      "learning_rate": 0.00019832600128811986,
+      "loss": 0.8246012330055237,
+      "step": 1902
+    },
+    {
+      "epoch": 0.8033755274261604,
+      "grad_norm": 0.4089467227458954,
+      "learning_rate": 0.00019831723863663285,
+      "loss": 0.8387641310691833,
+      "step": 1904
+    },
+    {
+      "epoch": 0.8042194092827004,
+      "grad_norm": 0.4871654510498047,
+      "learning_rate": 0.0001983084533052471,
+      "loss": 0.7891429662704468,
+      "step": 1906
+    },
+    {
+      "epoch": 0.8050632911392405,
+      "grad_norm": 0.40085604786872864,
+      "learning_rate": 0.00019829964529598923,
+      "loss": 0.8008774518966675,
+      "step": 1908
+    },
+    {
+      "epoch": 0.8059071729957806,
+      "grad_norm": 0.38775014877319336,
+      "learning_rate": 0.00019829081461089109,
+      "loss": 0.7921834588050842,
+      "step": 1910
+    },
+    {
+      "epoch": 0.8067510548523207,
+      "grad_norm": 0.37830594182014465,
+      "learning_rate": 0.0001982819612519897,
+      "loss": 0.7752519845962524,
+      "step": 1912
+    },
+    {
+      "epoch": 0.8075949367088607,
+      "grad_norm": 0.41137194633483887,
+      "learning_rate": 0.0001982730852213274,
+      "loss": 0.7958255410194397,
+      "step": 1914
+    },
+    {
+      "epoch": 0.8084388185654009,
+      "grad_norm": 0.4871830940246582,
+      "learning_rate": 0.0001982641865209517,
+      "loss": 0.8177281618118286,
+      "step": 1916
+    },
+    {
+      "epoch": 0.809282700421941,
+      "grad_norm": 0.4157388210296631,
+      "learning_rate": 0.0001982552651529154,
+      "loss": 0.7726616859436035,
+      "step": 1918
+    },
+    {
+      "epoch": 0.810126582278481,
+      "grad_norm": 0.32338014245033264,
+      "learning_rate": 0.00019824632111927645,
+      "loss": 0.6601400375366211,
+      "step": 1920
+    },
+    {
+      "epoch": 0.810970464135021,
+      "grad_norm": 0.4642949104309082,
+      "learning_rate": 0.00019823735442209804,
+      "loss": 0.7866622805595398,
+      "step": 1922
+    },
+    {
+      "epoch": 0.8118143459915612,
+      "grad_norm": 0.43558797240257263,
+      "learning_rate": 0.00019822836506344865,
+      "loss": 0.7857986092567444,
+      "step": 1924
+    },
+    {
+      "epoch": 0.8126582278481013,
+      "grad_norm": 0.3996953070163727,
+      "learning_rate": 0.000198219353045402,
+      "loss": 0.8407763838768005,
+      "step": 1926
+    },
+    {
+      "epoch": 0.8135021097046413,
+      "grad_norm": 0.4415414333343506,
+      "learning_rate": 0.00019821031837003686,
+      "loss": 0.8433752655982971,
+      "step": 1928
+    },
+    {
+      "epoch": 0.8143459915611815,
+      "grad_norm": 0.37329310178756714,
+      "learning_rate": 0.00019820126103943752,
+      "loss": 0.6941158175468445,
+      "step": 1930
+    },
+    {
+      "epoch": 0.8151898734177215,
+      "grad_norm": 0.4482601284980774,
+      "learning_rate": 0.00019819218105569323,
+      "loss": 0.775604248046875,
+      "step": 1932
+    },
+    {
+      "epoch": 0.8160337552742616,
+      "grad_norm": 0.426194429397583,
+      "learning_rate": 0.00019818307842089854,
+      "loss": 0.8496418595314026,
+      "step": 1934
+    },
+    {
+      "epoch": 0.8168776371308016,
+      "grad_norm": 0.42379093170166016,
+      "learning_rate": 0.00019817395313715335,
+      "loss": 0.7767958045005798,
+      "step": 1936
+    },
+    {
+      "epoch": 0.8177215189873418,
+      "grad_norm": 0.4149782061576843,
+      "learning_rate": 0.0001981648052065626,
+      "loss": 0.7347666025161743,
+      "step": 1938
+    },
+    {
+      "epoch": 0.8185654008438819,
+      "grad_norm": 0.3927431106567383,
+      "learning_rate": 0.0001981556346312366,
+      "loss": 0.7839647531509399,
+      "step": 1940
+    },
+    {
+      "epoch": 0.8194092827004219,
+      "grad_norm": 0.4483136236667633,
+      "learning_rate": 0.00019814644141329083,
+      "loss": 0.8870531916618347,
+      "step": 1942
+    },
+    {
+      "epoch": 0.8202531645569621,
+      "grad_norm": 0.4400147497653961,
+      "learning_rate": 0.00019813722555484594,
+      "loss": 0.8646620512008667,
+      "step": 1944
+    },
+    {
+      "epoch": 0.8210970464135021,
+      "grad_norm": 0.469971626996994,
+      "learning_rate": 0.00019812798705802785,
+      "loss": 0.9576541781425476,
+      "step": 1946
+    },
+    {
+      "epoch": 0.8219409282700422,
+      "grad_norm": 0.388113409280777,
+      "learning_rate": 0.00019811872592496776,
+      "loss": 0.7956477403640747,
+      "step": 1948
+    },
+    {
+      "epoch": 0.8227848101265823,
+      "grad_norm": 0.4271424114704132,
+      "learning_rate": 0.00019810944215780201,
+      "loss": 0.8440352082252502,
+      "step": 1950
+    },
+    {
+      "epoch": 0.8236286919831224,
+      "grad_norm": 0.4097966253757477,
+      "learning_rate": 0.00019810013575867217,
+      "loss": 0.8589065074920654,
+      "step": 1952
+    },
+    {
+      "epoch": 0.8244725738396624,
+      "grad_norm": 0.41550710797309875,
+      "learning_rate": 0.00019809080672972504,
+      "loss": 0.8606626987457275,
+      "step": 1954
+    },
+    {
+      "epoch": 0.8253164556962025,
+      "grad_norm": 0.3680916130542755,
+      "learning_rate": 0.0001980814550731127,
+      "loss": 0.7582436800003052,
+      "step": 1956
+    },
+    {
+      "epoch": 0.8261603375527427,
+      "grad_norm": 0.573297917842865,
+      "learning_rate": 0.00019807208079099234,
+      "loss": 0.8345186710357666,
+      "step": 1958
+    },
+    {
+      "epoch": 0.8270042194092827,
+      "grad_norm": 0.43631303310394287,
+      "learning_rate": 0.00019806268388552646,
+      "loss": 0.889683723449707,
+      "step": 1960
+    },
+    {
+      "epoch": 0.8278481012658228,
+      "grad_norm": 0.442531019449234,
+      "learning_rate": 0.00019805326435888275,
+      "loss": 0.8092973232269287,
+      "step": 1962
+    },
+    {
+      "epoch": 0.8286919831223629,
+      "grad_norm": 0.45020824670791626,
+      "learning_rate": 0.0001980438222132341,
+      "loss": 0.8919535279273987,
+      "step": 1964
+    },
+    {
+      "epoch": 0.829535864978903,
+      "grad_norm": 0.43131786584854126,
+      "learning_rate": 0.0001980343574507586,
+      "loss": 0.7652381658554077,
+      "step": 1966
+    },
+    {
+      "epoch": 0.830379746835443,
+      "grad_norm": 0.4461326599121094,
+      "learning_rate": 0.00019802487007363967,
+      "loss": 0.8512389659881592,
+      "step": 1968
+    },
+    {
+      "epoch": 0.8312236286919831,
+      "grad_norm": 0.4676894247531891,
+      "learning_rate": 0.0001980153600840658,
+      "loss": 0.8681327700614929,
+      "step": 1970
+    },
+    {
+      "epoch": 0.8320675105485232,
+      "grad_norm": 0.38376888632774353,
+      "learning_rate": 0.00019800582748423075,
+      "loss": 0.7476955652236938,
+      "step": 1972
+    },
+    {
+      "epoch": 0.8329113924050633,
+      "grad_norm": 0.37173229455947876,
+      "learning_rate": 0.00019799627227633354,
+      "loss": 0.814192533493042,
+      "step": 1974
+    },
+    {
+      "epoch": 0.8337552742616033,
+      "grad_norm": 0.4345237910747528,
+      "learning_rate": 0.00019798669446257844,
+      "loss": 0.7668994069099426,
+      "step": 1976
+    },
+    {
+      "epoch": 0.8345991561181435,
+      "grad_norm": 0.45872583985328674,
+      "learning_rate": 0.0001979770940451747,
+      "loss": 0.8321775197982788,
+      "step": 1978
+    },
+    {
+      "epoch": 0.8354430379746836,
+      "grad_norm": 0.40618205070495605,
+      "learning_rate": 0.0001979674710263371,
+      "loss": 0.7824342250823975,
+      "step": 1980
+    },
+    {
+      "epoch": 0.8362869198312236,
+      "grad_norm": 0.38377392292022705,
+      "learning_rate": 0.00019795782540828544,
+      "loss": 0.8752480745315552,
+      "step": 1982
+    },
+    {
+      "epoch": 0.8371308016877637,
+      "grad_norm": 0.36340072751045227,
+      "learning_rate": 0.00019794815719324476,
+      "loss": 0.66700679063797,
+      "step": 1984
+    },
+    {
+      "epoch": 0.8379746835443038,
+      "grad_norm": 0.4638384282588959,
+      "learning_rate": 0.00019793846638344536,
+      "loss": 0.8582209944725037,
+      "step": 1986
+    },
+    {
+      "epoch": 0.8388185654008439,
+      "grad_norm": 0.38364699482917786,
+      "learning_rate": 0.00019792875298112268,
+      "loss": 0.7724968791007996,
+      "step": 1988
+    },
+    {
+      "epoch": 0.8396624472573839,
+      "grad_norm": 0.439730703830719,
+      "learning_rate": 0.00019791901698851749,
+      "loss": 0.8542404174804688,
+      "step": 1990
+    },
+    {
+      "epoch": 0.8405063291139241,
+      "grad_norm": 0.4622722566127777,
+      "learning_rate": 0.00019790925840787563,
+      "loss": 0.8424296975135803,
+      "step": 1992
+    },
+    {
+      "epoch": 0.8413502109704641,
+      "grad_norm": 0.4491989314556122,
+      "learning_rate": 0.0001978994772414482,
+      "loss": 0.8400710225105286,
+      "step": 1994
+    },
+    {
+      "epoch": 0.8421940928270042,
+      "grad_norm": 0.372295081615448,
+      "learning_rate": 0.0001978896734914916,
+      "loss": 0.7265452146530151,
+      "step": 1996
+    },
+    {
+      "epoch": 0.8430379746835444,
+      "grad_norm": 0.4328666925430298,
+      "learning_rate": 0.00019787984716026732,
+      "loss": 0.8458228707313538,
+      "step": 1998
+    },
+    {
+      "epoch": 0.8438818565400844,
+      "grad_norm": 0.42246317863464355,
+      "learning_rate": 0.00019786999825004216,
+      "loss": 0.8517491817474365,
+      "step": 2000
+    },
+    {
+      "epoch": 0.8438818565400844,
+      "eval_loss": 0.8460908532142639,
+      "eval_runtime": 685.2518,
+      "eval_samples_per_second": 3.075,
+      "eval_steps_per_second": 3.075,
+      "step": 2000
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 14220,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 6,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 5,
+        "early_stopping_threshold": 0.001
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 0
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.195087286365225e+18,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-2000/training_args.bin b/sft_qwen_14B_v2/checkpoints/checkpoint-2000/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..0bd5f45757bdc6c7fc2d67718b18675b51983e0e
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-2000/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:afae4a8ce1391f149eb00b14eed8f891c715e892ea546bf754d22db2c2bc5969
+size 4792
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-2500/README.md b/sft_qwen_14B_v2/checkpoints/checkpoint-2500/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..134a3a8531b68adda3fed2a98db824f7c1b7dcfd
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-2500/README.md
@@ -0,0 +1,207 @@
+---
+base_model: Models/Qwen2.5-Coder-14B-CPT
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Models/Qwen2.5-Coder-14B-CPT
+- lora
+- transformers
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.0
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-2500/adapter_config.json b/sft_qwen_14B_v2/checkpoints/checkpoint-2500/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..d6316742bc71d30704e2ccdb9080ed0b8062e454
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-2500/adapter_config.json
@@ -0,0 +1,43 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Models/Qwen2.5-Coder-14B-CPT",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.0",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "v_proj",
+    "k_proj",
+    "q_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-2500/adapter_model.safetensors b/sft_qwen_14B_v2/checkpoints/checkpoint-2500/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..d5c0a484ce6e53cca8831b3a3632ae20a17d2ce2
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-2500/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a636e1d293ef747a9d4ba56f26013a4c53586a2f6ea38007f2f9ae4160279b10
+size 201378736
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-2500/optimizer.pt b/sft_qwen_14B_v2/checkpoints/checkpoint-2500/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..96428d75882c206e93c1d9ae176b46c9117a7bf1
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-2500/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c28f62d25813fcda2055109e78038dad11c58245a033d42ff1be2f0c8ace5d2
+size 402976786
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-2500/rng_state.pth b/sft_qwen_14B_v2/checkpoints/checkpoint-2500/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..9153a76813c7e8687571bc85a9719725a6bdc214
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-2500/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c52cacce7e167165387adaceb3af68d71fcfa2d3727df90524af91c93858d7af
+size 14244
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-2500/scheduler.pt b/sft_qwen_14B_v2/checkpoints/checkpoint-2500/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..7a44a2c2c1f7b9179509254c7a7a3efdcc43044c
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-2500/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:efc1d8a0a2b86aa25004510e00c98c486065b74b7c5a64dc37307cefa17ed157
+size 1064
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-2500/trainer_state.json b/sft_qwen_14B_v2/checkpoints/checkpoint-2500/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..a8ecc741c3bbfdeccd0859e71f511ef4b1401e7d
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-2500/trainer_state.json
@@ -0,0 +1,8993 @@
+{
+  "best_global_step": 2500,
+  "best_metric": 0.8249453902244568,
+  "best_model_checkpoint": "task2file/sft_qwen_14B_v2/checkpoints/checkpoint-2500",
+  "epoch": 1.0548523206751055,
+  "eval_steps": 100,
+  "global_step": 2500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0008438818565400844,
+      "grad_norm": 0.5386583805084229,
+      "learning_rate": 1.7574692442882248e-07,
+      "loss": 1.6941628456115723,
+      "step": 2
+    },
+    {
+      "epoch": 0.0016877637130801688,
+      "grad_norm": 0.5477277636528015,
+      "learning_rate": 5.272407732864675e-07,
+      "loss": 1.7132279872894287,
+      "step": 4
+    },
+    {
+      "epoch": 0.002531645569620253,
+      "grad_norm": 0.5390765070915222,
+      "learning_rate": 8.787346221441126e-07,
+      "loss": 1.641180396080017,
+      "step": 6
+    },
+    {
+      "epoch": 0.0033755274261603376,
+      "grad_norm": 0.5023683905601501,
+      "learning_rate": 1.2302284710017575e-06,
+      "loss": 1.5616240501403809,
+      "step": 8
+    },
+    {
+      "epoch": 0.004219409282700422,
+      "grad_norm": 0.4899154603481293,
+      "learning_rate": 1.5817223198594026e-06,
+      "loss": 1.572033405303955,
+      "step": 10
+    },
+    {
+      "epoch": 0.005063291139240506,
+      "grad_norm": 0.5239788293838501,
+      "learning_rate": 1.9332161687170474e-06,
+      "loss": 1.6242921352386475,
+      "step": 12
+    },
+    {
+      "epoch": 0.00590717299578059,
+      "grad_norm": 0.5172926783561707,
+      "learning_rate": 2.2847100175746925e-06,
+      "loss": 1.6800041198730469,
+      "step": 14
+    },
+    {
+      "epoch": 0.006751054852320675,
+      "grad_norm": 0.5539224743843079,
+      "learning_rate": 2.6362038664323376e-06,
+      "loss": 1.6450834274291992,
+      "step": 16
+    },
+    {
+      "epoch": 0.007594936708860759,
+      "grad_norm": 0.5255337953567505,
+      "learning_rate": 2.9876977152899827e-06,
+      "loss": 1.6673263311386108,
+      "step": 18
+    },
+    {
+      "epoch": 0.008438818565400843,
+      "grad_norm": 0.5074548721313477,
+      "learning_rate": 3.3391915641476277e-06,
+      "loss": 1.531802773475647,
+      "step": 20
+    },
+    {
+      "epoch": 0.009282700421940928,
+      "grad_norm": 0.4160279333591461,
+      "learning_rate": 3.6906854130052724e-06,
+      "loss": 1.599354863166809,
+      "step": 22
+    },
+    {
+      "epoch": 0.010126582278481013,
+      "grad_norm": 0.5716474652290344,
+      "learning_rate": 4.0421792618629174e-06,
+      "loss": 1.6700962781906128,
+      "step": 24
+    },
+    {
+      "epoch": 0.010970464135021098,
+      "grad_norm": 0.5148899555206299,
+      "learning_rate": 4.3936731107205625e-06,
+      "loss": 1.66217839717865,
+      "step": 26
+    },
+    {
+      "epoch": 0.01181434599156118,
+      "grad_norm": 0.575722336769104,
+      "learning_rate": 4.7451669595782076e-06,
+      "loss": 1.6692266464233398,
+      "step": 28
+    },
+    {
+      "epoch": 0.012658227848101266,
+      "grad_norm": 0.5345953106880188,
+      "learning_rate": 5.096660808435853e-06,
+      "loss": 1.5518689155578613,
+      "step": 30
+    },
+    {
+      "epoch": 0.01350210970464135,
+      "grad_norm": 0.4462043344974518,
+      "learning_rate": 5.448154657293498e-06,
+      "loss": 1.5930007696151733,
+      "step": 32
+    },
+    {
+      "epoch": 0.014345991561181435,
+      "grad_norm": 0.5119605660438538,
+      "learning_rate": 5.799648506151143e-06,
+      "loss": 1.6069684028625488,
+      "step": 34
+    },
+    {
+      "epoch": 0.015189873417721518,
+      "grad_norm": 0.5328608751296997,
+      "learning_rate": 6.151142355008788e-06,
+      "loss": 1.5838109254837036,
+      "step": 36
+    },
+    {
+      "epoch": 0.016033755274261603,
+      "grad_norm": 0.5065920352935791,
+      "learning_rate": 6.502636203866433e-06,
+      "loss": 1.608130931854248,
+      "step": 38
+    },
+    {
+      "epoch": 0.016877637130801686,
+      "grad_norm": 0.4479359984397888,
+      "learning_rate": 6.854130052724078e-06,
+      "loss": 1.5942182540893555,
+      "step": 40
+    },
+    {
+      "epoch": 0.017721518987341773,
+      "grad_norm": 0.42844903469085693,
+      "learning_rate": 7.205623901581722e-06,
+      "loss": 1.6441553831100464,
+      "step": 42
+    },
+    {
+      "epoch": 0.018565400843881856,
+      "grad_norm": 0.476630836725235,
+      "learning_rate": 7.557117750439367e-06,
+      "loss": 1.6068111658096313,
+      "step": 44
+    },
+    {
+      "epoch": 0.019409282700421943,
+      "grad_norm": 0.4532654881477356,
+      "learning_rate": 7.908611599297012e-06,
+      "loss": 1.6618021726608276,
+      "step": 46
+    },
+    {
+      "epoch": 0.020253164556962026,
+      "grad_norm": 0.3701118230819702,
+      "learning_rate": 8.260105448154657e-06,
+      "loss": 1.4730033874511719,
+      "step": 48
+    },
+    {
+      "epoch": 0.02109704641350211,
+      "grad_norm": 0.38471561670303345,
+      "learning_rate": 8.611599297012302e-06,
+      "loss": 1.4828267097473145,
+      "step": 50
+    },
+    {
+      "epoch": 0.021940928270042195,
+      "grad_norm": 0.3602336347103119,
+      "learning_rate": 8.963093145869948e-06,
+      "loss": 1.3877452611923218,
+      "step": 52
+    },
+    {
+      "epoch": 0.02278481012658228,
+      "grad_norm": 0.40318572521209717,
+      "learning_rate": 9.314586994727593e-06,
+      "loss": 1.49052894115448,
+      "step": 54
+    },
+    {
+      "epoch": 0.02362869198312236,
+      "grad_norm": 0.3223826587200165,
+      "learning_rate": 9.666080843585238e-06,
+      "loss": 1.4912524223327637,
+      "step": 56
+    },
+    {
+      "epoch": 0.024472573839662448,
+      "grad_norm": 0.3873065114021301,
+      "learning_rate": 1.0017574692442883e-05,
+      "loss": 1.526674509048462,
+      "step": 58
+    },
+    {
+      "epoch": 0.02531645569620253,
+      "grad_norm": 0.410159707069397,
+      "learning_rate": 1.0369068541300528e-05,
+      "loss": 1.4480271339416504,
+      "step": 60
+    },
+    {
+      "epoch": 0.026160337552742614,
+      "grad_norm": 0.3632003962993622,
+      "learning_rate": 1.0720562390158173e-05,
+      "loss": 1.4222990274429321,
+      "step": 62
+    },
+    {
+      "epoch": 0.0270042194092827,
+      "grad_norm": 0.33118435740470886,
+      "learning_rate": 1.1072056239015818e-05,
+      "loss": 1.387171745300293,
+      "step": 64
+    },
+    {
+      "epoch": 0.027848101265822784,
+      "grad_norm": 0.3301764726638794,
+      "learning_rate": 1.1423550087873463e-05,
+      "loss": 1.3523777723312378,
+      "step": 66
+    },
+    {
+      "epoch": 0.02869198312236287,
+      "grad_norm": 0.34342435002326965,
+      "learning_rate": 1.1775043936731108e-05,
+      "loss": 1.4515162706375122,
+      "step": 68
+    },
+    {
+      "epoch": 0.029535864978902954,
+      "grad_norm": 0.3243122100830078,
+      "learning_rate": 1.2126537785588753e-05,
+      "loss": 1.3509243726730347,
+      "step": 70
+    },
+    {
+      "epoch": 0.030379746835443037,
+      "grad_norm": 0.3450150787830353,
+      "learning_rate": 1.2478031634446398e-05,
+      "loss": 1.4936245679855347,
+      "step": 72
+    },
+    {
+      "epoch": 0.031223628691983123,
+      "grad_norm": 0.38912028074264526,
+      "learning_rate": 1.2829525483304042e-05,
+      "loss": 1.3419109582901,
+      "step": 74
+    },
+    {
+      "epoch": 0.032067510548523206,
+      "grad_norm": 0.3019310235977173,
+      "learning_rate": 1.3181019332161687e-05,
+      "loss": 1.4284154176712036,
+      "step": 76
+    },
+    {
+      "epoch": 0.03291139240506329,
+      "grad_norm": 0.37803682684898376,
+      "learning_rate": 1.3532513181019332e-05,
+      "loss": 1.4256561994552612,
+      "step": 78
+    },
+    {
+      "epoch": 0.03375527426160337,
+      "grad_norm": 0.34191736578941345,
+      "learning_rate": 1.3884007029876977e-05,
+      "loss": 1.3256909847259521,
+      "step": 80
+    },
+    {
+      "epoch": 0.03459915611814346,
+      "grad_norm": 0.35242700576782227,
+      "learning_rate": 1.4235500878734624e-05,
+      "loss": 1.2710685729980469,
+      "step": 82
+    },
+    {
+      "epoch": 0.035443037974683546,
+      "grad_norm": 0.38094228506088257,
+      "learning_rate": 1.4586994727592269e-05,
+      "loss": 1.253411889076233,
+      "step": 84
+    },
+    {
+      "epoch": 0.036286919831223625,
+      "grad_norm": 0.36837366223335266,
+      "learning_rate": 1.4938488576449914e-05,
+      "loss": 1.3064342737197876,
+      "step": 86
+    },
+    {
+      "epoch": 0.03713080168776371,
+      "grad_norm": 0.3443569242954254,
+      "learning_rate": 1.5289982425307557e-05,
+      "loss": 1.293562412261963,
+      "step": 88
+    },
+    {
+      "epoch": 0.0379746835443038,
+      "grad_norm": 0.3799338936805725,
+      "learning_rate": 1.5641476274165202e-05,
+      "loss": 1.3382648229599,
+      "step": 90
+    },
+    {
+      "epoch": 0.038818565400843885,
+      "grad_norm": 0.40501922369003296,
+      "learning_rate": 1.599297012302285e-05,
+      "loss": 1.3925724029541016,
+      "step": 92
+    },
+    {
+      "epoch": 0.039662447257383965,
+      "grad_norm": 0.4419630467891693,
+      "learning_rate": 1.6344463971880492e-05,
+      "loss": 1.357171893119812,
+      "step": 94
+    },
+    {
+      "epoch": 0.04050632911392405,
+      "grad_norm": 0.3619817793369293,
+      "learning_rate": 1.6695957820738137e-05,
+      "loss": 1.3029985427856445,
+      "step": 96
+    },
+    {
+      "epoch": 0.04135021097046414,
+      "grad_norm": 0.4851357340812683,
+      "learning_rate": 1.7047451669595782e-05,
+      "loss": 1.3498191833496094,
+      "step": 98
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "grad_norm": 0.418658584356308,
+      "learning_rate": 1.7398945518453427e-05,
+      "loss": 1.185287356376648,
+      "step": 100
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "eval_loss": 1.2979938983917236,
+      "eval_runtime": 682.1979,
+      "eval_samples_per_second": 3.089,
+      "eval_steps_per_second": 3.089,
+      "step": 100
+    },
+    {
+      "epoch": 0.043037974683544304,
+      "grad_norm": 0.4464418888092041,
+      "learning_rate": 1.7750439367311073e-05,
+      "loss": 1.2217272520065308,
+      "step": 102
+    },
+    {
+      "epoch": 0.04388185654008439,
+      "grad_norm": 0.4706237316131592,
+      "learning_rate": 1.8101933216168718e-05,
+      "loss": 1.2052050828933716,
+      "step": 104
+    },
+    {
+      "epoch": 0.04472573839662447,
+      "grad_norm": 0.46394404768943787,
+      "learning_rate": 1.8453427065026363e-05,
+      "loss": 1.221343994140625,
+      "step": 106
+    },
+    {
+      "epoch": 0.04556962025316456,
+      "grad_norm": 0.4726889431476593,
+      "learning_rate": 1.8804920913884008e-05,
+      "loss": 1.2387475967407227,
+      "step": 108
+    },
+    {
+      "epoch": 0.046413502109704644,
+      "grad_norm": 0.42130985856056213,
+      "learning_rate": 1.9156414762741653e-05,
+      "loss": 1.2851309776306152,
+      "step": 110
+    },
+    {
+      "epoch": 0.04725738396624472,
+      "grad_norm": 0.4504576623439789,
+      "learning_rate": 1.9507908611599298e-05,
+      "loss": 1.2753145694732666,
+      "step": 112
+    },
+    {
+      "epoch": 0.04810126582278481,
+      "grad_norm": 0.396085262298584,
+      "learning_rate": 1.9859402460456943e-05,
+      "loss": 1.2427717447280884,
+      "step": 114
+    },
+    {
+      "epoch": 0.048945147679324896,
+      "grad_norm": 0.5106491446495056,
+      "learning_rate": 2.0210896309314588e-05,
+      "loss": 1.2943825721740723,
+      "step": 116
+    },
+    {
+      "epoch": 0.049789029535864976,
+      "grad_norm": 0.42351317405700684,
+      "learning_rate": 2.0562390158172233e-05,
+      "loss": 1.263301134109497,
+      "step": 118
+    },
+    {
+      "epoch": 0.05063291139240506,
+      "grad_norm": 0.4403539299964905,
+      "learning_rate": 2.0913884007029878e-05,
+      "loss": 1.2647849321365356,
+      "step": 120
+    },
+    {
+      "epoch": 0.05147679324894515,
+      "grad_norm": 0.5260752439498901,
+      "learning_rate": 2.1265377855887523e-05,
+      "loss": 1.2351393699645996,
+      "step": 122
+    },
+    {
+      "epoch": 0.05232067510548523,
+      "grad_norm": 0.44978851079940796,
+      "learning_rate": 2.1616871704745168e-05,
+      "loss": 1.0384471416473389,
+      "step": 124
+    },
+    {
+      "epoch": 0.053164556962025315,
+      "grad_norm": 0.47732362151145935,
+      "learning_rate": 2.1968365553602813e-05,
+      "loss": 1.1518068313598633,
+      "step": 126
+    },
+    {
+      "epoch": 0.0540084388185654,
+      "grad_norm": 0.5473551750183105,
+      "learning_rate": 2.231985940246046e-05,
+      "loss": 1.2264912128448486,
+      "step": 128
+    },
+    {
+      "epoch": 0.05485232067510549,
+      "grad_norm": 0.4473855197429657,
+      "learning_rate": 2.2671353251318103e-05,
+      "loss": 1.1615246534347534,
+      "step": 130
+    },
+    {
+      "epoch": 0.05569620253164557,
+      "grad_norm": 0.5980377197265625,
+      "learning_rate": 2.302284710017575e-05,
+      "loss": 1.1334880590438843,
+      "step": 132
+    },
+    {
+      "epoch": 0.056540084388185655,
+      "grad_norm": 0.5987792015075684,
+      "learning_rate": 2.3374340949033394e-05,
+      "loss": 1.1546804904937744,
+      "step": 134
+    },
+    {
+      "epoch": 0.05738396624472574,
+      "grad_norm": 0.45355498790740967,
+      "learning_rate": 2.372583479789104e-05,
+      "loss": 1.194953441619873,
+      "step": 136
+    },
+    {
+      "epoch": 0.05822784810126582,
+      "grad_norm": 0.5373698472976685,
+      "learning_rate": 2.4077328646748684e-05,
+      "loss": 1.1067466735839844,
+      "step": 138
+    },
+    {
+      "epoch": 0.05907172995780591,
+      "grad_norm": 0.48734328150749207,
+      "learning_rate": 2.442882249560633e-05,
+      "loss": 1.188468098640442,
+      "step": 140
+    },
+    {
+      "epoch": 0.059915611814345994,
+      "grad_norm": 0.4692173898220062,
+      "learning_rate": 2.478031634446397e-05,
+      "loss": 1.1624362468719482,
+      "step": 142
+    },
+    {
+      "epoch": 0.060759493670886074,
+      "grad_norm": 0.532554030418396,
+      "learning_rate": 2.513181019332162e-05,
+      "loss": 1.0978907346725464,
+      "step": 144
+    },
+    {
+      "epoch": 0.06160337552742616,
+      "grad_norm": 0.5853802561759949,
+      "learning_rate": 2.5483304042179264e-05,
+      "loss": 1.2030781507492065,
+      "step": 146
+    },
+    {
+      "epoch": 0.06244725738396625,
+      "grad_norm": 0.5061611533164978,
+      "learning_rate": 2.583479789103691e-05,
+      "loss": 1.082366943359375,
+      "step": 148
+    },
+    {
+      "epoch": 0.06329113924050633,
+      "grad_norm": 0.49426141381263733,
+      "learning_rate": 2.6186291739894554e-05,
+      "loss": 1.10564386844635,
+      "step": 150
+    },
+    {
+      "epoch": 0.06413502109704641,
+      "grad_norm": 0.5846618413925171,
+      "learning_rate": 2.6537785588752196e-05,
+      "loss": 1.1992807388305664,
+      "step": 152
+    },
+    {
+      "epoch": 0.06497890295358649,
+      "grad_norm": 0.5517552495002747,
+      "learning_rate": 2.6889279437609844e-05,
+      "loss": 1.1757566928863525,
+      "step": 154
+    },
+    {
+      "epoch": 0.06582278481012659,
+      "grad_norm": 0.5667305588722229,
+      "learning_rate": 2.7240773286467486e-05,
+      "loss": 1.0548783540725708,
+      "step": 156
+    },
+    {
+      "epoch": 0.06666666666666667,
+      "grad_norm": 0.6760414242744446,
+      "learning_rate": 2.7592267135325134e-05,
+      "loss": 1.184364914894104,
+      "step": 158
+    },
+    {
+      "epoch": 0.06751054852320675,
+      "grad_norm": 0.5261430740356445,
+      "learning_rate": 2.7943760984182776e-05,
+      "loss": 1.1945042610168457,
+      "step": 160
+    },
+    {
+      "epoch": 0.06835443037974684,
+      "grad_norm": 0.6155015230178833,
+      "learning_rate": 2.8295254833040425e-05,
+      "loss": 1.2021973133087158,
+      "step": 162
+    },
+    {
+      "epoch": 0.06919831223628692,
+      "grad_norm": 0.6131619215011597,
+      "learning_rate": 2.8646748681898066e-05,
+      "loss": 1.144123911857605,
+      "step": 164
+    },
+    {
+      "epoch": 0.070042194092827,
+      "grad_norm": 0.5749185681343079,
+      "learning_rate": 2.8998242530755715e-05,
+      "loss": 1.1329256296157837,
+      "step": 166
+    },
+    {
+      "epoch": 0.07088607594936709,
+      "grad_norm": 0.5243118405342102,
+      "learning_rate": 2.9349736379613356e-05,
+      "loss": 1.0892387628555298,
+      "step": 168
+    },
+    {
+      "epoch": 0.07172995780590717,
+      "grad_norm": 0.7190104722976685,
+      "learning_rate": 2.9701230228471005e-05,
+      "loss": 1.163260817527771,
+      "step": 170
+    },
+    {
+      "epoch": 0.07257383966244725,
+      "grad_norm": 0.5486982464790344,
+      "learning_rate": 3.0052724077328647e-05,
+      "loss": 1.0880777835845947,
+      "step": 172
+    },
+    {
+      "epoch": 0.07341772151898734,
+      "grad_norm": 0.5020889043807983,
+      "learning_rate": 3.0404217926186295e-05,
+      "loss": 1.0433368682861328,
+      "step": 174
+    },
+    {
+      "epoch": 0.07426160337552742,
+      "grad_norm": 0.47329774498939514,
+      "learning_rate": 3.075571177504394e-05,
+      "loss": 1.0528991222381592,
+      "step": 176
+    },
+    {
+      "epoch": 0.0751054852320675,
+      "grad_norm": 0.6635547876358032,
+      "learning_rate": 3.110720562390158e-05,
+      "loss": 1.1627811193466187,
+      "step": 178
+    },
+    {
+      "epoch": 0.0759493670886076,
+      "grad_norm": 0.5624618530273438,
+      "learning_rate": 3.145869947275923e-05,
+      "loss": 1.084869384765625,
+      "step": 180
+    },
+    {
+      "epoch": 0.07679324894514768,
+      "grad_norm": 0.6029536724090576,
+      "learning_rate": 3.181019332161687e-05,
+      "loss": 1.2227671146392822,
+      "step": 182
+    },
+    {
+      "epoch": 0.07763713080168777,
+      "grad_norm": 0.930959939956665,
+      "learning_rate": 3.216168717047452e-05,
+      "loss": 1.0955452919006348,
+      "step": 184
+    },
+    {
+      "epoch": 0.07848101265822785,
+      "grad_norm": 0.5326952338218689,
+      "learning_rate": 3.251318101933216e-05,
+      "loss": 1.0640798807144165,
+      "step": 186
+    },
+    {
+      "epoch": 0.07932489451476793,
+      "grad_norm": 0.5484727621078491,
+      "learning_rate": 3.286467486818981e-05,
+      "loss": 1.0700589418411255,
+      "step": 188
+    },
+    {
+      "epoch": 0.08016877637130802,
+      "grad_norm": 0.605273425579071,
+      "learning_rate": 3.3216168717047456e-05,
+      "loss": 1.1593081951141357,
+      "step": 190
+    },
+    {
+      "epoch": 0.0810126582278481,
+      "grad_norm": 0.5704394578933716,
+      "learning_rate": 3.35676625659051e-05,
+      "loss": 1.1617076396942139,
+      "step": 192
+    },
+    {
+      "epoch": 0.08185654008438818,
+      "grad_norm": 0.5929452180862427,
+      "learning_rate": 3.3919156414762746e-05,
+      "loss": 1.1346839666366577,
+      "step": 194
+    },
+    {
+      "epoch": 0.08270042194092828,
+      "grad_norm": 0.5624077916145325,
+      "learning_rate": 3.427065026362039e-05,
+      "loss": 1.0934710502624512,
+      "step": 196
+    },
+    {
+      "epoch": 0.08354430379746836,
+      "grad_norm": 0.6717425584793091,
+      "learning_rate": 3.4622144112478036e-05,
+      "loss": 1.1810534000396729,
+      "step": 198
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "grad_norm": 0.5120199918746948,
+      "learning_rate": 3.4973637961335674e-05,
+      "loss": 1.1525514125823975,
+      "step": 200
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "eval_loss": 1.142486810684204,
+      "eval_runtime": 668.2356,
+      "eval_samples_per_second": 3.153,
+      "eval_steps_per_second": 3.153,
+      "step": 200
+    },
+    {
+      "epoch": 0.08523206751054853,
+      "grad_norm": 0.5144487023353577,
+      "learning_rate": 3.5325131810193326e-05,
+      "loss": 1.0243735313415527,
+      "step": 202
+    },
+    {
+      "epoch": 0.08607594936708861,
+      "grad_norm": 0.6325069069862366,
+      "learning_rate": 3.5676625659050964e-05,
+      "loss": 1.118743896484375,
+      "step": 204
+    },
+    {
+      "epoch": 0.08691983122362869,
+      "grad_norm": 0.5501633882522583,
+      "learning_rate": 3.6028119507908616e-05,
+      "loss": 1.0380504131317139,
+      "step": 206
+    },
+    {
+      "epoch": 0.08776371308016878,
+      "grad_norm": 0.6133899688720703,
+      "learning_rate": 3.6379613356766254e-05,
+      "loss": 0.9837555885314941,
+      "step": 208
+    },
+    {
+      "epoch": 0.08860759493670886,
+      "grad_norm": 0.5799810886383057,
+      "learning_rate": 3.6731107205623906e-05,
+      "loss": 1.090720295906067,
+      "step": 210
+    },
+    {
+      "epoch": 0.08945147679324894,
+      "grad_norm": 0.6039511561393738,
+      "learning_rate": 3.7082601054481544e-05,
+      "loss": 1.120232343673706,
+      "step": 212
+    },
+    {
+      "epoch": 0.09029535864978903,
+      "grad_norm": 0.5983024835586548,
+      "learning_rate": 3.7434094903339196e-05,
+      "loss": 1.096949815750122,
+      "step": 214
+    },
+    {
+      "epoch": 0.09113924050632911,
+      "grad_norm": 0.5641079545021057,
+      "learning_rate": 3.7785588752196835e-05,
+      "loss": 1.1226298809051514,
+      "step": 216
+    },
+    {
+      "epoch": 0.0919831223628692,
+      "grad_norm": 0.655717134475708,
+      "learning_rate": 3.8137082601054486e-05,
+      "loss": 1.1260643005371094,
+      "step": 218
+    },
+    {
+      "epoch": 0.09282700421940929,
+      "grad_norm": 0.6111898422241211,
+      "learning_rate": 3.848857644991213e-05,
+      "loss": 1.0777709484100342,
+      "step": 220
+    },
+    {
+      "epoch": 0.09367088607594937,
+      "grad_norm": 0.6821302771568298,
+      "learning_rate": 3.884007029876977e-05,
+      "loss": 1.10588800907135,
+      "step": 222
+    },
+    {
+      "epoch": 0.09451476793248945,
+      "grad_norm": 0.693175733089447,
+      "learning_rate": 3.919156414762742e-05,
+      "loss": 1.1498671770095825,
+      "step": 224
+    },
+    {
+      "epoch": 0.09535864978902954,
+      "grad_norm": 0.5288166403770447,
+      "learning_rate": 3.954305799648506e-05,
+      "loss": 1.0587562322616577,
+      "step": 226
+    },
+    {
+      "epoch": 0.09620253164556962,
+      "grad_norm": 0.6882867813110352,
+      "learning_rate": 3.989455184534271e-05,
+      "loss": 1.1107512712478638,
+      "step": 228
+    },
+    {
+      "epoch": 0.0970464135021097,
+      "grad_norm": 0.5834154486656189,
+      "learning_rate": 4.024604569420035e-05,
+      "loss": 1.020510196685791,
+      "step": 230
+    },
+    {
+      "epoch": 0.09789029535864979,
+      "grad_norm": 0.7157064080238342,
+      "learning_rate": 4.0597539543058e-05,
+      "loss": 1.0642449855804443,
+      "step": 232
+    },
+    {
+      "epoch": 0.09873417721518987,
+      "grad_norm": 0.6530708074569702,
+      "learning_rate": 4.094903339191564e-05,
+      "loss": 1.0359872579574585,
+      "step": 234
+    },
+    {
+      "epoch": 0.09957805907172995,
+      "grad_norm": 0.6329686045646667,
+      "learning_rate": 4.130052724077329e-05,
+      "loss": 1.050504446029663,
+      "step": 236
+    },
+    {
+      "epoch": 0.10042194092827005,
+      "grad_norm": 0.6597026586532593,
+      "learning_rate": 4.165202108963093e-05,
+      "loss": 1.2621175050735474,
+      "step": 238
+    },
+    {
+      "epoch": 0.10126582278481013,
+      "grad_norm": 0.6195225119590759,
+      "learning_rate": 4.200351493848858e-05,
+      "loss": 1.1218310594558716,
+      "step": 240
+    },
+    {
+      "epoch": 0.1021097046413502,
+      "grad_norm": 0.6764137744903564,
+      "learning_rate": 4.235500878734622e-05,
+      "loss": 1.1250728368759155,
+      "step": 242
+    },
+    {
+      "epoch": 0.1029535864978903,
+      "grad_norm": 0.552363395690918,
+      "learning_rate": 4.270650263620387e-05,
+      "loss": 1.028212308883667,
+      "step": 244
+    },
+    {
+      "epoch": 0.10379746835443038,
+      "grad_norm": 0.5620495676994324,
+      "learning_rate": 4.305799648506151e-05,
+      "loss": 1.0425450801849365,
+      "step": 246
+    },
+    {
+      "epoch": 0.10464135021097046,
+      "grad_norm": 0.6860032081604004,
+      "learning_rate": 4.3409490333919156e-05,
+      "loss": 1.144278883934021,
+      "step": 248
+    },
+    {
+      "epoch": 0.10548523206751055,
+      "grad_norm": 0.6033259034156799,
+      "learning_rate": 4.37609841827768e-05,
+      "loss": 1.1223982572555542,
+      "step": 250
+    },
+    {
+      "epoch": 0.10632911392405063,
+      "grad_norm": 0.6292146444320679,
+      "learning_rate": 4.4112478031634446e-05,
+      "loss": 1.1609960794448853,
+      "step": 252
+    },
+    {
+      "epoch": 0.10717299578059072,
+      "grad_norm": 0.7982883453369141,
+      "learning_rate": 4.44639718804921e-05,
+      "loss": 1.063547968864441,
+      "step": 254
+    },
+    {
+      "epoch": 0.1080168776371308,
+      "grad_norm": 0.7719110250473022,
+      "learning_rate": 4.4815465729349736e-05,
+      "loss": 1.0719804763793945,
+      "step": 256
+    },
+    {
+      "epoch": 0.10886075949367088,
+      "grad_norm": 0.6101011633872986,
+      "learning_rate": 4.516695957820739e-05,
+      "loss": 1.0778400897979736,
+      "step": 258
+    },
+    {
+      "epoch": 0.10970464135021098,
+      "grad_norm": 0.7300994396209717,
+      "learning_rate": 4.5518453427065026e-05,
+      "loss": 1.2129558324813843,
+      "step": 260
+    },
+    {
+      "epoch": 0.11054852320675106,
+      "grad_norm": 0.8348747491836548,
+      "learning_rate": 4.586994727592268e-05,
+      "loss": 1.221714735031128,
+      "step": 262
+    },
+    {
+      "epoch": 0.11139240506329114,
+      "grad_norm": 0.5445612072944641,
+      "learning_rate": 4.6221441124780316e-05,
+      "loss": 1.0187978744506836,
+      "step": 264
+    },
+    {
+      "epoch": 0.11223628691983123,
+      "grad_norm": 0.6230319738388062,
+      "learning_rate": 4.657293497363797e-05,
+      "loss": 1.096561312675476,
+      "step": 266
+    },
+    {
+      "epoch": 0.11308016877637131,
+      "grad_norm": 0.6231237649917603,
+      "learning_rate": 4.6924428822495606e-05,
+      "loss": 1.089842438697815,
+      "step": 268
+    },
+    {
+      "epoch": 0.11392405063291139,
+      "grad_norm": 0.7178627252578735,
+      "learning_rate": 4.727592267135325e-05,
+      "loss": 1.0696645975112915,
+      "step": 270
+    },
+    {
+      "epoch": 0.11476793248945148,
+      "grad_norm": 0.6895854473114014,
+      "learning_rate": 4.7627416520210896e-05,
+      "loss": 1.0511361360549927,
+      "step": 272
+    },
+    {
+      "epoch": 0.11561181434599156,
+      "grad_norm": 0.6046878695487976,
+      "learning_rate": 4.797891036906854e-05,
+      "loss": 1.1373958587646484,
+      "step": 274
+    },
+    {
+      "epoch": 0.11645569620253164,
+      "grad_norm": 0.6524552702903748,
+      "learning_rate": 4.833040421792619e-05,
+      "loss": 1.0734186172485352,
+      "step": 276
+    },
+    {
+      "epoch": 0.11729957805907174,
+      "grad_norm": 0.6331019997596741,
+      "learning_rate": 4.868189806678383e-05,
+      "loss": 1.123913049697876,
+      "step": 278
+    },
+    {
+      "epoch": 0.11814345991561181,
+      "grad_norm": 0.5919018983840942,
+      "learning_rate": 4.903339191564148e-05,
+      "loss": 1.0635710954666138,
+      "step": 280
+    },
+    {
+      "epoch": 0.1189873417721519,
+      "grad_norm": 0.6067633032798767,
+      "learning_rate": 4.938488576449912e-05,
+      "loss": 1.0429247617721558,
+      "step": 282
+    },
+    {
+      "epoch": 0.11983122362869199,
+      "grad_norm": 0.6583750247955322,
+      "learning_rate": 4.9736379613356774e-05,
+      "loss": 1.1397464275360107,
+      "step": 284
+    },
+    {
+      "epoch": 0.12067510548523207,
+      "grad_norm": 0.6200069785118103,
+      "learning_rate": 5.008787346221442e-05,
+      "loss": 1.0590803623199463,
+      "step": 286
+    },
+    {
+      "epoch": 0.12151898734177215,
+      "grad_norm": 0.6798665523529053,
+      "learning_rate": 5.0439367311072064e-05,
+      "loss": 1.1318789720535278,
+      "step": 288
+    },
+    {
+      "epoch": 0.12236286919831224,
+      "grad_norm": 0.7508794069290161,
+      "learning_rate": 5.07908611599297e-05,
+      "loss": 1.0934956073760986,
+      "step": 290
+    },
+    {
+      "epoch": 0.12320675105485232,
+      "grad_norm": 0.6901452541351318,
+      "learning_rate": 5.114235500878735e-05,
+      "loss": 1.163407802581787,
+      "step": 292
+    },
+    {
+      "epoch": 0.1240506329113924,
+      "grad_norm": 0.6423285603523254,
+      "learning_rate": 5.1493848857644985e-05,
+      "loss": 1.09059476852417,
+      "step": 294
+    },
+    {
+      "epoch": 0.1248945147679325,
+      "grad_norm": 0.6839275360107422,
+      "learning_rate": 5.1845342706502644e-05,
+      "loss": 1.0690211057662964,
+      "step": 296
+    },
+    {
+      "epoch": 0.1257383966244726,
+      "grad_norm": 0.6350128054618835,
+      "learning_rate": 5.219683655536028e-05,
+      "loss": 0.982322096824646,
+      "step": 298
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "grad_norm": 0.7136530876159668,
+      "learning_rate": 5.254833040421793e-05,
+      "loss": 1.1132930517196655,
+      "step": 300
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "eval_loss": 1.0952109098434448,
+      "eval_runtime": 677.0652,
+      "eval_samples_per_second": 3.112,
+      "eval_steps_per_second": 3.112,
+      "step": 300
+    },
+    {
+      "epoch": 0.12742616033755275,
+      "grad_norm": 0.7339721322059631,
+      "learning_rate": 5.289982425307557e-05,
+      "loss": 0.973595917224884,
+      "step": 302
+    },
+    {
+      "epoch": 0.12827004219409283,
+      "grad_norm": 0.5941481590270996,
+      "learning_rate": 5.3251318101933224e-05,
+      "loss": 0.9819849729537964,
+      "step": 304
+    },
+    {
+      "epoch": 0.1291139240506329,
+      "grad_norm": 0.7153938412666321,
+      "learning_rate": 5.360281195079086e-05,
+      "loss": 1.0315470695495605,
+      "step": 306
+    },
+    {
+      "epoch": 0.12995780590717299,
+      "grad_norm": 0.5167180299758911,
+      "learning_rate": 5.395430579964851e-05,
+      "loss": 0.9492001533508301,
+      "step": 308
+    },
+    {
+      "epoch": 0.1308016877637131,
+      "grad_norm": 0.6055944561958313,
+      "learning_rate": 5.430579964850615e-05,
+      "loss": 1.0156209468841553,
+      "step": 310
+    },
+    {
+      "epoch": 0.13164556962025317,
+      "grad_norm": 0.7662386298179626,
+      "learning_rate": 5.4657293497363805e-05,
+      "loss": 1.1791651248931885,
+      "step": 312
+    },
+    {
+      "epoch": 0.13248945147679325,
+      "grad_norm": 0.6065546274185181,
+      "learning_rate": 5.500878734622145e-05,
+      "loss": 1.0009297132492065,
+      "step": 314
+    },
+    {
+      "epoch": 0.13333333333333333,
+      "grad_norm": 0.604225754737854,
+      "learning_rate": 5.536028119507909e-05,
+      "loss": 1.0208244323730469,
+      "step": 316
+    },
+    {
+      "epoch": 0.1341772151898734,
+      "grad_norm": 0.6186763048171997,
+      "learning_rate": 5.571177504393673e-05,
+      "loss": 0.9968416690826416,
+      "step": 318
+    },
+    {
+      "epoch": 0.1350210970464135,
+      "grad_norm": 0.7100363969802856,
+      "learning_rate": 5.606326889279437e-05,
+      "loss": 0.9540256857872009,
+      "step": 320
+    },
+    {
+      "epoch": 0.1358649789029536,
+      "grad_norm": 0.6979711055755615,
+      "learning_rate": 5.641476274165203e-05,
+      "loss": 1.0631953477859497,
+      "step": 322
+    },
+    {
+      "epoch": 0.13670886075949368,
+      "grad_norm": 0.6237109303474426,
+      "learning_rate": 5.676625659050967e-05,
+      "loss": 1.0170501470565796,
+      "step": 324
+    },
+    {
+      "epoch": 0.13755274261603376,
+      "grad_norm": 0.6525548696517944,
+      "learning_rate": 5.711775043936731e-05,
+      "loss": 1.0715603828430176,
+      "step": 326
+    },
+    {
+      "epoch": 0.13839662447257384,
+      "grad_norm": 0.6869221329689026,
+      "learning_rate": 5.746924428822495e-05,
+      "loss": 1.0111541748046875,
+      "step": 328
+    },
+    {
+      "epoch": 0.13924050632911392,
+      "grad_norm": 0.553188145160675,
+      "learning_rate": 5.782073813708261e-05,
+      "loss": 1.0311682224273682,
+      "step": 330
+    },
+    {
+      "epoch": 0.140084388185654,
+      "grad_norm": 0.6760852932929993,
+      "learning_rate": 5.817223198594025e-05,
+      "loss": 1.0213634967803955,
+      "step": 332
+    },
+    {
+      "epoch": 0.1409282700421941,
+      "grad_norm": 0.5907419919967651,
+      "learning_rate": 5.8523725834797894e-05,
+      "loss": 0.9748594164848328,
+      "step": 334
+    },
+    {
+      "epoch": 0.14177215189873418,
+      "grad_norm": 0.7044920921325684,
+      "learning_rate": 5.887521968365554e-05,
+      "loss": 1.05863356590271,
+      "step": 336
+    },
+    {
+      "epoch": 0.14261603375527426,
+      "grad_norm": 0.679073691368103,
+      "learning_rate": 5.922671353251318e-05,
+      "loss": 1.1341127157211304,
+      "step": 338
+    },
+    {
+      "epoch": 0.14345991561181434,
+      "grad_norm": 0.7676237225532532,
+      "learning_rate": 5.957820738137083e-05,
+      "loss": 0.9540836215019226,
+      "step": 340
+    },
+    {
+      "epoch": 0.14430379746835442,
+      "grad_norm": 0.6313899755477905,
+      "learning_rate": 5.9929701230228474e-05,
+      "loss": 1.0585911273956299,
+      "step": 342
+    },
+    {
+      "epoch": 0.1451476793248945,
+      "grad_norm": 0.7123099565505981,
+      "learning_rate": 6.028119507908612e-05,
+      "loss": 1.0760118961334229,
+      "step": 344
+    },
+    {
+      "epoch": 0.1459915611814346,
+      "grad_norm": 0.585935652256012,
+      "learning_rate": 6.063268892794376e-05,
+      "loss": 1.036866307258606,
+      "step": 346
+    },
+    {
+      "epoch": 0.1468354430379747,
+      "grad_norm": 0.5643263459205627,
+      "learning_rate": 6.0984182776801416e-05,
+      "loss": 1.0242938995361328,
+      "step": 348
+    },
+    {
+      "epoch": 0.14767932489451477,
+      "grad_norm": 0.626761794090271,
+      "learning_rate": 6.133567662565906e-05,
+      "loss": 1.0497376918792725,
+      "step": 350
+    },
+    {
+      "epoch": 0.14852320675105485,
+      "grad_norm": 0.5106956958770752,
+      "learning_rate": 6.16871704745167e-05,
+      "loss": 0.9811885356903076,
+      "step": 352
+    },
+    {
+      "epoch": 0.14936708860759493,
+      "grad_norm": 0.6948089003562927,
+      "learning_rate": 6.203866432337434e-05,
+      "loss": 1.0715330839157104,
+      "step": 354
+    },
+    {
+      "epoch": 0.150210970464135,
+      "grad_norm": 0.699713945388794,
+      "learning_rate": 6.239015817223199e-05,
+      "loss": 1.0405226945877075,
+      "step": 356
+    },
+    {
+      "epoch": 0.15105485232067511,
+      "grad_norm": 0.6437667012214661,
+      "learning_rate": 6.274165202108964e-05,
+      "loss": 1.0490930080413818,
+      "step": 358
+    },
+    {
+      "epoch": 0.1518987341772152,
+      "grad_norm": 0.6952699422836304,
+      "learning_rate": 6.309314586994728e-05,
+      "loss": 0.9267548322677612,
+      "step": 360
+    },
+    {
+      "epoch": 0.15274261603375527,
+      "grad_norm": 0.6713186502456665,
+      "learning_rate": 6.344463971880492e-05,
+      "loss": 1.0427420139312744,
+      "step": 362
+    },
+    {
+      "epoch": 0.15358649789029535,
+      "grad_norm": 0.6750379800796509,
+      "learning_rate": 6.379613356766257e-05,
+      "loss": 1.048950433731079,
+      "step": 364
+    },
+    {
+      "epoch": 0.15443037974683543,
+      "grad_norm": 0.6053379774093628,
+      "learning_rate": 6.414762741652022e-05,
+      "loss": 1.0156004428863525,
+      "step": 366
+    },
+    {
+      "epoch": 0.15527426160337554,
+      "grad_norm": 0.8063633441925049,
+      "learning_rate": 6.449912126537786e-05,
+      "loss": 1.0020819902420044,
+      "step": 368
+    },
+    {
+      "epoch": 0.15611814345991562,
+      "grad_norm": 0.8027494549751282,
+      "learning_rate": 6.48506151142355e-05,
+      "loss": 1.055633783340454,
+      "step": 370
+    },
+    {
+      "epoch": 0.1569620253164557,
+      "grad_norm": 0.6580121517181396,
+      "learning_rate": 6.520210896309315e-05,
+      "loss": 1.0149940252304077,
+      "step": 372
+    },
+    {
+      "epoch": 0.15780590717299578,
+      "grad_norm": 0.6561233997344971,
+      "learning_rate": 6.55536028119508e-05,
+      "loss": 0.9769611954689026,
+      "step": 374
+    },
+    {
+      "epoch": 0.15864978902953586,
+      "grad_norm": 0.6444346308708191,
+      "learning_rate": 6.590509666080844e-05,
+      "loss": 0.9099349975585938,
+      "step": 376
+    },
+    {
+      "epoch": 0.15949367088607594,
+      "grad_norm": 0.5879359245300293,
+      "learning_rate": 6.625659050966608e-05,
+      "loss": 1.0797548294067383,
+      "step": 378
+    },
+    {
+      "epoch": 0.16033755274261605,
+      "grad_norm": 0.6994144916534424,
+      "learning_rate": 6.660808435852373e-05,
+      "loss": 1.0336791276931763,
+      "step": 380
+    },
+    {
+      "epoch": 0.16118143459915613,
+      "grad_norm": 0.6128669381141663,
+      "learning_rate": 6.695957820738138e-05,
+      "loss": 1.018118143081665,
+      "step": 382
+    },
+    {
+      "epoch": 0.1620253164556962,
+      "grad_norm": 1.0237540006637573,
+      "learning_rate": 6.731107205623902e-05,
+      "loss": 1.1405497789382935,
+      "step": 384
+    },
+    {
+      "epoch": 0.16286919831223629,
+      "grad_norm": 0.6091578006744385,
+      "learning_rate": 6.766256590509666e-05,
+      "loss": 1.0314189195632935,
+      "step": 386
+    },
+    {
+      "epoch": 0.16371308016877636,
+      "grad_norm": 0.5916037559509277,
+      "learning_rate": 6.801405975395431e-05,
+      "loss": 0.9564052820205688,
+      "step": 388
+    },
+    {
+      "epoch": 0.16455696202531644,
+      "grad_norm": 0.771653950214386,
+      "learning_rate": 6.836555360281195e-05,
+      "loss": 1.0023859739303589,
+      "step": 390
+    },
+    {
+      "epoch": 0.16540084388185655,
+      "grad_norm": 0.654658317565918,
+      "learning_rate": 6.87170474516696e-05,
+      "loss": 1.07024085521698,
+      "step": 392
+    },
+    {
+      "epoch": 0.16624472573839663,
+      "grad_norm": 0.6611968874931335,
+      "learning_rate": 6.906854130052724e-05,
+      "loss": 1.0552500486373901,
+      "step": 394
+    },
+    {
+      "epoch": 0.1670886075949367,
+      "grad_norm": 0.6955893039703369,
+      "learning_rate": 6.942003514938489e-05,
+      "loss": 1.0562875270843506,
+      "step": 396
+    },
+    {
+      "epoch": 0.1679324894514768,
+      "grad_norm": 0.6666058301925659,
+      "learning_rate": 6.977152899824253e-05,
+      "loss": 0.9850592017173767,
+      "step": 398
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "grad_norm": 0.6131711006164551,
+      "learning_rate": 7.012302284710018e-05,
+      "loss": 1.0077755451202393,
+      "step": 400
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "eval_loss": 1.0625108480453491,
+      "eval_runtime": 691.0068,
+      "eval_samples_per_second": 3.049,
+      "eval_steps_per_second": 3.049,
+      "step": 400
+    },
+    {
+      "epoch": 0.16962025316455695,
+      "grad_norm": 0.6286499500274658,
+      "learning_rate": 7.047451669595783e-05,
+      "loss": 1.1012427806854248,
+      "step": 402
+    },
+    {
+      "epoch": 0.17046413502109706,
+      "grad_norm": 0.6639351844787598,
+      "learning_rate": 7.082601054481547e-05,
+      "loss": 1.0379719734191895,
+      "step": 404
+    },
+    {
+      "epoch": 0.17130801687763714,
+      "grad_norm": 0.750401496887207,
+      "learning_rate": 7.117750439367311e-05,
+      "loss": 1.031856656074524,
+      "step": 406
+    },
+    {
+      "epoch": 0.17215189873417722,
+      "grad_norm": 0.8084847331047058,
+      "learning_rate": 7.152899824253075e-05,
+      "loss": 1.0493193864822388,
+      "step": 408
+    },
+    {
+      "epoch": 0.1729957805907173,
+      "grad_norm": 0.7448462247848511,
+      "learning_rate": 7.188049209138841e-05,
+      "loss": 1.1012418270111084,
+      "step": 410
+    },
+    {
+      "epoch": 0.17383966244725738,
+      "grad_norm": 0.5841867923736572,
+      "learning_rate": 7.223198594024605e-05,
+      "loss": 0.9926692247390747,
+      "step": 412
+    },
+    {
+      "epoch": 0.17468354430379746,
+      "grad_norm": 0.7125606536865234,
+      "learning_rate": 7.258347978910369e-05,
+      "loss": 1.0588877201080322,
+      "step": 414
+    },
+    {
+      "epoch": 0.17552742616033756,
+      "grad_norm": 0.5750942230224609,
+      "learning_rate": 7.293497363796134e-05,
+      "loss": 1.038270354270935,
+      "step": 416
+    },
+    {
+      "epoch": 0.17637130801687764,
+      "grad_norm": 0.565444827079773,
+      "learning_rate": 7.328646748681899e-05,
+      "loss": 0.9843021035194397,
+      "step": 418
+    },
+    {
+      "epoch": 0.17721518987341772,
+      "grad_norm": 0.5825693011283875,
+      "learning_rate": 7.363796133567663e-05,
+      "loss": 1.0731632709503174,
+      "step": 420
+    },
+    {
+      "epoch": 0.1780590717299578,
+      "grad_norm": 0.6267391443252563,
+      "learning_rate": 7.398945518453427e-05,
+      "loss": 1.0061273574829102,
+      "step": 422
+    },
+    {
+      "epoch": 0.17890295358649788,
+      "grad_norm": 0.6621372103691101,
+      "learning_rate": 7.434094903339192e-05,
+      "loss": 1.0461612939834595,
+      "step": 424
+    },
+    {
+      "epoch": 0.17974683544303796,
+      "grad_norm": 0.6635435223579407,
+      "learning_rate": 7.469244288224957e-05,
+      "loss": 0.9789207577705383,
+      "step": 426
+    },
+    {
+      "epoch": 0.18059071729957807,
+      "grad_norm": 0.6342346668243408,
+      "learning_rate": 7.504393673110721e-05,
+      "loss": 1.0527069568634033,
+      "step": 428
+    },
+    {
+      "epoch": 0.18143459915611815,
+      "grad_norm": 0.6762149930000305,
+      "learning_rate": 7.539543057996485e-05,
+      "loss": 0.9708702564239502,
+      "step": 430
+    },
+    {
+      "epoch": 0.18227848101265823,
+      "grad_norm": 0.7073282599449158,
+      "learning_rate": 7.57469244288225e-05,
+      "loss": 1.0509834289550781,
+      "step": 432
+    },
+    {
+      "epoch": 0.1831223628691983,
+      "grad_norm": 0.6917856931686401,
+      "learning_rate": 7.609841827768014e-05,
+      "loss": 1.0128819942474365,
+      "step": 434
+    },
+    {
+      "epoch": 0.1839662447257384,
+      "grad_norm": 0.5574942231178284,
+      "learning_rate": 7.644991212653779e-05,
+      "loss": 0.989395797252655,
+      "step": 436
+    },
+    {
+      "epoch": 0.1848101265822785,
+      "grad_norm": 0.640765905380249,
+      "learning_rate": 7.680140597539543e-05,
+      "loss": 0.9846042990684509,
+      "step": 438
+    },
+    {
+      "epoch": 0.18565400843881857,
+      "grad_norm": 0.6699127554893494,
+      "learning_rate": 7.715289982425308e-05,
+      "loss": 1.0344442129135132,
+      "step": 440
+    },
+    {
+      "epoch": 0.18649789029535865,
+      "grad_norm": 0.6164930462837219,
+      "learning_rate": 7.750439367311072e-05,
+      "loss": 1.0179373025894165,
+      "step": 442
+    },
+    {
+      "epoch": 0.18734177215189873,
+      "grad_norm": 0.6880720853805542,
+      "learning_rate": 7.785588752196837e-05,
+      "loss": 1.0518895387649536,
+      "step": 444
+    },
+    {
+      "epoch": 0.1881856540084388,
+      "grad_norm": 0.6501413583755493,
+      "learning_rate": 7.820738137082601e-05,
+      "loss": 1.0442606210708618,
+      "step": 446
+    },
+    {
+      "epoch": 0.1890295358649789,
+      "grad_norm": 0.6076085567474365,
+      "learning_rate": 7.855887521968366e-05,
+      "loss": 0.9828442335128784,
+      "step": 448
+    },
+    {
+      "epoch": 0.189873417721519,
+      "grad_norm": 0.6418202519416809,
+      "learning_rate": 7.89103690685413e-05,
+      "loss": 1.0573710203170776,
+      "step": 450
+    },
+    {
+      "epoch": 0.19071729957805908,
+      "grad_norm": 0.7055076360702515,
+      "learning_rate": 7.926186291739895e-05,
+      "loss": 1.0216103792190552,
+      "step": 452
+    },
+    {
+      "epoch": 0.19156118143459916,
+      "grad_norm": 0.5668330192565918,
+      "learning_rate": 7.961335676625659e-05,
+      "loss": 0.9837722778320312,
+      "step": 454
+    },
+    {
+      "epoch": 0.19240506329113924,
+      "grad_norm": 0.6419380307197571,
+      "learning_rate": 7.996485061511424e-05,
+      "loss": 1.0003894567489624,
+      "step": 456
+    },
+    {
+      "epoch": 0.19324894514767932,
+      "grad_norm": 0.5949198007583618,
+      "learning_rate": 8.031634446397188e-05,
+      "loss": 1.0609031915664673,
+      "step": 458
+    },
+    {
+      "epoch": 0.1940928270042194,
+      "grad_norm": 0.7032039761543274,
+      "learning_rate": 8.066783831282952e-05,
+      "loss": 1.0543403625488281,
+      "step": 460
+    },
+    {
+      "epoch": 0.1949367088607595,
+      "grad_norm": 0.5775868892669678,
+      "learning_rate": 8.101933216168718e-05,
+      "loss": 0.9819303154945374,
+      "step": 462
+    },
+    {
+      "epoch": 0.19578059071729959,
+      "grad_norm": 0.9301062226295471,
+      "learning_rate": 8.137082601054482e-05,
+      "loss": 1.0542067289352417,
+      "step": 464
+    },
+    {
+      "epoch": 0.19662447257383966,
+      "grad_norm": 0.6193217039108276,
+      "learning_rate": 8.172231985940246e-05,
+      "loss": 0.9966341257095337,
+      "step": 466
+    },
+    {
+      "epoch": 0.19746835443037974,
+      "grad_norm": 0.6286146640777588,
+      "learning_rate": 8.20738137082601e-05,
+      "loss": 1.0474121570587158,
+      "step": 468
+    },
+    {
+      "epoch": 0.19831223628691982,
+      "grad_norm": 0.7418972253799438,
+      "learning_rate": 8.242530755711776e-05,
+      "loss": 0.9549239277839661,
+      "step": 470
+    },
+    {
+      "epoch": 0.1991561181434599,
+      "grad_norm": 0.6122808456420898,
+      "learning_rate": 8.27768014059754e-05,
+      "loss": 1.0191338062286377,
+      "step": 472
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.6375362277030945,
+      "learning_rate": 8.312829525483304e-05,
+      "loss": 1.0987539291381836,
+      "step": 474
+    },
+    {
+      "epoch": 0.2008438818565401,
+      "grad_norm": 0.6459513306617737,
+      "learning_rate": 8.347978910369068e-05,
+      "loss": 1.0369136333465576,
+      "step": 476
+    },
+    {
+      "epoch": 0.20168776371308017,
+      "grad_norm": 0.7029640674591064,
+      "learning_rate": 8.383128295254833e-05,
+      "loss": 1.0582096576690674,
+      "step": 478
+    },
+    {
+      "epoch": 0.20253164556962025,
+      "grad_norm": 0.6345387697219849,
+      "learning_rate": 8.418277680140598e-05,
+      "loss": 1.022916316986084,
+      "step": 480
+    },
+    {
+      "epoch": 0.20337552742616033,
+      "grad_norm": 0.5764590501785278,
+      "learning_rate": 8.453427065026362e-05,
+      "loss": 0.973024308681488,
+      "step": 482
+    },
+    {
+      "epoch": 0.2042194092827004,
+      "grad_norm": 0.5884482860565186,
+      "learning_rate": 8.488576449912127e-05,
+      "loss": 1.0292812585830688,
+      "step": 484
+    },
+    {
+      "epoch": 0.20506329113924052,
+      "grad_norm": 0.616357147693634,
+      "learning_rate": 8.523725834797891e-05,
+      "loss": 1.0083447694778442,
+      "step": 486
+    },
+    {
+      "epoch": 0.2059071729957806,
+      "grad_norm": 0.7671196460723877,
+      "learning_rate": 8.558875219683656e-05,
+      "loss": 0.9936985373497009,
+      "step": 488
+    },
+    {
+      "epoch": 0.20675105485232068,
+      "grad_norm": 0.6197299957275391,
+      "learning_rate": 8.59402460456942e-05,
+      "loss": 1.051513910293579,
+      "step": 490
+    },
+    {
+      "epoch": 0.20759493670886076,
+      "grad_norm": 0.6912890672683716,
+      "learning_rate": 8.629173989455185e-05,
+      "loss": 0.9474978446960449,
+      "step": 492
+    },
+    {
+      "epoch": 0.20843881856540084,
+      "grad_norm": 0.6941592693328857,
+      "learning_rate": 8.664323374340949e-05,
+      "loss": 1.0671660900115967,
+      "step": 494
+    },
+    {
+      "epoch": 0.20928270042194091,
+      "grad_norm": 0.5889528393745422,
+      "learning_rate": 8.699472759226714e-05,
+      "loss": 1.0020159482955933,
+      "step": 496
+    },
+    {
+      "epoch": 0.21012658227848102,
+      "grad_norm": 0.6478549838066101,
+      "learning_rate": 8.734622144112478e-05,
+      "loss": 1.0165860652923584,
+      "step": 498
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "grad_norm": 0.6018255949020386,
+      "learning_rate": 8.769771528998243e-05,
+      "loss": 0.8798263072967529,
+      "step": 500
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "eval_loss": 1.042096495628357,
+      "eval_runtime": 692.4361,
+      "eval_samples_per_second": 3.043,
+      "eval_steps_per_second": 3.043,
+      "step": 500
+    },
+    {
+      "epoch": 0.21181434599156118,
+      "grad_norm": 0.578990638256073,
+      "learning_rate": 8.804920913884007e-05,
+      "loss": 1.092096209526062,
+      "step": 502
+    },
+    {
+      "epoch": 0.21265822784810126,
+      "grad_norm": 0.6597883701324463,
+      "learning_rate": 8.840070298769771e-05,
+      "loss": 1.0413451194763184,
+      "step": 504
+    },
+    {
+      "epoch": 0.21350210970464134,
+      "grad_norm": 0.6660305261611938,
+      "learning_rate": 8.875219683655536e-05,
+      "loss": 1.0073142051696777,
+      "step": 506
+    },
+    {
+      "epoch": 0.21434599156118145,
+      "grad_norm": 0.6283115148544312,
+      "learning_rate": 8.910369068541301e-05,
+      "loss": 1.0319768190383911,
+      "step": 508
+    },
+    {
+      "epoch": 0.21518987341772153,
+      "grad_norm": 0.6257343292236328,
+      "learning_rate": 8.945518453427065e-05,
+      "loss": 1.0046353340148926,
+      "step": 510
+    },
+    {
+      "epoch": 0.2160337552742616,
+      "grad_norm": 0.5530875325202942,
+      "learning_rate": 8.980667838312829e-05,
+      "loss": 0.9169099926948547,
+      "step": 512
+    },
+    {
+      "epoch": 0.2168776371308017,
+      "grad_norm": 0.5369633436203003,
+      "learning_rate": 9.015817223198594e-05,
+      "loss": 0.9081505537033081,
+      "step": 514
+    },
+    {
+      "epoch": 0.21772151898734177,
+      "grad_norm": 0.6618232131004333,
+      "learning_rate": 9.05096660808436e-05,
+      "loss": 0.9165045022964478,
+      "step": 516
+    },
+    {
+      "epoch": 0.21856540084388185,
+      "grad_norm": 0.600666344165802,
+      "learning_rate": 9.086115992970123e-05,
+      "loss": 0.91348797082901,
+      "step": 518
+    },
+    {
+      "epoch": 0.21940928270042195,
+      "grad_norm": 0.5919831991195679,
+      "learning_rate": 9.121265377855887e-05,
+      "loss": 1.006508469581604,
+      "step": 520
+    },
+    {
+      "epoch": 0.22025316455696203,
+      "grad_norm": 0.688058614730835,
+      "learning_rate": 9.156414762741654e-05,
+      "loss": 1.0013236999511719,
+      "step": 522
+    },
+    {
+      "epoch": 0.2210970464135021,
+      "grad_norm": 0.6721227765083313,
+      "learning_rate": 9.191564147627418e-05,
+      "loss": 1.0909923315048218,
+      "step": 524
+    },
+    {
+      "epoch": 0.2219409282700422,
+      "grad_norm": 0.5987313389778137,
+      "learning_rate": 9.226713532513181e-05,
+      "loss": 1.0117096900939941,
+      "step": 526
+    },
+    {
+      "epoch": 0.22278481012658227,
+      "grad_norm": 0.6191489696502686,
+      "learning_rate": 9.261862917398945e-05,
+      "loss": 1.0153647661209106,
+      "step": 528
+    },
+    {
+      "epoch": 0.22362869198312235,
+      "grad_norm": 0.6821563243865967,
+      "learning_rate": 9.29701230228471e-05,
+      "loss": 0.9649755954742432,
+      "step": 530
+    },
+    {
+      "epoch": 0.22447257383966246,
+      "grad_norm": 1.760398268699646,
+      "learning_rate": 9.332161687170476e-05,
+      "loss": 0.8673232197761536,
+      "step": 532
+    },
+    {
+      "epoch": 0.22531645569620254,
+      "grad_norm": 0.6670058369636536,
+      "learning_rate": 9.36731107205624e-05,
+      "loss": 0.9942440986633301,
+      "step": 534
+    },
+    {
+      "epoch": 0.22616033755274262,
+      "grad_norm": 0.7345916032791138,
+      "learning_rate": 9.402460456942003e-05,
+      "loss": 1.0364389419555664,
+      "step": 536
+    },
+    {
+      "epoch": 0.2270042194092827,
+      "grad_norm": 0.5946128964424133,
+      "learning_rate": 9.437609841827768e-05,
+      "loss": 0.9314924478530884,
+      "step": 538
+    },
+    {
+      "epoch": 0.22784810126582278,
+      "grad_norm": 0.5800848603248596,
+      "learning_rate": 9.472759226713534e-05,
+      "loss": 1.0694862604141235,
+      "step": 540
+    },
+    {
+      "epoch": 0.22869198312236286,
+      "grad_norm": 0.6712192893028259,
+      "learning_rate": 9.507908611599297e-05,
+      "loss": 1.03531014919281,
+      "step": 542
+    },
+    {
+      "epoch": 0.22953586497890296,
+      "grad_norm": 0.5641416311264038,
+      "learning_rate": 9.543057996485063e-05,
+      "loss": 0.9795235991477966,
+      "step": 544
+    },
+    {
+      "epoch": 0.23037974683544304,
+      "grad_norm": 0.50412517786026,
+      "learning_rate": 9.578207381370826e-05,
+      "loss": 0.9641494750976562,
+      "step": 546
+    },
+    {
+      "epoch": 0.23122362869198312,
+      "grad_norm": 0.579118549823761,
+      "learning_rate": 9.61335676625659e-05,
+      "loss": 0.9375281929969788,
+      "step": 548
+    },
+    {
+      "epoch": 0.2320675105485232,
+      "grad_norm": 0.5888341665267944,
+      "learning_rate": 9.648506151142355e-05,
+      "loss": 0.9414046406745911,
+      "step": 550
+    },
+    {
+      "epoch": 0.23291139240506328,
+      "grad_norm": 0.5595056414604187,
+      "learning_rate": 9.68365553602812e-05,
+      "loss": 0.9005617499351501,
+      "step": 552
+    },
+    {
+      "epoch": 0.23375527426160336,
+      "grad_norm": 0.6605326533317566,
+      "learning_rate": 9.718804920913884e-05,
+      "loss": 1.0283968448638916,
+      "step": 554
+    },
+    {
+      "epoch": 0.23459915611814347,
+      "grad_norm": 0.5657313466072083,
+      "learning_rate": 9.753954305799648e-05,
+      "loss": 1.0058249235153198,
+      "step": 556
+    },
+    {
+      "epoch": 0.23544303797468355,
+      "grad_norm": 0.5433364510536194,
+      "learning_rate": 9.789103690685413e-05,
+      "loss": 0.9835494756698608,
+      "step": 558
+    },
+    {
+      "epoch": 0.23628691983122363,
+      "grad_norm": 0.6129802465438843,
+      "learning_rate": 9.824253075571179e-05,
+      "loss": 1.054532527923584,
+      "step": 560
+    },
+    {
+      "epoch": 0.2371308016877637,
+      "grad_norm": 0.6496239304542542,
+      "learning_rate": 9.859402460456942e-05,
+      "loss": 1.0240973234176636,
+      "step": 562
+    },
+    {
+      "epoch": 0.2379746835443038,
+      "grad_norm": 0.6380873918533325,
+      "learning_rate": 9.894551845342706e-05,
+      "loss": 1.0229179859161377,
+      "step": 564
+    },
+    {
+      "epoch": 0.23881856540084387,
+      "grad_norm": 0.6151993870735168,
+      "learning_rate": 9.929701230228471e-05,
+      "loss": 1.0111570358276367,
+      "step": 566
+    },
+    {
+      "epoch": 0.23966244725738398,
+      "grad_norm": 0.5727584958076477,
+      "learning_rate": 9.964850615114237e-05,
+      "loss": 0.9450829029083252,
+      "step": 568
+    },
+    {
+      "epoch": 0.24050632911392406,
+      "grad_norm": 0.6620725989341736,
+      "learning_rate": 0.0001,
+      "loss": 0.9800319075584412,
+      "step": 570
+    },
+    {
+      "epoch": 0.24135021097046414,
+      "grad_norm": 0.6151163578033447,
+      "learning_rate": 0.00010035149384885764,
+      "loss": 0.9757438898086548,
+      "step": 572
+    },
+    {
+      "epoch": 0.24219409282700421,
+      "grad_norm": 0.5672140717506409,
+      "learning_rate": 0.0001007029876977153,
+      "loss": 0.9104921817779541,
+      "step": 574
+    },
+    {
+      "epoch": 0.2430379746835443,
+      "grad_norm": 0.5697256326675415,
+      "learning_rate": 0.00010105448154657293,
+      "loss": 1.1027376651763916,
+      "step": 576
+    },
+    {
+      "epoch": 0.2438818565400844,
+      "grad_norm": 0.5590381622314453,
+      "learning_rate": 0.00010140597539543057,
+      "loss": 1.0055404901504517,
+      "step": 578
+    },
+    {
+      "epoch": 0.24472573839662448,
+      "grad_norm": 0.5518567562103271,
+      "learning_rate": 0.00010175746924428824,
+      "loss": 1.020835518836975,
+      "step": 580
+    },
+    {
+      "epoch": 0.24556962025316456,
+      "grad_norm": 0.6338496208190918,
+      "learning_rate": 0.00010210896309314588,
+      "loss": 0.9528344869613647,
+      "step": 582
+    },
+    {
+      "epoch": 0.24641350210970464,
+      "grad_norm": 0.6497329473495483,
+      "learning_rate": 0.00010246045694200353,
+      "loss": 1.0088670253753662,
+      "step": 584
+    },
+    {
+      "epoch": 0.24725738396624472,
+      "grad_norm": 0.49888095259666443,
+      "learning_rate": 0.00010281195079086117,
+      "loss": 0.9961200952529907,
+      "step": 586
+    },
+    {
+      "epoch": 0.2481012658227848,
+      "grad_norm": 0.5680158734321594,
+      "learning_rate": 0.0001031634446397188,
+      "loss": 0.9635610580444336,
+      "step": 588
+    },
+    {
+      "epoch": 0.2489451476793249,
+      "grad_norm": 0.658168375492096,
+      "learning_rate": 0.00010351493848857646,
+      "loss": 0.9392287135124207,
+      "step": 590
+    },
+    {
+      "epoch": 0.249789029535865,
+      "grad_norm": 0.618262767791748,
+      "learning_rate": 0.0001038664323374341,
+      "loss": 0.9600516557693481,
+      "step": 592
+    },
+    {
+      "epoch": 0.25063291139240507,
+      "grad_norm": 0.6003909111022949,
+      "learning_rate": 0.00010421792618629173,
+      "loss": 1.005476713180542,
+      "step": 594
+    },
+    {
+      "epoch": 0.2514767932489452,
+      "grad_norm": 0.5437078475952148,
+      "learning_rate": 0.00010456942003514938,
+      "loss": 0.9523017406463623,
+      "step": 596
+    },
+    {
+      "epoch": 0.2523206751054852,
+      "grad_norm": 0.5524541735649109,
+      "learning_rate": 0.00010492091388400705,
+      "loss": 0.9526668787002563,
+      "step": 598
+    },
+    {
+      "epoch": 0.25316455696202533,
+      "grad_norm": 0.679504930973053,
+      "learning_rate": 0.00010527240773286469,
+      "loss": 1.019660472869873,
+      "step": 600
+    },
+    {
+      "epoch": 0.25316455696202533,
+      "eval_loss": 1.0193854570388794,
+      "eval_runtime": 677.9523,
+      "eval_samples_per_second": 3.108,
+      "eval_steps_per_second": 3.108,
+      "step": 600
+    },
+    {
+      "epoch": 0.2540084388185654,
+      "grad_norm": 0.5646136999130249,
+      "learning_rate": 0.00010562390158172233,
+      "loss": 0.9910882711410522,
+      "step": 602
+    },
+    {
+      "epoch": 0.2548523206751055,
+      "grad_norm": 0.5238093137741089,
+      "learning_rate": 0.00010597539543057998,
+      "loss": 0.9616432785987854,
+      "step": 604
+    },
+    {
+      "epoch": 0.25569620253164554,
+      "grad_norm": 0.7483857274055481,
+      "learning_rate": 0.00010632688927943762,
+      "loss": 1.0078275203704834,
+      "step": 606
+    },
+    {
+      "epoch": 0.25654008438818565,
+      "grad_norm": 0.578948974609375,
+      "learning_rate": 0.00010667838312829525,
+      "loss": 0.9827103018760681,
+      "step": 608
+    },
+    {
+      "epoch": 0.25738396624472576,
+      "grad_norm": 0.5525906085968018,
+      "learning_rate": 0.00010702987697715289,
+      "loss": 1.0423277616500854,
+      "step": 610
+    },
+    {
+      "epoch": 0.2582278481012658,
+      "grad_norm": 0.6721326112747192,
+      "learning_rate": 0.00010738137082601054,
+      "loss": 0.9561693072319031,
+      "step": 612
+    },
+    {
+      "epoch": 0.2590717299578059,
+      "grad_norm": 0.5701051354408264,
+      "learning_rate": 0.00010773286467486821,
+      "loss": 0.9602992534637451,
+      "step": 614
+    },
+    {
+      "epoch": 0.25991561181434597,
+      "grad_norm": 0.6349860429763794,
+      "learning_rate": 0.00010808435852372585,
+      "loss": 1.1422650814056396,
+      "step": 616
+    },
+    {
+      "epoch": 0.2607594936708861,
+      "grad_norm": 0.5496085286140442,
+      "learning_rate": 0.00010843585237258349,
+      "loss": 0.9762773513793945,
+      "step": 618
+    },
+    {
+      "epoch": 0.2616033755274262,
+      "grad_norm": 0.6080722808837891,
+      "learning_rate": 0.00010878734622144114,
+      "loss": 1.0133616924285889,
+      "step": 620
+    },
+    {
+      "epoch": 0.26244725738396624,
+      "grad_norm": 0.5450218915939331,
+      "learning_rate": 0.00010913884007029878,
+      "loss": 0.9385587573051453,
+      "step": 622
+    },
+    {
+      "epoch": 0.26329113924050634,
+      "grad_norm": 0.592106819152832,
+      "learning_rate": 0.00010949033391915641,
+      "loss": 0.9359989762306213,
+      "step": 624
+    },
+    {
+      "epoch": 0.2641350210970464,
+      "grad_norm": 0.6449427604675293,
+      "learning_rate": 0.00010984182776801407,
+      "loss": 1.0266027450561523,
+      "step": 626
+    },
+    {
+      "epoch": 0.2649789029535865,
+      "grad_norm": 0.538299560546875,
+      "learning_rate": 0.0001101933216168717,
+      "loss": 0.9303187131881714,
+      "step": 628
+    },
+    {
+      "epoch": 0.26582278481012656,
+      "grad_norm": 0.546316921710968,
+      "learning_rate": 0.00011054481546572934,
+      "loss": 0.9368857145309448,
+      "step": 630
+    },
+    {
+      "epoch": 0.26666666666666666,
+      "grad_norm": 0.5818730592727661,
+      "learning_rate": 0.00011089630931458701,
+      "loss": 0.9573145508766174,
+      "step": 632
+    },
+    {
+      "epoch": 0.26751054852320677,
+      "grad_norm": 0.5958262085914612,
+      "learning_rate": 0.00011124780316344465,
+      "loss": 0.9345449805259705,
+      "step": 634
+    },
+    {
+      "epoch": 0.2683544303797468,
+      "grad_norm": 0.6259077787399292,
+      "learning_rate": 0.0001115992970123023,
+      "loss": 1.0906590223312378,
+      "step": 636
+    },
+    {
+      "epoch": 0.26919831223628693,
+      "grad_norm": 0.589672863483429,
+      "learning_rate": 0.00011195079086115994,
+      "loss": 1.0757447481155396,
+      "step": 638
+    },
+    {
+      "epoch": 0.270042194092827,
+      "grad_norm": 0.5714080333709717,
+      "learning_rate": 0.00011230228471001758,
+      "loss": 0.9310855269432068,
+      "step": 640
+    },
+    {
+      "epoch": 0.2708860759493671,
+      "grad_norm": 0.45342639088630676,
+      "learning_rate": 0.00011265377855887523,
+      "loss": 0.9276360273361206,
+      "step": 642
+    },
+    {
+      "epoch": 0.2717299578059072,
+      "grad_norm": 0.6386750340461731,
+      "learning_rate": 0.00011300527240773287,
+      "loss": 1.084719181060791,
+      "step": 644
+    },
+    {
+      "epoch": 0.27257383966244725,
+      "grad_norm": 0.6446163654327393,
+      "learning_rate": 0.0001133567662565905,
+      "loss": 0.9763918519020081,
+      "step": 646
+    },
+    {
+      "epoch": 0.27341772151898736,
+      "grad_norm": 0.5925686359405518,
+      "learning_rate": 0.00011370826010544816,
+      "loss": 0.9517921805381775,
+      "step": 648
+    },
+    {
+      "epoch": 0.2742616033755274,
+      "grad_norm": 0.5399773716926575,
+      "learning_rate": 0.00011405975395430582,
+      "loss": 1.0587927103042603,
+      "step": 650
+    },
+    {
+      "epoch": 0.2751054852320675,
+      "grad_norm": 0.5872456431388855,
+      "learning_rate": 0.00011441124780316346,
+      "loss": 0.883341908454895,
+      "step": 652
+    },
+    {
+      "epoch": 0.2759493670886076,
+      "grad_norm": 0.5574564337730408,
+      "learning_rate": 0.0001147627416520211,
+      "loss": 1.0306891202926636,
+      "step": 654
+    },
+    {
+      "epoch": 0.2767932489451477,
+      "grad_norm": 0.47789013385772705,
+      "learning_rate": 0.00011511423550087874,
+      "loss": 0.8814032077789307,
+      "step": 656
+    },
+    {
+      "epoch": 0.2776371308016878,
+      "grad_norm": 0.5565530061721802,
+      "learning_rate": 0.00011546572934973639,
+      "loss": 0.9460552334785461,
+      "step": 658
+    },
+    {
+      "epoch": 0.27848101265822783,
+      "grad_norm": 0.5299761295318604,
+      "learning_rate": 0.00011581722319859403,
+      "loss": 0.9475110769271851,
+      "step": 660
+    },
+    {
+      "epoch": 0.27932489451476794,
+      "grad_norm": 0.6503344178199768,
+      "learning_rate": 0.00011616871704745166,
+      "loss": 1.0630913972854614,
+      "step": 662
+    },
+    {
+      "epoch": 0.280168776371308,
+      "grad_norm": 0.5794585943222046,
+      "learning_rate": 0.00011652021089630932,
+      "loss": 0.9389138221740723,
+      "step": 664
+    },
+    {
+      "epoch": 0.2810126582278481,
+      "grad_norm": 0.5762867331504822,
+      "learning_rate": 0.00011687170474516695,
+      "loss": 0.8934136033058167,
+      "step": 666
+    },
+    {
+      "epoch": 0.2818565400843882,
+      "grad_norm": 0.6565435528755188,
+      "learning_rate": 0.00011722319859402462,
+      "loss": 1.1072614192962646,
+      "step": 668
+    },
+    {
+      "epoch": 0.28270042194092826,
+      "grad_norm": 0.5819830298423767,
+      "learning_rate": 0.00011757469244288226,
+      "loss": 1.0501434803009033,
+      "step": 670
+    },
+    {
+      "epoch": 0.28354430379746837,
+      "grad_norm": 0.6071487069129944,
+      "learning_rate": 0.00011792618629173991,
+      "loss": 0.9880793690681458,
+      "step": 672
+    },
+    {
+      "epoch": 0.2843881856540084,
+      "grad_norm": 0.5765058398246765,
+      "learning_rate": 0.00011827768014059755,
+      "loss": 0.9670693874359131,
+      "step": 674
+    },
+    {
+      "epoch": 0.2852320675105485,
+      "grad_norm": 0.5245351791381836,
+      "learning_rate": 0.00011862917398945519,
+      "loss": 0.9602360725402832,
+      "step": 676
+    },
+    {
+      "epoch": 0.28607594936708863,
+      "grad_norm": 0.6189922094345093,
+      "learning_rate": 0.00011898066783831282,
+      "loss": 0.9684560894966125,
+      "step": 678
+    },
+    {
+      "epoch": 0.2869198312236287,
+      "grad_norm": 0.6138690710067749,
+      "learning_rate": 0.00011933216168717048,
+      "loss": 0.9465792775154114,
+      "step": 680
+    },
+    {
+      "epoch": 0.2877637130801688,
+      "grad_norm": 0.5371595621109009,
+      "learning_rate": 0.00011968365553602812,
+      "loss": 0.8495944738388062,
+      "step": 682
+    },
+    {
+      "epoch": 0.28860759493670884,
+      "grad_norm": 0.5549944639205933,
+      "learning_rate": 0.00012003514938488578,
+      "loss": 0.9663267135620117,
+      "step": 684
+    },
+    {
+      "epoch": 0.28945147679324895,
+      "grad_norm": 0.6484189033508301,
+      "learning_rate": 0.00012038664323374342,
+      "loss": 0.9736058712005615,
+      "step": 686
+    },
+    {
+      "epoch": 0.290295358649789,
+      "grad_norm": 0.540351390838623,
+      "learning_rate": 0.00012073813708260107,
+      "loss": 1.0591845512390137,
+      "step": 688
+    },
+    {
+      "epoch": 0.2911392405063291,
+      "grad_norm": 0.5657922029495239,
+      "learning_rate": 0.00012108963093145871,
+      "loss": 0.944908618927002,
+      "step": 690
+    },
+    {
+      "epoch": 0.2919831223628692,
+      "grad_norm": 0.6040505170822144,
+      "learning_rate": 0.00012144112478031635,
+      "loss": 1.0018219947814941,
+      "step": 692
+    },
+    {
+      "epoch": 0.29282700421940927,
+      "grad_norm": 0.5435477495193481,
+      "learning_rate": 0.000121792618629174,
+      "loss": 1.0351502895355225,
+      "step": 694
+    },
+    {
+      "epoch": 0.2936708860759494,
+      "grad_norm": 0.5712518692016602,
+      "learning_rate": 0.00012214411247803164,
+      "loss": 0.9935672283172607,
+      "step": 696
+    },
+    {
+      "epoch": 0.29451476793248943,
+      "grad_norm": 0.6138222813606262,
+      "learning_rate": 0.00012249560632688928,
+      "loss": 1.0165108442306519,
+      "step": 698
+    },
+    {
+      "epoch": 0.29535864978902954,
+      "grad_norm": 0.4495212435722351,
+      "learning_rate": 0.00012284710017574691,
+      "loss": 0.9334425926208496,
+      "step": 700
+    },
+    {
+      "epoch": 0.29535864978902954,
+      "eval_loss": 0.996929407119751,
+      "eval_runtime": 668.6398,
+      "eval_samples_per_second": 3.151,
+      "eval_steps_per_second": 3.151,
+      "step": 700
+    },
+    {
+      "epoch": 0.29620253164556964,
+      "grad_norm": 0.5321539044380188,
+      "learning_rate": 0.00012319859402460458,
+      "loss": 1.0516537427902222,
+      "step": 702
+    },
+    {
+      "epoch": 0.2970464135021097,
+      "grad_norm": 0.5716516971588135,
+      "learning_rate": 0.00012355008787346222,
+      "loss": 0.9387198686599731,
+      "step": 704
+    },
+    {
+      "epoch": 0.2978902953586498,
+      "grad_norm": 0.5617920160293579,
+      "learning_rate": 0.00012390158172231988,
+      "loss": 0.99737948179245,
+      "step": 706
+    },
+    {
+      "epoch": 0.29873417721518986,
+      "grad_norm": 0.4922899007797241,
+      "learning_rate": 0.00012425307557117752,
+      "loss": 0.9955025911331177,
+      "step": 708
+    },
+    {
+      "epoch": 0.29957805907172996,
+      "grad_norm": 0.543501615524292,
+      "learning_rate": 0.00012460456942003516,
+      "loss": 0.9124280214309692,
+      "step": 710
+    },
+    {
+      "epoch": 0.30042194092827,
+      "grad_norm": 0.49590054154396057,
+      "learning_rate": 0.0001249560632688928,
+      "loss": 0.9820216298103333,
+      "step": 712
+    },
+    {
+      "epoch": 0.3012658227848101,
+      "grad_norm": 0.5984305739402771,
+      "learning_rate": 0.00012530755711775044,
+      "loss": 1.0152074098587036,
+      "step": 714
+    },
+    {
+      "epoch": 0.30210970464135023,
+      "grad_norm": 0.9343504905700684,
+      "learning_rate": 0.00012565905096660807,
+      "loss": 1.0577725172042847,
+      "step": 716
+    },
+    {
+      "epoch": 0.3029535864978903,
+      "grad_norm": 0.5118702054023743,
+      "learning_rate": 0.0001260105448154657,
+      "loss": 0.9830358028411865,
+      "step": 718
+    },
+    {
+      "epoch": 0.3037974683544304,
+      "grad_norm": 0.4940392076969147,
+      "learning_rate": 0.00012636203866432338,
+      "loss": 0.9466043710708618,
+      "step": 720
+    },
+    {
+      "epoch": 0.30464135021097044,
+      "grad_norm": 0.5965693593025208,
+      "learning_rate": 0.00012671353251318104,
+      "loss": 1.015270709991455,
+      "step": 722
+    },
+    {
+      "epoch": 0.30548523206751055,
+      "grad_norm": 0.5020529627799988,
+      "learning_rate": 0.00012706502636203868,
+      "loss": 0.9703927636146545,
+      "step": 724
+    },
+    {
+      "epoch": 0.30632911392405066,
+      "grad_norm": 0.6067010164260864,
+      "learning_rate": 0.00012741652021089632,
+      "loss": 1.0255526304244995,
+      "step": 726
+    },
+    {
+      "epoch": 0.3071729957805907,
+      "grad_norm": 0.5931884050369263,
+      "learning_rate": 0.00012776801405975396,
+      "loss": 0.9335633516311646,
+      "step": 728
+    },
+    {
+      "epoch": 0.3080168776371308,
+      "grad_norm": 0.5938752293586731,
+      "learning_rate": 0.0001281195079086116,
+      "loss": 1.0921578407287598,
+      "step": 730
+    },
+    {
+      "epoch": 0.30886075949367087,
+      "grad_norm": 0.49728086590766907,
+      "learning_rate": 0.00012847100175746923,
+      "loss": 0.963066041469574,
+      "step": 732
+    },
+    {
+      "epoch": 0.309704641350211,
+      "grad_norm": 0.5452080965042114,
+      "learning_rate": 0.0001288224956063269,
+      "loss": 0.9513075351715088,
+      "step": 734
+    },
+    {
+      "epoch": 0.3105485232067511,
+      "grad_norm": 0.5497731566429138,
+      "learning_rate": 0.00012917398945518454,
+      "loss": 0.8576077222824097,
+      "step": 736
+    },
+    {
+      "epoch": 0.31139240506329113,
+      "grad_norm": 0.5580397248268127,
+      "learning_rate": 0.0001295254833040422,
+      "loss": 0.9542577862739563,
+      "step": 738
+    },
+    {
+      "epoch": 0.31223628691983124,
+      "grad_norm": 0.5890427827835083,
+      "learning_rate": 0.00012987697715289984,
+      "loss": 0.8992732167243958,
+      "step": 740
+    },
+    {
+      "epoch": 0.3130801687763713,
+      "grad_norm": 0.5942965745925903,
+      "learning_rate": 0.00013022847100175748,
+      "loss": 1.0322896242141724,
+      "step": 742
+    },
+    {
+      "epoch": 0.3139240506329114,
+      "grad_norm": 0.6341713070869446,
+      "learning_rate": 0.00013057996485061512,
+      "loss": 0.9217103719711304,
+      "step": 744
+    },
+    {
+      "epoch": 0.31476793248945145,
+      "grad_norm": 0.5294105410575867,
+      "learning_rate": 0.00013093145869947276,
+      "loss": 0.951789915561676,
+      "step": 746
+    },
+    {
+      "epoch": 0.31561181434599156,
+      "grad_norm": 0.6372058391571045,
+      "learning_rate": 0.0001312829525483304,
+      "loss": 0.9459875226020813,
+      "step": 748
+    },
+    {
+      "epoch": 0.31645569620253167,
+      "grad_norm": 0.5979796648025513,
+      "learning_rate": 0.00013163444639718806,
+      "loss": 0.9626097679138184,
+      "step": 750
+    },
+    {
+      "epoch": 0.3172995780590717,
+      "grad_norm": 0.5682399868965149,
+      "learning_rate": 0.0001319859402460457,
+      "loss": 1.0261781215667725,
+      "step": 752
+    },
+    {
+      "epoch": 0.3181434599156118,
+      "grad_norm": 0.5349125266075134,
+      "learning_rate": 0.00013233743409490336,
+      "loss": 0.9319828152656555,
+      "step": 754
+    },
+    {
+      "epoch": 0.3189873417721519,
+      "grad_norm": 0.6093934178352356,
+      "learning_rate": 0.000132688927943761,
+      "loss": 0.9216550588607788,
+      "step": 756
+    },
+    {
+      "epoch": 0.319831223628692,
+      "grad_norm": 0.5188612341880798,
+      "learning_rate": 0.00013304042179261864,
+      "loss": 0.901739776134491,
+      "step": 758
+    },
+    {
+      "epoch": 0.3206751054852321,
+      "grad_norm": 0.5877130627632141,
+      "learning_rate": 0.00013339191564147628,
+      "loss": 1.0362589359283447,
+      "step": 760
+    },
+    {
+      "epoch": 0.32151898734177214,
+      "grad_norm": 0.5542771816253662,
+      "learning_rate": 0.00013374340949033392,
+      "loss": 0.8787116408348083,
+      "step": 762
+    },
+    {
+      "epoch": 0.32236286919831225,
+      "grad_norm": 0.5084902048110962,
+      "learning_rate": 0.00013409490333919156,
+      "loss": 0.9237037301063538,
+      "step": 764
+    },
+    {
+      "epoch": 0.3232067510548523,
+      "grad_norm": 0.5461528301239014,
+      "learning_rate": 0.00013444639718804922,
+      "loss": 1.0150731801986694,
+      "step": 766
+    },
+    {
+      "epoch": 0.3240506329113924,
+      "grad_norm": 0.53483647108078,
+      "learning_rate": 0.00013479789103690686,
+      "loss": 0.8985214829444885,
+      "step": 768
+    },
+    {
+      "epoch": 0.32489451476793246,
+      "grad_norm": 0.5580531358718872,
+      "learning_rate": 0.0001351493848857645,
+      "loss": 1.0225775241851807,
+      "step": 770
+    },
+    {
+      "epoch": 0.32573839662447257,
+      "grad_norm": 0.5203377604484558,
+      "learning_rate": 0.00013550087873462216,
+      "loss": 0.9571293592453003,
+      "step": 772
+    },
+    {
+      "epoch": 0.3265822784810127,
+      "grad_norm": 0.5049671530723572,
+      "learning_rate": 0.0001358523725834798,
+      "loss": 1.0468909740447998,
+      "step": 774
+    },
+    {
+      "epoch": 0.32742616033755273,
+      "grad_norm": 0.4723063111305237,
+      "learning_rate": 0.00013620386643233744,
+      "loss": 0.7743215560913086,
+      "step": 776
+    },
+    {
+      "epoch": 0.32827004219409284,
+      "grad_norm": 0.6310980916023254,
+      "learning_rate": 0.00013655536028119508,
+      "loss": 1.021510362625122,
+      "step": 778
+    },
+    {
+      "epoch": 0.3291139240506329,
+      "grad_norm": 0.47066664695739746,
+      "learning_rate": 0.00013690685413005274,
+      "loss": 0.9134382605552673,
+      "step": 780
+    },
+    {
+      "epoch": 0.329957805907173,
+      "grad_norm": 0.5725092887878418,
+      "learning_rate": 0.00013725834797891038,
+      "loss": 0.9797834753990173,
+      "step": 782
+    },
+    {
+      "epoch": 0.3308016877637131,
+      "grad_norm": 0.5139563083648682,
+      "learning_rate": 0.00013760984182776802,
+      "loss": 0.9372621178627014,
+      "step": 784
+    },
+    {
+      "epoch": 0.33164556962025316,
+      "grad_norm": 0.5275821685791016,
+      "learning_rate": 0.00013796133567662566,
+      "loss": 0.9528245329856873,
+      "step": 786
+    },
+    {
+      "epoch": 0.33248945147679326,
+      "grad_norm": 0.5702582001686096,
+      "learning_rate": 0.0001383128295254833,
+      "loss": 0.9750176072120667,
+      "step": 788
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.5281293392181396,
+      "learning_rate": 0.00013866432337434096,
+      "loss": 0.9412306547164917,
+      "step": 790
+    },
+    {
+      "epoch": 0.3341772151898734,
+      "grad_norm": 0.5578986406326294,
+      "learning_rate": 0.0001390158172231986,
+      "loss": 0.997580885887146,
+      "step": 792
+    },
+    {
+      "epoch": 0.33502109704641353,
+      "grad_norm": 0.50461345911026,
+      "learning_rate": 0.00013936731107205624,
+      "loss": 0.9082320928573608,
+      "step": 794
+    },
+    {
+      "epoch": 0.3358649789029536,
+      "grad_norm": 0.5258530378341675,
+      "learning_rate": 0.0001397188049209139,
+      "loss": 1.0082844495773315,
+      "step": 796
+    },
+    {
+      "epoch": 0.3367088607594937,
+      "grad_norm": 0.5548169016838074,
+      "learning_rate": 0.00014007029876977154,
+      "loss": 0.9729003310203552,
+      "step": 798
+    },
+    {
+      "epoch": 0.33755274261603374,
+      "grad_norm": 0.48601076006889343,
+      "learning_rate": 0.00014042179261862918,
+      "loss": 0.9099526405334473,
+      "step": 800
+    },
+    {
+      "epoch": 0.33755274261603374,
+      "eval_loss": 0.9800403714179993,
+      "eval_runtime": 678.8306,
+      "eval_samples_per_second": 3.104,
+      "eval_steps_per_second": 3.104,
+      "step": 800
+    },
+    {
+      "epoch": 0.33839662447257385,
+      "grad_norm": 0.5413158535957336,
+      "learning_rate": 0.00014077328646748682,
+      "loss": 0.8610644936561584,
+      "step": 802
+    },
+    {
+      "epoch": 0.3392405063291139,
+      "grad_norm": 0.5147035717964172,
+      "learning_rate": 0.00014112478031634446,
+      "loss": 0.9584825038909912,
+      "step": 804
+    },
+    {
+      "epoch": 0.340084388185654,
+      "grad_norm": 0.5931771397590637,
+      "learning_rate": 0.0001414762741652021,
+      "loss": 1.0142558813095093,
+      "step": 806
+    },
+    {
+      "epoch": 0.3409282700421941,
+      "grad_norm": 0.5178377032279968,
+      "learning_rate": 0.00014182776801405976,
+      "loss": 1.0078763961791992,
+      "step": 808
+    },
+    {
+      "epoch": 0.34177215189873417,
+      "grad_norm": 0.5453237295150757,
+      "learning_rate": 0.0001421792618629174,
+      "loss": 0.9107215404510498,
+      "step": 810
+    },
+    {
+      "epoch": 0.3426160337552743,
+      "grad_norm": 0.5886152982711792,
+      "learning_rate": 0.00014253075571177506,
+      "loss": 0.9981362819671631,
+      "step": 812
+    },
+    {
+      "epoch": 0.3434599156118143,
+      "grad_norm": 0.48040178418159485,
+      "learning_rate": 0.0001428822495606327,
+      "loss": 0.9636131525039673,
+      "step": 814
+    },
+    {
+      "epoch": 0.34430379746835443,
+      "grad_norm": 0.5011753439903259,
+      "learning_rate": 0.00014323374340949034,
+      "loss": 0.9590586423873901,
+      "step": 816
+    },
+    {
+      "epoch": 0.34514767932489454,
+      "grad_norm": 0.57858806848526,
+      "learning_rate": 0.00014358523725834798,
+      "loss": 0.978246808052063,
+      "step": 818
+    },
+    {
+      "epoch": 0.3459915611814346,
+      "grad_norm": 0.46092939376831055,
+      "learning_rate": 0.00014393673110720562,
+      "loss": 0.8549934029579163,
+      "step": 820
+    },
+    {
+      "epoch": 0.3468354430379747,
+      "grad_norm": 0.5756489038467407,
+      "learning_rate": 0.00014428822495606326,
+      "loss": 0.9771265387535095,
+      "step": 822
+    },
+    {
+      "epoch": 0.34767932489451475,
+      "grad_norm": 0.5501731634140015,
+      "learning_rate": 0.00014463971880492092,
+      "loss": 0.8739748001098633,
+      "step": 824
+    },
+    {
+      "epoch": 0.34852320675105486,
+      "grad_norm": 0.5451868176460266,
+      "learning_rate": 0.0001449912126537786,
+      "loss": 0.9129468202590942,
+      "step": 826
+    },
+    {
+      "epoch": 0.3493670886075949,
+      "grad_norm": 0.4624619781970978,
+      "learning_rate": 0.00014534270650263623,
+      "loss": 0.9196704030036926,
+      "step": 828
+    },
+    {
+      "epoch": 0.350210970464135,
+      "grad_norm": 0.520878791809082,
+      "learning_rate": 0.00014569420035149386,
+      "loss": 0.9976527690887451,
+      "step": 830
+    },
+    {
+      "epoch": 0.3510548523206751,
+      "grad_norm": 0.4469171464443207,
+      "learning_rate": 0.0001460456942003515,
+      "loss": 0.7753443717956543,
+      "step": 832
+    },
+    {
+      "epoch": 0.3518987341772152,
+      "grad_norm": 0.5105249881744385,
+      "learning_rate": 0.00014639718804920914,
+      "loss": 0.9584846496582031,
+      "step": 834
+    },
+    {
+      "epoch": 0.3527426160337553,
+      "grad_norm": 0.5043913125991821,
+      "learning_rate": 0.00014674868189806678,
+      "loss": 0.9013500213623047,
+      "step": 836
+    },
+    {
+      "epoch": 0.35358649789029534,
+      "grad_norm": 0.575850784778595,
+      "learning_rate": 0.00014710017574692442,
+      "loss": 0.9775562286376953,
+      "step": 838
+    },
+    {
+      "epoch": 0.35443037974683544,
+      "grad_norm": 0.5128876566886902,
+      "learning_rate": 0.00014745166959578208,
+      "loss": 0.9278940558433533,
+      "step": 840
+    },
+    {
+      "epoch": 0.35527426160337555,
+      "grad_norm": 0.5757885575294495,
+      "learning_rate": 0.00014780316344463975,
+      "loss": 1.0091488361358643,
+      "step": 842
+    },
+    {
+      "epoch": 0.3561181434599156,
+      "grad_norm": 0.500934898853302,
+      "learning_rate": 0.00014815465729349739,
+      "loss": 0.9286836981773376,
+      "step": 844
+    },
+    {
+      "epoch": 0.3569620253164557,
+      "grad_norm": 0.5220686197280884,
+      "learning_rate": 0.00014850615114235502,
+      "loss": 0.9484171867370605,
+      "step": 846
+    },
+    {
+      "epoch": 0.35780590717299576,
+      "grad_norm": 0.5494697690010071,
+      "learning_rate": 0.00014885764499121266,
+      "loss": 1.0556397438049316,
+      "step": 848
+    },
+    {
+      "epoch": 0.35864978902953587,
+      "grad_norm": 0.46633943915367126,
+      "learning_rate": 0.0001492091388400703,
+      "loss": 0.933089017868042,
+      "step": 850
+    },
+    {
+      "epoch": 0.3594936708860759,
+      "grad_norm": 0.4864962697029114,
+      "learning_rate": 0.00014956063268892794,
+      "loss": 1.0016963481903076,
+      "step": 852
+    },
+    {
+      "epoch": 0.36033755274261603,
+      "grad_norm": 0.5032764673233032,
+      "learning_rate": 0.0001499121265377856,
+      "loss": 0.9085348844528198,
+      "step": 854
+    },
+    {
+      "epoch": 0.36118143459915614,
+      "grad_norm": 0.5424998998641968,
+      "learning_rate": 0.00015026362038664324,
+      "loss": 0.923502504825592,
+      "step": 856
+    },
+    {
+      "epoch": 0.3620253164556962,
+      "grad_norm": 0.5181655287742615,
+      "learning_rate": 0.00015061511423550088,
+      "loss": 0.919174313545227,
+      "step": 858
+    },
+    {
+      "epoch": 0.3628691983122363,
+      "grad_norm": 0.5129443407058716,
+      "learning_rate": 0.00015096660808435855,
+      "loss": 0.851981520652771,
+      "step": 860
+    },
+    {
+      "epoch": 0.36371308016877635,
+      "grad_norm": 0.49540698528289795,
+      "learning_rate": 0.00015131810193321618,
+      "loss": 0.8633858561515808,
+      "step": 862
+    },
+    {
+      "epoch": 0.36455696202531646,
+      "grad_norm": 0.4706701934337616,
+      "learning_rate": 0.00015166959578207382,
+      "loss": 0.9473677277565002,
+      "step": 864
+    },
+    {
+      "epoch": 0.36540084388185656,
+      "grad_norm": 0.587704598903656,
+      "learning_rate": 0.00015202108963093146,
+      "loss": 0.942383885383606,
+      "step": 866
+    },
+    {
+      "epoch": 0.3662447257383966,
+      "grad_norm": 0.5851273536682129,
+      "learning_rate": 0.0001523725834797891,
+      "loss": 0.9811251163482666,
+      "step": 868
+    },
+    {
+      "epoch": 0.3670886075949367,
+      "grad_norm": 0.46357613801956177,
+      "learning_rate": 0.00015272407732864676,
+      "loss": 0.8151084184646606,
+      "step": 870
+    },
+    {
+      "epoch": 0.3679324894514768,
+      "grad_norm": 0.47250670194625854,
+      "learning_rate": 0.0001530755711775044,
+      "loss": 0.9023333191871643,
+      "step": 872
+    },
+    {
+      "epoch": 0.3687763713080169,
+      "grad_norm": 0.47510042786598206,
+      "learning_rate": 0.00015342706502636204,
+      "loss": 0.8758499622344971,
+      "step": 874
+    },
+    {
+      "epoch": 0.369620253164557,
+      "grad_norm": 0.5687124729156494,
+      "learning_rate": 0.00015377855887521968,
+      "loss": 0.9837421774864197,
+      "step": 876
+    },
+    {
+      "epoch": 0.37046413502109704,
+      "grad_norm": 0.49064236879348755,
+      "learning_rate": 0.00015413005272407735,
+      "loss": 0.9084216356277466,
+      "step": 878
+    },
+    {
+      "epoch": 0.37130801687763715,
+      "grad_norm": 0.5288164615631104,
+      "learning_rate": 0.00015448154657293498,
+      "loss": 0.8425542712211609,
+      "step": 880
+    },
+    {
+      "epoch": 0.3721518987341772,
+      "grad_norm": 0.5446951985359192,
+      "learning_rate": 0.00015483304042179262,
+      "loss": 1.006197452545166,
+      "step": 882
+    },
+    {
+      "epoch": 0.3729957805907173,
+      "grad_norm": 0.46872642636299133,
+      "learning_rate": 0.00015518453427065026,
+      "loss": 0.9779444932937622,
+      "step": 884
+    },
+    {
+      "epoch": 0.37383966244725736,
+      "grad_norm": 0.4852714240550995,
+      "learning_rate": 0.00015553602811950793,
+      "loss": 0.9099963307380676,
+      "step": 886
+    },
+    {
+      "epoch": 0.37468354430379747,
+      "grad_norm": 0.5219841003417969,
+      "learning_rate": 0.00015588752196836556,
+      "loss": 0.9730570316314697,
+      "step": 888
+    },
+    {
+      "epoch": 0.3755274261603376,
+      "grad_norm": 0.5258626341819763,
+      "learning_rate": 0.0001562390158172232,
+      "loss": 0.9304586052894592,
+      "step": 890
+    },
+    {
+      "epoch": 0.3763713080168776,
+      "grad_norm": 0.4266716241836548,
+      "learning_rate": 0.00015659050966608084,
+      "loss": 0.8680716156959534,
+      "step": 892
+    },
+    {
+      "epoch": 0.37721518987341773,
+      "grad_norm": 0.46361327171325684,
+      "learning_rate": 0.0001569420035149385,
+      "loss": 0.8428018093109131,
+      "step": 894
+    },
+    {
+      "epoch": 0.3780590717299578,
+      "grad_norm": 0.5313687920570374,
+      "learning_rate": 0.00015729349736379614,
+      "loss": 0.8465750217437744,
+      "step": 896
+    },
+    {
+      "epoch": 0.3789029535864979,
+      "grad_norm": 0.47166210412979126,
+      "learning_rate": 0.00015764499121265378,
+      "loss": 0.9120327234268188,
+      "step": 898
+    },
+    {
+      "epoch": 0.379746835443038,
+      "grad_norm": 0.4647318720817566,
+      "learning_rate": 0.00015799648506151145,
+      "loss": 0.8950425982475281,
+      "step": 900
+    },
+    {
+      "epoch": 0.379746835443038,
+      "eval_loss": 0.9643027186393738,
+      "eval_runtime": 691.7929,
+      "eval_samples_per_second": 3.046,
+      "eval_steps_per_second": 3.046,
+      "step": 900
+    },
+    {
+      "epoch": 0.38059071729957805,
+      "grad_norm": 0.5445119738578796,
+      "learning_rate": 0.00015834797891036909,
+      "loss": 0.908163845539093,
+      "step": 902
+    },
+    {
+      "epoch": 0.38143459915611816,
+      "grad_norm": 0.4311858117580414,
+      "learning_rate": 0.00015869947275922672,
+      "loss": 0.8945821523666382,
+      "step": 904
+    },
+    {
+      "epoch": 0.3822784810126582,
+      "grad_norm": 0.5590984225273132,
+      "learning_rate": 0.00015905096660808436,
+      "loss": 0.9478458762168884,
+      "step": 906
+    },
+    {
+      "epoch": 0.3831223628691983,
+      "grad_norm": 0.5470241904258728,
+      "learning_rate": 0.000159402460456942,
+      "loss": 0.9259957671165466,
+      "step": 908
+    },
+    {
+      "epoch": 0.38396624472573837,
+      "grad_norm": 0.5498791337013245,
+      "learning_rate": 0.00015975395430579964,
+      "loss": 0.8824930787086487,
+      "step": 910
+    },
+    {
+      "epoch": 0.3848101265822785,
+      "grad_norm": 0.4779198467731476,
+      "learning_rate": 0.0001601054481546573,
+      "loss": 0.8842340707778931,
+      "step": 912
+    },
+    {
+      "epoch": 0.3856540084388186,
+      "grad_norm": 0.5390620827674866,
+      "learning_rate": 0.00016045694200351494,
+      "loss": 0.92950040102005,
+      "step": 914
+    },
+    {
+      "epoch": 0.38649789029535864,
+      "grad_norm": 0.505519688129425,
+      "learning_rate": 0.0001608084358523726,
+      "loss": 0.8420897126197815,
+      "step": 916
+    },
+    {
+      "epoch": 0.38734177215189874,
+      "grad_norm": 0.4463907778263092,
+      "learning_rate": 0.00016115992970123025,
+      "loss": 0.8256624341011047,
+      "step": 918
+    },
+    {
+      "epoch": 0.3881856540084388,
+      "grad_norm": 0.5321422219276428,
+      "learning_rate": 0.00016151142355008788,
+      "loss": 0.8701168298721313,
+      "step": 920
+    },
+    {
+      "epoch": 0.3890295358649789,
+      "grad_norm": 0.4343073070049286,
+      "learning_rate": 0.00016186291739894552,
+      "loss": 0.9019309878349304,
+      "step": 922
+    },
+    {
+      "epoch": 0.389873417721519,
+      "grad_norm": 0.5311984419822693,
+      "learning_rate": 0.00016221441124780316,
+      "loss": 0.8560551404953003,
+      "step": 924
+    },
+    {
+      "epoch": 0.39071729957805906,
+      "grad_norm": 0.588691771030426,
+      "learning_rate": 0.0001625659050966608,
+      "loss": 0.9700050354003906,
+      "step": 926
+    },
+    {
+      "epoch": 0.39156118143459917,
+      "grad_norm": 0.5425586104393005,
+      "learning_rate": 0.00016291739894551844,
+      "loss": 1.043768048286438,
+      "step": 928
+    },
+    {
+      "epoch": 0.3924050632911392,
+      "grad_norm": 0.5228736996650696,
+      "learning_rate": 0.0001632688927943761,
+      "loss": 0.9501712918281555,
+      "step": 930
+    },
+    {
+      "epoch": 0.39324894514767933,
+      "grad_norm": 0.48960360884666443,
+      "learning_rate": 0.00016362038664323377,
+      "loss": 0.9223058223724365,
+      "step": 932
+    },
+    {
+      "epoch": 0.39409282700421944,
+      "grad_norm": 0.45204755663871765,
+      "learning_rate": 0.0001639718804920914,
+      "loss": 0.9692960977554321,
+      "step": 934
+    },
+    {
+      "epoch": 0.3949367088607595,
+      "grad_norm": 0.5299274921417236,
+      "learning_rate": 0.00016432337434094905,
+      "loss": 0.9467466473579407,
+      "step": 936
+    },
+    {
+      "epoch": 0.3957805907172996,
+      "grad_norm": 0.5607715249061584,
+      "learning_rate": 0.00016467486818980668,
+      "loss": 0.9118053317070007,
+      "step": 938
+    },
+    {
+      "epoch": 0.39662447257383965,
+      "grad_norm": 0.5271831154823303,
+      "learning_rate": 0.00016502636203866432,
+      "loss": 0.9131460189819336,
+      "step": 940
+    },
+    {
+      "epoch": 0.39746835443037976,
+      "grad_norm": 0.5075286030769348,
+      "learning_rate": 0.00016537785588752196,
+      "loss": 0.9358300566673279,
+      "step": 942
+    },
+    {
+      "epoch": 0.3983122362869198,
+      "grad_norm": 0.515731155872345,
+      "learning_rate": 0.00016572934973637963,
+      "loss": 0.8908210396766663,
+      "step": 944
+    },
+    {
+      "epoch": 0.3991561181434599,
+      "grad_norm": 0.4856977164745331,
+      "learning_rate": 0.00016608084358523726,
+      "loss": 0.9775290489196777,
+      "step": 946
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.48846355080604553,
+      "learning_rate": 0.00016643233743409493,
+      "loss": 0.8957490921020508,
+      "step": 948
+    },
+    {
+      "epoch": 0.4008438818565401,
+      "grad_norm": 0.42990800738334656,
+      "learning_rate": 0.00016678383128295257,
+      "loss": 0.9036174416542053,
+      "step": 950
+    },
+    {
+      "epoch": 0.4016877637130802,
+      "grad_norm": 0.49552062153816223,
+      "learning_rate": 0.0001671353251318102,
+      "loss": 0.991032600402832,
+      "step": 952
+    },
+    {
+      "epoch": 0.40253164556962023,
+      "grad_norm": 0.4565040171146393,
+      "learning_rate": 0.00016748681898066784,
+      "loss": 0.823063313961029,
+      "step": 954
+    },
+    {
+      "epoch": 0.40337552742616034,
+      "grad_norm": 0.4290153682231903,
+      "learning_rate": 0.00016783831282952548,
+      "loss": 0.8785063624382019,
+      "step": 956
+    },
+    {
+      "epoch": 0.40421940928270045,
+      "grad_norm": 0.5419702529907227,
+      "learning_rate": 0.00016818980667838312,
+      "loss": 0.8763971924781799,
+      "step": 958
+    },
+    {
+      "epoch": 0.4050632911392405,
+      "grad_norm": 0.5177501440048218,
+      "learning_rate": 0.00016854130052724079,
+      "loss": 0.9470553398132324,
+      "step": 960
+    },
+    {
+      "epoch": 0.4059071729957806,
+      "grad_norm": 0.539725661277771,
+      "learning_rate": 0.00016889279437609842,
+      "loss": 0.9235025644302368,
+      "step": 962
+    },
+    {
+      "epoch": 0.40675105485232066,
+      "grad_norm": 0.5324983596801758,
+      "learning_rate": 0.0001692442882249561,
+      "loss": 1.0248996019363403,
+      "step": 964
+    },
+    {
+      "epoch": 0.40759493670886077,
+      "grad_norm": 0.4936407506465912,
+      "learning_rate": 0.00016959578207381373,
+      "loss": 0.9076873660087585,
+      "step": 966
+    },
+    {
+      "epoch": 0.4084388185654008,
+      "grad_norm": 0.4960501194000244,
+      "learning_rate": 0.00016994727592267137,
+      "loss": 0.9162673950195312,
+      "step": 968
+    },
+    {
+      "epoch": 0.4092827004219409,
+      "grad_norm": 0.45093682408332825,
+      "learning_rate": 0.000170298769771529,
+      "loss": 0.904100775718689,
+      "step": 970
+    },
+    {
+      "epoch": 0.41012658227848103,
+      "grad_norm": 0.4560275077819824,
+      "learning_rate": 0.00017065026362038664,
+      "loss": 0.8633337020874023,
+      "step": 972
+    },
+    {
+      "epoch": 0.4109704641350211,
+      "grad_norm": 0.44885000586509705,
+      "learning_rate": 0.00017100175746924428,
+      "loss": 0.8454209566116333,
+      "step": 974
+    },
+    {
+      "epoch": 0.4118143459915612,
+      "grad_norm": 0.47251659631729126,
+      "learning_rate": 0.00017135325131810195,
+      "loss": 0.824730396270752,
+      "step": 976
+    },
+    {
+      "epoch": 0.41265822784810124,
+      "grad_norm": 0.6597666144371033,
+      "learning_rate": 0.00017170474516695959,
+      "loss": 0.9496501684188843,
+      "step": 978
+    },
+    {
+      "epoch": 0.41350210970464135,
+      "grad_norm": 0.39806297421455383,
+      "learning_rate": 0.00017205623901581722,
+      "loss": 0.9419087171554565,
+      "step": 980
+    },
+    {
+      "epoch": 0.41434599156118146,
+      "grad_norm": 0.48231109976768494,
+      "learning_rate": 0.0001724077328646749,
+      "loss": 0.9182976484298706,
+      "step": 982
+    },
+    {
+      "epoch": 0.4151898734177215,
+      "grad_norm": 0.5438776612281799,
+      "learning_rate": 0.00017275922671353253,
+      "loss": 0.9386967420578003,
+      "step": 984
+    },
+    {
+      "epoch": 0.4160337552742616,
+      "grad_norm": 0.4959667921066284,
+      "learning_rate": 0.00017311072056239017,
+      "loss": 0.897849440574646,
+      "step": 986
+    },
+    {
+      "epoch": 0.41687763713080167,
+      "grad_norm": 0.43533357977867126,
+      "learning_rate": 0.0001734622144112478,
+      "loss": 0.8776953816413879,
+      "step": 988
+    },
+    {
+      "epoch": 0.4177215189873418,
+      "grad_norm": 0.47513946890830994,
+      "learning_rate": 0.00017381370826010547,
+      "loss": 0.9162989854812622,
+      "step": 990
+    },
+    {
+      "epoch": 0.41856540084388183,
+      "grad_norm": 0.4907188415527344,
+      "learning_rate": 0.0001741652021089631,
+      "loss": 0.9482660889625549,
+      "step": 992
+    },
+    {
+      "epoch": 0.41940928270042194,
+      "grad_norm": 0.44499966502189636,
+      "learning_rate": 0.00017451669595782075,
+      "loss": 0.8812930583953857,
+      "step": 994
+    },
+    {
+      "epoch": 0.42025316455696204,
+      "grad_norm": 0.4535730481147766,
+      "learning_rate": 0.00017486818980667838,
+      "loss": 0.9439874887466431,
+      "step": 996
+    },
+    {
+      "epoch": 0.4210970464135021,
+      "grad_norm": 0.5240745544433594,
+      "learning_rate": 0.00017521968365553602,
+      "loss": 0.8818395137786865,
+      "step": 998
+    },
+    {
+      "epoch": 0.4219409282700422,
+      "grad_norm": 0.5301211476325989,
+      "learning_rate": 0.0001755711775043937,
+      "loss": 0.886186957359314,
+      "step": 1000
+    },
+    {
+      "epoch": 0.4219409282700422,
+      "eval_loss": 0.9487298727035522,
+      "eval_runtime": 689.4288,
+      "eval_samples_per_second": 3.056,
+      "eval_steps_per_second": 3.056,
+      "step": 1000
+    },
+    {
+      "epoch": 0.42278481012658226,
+      "grad_norm": 0.47876957058906555,
+      "learning_rate": 0.00017592267135325133,
+      "loss": 0.8814021348953247,
+      "step": 1002
+    },
+    {
+      "epoch": 0.42362869198312236,
+      "grad_norm": 0.4929780960083008,
+      "learning_rate": 0.00017627416520210896,
+      "loss": 0.8295068740844727,
+      "step": 1004
+    },
+    {
+      "epoch": 0.42447257383966247,
+      "grad_norm": 0.41888436675071716,
+      "learning_rate": 0.00017662565905096663,
+      "loss": 0.8364827036857605,
+      "step": 1006
+    },
+    {
+      "epoch": 0.4253164556962025,
+      "grad_norm": 0.5175151824951172,
+      "learning_rate": 0.00017697715289982427,
+      "loss": 1.0167189836502075,
+      "step": 1008
+    },
+    {
+      "epoch": 0.42616033755274263,
+      "grad_norm": 0.4815356135368347,
+      "learning_rate": 0.0001773286467486819,
+      "loss": 0.8460752964019775,
+      "step": 1010
+    },
+    {
+      "epoch": 0.4270042194092827,
+      "grad_norm": 0.5210875272750854,
+      "learning_rate": 0.00017768014059753954,
+      "loss": 0.8595574498176575,
+      "step": 1012
+    },
+    {
+      "epoch": 0.4278481012658228,
+      "grad_norm": 0.4400486946105957,
+      "learning_rate": 0.00017803163444639718,
+      "loss": 0.8764723539352417,
+      "step": 1014
+    },
+    {
+      "epoch": 0.4286919831223629,
+      "grad_norm": 0.5282127857208252,
+      "learning_rate": 0.00017838312829525482,
+      "loss": 0.9706798791885376,
+      "step": 1016
+    },
+    {
+      "epoch": 0.42953586497890295,
+      "grad_norm": 0.47523441910743713,
+      "learning_rate": 0.00017873462214411249,
+      "loss": 0.8912002444267273,
+      "step": 1018
+    },
+    {
+      "epoch": 0.43037974683544306,
+      "grad_norm": 0.49640706181526184,
+      "learning_rate": 0.00017908611599297012,
+      "loss": 0.8835636973381042,
+      "step": 1020
+    },
+    {
+      "epoch": 0.4312236286919831,
+      "grad_norm": 0.5253039002418518,
+      "learning_rate": 0.0001794376098418278,
+      "loss": 0.8711735606193542,
+      "step": 1022
+    },
+    {
+      "epoch": 0.4320675105485232,
+      "grad_norm": 0.49285009503364563,
+      "learning_rate": 0.00017978910369068543,
+      "loss": 0.9064869284629822,
+      "step": 1024
+    },
+    {
+      "epoch": 0.43291139240506327,
+      "grad_norm": 0.48758041858673096,
+      "learning_rate": 0.00018014059753954307,
+      "loss": 0.8993359208106995,
+      "step": 1026
+    },
+    {
+      "epoch": 0.4337552742616034,
+      "grad_norm": 0.4815461039543152,
+      "learning_rate": 0.0001804920913884007,
+      "loss": 0.8747937679290771,
+      "step": 1028
+    },
+    {
+      "epoch": 0.4345991561181435,
+      "grad_norm": 0.4930349886417389,
+      "learning_rate": 0.00018084358523725834,
+      "loss": 0.8879084587097168,
+      "step": 1030
+    },
+    {
+      "epoch": 0.43544303797468353,
+      "grad_norm": 0.49303027987480164,
+      "learning_rate": 0.00018119507908611598,
+      "loss": 1.0023083686828613,
+      "step": 1032
+    },
+    {
+      "epoch": 0.43628691983122364,
+      "grad_norm": 0.5312249064445496,
+      "learning_rate": 0.00018154657293497365,
+      "loss": 0.938680112361908,
+      "step": 1034
+    },
+    {
+      "epoch": 0.4371308016877637,
+      "grad_norm": 0.5174582004547119,
+      "learning_rate": 0.0001818980667838313,
+      "loss": 0.854195773601532,
+      "step": 1036
+    },
+    {
+      "epoch": 0.4379746835443038,
+      "grad_norm": 0.5452545881271362,
+      "learning_rate": 0.00018224956063268895,
+      "loss": 0.9059375524520874,
+      "step": 1038
+    },
+    {
+      "epoch": 0.4388185654008439,
+      "grad_norm": 0.5480839014053345,
+      "learning_rate": 0.0001826010544815466,
+      "loss": 0.9708920121192932,
+      "step": 1040
+    },
+    {
+      "epoch": 0.43966244725738396,
+      "grad_norm": 0.44693151116371155,
+      "learning_rate": 0.00018295254833040423,
+      "loss": 0.8561046123504639,
+      "step": 1042
+    },
+    {
+      "epoch": 0.44050632911392407,
+      "grad_norm": 0.4024234116077423,
+      "learning_rate": 0.00018330404217926187,
+      "loss": 0.8811968564987183,
+      "step": 1044
+    },
+    {
+      "epoch": 0.4413502109704641,
+      "grad_norm": 0.5586408376693726,
+      "learning_rate": 0.0001836555360281195,
+      "loss": 0.978068470954895,
+      "step": 1046
+    },
+    {
+      "epoch": 0.4421940928270042,
+      "grad_norm": 0.4918624758720398,
+      "learning_rate": 0.00018400702987697714,
+      "loss": 0.8640981316566467,
+      "step": 1048
+    },
+    {
+      "epoch": 0.4430379746835443,
+      "grad_norm": 0.4335230588912964,
+      "learning_rate": 0.0001843585237258348,
+      "loss": 0.9077964425086975,
+      "step": 1050
+    },
+    {
+      "epoch": 0.4438818565400844,
+      "grad_norm": 0.5275123715400696,
+      "learning_rate": 0.00018471001757469247,
+      "loss": 0.960682213306427,
+      "step": 1052
+    },
+    {
+      "epoch": 0.4447257383966245,
+      "grad_norm": 0.5144415497779846,
+      "learning_rate": 0.0001850615114235501,
+      "loss": 0.8549577593803406,
+      "step": 1054
+    },
+    {
+      "epoch": 0.44556962025316454,
+      "grad_norm": 0.5166662931442261,
+      "learning_rate": 0.00018541300527240775,
+      "loss": 0.9320827126502991,
+      "step": 1056
+    },
+    {
+      "epoch": 0.44641350210970465,
+      "grad_norm": 0.5011980533599854,
+      "learning_rate": 0.0001857644991212654,
+      "loss": 0.9632431864738464,
+      "step": 1058
+    },
+    {
+      "epoch": 0.4472573839662447,
+      "grad_norm": 0.49312469363212585,
+      "learning_rate": 0.00018611599297012303,
+      "loss": 0.9442946910858154,
+      "step": 1060
+    },
+    {
+      "epoch": 0.4481012658227848,
+      "grad_norm": 0.49958568811416626,
+      "learning_rate": 0.00018646748681898066,
+      "loss": 0.952802300453186,
+      "step": 1062
+    },
+    {
+      "epoch": 0.4489451476793249,
+      "grad_norm": 0.49827462434768677,
+      "learning_rate": 0.00018681898066783833,
+      "loss": 0.9630650877952576,
+      "step": 1064
+    },
+    {
+      "epoch": 0.44978902953586497,
+      "grad_norm": 0.523980438709259,
+      "learning_rate": 0.00018717047451669597,
+      "loss": 0.9081395268440247,
+      "step": 1066
+    },
+    {
+      "epoch": 0.4506329113924051,
+      "grad_norm": 0.5108568668365479,
+      "learning_rate": 0.0001875219683655536,
+      "loss": 0.9398958683013916,
+      "step": 1068
+    },
+    {
+      "epoch": 0.45147679324894513,
+      "grad_norm": 0.4453965127468109,
+      "learning_rate": 0.00018787346221441127,
+      "loss": 0.93592768907547,
+      "step": 1070
+    },
+    {
+      "epoch": 0.45232067510548524,
+      "grad_norm": 0.4675683081150055,
+      "learning_rate": 0.0001882249560632689,
+      "loss": 0.8879633545875549,
+      "step": 1072
+    },
+    {
+      "epoch": 0.4531645569620253,
+      "grad_norm": 0.47398847341537476,
+      "learning_rate": 0.00018857644991212655,
+      "loss": 0.9966004490852356,
+      "step": 1074
+    },
+    {
+      "epoch": 0.4540084388185654,
+      "grad_norm": 0.46053192019462585,
+      "learning_rate": 0.0001889279437609842,
+      "loss": 0.9184179902076721,
+      "step": 1076
+    },
+    {
+      "epoch": 0.4548523206751055,
+      "grad_norm": 0.5601398348808289,
+      "learning_rate": 0.00018927943760984182,
+      "loss": 0.9434974193572998,
+      "step": 1078
+    },
+    {
+      "epoch": 0.45569620253164556,
+      "grad_norm": 0.48422637581825256,
+      "learning_rate": 0.0001896309314586995,
+      "loss": 0.9522465467453003,
+      "step": 1080
+    },
+    {
+      "epoch": 0.45654008438818566,
+      "grad_norm": 0.46280911564826965,
+      "learning_rate": 0.00018998242530755713,
+      "loss": 0.9315434694290161,
+      "step": 1082
+    },
+    {
+      "epoch": 0.4573839662447257,
+      "grad_norm": 0.4658683240413666,
+      "learning_rate": 0.00019033391915641477,
+      "loss": 0.8165783286094666,
+      "step": 1084
+    },
+    {
+      "epoch": 0.4582278481012658,
+      "grad_norm": 0.4192182123661041,
+      "learning_rate": 0.0001906854130052724,
+      "loss": 0.8024274110794067,
+      "step": 1086
+    },
+    {
+      "epoch": 0.45907172995780593,
+      "grad_norm": 0.4178735613822937,
+      "learning_rate": 0.00019103690685413007,
+      "loss": 0.8620653748512268,
+      "step": 1088
+    },
+    {
+      "epoch": 0.459915611814346,
+      "grad_norm": 0.4488574266433716,
+      "learning_rate": 0.0001913884007029877,
+      "loss": 0.9121530055999756,
+      "step": 1090
+    },
+    {
+      "epoch": 0.4607594936708861,
+      "grad_norm": 0.5164965987205505,
+      "learning_rate": 0.00019173989455184535,
+      "loss": 0.9496700763702393,
+      "step": 1092
+    },
+    {
+      "epoch": 0.46160337552742614,
+      "grad_norm": 0.41563132405281067,
+      "learning_rate": 0.00019209138840070299,
+      "loss": 0.8757708668708801,
+      "step": 1094
+    },
+    {
+      "epoch": 0.46244725738396625,
+      "grad_norm": 2.3222429752349854,
+      "learning_rate": 0.00019244288224956065,
+      "loss": 0.8406533002853394,
+      "step": 1096
+    },
+    {
+      "epoch": 0.46329113924050636,
+      "grad_norm": 0.4813845157623291,
+      "learning_rate": 0.0001927943760984183,
+      "loss": 0.9459465742111206,
+      "step": 1098
+    },
+    {
+      "epoch": 0.4641350210970464,
+      "grad_norm": 0.6233882308006287,
+      "learning_rate": 0.00019314586994727593,
+      "loss": 0.9376904368400574,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4641350210970464,
+      "eval_loss": 0.9357889294624329,
+      "eval_runtime": 676.9573,
+      "eval_samples_per_second": 3.112,
+      "eval_steps_per_second": 3.112,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4649789029535865,
+      "grad_norm": 0.5125579237937927,
+      "learning_rate": 0.00019349736379613357,
+      "loss": 0.8998825550079346,
+      "step": 1102
+    },
+    {
+      "epoch": 0.46582278481012657,
+      "grad_norm": 0.4534320831298828,
+      "learning_rate": 0.00019384885764499123,
+      "loss": 0.9213768839836121,
+      "step": 1104
+    },
+    {
+      "epoch": 0.4666666666666667,
+      "grad_norm": 0.4715143144130707,
+      "learning_rate": 0.00019420035149384887,
+      "loss": 0.9739661812782288,
+      "step": 1106
+    },
+    {
+      "epoch": 0.4675105485232067,
+      "grad_norm": 0.45529672503471375,
+      "learning_rate": 0.0001945518453427065,
+      "loss": 0.9113216400146484,
+      "step": 1108
+    },
+    {
+      "epoch": 0.46835443037974683,
+      "grad_norm": 0.48199015855789185,
+      "learning_rate": 0.00019490333919156417,
+      "loss": 0.92528235912323,
+      "step": 1110
+    },
+    {
+      "epoch": 0.46919831223628694,
+      "grad_norm": 0.4425188899040222,
+      "learning_rate": 0.0001952548330404218,
+      "loss": 0.8612716197967529,
+      "step": 1112
+    },
+    {
+      "epoch": 0.470042194092827,
+      "grad_norm": 0.4311593770980835,
+      "learning_rate": 0.00019560632688927945,
+      "loss": 0.8901699185371399,
+      "step": 1114
+    },
+    {
+      "epoch": 0.4708860759493671,
+      "grad_norm": 0.497806578874588,
+      "learning_rate": 0.0001959578207381371,
+      "loss": 0.9271994829177856,
+      "step": 1116
+    },
+    {
+      "epoch": 0.47172995780590715,
+      "grad_norm": 0.47149473428726196,
+      "learning_rate": 0.00019630931458699473,
+      "loss": 0.8740925788879395,
+      "step": 1118
+    },
+    {
+      "epoch": 0.47257383966244726,
+      "grad_norm": 0.5029966235160828,
+      "learning_rate": 0.00019666080843585236,
+      "loss": 0.9190115928649902,
+      "step": 1120
+    },
+    {
+      "epoch": 0.47341772151898737,
+      "grad_norm": 0.47972601652145386,
+      "learning_rate": 0.00019701230228471003,
+      "loss": 1.0043057203292847,
+      "step": 1122
+    },
+    {
+      "epoch": 0.4742616033755274,
+      "grad_norm": 0.45096471905708313,
+      "learning_rate": 0.00019736379613356767,
+      "loss": 0.9472925066947937,
+      "step": 1124
+    },
+    {
+      "epoch": 0.4751054852320675,
+      "grad_norm": 0.459852933883667,
+      "learning_rate": 0.00019771528998242533,
+      "loss": 0.8043124079704285,
+      "step": 1126
+    },
+    {
+      "epoch": 0.4759493670886076,
+      "grad_norm": 0.5313422679901123,
+      "learning_rate": 0.00019806678383128297,
+      "loss": 0.9662142395973206,
+      "step": 1128
+    },
+    {
+      "epoch": 0.4767932489451477,
+      "grad_norm": 0.43474531173706055,
+      "learning_rate": 0.0001984182776801406,
+      "loss": 0.8691151738166809,
+      "step": 1130
+    },
+    {
+      "epoch": 0.47763713080168774,
+      "grad_norm": 0.5232312083244324,
+      "learning_rate": 0.00019876977152899825,
+      "loss": 0.9024254083633423,
+      "step": 1132
+    },
+    {
+      "epoch": 0.47848101265822784,
+      "grad_norm": 0.47671905159950256,
+      "learning_rate": 0.0001991212653778559,
+      "loss": 0.8996873497962952,
+      "step": 1134
+    },
+    {
+      "epoch": 0.47932489451476795,
+      "grad_norm": 0.42289480566978455,
+      "learning_rate": 0.00019947275922671353,
+      "loss": 0.8336917757987976,
+      "step": 1136
+    },
+    {
+      "epoch": 0.480168776371308,
+      "grad_norm": 0.4700844883918762,
+      "learning_rate": 0.0001998242530755712,
+      "loss": 0.8943206071853638,
+      "step": 1138
+    },
+    {
+      "epoch": 0.4810126582278481,
+      "grad_norm": 0.5341399312019348,
+      "learning_rate": 0.00019999999711649004,
+      "loss": 0.9340365529060364,
+      "step": 1140
+    },
+    {
+      "epoch": 0.48185654008438816,
+      "grad_norm": 0.46169522404670715,
+      "learning_rate": 0.00019999997404841123,
+      "loss": 0.8241778016090393,
+      "step": 1142
+    },
+    {
+      "epoch": 0.48270042194092827,
+      "grad_norm": 0.5475223064422607,
+      "learning_rate": 0.00019999992791225896,
+      "loss": 1.0096158981323242,
+      "step": 1144
+    },
+    {
+      "epoch": 0.4835443037974684,
+      "grad_norm": 0.524641215801239,
+      "learning_rate": 0.00019999985870804385,
+      "loss": 0.9650378227233887,
+      "step": 1146
+    },
+    {
+      "epoch": 0.48438818565400843,
+      "grad_norm": 0.4326174855232239,
+      "learning_rate": 0.00019999976643578186,
+      "loss": 0.9003009796142578,
+      "step": 1148
+    },
+    {
+      "epoch": 0.48523206751054854,
+      "grad_norm": 0.49034059047698975,
+      "learning_rate": 0.0001999996510954943,
+      "loss": 0.9793432950973511,
+      "step": 1150
+    },
+    {
+      "epoch": 0.4860759493670886,
+      "grad_norm": 0.4760092496871948,
+      "learning_rate": 0.00019999951268720776,
+      "loss": 0.8793007731437683,
+      "step": 1152
+    },
+    {
+      "epoch": 0.4869198312236287,
+      "grad_norm": 0.44606879353523254,
+      "learning_rate": 0.00019999935121095417,
+      "loss": 0.888630747795105,
+      "step": 1154
+    },
+    {
+      "epoch": 0.4877637130801688,
+      "grad_norm": 0.5030332207679749,
+      "learning_rate": 0.0001999991666667708,
+      "loss": 0.9660000801086426,
+      "step": 1156
+    },
+    {
+      "epoch": 0.48860759493670886,
+      "grad_norm": 0.4295555055141449,
+      "learning_rate": 0.00019999895905470014,
+      "loss": 0.824654757976532,
+      "step": 1158
+    },
+    {
+      "epoch": 0.48945147679324896,
+      "grad_norm": 0.392167866230011,
+      "learning_rate": 0.0001999987283747902,
+      "loss": 0.8159562349319458,
+      "step": 1160
+    },
+    {
+      "epoch": 0.490295358649789,
+      "grad_norm": 0.484611839056015,
+      "learning_rate": 0.00019999847462709412,
+      "loss": 0.9630686044692993,
+      "step": 1162
+    },
+    {
+      "epoch": 0.4911392405063291,
+      "grad_norm": 0.45891445875167847,
+      "learning_rate": 0.00019999819781167042,
+      "loss": 0.8396129608154297,
+      "step": 1164
+    },
+    {
+      "epoch": 0.4919831223628692,
+      "grad_norm": 0.5553452968597412,
+      "learning_rate": 0.00019999789792858304,
+      "loss": 0.8194513916969299,
+      "step": 1166
+    },
+    {
+      "epoch": 0.4928270042194093,
+      "grad_norm": 0.38998672366142273,
+      "learning_rate": 0.00019999757497790106,
+      "loss": 0.9268721342086792,
+      "step": 1168
+    },
+    {
+      "epoch": 0.4936708860759494,
+      "grad_norm": 0.45445355772972107,
+      "learning_rate": 0.00019999722895969904,
+      "loss": 0.9204684495925903,
+      "step": 1170
+    },
+    {
+      "epoch": 0.49451476793248944,
+      "grad_norm": 0.484225332736969,
+      "learning_rate": 0.00019999685987405678,
+      "loss": 0.9270301461219788,
+      "step": 1172
+    },
+    {
+      "epoch": 0.49535864978902955,
+      "grad_norm": 0.45215415954589844,
+      "learning_rate": 0.00019999646772105942,
+      "loss": 0.8782645463943481,
+      "step": 1174
+    },
+    {
+      "epoch": 0.4962025316455696,
+      "grad_norm": 0.41578832268714905,
+      "learning_rate": 0.00019999605250079744,
+      "loss": 0.873112678527832,
+      "step": 1176
+    },
+    {
+      "epoch": 0.4970464135021097,
+      "grad_norm": 0.4632788300514221,
+      "learning_rate": 0.0001999956142133666,
+      "loss": 0.9593189358711243,
+      "step": 1178
+    },
+    {
+      "epoch": 0.4978902953586498,
+      "grad_norm": 0.5007622838020325,
+      "learning_rate": 0.000199995152858868,
+      "loss": 0.9215621948242188,
+      "step": 1180
+    },
+    {
+      "epoch": 0.49873417721518987,
+      "grad_norm": 0.5201935768127441,
+      "learning_rate": 0.0001999946684374081,
+      "loss": 0.8964219689369202,
+      "step": 1182
+    },
+    {
+      "epoch": 0.49957805907173,
+      "grad_norm": 0.5053967237472534,
+      "learning_rate": 0.0001999941609490986,
+      "loss": 0.8528663516044617,
+      "step": 1184
+    },
+    {
+      "epoch": 0.5004219409282701,
+      "grad_norm": 0.44397974014282227,
+      "learning_rate": 0.0001999936303940566,
+      "loss": 0.8214734792709351,
+      "step": 1186
+    },
+    {
+      "epoch": 0.5012658227848101,
+      "grad_norm": 0.4563101530075073,
+      "learning_rate": 0.0001999930767724045,
+      "loss": 0.8882166743278503,
+      "step": 1188
+    },
+    {
+      "epoch": 0.5021097046413502,
+      "grad_norm": 0.4666728675365448,
+      "learning_rate": 0.00019999250008426997,
+      "loss": 0.8882588148117065,
+      "step": 1190
+    },
+    {
+      "epoch": 0.5029535864978903,
+      "grad_norm": 0.4423629939556122,
+      "learning_rate": 0.00019999190032978607,
+      "loss": 0.8353691697120667,
+      "step": 1192
+    },
+    {
+      "epoch": 0.5037974683544304,
+      "grad_norm": 0.4514595866203308,
+      "learning_rate": 0.00019999127750909118,
+      "loss": 0.8721219301223755,
+      "step": 1194
+    },
+    {
+      "epoch": 0.5046413502109705,
+      "grad_norm": 0.48469996452331543,
+      "learning_rate": 0.00019999063162232886,
+      "loss": 0.902795135974884,
+      "step": 1196
+    },
+    {
+      "epoch": 0.5054852320675105,
+      "grad_norm": 0.4931983947753906,
+      "learning_rate": 0.00019998996266964823,
+      "loss": 0.8843175768852234,
+      "step": 1198
+    },
+    {
+      "epoch": 0.5063291139240507,
+      "grad_norm": 0.507408082485199,
+      "learning_rate": 0.00019998927065120357,
+      "loss": 0.9179208278656006,
+      "step": 1200
+    },
+    {
+      "epoch": 0.5063291139240507,
+      "eval_loss": 0.9224098324775696,
+      "eval_runtime": 669.7542,
+      "eval_samples_per_second": 3.146,
+      "eval_steps_per_second": 3.146,
+      "step": 1200
+    },
+    {
+      "epoch": 0.5071729957805907,
+      "grad_norm": 0.4024188220500946,
+      "learning_rate": 0.00019998855556715447,
+      "loss": 0.9078981876373291,
+      "step": 1202
+    },
+    {
+      "epoch": 0.5080168776371308,
+      "grad_norm": 0.6558667421340942,
+      "learning_rate": 0.00019998781741766594,
+      "loss": 0.8089252710342407,
+      "step": 1204
+    },
+    {
+      "epoch": 0.5088607594936709,
+      "grad_norm": 0.45128145813941956,
+      "learning_rate": 0.00019998705620290823,
+      "loss": 0.9316248297691345,
+      "step": 1206
+    },
+    {
+      "epoch": 0.509704641350211,
+      "grad_norm": 0.4733511507511139,
+      "learning_rate": 0.00019998627192305694,
+      "loss": 0.9047867655754089,
+      "step": 1208
+    },
+    {
+      "epoch": 0.510548523206751,
+      "grad_norm": 0.4026021659374237,
+      "learning_rate": 0.00019998546457829298,
+      "loss": 0.8507166504859924,
+      "step": 1210
+    },
+    {
+      "epoch": 0.5113924050632911,
+      "grad_norm": 0.43416184186935425,
+      "learning_rate": 0.00019998463416880262,
+      "loss": 0.8490090370178223,
+      "step": 1212
+    },
+    {
+      "epoch": 0.5122362869198313,
+      "grad_norm": 0.3678364157676697,
+      "learning_rate": 0.0001999837806947774,
+      "loss": 0.7730492949485779,
+      "step": 1214
+    },
+    {
+      "epoch": 0.5130801687763713,
+      "grad_norm": 0.49042677879333496,
+      "learning_rate": 0.00019998290415641415,
+      "loss": 0.9625269174575806,
+      "step": 1216
+    },
+    {
+      "epoch": 0.5139240506329114,
+      "grad_norm": 0.4576701819896698,
+      "learning_rate": 0.00019998200455391516,
+      "loss": 0.8961732983589172,
+      "step": 1218
+    },
+    {
+      "epoch": 0.5147679324894515,
+      "grad_norm": 0.4465518593788147,
+      "learning_rate": 0.00019998108188748793,
+      "loss": 0.8508996367454529,
+      "step": 1220
+    },
+    {
+      "epoch": 0.5156118143459916,
+      "grad_norm": 0.4267960488796234,
+      "learning_rate": 0.00019998013615734524,
+      "loss": 0.8664930462837219,
+      "step": 1222
+    },
+    {
+      "epoch": 0.5164556962025316,
+      "grad_norm": 0.39936602115631104,
+      "learning_rate": 0.0001999791673637053,
+      "loss": 0.8841317892074585,
+      "step": 1224
+    },
+    {
+      "epoch": 0.5172995780590718,
+      "grad_norm": 0.4442414343357086,
+      "learning_rate": 0.0001999781755067916,
+      "loss": 0.8601276874542236,
+      "step": 1226
+    },
+    {
+      "epoch": 0.5181434599156118,
+      "grad_norm": 0.4397028088569641,
+      "learning_rate": 0.00019997716058683292,
+      "loss": 0.8377046585083008,
+      "step": 1228
+    },
+    {
+      "epoch": 0.5189873417721519,
+      "grad_norm": 0.5327648520469666,
+      "learning_rate": 0.0001999761226040634,
+      "loss": 0.9853615164756775,
+      "step": 1230
+    },
+    {
+      "epoch": 0.5198312236286919,
+      "grad_norm": 0.40931421518325806,
+      "learning_rate": 0.00019997506155872244,
+      "loss": 0.9235715866088867,
+      "step": 1232
+    },
+    {
+      "epoch": 0.5206751054852321,
+      "grad_norm": 0.4599161148071289,
+      "learning_rate": 0.00019997397745105487,
+      "loss": 0.868396520614624,
+      "step": 1234
+    },
+    {
+      "epoch": 0.5215189873417722,
+      "grad_norm": 0.4813833236694336,
+      "learning_rate": 0.0001999728702813107,
+      "loss": 0.9946733117103577,
+      "step": 1236
+    },
+    {
+      "epoch": 0.5223628691983122,
+      "grad_norm": 0.43518269062042236,
+      "learning_rate": 0.00019997174004974543,
+      "loss": 0.8608635067939758,
+      "step": 1238
+    },
+    {
+      "epoch": 0.5232067510548524,
+      "grad_norm": 0.45646214485168457,
+      "learning_rate": 0.0001999705867566197,
+      "loss": 0.819354772567749,
+      "step": 1240
+    },
+    {
+      "epoch": 0.5240506329113924,
+      "grad_norm": 0.46834197640419006,
+      "learning_rate": 0.00019996941040219954,
+      "loss": 0.8934658765792847,
+      "step": 1242
+    },
+    {
+      "epoch": 0.5248945147679325,
+      "grad_norm": 0.46065032482147217,
+      "learning_rate": 0.0001999682109867564,
+      "loss": 0.8624778985977173,
+      "step": 1244
+    },
+    {
+      "epoch": 0.5257383966244725,
+      "grad_norm": 0.4506741166114807,
+      "learning_rate": 0.00019996698851056688,
+      "loss": 0.8540882468223572,
+      "step": 1246
+    },
+    {
+      "epoch": 0.5265822784810127,
+      "grad_norm": 0.4550519287586212,
+      "learning_rate": 0.00019996574297391302,
+      "loss": 0.9003006219863892,
+      "step": 1248
+    },
+    {
+      "epoch": 0.5274261603375527,
+      "grad_norm": 0.4097813367843628,
+      "learning_rate": 0.00019996447437708214,
+      "loss": 0.8490248918533325,
+      "step": 1250
+    },
+    {
+      "epoch": 0.5282700421940928,
+      "grad_norm": 0.4375711679458618,
+      "learning_rate": 0.00019996318272036688,
+      "loss": 0.9204569458961487,
+      "step": 1252
+    },
+    {
+      "epoch": 0.529113924050633,
+      "grad_norm": 0.4335198998451233,
+      "learning_rate": 0.00019996186800406518,
+      "loss": 0.8831789493560791,
+      "step": 1254
+    },
+    {
+      "epoch": 0.529957805907173,
+      "grad_norm": 0.4840933084487915,
+      "learning_rate": 0.00019996053022848035,
+      "loss": 0.8578064441680908,
+      "step": 1256
+    },
+    {
+      "epoch": 0.5308016877637131,
+      "grad_norm": 0.48911216855049133,
+      "learning_rate": 0.00019995916939392097,
+      "loss": 0.9235416650772095,
+      "step": 1258
+    },
+    {
+      "epoch": 0.5316455696202531,
+      "grad_norm": 0.41106730699539185,
+      "learning_rate": 0.00019995778550070098,
+      "loss": 0.7897764444351196,
+      "step": 1260
+    },
+    {
+      "epoch": 0.5324894514767933,
+      "grad_norm": 0.42233356833457947,
+      "learning_rate": 0.00019995637854913957,
+      "loss": 0.9392989873886108,
+      "step": 1262
+    },
+    {
+      "epoch": 0.5333333333333333,
+      "grad_norm": 0.39145180583000183,
+      "learning_rate": 0.0001999549485395613,
+      "loss": 0.9130091667175293,
+      "step": 1264
+    },
+    {
+      "epoch": 0.5341772151898734,
+      "grad_norm": 0.42780739068984985,
+      "learning_rate": 0.00019995349547229614,
+      "loss": 0.9690561890602112,
+      "step": 1266
+    },
+    {
+      "epoch": 0.5350210970464135,
+      "grad_norm": 0.4823327362537384,
+      "learning_rate": 0.0001999520193476792,
+      "loss": 0.9990131258964539,
+      "step": 1268
+    },
+    {
+      "epoch": 0.5358649789029536,
+      "grad_norm": 0.4669715166091919,
+      "learning_rate": 0.00019995052016605097,
+      "loss": 0.8343052864074707,
+      "step": 1270
+    },
+    {
+      "epoch": 0.5367088607594936,
+      "grad_norm": 0.4024743437767029,
+      "learning_rate": 0.0001999489979277574,
+      "loss": 0.9605970978736877,
+      "step": 1272
+    },
+    {
+      "epoch": 0.5375527426160338,
+      "grad_norm": 0.42063432931900024,
+      "learning_rate": 0.0001999474526331495,
+      "loss": 0.8164438009262085,
+      "step": 1274
+    },
+    {
+      "epoch": 0.5383966244725739,
+      "grad_norm": 0.47953495383262634,
+      "learning_rate": 0.00019994588428258383,
+      "loss": 0.879767656326294,
+      "step": 1276
+    },
+    {
+      "epoch": 0.5392405063291139,
+      "grad_norm": 0.39668557047843933,
+      "learning_rate": 0.00019994429287642214,
+      "loss": 0.7658726572990417,
+      "step": 1278
+    },
+    {
+      "epoch": 0.540084388185654,
+      "grad_norm": 0.5136121511459351,
+      "learning_rate": 0.00019994267841503155,
+      "loss": 0.9477384686470032,
+      "step": 1280
+    },
+    {
+      "epoch": 0.5409282700421941,
+      "grad_norm": 0.4039286673069,
+      "learning_rate": 0.00019994104089878452,
+      "loss": 0.7883896827697754,
+      "step": 1282
+    },
+    {
+      "epoch": 0.5417721518987342,
+      "grad_norm": 0.4425487816333771,
+      "learning_rate": 0.0001999393803280587,
+      "loss": 0.8526129126548767,
+      "step": 1284
+    },
+    {
+      "epoch": 0.5426160337552742,
+      "grad_norm": 0.4544559121131897,
+      "learning_rate": 0.00019993769670323725,
+      "loss": 0.8694683909416199,
+      "step": 1286
+    },
+    {
+      "epoch": 0.5434599156118144,
+      "grad_norm": 0.42872917652130127,
+      "learning_rate": 0.0001999359900247085,
+      "loss": 0.8211527466773987,
+      "step": 1288
+    },
+    {
+      "epoch": 0.5443037974683544,
+      "grad_norm": 0.41549673676490784,
+      "learning_rate": 0.00019993426029286617,
+      "loss": 0.7967932224273682,
+      "step": 1290
+    },
+    {
+      "epoch": 0.5451476793248945,
+      "grad_norm": 0.42888355255126953,
+      "learning_rate": 0.00019993250750810926,
+      "loss": 0.8291563987731934,
+      "step": 1292
+    },
+    {
+      "epoch": 0.5459915611814345,
+      "grad_norm": 0.44985485076904297,
+      "learning_rate": 0.0001999307316708421,
+      "loss": 0.8787803649902344,
+      "step": 1294
+    },
+    {
+      "epoch": 0.5468354430379747,
+      "grad_norm": 0.43326008319854736,
+      "learning_rate": 0.00019992893278147436,
+      "loss": 0.8375519514083862,
+      "step": 1296
+    },
+    {
+      "epoch": 0.5476793248945148,
+      "grad_norm": 0.406556099653244,
+      "learning_rate": 0.00019992711084042096,
+      "loss": 0.8538051247596741,
+      "step": 1298
+    },
+    {
+      "epoch": 0.5485232067510548,
+      "grad_norm": 0.46090877056121826,
+      "learning_rate": 0.00019992526584810226,
+      "loss": 0.9722132086753845,
+      "step": 1300
+    },
+    {
+      "epoch": 0.5485232067510548,
+      "eval_loss": 0.9068717360496521,
+      "eval_runtime": 680.7718,
+      "eval_samples_per_second": 3.095,
+      "eval_steps_per_second": 3.095,
+      "step": 1300
+    },
+    {
+      "epoch": 0.549367088607595,
+      "grad_norm": 0.44932207465171814,
+      "learning_rate": 0.00019992339780494382,
+      "loss": 0.8474152684211731,
+      "step": 1302
+    },
+    {
+      "epoch": 0.550210970464135,
+      "grad_norm": 0.5185276865959167,
+      "learning_rate": 0.0001999215067113766,
+      "loss": 0.8440775871276855,
+      "step": 1304
+    },
+    {
+      "epoch": 0.5510548523206751,
+      "grad_norm": 0.43811365962028503,
+      "learning_rate": 0.0001999195925678368,
+      "loss": 0.8848010897636414,
+      "step": 1306
+    },
+    {
+      "epoch": 0.5518987341772152,
+      "grad_norm": 0.4861660301685333,
+      "learning_rate": 0.00019991765537476596,
+      "loss": 0.8265100121498108,
+      "step": 1308
+    },
+    {
+      "epoch": 0.5527426160337553,
+      "grad_norm": 0.5068721771240234,
+      "learning_rate": 0.000199915695132611,
+      "loss": 0.9327963590621948,
+      "step": 1310
+    },
+    {
+      "epoch": 0.5535864978902953,
+      "grad_norm": 0.4902805685997009,
+      "learning_rate": 0.0001999137118418241,
+      "loss": 0.9315155744552612,
+      "step": 1312
+    },
+    {
+      "epoch": 0.5544303797468354,
+      "grad_norm": 0.4530964493751526,
+      "learning_rate": 0.00019991170550286272,
+      "loss": 0.9344852566719055,
+      "step": 1314
+    },
+    {
+      "epoch": 0.5552742616033756,
+      "grad_norm": 0.4106673300266266,
+      "learning_rate": 0.00019990967611618974,
+      "loss": 0.8326173424720764,
+      "step": 1316
+    },
+    {
+      "epoch": 0.5561181434599156,
+      "grad_norm": 0.4439711570739746,
+      "learning_rate": 0.0001999076236822733,
+      "loss": 0.8934909105300903,
+      "step": 1318
+    },
+    {
+      "epoch": 0.5569620253164557,
+      "grad_norm": 0.40245553851127625,
+      "learning_rate": 0.0001999055482015868,
+      "loss": 0.6691107153892517,
+      "step": 1320
+    },
+    {
+      "epoch": 0.5578059071729958,
+      "grad_norm": 0.43904435634613037,
+      "learning_rate": 0.0001999034496746091,
+      "loss": 0.8560307621955872,
+      "step": 1322
+    },
+    {
+      "epoch": 0.5586497890295359,
+      "grad_norm": 0.5131705403327942,
+      "learning_rate": 0.00019990132810182422,
+      "loss": 0.9024442434310913,
+      "step": 1324
+    },
+    {
+      "epoch": 0.5594936708860759,
+      "grad_norm": 0.4539531171321869,
+      "learning_rate": 0.00019989918348372156,
+      "loss": 0.8675428628921509,
+      "step": 1326
+    },
+    {
+      "epoch": 0.560337552742616,
+      "grad_norm": 0.44469985365867615,
+      "learning_rate": 0.00019989701582079591,
+      "loss": 0.8853429555892944,
+      "step": 1328
+    },
+    {
+      "epoch": 0.5611814345991561,
+      "grad_norm": 0.4485546946525574,
+      "learning_rate": 0.00019989482511354725,
+      "loss": 0.9675378799438477,
+      "step": 1330
+    },
+    {
+      "epoch": 0.5620253164556962,
+      "grad_norm": 0.43316468596458435,
+      "learning_rate": 0.00019989261136248097,
+      "loss": 0.8256655335426331,
+      "step": 1332
+    },
+    {
+      "epoch": 0.5628691983122363,
+      "grad_norm": 0.4315780699253082,
+      "learning_rate": 0.00019989037456810772,
+      "loss": 0.7727690935134888,
+      "step": 1334
+    },
+    {
+      "epoch": 0.5637130801687764,
+      "grad_norm": 0.45147502422332764,
+      "learning_rate": 0.00019988811473094348,
+      "loss": 0.9240618348121643,
+      "step": 1336
+    },
+    {
+      "epoch": 0.5645569620253165,
+      "grad_norm": 0.4606908857822418,
+      "learning_rate": 0.00019988583185150957,
+      "loss": 0.8005949258804321,
+      "step": 1338
+    },
+    {
+      "epoch": 0.5654008438818565,
+      "grad_norm": 0.48932451009750366,
+      "learning_rate": 0.00019988352593033255,
+      "loss": 0.9042718410491943,
+      "step": 1340
+    },
+    {
+      "epoch": 0.5662447257383966,
+      "grad_norm": 0.46022218465805054,
+      "learning_rate": 0.00019988119696794443,
+      "loss": 0.8724613785743713,
+      "step": 1342
+    },
+    {
+      "epoch": 0.5670886075949367,
+      "grad_norm": 0.4414621889591217,
+      "learning_rate": 0.00019987884496488242,
+      "loss": 0.8653066158294678,
+      "step": 1344
+    },
+    {
+      "epoch": 0.5679324894514768,
+      "grad_norm": 0.4490342140197754,
+      "learning_rate": 0.0001998764699216891,
+      "loss": 0.8843849301338196,
+      "step": 1346
+    },
+    {
+      "epoch": 0.5687763713080168,
+      "grad_norm": 0.47690069675445557,
+      "learning_rate": 0.00019987407183891232,
+      "loss": 0.8385393023490906,
+      "step": 1348
+    },
+    {
+      "epoch": 0.569620253164557,
+      "grad_norm": 0.41409361362457275,
+      "learning_rate": 0.00019987165071710527,
+      "loss": 0.77492755651474,
+      "step": 1350
+    },
+    {
+      "epoch": 0.570464135021097,
+      "grad_norm": 0.4151647984981537,
+      "learning_rate": 0.0001998692065568265,
+      "loss": 0.8368680477142334,
+      "step": 1352
+    },
+    {
+      "epoch": 0.5713080168776371,
+      "grad_norm": 0.4105067849159241,
+      "learning_rate": 0.0001998667393586398,
+      "loss": 0.833220362663269,
+      "step": 1354
+    },
+    {
+      "epoch": 0.5721518987341773,
+      "grad_norm": 0.44701820611953735,
+      "learning_rate": 0.0001998642491231143,
+      "loss": 0.8041552305221558,
+      "step": 1356
+    },
+    {
+      "epoch": 0.5729957805907173,
+      "grad_norm": 0.4621582329273224,
+      "learning_rate": 0.00019986173585082444,
+      "loss": 0.9052709937095642,
+      "step": 1358
+    },
+    {
+      "epoch": 0.5738396624472574,
+      "grad_norm": 0.4934713840484619,
+      "learning_rate": 0.00019985919954235005,
+      "loss": 0.8036054372787476,
+      "step": 1360
+    },
+    {
+      "epoch": 0.5746835443037974,
+      "grad_norm": 0.4557384252548218,
+      "learning_rate": 0.0001998566401982761,
+      "loss": 0.8340095281600952,
+      "step": 1362
+    },
+    {
+      "epoch": 0.5755274261603376,
+      "grad_norm": 0.5048403739929199,
+      "learning_rate": 0.00019985405781919305,
+      "loss": 0.8703633546829224,
+      "step": 1364
+    },
+    {
+      "epoch": 0.5763713080168776,
+      "grad_norm": 0.4707714319229126,
+      "learning_rate": 0.0001998514524056966,
+      "loss": 0.9115830659866333,
+      "step": 1366
+    },
+    {
+      "epoch": 0.5772151898734177,
+      "grad_norm": 0.4502186179161072,
+      "learning_rate": 0.00019984882395838777,
+      "loss": 0.8603149056434631,
+      "step": 1368
+    },
+    {
+      "epoch": 0.5780590717299579,
+      "grad_norm": 0.4229136109352112,
+      "learning_rate": 0.00019984617247787288,
+      "loss": 0.7482197284698486,
+      "step": 1370
+    },
+    {
+      "epoch": 0.5789029535864979,
+      "grad_norm": 0.49208030104637146,
+      "learning_rate": 0.0001998434979647636,
+      "loss": 0.9071239233016968,
+      "step": 1372
+    },
+    {
+      "epoch": 0.579746835443038,
+      "grad_norm": 0.47393161058425903,
+      "learning_rate": 0.00019984080041967687,
+      "loss": 0.9370644092559814,
+      "step": 1374
+    },
+    {
+      "epoch": 0.580590717299578,
+      "grad_norm": 0.4483324885368347,
+      "learning_rate": 0.00019983807984323492,
+      "loss": 0.8536180257797241,
+      "step": 1376
+    },
+    {
+      "epoch": 0.5814345991561182,
+      "grad_norm": 0.4163796007633209,
+      "learning_rate": 0.00019983533623606543,
+      "loss": 0.7360405325889587,
+      "step": 1378
+    },
+    {
+      "epoch": 0.5822784810126582,
+      "grad_norm": 0.44147446751594543,
+      "learning_rate": 0.00019983256959880122,
+      "loss": 0.8162824511528015,
+      "step": 1380
+    },
+    {
+      "epoch": 0.5831223628691983,
+      "grad_norm": 0.44443491101264954,
+      "learning_rate": 0.0001998297799320805,
+      "loss": 0.9169327616691589,
+      "step": 1382
+    },
+    {
+      "epoch": 0.5839662447257384,
+      "grad_norm": 0.4641949236392975,
+      "learning_rate": 0.00019982696723654686,
+      "loss": 0.8875693678855896,
+      "step": 1384
+    },
+    {
+      "epoch": 0.5848101265822785,
+      "grad_norm": 0.4327974319458008,
+      "learning_rate": 0.00019982413151284906,
+      "loss": 0.7747344970703125,
+      "step": 1386
+    },
+    {
+      "epoch": 0.5856540084388185,
+      "grad_norm": 0.44200772047042847,
+      "learning_rate": 0.0001998212727616413,
+      "loss": 0.8490481972694397,
+      "step": 1388
+    },
+    {
+      "epoch": 0.5864978902953587,
+      "grad_norm": 0.39084959030151367,
+      "learning_rate": 0.000199818390983583,
+      "loss": 0.8014808297157288,
+      "step": 1390
+    },
+    {
+      "epoch": 0.5873417721518988,
+      "grad_norm": 0.3691277801990509,
+      "learning_rate": 0.00019981548617933897,
+      "loss": 0.8650928139686584,
+      "step": 1392
+    },
+    {
+      "epoch": 0.5881856540084388,
+      "grad_norm": 0.480002760887146,
+      "learning_rate": 0.00019981255834957925,
+      "loss": 0.7713267803192139,
+      "step": 1394
+    },
+    {
+      "epoch": 0.5890295358649789,
+      "grad_norm": 0.4397226572036743,
+      "learning_rate": 0.00019980960749497927,
+      "loss": 0.9053656458854675,
+      "step": 1396
+    },
+    {
+      "epoch": 0.589873417721519,
+      "grad_norm": 0.4263727366924286,
+      "learning_rate": 0.00019980663361621973,
+      "loss": 0.8737669587135315,
+      "step": 1398
+    },
+    {
+      "epoch": 0.5907172995780591,
+      "grad_norm": 0.4468817710876465,
+      "learning_rate": 0.0001998036367139866,
+      "loss": 0.9396650791168213,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5907172995780591,
+      "eval_loss": 0.8971880674362183,
+      "eval_runtime": 692.8046,
+      "eval_samples_per_second": 3.041,
+      "eval_steps_per_second": 3.041,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5915611814345991,
+      "grad_norm": 0.4128144383430481,
+      "learning_rate": 0.00019980061678897127,
+      "loss": 0.926134467124939,
+      "step": 1402
+    },
+    {
+      "epoch": 0.5924050632911393,
+      "grad_norm": 0.45524629950523376,
+      "learning_rate": 0.00019979757384187035,
+      "loss": 0.826360285282135,
+      "step": 1404
+    },
+    {
+      "epoch": 0.5932489451476793,
+      "grad_norm": 0.42521336674690247,
+      "learning_rate": 0.0001997945078733858,
+      "loss": 0.8147702813148499,
+      "step": 1406
+    },
+    {
+      "epoch": 0.5940928270042194,
+      "grad_norm": 0.4172525107860565,
+      "learning_rate": 0.00019979141888422489,
+      "loss": 0.7711596488952637,
+      "step": 1408
+    },
+    {
+      "epoch": 0.5949367088607594,
+      "grad_norm": 0.46193623542785645,
+      "learning_rate": 0.00019978830687510013,
+      "loss": 0.7875980138778687,
+      "step": 1410
+    },
+    {
+      "epoch": 0.5957805907172996,
+      "grad_norm": 0.4455825686454773,
+      "learning_rate": 0.00019978517184672946,
+      "loss": 0.8590195178985596,
+      "step": 1412
+    },
+    {
+      "epoch": 0.5966244725738397,
+      "grad_norm": 0.4266505539417267,
+      "learning_rate": 0.00019978201379983608,
+      "loss": 0.9318227767944336,
+      "step": 1414
+    },
+    {
+      "epoch": 0.5974683544303797,
+      "grad_norm": 0.42627251148223877,
+      "learning_rate": 0.00019977883273514843,
+      "loss": 0.8644474744796753,
+      "step": 1416
+    },
+    {
+      "epoch": 0.5983122362869199,
+      "grad_norm": 0.393873393535614,
+      "learning_rate": 0.00019977562865340038,
+      "loss": 0.7760446071624756,
+      "step": 1418
+    },
+    {
+      "epoch": 0.5991561181434599,
+      "grad_norm": 0.44334208965301514,
+      "learning_rate": 0.000199772401555331,
+      "loss": 0.8071584701538086,
+      "step": 1420
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.39390063285827637,
+      "learning_rate": 0.0001997691514416848,
+      "loss": 0.9076889753341675,
+      "step": 1422
+    },
+    {
+      "epoch": 0.60084388185654,
+      "grad_norm": 0.46753421425819397,
+      "learning_rate": 0.00019976587831321144,
+      "loss": 0.8534318804740906,
+      "step": 1424
+    },
+    {
+      "epoch": 0.6016877637130802,
+      "grad_norm": 0.4840681254863739,
+      "learning_rate": 0.00019976258217066598,
+      "loss": 0.9049596190452576,
+      "step": 1426
+    },
+    {
+      "epoch": 0.6025316455696202,
+      "grad_norm": 0.4740557372570038,
+      "learning_rate": 0.00019975926301480878,
+      "loss": 0.8520918488502502,
+      "step": 1428
+    },
+    {
+      "epoch": 0.6033755274261603,
+      "grad_norm": 0.46498584747314453,
+      "learning_rate": 0.00019975592084640553,
+      "loss": 0.8221305012702942,
+      "step": 1430
+    },
+    {
+      "epoch": 0.6042194092827005,
+      "grad_norm": 0.40885692834854126,
+      "learning_rate": 0.0001997525556662272,
+      "loss": 0.8132198452949524,
+      "step": 1432
+    },
+    {
+      "epoch": 0.6050632911392405,
+      "grad_norm": 0.4375821053981781,
+      "learning_rate": 0.00019974916747505006,
+      "loss": 0.870534360408783,
+      "step": 1434
+    },
+    {
+      "epoch": 0.6059071729957806,
+      "grad_norm": 0.4780935049057007,
+      "learning_rate": 0.00019974575627365569,
+      "loss": 0.8825759291648865,
+      "step": 1436
+    },
+    {
+      "epoch": 0.6067510548523207,
+      "grad_norm": 0.4623970687389374,
+      "learning_rate": 0.00019974232206283098,
+      "loss": 0.8187641501426697,
+      "step": 1438
+    },
+    {
+      "epoch": 0.6075949367088608,
+      "grad_norm": 0.42172595858573914,
+      "learning_rate": 0.0001997388648433682,
+      "loss": 0.8569821119308472,
+      "step": 1440
+    },
+    {
+      "epoch": 0.6084388185654008,
+      "grad_norm": 0.49386659264564514,
+      "learning_rate": 0.0001997353846160648,
+      "loss": 0.9173614382743835,
+      "step": 1442
+    },
+    {
+      "epoch": 0.6092827004219409,
+      "grad_norm": 0.47122615575790405,
+      "learning_rate": 0.00019973188138172363,
+      "loss": 0.9444975256919861,
+      "step": 1444
+    },
+    {
+      "epoch": 0.610126582278481,
+      "grad_norm": 0.4831530451774597,
+      "learning_rate": 0.0001997283551411528,
+      "loss": 0.9537245631217957,
+      "step": 1446
+    },
+    {
+      "epoch": 0.6109704641350211,
+      "grad_norm": 0.44460317492485046,
+      "learning_rate": 0.0001997248058951658,
+      "loss": 0.8481367230415344,
+      "step": 1448
+    },
+    {
+      "epoch": 0.6118143459915611,
+      "grad_norm": 0.4315263628959656,
+      "learning_rate": 0.00019972123364458128,
+      "loss": 0.8572371006011963,
+      "step": 1450
+    },
+    {
+      "epoch": 0.6126582278481013,
+      "grad_norm": 0.38559582829475403,
+      "learning_rate": 0.00019971763839022336,
+      "loss": 0.762590765953064,
+      "step": 1452
+    },
+    {
+      "epoch": 0.6135021097046414,
+      "grad_norm": 0.3820290267467499,
+      "learning_rate": 0.00019971402013292138,
+      "loss": 0.7879500389099121,
+      "step": 1454
+    },
+    {
+      "epoch": 0.6143459915611814,
+      "grad_norm": 0.42509570717811584,
+      "learning_rate": 0.00019971037887351,
+      "loss": 0.8696863055229187,
+      "step": 1456
+    },
+    {
+      "epoch": 0.6151898734177215,
+      "grad_norm": 0.46750548481941223,
+      "learning_rate": 0.00019970671461282916,
+      "loss": 0.8425862789154053,
+      "step": 1458
+    },
+    {
+      "epoch": 0.6160337552742616,
+      "grad_norm": 0.3937121033668518,
+      "learning_rate": 0.0001997030273517242,
+      "loss": 0.8028931021690369,
+      "step": 1460
+    },
+    {
+      "epoch": 0.6168776371308017,
+      "grad_norm": 0.4361220896244049,
+      "learning_rate": 0.00019969931709104565,
+      "loss": 0.8662509918212891,
+      "step": 1462
+    },
+    {
+      "epoch": 0.6177215189873417,
+      "grad_norm": 0.4737963378429413,
+      "learning_rate": 0.00019969558383164943,
+      "loss": 0.8962596654891968,
+      "step": 1464
+    },
+    {
+      "epoch": 0.6185654008438819,
+      "grad_norm": 0.3952867388725281,
+      "learning_rate": 0.00019969182757439668,
+      "loss": 0.8630867600440979,
+      "step": 1466
+    },
+    {
+      "epoch": 0.619409282700422,
+      "grad_norm": 0.41688743233680725,
+      "learning_rate": 0.00019968804832015393,
+      "loss": 0.8759240508079529,
+      "step": 1468
+    },
+    {
+      "epoch": 0.620253164556962,
+      "grad_norm": 0.4599224627017975,
+      "learning_rate": 0.00019968424606979298,
+      "loss": 0.8459385633468628,
+      "step": 1470
+    },
+    {
+      "epoch": 0.6210970464135022,
+      "grad_norm": 0.5476765632629395,
+      "learning_rate": 0.00019968042082419094,
+      "loss": 0.8844659328460693,
+      "step": 1472
+    },
+    {
+      "epoch": 0.6219409282700422,
+      "grad_norm": 0.46202269196510315,
+      "learning_rate": 0.00019967657258423022,
+      "loss": 0.860946536064148,
+      "step": 1474
+    },
+    {
+      "epoch": 0.6227848101265823,
+      "grad_norm": 0.39446666836738586,
+      "learning_rate": 0.00019967270135079853,
+      "loss": 0.7587860822677612,
+      "step": 1476
+    },
+    {
+      "epoch": 0.6236286919831223,
+      "grad_norm": 0.4500375986099243,
+      "learning_rate": 0.0001996688071247889,
+      "loss": 0.9143000245094299,
+      "step": 1478
+    },
+    {
+      "epoch": 0.6244725738396625,
+      "grad_norm": 0.4203545153141022,
+      "learning_rate": 0.0001996648899070996,
+      "loss": 0.7865519523620605,
+      "step": 1480
+    },
+    {
+      "epoch": 0.6253164556962025,
+      "grad_norm": 0.45623889565467834,
+      "learning_rate": 0.00019966094969863432,
+      "loss": 0.9192912578582764,
+      "step": 1482
+    },
+    {
+      "epoch": 0.6261603375527426,
+      "grad_norm": 0.4139637053012848,
+      "learning_rate": 0.00019965698650030195,
+      "loss": 0.8459161520004272,
+      "step": 1484
+    },
+    {
+      "epoch": 0.6270042194092827,
+      "grad_norm": 0.4410824179649353,
+      "learning_rate": 0.00019965300031301678,
+      "loss": 0.8159077763557434,
+      "step": 1486
+    },
+    {
+      "epoch": 0.6278481012658228,
+      "grad_norm": 0.43821197748184204,
+      "learning_rate": 0.00019964899113769828,
+      "loss": 0.884467363357544,
+      "step": 1488
+    },
+    {
+      "epoch": 0.6286919831223629,
+      "grad_norm": 0.4692091643810272,
+      "learning_rate": 0.00019964495897527133,
+      "loss": 0.8693601489067078,
+      "step": 1490
+    },
+    {
+      "epoch": 0.6295358649789029,
+      "grad_norm": 0.40715447068214417,
+      "learning_rate": 0.00019964090382666608,
+      "loss": 0.8897743225097656,
+      "step": 1492
+    },
+    {
+      "epoch": 0.6303797468354431,
+      "grad_norm": 0.4123334586620331,
+      "learning_rate": 0.0001996368256928179,
+      "loss": 0.8274733424186707,
+      "step": 1494
+    },
+    {
+      "epoch": 0.6312236286919831,
+      "grad_norm": 0.44347211718559265,
+      "learning_rate": 0.00019963272457466767,
+      "loss": 0.8158749938011169,
+      "step": 1496
+    },
+    {
+      "epoch": 0.6320675105485232,
+      "grad_norm": 0.4700329601764679,
+      "learning_rate": 0.00019962860047316135,
+      "loss": 0.8439569473266602,
+      "step": 1498
+    },
+    {
+      "epoch": 0.6329113924050633,
+      "grad_norm": 0.46415844559669495,
+      "learning_rate": 0.00019962445338925027,
+      "loss": 0.8316822648048401,
+      "step": 1500
+    },
+    {
+      "epoch": 0.6329113924050633,
+      "eval_loss": 0.887488842010498,
+      "eval_runtime": 686.2804,
+      "eval_samples_per_second": 3.07,
+      "eval_steps_per_second": 3.07,
+      "step": 1500
+    },
+    {
+      "epoch": 0.6337552742616034,
+      "grad_norm": 0.44183167815208435,
+      "learning_rate": 0.0001996202833238911,
+      "loss": 0.9121994972229004,
+      "step": 1502
+    },
+    {
+      "epoch": 0.6345991561181434,
+      "grad_norm": 0.4279208779335022,
+      "learning_rate": 0.00019961609027804587,
+      "loss": 0.9160211086273193,
+      "step": 1504
+    },
+    {
+      "epoch": 0.6354430379746835,
+      "grad_norm": 0.3967169523239136,
+      "learning_rate": 0.00019961187425268176,
+      "loss": 0.861677348613739,
+      "step": 1506
+    },
+    {
+      "epoch": 0.6362869198312237,
+      "grad_norm": 0.4796451926231384,
+      "learning_rate": 0.0001996076352487713,
+      "loss": 0.8478423357009888,
+      "step": 1508
+    },
+    {
+      "epoch": 0.6371308016877637,
+      "grad_norm": 0.3861426115036011,
+      "learning_rate": 0.00019960337326729245,
+      "loss": 0.8739159107208252,
+      "step": 1510
+    },
+    {
+      "epoch": 0.6379746835443038,
+      "grad_norm": 0.4567820429801941,
+      "learning_rate": 0.00019959908830922824,
+      "loss": 0.7693920731544495,
+      "step": 1512
+    },
+    {
+      "epoch": 0.6388185654008439,
+      "grad_norm": 0.4457108676433563,
+      "learning_rate": 0.00019959478037556724,
+      "loss": 0.8482301831245422,
+      "step": 1514
+    },
+    {
+      "epoch": 0.639662447257384,
+      "grad_norm": 0.43918928503990173,
+      "learning_rate": 0.00019959044946730314,
+      "loss": 0.8626812696456909,
+      "step": 1516
+    },
+    {
+      "epoch": 0.640506329113924,
+      "grad_norm": 0.4095900058746338,
+      "learning_rate": 0.00019958609558543504,
+      "loss": 0.8342230916023254,
+      "step": 1518
+    },
+    {
+      "epoch": 0.6413502109704642,
+      "grad_norm": 0.4181270897388458,
+      "learning_rate": 0.00019958171873096724,
+      "loss": 0.8794118165969849,
+      "step": 1520
+    },
+    {
+      "epoch": 0.6421940928270042,
+      "grad_norm": 0.4750959277153015,
+      "learning_rate": 0.00019957731890490947,
+      "loss": 0.7787677049636841,
+      "step": 1522
+    },
+    {
+      "epoch": 0.6430379746835443,
+      "grad_norm": 0.5129296779632568,
+      "learning_rate": 0.00019957289610827663,
+      "loss": 0.9084368348121643,
+      "step": 1524
+    },
+    {
+      "epoch": 0.6438818565400843,
+      "grad_norm": 0.4829029142856598,
+      "learning_rate": 0.000199568450342089,
+      "loss": 0.9066952466964722,
+      "step": 1526
+    },
+    {
+      "epoch": 0.6447257383966245,
+      "grad_norm": 0.4225057363510132,
+      "learning_rate": 0.00019956398160737214,
+      "loss": 0.8127874732017517,
+      "step": 1528
+    },
+    {
+      "epoch": 0.6455696202531646,
+      "grad_norm": 0.4231826364994049,
+      "learning_rate": 0.00019955948990515684,
+      "loss": 0.8781921863555908,
+      "step": 1530
+    },
+    {
+      "epoch": 0.6464135021097046,
+      "grad_norm": 0.4436216950416565,
+      "learning_rate": 0.00019955497523647933,
+      "loss": 0.8136996626853943,
+      "step": 1532
+    },
+    {
+      "epoch": 0.6472573839662448,
+      "grad_norm": 0.4018244743347168,
+      "learning_rate": 0.00019955043760238098,
+      "loss": 0.8003877401351929,
+      "step": 1534
+    },
+    {
+      "epoch": 0.6481012658227848,
+      "grad_norm": 0.40024611353874207,
+      "learning_rate": 0.0001995458770039086,
+      "loss": 0.8996267914772034,
+      "step": 1536
+    },
+    {
+      "epoch": 0.6489451476793249,
+      "grad_norm": 0.42845702171325684,
+      "learning_rate": 0.00019954129344211424,
+      "loss": 0.8610522150993347,
+      "step": 1538
+    },
+    {
+      "epoch": 0.6497890295358649,
+      "grad_norm": 0.4258182644844055,
+      "learning_rate": 0.00019953668691805517,
+      "loss": 0.8991837501525879,
+      "step": 1540
+    },
+    {
+      "epoch": 0.6506329113924051,
+      "grad_norm": 0.4483257830142975,
+      "learning_rate": 0.0001995320574327941,
+      "loss": 0.8150189518928528,
+      "step": 1542
+    },
+    {
+      "epoch": 0.6514767932489451,
+      "grad_norm": 0.4320627748966217,
+      "learning_rate": 0.0001995274049873989,
+      "loss": 0.8572840690612793,
+      "step": 1544
+    },
+    {
+      "epoch": 0.6523206751054852,
+      "grad_norm": 0.39950594305992126,
+      "learning_rate": 0.00019952272958294288,
+      "loss": 0.7766129374504089,
+      "step": 1546
+    },
+    {
+      "epoch": 0.6531645569620254,
+      "grad_norm": 0.4130117893218994,
+      "learning_rate": 0.00019951803122050446,
+      "loss": 0.9032199382781982,
+      "step": 1548
+    },
+    {
+      "epoch": 0.6540084388185654,
+      "grad_norm": 0.4436919391155243,
+      "learning_rate": 0.00019951330990116754,
+      "loss": 0.8361048102378845,
+      "step": 1550
+    },
+    {
+      "epoch": 0.6548523206751055,
+      "grad_norm": 0.4229227900505066,
+      "learning_rate": 0.00019950856562602121,
+      "loss": 0.8661768436431885,
+      "step": 1552
+    },
+    {
+      "epoch": 0.6556962025316456,
+      "grad_norm": 0.4839727580547333,
+      "learning_rate": 0.00019950379839615988,
+      "loss": 0.9737826585769653,
+      "step": 1554
+    },
+    {
+      "epoch": 0.6565400843881857,
+      "grad_norm": 0.3859386742115021,
+      "learning_rate": 0.00019949900821268328,
+      "loss": 0.7542453408241272,
+      "step": 1556
+    },
+    {
+      "epoch": 0.6573839662447257,
+      "grad_norm": 0.39643988013267517,
+      "learning_rate": 0.0001994941950766964,
+      "loss": 0.9226290583610535,
+      "step": 1558
+    },
+    {
+      "epoch": 0.6582278481012658,
+      "grad_norm": 0.4261355400085449,
+      "learning_rate": 0.00019948935898930954,
+      "loss": 0.8375602960586548,
+      "step": 1560
+    },
+    {
+      "epoch": 0.6590717299578059,
+      "grad_norm": 0.445931077003479,
+      "learning_rate": 0.00019948449995163826,
+      "loss": 0.7847310304641724,
+      "step": 1562
+    },
+    {
+      "epoch": 0.659915611814346,
+      "grad_norm": 0.4680459797382355,
+      "learning_rate": 0.00019947961796480353,
+      "loss": 0.851150631904602,
+      "step": 1564
+    },
+    {
+      "epoch": 0.660759493670886,
+      "grad_norm": 0.6826074719429016,
+      "learning_rate": 0.00019947471302993143,
+      "loss": 0.8259562253952026,
+      "step": 1566
+    },
+    {
+      "epoch": 0.6616033755274262,
+      "grad_norm": 0.47369205951690674,
+      "learning_rate": 0.00019946978514815352,
+      "loss": 0.879021406173706,
+      "step": 1568
+    },
+    {
+      "epoch": 0.6624472573839663,
+      "grad_norm": 0.41111937165260315,
+      "learning_rate": 0.0001994648343206065,
+      "loss": 0.9296056032180786,
+      "step": 1570
+    },
+    {
+      "epoch": 0.6632911392405063,
+      "grad_norm": 0.4297783076763153,
+      "learning_rate": 0.00019945986054843248,
+      "loss": 0.873822808265686,
+      "step": 1572
+    },
+    {
+      "epoch": 0.6641350210970464,
+      "grad_norm": 0.48124316334724426,
+      "learning_rate": 0.00019945486383277878,
+      "loss": 0.937151312828064,
+      "step": 1574
+    },
+    {
+      "epoch": 0.6649789029535865,
+      "grad_norm": 0.4048190116882324,
+      "learning_rate": 0.00019944984417479805,
+      "loss": 0.8398929834365845,
+      "step": 1576
+    },
+    {
+      "epoch": 0.6658227848101266,
+      "grad_norm": 0.4023774266242981,
+      "learning_rate": 0.00019944480157564826,
+      "loss": 0.8269520998001099,
+      "step": 1578
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.3784184157848358,
+      "learning_rate": 0.00019943973603649262,
+      "loss": 0.8237608671188354,
+      "step": 1580
+    },
+    {
+      "epoch": 0.6675105485232068,
+      "grad_norm": 0.4448573887348175,
+      "learning_rate": 0.00019943464755849965,
+      "loss": 0.8469099998474121,
+      "step": 1582
+    },
+    {
+      "epoch": 0.6683544303797468,
+      "grad_norm": 0.4690774083137512,
+      "learning_rate": 0.00019942953614284317,
+      "loss": 0.8630822896957397,
+      "step": 1584
+    },
+    {
+      "epoch": 0.6691983122362869,
+      "grad_norm": 0.43670853972435,
+      "learning_rate": 0.00019942440179070227,
+      "loss": 0.8904082179069519,
+      "step": 1586
+    },
+    {
+      "epoch": 0.6700421940928271,
+      "grad_norm": 0.4236849844455719,
+      "learning_rate": 0.00019941924450326138,
+      "loss": 0.8355640172958374,
+      "step": 1588
+    },
+    {
+      "epoch": 0.6708860759493671,
+      "grad_norm": 0.48624539375305176,
+      "learning_rate": 0.00019941406428171013,
+      "loss": 0.8898435831069946,
+      "step": 1590
+    },
+    {
+      "epoch": 0.6717299578059072,
+      "grad_norm": 0.4453780949115753,
+      "learning_rate": 0.00019940886112724355,
+      "loss": 0.9000012278556824,
+      "step": 1592
+    },
+    {
+      "epoch": 0.6725738396624472,
+      "grad_norm": 0.4417199194431305,
+      "learning_rate": 0.00019940363504106193,
+      "loss": 0.8171058893203735,
+      "step": 1594
+    },
+    {
+      "epoch": 0.6734177215189874,
+      "grad_norm": 0.45055362582206726,
+      "learning_rate": 0.00019939838602437074,
+      "loss": 0.8703798055648804,
+      "step": 1596
+    },
+    {
+      "epoch": 0.6742616033755274,
+      "grad_norm": 0.42316532135009766,
+      "learning_rate": 0.00019939311407838088,
+      "loss": 0.8677281737327576,
+      "step": 1598
+    },
+    {
+      "epoch": 0.6751054852320675,
+      "grad_norm": 0.3935781717300415,
+      "learning_rate": 0.0001993878192043085,
+      "loss": 0.7314013242721558,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6751054852320675,
+      "eval_loss": 0.8769772052764893,
+      "eval_runtime": 677.9338,
+      "eval_samples_per_second": 3.108,
+      "eval_steps_per_second": 3.108,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6759493670886076,
+      "grad_norm": 0.4197012186050415,
+      "learning_rate": 0.000199382501403375,
+      "loss": 0.8075380325317383,
+      "step": 1602
+    },
+    {
+      "epoch": 0.6767932489451477,
+      "grad_norm": 0.45227327942848206,
+      "learning_rate": 0.00019937716067680713,
+      "loss": 0.7885794639587402,
+      "step": 1604
+    },
+    {
+      "epoch": 0.6776371308016877,
+      "grad_norm": 0.45660385489463806,
+      "learning_rate": 0.00019937179702583682,
+      "loss": 0.868261992931366,
+      "step": 1606
+    },
+    {
+      "epoch": 0.6784810126582278,
+      "grad_norm": 0.3955288231372833,
+      "learning_rate": 0.00019936641045170144,
+      "loss": 0.8306655287742615,
+      "step": 1608
+    },
+    {
+      "epoch": 0.679324894514768,
+      "grad_norm": 0.3651324510574341,
+      "learning_rate": 0.00019936100095564353,
+      "loss": 0.7505315542221069,
+      "step": 1610
+    },
+    {
+      "epoch": 0.680168776371308,
+      "grad_norm": 0.4240955412387848,
+      "learning_rate": 0.00019935556853891096,
+      "loss": 0.8792756795883179,
+      "step": 1612
+    },
+    {
+      "epoch": 0.6810126582278481,
+      "grad_norm": 0.4142507314682007,
+      "learning_rate": 0.00019935011320275687,
+      "loss": 0.7328272461891174,
+      "step": 1614
+    },
+    {
+      "epoch": 0.6818565400843882,
+      "grad_norm": 0.48339322209358215,
+      "learning_rate": 0.00019934463494843975,
+      "loss": 0.8414849042892456,
+      "step": 1616
+    },
+    {
+      "epoch": 0.6827004219409283,
+      "grad_norm": 0.4312443435192108,
+      "learning_rate": 0.00019933913377722328,
+      "loss": 0.8588716983795166,
+      "step": 1618
+    },
+    {
+      "epoch": 0.6835443037974683,
+      "grad_norm": 0.43849512934684753,
+      "learning_rate": 0.0001993336096903765,
+      "loss": 0.846511960029602,
+      "step": 1620
+    },
+    {
+      "epoch": 0.6843881856540084,
+      "grad_norm": 0.45767566561698914,
+      "learning_rate": 0.0001993280626891737,
+      "loss": 0.9352323412895203,
+      "step": 1622
+    },
+    {
+      "epoch": 0.6852320675105485,
+      "grad_norm": 0.438412606716156,
+      "learning_rate": 0.00019932249277489444,
+      "loss": 0.8884757161140442,
+      "step": 1624
+    },
+    {
+      "epoch": 0.6860759493670886,
+      "grad_norm": 0.45893922448158264,
+      "learning_rate": 0.0001993168999488237,
+      "loss": 0.7739649415016174,
+      "step": 1626
+    },
+    {
+      "epoch": 0.6869198312236287,
+      "grad_norm": 0.38631772994995117,
+      "learning_rate": 0.0001993112842122515,
+      "loss": 0.7948258519172668,
+      "step": 1628
+    },
+    {
+      "epoch": 0.6877637130801688,
+      "grad_norm": 0.3824027180671692,
+      "learning_rate": 0.00019930564556647334,
+      "loss": 0.750511884689331,
+      "step": 1630
+    },
+    {
+      "epoch": 0.6886075949367089,
+      "grad_norm": 0.3980776071548462,
+      "learning_rate": 0.00019929998401278996,
+      "loss": 0.9006738662719727,
+      "step": 1632
+    },
+    {
+      "epoch": 0.6894514767932489,
+      "grad_norm": 0.41419896483421326,
+      "learning_rate": 0.00019929429955250734,
+      "loss": 0.6996869444847107,
+      "step": 1634
+    },
+    {
+      "epoch": 0.6902953586497891,
+      "grad_norm": 0.43783676624298096,
+      "learning_rate": 0.00019928859218693682,
+      "loss": 0.8523393273353577,
+      "step": 1636
+    },
+    {
+      "epoch": 0.6911392405063291,
+      "grad_norm": 0.4378126561641693,
+      "learning_rate": 0.00019928286191739497,
+      "loss": 0.8885987401008606,
+      "step": 1638
+    },
+    {
+      "epoch": 0.6919831223628692,
+      "grad_norm": 0.4276133179664612,
+      "learning_rate": 0.00019927710874520363,
+      "loss": 0.8563809990882874,
+      "step": 1640
+    },
+    {
+      "epoch": 0.6928270042194092,
+      "grad_norm": 0.4600282311439514,
+      "learning_rate": 0.00019927133267168992,
+      "loss": 0.854114294052124,
+      "step": 1642
+    },
+    {
+      "epoch": 0.6936708860759494,
+      "grad_norm": 0.5058461427688599,
+      "learning_rate": 0.00019926553369818635,
+      "loss": 0.8557889461517334,
+      "step": 1644
+    },
+    {
+      "epoch": 0.6945147679324895,
+      "grad_norm": 0.4947412312030792,
+      "learning_rate": 0.00019925971182603057,
+      "loss": 0.8622767925262451,
+      "step": 1646
+    },
+    {
+      "epoch": 0.6953586497890295,
+      "grad_norm": 0.42555496096611023,
+      "learning_rate": 0.00019925386705656558,
+      "loss": 0.7836558818817139,
+      "step": 1648
+    },
+    {
+      "epoch": 0.6962025316455697,
+      "grad_norm": 0.47560831904411316,
+      "learning_rate": 0.00019924799939113967,
+      "loss": 0.7877846956253052,
+      "step": 1650
+    },
+    {
+      "epoch": 0.6970464135021097,
+      "grad_norm": 0.391702264547348,
+      "learning_rate": 0.0001992421088311064,
+      "loss": 0.775681734085083,
+      "step": 1652
+    },
+    {
+      "epoch": 0.6978902953586498,
+      "grad_norm": 0.463679701089859,
+      "learning_rate": 0.0001992361953778246,
+      "loss": 0.8809882402420044,
+      "step": 1654
+    },
+    {
+      "epoch": 0.6987341772151898,
+      "grad_norm": 0.44096484780311584,
+      "learning_rate": 0.00019923025903265836,
+      "loss": 0.8513299226760864,
+      "step": 1656
+    },
+    {
+      "epoch": 0.69957805907173,
+      "grad_norm": 0.47818854451179504,
+      "learning_rate": 0.0001992242997969772,
+      "loss": 0.8309667706489563,
+      "step": 1658
+    },
+    {
+      "epoch": 0.70042194092827,
+      "grad_norm": 0.4192790985107422,
+      "learning_rate": 0.00019921831767215565,
+      "loss": 0.8817558884620667,
+      "step": 1660
+    },
+    {
+      "epoch": 0.7012658227848101,
+      "grad_norm": 0.45285239815711975,
+      "learning_rate": 0.00019921231265957373,
+      "loss": 0.8315755724906921,
+      "step": 1662
+    },
+    {
+      "epoch": 0.7021097046413503,
+      "grad_norm": 0.47118625044822693,
+      "learning_rate": 0.00019920628476061673,
+      "loss": 0.8698058724403381,
+      "step": 1664
+    },
+    {
+      "epoch": 0.7029535864978903,
+      "grad_norm": 0.4769238829612732,
+      "learning_rate": 0.0001992002339766751,
+      "loss": 0.953620970249176,
+      "step": 1666
+    },
+    {
+      "epoch": 0.7037974683544304,
+      "grad_norm": 0.39397355914115906,
+      "learning_rate": 0.00019919416030914468,
+      "loss": 0.8536615371704102,
+      "step": 1668
+    },
+    {
+      "epoch": 0.7046413502109705,
+      "grad_norm": 0.3659648299217224,
+      "learning_rate": 0.00019918806375942655,
+      "loss": 0.8159828782081604,
+      "step": 1670
+    },
+    {
+      "epoch": 0.7054852320675106,
+      "grad_norm": 0.4381789565086365,
+      "learning_rate": 0.00019918194432892703,
+      "loss": 0.782463550567627,
+      "step": 1672
+    },
+    {
+      "epoch": 0.7063291139240506,
+      "grad_norm": 0.44745752215385437,
+      "learning_rate": 0.00019917580201905778,
+      "loss": 0.9314272999763489,
+      "step": 1674
+    },
+    {
+      "epoch": 0.7071729957805907,
+      "grad_norm": 0.4515164792537689,
+      "learning_rate": 0.00019916963683123574,
+      "loss": 0.860816240310669,
+      "step": 1676
+    },
+    {
+      "epoch": 0.7080168776371308,
+      "grad_norm": 0.42735886573791504,
+      "learning_rate": 0.00019916344876688303,
+      "loss": 0.8008456826210022,
+      "step": 1678
+    },
+    {
+      "epoch": 0.7088607594936709,
+      "grad_norm": 0.418275386095047,
+      "learning_rate": 0.00019915723782742718,
+      "loss": 0.7796595096588135,
+      "step": 1680
+    },
+    {
+      "epoch": 0.7097046413502109,
+      "grad_norm": 0.44347989559173584,
+      "learning_rate": 0.00019915100401430089,
+      "loss": 0.9006506204605103,
+      "step": 1682
+    },
+    {
+      "epoch": 0.7105485232067511,
+      "grad_norm": 0.3932702839374542,
+      "learning_rate": 0.00019914474732894225,
+      "loss": 0.7750146985054016,
+      "step": 1684
+    },
+    {
+      "epoch": 0.7113924050632912,
+      "grad_norm": 0.4810502231121063,
+      "learning_rate": 0.00019913846777279447,
+      "loss": 0.7508097290992737,
+      "step": 1686
+    },
+    {
+      "epoch": 0.7122362869198312,
+      "grad_norm": 0.4166778028011322,
+      "learning_rate": 0.00019913216534730622,
+      "loss": 0.7926796078681946,
+      "step": 1688
+    },
+    {
+      "epoch": 0.7130801687763713,
+      "grad_norm": 0.47900474071502686,
+      "learning_rate": 0.00019912584005393123,
+      "loss": 0.8363928198814392,
+      "step": 1690
+    },
+    {
+      "epoch": 0.7139240506329114,
+      "grad_norm": 0.435143381357193,
+      "learning_rate": 0.00019911949189412872,
+      "loss": 0.8649623990058899,
+      "step": 1692
+    },
+    {
+      "epoch": 0.7147679324894515,
+      "grad_norm": 0.4070208966732025,
+      "learning_rate": 0.00019911312086936305,
+      "loss": 0.8196499943733215,
+      "step": 1694
+    },
+    {
+      "epoch": 0.7156118143459915,
+      "grad_norm": 0.4717724621295929,
+      "learning_rate": 0.0001991067269811039,
+      "loss": 0.9157413840293884,
+      "step": 1696
+    },
+    {
+      "epoch": 0.7164556962025317,
+      "grad_norm": 0.3942880928516388,
+      "learning_rate": 0.0001991003102308262,
+      "loss": 0.7880456447601318,
+      "step": 1698
+    },
+    {
+      "epoch": 0.7172995780590717,
+      "grad_norm": 0.4037776589393616,
+      "learning_rate": 0.0001990938706200102,
+      "loss": 0.8088646531105042,
+      "step": 1700
+    },
+    {
+      "epoch": 0.7172995780590717,
+      "eval_loss": 0.8708170056343079,
+      "eval_runtime": 670.3019,
+      "eval_samples_per_second": 3.143,
+      "eval_steps_per_second": 3.143,
+      "step": 1700
+    },
+    {
+      "epoch": 0.7181434599156118,
+      "grad_norm": 0.4443536400794983,
+      "learning_rate": 0.0001990874081501414,
+      "loss": 0.8740324378013611,
+      "step": 1702
+    },
+    {
+      "epoch": 0.7189873417721518,
+      "grad_norm": 0.45089733600616455,
+      "learning_rate": 0.00019908092282271053,
+      "loss": 0.8305023908615112,
+      "step": 1704
+    },
+    {
+      "epoch": 0.719831223628692,
+      "grad_norm": 0.3903036415576935,
+      "learning_rate": 0.00019907441463921366,
+      "loss": 0.7718394994735718,
+      "step": 1706
+    },
+    {
+      "epoch": 0.7206751054852321,
+      "grad_norm": 0.42990729212760925,
+      "learning_rate": 0.00019906788360115208,
+      "loss": 0.843289315700531,
+      "step": 1708
+    },
+    {
+      "epoch": 0.7215189873417721,
+      "grad_norm": 0.4287833571434021,
+      "learning_rate": 0.0001990613297100324,
+      "loss": 0.8332282900810242,
+      "step": 1710
+    },
+    {
+      "epoch": 0.7223628691983123,
+      "grad_norm": 0.4557269513607025,
+      "learning_rate": 0.00019905475296736646,
+      "loss": 0.784180223941803,
+      "step": 1712
+    },
+    {
+      "epoch": 0.7232067510548523,
+      "grad_norm": 0.4001525640487671,
+      "learning_rate": 0.0001990481533746714,
+      "loss": 0.7807716131210327,
+      "step": 1714
+    },
+    {
+      "epoch": 0.7240506329113924,
+      "grad_norm": 0.5208709836006165,
+      "learning_rate": 0.0001990415309334696,
+      "loss": 0.7936414480209351,
+      "step": 1716
+    },
+    {
+      "epoch": 0.7248945147679325,
+      "grad_norm": 0.46469953656196594,
+      "learning_rate": 0.00019903488564528875,
+      "loss": 0.7964845895767212,
+      "step": 1718
+    },
+    {
+      "epoch": 0.7257383966244726,
+      "grad_norm": 0.405773401260376,
+      "learning_rate": 0.00019902821751166178,
+      "loss": 0.8302215337753296,
+      "step": 1720
+    },
+    {
+      "epoch": 0.7265822784810126,
+      "grad_norm": 0.3874973654747009,
+      "learning_rate": 0.00019902152653412692,
+      "loss": 0.8486512899398804,
+      "step": 1722
+    },
+    {
+      "epoch": 0.7274261603375527,
+      "grad_norm": 0.46865755319595337,
+      "learning_rate": 0.00019901481271422758,
+      "loss": 0.837841272354126,
+      "step": 1724
+    },
+    {
+      "epoch": 0.7282700421940929,
+      "grad_norm": 0.4139123558998108,
+      "learning_rate": 0.0001990080760535126,
+      "loss": 0.7640942335128784,
+      "step": 1726
+    },
+    {
+      "epoch": 0.7291139240506329,
+      "grad_norm": 0.5003024935722351,
+      "learning_rate": 0.000199001316553536,
+      "loss": 0.8236244320869446,
+      "step": 1728
+    },
+    {
+      "epoch": 0.729957805907173,
+      "grad_norm": 0.38495776057243347,
+      "learning_rate": 0.00019899453421585696,
+      "loss": 0.7792612314224243,
+      "step": 1730
+    },
+    {
+      "epoch": 0.7308016877637131,
+      "grad_norm": 0.4462733268737793,
+      "learning_rate": 0.00019898772904204014,
+      "loss": 0.7733872532844543,
+      "step": 1732
+    },
+    {
+      "epoch": 0.7316455696202532,
+      "grad_norm": 0.398248553276062,
+      "learning_rate": 0.00019898090103365532,
+      "loss": 0.8933147192001343,
+      "step": 1734
+    },
+    {
+      "epoch": 0.7324894514767932,
+      "grad_norm": 0.4231175482273102,
+      "learning_rate": 0.00019897405019227756,
+      "loss": 0.8198356628417969,
+      "step": 1736
+    },
+    {
+      "epoch": 0.7333333333333333,
+      "grad_norm": 0.43639084696769714,
+      "learning_rate": 0.0001989671765194873,
+      "loss": 0.8266391158103943,
+      "step": 1738
+    },
+    {
+      "epoch": 0.7341772151898734,
+      "grad_norm": 0.4235835671424866,
+      "learning_rate": 0.00019896028001687008,
+      "loss": 0.7839997410774231,
+      "step": 1740
+    },
+    {
+      "epoch": 0.7350210970464135,
+      "grad_norm": 0.4033111035823822,
+      "learning_rate": 0.00019895336068601687,
+      "loss": 0.7153518199920654,
+      "step": 1742
+    },
+    {
+      "epoch": 0.7358649789029535,
+      "grad_norm": 0.44281265139579773,
+      "learning_rate": 0.00019894641852852377,
+      "loss": 0.9079289436340332,
+      "step": 1744
+    },
+    {
+      "epoch": 0.7367088607594937,
+      "grad_norm": 0.4882747232913971,
+      "learning_rate": 0.0001989394535459922,
+      "loss": 0.8645254373550415,
+      "step": 1746
+    },
+    {
+      "epoch": 0.7375527426160338,
+      "grad_norm": 0.34858641028404236,
+      "learning_rate": 0.00019893246574002887,
+      "loss": 0.8134214878082275,
+      "step": 1748
+    },
+    {
+      "epoch": 0.7383966244725738,
+      "grad_norm": 0.4273303747177124,
+      "learning_rate": 0.00019892545511224574,
+      "loss": 0.8142663240432739,
+      "step": 1750
+    },
+    {
+      "epoch": 0.739240506329114,
+      "grad_norm": 0.4289957284927368,
+      "learning_rate": 0.00019891842166426004,
+      "loss": 0.7957643866539001,
+      "step": 1752
+    },
+    {
+      "epoch": 0.740084388185654,
+      "grad_norm": 0.4812224805355072,
+      "learning_rate": 0.00019891136539769416,
+      "loss": 0.9000363945960999,
+      "step": 1754
+    },
+    {
+      "epoch": 0.7409282700421941,
+      "grad_norm": 0.4078022837638855,
+      "learning_rate": 0.00019890428631417595,
+      "loss": 0.7719792723655701,
+      "step": 1756
+    },
+    {
+      "epoch": 0.7417721518987341,
+      "grad_norm": 0.40169858932495117,
+      "learning_rate": 0.00019889718441533838,
+      "loss": 0.7681847214698792,
+      "step": 1758
+    },
+    {
+      "epoch": 0.7426160337552743,
+      "grad_norm": 0.3713594377040863,
+      "learning_rate": 0.00019889005970281972,
+      "loss": 0.7536827325820923,
+      "step": 1760
+    },
+    {
+      "epoch": 0.7434599156118143,
+      "grad_norm": 0.4183627665042877,
+      "learning_rate": 0.0001988829121782635,
+      "loss": 0.9000160694122314,
+      "step": 1762
+    },
+    {
+      "epoch": 0.7443037974683544,
+      "grad_norm": 0.4241594672203064,
+      "learning_rate": 0.00019887574184331851,
+      "loss": 0.8047307133674622,
+      "step": 1764
+    },
+    {
+      "epoch": 0.7451476793248946,
+      "grad_norm": 0.4249233603477478,
+      "learning_rate": 0.00019886854869963883,
+      "loss": 0.8932583928108215,
+      "step": 1766
+    },
+    {
+      "epoch": 0.7459915611814346,
+      "grad_norm": 0.35815340280532837,
+      "learning_rate": 0.00019886133274888378,
+      "loss": 0.7684977054595947,
+      "step": 1768
+    },
+    {
+      "epoch": 0.7468354430379747,
+      "grad_norm": 0.41035401821136475,
+      "learning_rate": 0.00019885409399271795,
+      "loss": 0.8644338250160217,
+      "step": 1770
+    },
+    {
+      "epoch": 0.7476793248945147,
+      "grad_norm": 0.5450286269187927,
+      "learning_rate": 0.00019884683243281116,
+      "loss": 0.8035860657691956,
+      "step": 1772
+    },
+    {
+      "epoch": 0.7485232067510549,
+      "grad_norm": 0.5280999541282654,
+      "learning_rate": 0.0001988395480708385,
+      "loss": 0.8861207962036133,
+      "step": 1774
+    },
+    {
+      "epoch": 0.7493670886075949,
+      "grad_norm": 0.42665231227874756,
+      "learning_rate": 0.00019883224090848036,
+      "loss": 0.7506847977638245,
+      "step": 1776
+    },
+    {
+      "epoch": 0.750210970464135,
+      "grad_norm": 0.39029282331466675,
+      "learning_rate": 0.00019882491094742232,
+      "loss": 0.7662046551704407,
+      "step": 1778
+    },
+    {
+      "epoch": 0.7510548523206751,
+      "grad_norm": 0.43870019912719727,
+      "learning_rate": 0.00019881755818935534,
+      "loss": 0.7590143084526062,
+      "step": 1780
+    },
+    {
+      "epoch": 0.7518987341772152,
+      "grad_norm": 0.4195050001144409,
+      "learning_rate": 0.0001988101826359755,
+      "loss": 0.9340365529060364,
+      "step": 1782
+    },
+    {
+      "epoch": 0.7527426160337553,
+      "grad_norm": 0.4236123263835907,
+      "learning_rate": 0.0001988027842889842,
+      "loss": 0.7969209551811218,
+      "step": 1784
+    },
+    {
+      "epoch": 0.7535864978902953,
+      "grad_norm": 0.4274357259273529,
+      "learning_rate": 0.00019879536315008808,
+      "loss": 0.7892382144927979,
+      "step": 1786
+    },
+    {
+      "epoch": 0.7544303797468355,
+      "grad_norm": 0.4622916579246521,
+      "learning_rate": 0.00019878791922099912,
+      "loss": 0.8752562403678894,
+      "step": 1788
+    },
+    {
+      "epoch": 0.7552742616033755,
+      "grad_norm": 0.3648734986782074,
+      "learning_rate": 0.00019878045250343445,
+      "loss": 0.813011884689331,
+      "step": 1790
+    },
+    {
+      "epoch": 0.7561181434599156,
+      "grad_norm": 0.41361239552497864,
+      "learning_rate": 0.00019877296299911648,
+      "loss": 0.7310198545455933,
+      "step": 1792
+    },
+    {
+      "epoch": 0.7569620253164557,
+      "grad_norm": 0.4113222062587738,
+      "learning_rate": 0.00019876545070977294,
+      "loss": 0.858386754989624,
+      "step": 1794
+    },
+    {
+      "epoch": 0.7578059071729958,
+      "grad_norm": 0.4295370876789093,
+      "learning_rate": 0.00019875791563713676,
+      "loss": 0.8377325534820557,
+      "step": 1796
+    },
+    {
+      "epoch": 0.7586497890295358,
+      "grad_norm": 0.4227522015571594,
+      "learning_rate": 0.0001987503577829461,
+      "loss": 0.9139418005943298,
+      "step": 1798
+    },
+    {
+      "epoch": 0.759493670886076,
+      "grad_norm": 0.3638151288032532,
+      "learning_rate": 0.00019874277714894442,
+      "loss": 0.6994872689247131,
+      "step": 1800
+    },
+    {
+      "epoch": 0.759493670886076,
+      "eval_loss": 0.8625519275665283,
+      "eval_runtime": 686.4271,
+      "eval_samples_per_second": 3.07,
+      "eval_steps_per_second": 3.07,
+      "step": 1800
+    },
+    {
+      "epoch": 0.760337552742616,
+      "grad_norm": 0.46954411268234253,
+      "learning_rate": 0.00019873517373688047,
+      "loss": 0.8681167960166931,
+      "step": 1802
+    },
+    {
+      "epoch": 0.7611814345991561,
+      "grad_norm": 0.41958674788475037,
+      "learning_rate": 0.00019872754754850819,
+      "loss": 0.7420852184295654,
+      "step": 1804
+    },
+    {
+      "epoch": 0.7620253164556962,
+      "grad_norm": 0.431226521730423,
+      "learning_rate": 0.00019871989858558678,
+      "loss": 0.8018608093261719,
+      "step": 1806
+    },
+    {
+      "epoch": 0.7628691983122363,
+      "grad_norm": 0.45129454135894775,
+      "learning_rate": 0.0001987122268498807,
+      "loss": 0.8793904781341553,
+      "step": 1808
+    },
+    {
+      "epoch": 0.7637130801687764,
+      "grad_norm": 0.44278961420059204,
+      "learning_rate": 0.00019870453234315972,
+      "loss": 0.8416730761528015,
+      "step": 1810
+    },
+    {
+      "epoch": 0.7645569620253164,
+      "grad_norm": 0.4689098000526428,
+      "learning_rate": 0.00019869681506719876,
+      "loss": 0.8225743174552917,
+      "step": 1812
+    },
+    {
+      "epoch": 0.7654008438818566,
+      "grad_norm": 0.4413386881351471,
+      "learning_rate": 0.00019868907502377806,
+      "loss": 0.7871913313865662,
+      "step": 1814
+    },
+    {
+      "epoch": 0.7662447257383966,
+      "grad_norm": 0.48395273089408875,
+      "learning_rate": 0.00019868131221468315,
+      "loss": 0.8616237044334412,
+      "step": 1816
+    },
+    {
+      "epoch": 0.7670886075949367,
+      "grad_norm": 0.370263934135437,
+      "learning_rate": 0.0001986735266417047,
+      "loss": 0.7243452668190002,
+      "step": 1818
+    },
+    {
+      "epoch": 0.7679324894514767,
+      "grad_norm": 0.4522445797920227,
+      "learning_rate": 0.00019866571830663875,
+      "loss": 0.8623812198638916,
+      "step": 1820
+    },
+    {
+      "epoch": 0.7687763713080169,
+      "grad_norm": 0.43821004033088684,
+      "learning_rate": 0.00019865788721128643,
+      "loss": 0.776618480682373,
+      "step": 1822
+    },
+    {
+      "epoch": 0.769620253164557,
+      "grad_norm": 0.3825650215148926,
+      "learning_rate": 0.00019865003335745436,
+      "loss": 0.8722774982452393,
+      "step": 1824
+    },
+    {
+      "epoch": 0.770464135021097,
+      "grad_norm": 0.43684205412864685,
+      "learning_rate": 0.00019864215674695416,
+      "loss": 0.907802939414978,
+      "step": 1826
+    },
+    {
+      "epoch": 0.7713080168776372,
+      "grad_norm": 0.4128975570201874,
+      "learning_rate": 0.00019863425738160287,
+      "loss": 0.8505539298057556,
+      "step": 1828
+    },
+    {
+      "epoch": 0.7721518987341772,
+      "grad_norm": 0.4254065454006195,
+      "learning_rate": 0.0001986263352632227,
+      "loss": 0.83274906873703,
+      "step": 1830
+    },
+    {
+      "epoch": 0.7729957805907173,
+      "grad_norm": 0.4137375056743622,
+      "learning_rate": 0.00019861839039364118,
+      "loss": 0.8172078728675842,
+      "step": 1832
+    },
+    {
+      "epoch": 0.7738396624472574,
+      "grad_norm": 0.4387703239917755,
+      "learning_rate": 0.00019861042277469096,
+      "loss": 0.8610842227935791,
+      "step": 1834
+    },
+    {
+      "epoch": 0.7746835443037975,
+      "grad_norm": 0.398952454328537,
+      "learning_rate": 0.00019860243240821005,
+      "loss": 0.7961188554763794,
+      "step": 1836
+    },
+    {
+      "epoch": 0.7755274261603375,
+      "grad_norm": 0.4079921543598175,
+      "learning_rate": 0.0001985944192960417,
+      "loss": 0.8078321218490601,
+      "step": 1838
+    },
+    {
+      "epoch": 0.7763713080168776,
+      "grad_norm": 0.4701489508152008,
+      "learning_rate": 0.00019858638344003433,
+      "loss": 0.8743909001350403,
+      "step": 1840
+    },
+    {
+      "epoch": 0.7772151898734178,
+      "grad_norm": 0.5391269326210022,
+      "learning_rate": 0.00019857832484204173,
+      "loss": 0.7681707143783569,
+      "step": 1842
+    },
+    {
+      "epoch": 0.7780590717299578,
+      "grad_norm": 0.36250734329223633,
+      "learning_rate": 0.00019857024350392277,
+      "loss": 0.7817525863647461,
+      "step": 1844
+    },
+    {
+      "epoch": 0.7789029535864979,
+      "grad_norm": 0.41237136721611023,
+      "learning_rate": 0.00019856213942754175,
+      "loss": 0.7614551782608032,
+      "step": 1846
+    },
+    {
+      "epoch": 0.779746835443038,
+      "grad_norm": 0.3968869745731354,
+      "learning_rate": 0.00019855401261476807,
+      "loss": 0.7492313981056213,
+      "step": 1848
+    },
+    {
+      "epoch": 0.7805907172995781,
+      "grad_norm": 0.441476434469223,
+      "learning_rate": 0.00019854586306747646,
+      "loss": 0.8615695238113403,
+      "step": 1850
+    },
+    {
+      "epoch": 0.7814345991561181,
+      "grad_norm": 0.4486154317855835,
+      "learning_rate": 0.00019853769078754686,
+      "loss": 0.8416724801063538,
+      "step": 1852
+    },
+    {
+      "epoch": 0.7822784810126582,
+      "grad_norm": 0.45148879289627075,
+      "learning_rate": 0.0001985294957768644,
+      "loss": 0.8987806439399719,
+      "step": 1854
+    },
+    {
+      "epoch": 0.7831223628691983,
+      "grad_norm": 0.39330095052719116,
+      "learning_rate": 0.0001985212780373196,
+      "loss": 0.7597590088844299,
+      "step": 1856
+    },
+    {
+      "epoch": 0.7839662447257384,
+      "grad_norm": 0.4260408580303192,
+      "learning_rate": 0.00019851303757080808,
+      "loss": 0.8429927229881287,
+      "step": 1858
+    },
+    {
+      "epoch": 0.7848101265822784,
+      "grad_norm": 0.37519171833992004,
+      "learning_rate": 0.00019850477437923075,
+      "loss": 0.7443564534187317,
+      "step": 1860
+    },
+    {
+      "epoch": 0.7856540084388186,
+      "grad_norm": 0.4991084337234497,
+      "learning_rate": 0.00019849648846449383,
+      "loss": 0.934238612651825,
+      "step": 1862
+    },
+    {
+      "epoch": 0.7864978902953587,
+      "grad_norm": 0.4711335003376007,
+      "learning_rate": 0.00019848817982850867,
+      "loss": 0.8938905000686646,
+      "step": 1864
+    },
+    {
+      "epoch": 0.7873417721518987,
+      "grad_norm": 0.4145370423793793,
+      "learning_rate": 0.00019847984847319193,
+      "loss": 0.8745643496513367,
+      "step": 1866
+    },
+    {
+      "epoch": 0.7881856540084389,
+      "grad_norm": 0.3444044888019562,
+      "learning_rate": 0.00019847149440046548,
+      "loss": 0.7178786396980286,
+      "step": 1868
+    },
+    {
+      "epoch": 0.7890295358649789,
+      "grad_norm": 0.34632962942123413,
+      "learning_rate": 0.00019846311761225646,
+      "loss": 0.9114303588867188,
+      "step": 1870
+    },
+    {
+      "epoch": 0.789873417721519,
+      "grad_norm": 0.38630741834640503,
+      "learning_rate": 0.0001984547181104972,
+      "loss": 0.836281955242157,
+      "step": 1872
+    },
+    {
+      "epoch": 0.790717299578059,
+      "grad_norm": 0.40947434306144714,
+      "learning_rate": 0.00019844629589712534,
+      "loss": 0.8247858881950378,
+      "step": 1874
+    },
+    {
+      "epoch": 0.7915611814345992,
+      "grad_norm": 0.45878684520721436,
+      "learning_rate": 0.00019843785097408372,
+      "loss": 0.8446351289749146,
+      "step": 1876
+    },
+    {
+      "epoch": 0.7924050632911392,
+      "grad_norm": 0.37202852964401245,
+      "learning_rate": 0.0001984293833433204,
+      "loss": 0.8567686676979065,
+      "step": 1878
+    },
+    {
+      "epoch": 0.7932489451476793,
+      "grad_norm": 0.3839069902896881,
+      "learning_rate": 0.00019842089300678876,
+      "loss": 0.8134047985076904,
+      "step": 1880
+    },
+    {
+      "epoch": 0.7940928270042195,
+      "grad_norm": 0.4018687605857849,
+      "learning_rate": 0.00019841237996644727,
+      "loss": 0.8670110106468201,
+      "step": 1882
+    },
+    {
+      "epoch": 0.7949367088607595,
+      "grad_norm": 0.372089684009552,
+      "learning_rate": 0.0001984038442242598,
+      "loss": 0.7975355386734009,
+      "step": 1884
+    },
+    {
+      "epoch": 0.7957805907172996,
+      "grad_norm": 0.4492705166339874,
+      "learning_rate": 0.0001983952857821953,
+      "loss": 0.8755611777305603,
+      "step": 1886
+    },
+    {
+      "epoch": 0.7966244725738396,
+      "grad_norm": 0.451054185628891,
+      "learning_rate": 0.00019838670464222813,
+      "loss": 0.8680241107940674,
+      "step": 1888
+    },
+    {
+      "epoch": 0.7974683544303798,
+      "grad_norm": 0.4036900997161865,
+      "learning_rate": 0.00019837810080633773,
+      "loss": 0.7990990877151489,
+      "step": 1890
+    },
+    {
+      "epoch": 0.7983122362869198,
+      "grad_norm": 0.46125802397727966,
+      "learning_rate": 0.0001983694742765089,
+      "loss": 0.8598953485488892,
+      "step": 1892
+    },
+    {
+      "epoch": 0.7991561181434599,
+      "grad_norm": 0.4458785951137543,
+      "learning_rate": 0.00019836082505473153,
+      "loss": 0.7707474827766418,
+      "step": 1894
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.44501590728759766,
+      "learning_rate": 0.0001983521531430009,
+      "loss": 0.8069534301757812,
+      "step": 1896
+    },
+    {
+      "epoch": 0.8008438818565401,
+      "grad_norm": 0.400771826505661,
+      "learning_rate": 0.0001983434585433174,
+      "loss": 0.862334668636322,
+      "step": 1898
+    },
+    {
+      "epoch": 0.8016877637130801,
+      "grad_norm": 0.39193347096443176,
+      "learning_rate": 0.00019833474125768676,
+      "loss": 0.8125433325767517,
+      "step": 1900
+    },
+    {
+      "epoch": 0.8016877637130801,
+      "eval_loss": 0.8546335697174072,
+      "eval_runtime": 688.5301,
+      "eval_samples_per_second": 3.06,
+      "eval_steps_per_second": 3.06,
+      "step": 1900
+    },
+    {
+      "epoch": 0.8025316455696202,
+      "grad_norm": 0.441129207611084,
+      "learning_rate": 0.00019832600128811986,
+      "loss": 0.8246012330055237,
+      "step": 1902
+    },
+    {
+      "epoch": 0.8033755274261604,
+      "grad_norm": 0.4089467227458954,
+      "learning_rate": 0.00019831723863663285,
+      "loss": 0.8387641310691833,
+      "step": 1904
+    },
+    {
+      "epoch": 0.8042194092827004,
+      "grad_norm": 0.4871654510498047,
+      "learning_rate": 0.0001983084533052471,
+      "loss": 0.7891429662704468,
+      "step": 1906
+    },
+    {
+      "epoch": 0.8050632911392405,
+      "grad_norm": 0.40085604786872864,
+      "learning_rate": 0.00019829964529598923,
+      "loss": 0.8008774518966675,
+      "step": 1908
+    },
+    {
+      "epoch": 0.8059071729957806,
+      "grad_norm": 0.38775014877319336,
+      "learning_rate": 0.00019829081461089109,
+      "loss": 0.7921834588050842,
+      "step": 1910
+    },
+    {
+      "epoch": 0.8067510548523207,
+      "grad_norm": 0.37830594182014465,
+      "learning_rate": 0.0001982819612519897,
+      "loss": 0.7752519845962524,
+      "step": 1912
+    },
+    {
+      "epoch": 0.8075949367088607,
+      "grad_norm": 0.41137194633483887,
+      "learning_rate": 0.0001982730852213274,
+      "loss": 0.7958255410194397,
+      "step": 1914
+    },
+    {
+      "epoch": 0.8084388185654009,
+      "grad_norm": 0.4871830940246582,
+      "learning_rate": 0.0001982641865209517,
+      "loss": 0.8177281618118286,
+      "step": 1916
+    },
+    {
+      "epoch": 0.809282700421941,
+      "grad_norm": 0.4157388210296631,
+      "learning_rate": 0.0001982552651529154,
+      "loss": 0.7726616859436035,
+      "step": 1918
+    },
+    {
+      "epoch": 0.810126582278481,
+      "grad_norm": 0.32338014245033264,
+      "learning_rate": 0.00019824632111927645,
+      "loss": 0.6601400375366211,
+      "step": 1920
+    },
+    {
+      "epoch": 0.810970464135021,
+      "grad_norm": 0.4642949104309082,
+      "learning_rate": 0.00019823735442209804,
+      "loss": 0.7866622805595398,
+      "step": 1922
+    },
+    {
+      "epoch": 0.8118143459915612,
+      "grad_norm": 0.43558797240257263,
+      "learning_rate": 0.00019822836506344865,
+      "loss": 0.7857986092567444,
+      "step": 1924
+    },
+    {
+      "epoch": 0.8126582278481013,
+      "grad_norm": 0.3996953070163727,
+      "learning_rate": 0.000198219353045402,
+      "loss": 0.8407763838768005,
+      "step": 1926
+    },
+    {
+      "epoch": 0.8135021097046413,
+      "grad_norm": 0.4415414333343506,
+      "learning_rate": 0.00019821031837003686,
+      "loss": 0.8433752655982971,
+      "step": 1928
+    },
+    {
+      "epoch": 0.8143459915611815,
+      "grad_norm": 0.37329310178756714,
+      "learning_rate": 0.00019820126103943752,
+      "loss": 0.6941158175468445,
+      "step": 1930
+    },
+    {
+      "epoch": 0.8151898734177215,
+      "grad_norm": 0.4482601284980774,
+      "learning_rate": 0.00019819218105569323,
+      "loss": 0.775604248046875,
+      "step": 1932
+    },
+    {
+      "epoch": 0.8160337552742616,
+      "grad_norm": 0.426194429397583,
+      "learning_rate": 0.00019818307842089854,
+      "loss": 0.8496418595314026,
+      "step": 1934
+    },
+    {
+      "epoch": 0.8168776371308016,
+      "grad_norm": 0.42379093170166016,
+      "learning_rate": 0.00019817395313715335,
+      "loss": 0.7767958045005798,
+      "step": 1936
+    },
+    {
+      "epoch": 0.8177215189873418,
+      "grad_norm": 0.4149782061576843,
+      "learning_rate": 0.0001981648052065626,
+      "loss": 0.7347666025161743,
+      "step": 1938
+    },
+    {
+      "epoch": 0.8185654008438819,
+      "grad_norm": 0.3927431106567383,
+      "learning_rate": 0.0001981556346312366,
+      "loss": 0.7839647531509399,
+      "step": 1940
+    },
+    {
+      "epoch": 0.8194092827004219,
+      "grad_norm": 0.4483136236667633,
+      "learning_rate": 0.00019814644141329083,
+      "loss": 0.8870531916618347,
+      "step": 1942
+    },
+    {
+      "epoch": 0.8202531645569621,
+      "grad_norm": 0.4400147497653961,
+      "learning_rate": 0.00019813722555484594,
+      "loss": 0.8646620512008667,
+      "step": 1944
+    },
+    {
+      "epoch": 0.8210970464135021,
+      "grad_norm": 0.469971626996994,
+      "learning_rate": 0.00019812798705802785,
+      "loss": 0.9576541781425476,
+      "step": 1946
+    },
+    {
+      "epoch": 0.8219409282700422,
+      "grad_norm": 0.388113409280777,
+      "learning_rate": 0.00019811872592496776,
+      "loss": 0.7956477403640747,
+      "step": 1948
+    },
+    {
+      "epoch": 0.8227848101265823,
+      "grad_norm": 0.4271424114704132,
+      "learning_rate": 0.00019810944215780201,
+      "loss": 0.8440352082252502,
+      "step": 1950
+    },
+    {
+      "epoch": 0.8236286919831224,
+      "grad_norm": 0.4097966253757477,
+      "learning_rate": 0.00019810013575867217,
+      "loss": 0.8589065074920654,
+      "step": 1952
+    },
+    {
+      "epoch": 0.8244725738396624,
+      "grad_norm": 0.41550710797309875,
+      "learning_rate": 0.00019809080672972504,
+      "loss": 0.8606626987457275,
+      "step": 1954
+    },
+    {
+      "epoch": 0.8253164556962025,
+      "grad_norm": 0.3680916130542755,
+      "learning_rate": 0.0001980814550731127,
+      "loss": 0.7582436800003052,
+      "step": 1956
+    },
+    {
+      "epoch": 0.8261603375527427,
+      "grad_norm": 0.573297917842865,
+      "learning_rate": 0.00019807208079099234,
+      "loss": 0.8345186710357666,
+      "step": 1958
+    },
+    {
+      "epoch": 0.8270042194092827,
+      "grad_norm": 0.43631303310394287,
+      "learning_rate": 0.00019806268388552646,
+      "loss": 0.889683723449707,
+      "step": 1960
+    },
+    {
+      "epoch": 0.8278481012658228,
+      "grad_norm": 0.442531019449234,
+      "learning_rate": 0.00019805326435888275,
+      "loss": 0.8092973232269287,
+      "step": 1962
+    },
+    {
+      "epoch": 0.8286919831223629,
+      "grad_norm": 0.45020824670791626,
+      "learning_rate": 0.0001980438222132341,
+      "loss": 0.8919535279273987,
+      "step": 1964
+    },
+    {
+      "epoch": 0.829535864978903,
+      "grad_norm": 0.43131786584854126,
+      "learning_rate": 0.0001980343574507586,
+      "loss": 0.7652381658554077,
+      "step": 1966
+    },
+    {
+      "epoch": 0.830379746835443,
+      "grad_norm": 0.4461326599121094,
+      "learning_rate": 0.00019802487007363967,
+      "loss": 0.8512389659881592,
+      "step": 1968
+    },
+    {
+      "epoch": 0.8312236286919831,
+      "grad_norm": 0.4676894247531891,
+      "learning_rate": 0.0001980153600840658,
+      "loss": 0.8681327700614929,
+      "step": 1970
+    },
+    {
+      "epoch": 0.8320675105485232,
+      "grad_norm": 0.38376888632774353,
+      "learning_rate": 0.00019800582748423075,
+      "loss": 0.7476955652236938,
+      "step": 1972
+    },
+    {
+      "epoch": 0.8329113924050633,
+      "grad_norm": 0.37173229455947876,
+      "learning_rate": 0.00019799627227633354,
+      "loss": 0.814192533493042,
+      "step": 1974
+    },
+    {
+      "epoch": 0.8337552742616033,
+      "grad_norm": 0.4345237910747528,
+      "learning_rate": 0.00019798669446257844,
+      "loss": 0.7668994069099426,
+      "step": 1976
+    },
+    {
+      "epoch": 0.8345991561181435,
+      "grad_norm": 0.45872583985328674,
+      "learning_rate": 0.0001979770940451747,
+      "loss": 0.8321775197982788,
+      "step": 1978
+    },
+    {
+      "epoch": 0.8354430379746836,
+      "grad_norm": 0.40618205070495605,
+      "learning_rate": 0.0001979674710263371,
+      "loss": 0.7824342250823975,
+      "step": 1980
+    },
+    {
+      "epoch": 0.8362869198312236,
+      "grad_norm": 0.38377392292022705,
+      "learning_rate": 0.00019795782540828544,
+      "loss": 0.8752480745315552,
+      "step": 1982
+    },
+    {
+      "epoch": 0.8371308016877637,
+      "grad_norm": 0.36340072751045227,
+      "learning_rate": 0.00019794815719324476,
+      "loss": 0.66700679063797,
+      "step": 1984
+    },
+    {
+      "epoch": 0.8379746835443038,
+      "grad_norm": 0.4638384282588959,
+      "learning_rate": 0.00019793846638344536,
+      "loss": 0.8582209944725037,
+      "step": 1986
+    },
+    {
+      "epoch": 0.8388185654008439,
+      "grad_norm": 0.38364699482917786,
+      "learning_rate": 0.00019792875298112268,
+      "loss": 0.7724968791007996,
+      "step": 1988
+    },
+    {
+      "epoch": 0.8396624472573839,
+      "grad_norm": 0.439730703830719,
+      "learning_rate": 0.00019791901698851749,
+      "loss": 0.8542404174804688,
+      "step": 1990
+    },
+    {
+      "epoch": 0.8405063291139241,
+      "grad_norm": 0.4622722566127777,
+      "learning_rate": 0.00019790925840787563,
+      "loss": 0.8424296975135803,
+      "step": 1992
+    },
+    {
+      "epoch": 0.8413502109704641,
+      "grad_norm": 0.4491989314556122,
+      "learning_rate": 0.0001978994772414482,
+      "loss": 0.8400710225105286,
+      "step": 1994
+    },
+    {
+      "epoch": 0.8421940928270042,
+      "grad_norm": 0.372295081615448,
+      "learning_rate": 0.0001978896734914916,
+      "loss": 0.7265452146530151,
+      "step": 1996
+    },
+    {
+      "epoch": 0.8430379746835444,
+      "grad_norm": 0.4328666925430298,
+      "learning_rate": 0.00019787984716026732,
+      "loss": 0.8458228707313538,
+      "step": 1998
+    },
+    {
+      "epoch": 0.8438818565400844,
+      "grad_norm": 0.42246317863464355,
+      "learning_rate": 0.00019786999825004216,
+      "loss": 0.8517491817474365,
+      "step": 2000
+    },
+    {
+      "epoch": 0.8438818565400844,
+      "eval_loss": 0.8460908532142639,
+      "eval_runtime": 685.2518,
+      "eval_samples_per_second": 3.075,
+      "eval_steps_per_second": 3.075,
+      "step": 2000
+    },
+    {
+      "epoch": 0.8447257383966245,
+      "grad_norm": 0.3927323818206787,
+      "learning_rate": 0.00019786012676308802,
+      "loss": 0.8045481443405151,
+      "step": 2002
+    },
+    {
+      "epoch": 0.8455696202531645,
+      "grad_norm": 0.38189369440078735,
+      "learning_rate": 0.00019785023270168202,
+      "loss": 0.7509838938713074,
+      "step": 2004
+    },
+    {
+      "epoch": 0.8464135021097047,
+      "grad_norm": 0.40383535623550415,
+      "learning_rate": 0.00019784031606810663,
+      "loss": 0.7730163335800171,
+      "step": 2006
+    },
+    {
+      "epoch": 0.8472573839662447,
+      "grad_norm": 0.3665419816970825,
+      "learning_rate": 0.00019783037686464935,
+      "loss": 0.7096828818321228,
+      "step": 2008
+    },
+    {
+      "epoch": 0.8481012658227848,
+      "grad_norm": 0.36623379588127136,
+      "learning_rate": 0.00019782041509360302,
+      "loss": 0.8186401128768921,
+      "step": 2010
+    },
+    {
+      "epoch": 0.8489451476793249,
+      "grad_norm": 0.477255642414093,
+      "learning_rate": 0.0001978104307572656,
+      "loss": 0.8502879738807678,
+      "step": 2012
+    },
+    {
+      "epoch": 0.849789029535865,
+      "grad_norm": 0.44456636905670166,
+      "learning_rate": 0.00019780042385794032,
+      "loss": 0.9010884761810303,
+      "step": 2014
+    },
+    {
+      "epoch": 0.850632911392405,
+      "grad_norm": 0.42378175258636475,
+      "learning_rate": 0.00019779039439793552,
+      "loss": 0.774183452129364,
+      "step": 2016
+    },
+    {
+      "epoch": 0.8514767932489451,
+      "grad_norm": 0.43021997809410095,
+      "learning_rate": 0.00019778034237956482,
+      "loss": 0.8487694263458252,
+      "step": 2018
+    },
+    {
+      "epoch": 0.8523206751054853,
+      "grad_norm": 0.38188812136650085,
+      "learning_rate": 0.00019777026780514704,
+      "loss": 0.8004868030548096,
+      "step": 2020
+    },
+    {
+      "epoch": 0.8531645569620253,
+      "grad_norm": 0.3684130609035492,
+      "learning_rate": 0.0001977601706770062,
+      "loss": 0.7680953145027161,
+      "step": 2022
+    },
+    {
+      "epoch": 0.8540084388185654,
+      "grad_norm": 0.444844514131546,
+      "learning_rate": 0.00019775005099747153,
+      "loss": 0.8990264534950256,
+      "step": 2024
+    },
+    {
+      "epoch": 0.8548523206751055,
+      "grad_norm": 0.3432365655899048,
+      "learning_rate": 0.00019773990876887737,
+      "loss": 0.7676048874855042,
+      "step": 2026
+    },
+    {
+      "epoch": 0.8556962025316456,
+      "grad_norm": 0.3728618025779724,
+      "learning_rate": 0.00019772974399356343,
+      "loss": 0.8546769618988037,
+      "step": 2028
+    },
+    {
+      "epoch": 0.8565400843881856,
+      "grad_norm": 0.42767980694770813,
+      "learning_rate": 0.00019771955667387449,
+      "loss": 0.8009977340698242,
+      "step": 2030
+    },
+    {
+      "epoch": 0.8573839662447258,
+      "grad_norm": 0.4125710129737854,
+      "learning_rate": 0.00019770934681216055,
+      "loss": 0.8424299955368042,
+      "step": 2032
+    },
+    {
+      "epoch": 0.8582278481012658,
+      "grad_norm": 0.4399266839027405,
+      "learning_rate": 0.00019769911441077685,
+      "loss": 0.7940502762794495,
+      "step": 2034
+    },
+    {
+      "epoch": 0.8590717299578059,
+      "grad_norm": 0.41797709465026855,
+      "learning_rate": 0.0001976888594720838,
+      "loss": 0.9424314498901367,
+      "step": 2036
+    },
+    {
+      "epoch": 0.859915611814346,
+      "grad_norm": 0.5245009660720825,
+      "learning_rate": 0.00019767858199844698,
+      "loss": 0.9102896451950073,
+      "step": 2038
+    },
+    {
+      "epoch": 0.8607594936708861,
+      "grad_norm": 0.4092593193054199,
+      "learning_rate": 0.00019766828199223728,
+      "loss": 0.7434720396995544,
+      "step": 2040
+    },
+    {
+      "epoch": 0.8616033755274262,
+      "grad_norm": 0.40787473320961,
+      "learning_rate": 0.00019765795945583068,
+      "loss": 0.7441987991333008,
+      "step": 2042
+    },
+    {
+      "epoch": 0.8624472573839662,
+      "grad_norm": 0.42438021302223206,
+      "learning_rate": 0.0001976476143916084,
+      "loss": 0.8317872881889343,
+      "step": 2044
+    },
+    {
+      "epoch": 0.8632911392405064,
+      "grad_norm": 0.38261914253234863,
+      "learning_rate": 0.00019763724680195682,
+      "loss": 0.764509379863739,
+      "step": 2046
+    },
+    {
+      "epoch": 0.8641350210970464,
+      "grad_norm": 0.43295741081237793,
+      "learning_rate": 0.00019762685668926758,
+      "loss": 0.8666167855262756,
+      "step": 2048
+    },
+    {
+      "epoch": 0.8649789029535865,
+      "grad_norm": 0.36384403705596924,
+      "learning_rate": 0.00019761644405593743,
+      "loss": 0.7606754302978516,
+      "step": 2050
+    },
+    {
+      "epoch": 0.8658227848101265,
+      "grad_norm": 0.4296814501285553,
+      "learning_rate": 0.0001976060089043684,
+      "loss": 0.827495813369751,
+      "step": 2052
+    },
+    {
+      "epoch": 0.8666666666666667,
+      "grad_norm": 0.3831455111503601,
+      "learning_rate": 0.0001975955512369677,
+      "loss": 0.7418577075004578,
+      "step": 2054
+    },
+    {
+      "epoch": 0.8675105485232067,
+      "grad_norm": 0.44710344076156616,
+      "learning_rate": 0.0001975850710561477,
+      "loss": 0.8603045344352722,
+      "step": 2056
+    },
+    {
+      "epoch": 0.8683544303797468,
+      "grad_norm": 0.44012558460235596,
+      "learning_rate": 0.00019757456836432593,
+      "loss": 0.900658905506134,
+      "step": 2058
+    },
+    {
+      "epoch": 0.869198312236287,
+      "grad_norm": 0.42166298627853394,
+      "learning_rate": 0.0001975640431639252,
+      "loss": 0.8416497707366943,
+      "step": 2060
+    },
+    {
+      "epoch": 0.870042194092827,
+      "grad_norm": 0.40824490785598755,
+      "learning_rate": 0.00019755349545737345,
+      "loss": 0.8281066417694092,
+      "step": 2062
+    },
+    {
+      "epoch": 0.8708860759493671,
+      "grad_norm": 0.4272223711013794,
+      "learning_rate": 0.00019754292524710388,
+      "loss": 0.83869469165802,
+      "step": 2064
+    },
+    {
+      "epoch": 0.8717299578059071,
+      "grad_norm": 0.3732798397541046,
+      "learning_rate": 0.00019753233253555478,
+      "loss": 0.8337644934654236,
+      "step": 2066
+    },
+    {
+      "epoch": 0.8725738396624473,
+      "grad_norm": 0.4048069417476654,
+      "learning_rate": 0.00019752171732516973,
+      "loss": 0.7677452564239502,
+      "step": 2068
+    },
+    {
+      "epoch": 0.8734177215189873,
+      "grad_norm": 0.43628430366516113,
+      "learning_rate": 0.00019751107961839742,
+      "loss": 0.8320774435997009,
+      "step": 2070
+    },
+    {
+      "epoch": 0.8742616033755274,
+      "grad_norm": 0.43674105405807495,
+      "learning_rate": 0.00019750041941769174,
+      "loss": 0.8480262160301208,
+      "step": 2072
+    },
+    {
+      "epoch": 0.8751054852320675,
+      "grad_norm": 0.3678584098815918,
+      "learning_rate": 0.0001974897367255119,
+      "loss": 0.7598503828048706,
+      "step": 2074
+    },
+    {
+      "epoch": 0.8759493670886076,
+      "grad_norm": 0.4316411316394806,
+      "learning_rate": 0.0001974790315443221,
+      "loss": 0.7907084822654724,
+      "step": 2076
+    },
+    {
+      "epoch": 0.8767932489451477,
+      "grad_norm": 0.426465779542923,
+      "learning_rate": 0.00019746830387659182,
+      "loss": 0.8287727236747742,
+      "step": 2078
+    },
+    {
+      "epoch": 0.8776371308016878,
+      "grad_norm": 0.447337806224823,
+      "learning_rate": 0.00019745755372479578,
+      "loss": 0.8464577794075012,
+      "step": 2080
+    },
+    {
+      "epoch": 0.8784810126582279,
+      "grad_norm": 0.37877580523490906,
+      "learning_rate": 0.0001974467810914138,
+      "loss": 0.7383086681365967,
+      "step": 2082
+    },
+    {
+      "epoch": 0.8793248945147679,
+      "grad_norm": 0.3914664089679718,
+      "learning_rate": 0.0001974359859789309,
+      "loss": 0.729221522808075,
+      "step": 2084
+    },
+    {
+      "epoch": 0.880168776371308,
+      "grad_norm": 0.39767366647720337,
+      "learning_rate": 0.00019742516838983733,
+      "loss": 0.692332923412323,
+      "step": 2086
+    },
+    {
+      "epoch": 0.8810126582278481,
+      "grad_norm": 0.4396754205226898,
+      "learning_rate": 0.0001974143283266285,
+      "loss": 0.8540068864822388,
+      "step": 2088
+    },
+    {
+      "epoch": 0.8818565400843882,
+      "grad_norm": 0.4968172609806061,
+      "learning_rate": 0.00019740346579180502,
+      "loss": 0.8570998907089233,
+      "step": 2090
+    },
+    {
+      "epoch": 0.8827004219409282,
+      "grad_norm": 0.41686809062957764,
+      "learning_rate": 0.00019739258078787264,
+      "loss": 0.784139096736908,
+      "step": 2092
+    },
+    {
+      "epoch": 0.8835443037974684,
+      "grad_norm": 0.38788047432899475,
+      "learning_rate": 0.0001973816733173423,
+      "loss": 0.8521397113800049,
+      "step": 2094
+    },
+    {
+      "epoch": 0.8843881856540085,
+      "grad_norm": 0.38369178771972656,
+      "learning_rate": 0.00019737074338273023,
+      "loss": 0.8457935452461243,
+      "step": 2096
+    },
+    {
+      "epoch": 0.8852320675105485,
+      "grad_norm": 0.38869398832321167,
+      "learning_rate": 0.0001973597909865577,
+      "loss": 0.7425959706306458,
+      "step": 2098
+    },
+    {
+      "epoch": 0.8860759493670886,
+      "grad_norm": 0.4067917466163635,
+      "learning_rate": 0.00019734881613135122,
+      "loss": 0.7742393016815186,
+      "step": 2100
+    },
+    {
+      "epoch": 0.8860759493670886,
+      "eval_loss": 0.8401098847389221,
+      "eval_runtime": 669.1149,
+      "eval_samples_per_second": 3.149,
+      "eval_steps_per_second": 3.149,
+      "step": 2100
+    },
+    {
+      "epoch": 0.8869198312236287,
+      "grad_norm": 0.4282098412513733,
+      "learning_rate": 0.00019733781881964242,
+      "loss": 0.9099814295768738,
+      "step": 2102
+    },
+    {
+      "epoch": 0.8877637130801688,
+      "grad_norm": 0.33882173895835876,
+      "learning_rate": 0.00019732679905396826,
+      "loss": 0.7458856105804443,
+      "step": 2104
+    },
+    {
+      "epoch": 0.8886075949367088,
+      "grad_norm": 0.43959516286849976,
+      "learning_rate": 0.00019731575683687078,
+      "loss": 0.7829693555831909,
+      "step": 2106
+    },
+    {
+      "epoch": 0.889451476793249,
+      "grad_norm": 0.40046483278274536,
+      "learning_rate": 0.00019730469217089716,
+      "loss": 0.7898997068405151,
+      "step": 2108
+    },
+    {
+      "epoch": 0.890295358649789,
+      "grad_norm": 0.40572553873062134,
+      "learning_rate": 0.00019729360505859984,
+      "loss": 0.7874186635017395,
+      "step": 2110
+    },
+    {
+      "epoch": 0.8911392405063291,
+      "grad_norm": 0.42086347937583923,
+      "learning_rate": 0.0001972824955025364,
+      "loss": 0.8499898910522461,
+      "step": 2112
+    },
+    {
+      "epoch": 0.8919831223628693,
+      "grad_norm": 0.38845229148864746,
+      "learning_rate": 0.00019727136350526956,
+      "loss": 0.7940409779548645,
+      "step": 2114
+    },
+    {
+      "epoch": 0.8928270042194093,
+      "grad_norm": 0.41575267910957336,
+      "learning_rate": 0.00019726020906936728,
+      "loss": 0.8279082775115967,
+      "step": 2116
+    },
+    {
+      "epoch": 0.8936708860759494,
+      "grad_norm": 0.37615057826042175,
+      "learning_rate": 0.0001972490321974027,
+      "loss": 0.8217757940292358,
+      "step": 2118
+    },
+    {
+      "epoch": 0.8945147679324894,
+      "grad_norm": 0.9318385124206543,
+      "learning_rate": 0.00019723783289195415,
+      "loss": 0.7637750506401062,
+      "step": 2120
+    },
+    {
+      "epoch": 0.8953586497890296,
+      "grad_norm": 0.41948962211608887,
+      "learning_rate": 0.000197226611155605,
+      "loss": 0.8702366948127747,
+      "step": 2122
+    },
+    {
+      "epoch": 0.8962025316455696,
+      "grad_norm": 0.4718417525291443,
+      "learning_rate": 0.00019721536699094392,
+      "loss": 0.8146984577178955,
+      "step": 2124
+    },
+    {
+      "epoch": 0.8970464135021097,
+      "grad_norm": 0.4470657408237457,
+      "learning_rate": 0.00019720410040056474,
+      "loss": 0.8646075129508972,
+      "step": 2126
+    },
+    {
+      "epoch": 0.8978902953586498,
+      "grad_norm": 0.4176540970802307,
+      "learning_rate": 0.00019719281138706643,
+      "loss": 0.7925992012023926,
+      "step": 2128
+    },
+    {
+      "epoch": 0.8987341772151899,
+      "grad_norm": 0.4854719042778015,
+      "learning_rate": 0.00019718149995305315,
+      "loss": 0.8142143487930298,
+      "step": 2130
+    },
+    {
+      "epoch": 0.8995780590717299,
+      "grad_norm": 0.4270440340042114,
+      "learning_rate": 0.00019717016610113427,
+      "loss": 0.8050890564918518,
+      "step": 2132
+    },
+    {
+      "epoch": 0.90042194092827,
+      "grad_norm": 0.37710970640182495,
+      "learning_rate": 0.00019715880983392422,
+      "loss": 0.7888354659080505,
+      "step": 2134
+    },
+    {
+      "epoch": 0.9012658227848102,
+      "grad_norm": 0.4181252717971802,
+      "learning_rate": 0.00019714743115404273,
+      "loss": 0.7348225712776184,
+      "step": 2136
+    },
+    {
+      "epoch": 0.9021097046413502,
+      "grad_norm": 0.4475359618663788,
+      "learning_rate": 0.0001971360300641146,
+      "loss": 0.7729623913764954,
+      "step": 2138
+    },
+    {
+      "epoch": 0.9029535864978903,
+      "grad_norm": 0.4052332937717438,
+      "learning_rate": 0.00019712460656676992,
+      "loss": 0.8066384792327881,
+      "step": 2140
+    },
+    {
+      "epoch": 0.9037974683544304,
+      "grad_norm": 0.4567393362522125,
+      "learning_rate": 0.0001971131606646438,
+      "loss": 0.8053154349327087,
+      "step": 2142
+    },
+    {
+      "epoch": 0.9046413502109705,
+      "grad_norm": 0.4304409325122833,
+      "learning_rate": 0.00019710169236037656,
+      "loss": 0.7242560386657715,
+      "step": 2144
+    },
+    {
+      "epoch": 0.9054852320675105,
+      "grad_norm": 0.4785008132457733,
+      "learning_rate": 0.0001970902016566138,
+      "loss": 0.8475301265716553,
+      "step": 2146
+    },
+    {
+      "epoch": 0.9063291139240506,
+      "grad_norm": 0.45226508378982544,
+      "learning_rate": 0.00019707868855600618,
+      "loss": 0.928544282913208,
+      "step": 2148
+    },
+    {
+      "epoch": 0.9071729957805907,
+      "grad_norm": 0.3794468641281128,
+      "learning_rate": 0.00019706715306120956,
+      "loss": 0.8135722875595093,
+      "step": 2150
+    },
+    {
+      "epoch": 0.9080168776371308,
+      "grad_norm": 0.4005141258239746,
+      "learning_rate": 0.0001970555951748849,
+      "loss": 0.7808290719985962,
+      "step": 2152
+    },
+    {
+      "epoch": 0.9088607594936708,
+      "grad_norm": 0.37929919362068176,
+      "learning_rate": 0.00019704401489969842,
+      "loss": 0.8292917013168335,
+      "step": 2154
+    },
+    {
+      "epoch": 0.909704641350211,
+      "grad_norm": 0.4458678066730499,
+      "learning_rate": 0.0001970324122383215,
+      "loss": 0.772884726524353,
+      "step": 2156
+    },
+    {
+      "epoch": 0.9105485232067511,
+      "grad_norm": 0.41907885670661926,
+      "learning_rate": 0.0001970207871934306,
+      "loss": 0.8204470872879028,
+      "step": 2158
+    },
+    {
+      "epoch": 0.9113924050632911,
+      "grad_norm": 0.40610212087631226,
+      "learning_rate": 0.00019700913976770741,
+      "loss": 0.8361451625823975,
+      "step": 2160
+    },
+    {
+      "epoch": 0.9122362869198313,
+      "grad_norm": 0.4213273227214813,
+      "learning_rate": 0.00019699746996383878,
+      "loss": 0.8456075191497803,
+      "step": 2162
+    },
+    {
+      "epoch": 0.9130801687763713,
+      "grad_norm": 0.49264612793922424,
+      "learning_rate": 0.0001969857777845167,
+      "loss": 0.8731603622436523,
+      "step": 2164
+    },
+    {
+      "epoch": 0.9139240506329114,
+      "grad_norm": 0.42002132534980774,
+      "learning_rate": 0.00019697406323243834,
+      "loss": 0.8149666786193848,
+      "step": 2166
+    },
+    {
+      "epoch": 0.9147679324894514,
+      "grad_norm": 0.37501421570777893,
+      "learning_rate": 0.000196962326310306,
+      "loss": 0.7508546710014343,
+      "step": 2168
+    },
+    {
+      "epoch": 0.9156118143459916,
+      "grad_norm": 0.3706159293651581,
+      "learning_rate": 0.00019695056702082717,
+      "loss": 0.830280601978302,
+      "step": 2170
+    },
+    {
+      "epoch": 0.9164556962025316,
+      "grad_norm": 0.3873467445373535,
+      "learning_rate": 0.00019693878536671452,
+      "loss": 0.7297167778015137,
+      "step": 2172
+    },
+    {
+      "epoch": 0.9172995780590717,
+      "grad_norm": 0.42876651883125305,
+      "learning_rate": 0.00019692698135068582,
+      "loss": 0.7924292087554932,
+      "step": 2174
+    },
+    {
+      "epoch": 0.9181434599156119,
+      "grad_norm": 0.4259159564971924,
+      "learning_rate": 0.00019691515497546406,
+      "loss": 0.8098247051239014,
+      "step": 2176
+    },
+    {
+      "epoch": 0.9189873417721519,
+      "grad_norm": 0.38152721524238586,
+      "learning_rate": 0.00019690330624377728,
+      "loss": 0.6789063215255737,
+      "step": 2178
+    },
+    {
+      "epoch": 0.919831223628692,
+      "grad_norm": 0.4135437607765198,
+      "learning_rate": 0.00019689143515835887,
+      "loss": 0.8582620620727539,
+      "step": 2180
+    },
+    {
+      "epoch": 0.920675105485232,
+      "grad_norm": 0.3818798065185547,
+      "learning_rate": 0.0001968795417219472,
+      "loss": 0.8059743642807007,
+      "step": 2182
+    },
+    {
+      "epoch": 0.9215189873417722,
+      "grad_norm": 0.39021286368370056,
+      "learning_rate": 0.00019686762593728582,
+      "loss": 0.7432259917259216,
+      "step": 2184
+    },
+    {
+      "epoch": 0.9223628691983122,
+      "grad_norm": 0.37864068150520325,
+      "learning_rate": 0.00019685568780712355,
+      "loss": 0.7561086416244507,
+      "step": 2186
+    },
+    {
+      "epoch": 0.9232067510548523,
+      "grad_norm": 0.450847864151001,
+      "learning_rate": 0.00019684372733421425,
+      "loss": 0.7552722692489624,
+      "step": 2188
+    },
+    {
+      "epoch": 0.9240506329113924,
+      "grad_norm": 0.37909507751464844,
+      "learning_rate": 0.00019683174452131695,
+      "loss": 0.6905336380004883,
+      "step": 2190
+    },
+    {
+      "epoch": 0.9248945147679325,
+      "grad_norm": 0.4439173638820648,
+      "learning_rate": 0.0001968197393711959,
+      "loss": 0.7710593938827515,
+      "step": 2192
+    },
+    {
+      "epoch": 0.9257383966244725,
+      "grad_norm": 0.3908376395702362,
+      "learning_rate": 0.00019680771188662044,
+      "loss": 0.8212814331054688,
+      "step": 2194
+    },
+    {
+      "epoch": 0.9265822784810127,
+      "grad_norm": 0.3528299033641815,
+      "learning_rate": 0.00019679566207036507,
+      "loss": 0.751355767250061,
+      "step": 2196
+    },
+    {
+      "epoch": 0.9274261603375528,
+      "grad_norm": 0.4351397156715393,
+      "learning_rate": 0.00019678358992520945,
+      "loss": 0.8204057812690735,
+      "step": 2198
+    },
+    {
+      "epoch": 0.9282700421940928,
+      "grad_norm": 0.45566943287849426,
+      "learning_rate": 0.0001967714954539384,
+      "loss": 0.763841450214386,
+      "step": 2200
+    },
+    {
+      "epoch": 0.9282700421940928,
+      "eval_loss": 0.8336610198020935,
+      "eval_runtime": 674.5134,
+      "eval_samples_per_second": 3.124,
+      "eval_steps_per_second": 3.124,
+      "step": 2200
+    },
+    {
+      "epoch": 0.9291139240506329,
+      "grad_norm": 0.4516221284866333,
+      "learning_rate": 0.00019675937865934188,
+      "loss": 0.861711323261261,
+      "step": 2202
+    },
+    {
+      "epoch": 0.929957805907173,
+      "grad_norm": 0.43421855568885803,
+      "learning_rate": 0.00019674723954421502,
+      "loss": 0.8062718510627747,
+      "step": 2204
+    },
+    {
+      "epoch": 0.9308016877637131,
+      "grad_norm": 0.42442476749420166,
+      "learning_rate": 0.00019673507811135803,
+      "loss": 0.8069695234298706,
+      "step": 2206
+    },
+    {
+      "epoch": 0.9316455696202531,
+      "grad_norm": 0.3976045250892639,
+      "learning_rate": 0.00019672289436357637,
+      "loss": 0.7336108684539795,
+      "step": 2208
+    },
+    {
+      "epoch": 0.9324894514767933,
+      "grad_norm": 0.4131270945072174,
+      "learning_rate": 0.00019671068830368058,
+      "loss": 0.7559890151023865,
+      "step": 2210
+    },
+    {
+      "epoch": 0.9333333333333333,
+      "grad_norm": 0.4599294364452362,
+      "learning_rate": 0.00019669845993448634,
+      "loss": 0.8447075486183167,
+      "step": 2212
+    },
+    {
+      "epoch": 0.9341772151898734,
+      "grad_norm": 0.4291204810142517,
+      "learning_rate": 0.00019668620925881455,
+      "loss": 0.8329233527183533,
+      "step": 2214
+    },
+    {
+      "epoch": 0.9350210970464135,
+      "grad_norm": 0.4226539731025696,
+      "learning_rate": 0.00019667393627949116,
+      "loss": 0.749232292175293,
+      "step": 2216
+    },
+    {
+      "epoch": 0.9358649789029536,
+      "grad_norm": 0.4159580171108246,
+      "learning_rate": 0.00019666164099934732,
+      "loss": 0.7470011115074158,
+      "step": 2218
+    },
+    {
+      "epoch": 0.9367088607594937,
+      "grad_norm": 0.4160766303539276,
+      "learning_rate": 0.00019664932342121935,
+      "loss": 0.8094202280044556,
+      "step": 2220
+    },
+    {
+      "epoch": 0.9375527426160337,
+      "grad_norm": 0.45284298062324524,
+      "learning_rate": 0.00019663698354794863,
+      "loss": 0.8398169875144958,
+      "step": 2222
+    },
+    {
+      "epoch": 0.9383966244725739,
+      "grad_norm": 0.3960540294647217,
+      "learning_rate": 0.00019662462138238178,
+      "loss": 0.7652443051338196,
+      "step": 2224
+    },
+    {
+      "epoch": 0.9392405063291139,
+      "grad_norm": 0.4269275367259979,
+      "learning_rate": 0.00019661223692737047,
+      "loss": 0.7909489870071411,
+      "step": 2226
+    },
+    {
+      "epoch": 0.940084388185654,
+      "grad_norm": 0.38187873363494873,
+      "learning_rate": 0.00019659983018577157,
+      "loss": 0.7821935415267944,
+      "step": 2228
+    },
+    {
+      "epoch": 0.9409282700421941,
+      "grad_norm": 0.45898866653442383,
+      "learning_rate": 0.0001965874011604471,
+      "loss": 0.7792721390724182,
+      "step": 2230
+    },
+    {
+      "epoch": 0.9417721518987342,
+      "grad_norm": 0.4740699827671051,
+      "learning_rate": 0.00019657494985426418,
+      "loss": 0.885725200176239,
+      "step": 2232
+    },
+    {
+      "epoch": 0.9426160337552743,
+      "grad_norm": 0.3912813365459442,
+      "learning_rate": 0.00019656247627009506,
+      "loss": 0.7898101806640625,
+      "step": 2234
+    },
+    {
+      "epoch": 0.9434599156118143,
+      "grad_norm": 0.4552418291568756,
+      "learning_rate": 0.0001965499804108172,
+      "loss": 0.8646330833435059,
+      "step": 2236
+    },
+    {
+      "epoch": 0.9443037974683545,
+      "grad_norm": 0.4178548753261566,
+      "learning_rate": 0.0001965374622793131,
+      "loss": 0.8068296313285828,
+      "step": 2238
+    },
+    {
+      "epoch": 0.9451476793248945,
+      "grad_norm": 0.38649076223373413,
+      "learning_rate": 0.00019652492187847057,
+      "loss": 0.7953271865844727,
+      "step": 2240
+    },
+    {
+      "epoch": 0.9459915611814346,
+      "grad_norm": 0.39785075187683105,
+      "learning_rate": 0.0001965123592111823,
+      "loss": 0.8423402309417725,
+      "step": 2242
+    },
+    {
+      "epoch": 0.9468354430379747,
+      "grad_norm": 0.47455263137817383,
+      "learning_rate": 0.00019649977428034632,
+      "loss": 0.8883589506149292,
+      "step": 2244
+    },
+    {
+      "epoch": 0.9476793248945148,
+      "grad_norm": 0.3866841197013855,
+      "learning_rate": 0.0001964871670888657,
+      "loss": 0.7232863903045654,
+      "step": 2246
+    },
+    {
+      "epoch": 0.9485232067510548,
+      "grad_norm": 0.3950609564781189,
+      "learning_rate": 0.00019647453763964876,
+      "loss": 0.7265347838401794,
+      "step": 2248
+    },
+    {
+      "epoch": 0.9493670886075949,
+      "grad_norm": 0.4315829575061798,
+      "learning_rate": 0.0001964618859356088,
+      "loss": 0.8000131249427795,
+      "step": 2250
+    },
+    {
+      "epoch": 0.950210970464135,
+      "grad_norm": 0.403984934091568,
+      "learning_rate": 0.00019644921197966433,
+      "loss": 0.823086678981781,
+      "step": 2252
+    },
+    {
+      "epoch": 0.9510548523206751,
+      "grad_norm": 0.4311080574989319,
+      "learning_rate": 0.00019643651577473897,
+      "loss": 0.8457553386688232,
+      "step": 2254
+    },
+    {
+      "epoch": 0.9518987341772152,
+      "grad_norm": 0.3884856104850769,
+      "learning_rate": 0.00019642379732376158,
+      "loss": 0.7866189479827881,
+      "step": 2256
+    },
+    {
+      "epoch": 0.9527426160337553,
+      "grad_norm": 0.3384067714214325,
+      "learning_rate": 0.00019641105662966595,
+      "loss": 0.7095813751220703,
+      "step": 2258
+    },
+    {
+      "epoch": 0.9535864978902954,
+      "grad_norm": 0.40426746010780334,
+      "learning_rate": 0.00019639829369539117,
+      "loss": 0.8271632194519043,
+      "step": 2260
+    },
+    {
+      "epoch": 0.9544303797468354,
+      "grad_norm": 0.45495131611824036,
+      "learning_rate": 0.00019638550852388144,
+      "loss": 0.8204861283302307,
+      "step": 2262
+    },
+    {
+      "epoch": 0.9552742616033755,
+      "grad_norm": 0.47201067209243774,
+      "learning_rate": 0.00019637270111808598,
+      "loss": 0.884763777256012,
+      "step": 2264
+    },
+    {
+      "epoch": 0.9561181434599156,
+      "grad_norm": 0.3959396779537201,
+      "learning_rate": 0.0001963598714809592,
+      "loss": 0.7630926370620728,
+      "step": 2266
+    },
+    {
+      "epoch": 0.9569620253164557,
+      "grad_norm": 0.4197136461734772,
+      "learning_rate": 0.00019634701961546077,
+      "loss": 0.797870397567749,
+      "step": 2268
+    },
+    {
+      "epoch": 0.9578059071729957,
+      "grad_norm": 0.381290078163147,
+      "learning_rate": 0.00019633414552455526,
+      "loss": 0.7755846977233887,
+      "step": 2270
+    },
+    {
+      "epoch": 0.9586497890295359,
+      "grad_norm": 0.3868124783039093,
+      "learning_rate": 0.00019632124921121255,
+      "loss": 0.7553264498710632,
+      "step": 2272
+    },
+    {
+      "epoch": 0.959493670886076,
+      "grad_norm": 0.4350631833076477,
+      "learning_rate": 0.00019630833067840747,
+      "loss": 0.7792160511016846,
+      "step": 2274
+    },
+    {
+      "epoch": 0.960337552742616,
+      "grad_norm": 0.4717157781124115,
+      "learning_rate": 0.00019629538992912016,
+      "loss": 0.878220796585083,
+      "step": 2276
+    },
+    {
+      "epoch": 0.9611814345991562,
+      "grad_norm": 0.45545122027397156,
+      "learning_rate": 0.00019628242696633584,
+      "loss": 0.8891014456748962,
+      "step": 2278
+    },
+    {
+      "epoch": 0.9620253164556962,
+      "grad_norm": 0.4421025514602661,
+      "learning_rate": 0.00019626944179304467,
+      "loss": 0.8024723529815674,
+      "step": 2280
+    },
+    {
+      "epoch": 0.9628691983122363,
+      "grad_norm": 0.40385812520980835,
+      "learning_rate": 0.00019625643441224222,
+      "loss": 0.8036280274391174,
+      "step": 2282
+    },
+    {
+      "epoch": 0.9637130801687763,
+      "grad_norm": 0.3866552412509918,
+      "learning_rate": 0.00019624340482692898,
+      "loss": 0.8072395324707031,
+      "step": 2284
+    },
+    {
+      "epoch": 0.9645569620253165,
+      "grad_norm": 0.4113233685493469,
+      "learning_rate": 0.0001962303530401106,
+      "loss": 0.8046114444732666,
+      "step": 2286
+    },
+    {
+      "epoch": 0.9654008438818565,
+      "grad_norm": 0.40720227360725403,
+      "learning_rate": 0.00019621727905479798,
+      "loss": 0.8434037566184998,
+      "step": 2288
+    },
+    {
+      "epoch": 0.9662447257383966,
+      "grad_norm": 0.35759273171424866,
+      "learning_rate": 0.00019620418287400693,
+      "loss": 0.7832419276237488,
+      "step": 2290
+    },
+    {
+      "epoch": 0.9670886075949368,
+      "grad_norm": 0.4170442223548889,
+      "learning_rate": 0.0001961910645007585,
+      "loss": 0.7975693345069885,
+      "step": 2292
+    },
+    {
+      "epoch": 0.9679324894514768,
+      "grad_norm": 0.4126700162887573,
+      "learning_rate": 0.00019617792393807895,
+      "loss": 0.8554494976997375,
+      "step": 2294
+    },
+    {
+      "epoch": 0.9687763713080169,
+      "grad_norm": 0.4231695830821991,
+      "learning_rate": 0.00019616476118899942,
+      "loss": 0.824539065361023,
+      "step": 2296
+    },
+    {
+      "epoch": 0.9696202531645569,
+      "grad_norm": 0.3556209206581116,
+      "learning_rate": 0.0001961515762565564,
+      "loss": 0.7008040547370911,
+      "step": 2298
+    },
+    {
+      "epoch": 0.9704641350210971,
+      "grad_norm": 0.4370400309562683,
+      "learning_rate": 0.00019613836914379132,
+      "loss": 0.8192558288574219,
+      "step": 2300
+    },
+    {
+      "epoch": 0.9704641350210971,
+      "eval_loss": 0.8281980156898499,
+      "eval_runtime": 688.6136,
+      "eval_samples_per_second": 3.06,
+      "eval_steps_per_second": 3.06,
+      "step": 2300
+    },
+    {
+      "epoch": 0.9713080168776371,
+      "grad_norm": 0.4207614064216614,
+      "learning_rate": 0.00019612513985375088,
+      "loss": 0.878928005695343,
+      "step": 2302
+    },
+    {
+      "epoch": 0.9721518987341772,
+      "grad_norm": 0.37607815861701965,
+      "learning_rate": 0.0001961118883894868,
+      "loss": 0.736182689666748,
+      "step": 2304
+    },
+    {
+      "epoch": 0.9729957805907173,
+      "grad_norm": 0.43311476707458496,
+      "learning_rate": 0.00019609861475405592,
+      "loss": 0.8403618335723877,
+      "step": 2306
+    },
+    {
+      "epoch": 0.9738396624472574,
+      "grad_norm": 0.4533176124095917,
+      "learning_rate": 0.00019608531895052021,
+      "loss": 0.8414564728736877,
+      "step": 2308
+    },
+    {
+      "epoch": 0.9746835443037974,
+      "grad_norm": 0.4025697410106659,
+      "learning_rate": 0.00019607200098194677,
+      "loss": 0.8105775117874146,
+      "step": 2310
+    },
+    {
+      "epoch": 0.9755274261603376,
+      "grad_norm": 0.37894198298454285,
+      "learning_rate": 0.00019605866085140784,
+      "loss": 0.8053821325302124,
+      "step": 2312
+    },
+    {
+      "epoch": 0.9763713080168777,
+      "grad_norm": 0.46265891194343567,
+      "learning_rate": 0.0001960452985619807,
+      "loss": 0.8092884421348572,
+      "step": 2314
+    },
+    {
+      "epoch": 0.9772151898734177,
+      "grad_norm": 0.3770054280757904,
+      "learning_rate": 0.0001960319141167477,
+      "loss": 0.7456095218658447,
+      "step": 2316
+    },
+    {
+      "epoch": 0.9780590717299578,
+      "grad_norm": 0.44084450602531433,
+      "learning_rate": 0.0001960185075187965,
+      "loss": 0.7739883661270142,
+      "step": 2318
+    },
+    {
+      "epoch": 0.9789029535864979,
+      "grad_norm": 0.3880341351032257,
+      "learning_rate": 0.00019600507877121965,
+      "loss": 0.6961302757263184,
+      "step": 2320
+    },
+    {
+      "epoch": 0.979746835443038,
+      "grad_norm": 0.468037486076355,
+      "learning_rate": 0.00019599162787711495,
+      "loss": 0.8447984457015991,
+      "step": 2322
+    },
+    {
+      "epoch": 0.980590717299578,
+      "grad_norm": 0.4123631417751312,
+      "learning_rate": 0.00019597815483958527,
+      "loss": 0.8155615329742432,
+      "step": 2324
+    },
+    {
+      "epoch": 0.9814345991561182,
+      "grad_norm": 0.3827875554561615,
+      "learning_rate": 0.00019596465966173853,
+      "loss": 0.7990952730178833,
+      "step": 2326
+    },
+    {
+      "epoch": 0.9822784810126582,
+      "grad_norm": 0.4644222557544708,
+      "learning_rate": 0.00019595114234668788,
+      "loss": 0.8899750709533691,
+      "step": 2328
+    },
+    {
+      "epoch": 0.9831223628691983,
+      "grad_norm": 0.43699249625205994,
+      "learning_rate": 0.00019593760289755144,
+      "loss": 0.7863160371780396,
+      "step": 2330
+    },
+    {
+      "epoch": 0.9839662447257383,
+      "grad_norm": 0.40713244676589966,
+      "learning_rate": 0.00019592404131745254,
+      "loss": 0.7960637807846069,
+      "step": 2332
+    },
+    {
+      "epoch": 0.9848101265822785,
+      "grad_norm": 0.37364068627357483,
+      "learning_rate": 0.00019591045760951957,
+      "loss": 0.714144229888916,
+      "step": 2334
+    },
+    {
+      "epoch": 0.9856540084388186,
+      "grad_norm": 0.4094736576080322,
+      "learning_rate": 0.00019589685177688599,
+      "loss": 0.7311791777610779,
+      "step": 2336
+    },
+    {
+      "epoch": 0.9864978902953586,
+      "grad_norm": 0.4275180697441101,
+      "learning_rate": 0.0001958832238226905,
+      "loss": 0.8191342353820801,
+      "step": 2338
+    },
+    {
+      "epoch": 0.9873417721518988,
+      "grad_norm": 0.4627651274204254,
+      "learning_rate": 0.00019586957375007668,
+      "loss": 0.7476106286048889,
+      "step": 2340
+    },
+    {
+      "epoch": 0.9881856540084388,
+      "grad_norm": 0.44676414132118225,
+      "learning_rate": 0.00019585590156219346,
+      "loss": 0.8257967233657837,
+      "step": 2342
+    },
+    {
+      "epoch": 0.9890295358649789,
+      "grad_norm": 0.40667134523391724,
+      "learning_rate": 0.00019584220726219466,
+      "loss": 0.7922079563140869,
+      "step": 2344
+    },
+    {
+      "epoch": 0.9898734177215189,
+      "grad_norm": 0.40865346789360046,
+      "learning_rate": 0.00019582849085323934,
+      "loss": 0.7892434597015381,
+      "step": 2346
+    },
+    {
+      "epoch": 0.9907172995780591,
+      "grad_norm": 0.43309709429740906,
+      "learning_rate": 0.00019581475233849163,
+      "loss": 0.8766644597053528,
+      "step": 2348
+    },
+    {
+      "epoch": 0.9915611814345991,
+      "grad_norm": 0.41558781266212463,
+      "learning_rate": 0.00019580099172112069,
+      "loss": 0.9183312058448792,
+      "step": 2350
+    },
+    {
+      "epoch": 0.9924050632911392,
+      "grad_norm": 0.432748019695282,
+      "learning_rate": 0.00019578720900430082,
+      "loss": 0.7905668020248413,
+      "step": 2352
+    },
+    {
+      "epoch": 0.9932489451476794,
+      "grad_norm": 0.4116985499858856,
+      "learning_rate": 0.0001957734041912115,
+      "loss": 0.8279712200164795,
+      "step": 2354
+    },
+    {
+      "epoch": 0.9940928270042194,
+      "grad_norm": 0.3927275240421295,
+      "learning_rate": 0.00019575957728503718,
+      "loss": 0.7451362609863281,
+      "step": 2356
+    },
+    {
+      "epoch": 0.9949367088607595,
+      "grad_norm": 0.3389509320259094,
+      "learning_rate": 0.0001957457282889675,
+      "loss": 0.6769390106201172,
+      "step": 2358
+    },
+    {
+      "epoch": 0.9957805907172996,
+      "grad_norm": 0.4220774471759796,
+      "learning_rate": 0.00019573185720619712,
+      "loss": 0.7590897083282471,
+      "step": 2360
+    },
+    {
+      "epoch": 0.9966244725738397,
+      "grad_norm": 0.4299897849559784,
+      "learning_rate": 0.00019571796403992583,
+      "loss": 0.8301284313201904,
+      "step": 2362
+    },
+    {
+      "epoch": 0.9974683544303797,
+      "grad_norm": 0.4031789302825928,
+      "learning_rate": 0.00019570404879335856,
+      "loss": 0.8019471764564514,
+      "step": 2364
+    },
+    {
+      "epoch": 0.9983122362869198,
+      "grad_norm": 0.3957567811012268,
+      "learning_rate": 0.00019569011146970526,
+      "loss": 0.7427027821540833,
+      "step": 2366
+    },
+    {
+      "epoch": 0.99915611814346,
+      "grad_norm": 0.41741347312927246,
+      "learning_rate": 0.000195676152072181,
+      "loss": 0.8221617341041565,
+      "step": 2368
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.47121551632881165,
+      "learning_rate": 0.00019566217060400595,
+      "loss": 0.8143455982208252,
+      "step": 2370
+    },
+    {
+      "epoch": 1.0008438818565402,
+      "grad_norm": 0.3854716122150421,
+      "learning_rate": 0.0001956481670684054,
+      "loss": 0.6971694231033325,
+      "step": 2372
+    },
+    {
+      "epoch": 1.00168776371308,
+      "grad_norm": 0.4029138386249542,
+      "learning_rate": 0.00019563414146860964,
+      "loss": 0.6719155311584473,
+      "step": 2374
+    },
+    {
+      "epoch": 1.0025316455696203,
+      "grad_norm": 0.40136656165122986,
+      "learning_rate": 0.00019562009380785412,
+      "loss": 0.7161872386932373,
+      "step": 2376
+    },
+    {
+      "epoch": 1.0033755274261604,
+      "grad_norm": 0.40685850381851196,
+      "learning_rate": 0.0001956060240893794,
+      "loss": 0.71699059009552,
+      "step": 2378
+    },
+    {
+      "epoch": 1.0042194092827004,
+      "grad_norm": 0.46153366565704346,
+      "learning_rate": 0.00019559193231643104,
+      "loss": 0.8030059337615967,
+      "step": 2380
+    },
+    {
+      "epoch": 1.0050632911392405,
+      "grad_norm": 0.45048823952674866,
+      "learning_rate": 0.0001955778184922598,
+      "loss": 0.7666479349136353,
+      "step": 2382
+    },
+    {
+      "epoch": 1.0059071729957807,
+      "grad_norm": 0.4260413348674774,
+      "learning_rate": 0.00019556368262012142,
+      "loss": 0.7481234073638916,
+      "step": 2384
+    },
+    {
+      "epoch": 1.0067510548523206,
+      "grad_norm": 0.4819575548171997,
+      "learning_rate": 0.0001955495247032768,
+      "loss": 0.7943820953369141,
+      "step": 2386
+    },
+    {
+      "epoch": 1.0075949367088608,
+      "grad_norm": 0.45789605379104614,
+      "learning_rate": 0.0001955353447449919,
+      "loss": 0.6401634812355042,
+      "step": 2388
+    },
+    {
+      "epoch": 1.0084388185654007,
+      "grad_norm": 0.37019097805023193,
+      "learning_rate": 0.00019552114274853778,
+      "loss": 0.7058886885643005,
+      "step": 2390
+    },
+    {
+      "epoch": 1.009282700421941,
+      "grad_norm": 0.44416263699531555,
+      "learning_rate": 0.0001955069187171905,
+      "loss": 0.7631242871284485,
+      "step": 2392
+    },
+    {
+      "epoch": 1.010126582278481,
+      "grad_norm": 0.4478379786014557,
+      "learning_rate": 0.00019549267265423138,
+      "loss": 0.7873892784118652,
+      "step": 2394
+    },
+    {
+      "epoch": 1.010970464135021,
+      "grad_norm": 0.4596130847930908,
+      "learning_rate": 0.00019547840456294658,
+      "loss": 0.7882059216499329,
+      "step": 2396
+    },
+    {
+      "epoch": 1.0118143459915612,
+      "grad_norm": 0.4062999188899994,
+      "learning_rate": 0.0001954641144466276,
+      "loss": 0.6358574628829956,
+      "step": 2398
+    },
+    {
+      "epoch": 1.0126582278481013,
+      "grad_norm": 0.47627049684524536,
+      "learning_rate": 0.0001954498023085708,
+      "loss": 0.7803886532783508,
+      "step": 2400
+    },
+    {
+      "epoch": 1.0126582278481013,
+      "eval_loss": 0.8250564932823181,
+      "eval_runtime": 691.5833,
+      "eval_samples_per_second": 3.047,
+      "eval_steps_per_second": 3.047,
+      "step": 2400
+    },
+    {
+      "epoch": 1.0135021097046413,
+      "grad_norm": 0.40690696239471436,
+      "learning_rate": 0.00019543546815207777,
+      "loss": 0.7394700050354004,
+      "step": 2402
+    },
+    {
+      "epoch": 1.0143459915611814,
+      "grad_norm": 0.45543980598449707,
+      "learning_rate": 0.0001954211119804551,
+      "loss": 0.7096555829048157,
+      "step": 2404
+    },
+    {
+      "epoch": 1.0151898734177216,
+      "grad_norm": 0.4367314279079437,
+      "learning_rate": 0.00019540673379701454,
+      "loss": 0.7961843013763428,
+      "step": 2406
+    },
+    {
+      "epoch": 1.0160337552742615,
+      "grad_norm": 0.4458340108394623,
+      "learning_rate": 0.00019539233360507278,
+      "loss": 0.743843138217926,
+      "step": 2408
+    },
+    {
+      "epoch": 1.0168776371308017,
+      "grad_norm": 0.3741893470287323,
+      "learning_rate": 0.0001953779114079517,
+      "loss": 0.7273756265640259,
+      "step": 2410
+    },
+    {
+      "epoch": 1.0177215189873419,
+      "grad_norm": 0.4340079426765442,
+      "learning_rate": 0.00019536346720897823,
+      "loss": 0.7922506928443909,
+      "step": 2412
+    },
+    {
+      "epoch": 1.0185654008438818,
+      "grad_norm": 0.4518018662929535,
+      "learning_rate": 0.00019534900101148437,
+      "loss": 0.6492037773132324,
+      "step": 2414
+    },
+    {
+      "epoch": 1.019409282700422,
+      "grad_norm": 0.4160727560520172,
+      "learning_rate": 0.00019533451281880716,
+      "loss": 0.7422950863838196,
+      "step": 2416
+    },
+    {
+      "epoch": 1.0202531645569621,
+      "grad_norm": 0.35953211784362793,
+      "learning_rate": 0.00019532000263428883,
+      "loss": 0.6489207148551941,
+      "step": 2418
+    },
+    {
+      "epoch": 1.021097046413502,
+      "grad_norm": 0.38774433732032776,
+      "learning_rate": 0.0001953054704612765,
+      "loss": 0.7449417114257812,
+      "step": 2420
+    },
+    {
+      "epoch": 1.0219409282700422,
+      "grad_norm": 0.48185858130455017,
+      "learning_rate": 0.0001952909163031225,
+      "loss": 0.8481494188308716,
+      "step": 2422
+    },
+    {
+      "epoch": 1.0227848101265822,
+      "grad_norm": 0.4098491072654724,
+      "learning_rate": 0.00019527634016318425,
+      "loss": 0.7442291378974915,
+      "step": 2424
+    },
+    {
+      "epoch": 1.0236286919831223,
+      "grad_norm": 0.41284480690956116,
+      "learning_rate": 0.00019526174204482412,
+      "loss": 0.7564480900764465,
+      "step": 2426
+    },
+    {
+      "epoch": 1.0244725738396625,
+      "grad_norm": 0.44948479533195496,
+      "learning_rate": 0.00019524712195140965,
+      "loss": 0.7234854102134705,
+      "step": 2428
+    },
+    {
+      "epoch": 1.0253164556962024,
+      "grad_norm": 0.39939233660697937,
+      "learning_rate": 0.0001952324798863134,
+      "loss": 0.7638729810714722,
+      "step": 2430
+    },
+    {
+      "epoch": 1.0261603375527426,
+      "grad_norm": 0.40807482600212097,
+      "learning_rate": 0.00019521781585291298,
+      "loss": 0.6922630071640015,
+      "step": 2432
+    },
+    {
+      "epoch": 1.0270042194092828,
+      "grad_norm": 0.45735791325569153,
+      "learning_rate": 0.00019520312985459116,
+      "loss": 0.762996256351471,
+      "step": 2434
+    },
+    {
+      "epoch": 1.0278481012658227,
+      "grad_norm": 0.4280409812927246,
+      "learning_rate": 0.0001951884218947357,
+      "loss": 0.7811596989631653,
+      "step": 2436
+    },
+    {
+      "epoch": 1.0286919831223629,
+      "grad_norm": 0.41957560181617737,
+      "learning_rate": 0.00019517369197673943,
+      "loss": 0.7278305292129517,
+      "step": 2438
+    },
+    {
+      "epoch": 1.029535864978903,
+      "grad_norm": 0.4832586646080017,
+      "learning_rate": 0.00019515894010400028,
+      "loss": 0.807456910610199,
+      "step": 2440
+    },
+    {
+      "epoch": 1.030379746835443,
+      "grad_norm": 0.46448788046836853,
+      "learning_rate": 0.00019514416627992122,
+      "loss": 0.7421302795410156,
+      "step": 2442
+    },
+    {
+      "epoch": 1.0312236286919831,
+      "grad_norm": 0.41362619400024414,
+      "learning_rate": 0.00019512937050791024,
+      "loss": 0.6892213225364685,
+      "step": 2444
+    },
+    {
+      "epoch": 1.0320675105485233,
+      "grad_norm": 0.4158109128475189,
+      "learning_rate": 0.0001951145527913805,
+      "loss": 0.7305331826210022,
+      "step": 2446
+    },
+    {
+      "epoch": 1.0329113924050632,
+      "grad_norm": 0.4725087881088257,
+      "learning_rate": 0.00019509971313375011,
+      "loss": 0.7682753205299377,
+      "step": 2448
+    },
+    {
+      "epoch": 1.0337552742616034,
+      "grad_norm": 0.4341438114643097,
+      "learning_rate": 0.00019508485153844237,
+      "loss": 0.747041642665863,
+      "step": 2450
+    },
+    {
+      "epoch": 1.0345991561181433,
+      "grad_norm": 0.44515377283096313,
+      "learning_rate": 0.00019506996800888548,
+      "loss": 0.7312062382698059,
+      "step": 2452
+    },
+    {
+      "epoch": 1.0354430379746835,
+      "grad_norm": 0.5313915014266968,
+      "learning_rate": 0.00019505506254851283,
+      "loss": 0.7861107587814331,
+      "step": 2454
+    },
+    {
+      "epoch": 1.0362869198312237,
+      "grad_norm": 0.6094246506690979,
+      "learning_rate": 0.00019504013516076285,
+      "loss": 0.8458115458488464,
+      "step": 2456
+    },
+    {
+      "epoch": 1.0371308016877636,
+      "grad_norm": 0.36419153213500977,
+      "learning_rate": 0.00019502518584907895,
+      "loss": 0.6091256141662598,
+      "step": 2458
+    },
+    {
+      "epoch": 1.0379746835443038,
+      "grad_norm": 0.399271696805954,
+      "learning_rate": 0.00019501021461690966,
+      "loss": 0.6703370809555054,
+      "step": 2460
+    },
+    {
+      "epoch": 1.038818565400844,
+      "grad_norm": 0.38560932874679565,
+      "learning_rate": 0.00019499522146770856,
+      "loss": 0.7220943570137024,
+      "step": 2462
+    },
+    {
+      "epoch": 1.0396624472573839,
+      "grad_norm": 0.44199803471565247,
+      "learning_rate": 0.00019498020640493428,
+      "loss": 0.7501406073570251,
+      "step": 2464
+    },
+    {
+      "epoch": 1.040506329113924,
+      "grad_norm": 0.47394421696662903,
+      "learning_rate": 0.00019496516943205054,
+      "loss": 0.7957925796508789,
+      "step": 2466
+    },
+    {
+      "epoch": 1.0413502109704642,
+      "grad_norm": 0.5274425745010376,
+      "learning_rate": 0.00019495011055252603,
+      "loss": 0.7964824438095093,
+      "step": 2468
+    },
+    {
+      "epoch": 1.0421940928270041,
+      "grad_norm": 0.44004881381988525,
+      "learning_rate": 0.0001949350297698346,
+      "loss": 0.7478582262992859,
+      "step": 2470
+    },
+    {
+      "epoch": 1.0430379746835443,
+      "grad_norm": 0.40662670135498047,
+      "learning_rate": 0.000194919927087455,
+      "loss": 0.738339900970459,
+      "step": 2472
+    },
+    {
+      "epoch": 1.0438818565400845,
+      "grad_norm": 0.43934181332588196,
+      "learning_rate": 0.00019490480250887123,
+      "loss": 0.817171037197113,
+      "step": 2474
+    },
+    {
+      "epoch": 1.0447257383966244,
+      "grad_norm": 0.4111221432685852,
+      "learning_rate": 0.00019488965603757223,
+      "loss": 0.7413169145584106,
+      "step": 2476
+    },
+    {
+      "epoch": 1.0455696202531646,
+      "grad_norm": 0.3770996630191803,
+      "learning_rate": 0.00019487448767705194,
+      "loss": 0.7893102169036865,
+      "step": 2478
+    },
+    {
+      "epoch": 1.0464135021097047,
+      "grad_norm": 0.4524461627006531,
+      "learning_rate": 0.00019485929743080945,
+      "loss": 0.7037841081619263,
+      "step": 2480
+    },
+    {
+      "epoch": 1.0472573839662447,
+      "grad_norm": 0.4232378900051117,
+      "learning_rate": 0.00019484408530234885,
+      "loss": 0.7456525564193726,
+      "step": 2482
+    },
+    {
+      "epoch": 1.0481012658227848,
+      "grad_norm": 0.3842810392379761,
+      "learning_rate": 0.00019482885129517929,
+      "loss": 0.7263808250427246,
+      "step": 2484
+    },
+    {
+      "epoch": 1.048945147679325,
+      "grad_norm": 0.43203726410865784,
+      "learning_rate": 0.00019481359541281495,
+      "loss": 0.7101474404335022,
+      "step": 2486
+    },
+    {
+      "epoch": 1.049789029535865,
+      "grad_norm": 0.46944454312324524,
+      "learning_rate": 0.00019479831765877509,
+      "loss": 0.8561921715736389,
+      "step": 2488
+    },
+    {
+      "epoch": 1.0506329113924051,
+      "grad_norm": 0.4455019533634186,
+      "learning_rate": 0.00019478301803658397,
+      "loss": 0.7206679582595825,
+      "step": 2490
+    },
+    {
+      "epoch": 1.051476793248945,
+      "grad_norm": 0.4253813326358795,
+      "learning_rate": 0.00019476769654977095,
+      "loss": 0.7271745800971985,
+      "step": 2492
+    },
+    {
+      "epoch": 1.0523206751054852,
+      "grad_norm": 0.42758429050445557,
+      "learning_rate": 0.00019475235320187033,
+      "loss": 0.7358576059341431,
+      "step": 2494
+    },
+    {
+      "epoch": 1.0531645569620254,
+      "grad_norm": 0.400634765625,
+      "learning_rate": 0.00019473698799642164,
+      "loss": 0.6617647409439087,
+      "step": 2496
+    },
+    {
+      "epoch": 1.0540084388185653,
+      "grad_norm": 0.4132104814052582,
+      "learning_rate": 0.00019472160093696926,
+      "loss": 0.6900970935821533,
+      "step": 2498
+    },
+    {
+      "epoch": 1.0548523206751055,
+      "grad_norm": 0.5116495490074158,
+      "learning_rate": 0.00019470619202706268,
+      "loss": 0.7352861762046814,
+      "step": 2500
+    },
+    {
+      "epoch": 1.0548523206751055,
+      "eval_loss": 0.8249453902244568,
+      "eval_runtime": 679.4446,
+      "eval_samples_per_second": 3.101,
+      "eval_steps_per_second": 3.101,
+      "step": 2500
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 14220,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 6,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 5,
+        "early_stopping_threshold": 0.001
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 1
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.4948168708894147e+18,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-2500/training_args.bin b/sft_qwen_14B_v2/checkpoints/checkpoint-2500/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..0bd5f45757bdc6c7fc2d67718b18675b51983e0e
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-2500/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:afae4a8ce1391f149eb00b14eed8f891c715e892ea546bf754d22db2c2bc5969
+size 4792
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-3000/README.md b/sft_qwen_14B_v2/checkpoints/checkpoint-3000/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..134a3a8531b68adda3fed2a98db824f7c1b7dcfd
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-3000/README.md
@@ -0,0 +1,207 @@
+---
+base_model: Models/Qwen2.5-Coder-14B-CPT
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Models/Qwen2.5-Coder-14B-CPT
+- lora
+- transformers
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.0
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-3000/adapter_config.json b/sft_qwen_14B_v2/checkpoints/checkpoint-3000/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..d6316742bc71d30704e2ccdb9080ed0b8062e454
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-3000/adapter_config.json
@@ -0,0 +1,43 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Models/Qwen2.5-Coder-14B-CPT",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.0",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "v_proj",
+    "k_proj",
+    "q_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-3000/adapter_model.safetensors b/sft_qwen_14B_v2/checkpoints/checkpoint-3000/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..0560af4c177c87f98a0bcad344194f26b5433b54
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-3000/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84541ff7632f88bb0f7123c3f0560425ec3bb43bb106c7a7e32e03063e130ed0
+size 201378736
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-3000/optimizer.pt b/sft_qwen_14B_v2/checkpoints/checkpoint-3000/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..377551f3bfe4c64c7a8fab135edc06c8857fabba
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-3000/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b13b0455977b17d0e21b9257bce6ddb3bc7913419710555f0c98d7f56663e413
+size 402976786
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-3000/rng_state.pth b/sft_qwen_14B_v2/checkpoints/checkpoint-3000/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..c6a0eb8a6cdc768af3c7bc07b1bb735a8563ddaa
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-3000/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c0c19d6218afb9104df41c5b27e6edf90eb6e07a47e8b43e4f82ac1a8a44b8eb
+size 14244
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-3000/scheduler.pt b/sft_qwen_14B_v2/checkpoints/checkpoint-3000/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..846d0082b22e5db07b046e3d7b616641094be470
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-3000/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:608dbd365b0ca511f27dbb9deb12e1f149332571e34f6a5c04cfaf79a3d71f1f
+size 1064
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-3000/trainer_state.json b/sft_qwen_14B_v2/checkpoints/checkpoint-3000/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..3300ea033360b31f88100bef89f3fe03aa778306
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-3000/trainer_state.json
@@ -0,0 +1,10783 @@
+{
+  "best_global_step": 3000,
+  "best_metric": 0.8033165335655212,
+  "best_model_checkpoint": "task2file/sft_qwen_14B_v2/checkpoints/checkpoint-3000",
+  "epoch": 1.2658227848101267,
+  "eval_steps": 100,
+  "global_step": 3000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0008438818565400844,
+      "grad_norm": 0.5386583805084229,
+      "learning_rate": 1.7574692442882248e-07,
+      "loss": 1.6941628456115723,
+      "step": 2
+    },
+    {
+      "epoch": 0.0016877637130801688,
+      "grad_norm": 0.5477277636528015,
+      "learning_rate": 5.272407732864675e-07,
+      "loss": 1.7132279872894287,
+      "step": 4
+    },
+    {
+      "epoch": 0.002531645569620253,
+      "grad_norm": 0.5390765070915222,
+      "learning_rate": 8.787346221441126e-07,
+      "loss": 1.641180396080017,
+      "step": 6
+    },
+    {
+      "epoch": 0.0033755274261603376,
+      "grad_norm": 0.5023683905601501,
+      "learning_rate": 1.2302284710017575e-06,
+      "loss": 1.5616240501403809,
+      "step": 8
+    },
+    {
+      "epoch": 0.004219409282700422,
+      "grad_norm": 0.4899154603481293,
+      "learning_rate": 1.5817223198594026e-06,
+      "loss": 1.572033405303955,
+      "step": 10
+    },
+    {
+      "epoch": 0.005063291139240506,
+      "grad_norm": 0.5239788293838501,
+      "learning_rate": 1.9332161687170474e-06,
+      "loss": 1.6242921352386475,
+      "step": 12
+    },
+    {
+      "epoch": 0.00590717299578059,
+      "grad_norm": 0.5172926783561707,
+      "learning_rate": 2.2847100175746925e-06,
+      "loss": 1.6800041198730469,
+      "step": 14
+    },
+    {
+      "epoch": 0.006751054852320675,
+      "grad_norm": 0.5539224743843079,
+      "learning_rate": 2.6362038664323376e-06,
+      "loss": 1.6450834274291992,
+      "step": 16
+    },
+    {
+      "epoch": 0.007594936708860759,
+      "grad_norm": 0.5255337953567505,
+      "learning_rate": 2.9876977152899827e-06,
+      "loss": 1.6673263311386108,
+      "step": 18
+    },
+    {
+      "epoch": 0.008438818565400843,
+      "grad_norm": 0.5074548721313477,
+      "learning_rate": 3.3391915641476277e-06,
+      "loss": 1.531802773475647,
+      "step": 20
+    },
+    {
+      "epoch": 0.009282700421940928,
+      "grad_norm": 0.4160279333591461,
+      "learning_rate": 3.6906854130052724e-06,
+      "loss": 1.599354863166809,
+      "step": 22
+    },
+    {
+      "epoch": 0.010126582278481013,
+      "grad_norm": 0.5716474652290344,
+      "learning_rate": 4.0421792618629174e-06,
+      "loss": 1.6700962781906128,
+      "step": 24
+    },
+    {
+      "epoch": 0.010970464135021098,
+      "grad_norm": 0.5148899555206299,
+      "learning_rate": 4.3936731107205625e-06,
+      "loss": 1.66217839717865,
+      "step": 26
+    },
+    {
+      "epoch": 0.01181434599156118,
+      "grad_norm": 0.575722336769104,
+      "learning_rate": 4.7451669595782076e-06,
+      "loss": 1.6692266464233398,
+      "step": 28
+    },
+    {
+      "epoch": 0.012658227848101266,
+      "grad_norm": 0.5345953106880188,
+      "learning_rate": 5.096660808435853e-06,
+      "loss": 1.5518689155578613,
+      "step": 30
+    },
+    {
+      "epoch": 0.01350210970464135,
+      "grad_norm": 0.4462043344974518,
+      "learning_rate": 5.448154657293498e-06,
+      "loss": 1.5930007696151733,
+      "step": 32
+    },
+    {
+      "epoch": 0.014345991561181435,
+      "grad_norm": 0.5119605660438538,
+      "learning_rate": 5.799648506151143e-06,
+      "loss": 1.6069684028625488,
+      "step": 34
+    },
+    {
+      "epoch": 0.015189873417721518,
+      "grad_norm": 0.5328608751296997,
+      "learning_rate": 6.151142355008788e-06,
+      "loss": 1.5838109254837036,
+      "step": 36
+    },
+    {
+      "epoch": 0.016033755274261603,
+      "grad_norm": 0.5065920352935791,
+      "learning_rate": 6.502636203866433e-06,
+      "loss": 1.608130931854248,
+      "step": 38
+    },
+    {
+      "epoch": 0.016877637130801686,
+      "grad_norm": 0.4479359984397888,
+      "learning_rate": 6.854130052724078e-06,
+      "loss": 1.5942182540893555,
+      "step": 40
+    },
+    {
+      "epoch": 0.017721518987341773,
+      "grad_norm": 0.42844903469085693,
+      "learning_rate": 7.205623901581722e-06,
+      "loss": 1.6441553831100464,
+      "step": 42
+    },
+    {
+      "epoch": 0.018565400843881856,
+      "grad_norm": 0.476630836725235,
+      "learning_rate": 7.557117750439367e-06,
+      "loss": 1.6068111658096313,
+      "step": 44
+    },
+    {
+      "epoch": 0.019409282700421943,
+      "grad_norm": 0.4532654881477356,
+      "learning_rate": 7.908611599297012e-06,
+      "loss": 1.6618021726608276,
+      "step": 46
+    },
+    {
+      "epoch": 0.020253164556962026,
+      "grad_norm": 0.3701118230819702,
+      "learning_rate": 8.260105448154657e-06,
+      "loss": 1.4730033874511719,
+      "step": 48
+    },
+    {
+      "epoch": 0.02109704641350211,
+      "grad_norm": 0.38471561670303345,
+      "learning_rate": 8.611599297012302e-06,
+      "loss": 1.4828267097473145,
+      "step": 50
+    },
+    {
+      "epoch": 0.021940928270042195,
+      "grad_norm": 0.3602336347103119,
+      "learning_rate": 8.963093145869948e-06,
+      "loss": 1.3877452611923218,
+      "step": 52
+    },
+    {
+      "epoch": 0.02278481012658228,
+      "grad_norm": 0.40318572521209717,
+      "learning_rate": 9.314586994727593e-06,
+      "loss": 1.49052894115448,
+      "step": 54
+    },
+    {
+      "epoch": 0.02362869198312236,
+      "grad_norm": 0.3223826587200165,
+      "learning_rate": 9.666080843585238e-06,
+      "loss": 1.4912524223327637,
+      "step": 56
+    },
+    {
+      "epoch": 0.024472573839662448,
+      "grad_norm": 0.3873065114021301,
+      "learning_rate": 1.0017574692442883e-05,
+      "loss": 1.526674509048462,
+      "step": 58
+    },
+    {
+      "epoch": 0.02531645569620253,
+      "grad_norm": 0.410159707069397,
+      "learning_rate": 1.0369068541300528e-05,
+      "loss": 1.4480271339416504,
+      "step": 60
+    },
+    {
+      "epoch": 0.026160337552742614,
+      "grad_norm": 0.3632003962993622,
+      "learning_rate": 1.0720562390158173e-05,
+      "loss": 1.4222990274429321,
+      "step": 62
+    },
+    {
+      "epoch": 0.0270042194092827,
+      "grad_norm": 0.33118435740470886,
+      "learning_rate": 1.1072056239015818e-05,
+      "loss": 1.387171745300293,
+      "step": 64
+    },
+    {
+      "epoch": 0.027848101265822784,
+      "grad_norm": 0.3301764726638794,
+      "learning_rate": 1.1423550087873463e-05,
+      "loss": 1.3523777723312378,
+      "step": 66
+    },
+    {
+      "epoch": 0.02869198312236287,
+      "grad_norm": 0.34342435002326965,
+      "learning_rate": 1.1775043936731108e-05,
+      "loss": 1.4515162706375122,
+      "step": 68
+    },
+    {
+      "epoch": 0.029535864978902954,
+      "grad_norm": 0.3243122100830078,
+      "learning_rate": 1.2126537785588753e-05,
+      "loss": 1.3509243726730347,
+      "step": 70
+    },
+    {
+      "epoch": 0.030379746835443037,
+      "grad_norm": 0.3450150787830353,
+      "learning_rate": 1.2478031634446398e-05,
+      "loss": 1.4936245679855347,
+      "step": 72
+    },
+    {
+      "epoch": 0.031223628691983123,
+      "grad_norm": 0.38912028074264526,
+      "learning_rate": 1.2829525483304042e-05,
+      "loss": 1.3419109582901,
+      "step": 74
+    },
+    {
+      "epoch": 0.032067510548523206,
+      "grad_norm": 0.3019310235977173,
+      "learning_rate": 1.3181019332161687e-05,
+      "loss": 1.4284154176712036,
+      "step": 76
+    },
+    {
+      "epoch": 0.03291139240506329,
+      "grad_norm": 0.37803682684898376,
+      "learning_rate": 1.3532513181019332e-05,
+      "loss": 1.4256561994552612,
+      "step": 78
+    },
+    {
+      "epoch": 0.03375527426160337,
+      "grad_norm": 0.34191736578941345,
+      "learning_rate": 1.3884007029876977e-05,
+      "loss": 1.3256909847259521,
+      "step": 80
+    },
+    {
+      "epoch": 0.03459915611814346,
+      "grad_norm": 0.35242700576782227,
+      "learning_rate": 1.4235500878734624e-05,
+      "loss": 1.2710685729980469,
+      "step": 82
+    },
+    {
+      "epoch": 0.035443037974683546,
+      "grad_norm": 0.38094228506088257,
+      "learning_rate": 1.4586994727592269e-05,
+      "loss": 1.253411889076233,
+      "step": 84
+    },
+    {
+      "epoch": 0.036286919831223625,
+      "grad_norm": 0.36837366223335266,
+      "learning_rate": 1.4938488576449914e-05,
+      "loss": 1.3064342737197876,
+      "step": 86
+    },
+    {
+      "epoch": 0.03713080168776371,
+      "grad_norm": 0.3443569242954254,
+      "learning_rate": 1.5289982425307557e-05,
+      "loss": 1.293562412261963,
+      "step": 88
+    },
+    {
+      "epoch": 0.0379746835443038,
+      "grad_norm": 0.3799338936805725,
+      "learning_rate": 1.5641476274165202e-05,
+      "loss": 1.3382648229599,
+      "step": 90
+    },
+    {
+      "epoch": 0.038818565400843885,
+      "grad_norm": 0.40501922369003296,
+      "learning_rate": 1.599297012302285e-05,
+      "loss": 1.3925724029541016,
+      "step": 92
+    },
+    {
+      "epoch": 0.039662447257383965,
+      "grad_norm": 0.4419630467891693,
+      "learning_rate": 1.6344463971880492e-05,
+      "loss": 1.357171893119812,
+      "step": 94
+    },
+    {
+      "epoch": 0.04050632911392405,
+      "grad_norm": 0.3619817793369293,
+      "learning_rate": 1.6695957820738137e-05,
+      "loss": 1.3029985427856445,
+      "step": 96
+    },
+    {
+      "epoch": 0.04135021097046414,
+      "grad_norm": 0.4851357340812683,
+      "learning_rate": 1.7047451669595782e-05,
+      "loss": 1.3498191833496094,
+      "step": 98
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "grad_norm": 0.418658584356308,
+      "learning_rate": 1.7398945518453427e-05,
+      "loss": 1.185287356376648,
+      "step": 100
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "eval_loss": 1.2979938983917236,
+      "eval_runtime": 682.1979,
+      "eval_samples_per_second": 3.089,
+      "eval_steps_per_second": 3.089,
+      "step": 100
+    },
+    {
+      "epoch": 0.043037974683544304,
+      "grad_norm": 0.4464418888092041,
+      "learning_rate": 1.7750439367311073e-05,
+      "loss": 1.2217272520065308,
+      "step": 102
+    },
+    {
+      "epoch": 0.04388185654008439,
+      "grad_norm": 0.4706237316131592,
+      "learning_rate": 1.8101933216168718e-05,
+      "loss": 1.2052050828933716,
+      "step": 104
+    },
+    {
+      "epoch": 0.04472573839662447,
+      "grad_norm": 0.46394404768943787,
+      "learning_rate": 1.8453427065026363e-05,
+      "loss": 1.221343994140625,
+      "step": 106
+    },
+    {
+      "epoch": 0.04556962025316456,
+      "grad_norm": 0.4726889431476593,
+      "learning_rate": 1.8804920913884008e-05,
+      "loss": 1.2387475967407227,
+      "step": 108
+    },
+    {
+      "epoch": 0.046413502109704644,
+      "grad_norm": 0.42130985856056213,
+      "learning_rate": 1.9156414762741653e-05,
+      "loss": 1.2851309776306152,
+      "step": 110
+    },
+    {
+      "epoch": 0.04725738396624472,
+      "grad_norm": 0.4504576623439789,
+      "learning_rate": 1.9507908611599298e-05,
+      "loss": 1.2753145694732666,
+      "step": 112
+    },
+    {
+      "epoch": 0.04810126582278481,
+      "grad_norm": 0.396085262298584,
+      "learning_rate": 1.9859402460456943e-05,
+      "loss": 1.2427717447280884,
+      "step": 114
+    },
+    {
+      "epoch": 0.048945147679324896,
+      "grad_norm": 0.5106491446495056,
+      "learning_rate": 2.0210896309314588e-05,
+      "loss": 1.2943825721740723,
+      "step": 116
+    },
+    {
+      "epoch": 0.049789029535864976,
+      "grad_norm": 0.42351317405700684,
+      "learning_rate": 2.0562390158172233e-05,
+      "loss": 1.263301134109497,
+      "step": 118
+    },
+    {
+      "epoch": 0.05063291139240506,
+      "grad_norm": 0.4403539299964905,
+      "learning_rate": 2.0913884007029878e-05,
+      "loss": 1.2647849321365356,
+      "step": 120
+    },
+    {
+      "epoch": 0.05147679324894515,
+      "grad_norm": 0.5260752439498901,
+      "learning_rate": 2.1265377855887523e-05,
+      "loss": 1.2351393699645996,
+      "step": 122
+    },
+    {
+      "epoch": 0.05232067510548523,
+      "grad_norm": 0.44978851079940796,
+      "learning_rate": 2.1616871704745168e-05,
+      "loss": 1.0384471416473389,
+      "step": 124
+    },
+    {
+      "epoch": 0.053164556962025315,
+      "grad_norm": 0.47732362151145935,
+      "learning_rate": 2.1968365553602813e-05,
+      "loss": 1.1518068313598633,
+      "step": 126
+    },
+    {
+      "epoch": 0.0540084388185654,
+      "grad_norm": 0.5473551750183105,
+      "learning_rate": 2.231985940246046e-05,
+      "loss": 1.2264912128448486,
+      "step": 128
+    },
+    {
+      "epoch": 0.05485232067510549,
+      "grad_norm": 0.4473855197429657,
+      "learning_rate": 2.2671353251318103e-05,
+      "loss": 1.1615246534347534,
+      "step": 130
+    },
+    {
+      "epoch": 0.05569620253164557,
+      "grad_norm": 0.5980377197265625,
+      "learning_rate": 2.302284710017575e-05,
+      "loss": 1.1334880590438843,
+      "step": 132
+    },
+    {
+      "epoch": 0.056540084388185655,
+      "grad_norm": 0.5987792015075684,
+      "learning_rate": 2.3374340949033394e-05,
+      "loss": 1.1546804904937744,
+      "step": 134
+    },
+    {
+      "epoch": 0.05738396624472574,
+      "grad_norm": 0.45355498790740967,
+      "learning_rate": 2.372583479789104e-05,
+      "loss": 1.194953441619873,
+      "step": 136
+    },
+    {
+      "epoch": 0.05822784810126582,
+      "grad_norm": 0.5373698472976685,
+      "learning_rate": 2.4077328646748684e-05,
+      "loss": 1.1067466735839844,
+      "step": 138
+    },
+    {
+      "epoch": 0.05907172995780591,
+      "grad_norm": 0.48734328150749207,
+      "learning_rate": 2.442882249560633e-05,
+      "loss": 1.188468098640442,
+      "step": 140
+    },
+    {
+      "epoch": 0.059915611814345994,
+      "grad_norm": 0.4692173898220062,
+      "learning_rate": 2.478031634446397e-05,
+      "loss": 1.1624362468719482,
+      "step": 142
+    },
+    {
+      "epoch": 0.060759493670886074,
+      "grad_norm": 0.532554030418396,
+      "learning_rate": 2.513181019332162e-05,
+      "loss": 1.0978907346725464,
+      "step": 144
+    },
+    {
+      "epoch": 0.06160337552742616,
+      "grad_norm": 0.5853802561759949,
+      "learning_rate": 2.5483304042179264e-05,
+      "loss": 1.2030781507492065,
+      "step": 146
+    },
+    {
+      "epoch": 0.06244725738396625,
+      "grad_norm": 0.5061611533164978,
+      "learning_rate": 2.583479789103691e-05,
+      "loss": 1.082366943359375,
+      "step": 148
+    },
+    {
+      "epoch": 0.06329113924050633,
+      "grad_norm": 0.49426141381263733,
+      "learning_rate": 2.6186291739894554e-05,
+      "loss": 1.10564386844635,
+      "step": 150
+    },
+    {
+      "epoch": 0.06413502109704641,
+      "grad_norm": 0.5846618413925171,
+      "learning_rate": 2.6537785588752196e-05,
+      "loss": 1.1992807388305664,
+      "step": 152
+    },
+    {
+      "epoch": 0.06497890295358649,
+      "grad_norm": 0.5517552495002747,
+      "learning_rate": 2.6889279437609844e-05,
+      "loss": 1.1757566928863525,
+      "step": 154
+    },
+    {
+      "epoch": 0.06582278481012659,
+      "grad_norm": 0.5667305588722229,
+      "learning_rate": 2.7240773286467486e-05,
+      "loss": 1.0548783540725708,
+      "step": 156
+    },
+    {
+      "epoch": 0.06666666666666667,
+      "grad_norm": 0.6760414242744446,
+      "learning_rate": 2.7592267135325134e-05,
+      "loss": 1.184364914894104,
+      "step": 158
+    },
+    {
+      "epoch": 0.06751054852320675,
+      "grad_norm": 0.5261430740356445,
+      "learning_rate": 2.7943760984182776e-05,
+      "loss": 1.1945042610168457,
+      "step": 160
+    },
+    {
+      "epoch": 0.06835443037974684,
+      "grad_norm": 0.6155015230178833,
+      "learning_rate": 2.8295254833040425e-05,
+      "loss": 1.2021973133087158,
+      "step": 162
+    },
+    {
+      "epoch": 0.06919831223628692,
+      "grad_norm": 0.6131619215011597,
+      "learning_rate": 2.8646748681898066e-05,
+      "loss": 1.144123911857605,
+      "step": 164
+    },
+    {
+      "epoch": 0.070042194092827,
+      "grad_norm": 0.5749185681343079,
+      "learning_rate": 2.8998242530755715e-05,
+      "loss": 1.1329256296157837,
+      "step": 166
+    },
+    {
+      "epoch": 0.07088607594936709,
+      "grad_norm": 0.5243118405342102,
+      "learning_rate": 2.9349736379613356e-05,
+      "loss": 1.0892387628555298,
+      "step": 168
+    },
+    {
+      "epoch": 0.07172995780590717,
+      "grad_norm": 0.7190104722976685,
+      "learning_rate": 2.9701230228471005e-05,
+      "loss": 1.163260817527771,
+      "step": 170
+    },
+    {
+      "epoch": 0.07257383966244725,
+      "grad_norm": 0.5486982464790344,
+      "learning_rate": 3.0052724077328647e-05,
+      "loss": 1.0880777835845947,
+      "step": 172
+    },
+    {
+      "epoch": 0.07341772151898734,
+      "grad_norm": 0.5020889043807983,
+      "learning_rate": 3.0404217926186295e-05,
+      "loss": 1.0433368682861328,
+      "step": 174
+    },
+    {
+      "epoch": 0.07426160337552742,
+      "grad_norm": 0.47329774498939514,
+      "learning_rate": 3.075571177504394e-05,
+      "loss": 1.0528991222381592,
+      "step": 176
+    },
+    {
+      "epoch": 0.0751054852320675,
+      "grad_norm": 0.6635547876358032,
+      "learning_rate": 3.110720562390158e-05,
+      "loss": 1.1627811193466187,
+      "step": 178
+    },
+    {
+      "epoch": 0.0759493670886076,
+      "grad_norm": 0.5624618530273438,
+      "learning_rate": 3.145869947275923e-05,
+      "loss": 1.084869384765625,
+      "step": 180
+    },
+    {
+      "epoch": 0.07679324894514768,
+      "grad_norm": 0.6029536724090576,
+      "learning_rate": 3.181019332161687e-05,
+      "loss": 1.2227671146392822,
+      "step": 182
+    },
+    {
+      "epoch": 0.07763713080168777,
+      "grad_norm": 0.930959939956665,
+      "learning_rate": 3.216168717047452e-05,
+      "loss": 1.0955452919006348,
+      "step": 184
+    },
+    {
+      "epoch": 0.07848101265822785,
+      "grad_norm": 0.5326952338218689,
+      "learning_rate": 3.251318101933216e-05,
+      "loss": 1.0640798807144165,
+      "step": 186
+    },
+    {
+      "epoch": 0.07932489451476793,
+      "grad_norm": 0.5484727621078491,
+      "learning_rate": 3.286467486818981e-05,
+      "loss": 1.0700589418411255,
+      "step": 188
+    },
+    {
+      "epoch": 0.08016877637130802,
+      "grad_norm": 0.605273425579071,
+      "learning_rate": 3.3216168717047456e-05,
+      "loss": 1.1593081951141357,
+      "step": 190
+    },
+    {
+      "epoch": 0.0810126582278481,
+      "grad_norm": 0.5704394578933716,
+      "learning_rate": 3.35676625659051e-05,
+      "loss": 1.1617076396942139,
+      "step": 192
+    },
+    {
+      "epoch": 0.08185654008438818,
+      "grad_norm": 0.5929452180862427,
+      "learning_rate": 3.3919156414762746e-05,
+      "loss": 1.1346839666366577,
+      "step": 194
+    },
+    {
+      "epoch": 0.08270042194092828,
+      "grad_norm": 0.5624077916145325,
+      "learning_rate": 3.427065026362039e-05,
+      "loss": 1.0934710502624512,
+      "step": 196
+    },
+    {
+      "epoch": 0.08354430379746836,
+      "grad_norm": 0.6717425584793091,
+      "learning_rate": 3.4622144112478036e-05,
+      "loss": 1.1810534000396729,
+      "step": 198
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "grad_norm": 0.5120199918746948,
+      "learning_rate": 3.4973637961335674e-05,
+      "loss": 1.1525514125823975,
+      "step": 200
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "eval_loss": 1.142486810684204,
+      "eval_runtime": 668.2356,
+      "eval_samples_per_second": 3.153,
+      "eval_steps_per_second": 3.153,
+      "step": 200
+    },
+    {
+      "epoch": 0.08523206751054853,
+      "grad_norm": 0.5144487023353577,
+      "learning_rate": 3.5325131810193326e-05,
+      "loss": 1.0243735313415527,
+      "step": 202
+    },
+    {
+      "epoch": 0.08607594936708861,
+      "grad_norm": 0.6325069069862366,
+      "learning_rate": 3.5676625659050964e-05,
+      "loss": 1.118743896484375,
+      "step": 204
+    },
+    {
+      "epoch": 0.08691983122362869,
+      "grad_norm": 0.5501633882522583,
+      "learning_rate": 3.6028119507908616e-05,
+      "loss": 1.0380504131317139,
+      "step": 206
+    },
+    {
+      "epoch": 0.08776371308016878,
+      "grad_norm": 0.6133899688720703,
+      "learning_rate": 3.6379613356766254e-05,
+      "loss": 0.9837555885314941,
+      "step": 208
+    },
+    {
+      "epoch": 0.08860759493670886,
+      "grad_norm": 0.5799810886383057,
+      "learning_rate": 3.6731107205623906e-05,
+      "loss": 1.090720295906067,
+      "step": 210
+    },
+    {
+      "epoch": 0.08945147679324894,
+      "grad_norm": 0.6039511561393738,
+      "learning_rate": 3.7082601054481544e-05,
+      "loss": 1.120232343673706,
+      "step": 212
+    },
+    {
+      "epoch": 0.09029535864978903,
+      "grad_norm": 0.5983024835586548,
+      "learning_rate": 3.7434094903339196e-05,
+      "loss": 1.096949815750122,
+      "step": 214
+    },
+    {
+      "epoch": 0.09113924050632911,
+      "grad_norm": 0.5641079545021057,
+      "learning_rate": 3.7785588752196835e-05,
+      "loss": 1.1226298809051514,
+      "step": 216
+    },
+    {
+      "epoch": 0.0919831223628692,
+      "grad_norm": 0.655717134475708,
+      "learning_rate": 3.8137082601054486e-05,
+      "loss": 1.1260643005371094,
+      "step": 218
+    },
+    {
+      "epoch": 0.09282700421940929,
+      "grad_norm": 0.6111898422241211,
+      "learning_rate": 3.848857644991213e-05,
+      "loss": 1.0777709484100342,
+      "step": 220
+    },
+    {
+      "epoch": 0.09367088607594937,
+      "grad_norm": 0.6821302771568298,
+      "learning_rate": 3.884007029876977e-05,
+      "loss": 1.10588800907135,
+      "step": 222
+    },
+    {
+      "epoch": 0.09451476793248945,
+      "grad_norm": 0.693175733089447,
+      "learning_rate": 3.919156414762742e-05,
+      "loss": 1.1498671770095825,
+      "step": 224
+    },
+    {
+      "epoch": 0.09535864978902954,
+      "grad_norm": 0.5288166403770447,
+      "learning_rate": 3.954305799648506e-05,
+      "loss": 1.0587562322616577,
+      "step": 226
+    },
+    {
+      "epoch": 0.09620253164556962,
+      "grad_norm": 0.6882867813110352,
+      "learning_rate": 3.989455184534271e-05,
+      "loss": 1.1107512712478638,
+      "step": 228
+    },
+    {
+      "epoch": 0.0970464135021097,
+      "grad_norm": 0.5834154486656189,
+      "learning_rate": 4.024604569420035e-05,
+      "loss": 1.020510196685791,
+      "step": 230
+    },
+    {
+      "epoch": 0.09789029535864979,
+      "grad_norm": 0.7157064080238342,
+      "learning_rate": 4.0597539543058e-05,
+      "loss": 1.0642449855804443,
+      "step": 232
+    },
+    {
+      "epoch": 0.09873417721518987,
+      "grad_norm": 0.6530708074569702,
+      "learning_rate": 4.094903339191564e-05,
+      "loss": 1.0359872579574585,
+      "step": 234
+    },
+    {
+      "epoch": 0.09957805907172995,
+      "grad_norm": 0.6329686045646667,
+      "learning_rate": 4.130052724077329e-05,
+      "loss": 1.050504446029663,
+      "step": 236
+    },
+    {
+      "epoch": 0.10042194092827005,
+      "grad_norm": 0.6597026586532593,
+      "learning_rate": 4.165202108963093e-05,
+      "loss": 1.2621175050735474,
+      "step": 238
+    },
+    {
+      "epoch": 0.10126582278481013,
+      "grad_norm": 0.6195225119590759,
+      "learning_rate": 4.200351493848858e-05,
+      "loss": 1.1218310594558716,
+      "step": 240
+    },
+    {
+      "epoch": 0.1021097046413502,
+      "grad_norm": 0.6764137744903564,
+      "learning_rate": 4.235500878734622e-05,
+      "loss": 1.1250728368759155,
+      "step": 242
+    },
+    {
+      "epoch": 0.1029535864978903,
+      "grad_norm": 0.552363395690918,
+      "learning_rate": 4.270650263620387e-05,
+      "loss": 1.028212308883667,
+      "step": 244
+    },
+    {
+      "epoch": 0.10379746835443038,
+      "grad_norm": 0.5620495676994324,
+      "learning_rate": 4.305799648506151e-05,
+      "loss": 1.0425450801849365,
+      "step": 246
+    },
+    {
+      "epoch": 0.10464135021097046,
+      "grad_norm": 0.6860032081604004,
+      "learning_rate": 4.3409490333919156e-05,
+      "loss": 1.144278883934021,
+      "step": 248
+    },
+    {
+      "epoch": 0.10548523206751055,
+      "grad_norm": 0.6033259034156799,
+      "learning_rate": 4.37609841827768e-05,
+      "loss": 1.1223982572555542,
+      "step": 250
+    },
+    {
+      "epoch": 0.10632911392405063,
+      "grad_norm": 0.6292146444320679,
+      "learning_rate": 4.4112478031634446e-05,
+      "loss": 1.1609960794448853,
+      "step": 252
+    },
+    {
+      "epoch": 0.10717299578059072,
+      "grad_norm": 0.7982883453369141,
+      "learning_rate": 4.44639718804921e-05,
+      "loss": 1.063547968864441,
+      "step": 254
+    },
+    {
+      "epoch": 0.1080168776371308,
+      "grad_norm": 0.7719110250473022,
+      "learning_rate": 4.4815465729349736e-05,
+      "loss": 1.0719804763793945,
+      "step": 256
+    },
+    {
+      "epoch": 0.10886075949367088,
+      "grad_norm": 0.6101011633872986,
+      "learning_rate": 4.516695957820739e-05,
+      "loss": 1.0778400897979736,
+      "step": 258
+    },
+    {
+      "epoch": 0.10970464135021098,
+      "grad_norm": 0.7300994396209717,
+      "learning_rate": 4.5518453427065026e-05,
+      "loss": 1.2129558324813843,
+      "step": 260
+    },
+    {
+      "epoch": 0.11054852320675106,
+      "grad_norm": 0.8348747491836548,
+      "learning_rate": 4.586994727592268e-05,
+      "loss": 1.221714735031128,
+      "step": 262
+    },
+    {
+      "epoch": 0.11139240506329114,
+      "grad_norm": 0.5445612072944641,
+      "learning_rate": 4.6221441124780316e-05,
+      "loss": 1.0187978744506836,
+      "step": 264
+    },
+    {
+      "epoch": 0.11223628691983123,
+      "grad_norm": 0.6230319738388062,
+      "learning_rate": 4.657293497363797e-05,
+      "loss": 1.096561312675476,
+      "step": 266
+    },
+    {
+      "epoch": 0.11308016877637131,
+      "grad_norm": 0.6231237649917603,
+      "learning_rate": 4.6924428822495606e-05,
+      "loss": 1.089842438697815,
+      "step": 268
+    },
+    {
+      "epoch": 0.11392405063291139,
+      "grad_norm": 0.7178627252578735,
+      "learning_rate": 4.727592267135325e-05,
+      "loss": 1.0696645975112915,
+      "step": 270
+    },
+    {
+      "epoch": 0.11476793248945148,
+      "grad_norm": 0.6895854473114014,
+      "learning_rate": 4.7627416520210896e-05,
+      "loss": 1.0511361360549927,
+      "step": 272
+    },
+    {
+      "epoch": 0.11561181434599156,
+      "grad_norm": 0.6046878695487976,
+      "learning_rate": 4.797891036906854e-05,
+      "loss": 1.1373958587646484,
+      "step": 274
+    },
+    {
+      "epoch": 0.11645569620253164,
+      "grad_norm": 0.6524552702903748,
+      "learning_rate": 4.833040421792619e-05,
+      "loss": 1.0734186172485352,
+      "step": 276
+    },
+    {
+      "epoch": 0.11729957805907174,
+      "grad_norm": 0.6331019997596741,
+      "learning_rate": 4.868189806678383e-05,
+      "loss": 1.123913049697876,
+      "step": 278
+    },
+    {
+      "epoch": 0.11814345991561181,
+      "grad_norm": 0.5919018983840942,
+      "learning_rate": 4.903339191564148e-05,
+      "loss": 1.0635710954666138,
+      "step": 280
+    },
+    {
+      "epoch": 0.1189873417721519,
+      "grad_norm": 0.6067633032798767,
+      "learning_rate": 4.938488576449912e-05,
+      "loss": 1.0429247617721558,
+      "step": 282
+    },
+    {
+      "epoch": 0.11983122362869199,
+      "grad_norm": 0.6583750247955322,
+      "learning_rate": 4.9736379613356774e-05,
+      "loss": 1.1397464275360107,
+      "step": 284
+    },
+    {
+      "epoch": 0.12067510548523207,
+      "grad_norm": 0.6200069785118103,
+      "learning_rate": 5.008787346221442e-05,
+      "loss": 1.0590803623199463,
+      "step": 286
+    },
+    {
+      "epoch": 0.12151898734177215,
+      "grad_norm": 0.6798665523529053,
+      "learning_rate": 5.0439367311072064e-05,
+      "loss": 1.1318789720535278,
+      "step": 288
+    },
+    {
+      "epoch": 0.12236286919831224,
+      "grad_norm": 0.7508794069290161,
+      "learning_rate": 5.07908611599297e-05,
+      "loss": 1.0934956073760986,
+      "step": 290
+    },
+    {
+      "epoch": 0.12320675105485232,
+      "grad_norm": 0.6901452541351318,
+      "learning_rate": 5.114235500878735e-05,
+      "loss": 1.163407802581787,
+      "step": 292
+    },
+    {
+      "epoch": 0.1240506329113924,
+      "grad_norm": 0.6423285603523254,
+      "learning_rate": 5.1493848857644985e-05,
+      "loss": 1.09059476852417,
+      "step": 294
+    },
+    {
+      "epoch": 0.1248945147679325,
+      "grad_norm": 0.6839275360107422,
+      "learning_rate": 5.1845342706502644e-05,
+      "loss": 1.0690211057662964,
+      "step": 296
+    },
+    {
+      "epoch": 0.1257383966244726,
+      "grad_norm": 0.6350128054618835,
+      "learning_rate": 5.219683655536028e-05,
+      "loss": 0.982322096824646,
+      "step": 298
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "grad_norm": 0.7136530876159668,
+      "learning_rate": 5.254833040421793e-05,
+      "loss": 1.1132930517196655,
+      "step": 300
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "eval_loss": 1.0952109098434448,
+      "eval_runtime": 677.0652,
+      "eval_samples_per_second": 3.112,
+      "eval_steps_per_second": 3.112,
+      "step": 300
+    },
+    {
+      "epoch": 0.12742616033755275,
+      "grad_norm": 0.7339721322059631,
+      "learning_rate": 5.289982425307557e-05,
+      "loss": 0.973595917224884,
+      "step": 302
+    },
+    {
+      "epoch": 0.12827004219409283,
+      "grad_norm": 0.5941481590270996,
+      "learning_rate": 5.3251318101933224e-05,
+      "loss": 0.9819849729537964,
+      "step": 304
+    },
+    {
+      "epoch": 0.1291139240506329,
+      "grad_norm": 0.7153938412666321,
+      "learning_rate": 5.360281195079086e-05,
+      "loss": 1.0315470695495605,
+      "step": 306
+    },
+    {
+      "epoch": 0.12995780590717299,
+      "grad_norm": 0.5167180299758911,
+      "learning_rate": 5.395430579964851e-05,
+      "loss": 0.9492001533508301,
+      "step": 308
+    },
+    {
+      "epoch": 0.1308016877637131,
+      "grad_norm": 0.6055944561958313,
+      "learning_rate": 5.430579964850615e-05,
+      "loss": 1.0156209468841553,
+      "step": 310
+    },
+    {
+      "epoch": 0.13164556962025317,
+      "grad_norm": 0.7662386298179626,
+      "learning_rate": 5.4657293497363805e-05,
+      "loss": 1.1791651248931885,
+      "step": 312
+    },
+    {
+      "epoch": 0.13248945147679325,
+      "grad_norm": 0.6065546274185181,
+      "learning_rate": 5.500878734622145e-05,
+      "loss": 1.0009297132492065,
+      "step": 314
+    },
+    {
+      "epoch": 0.13333333333333333,
+      "grad_norm": 0.604225754737854,
+      "learning_rate": 5.536028119507909e-05,
+      "loss": 1.0208244323730469,
+      "step": 316
+    },
+    {
+      "epoch": 0.1341772151898734,
+      "grad_norm": 0.6186763048171997,
+      "learning_rate": 5.571177504393673e-05,
+      "loss": 0.9968416690826416,
+      "step": 318
+    },
+    {
+      "epoch": 0.1350210970464135,
+      "grad_norm": 0.7100363969802856,
+      "learning_rate": 5.606326889279437e-05,
+      "loss": 0.9540256857872009,
+      "step": 320
+    },
+    {
+      "epoch": 0.1358649789029536,
+      "grad_norm": 0.6979711055755615,
+      "learning_rate": 5.641476274165203e-05,
+      "loss": 1.0631953477859497,
+      "step": 322
+    },
+    {
+      "epoch": 0.13670886075949368,
+      "grad_norm": 0.6237109303474426,
+      "learning_rate": 5.676625659050967e-05,
+      "loss": 1.0170501470565796,
+      "step": 324
+    },
+    {
+      "epoch": 0.13755274261603376,
+      "grad_norm": 0.6525548696517944,
+      "learning_rate": 5.711775043936731e-05,
+      "loss": 1.0715603828430176,
+      "step": 326
+    },
+    {
+      "epoch": 0.13839662447257384,
+      "grad_norm": 0.6869221329689026,
+      "learning_rate": 5.746924428822495e-05,
+      "loss": 1.0111541748046875,
+      "step": 328
+    },
+    {
+      "epoch": 0.13924050632911392,
+      "grad_norm": 0.553188145160675,
+      "learning_rate": 5.782073813708261e-05,
+      "loss": 1.0311682224273682,
+      "step": 330
+    },
+    {
+      "epoch": 0.140084388185654,
+      "grad_norm": 0.6760852932929993,
+      "learning_rate": 5.817223198594025e-05,
+      "loss": 1.0213634967803955,
+      "step": 332
+    },
+    {
+      "epoch": 0.1409282700421941,
+      "grad_norm": 0.5907419919967651,
+      "learning_rate": 5.8523725834797894e-05,
+      "loss": 0.9748594164848328,
+      "step": 334
+    },
+    {
+      "epoch": 0.14177215189873418,
+      "grad_norm": 0.7044920921325684,
+      "learning_rate": 5.887521968365554e-05,
+      "loss": 1.05863356590271,
+      "step": 336
+    },
+    {
+      "epoch": 0.14261603375527426,
+      "grad_norm": 0.679073691368103,
+      "learning_rate": 5.922671353251318e-05,
+      "loss": 1.1341127157211304,
+      "step": 338
+    },
+    {
+      "epoch": 0.14345991561181434,
+      "grad_norm": 0.7676237225532532,
+      "learning_rate": 5.957820738137083e-05,
+      "loss": 0.9540836215019226,
+      "step": 340
+    },
+    {
+      "epoch": 0.14430379746835442,
+      "grad_norm": 0.6313899755477905,
+      "learning_rate": 5.9929701230228474e-05,
+      "loss": 1.0585911273956299,
+      "step": 342
+    },
+    {
+      "epoch": 0.1451476793248945,
+      "grad_norm": 0.7123099565505981,
+      "learning_rate": 6.028119507908612e-05,
+      "loss": 1.0760118961334229,
+      "step": 344
+    },
+    {
+      "epoch": 0.1459915611814346,
+      "grad_norm": 0.585935652256012,
+      "learning_rate": 6.063268892794376e-05,
+      "loss": 1.036866307258606,
+      "step": 346
+    },
+    {
+      "epoch": 0.1468354430379747,
+      "grad_norm": 0.5643263459205627,
+      "learning_rate": 6.0984182776801416e-05,
+      "loss": 1.0242938995361328,
+      "step": 348
+    },
+    {
+      "epoch": 0.14767932489451477,
+      "grad_norm": 0.626761794090271,
+      "learning_rate": 6.133567662565906e-05,
+      "loss": 1.0497376918792725,
+      "step": 350
+    },
+    {
+      "epoch": 0.14852320675105485,
+      "grad_norm": 0.5106956958770752,
+      "learning_rate": 6.16871704745167e-05,
+      "loss": 0.9811885356903076,
+      "step": 352
+    },
+    {
+      "epoch": 0.14936708860759493,
+      "grad_norm": 0.6948089003562927,
+      "learning_rate": 6.203866432337434e-05,
+      "loss": 1.0715330839157104,
+      "step": 354
+    },
+    {
+      "epoch": 0.150210970464135,
+      "grad_norm": 0.699713945388794,
+      "learning_rate": 6.239015817223199e-05,
+      "loss": 1.0405226945877075,
+      "step": 356
+    },
+    {
+      "epoch": 0.15105485232067511,
+      "grad_norm": 0.6437667012214661,
+      "learning_rate": 6.274165202108964e-05,
+      "loss": 1.0490930080413818,
+      "step": 358
+    },
+    {
+      "epoch": 0.1518987341772152,
+      "grad_norm": 0.6952699422836304,
+      "learning_rate": 6.309314586994728e-05,
+      "loss": 0.9267548322677612,
+      "step": 360
+    },
+    {
+      "epoch": 0.15274261603375527,
+      "grad_norm": 0.6713186502456665,
+      "learning_rate": 6.344463971880492e-05,
+      "loss": 1.0427420139312744,
+      "step": 362
+    },
+    {
+      "epoch": 0.15358649789029535,
+      "grad_norm": 0.6750379800796509,
+      "learning_rate": 6.379613356766257e-05,
+      "loss": 1.048950433731079,
+      "step": 364
+    },
+    {
+      "epoch": 0.15443037974683543,
+      "grad_norm": 0.6053379774093628,
+      "learning_rate": 6.414762741652022e-05,
+      "loss": 1.0156004428863525,
+      "step": 366
+    },
+    {
+      "epoch": 0.15527426160337554,
+      "grad_norm": 0.8063633441925049,
+      "learning_rate": 6.449912126537786e-05,
+      "loss": 1.0020819902420044,
+      "step": 368
+    },
+    {
+      "epoch": 0.15611814345991562,
+      "grad_norm": 0.8027494549751282,
+      "learning_rate": 6.48506151142355e-05,
+      "loss": 1.055633783340454,
+      "step": 370
+    },
+    {
+      "epoch": 0.1569620253164557,
+      "grad_norm": 0.6580121517181396,
+      "learning_rate": 6.520210896309315e-05,
+      "loss": 1.0149940252304077,
+      "step": 372
+    },
+    {
+      "epoch": 0.15780590717299578,
+      "grad_norm": 0.6561233997344971,
+      "learning_rate": 6.55536028119508e-05,
+      "loss": 0.9769611954689026,
+      "step": 374
+    },
+    {
+      "epoch": 0.15864978902953586,
+      "grad_norm": 0.6444346308708191,
+      "learning_rate": 6.590509666080844e-05,
+      "loss": 0.9099349975585938,
+      "step": 376
+    },
+    {
+      "epoch": 0.15949367088607594,
+      "grad_norm": 0.5879359245300293,
+      "learning_rate": 6.625659050966608e-05,
+      "loss": 1.0797548294067383,
+      "step": 378
+    },
+    {
+      "epoch": 0.16033755274261605,
+      "grad_norm": 0.6994144916534424,
+      "learning_rate": 6.660808435852373e-05,
+      "loss": 1.0336791276931763,
+      "step": 380
+    },
+    {
+      "epoch": 0.16118143459915613,
+      "grad_norm": 0.6128669381141663,
+      "learning_rate": 6.695957820738138e-05,
+      "loss": 1.018118143081665,
+      "step": 382
+    },
+    {
+      "epoch": 0.1620253164556962,
+      "grad_norm": 1.0237540006637573,
+      "learning_rate": 6.731107205623902e-05,
+      "loss": 1.1405497789382935,
+      "step": 384
+    },
+    {
+      "epoch": 0.16286919831223629,
+      "grad_norm": 0.6091578006744385,
+      "learning_rate": 6.766256590509666e-05,
+      "loss": 1.0314189195632935,
+      "step": 386
+    },
+    {
+      "epoch": 0.16371308016877636,
+      "grad_norm": 0.5916037559509277,
+      "learning_rate": 6.801405975395431e-05,
+      "loss": 0.9564052820205688,
+      "step": 388
+    },
+    {
+      "epoch": 0.16455696202531644,
+      "grad_norm": 0.771653950214386,
+      "learning_rate": 6.836555360281195e-05,
+      "loss": 1.0023859739303589,
+      "step": 390
+    },
+    {
+      "epoch": 0.16540084388185655,
+      "grad_norm": 0.654658317565918,
+      "learning_rate": 6.87170474516696e-05,
+      "loss": 1.07024085521698,
+      "step": 392
+    },
+    {
+      "epoch": 0.16624472573839663,
+      "grad_norm": 0.6611968874931335,
+      "learning_rate": 6.906854130052724e-05,
+      "loss": 1.0552500486373901,
+      "step": 394
+    },
+    {
+      "epoch": 0.1670886075949367,
+      "grad_norm": 0.6955893039703369,
+      "learning_rate": 6.942003514938489e-05,
+      "loss": 1.0562875270843506,
+      "step": 396
+    },
+    {
+      "epoch": 0.1679324894514768,
+      "grad_norm": 0.6666058301925659,
+      "learning_rate": 6.977152899824253e-05,
+      "loss": 0.9850592017173767,
+      "step": 398
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "grad_norm": 0.6131711006164551,
+      "learning_rate": 7.012302284710018e-05,
+      "loss": 1.0077755451202393,
+      "step": 400
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "eval_loss": 1.0625108480453491,
+      "eval_runtime": 691.0068,
+      "eval_samples_per_second": 3.049,
+      "eval_steps_per_second": 3.049,
+      "step": 400
+    },
+    {
+      "epoch": 0.16962025316455695,
+      "grad_norm": 0.6286499500274658,
+      "learning_rate": 7.047451669595783e-05,
+      "loss": 1.1012427806854248,
+      "step": 402
+    },
+    {
+      "epoch": 0.17046413502109706,
+      "grad_norm": 0.6639351844787598,
+      "learning_rate": 7.082601054481547e-05,
+      "loss": 1.0379719734191895,
+      "step": 404
+    },
+    {
+      "epoch": 0.17130801687763714,
+      "grad_norm": 0.750401496887207,
+      "learning_rate": 7.117750439367311e-05,
+      "loss": 1.031856656074524,
+      "step": 406
+    },
+    {
+      "epoch": 0.17215189873417722,
+      "grad_norm": 0.8084847331047058,
+      "learning_rate": 7.152899824253075e-05,
+      "loss": 1.0493193864822388,
+      "step": 408
+    },
+    {
+      "epoch": 0.1729957805907173,
+      "grad_norm": 0.7448462247848511,
+      "learning_rate": 7.188049209138841e-05,
+      "loss": 1.1012418270111084,
+      "step": 410
+    },
+    {
+      "epoch": 0.17383966244725738,
+      "grad_norm": 0.5841867923736572,
+      "learning_rate": 7.223198594024605e-05,
+      "loss": 0.9926692247390747,
+      "step": 412
+    },
+    {
+      "epoch": 0.17468354430379746,
+      "grad_norm": 0.7125606536865234,
+      "learning_rate": 7.258347978910369e-05,
+      "loss": 1.0588877201080322,
+      "step": 414
+    },
+    {
+      "epoch": 0.17552742616033756,
+      "grad_norm": 0.5750942230224609,
+      "learning_rate": 7.293497363796134e-05,
+      "loss": 1.038270354270935,
+      "step": 416
+    },
+    {
+      "epoch": 0.17637130801687764,
+      "grad_norm": 0.565444827079773,
+      "learning_rate": 7.328646748681899e-05,
+      "loss": 0.9843021035194397,
+      "step": 418
+    },
+    {
+      "epoch": 0.17721518987341772,
+      "grad_norm": 0.5825693011283875,
+      "learning_rate": 7.363796133567663e-05,
+      "loss": 1.0731632709503174,
+      "step": 420
+    },
+    {
+      "epoch": 0.1780590717299578,
+      "grad_norm": 0.6267391443252563,
+      "learning_rate": 7.398945518453427e-05,
+      "loss": 1.0061273574829102,
+      "step": 422
+    },
+    {
+      "epoch": 0.17890295358649788,
+      "grad_norm": 0.6621372103691101,
+      "learning_rate": 7.434094903339192e-05,
+      "loss": 1.0461612939834595,
+      "step": 424
+    },
+    {
+      "epoch": 0.17974683544303796,
+      "grad_norm": 0.6635435223579407,
+      "learning_rate": 7.469244288224957e-05,
+      "loss": 0.9789207577705383,
+      "step": 426
+    },
+    {
+      "epoch": 0.18059071729957807,
+      "grad_norm": 0.6342346668243408,
+      "learning_rate": 7.504393673110721e-05,
+      "loss": 1.0527069568634033,
+      "step": 428
+    },
+    {
+      "epoch": 0.18143459915611815,
+      "grad_norm": 0.6762149930000305,
+      "learning_rate": 7.539543057996485e-05,
+      "loss": 0.9708702564239502,
+      "step": 430
+    },
+    {
+      "epoch": 0.18227848101265823,
+      "grad_norm": 0.7073282599449158,
+      "learning_rate": 7.57469244288225e-05,
+      "loss": 1.0509834289550781,
+      "step": 432
+    },
+    {
+      "epoch": 0.1831223628691983,
+      "grad_norm": 0.6917856931686401,
+      "learning_rate": 7.609841827768014e-05,
+      "loss": 1.0128819942474365,
+      "step": 434
+    },
+    {
+      "epoch": 0.1839662447257384,
+      "grad_norm": 0.5574942231178284,
+      "learning_rate": 7.644991212653779e-05,
+      "loss": 0.989395797252655,
+      "step": 436
+    },
+    {
+      "epoch": 0.1848101265822785,
+      "grad_norm": 0.640765905380249,
+      "learning_rate": 7.680140597539543e-05,
+      "loss": 0.9846042990684509,
+      "step": 438
+    },
+    {
+      "epoch": 0.18565400843881857,
+      "grad_norm": 0.6699127554893494,
+      "learning_rate": 7.715289982425308e-05,
+      "loss": 1.0344442129135132,
+      "step": 440
+    },
+    {
+      "epoch": 0.18649789029535865,
+      "grad_norm": 0.6164930462837219,
+      "learning_rate": 7.750439367311072e-05,
+      "loss": 1.0179373025894165,
+      "step": 442
+    },
+    {
+      "epoch": 0.18734177215189873,
+      "grad_norm": 0.6880720853805542,
+      "learning_rate": 7.785588752196837e-05,
+      "loss": 1.0518895387649536,
+      "step": 444
+    },
+    {
+      "epoch": 0.1881856540084388,
+      "grad_norm": 0.6501413583755493,
+      "learning_rate": 7.820738137082601e-05,
+      "loss": 1.0442606210708618,
+      "step": 446
+    },
+    {
+      "epoch": 0.1890295358649789,
+      "grad_norm": 0.6076085567474365,
+      "learning_rate": 7.855887521968366e-05,
+      "loss": 0.9828442335128784,
+      "step": 448
+    },
+    {
+      "epoch": 0.189873417721519,
+      "grad_norm": 0.6418202519416809,
+      "learning_rate": 7.89103690685413e-05,
+      "loss": 1.0573710203170776,
+      "step": 450
+    },
+    {
+      "epoch": 0.19071729957805908,
+      "grad_norm": 0.7055076360702515,
+      "learning_rate": 7.926186291739895e-05,
+      "loss": 1.0216103792190552,
+      "step": 452
+    },
+    {
+      "epoch": 0.19156118143459916,
+      "grad_norm": 0.5668330192565918,
+      "learning_rate": 7.961335676625659e-05,
+      "loss": 0.9837722778320312,
+      "step": 454
+    },
+    {
+      "epoch": 0.19240506329113924,
+      "grad_norm": 0.6419380307197571,
+      "learning_rate": 7.996485061511424e-05,
+      "loss": 1.0003894567489624,
+      "step": 456
+    },
+    {
+      "epoch": 0.19324894514767932,
+      "grad_norm": 0.5949198007583618,
+      "learning_rate": 8.031634446397188e-05,
+      "loss": 1.0609031915664673,
+      "step": 458
+    },
+    {
+      "epoch": 0.1940928270042194,
+      "grad_norm": 0.7032039761543274,
+      "learning_rate": 8.066783831282952e-05,
+      "loss": 1.0543403625488281,
+      "step": 460
+    },
+    {
+      "epoch": 0.1949367088607595,
+      "grad_norm": 0.5775868892669678,
+      "learning_rate": 8.101933216168718e-05,
+      "loss": 0.9819303154945374,
+      "step": 462
+    },
+    {
+      "epoch": 0.19578059071729959,
+      "grad_norm": 0.9301062226295471,
+      "learning_rate": 8.137082601054482e-05,
+      "loss": 1.0542067289352417,
+      "step": 464
+    },
+    {
+      "epoch": 0.19662447257383966,
+      "grad_norm": 0.6193217039108276,
+      "learning_rate": 8.172231985940246e-05,
+      "loss": 0.9966341257095337,
+      "step": 466
+    },
+    {
+      "epoch": 0.19746835443037974,
+      "grad_norm": 0.6286146640777588,
+      "learning_rate": 8.20738137082601e-05,
+      "loss": 1.0474121570587158,
+      "step": 468
+    },
+    {
+      "epoch": 0.19831223628691982,
+      "grad_norm": 0.7418972253799438,
+      "learning_rate": 8.242530755711776e-05,
+      "loss": 0.9549239277839661,
+      "step": 470
+    },
+    {
+      "epoch": 0.1991561181434599,
+      "grad_norm": 0.6122808456420898,
+      "learning_rate": 8.27768014059754e-05,
+      "loss": 1.0191338062286377,
+      "step": 472
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.6375362277030945,
+      "learning_rate": 8.312829525483304e-05,
+      "loss": 1.0987539291381836,
+      "step": 474
+    },
+    {
+      "epoch": 0.2008438818565401,
+      "grad_norm": 0.6459513306617737,
+      "learning_rate": 8.347978910369068e-05,
+      "loss": 1.0369136333465576,
+      "step": 476
+    },
+    {
+      "epoch": 0.20168776371308017,
+      "grad_norm": 0.7029640674591064,
+      "learning_rate": 8.383128295254833e-05,
+      "loss": 1.0582096576690674,
+      "step": 478
+    },
+    {
+      "epoch": 0.20253164556962025,
+      "grad_norm": 0.6345387697219849,
+      "learning_rate": 8.418277680140598e-05,
+      "loss": 1.022916316986084,
+      "step": 480
+    },
+    {
+      "epoch": 0.20337552742616033,
+      "grad_norm": 0.5764590501785278,
+      "learning_rate": 8.453427065026362e-05,
+      "loss": 0.973024308681488,
+      "step": 482
+    },
+    {
+      "epoch": 0.2042194092827004,
+      "grad_norm": 0.5884482860565186,
+      "learning_rate": 8.488576449912127e-05,
+      "loss": 1.0292812585830688,
+      "step": 484
+    },
+    {
+      "epoch": 0.20506329113924052,
+      "grad_norm": 0.616357147693634,
+      "learning_rate": 8.523725834797891e-05,
+      "loss": 1.0083447694778442,
+      "step": 486
+    },
+    {
+      "epoch": 0.2059071729957806,
+      "grad_norm": 0.7671196460723877,
+      "learning_rate": 8.558875219683656e-05,
+      "loss": 0.9936985373497009,
+      "step": 488
+    },
+    {
+      "epoch": 0.20675105485232068,
+      "grad_norm": 0.6197299957275391,
+      "learning_rate": 8.59402460456942e-05,
+      "loss": 1.051513910293579,
+      "step": 490
+    },
+    {
+      "epoch": 0.20759493670886076,
+      "grad_norm": 0.6912890672683716,
+      "learning_rate": 8.629173989455185e-05,
+      "loss": 0.9474978446960449,
+      "step": 492
+    },
+    {
+      "epoch": 0.20843881856540084,
+      "grad_norm": 0.6941592693328857,
+      "learning_rate": 8.664323374340949e-05,
+      "loss": 1.0671660900115967,
+      "step": 494
+    },
+    {
+      "epoch": 0.20928270042194091,
+      "grad_norm": 0.5889528393745422,
+      "learning_rate": 8.699472759226714e-05,
+      "loss": 1.0020159482955933,
+      "step": 496
+    },
+    {
+      "epoch": 0.21012658227848102,
+      "grad_norm": 0.6478549838066101,
+      "learning_rate": 8.734622144112478e-05,
+      "loss": 1.0165860652923584,
+      "step": 498
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "grad_norm": 0.6018255949020386,
+      "learning_rate": 8.769771528998243e-05,
+      "loss": 0.8798263072967529,
+      "step": 500
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "eval_loss": 1.042096495628357,
+      "eval_runtime": 692.4361,
+      "eval_samples_per_second": 3.043,
+      "eval_steps_per_second": 3.043,
+      "step": 500
+    },
+    {
+      "epoch": 0.21181434599156118,
+      "grad_norm": 0.578990638256073,
+      "learning_rate": 8.804920913884007e-05,
+      "loss": 1.092096209526062,
+      "step": 502
+    },
+    {
+      "epoch": 0.21265822784810126,
+      "grad_norm": 0.6597883701324463,
+      "learning_rate": 8.840070298769771e-05,
+      "loss": 1.0413451194763184,
+      "step": 504
+    },
+    {
+      "epoch": 0.21350210970464134,
+      "grad_norm": 0.6660305261611938,
+      "learning_rate": 8.875219683655536e-05,
+      "loss": 1.0073142051696777,
+      "step": 506
+    },
+    {
+      "epoch": 0.21434599156118145,
+      "grad_norm": 0.6283115148544312,
+      "learning_rate": 8.910369068541301e-05,
+      "loss": 1.0319768190383911,
+      "step": 508
+    },
+    {
+      "epoch": 0.21518987341772153,
+      "grad_norm": 0.6257343292236328,
+      "learning_rate": 8.945518453427065e-05,
+      "loss": 1.0046353340148926,
+      "step": 510
+    },
+    {
+      "epoch": 0.2160337552742616,
+      "grad_norm": 0.5530875325202942,
+      "learning_rate": 8.980667838312829e-05,
+      "loss": 0.9169099926948547,
+      "step": 512
+    },
+    {
+      "epoch": 0.2168776371308017,
+      "grad_norm": 0.5369633436203003,
+      "learning_rate": 9.015817223198594e-05,
+      "loss": 0.9081505537033081,
+      "step": 514
+    },
+    {
+      "epoch": 0.21772151898734177,
+      "grad_norm": 0.6618232131004333,
+      "learning_rate": 9.05096660808436e-05,
+      "loss": 0.9165045022964478,
+      "step": 516
+    },
+    {
+      "epoch": 0.21856540084388185,
+      "grad_norm": 0.600666344165802,
+      "learning_rate": 9.086115992970123e-05,
+      "loss": 0.91348797082901,
+      "step": 518
+    },
+    {
+      "epoch": 0.21940928270042195,
+      "grad_norm": 0.5919831991195679,
+      "learning_rate": 9.121265377855887e-05,
+      "loss": 1.006508469581604,
+      "step": 520
+    },
+    {
+      "epoch": 0.22025316455696203,
+      "grad_norm": 0.688058614730835,
+      "learning_rate": 9.156414762741654e-05,
+      "loss": 1.0013236999511719,
+      "step": 522
+    },
+    {
+      "epoch": 0.2210970464135021,
+      "grad_norm": 0.6721227765083313,
+      "learning_rate": 9.191564147627418e-05,
+      "loss": 1.0909923315048218,
+      "step": 524
+    },
+    {
+      "epoch": 0.2219409282700422,
+      "grad_norm": 0.5987313389778137,
+      "learning_rate": 9.226713532513181e-05,
+      "loss": 1.0117096900939941,
+      "step": 526
+    },
+    {
+      "epoch": 0.22278481012658227,
+      "grad_norm": 0.6191489696502686,
+      "learning_rate": 9.261862917398945e-05,
+      "loss": 1.0153647661209106,
+      "step": 528
+    },
+    {
+      "epoch": 0.22362869198312235,
+      "grad_norm": 0.6821563243865967,
+      "learning_rate": 9.29701230228471e-05,
+      "loss": 0.9649755954742432,
+      "step": 530
+    },
+    {
+      "epoch": 0.22447257383966246,
+      "grad_norm": 1.760398268699646,
+      "learning_rate": 9.332161687170476e-05,
+      "loss": 0.8673232197761536,
+      "step": 532
+    },
+    {
+      "epoch": 0.22531645569620254,
+      "grad_norm": 0.6670058369636536,
+      "learning_rate": 9.36731107205624e-05,
+      "loss": 0.9942440986633301,
+      "step": 534
+    },
+    {
+      "epoch": 0.22616033755274262,
+      "grad_norm": 0.7345916032791138,
+      "learning_rate": 9.402460456942003e-05,
+      "loss": 1.0364389419555664,
+      "step": 536
+    },
+    {
+      "epoch": 0.2270042194092827,
+      "grad_norm": 0.5946128964424133,
+      "learning_rate": 9.437609841827768e-05,
+      "loss": 0.9314924478530884,
+      "step": 538
+    },
+    {
+      "epoch": 0.22784810126582278,
+      "grad_norm": 0.5800848603248596,
+      "learning_rate": 9.472759226713534e-05,
+      "loss": 1.0694862604141235,
+      "step": 540
+    },
+    {
+      "epoch": 0.22869198312236286,
+      "grad_norm": 0.6712192893028259,
+      "learning_rate": 9.507908611599297e-05,
+      "loss": 1.03531014919281,
+      "step": 542
+    },
+    {
+      "epoch": 0.22953586497890296,
+      "grad_norm": 0.5641416311264038,
+      "learning_rate": 9.543057996485063e-05,
+      "loss": 0.9795235991477966,
+      "step": 544
+    },
+    {
+      "epoch": 0.23037974683544304,
+      "grad_norm": 0.50412517786026,
+      "learning_rate": 9.578207381370826e-05,
+      "loss": 0.9641494750976562,
+      "step": 546
+    },
+    {
+      "epoch": 0.23122362869198312,
+      "grad_norm": 0.579118549823761,
+      "learning_rate": 9.61335676625659e-05,
+      "loss": 0.9375281929969788,
+      "step": 548
+    },
+    {
+      "epoch": 0.2320675105485232,
+      "grad_norm": 0.5888341665267944,
+      "learning_rate": 9.648506151142355e-05,
+      "loss": 0.9414046406745911,
+      "step": 550
+    },
+    {
+      "epoch": 0.23291139240506328,
+      "grad_norm": 0.5595056414604187,
+      "learning_rate": 9.68365553602812e-05,
+      "loss": 0.9005617499351501,
+      "step": 552
+    },
+    {
+      "epoch": 0.23375527426160336,
+      "grad_norm": 0.6605326533317566,
+      "learning_rate": 9.718804920913884e-05,
+      "loss": 1.0283968448638916,
+      "step": 554
+    },
+    {
+      "epoch": 0.23459915611814347,
+      "grad_norm": 0.5657313466072083,
+      "learning_rate": 9.753954305799648e-05,
+      "loss": 1.0058249235153198,
+      "step": 556
+    },
+    {
+      "epoch": 0.23544303797468355,
+      "grad_norm": 0.5433364510536194,
+      "learning_rate": 9.789103690685413e-05,
+      "loss": 0.9835494756698608,
+      "step": 558
+    },
+    {
+      "epoch": 0.23628691983122363,
+      "grad_norm": 0.6129802465438843,
+      "learning_rate": 9.824253075571179e-05,
+      "loss": 1.054532527923584,
+      "step": 560
+    },
+    {
+      "epoch": 0.2371308016877637,
+      "grad_norm": 0.6496239304542542,
+      "learning_rate": 9.859402460456942e-05,
+      "loss": 1.0240973234176636,
+      "step": 562
+    },
+    {
+      "epoch": 0.2379746835443038,
+      "grad_norm": 0.6380873918533325,
+      "learning_rate": 9.894551845342706e-05,
+      "loss": 1.0229179859161377,
+      "step": 564
+    },
+    {
+      "epoch": 0.23881856540084387,
+      "grad_norm": 0.6151993870735168,
+      "learning_rate": 9.929701230228471e-05,
+      "loss": 1.0111570358276367,
+      "step": 566
+    },
+    {
+      "epoch": 0.23966244725738398,
+      "grad_norm": 0.5727584958076477,
+      "learning_rate": 9.964850615114237e-05,
+      "loss": 0.9450829029083252,
+      "step": 568
+    },
+    {
+      "epoch": 0.24050632911392406,
+      "grad_norm": 0.6620725989341736,
+      "learning_rate": 0.0001,
+      "loss": 0.9800319075584412,
+      "step": 570
+    },
+    {
+      "epoch": 0.24135021097046414,
+      "grad_norm": 0.6151163578033447,
+      "learning_rate": 0.00010035149384885764,
+      "loss": 0.9757438898086548,
+      "step": 572
+    },
+    {
+      "epoch": 0.24219409282700421,
+      "grad_norm": 0.5672140717506409,
+      "learning_rate": 0.0001007029876977153,
+      "loss": 0.9104921817779541,
+      "step": 574
+    },
+    {
+      "epoch": 0.2430379746835443,
+      "grad_norm": 0.5697256326675415,
+      "learning_rate": 0.00010105448154657293,
+      "loss": 1.1027376651763916,
+      "step": 576
+    },
+    {
+      "epoch": 0.2438818565400844,
+      "grad_norm": 0.5590381622314453,
+      "learning_rate": 0.00010140597539543057,
+      "loss": 1.0055404901504517,
+      "step": 578
+    },
+    {
+      "epoch": 0.24472573839662448,
+      "grad_norm": 0.5518567562103271,
+      "learning_rate": 0.00010175746924428824,
+      "loss": 1.020835518836975,
+      "step": 580
+    },
+    {
+      "epoch": 0.24556962025316456,
+      "grad_norm": 0.6338496208190918,
+      "learning_rate": 0.00010210896309314588,
+      "loss": 0.9528344869613647,
+      "step": 582
+    },
+    {
+      "epoch": 0.24641350210970464,
+      "grad_norm": 0.6497329473495483,
+      "learning_rate": 0.00010246045694200353,
+      "loss": 1.0088670253753662,
+      "step": 584
+    },
+    {
+      "epoch": 0.24725738396624472,
+      "grad_norm": 0.49888095259666443,
+      "learning_rate": 0.00010281195079086117,
+      "loss": 0.9961200952529907,
+      "step": 586
+    },
+    {
+      "epoch": 0.2481012658227848,
+      "grad_norm": 0.5680158734321594,
+      "learning_rate": 0.0001031634446397188,
+      "loss": 0.9635610580444336,
+      "step": 588
+    },
+    {
+      "epoch": 0.2489451476793249,
+      "grad_norm": 0.658168375492096,
+      "learning_rate": 0.00010351493848857646,
+      "loss": 0.9392287135124207,
+      "step": 590
+    },
+    {
+      "epoch": 0.249789029535865,
+      "grad_norm": 0.618262767791748,
+      "learning_rate": 0.0001038664323374341,
+      "loss": 0.9600516557693481,
+      "step": 592
+    },
+    {
+      "epoch": 0.25063291139240507,
+      "grad_norm": 0.6003909111022949,
+      "learning_rate": 0.00010421792618629173,
+      "loss": 1.005476713180542,
+      "step": 594
+    },
+    {
+      "epoch": 0.2514767932489452,
+      "grad_norm": 0.5437078475952148,
+      "learning_rate": 0.00010456942003514938,
+      "loss": 0.9523017406463623,
+      "step": 596
+    },
+    {
+      "epoch": 0.2523206751054852,
+      "grad_norm": 0.5524541735649109,
+      "learning_rate": 0.00010492091388400705,
+      "loss": 0.9526668787002563,
+      "step": 598
+    },
+    {
+      "epoch": 0.25316455696202533,
+      "grad_norm": 0.679504930973053,
+      "learning_rate": 0.00010527240773286469,
+      "loss": 1.019660472869873,
+      "step": 600
+    },
+    {
+      "epoch": 0.25316455696202533,
+      "eval_loss": 1.0193854570388794,
+      "eval_runtime": 677.9523,
+      "eval_samples_per_second": 3.108,
+      "eval_steps_per_second": 3.108,
+      "step": 600
+    },
+    {
+      "epoch": 0.2540084388185654,
+      "grad_norm": 0.5646136999130249,
+      "learning_rate": 0.00010562390158172233,
+      "loss": 0.9910882711410522,
+      "step": 602
+    },
+    {
+      "epoch": 0.2548523206751055,
+      "grad_norm": 0.5238093137741089,
+      "learning_rate": 0.00010597539543057998,
+      "loss": 0.9616432785987854,
+      "step": 604
+    },
+    {
+      "epoch": 0.25569620253164554,
+      "grad_norm": 0.7483857274055481,
+      "learning_rate": 0.00010632688927943762,
+      "loss": 1.0078275203704834,
+      "step": 606
+    },
+    {
+      "epoch": 0.25654008438818565,
+      "grad_norm": 0.578948974609375,
+      "learning_rate": 0.00010667838312829525,
+      "loss": 0.9827103018760681,
+      "step": 608
+    },
+    {
+      "epoch": 0.25738396624472576,
+      "grad_norm": 0.5525906085968018,
+      "learning_rate": 0.00010702987697715289,
+      "loss": 1.0423277616500854,
+      "step": 610
+    },
+    {
+      "epoch": 0.2582278481012658,
+      "grad_norm": 0.6721326112747192,
+      "learning_rate": 0.00010738137082601054,
+      "loss": 0.9561693072319031,
+      "step": 612
+    },
+    {
+      "epoch": 0.2590717299578059,
+      "grad_norm": 0.5701051354408264,
+      "learning_rate": 0.00010773286467486821,
+      "loss": 0.9602992534637451,
+      "step": 614
+    },
+    {
+      "epoch": 0.25991561181434597,
+      "grad_norm": 0.6349860429763794,
+      "learning_rate": 0.00010808435852372585,
+      "loss": 1.1422650814056396,
+      "step": 616
+    },
+    {
+      "epoch": 0.2607594936708861,
+      "grad_norm": 0.5496085286140442,
+      "learning_rate": 0.00010843585237258349,
+      "loss": 0.9762773513793945,
+      "step": 618
+    },
+    {
+      "epoch": 0.2616033755274262,
+      "grad_norm": 0.6080722808837891,
+      "learning_rate": 0.00010878734622144114,
+      "loss": 1.0133616924285889,
+      "step": 620
+    },
+    {
+      "epoch": 0.26244725738396624,
+      "grad_norm": 0.5450218915939331,
+      "learning_rate": 0.00010913884007029878,
+      "loss": 0.9385587573051453,
+      "step": 622
+    },
+    {
+      "epoch": 0.26329113924050634,
+      "grad_norm": 0.592106819152832,
+      "learning_rate": 0.00010949033391915641,
+      "loss": 0.9359989762306213,
+      "step": 624
+    },
+    {
+      "epoch": 0.2641350210970464,
+      "grad_norm": 0.6449427604675293,
+      "learning_rate": 0.00010984182776801407,
+      "loss": 1.0266027450561523,
+      "step": 626
+    },
+    {
+      "epoch": 0.2649789029535865,
+      "grad_norm": 0.538299560546875,
+      "learning_rate": 0.0001101933216168717,
+      "loss": 0.9303187131881714,
+      "step": 628
+    },
+    {
+      "epoch": 0.26582278481012656,
+      "grad_norm": 0.546316921710968,
+      "learning_rate": 0.00011054481546572934,
+      "loss": 0.9368857145309448,
+      "step": 630
+    },
+    {
+      "epoch": 0.26666666666666666,
+      "grad_norm": 0.5818730592727661,
+      "learning_rate": 0.00011089630931458701,
+      "loss": 0.9573145508766174,
+      "step": 632
+    },
+    {
+      "epoch": 0.26751054852320677,
+      "grad_norm": 0.5958262085914612,
+      "learning_rate": 0.00011124780316344465,
+      "loss": 0.9345449805259705,
+      "step": 634
+    },
+    {
+      "epoch": 0.2683544303797468,
+      "grad_norm": 0.6259077787399292,
+      "learning_rate": 0.0001115992970123023,
+      "loss": 1.0906590223312378,
+      "step": 636
+    },
+    {
+      "epoch": 0.26919831223628693,
+      "grad_norm": 0.589672863483429,
+      "learning_rate": 0.00011195079086115994,
+      "loss": 1.0757447481155396,
+      "step": 638
+    },
+    {
+      "epoch": 0.270042194092827,
+      "grad_norm": 0.5714080333709717,
+      "learning_rate": 0.00011230228471001758,
+      "loss": 0.9310855269432068,
+      "step": 640
+    },
+    {
+      "epoch": 0.2708860759493671,
+      "grad_norm": 0.45342639088630676,
+      "learning_rate": 0.00011265377855887523,
+      "loss": 0.9276360273361206,
+      "step": 642
+    },
+    {
+      "epoch": 0.2717299578059072,
+      "grad_norm": 0.6386750340461731,
+      "learning_rate": 0.00011300527240773287,
+      "loss": 1.084719181060791,
+      "step": 644
+    },
+    {
+      "epoch": 0.27257383966244725,
+      "grad_norm": 0.6446163654327393,
+      "learning_rate": 0.0001133567662565905,
+      "loss": 0.9763918519020081,
+      "step": 646
+    },
+    {
+      "epoch": 0.27341772151898736,
+      "grad_norm": 0.5925686359405518,
+      "learning_rate": 0.00011370826010544816,
+      "loss": 0.9517921805381775,
+      "step": 648
+    },
+    {
+      "epoch": 0.2742616033755274,
+      "grad_norm": 0.5399773716926575,
+      "learning_rate": 0.00011405975395430582,
+      "loss": 1.0587927103042603,
+      "step": 650
+    },
+    {
+      "epoch": 0.2751054852320675,
+      "grad_norm": 0.5872456431388855,
+      "learning_rate": 0.00011441124780316346,
+      "loss": 0.883341908454895,
+      "step": 652
+    },
+    {
+      "epoch": 0.2759493670886076,
+      "grad_norm": 0.5574564337730408,
+      "learning_rate": 0.0001147627416520211,
+      "loss": 1.0306891202926636,
+      "step": 654
+    },
+    {
+      "epoch": 0.2767932489451477,
+      "grad_norm": 0.47789013385772705,
+      "learning_rate": 0.00011511423550087874,
+      "loss": 0.8814032077789307,
+      "step": 656
+    },
+    {
+      "epoch": 0.2776371308016878,
+      "grad_norm": 0.5565530061721802,
+      "learning_rate": 0.00011546572934973639,
+      "loss": 0.9460552334785461,
+      "step": 658
+    },
+    {
+      "epoch": 0.27848101265822783,
+      "grad_norm": 0.5299761295318604,
+      "learning_rate": 0.00011581722319859403,
+      "loss": 0.9475110769271851,
+      "step": 660
+    },
+    {
+      "epoch": 0.27932489451476794,
+      "grad_norm": 0.6503344178199768,
+      "learning_rate": 0.00011616871704745166,
+      "loss": 1.0630913972854614,
+      "step": 662
+    },
+    {
+      "epoch": 0.280168776371308,
+      "grad_norm": 0.5794585943222046,
+      "learning_rate": 0.00011652021089630932,
+      "loss": 0.9389138221740723,
+      "step": 664
+    },
+    {
+      "epoch": 0.2810126582278481,
+      "grad_norm": 0.5762867331504822,
+      "learning_rate": 0.00011687170474516695,
+      "loss": 0.8934136033058167,
+      "step": 666
+    },
+    {
+      "epoch": 0.2818565400843882,
+      "grad_norm": 0.6565435528755188,
+      "learning_rate": 0.00011722319859402462,
+      "loss": 1.1072614192962646,
+      "step": 668
+    },
+    {
+      "epoch": 0.28270042194092826,
+      "grad_norm": 0.5819830298423767,
+      "learning_rate": 0.00011757469244288226,
+      "loss": 1.0501434803009033,
+      "step": 670
+    },
+    {
+      "epoch": 0.28354430379746837,
+      "grad_norm": 0.6071487069129944,
+      "learning_rate": 0.00011792618629173991,
+      "loss": 0.9880793690681458,
+      "step": 672
+    },
+    {
+      "epoch": 0.2843881856540084,
+      "grad_norm": 0.5765058398246765,
+      "learning_rate": 0.00011827768014059755,
+      "loss": 0.9670693874359131,
+      "step": 674
+    },
+    {
+      "epoch": 0.2852320675105485,
+      "grad_norm": 0.5245351791381836,
+      "learning_rate": 0.00011862917398945519,
+      "loss": 0.9602360725402832,
+      "step": 676
+    },
+    {
+      "epoch": 0.28607594936708863,
+      "grad_norm": 0.6189922094345093,
+      "learning_rate": 0.00011898066783831282,
+      "loss": 0.9684560894966125,
+      "step": 678
+    },
+    {
+      "epoch": 0.2869198312236287,
+      "grad_norm": 0.6138690710067749,
+      "learning_rate": 0.00011933216168717048,
+      "loss": 0.9465792775154114,
+      "step": 680
+    },
+    {
+      "epoch": 0.2877637130801688,
+      "grad_norm": 0.5371595621109009,
+      "learning_rate": 0.00011968365553602812,
+      "loss": 0.8495944738388062,
+      "step": 682
+    },
+    {
+      "epoch": 0.28860759493670884,
+      "grad_norm": 0.5549944639205933,
+      "learning_rate": 0.00012003514938488578,
+      "loss": 0.9663267135620117,
+      "step": 684
+    },
+    {
+      "epoch": 0.28945147679324895,
+      "grad_norm": 0.6484189033508301,
+      "learning_rate": 0.00012038664323374342,
+      "loss": 0.9736058712005615,
+      "step": 686
+    },
+    {
+      "epoch": 0.290295358649789,
+      "grad_norm": 0.540351390838623,
+      "learning_rate": 0.00012073813708260107,
+      "loss": 1.0591845512390137,
+      "step": 688
+    },
+    {
+      "epoch": 0.2911392405063291,
+      "grad_norm": 0.5657922029495239,
+      "learning_rate": 0.00012108963093145871,
+      "loss": 0.944908618927002,
+      "step": 690
+    },
+    {
+      "epoch": 0.2919831223628692,
+      "grad_norm": 0.6040505170822144,
+      "learning_rate": 0.00012144112478031635,
+      "loss": 1.0018219947814941,
+      "step": 692
+    },
+    {
+      "epoch": 0.29282700421940927,
+      "grad_norm": 0.5435477495193481,
+      "learning_rate": 0.000121792618629174,
+      "loss": 1.0351502895355225,
+      "step": 694
+    },
+    {
+      "epoch": 0.2936708860759494,
+      "grad_norm": 0.5712518692016602,
+      "learning_rate": 0.00012214411247803164,
+      "loss": 0.9935672283172607,
+      "step": 696
+    },
+    {
+      "epoch": 0.29451476793248943,
+      "grad_norm": 0.6138222813606262,
+      "learning_rate": 0.00012249560632688928,
+      "loss": 1.0165108442306519,
+      "step": 698
+    },
+    {
+      "epoch": 0.29535864978902954,
+      "grad_norm": 0.4495212435722351,
+      "learning_rate": 0.00012284710017574691,
+      "loss": 0.9334425926208496,
+      "step": 700
+    },
+    {
+      "epoch": 0.29535864978902954,
+      "eval_loss": 0.996929407119751,
+      "eval_runtime": 668.6398,
+      "eval_samples_per_second": 3.151,
+      "eval_steps_per_second": 3.151,
+      "step": 700
+    },
+    {
+      "epoch": 0.29620253164556964,
+      "grad_norm": 0.5321539044380188,
+      "learning_rate": 0.00012319859402460458,
+      "loss": 1.0516537427902222,
+      "step": 702
+    },
+    {
+      "epoch": 0.2970464135021097,
+      "grad_norm": 0.5716516971588135,
+      "learning_rate": 0.00012355008787346222,
+      "loss": 0.9387198686599731,
+      "step": 704
+    },
+    {
+      "epoch": 0.2978902953586498,
+      "grad_norm": 0.5617920160293579,
+      "learning_rate": 0.00012390158172231988,
+      "loss": 0.99737948179245,
+      "step": 706
+    },
+    {
+      "epoch": 0.29873417721518986,
+      "grad_norm": 0.4922899007797241,
+      "learning_rate": 0.00012425307557117752,
+      "loss": 0.9955025911331177,
+      "step": 708
+    },
+    {
+      "epoch": 0.29957805907172996,
+      "grad_norm": 0.543501615524292,
+      "learning_rate": 0.00012460456942003516,
+      "loss": 0.9124280214309692,
+      "step": 710
+    },
+    {
+      "epoch": 0.30042194092827,
+      "grad_norm": 0.49590054154396057,
+      "learning_rate": 0.0001249560632688928,
+      "loss": 0.9820216298103333,
+      "step": 712
+    },
+    {
+      "epoch": 0.3012658227848101,
+      "grad_norm": 0.5984305739402771,
+      "learning_rate": 0.00012530755711775044,
+      "loss": 1.0152074098587036,
+      "step": 714
+    },
+    {
+      "epoch": 0.30210970464135023,
+      "grad_norm": 0.9343504905700684,
+      "learning_rate": 0.00012565905096660807,
+      "loss": 1.0577725172042847,
+      "step": 716
+    },
+    {
+      "epoch": 0.3029535864978903,
+      "grad_norm": 0.5118702054023743,
+      "learning_rate": 0.0001260105448154657,
+      "loss": 0.9830358028411865,
+      "step": 718
+    },
+    {
+      "epoch": 0.3037974683544304,
+      "grad_norm": 0.4940392076969147,
+      "learning_rate": 0.00012636203866432338,
+      "loss": 0.9466043710708618,
+      "step": 720
+    },
+    {
+      "epoch": 0.30464135021097044,
+      "grad_norm": 0.5965693593025208,
+      "learning_rate": 0.00012671353251318104,
+      "loss": 1.015270709991455,
+      "step": 722
+    },
+    {
+      "epoch": 0.30548523206751055,
+      "grad_norm": 0.5020529627799988,
+      "learning_rate": 0.00012706502636203868,
+      "loss": 0.9703927636146545,
+      "step": 724
+    },
+    {
+      "epoch": 0.30632911392405066,
+      "grad_norm": 0.6067010164260864,
+      "learning_rate": 0.00012741652021089632,
+      "loss": 1.0255526304244995,
+      "step": 726
+    },
+    {
+      "epoch": 0.3071729957805907,
+      "grad_norm": 0.5931884050369263,
+      "learning_rate": 0.00012776801405975396,
+      "loss": 0.9335633516311646,
+      "step": 728
+    },
+    {
+      "epoch": 0.3080168776371308,
+      "grad_norm": 0.5938752293586731,
+      "learning_rate": 0.0001281195079086116,
+      "loss": 1.0921578407287598,
+      "step": 730
+    },
+    {
+      "epoch": 0.30886075949367087,
+      "grad_norm": 0.49728086590766907,
+      "learning_rate": 0.00012847100175746923,
+      "loss": 0.963066041469574,
+      "step": 732
+    },
+    {
+      "epoch": 0.309704641350211,
+      "grad_norm": 0.5452080965042114,
+      "learning_rate": 0.0001288224956063269,
+      "loss": 0.9513075351715088,
+      "step": 734
+    },
+    {
+      "epoch": 0.3105485232067511,
+      "grad_norm": 0.5497731566429138,
+      "learning_rate": 0.00012917398945518454,
+      "loss": 0.8576077222824097,
+      "step": 736
+    },
+    {
+      "epoch": 0.31139240506329113,
+      "grad_norm": 0.5580397248268127,
+      "learning_rate": 0.0001295254833040422,
+      "loss": 0.9542577862739563,
+      "step": 738
+    },
+    {
+      "epoch": 0.31223628691983124,
+      "grad_norm": 0.5890427827835083,
+      "learning_rate": 0.00012987697715289984,
+      "loss": 0.8992732167243958,
+      "step": 740
+    },
+    {
+      "epoch": 0.3130801687763713,
+      "grad_norm": 0.5942965745925903,
+      "learning_rate": 0.00013022847100175748,
+      "loss": 1.0322896242141724,
+      "step": 742
+    },
+    {
+      "epoch": 0.3139240506329114,
+      "grad_norm": 0.6341713070869446,
+      "learning_rate": 0.00013057996485061512,
+      "loss": 0.9217103719711304,
+      "step": 744
+    },
+    {
+      "epoch": 0.31476793248945145,
+      "grad_norm": 0.5294105410575867,
+      "learning_rate": 0.00013093145869947276,
+      "loss": 0.951789915561676,
+      "step": 746
+    },
+    {
+      "epoch": 0.31561181434599156,
+      "grad_norm": 0.6372058391571045,
+      "learning_rate": 0.0001312829525483304,
+      "loss": 0.9459875226020813,
+      "step": 748
+    },
+    {
+      "epoch": 0.31645569620253167,
+      "grad_norm": 0.5979796648025513,
+      "learning_rate": 0.00013163444639718806,
+      "loss": 0.9626097679138184,
+      "step": 750
+    },
+    {
+      "epoch": 0.3172995780590717,
+      "grad_norm": 0.5682399868965149,
+      "learning_rate": 0.0001319859402460457,
+      "loss": 1.0261781215667725,
+      "step": 752
+    },
+    {
+      "epoch": 0.3181434599156118,
+      "grad_norm": 0.5349125266075134,
+      "learning_rate": 0.00013233743409490336,
+      "loss": 0.9319828152656555,
+      "step": 754
+    },
+    {
+      "epoch": 0.3189873417721519,
+      "grad_norm": 0.6093934178352356,
+      "learning_rate": 0.000132688927943761,
+      "loss": 0.9216550588607788,
+      "step": 756
+    },
+    {
+      "epoch": 0.319831223628692,
+      "grad_norm": 0.5188612341880798,
+      "learning_rate": 0.00013304042179261864,
+      "loss": 0.901739776134491,
+      "step": 758
+    },
+    {
+      "epoch": 0.3206751054852321,
+      "grad_norm": 0.5877130627632141,
+      "learning_rate": 0.00013339191564147628,
+      "loss": 1.0362589359283447,
+      "step": 760
+    },
+    {
+      "epoch": 0.32151898734177214,
+      "grad_norm": 0.5542771816253662,
+      "learning_rate": 0.00013374340949033392,
+      "loss": 0.8787116408348083,
+      "step": 762
+    },
+    {
+      "epoch": 0.32236286919831225,
+      "grad_norm": 0.5084902048110962,
+      "learning_rate": 0.00013409490333919156,
+      "loss": 0.9237037301063538,
+      "step": 764
+    },
+    {
+      "epoch": 0.3232067510548523,
+      "grad_norm": 0.5461528301239014,
+      "learning_rate": 0.00013444639718804922,
+      "loss": 1.0150731801986694,
+      "step": 766
+    },
+    {
+      "epoch": 0.3240506329113924,
+      "grad_norm": 0.53483647108078,
+      "learning_rate": 0.00013479789103690686,
+      "loss": 0.8985214829444885,
+      "step": 768
+    },
+    {
+      "epoch": 0.32489451476793246,
+      "grad_norm": 0.5580531358718872,
+      "learning_rate": 0.0001351493848857645,
+      "loss": 1.0225775241851807,
+      "step": 770
+    },
+    {
+      "epoch": 0.32573839662447257,
+      "grad_norm": 0.5203377604484558,
+      "learning_rate": 0.00013550087873462216,
+      "loss": 0.9571293592453003,
+      "step": 772
+    },
+    {
+      "epoch": 0.3265822784810127,
+      "grad_norm": 0.5049671530723572,
+      "learning_rate": 0.0001358523725834798,
+      "loss": 1.0468909740447998,
+      "step": 774
+    },
+    {
+      "epoch": 0.32742616033755273,
+      "grad_norm": 0.4723063111305237,
+      "learning_rate": 0.00013620386643233744,
+      "loss": 0.7743215560913086,
+      "step": 776
+    },
+    {
+      "epoch": 0.32827004219409284,
+      "grad_norm": 0.6310980916023254,
+      "learning_rate": 0.00013655536028119508,
+      "loss": 1.021510362625122,
+      "step": 778
+    },
+    {
+      "epoch": 0.3291139240506329,
+      "grad_norm": 0.47066664695739746,
+      "learning_rate": 0.00013690685413005274,
+      "loss": 0.9134382605552673,
+      "step": 780
+    },
+    {
+      "epoch": 0.329957805907173,
+      "grad_norm": 0.5725092887878418,
+      "learning_rate": 0.00013725834797891038,
+      "loss": 0.9797834753990173,
+      "step": 782
+    },
+    {
+      "epoch": 0.3308016877637131,
+      "grad_norm": 0.5139563083648682,
+      "learning_rate": 0.00013760984182776802,
+      "loss": 0.9372621178627014,
+      "step": 784
+    },
+    {
+      "epoch": 0.33164556962025316,
+      "grad_norm": 0.5275821685791016,
+      "learning_rate": 0.00013796133567662566,
+      "loss": 0.9528245329856873,
+      "step": 786
+    },
+    {
+      "epoch": 0.33248945147679326,
+      "grad_norm": 0.5702582001686096,
+      "learning_rate": 0.0001383128295254833,
+      "loss": 0.9750176072120667,
+      "step": 788
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.5281293392181396,
+      "learning_rate": 0.00013866432337434096,
+      "loss": 0.9412306547164917,
+      "step": 790
+    },
+    {
+      "epoch": 0.3341772151898734,
+      "grad_norm": 0.5578986406326294,
+      "learning_rate": 0.0001390158172231986,
+      "loss": 0.997580885887146,
+      "step": 792
+    },
+    {
+      "epoch": 0.33502109704641353,
+      "grad_norm": 0.50461345911026,
+      "learning_rate": 0.00013936731107205624,
+      "loss": 0.9082320928573608,
+      "step": 794
+    },
+    {
+      "epoch": 0.3358649789029536,
+      "grad_norm": 0.5258530378341675,
+      "learning_rate": 0.0001397188049209139,
+      "loss": 1.0082844495773315,
+      "step": 796
+    },
+    {
+      "epoch": 0.3367088607594937,
+      "grad_norm": 0.5548169016838074,
+      "learning_rate": 0.00014007029876977154,
+      "loss": 0.9729003310203552,
+      "step": 798
+    },
+    {
+      "epoch": 0.33755274261603374,
+      "grad_norm": 0.48601076006889343,
+      "learning_rate": 0.00014042179261862918,
+      "loss": 0.9099526405334473,
+      "step": 800
+    },
+    {
+      "epoch": 0.33755274261603374,
+      "eval_loss": 0.9800403714179993,
+      "eval_runtime": 678.8306,
+      "eval_samples_per_second": 3.104,
+      "eval_steps_per_second": 3.104,
+      "step": 800
+    },
+    {
+      "epoch": 0.33839662447257385,
+      "grad_norm": 0.5413158535957336,
+      "learning_rate": 0.00014077328646748682,
+      "loss": 0.8610644936561584,
+      "step": 802
+    },
+    {
+      "epoch": 0.3392405063291139,
+      "grad_norm": 0.5147035717964172,
+      "learning_rate": 0.00014112478031634446,
+      "loss": 0.9584825038909912,
+      "step": 804
+    },
+    {
+      "epoch": 0.340084388185654,
+      "grad_norm": 0.5931771397590637,
+      "learning_rate": 0.0001414762741652021,
+      "loss": 1.0142558813095093,
+      "step": 806
+    },
+    {
+      "epoch": 0.3409282700421941,
+      "grad_norm": 0.5178377032279968,
+      "learning_rate": 0.00014182776801405976,
+      "loss": 1.0078763961791992,
+      "step": 808
+    },
+    {
+      "epoch": 0.34177215189873417,
+      "grad_norm": 0.5453237295150757,
+      "learning_rate": 0.0001421792618629174,
+      "loss": 0.9107215404510498,
+      "step": 810
+    },
+    {
+      "epoch": 0.3426160337552743,
+      "grad_norm": 0.5886152982711792,
+      "learning_rate": 0.00014253075571177506,
+      "loss": 0.9981362819671631,
+      "step": 812
+    },
+    {
+      "epoch": 0.3434599156118143,
+      "grad_norm": 0.48040178418159485,
+      "learning_rate": 0.0001428822495606327,
+      "loss": 0.9636131525039673,
+      "step": 814
+    },
+    {
+      "epoch": 0.34430379746835443,
+      "grad_norm": 0.5011753439903259,
+      "learning_rate": 0.00014323374340949034,
+      "loss": 0.9590586423873901,
+      "step": 816
+    },
+    {
+      "epoch": 0.34514767932489454,
+      "grad_norm": 0.57858806848526,
+      "learning_rate": 0.00014358523725834798,
+      "loss": 0.978246808052063,
+      "step": 818
+    },
+    {
+      "epoch": 0.3459915611814346,
+      "grad_norm": 0.46092939376831055,
+      "learning_rate": 0.00014393673110720562,
+      "loss": 0.8549934029579163,
+      "step": 820
+    },
+    {
+      "epoch": 0.3468354430379747,
+      "grad_norm": 0.5756489038467407,
+      "learning_rate": 0.00014428822495606326,
+      "loss": 0.9771265387535095,
+      "step": 822
+    },
+    {
+      "epoch": 0.34767932489451475,
+      "grad_norm": 0.5501731634140015,
+      "learning_rate": 0.00014463971880492092,
+      "loss": 0.8739748001098633,
+      "step": 824
+    },
+    {
+      "epoch": 0.34852320675105486,
+      "grad_norm": 0.5451868176460266,
+      "learning_rate": 0.0001449912126537786,
+      "loss": 0.9129468202590942,
+      "step": 826
+    },
+    {
+      "epoch": 0.3493670886075949,
+      "grad_norm": 0.4624619781970978,
+      "learning_rate": 0.00014534270650263623,
+      "loss": 0.9196704030036926,
+      "step": 828
+    },
+    {
+      "epoch": 0.350210970464135,
+      "grad_norm": 0.520878791809082,
+      "learning_rate": 0.00014569420035149386,
+      "loss": 0.9976527690887451,
+      "step": 830
+    },
+    {
+      "epoch": 0.3510548523206751,
+      "grad_norm": 0.4469171464443207,
+      "learning_rate": 0.0001460456942003515,
+      "loss": 0.7753443717956543,
+      "step": 832
+    },
+    {
+      "epoch": 0.3518987341772152,
+      "grad_norm": 0.5105249881744385,
+      "learning_rate": 0.00014639718804920914,
+      "loss": 0.9584846496582031,
+      "step": 834
+    },
+    {
+      "epoch": 0.3527426160337553,
+      "grad_norm": 0.5043913125991821,
+      "learning_rate": 0.00014674868189806678,
+      "loss": 0.9013500213623047,
+      "step": 836
+    },
+    {
+      "epoch": 0.35358649789029534,
+      "grad_norm": 0.575850784778595,
+      "learning_rate": 0.00014710017574692442,
+      "loss": 0.9775562286376953,
+      "step": 838
+    },
+    {
+      "epoch": 0.35443037974683544,
+      "grad_norm": 0.5128876566886902,
+      "learning_rate": 0.00014745166959578208,
+      "loss": 0.9278940558433533,
+      "step": 840
+    },
+    {
+      "epoch": 0.35527426160337555,
+      "grad_norm": 0.5757885575294495,
+      "learning_rate": 0.00014780316344463975,
+      "loss": 1.0091488361358643,
+      "step": 842
+    },
+    {
+      "epoch": 0.3561181434599156,
+      "grad_norm": 0.500934898853302,
+      "learning_rate": 0.00014815465729349739,
+      "loss": 0.9286836981773376,
+      "step": 844
+    },
+    {
+      "epoch": 0.3569620253164557,
+      "grad_norm": 0.5220686197280884,
+      "learning_rate": 0.00014850615114235502,
+      "loss": 0.9484171867370605,
+      "step": 846
+    },
+    {
+      "epoch": 0.35780590717299576,
+      "grad_norm": 0.5494697690010071,
+      "learning_rate": 0.00014885764499121266,
+      "loss": 1.0556397438049316,
+      "step": 848
+    },
+    {
+      "epoch": 0.35864978902953587,
+      "grad_norm": 0.46633943915367126,
+      "learning_rate": 0.0001492091388400703,
+      "loss": 0.933089017868042,
+      "step": 850
+    },
+    {
+      "epoch": 0.3594936708860759,
+      "grad_norm": 0.4864962697029114,
+      "learning_rate": 0.00014956063268892794,
+      "loss": 1.0016963481903076,
+      "step": 852
+    },
+    {
+      "epoch": 0.36033755274261603,
+      "grad_norm": 0.5032764673233032,
+      "learning_rate": 0.0001499121265377856,
+      "loss": 0.9085348844528198,
+      "step": 854
+    },
+    {
+      "epoch": 0.36118143459915614,
+      "grad_norm": 0.5424998998641968,
+      "learning_rate": 0.00015026362038664324,
+      "loss": 0.923502504825592,
+      "step": 856
+    },
+    {
+      "epoch": 0.3620253164556962,
+      "grad_norm": 0.5181655287742615,
+      "learning_rate": 0.00015061511423550088,
+      "loss": 0.919174313545227,
+      "step": 858
+    },
+    {
+      "epoch": 0.3628691983122363,
+      "grad_norm": 0.5129443407058716,
+      "learning_rate": 0.00015096660808435855,
+      "loss": 0.851981520652771,
+      "step": 860
+    },
+    {
+      "epoch": 0.36371308016877635,
+      "grad_norm": 0.49540698528289795,
+      "learning_rate": 0.00015131810193321618,
+      "loss": 0.8633858561515808,
+      "step": 862
+    },
+    {
+      "epoch": 0.36455696202531646,
+      "grad_norm": 0.4706701934337616,
+      "learning_rate": 0.00015166959578207382,
+      "loss": 0.9473677277565002,
+      "step": 864
+    },
+    {
+      "epoch": 0.36540084388185656,
+      "grad_norm": 0.587704598903656,
+      "learning_rate": 0.00015202108963093146,
+      "loss": 0.942383885383606,
+      "step": 866
+    },
+    {
+      "epoch": 0.3662447257383966,
+      "grad_norm": 0.5851273536682129,
+      "learning_rate": 0.0001523725834797891,
+      "loss": 0.9811251163482666,
+      "step": 868
+    },
+    {
+      "epoch": 0.3670886075949367,
+      "grad_norm": 0.46357613801956177,
+      "learning_rate": 0.00015272407732864676,
+      "loss": 0.8151084184646606,
+      "step": 870
+    },
+    {
+      "epoch": 0.3679324894514768,
+      "grad_norm": 0.47250670194625854,
+      "learning_rate": 0.0001530755711775044,
+      "loss": 0.9023333191871643,
+      "step": 872
+    },
+    {
+      "epoch": 0.3687763713080169,
+      "grad_norm": 0.47510042786598206,
+      "learning_rate": 0.00015342706502636204,
+      "loss": 0.8758499622344971,
+      "step": 874
+    },
+    {
+      "epoch": 0.369620253164557,
+      "grad_norm": 0.5687124729156494,
+      "learning_rate": 0.00015377855887521968,
+      "loss": 0.9837421774864197,
+      "step": 876
+    },
+    {
+      "epoch": 0.37046413502109704,
+      "grad_norm": 0.49064236879348755,
+      "learning_rate": 0.00015413005272407735,
+      "loss": 0.9084216356277466,
+      "step": 878
+    },
+    {
+      "epoch": 0.37130801687763715,
+      "grad_norm": 0.5288164615631104,
+      "learning_rate": 0.00015448154657293498,
+      "loss": 0.8425542712211609,
+      "step": 880
+    },
+    {
+      "epoch": 0.3721518987341772,
+      "grad_norm": 0.5446951985359192,
+      "learning_rate": 0.00015483304042179262,
+      "loss": 1.006197452545166,
+      "step": 882
+    },
+    {
+      "epoch": 0.3729957805907173,
+      "grad_norm": 0.46872642636299133,
+      "learning_rate": 0.00015518453427065026,
+      "loss": 0.9779444932937622,
+      "step": 884
+    },
+    {
+      "epoch": 0.37383966244725736,
+      "grad_norm": 0.4852714240550995,
+      "learning_rate": 0.00015553602811950793,
+      "loss": 0.9099963307380676,
+      "step": 886
+    },
+    {
+      "epoch": 0.37468354430379747,
+      "grad_norm": 0.5219841003417969,
+      "learning_rate": 0.00015588752196836556,
+      "loss": 0.9730570316314697,
+      "step": 888
+    },
+    {
+      "epoch": 0.3755274261603376,
+      "grad_norm": 0.5258626341819763,
+      "learning_rate": 0.0001562390158172232,
+      "loss": 0.9304586052894592,
+      "step": 890
+    },
+    {
+      "epoch": 0.3763713080168776,
+      "grad_norm": 0.4266716241836548,
+      "learning_rate": 0.00015659050966608084,
+      "loss": 0.8680716156959534,
+      "step": 892
+    },
+    {
+      "epoch": 0.37721518987341773,
+      "grad_norm": 0.46361327171325684,
+      "learning_rate": 0.0001569420035149385,
+      "loss": 0.8428018093109131,
+      "step": 894
+    },
+    {
+      "epoch": 0.3780590717299578,
+      "grad_norm": 0.5313687920570374,
+      "learning_rate": 0.00015729349736379614,
+      "loss": 0.8465750217437744,
+      "step": 896
+    },
+    {
+      "epoch": 0.3789029535864979,
+      "grad_norm": 0.47166210412979126,
+      "learning_rate": 0.00015764499121265378,
+      "loss": 0.9120327234268188,
+      "step": 898
+    },
+    {
+      "epoch": 0.379746835443038,
+      "grad_norm": 0.4647318720817566,
+      "learning_rate": 0.00015799648506151145,
+      "loss": 0.8950425982475281,
+      "step": 900
+    },
+    {
+      "epoch": 0.379746835443038,
+      "eval_loss": 0.9643027186393738,
+      "eval_runtime": 691.7929,
+      "eval_samples_per_second": 3.046,
+      "eval_steps_per_second": 3.046,
+      "step": 900
+    },
+    {
+      "epoch": 0.38059071729957805,
+      "grad_norm": 0.5445119738578796,
+      "learning_rate": 0.00015834797891036909,
+      "loss": 0.908163845539093,
+      "step": 902
+    },
+    {
+      "epoch": 0.38143459915611816,
+      "grad_norm": 0.4311858117580414,
+      "learning_rate": 0.00015869947275922672,
+      "loss": 0.8945821523666382,
+      "step": 904
+    },
+    {
+      "epoch": 0.3822784810126582,
+      "grad_norm": 0.5590984225273132,
+      "learning_rate": 0.00015905096660808436,
+      "loss": 0.9478458762168884,
+      "step": 906
+    },
+    {
+      "epoch": 0.3831223628691983,
+      "grad_norm": 0.5470241904258728,
+      "learning_rate": 0.000159402460456942,
+      "loss": 0.9259957671165466,
+      "step": 908
+    },
+    {
+      "epoch": 0.38396624472573837,
+      "grad_norm": 0.5498791337013245,
+      "learning_rate": 0.00015975395430579964,
+      "loss": 0.8824930787086487,
+      "step": 910
+    },
+    {
+      "epoch": 0.3848101265822785,
+      "grad_norm": 0.4779198467731476,
+      "learning_rate": 0.0001601054481546573,
+      "loss": 0.8842340707778931,
+      "step": 912
+    },
+    {
+      "epoch": 0.3856540084388186,
+      "grad_norm": 0.5390620827674866,
+      "learning_rate": 0.00016045694200351494,
+      "loss": 0.92950040102005,
+      "step": 914
+    },
+    {
+      "epoch": 0.38649789029535864,
+      "grad_norm": 0.505519688129425,
+      "learning_rate": 0.0001608084358523726,
+      "loss": 0.8420897126197815,
+      "step": 916
+    },
+    {
+      "epoch": 0.38734177215189874,
+      "grad_norm": 0.4463907778263092,
+      "learning_rate": 0.00016115992970123025,
+      "loss": 0.8256624341011047,
+      "step": 918
+    },
+    {
+      "epoch": 0.3881856540084388,
+      "grad_norm": 0.5321422219276428,
+      "learning_rate": 0.00016151142355008788,
+      "loss": 0.8701168298721313,
+      "step": 920
+    },
+    {
+      "epoch": 0.3890295358649789,
+      "grad_norm": 0.4343073070049286,
+      "learning_rate": 0.00016186291739894552,
+      "loss": 0.9019309878349304,
+      "step": 922
+    },
+    {
+      "epoch": 0.389873417721519,
+      "grad_norm": 0.5311984419822693,
+      "learning_rate": 0.00016221441124780316,
+      "loss": 0.8560551404953003,
+      "step": 924
+    },
+    {
+      "epoch": 0.39071729957805906,
+      "grad_norm": 0.588691771030426,
+      "learning_rate": 0.0001625659050966608,
+      "loss": 0.9700050354003906,
+      "step": 926
+    },
+    {
+      "epoch": 0.39156118143459917,
+      "grad_norm": 0.5425586104393005,
+      "learning_rate": 0.00016291739894551844,
+      "loss": 1.043768048286438,
+      "step": 928
+    },
+    {
+      "epoch": 0.3924050632911392,
+      "grad_norm": 0.5228736996650696,
+      "learning_rate": 0.0001632688927943761,
+      "loss": 0.9501712918281555,
+      "step": 930
+    },
+    {
+      "epoch": 0.39324894514767933,
+      "grad_norm": 0.48960360884666443,
+      "learning_rate": 0.00016362038664323377,
+      "loss": 0.9223058223724365,
+      "step": 932
+    },
+    {
+      "epoch": 0.39409282700421944,
+      "grad_norm": 0.45204755663871765,
+      "learning_rate": 0.0001639718804920914,
+      "loss": 0.9692960977554321,
+      "step": 934
+    },
+    {
+      "epoch": 0.3949367088607595,
+      "grad_norm": 0.5299274921417236,
+      "learning_rate": 0.00016432337434094905,
+      "loss": 0.9467466473579407,
+      "step": 936
+    },
+    {
+      "epoch": 0.3957805907172996,
+      "grad_norm": 0.5607715249061584,
+      "learning_rate": 0.00016467486818980668,
+      "loss": 0.9118053317070007,
+      "step": 938
+    },
+    {
+      "epoch": 0.39662447257383965,
+      "grad_norm": 0.5271831154823303,
+      "learning_rate": 0.00016502636203866432,
+      "loss": 0.9131460189819336,
+      "step": 940
+    },
+    {
+      "epoch": 0.39746835443037976,
+      "grad_norm": 0.5075286030769348,
+      "learning_rate": 0.00016537785588752196,
+      "loss": 0.9358300566673279,
+      "step": 942
+    },
+    {
+      "epoch": 0.3983122362869198,
+      "grad_norm": 0.515731155872345,
+      "learning_rate": 0.00016572934973637963,
+      "loss": 0.8908210396766663,
+      "step": 944
+    },
+    {
+      "epoch": 0.3991561181434599,
+      "grad_norm": 0.4856977164745331,
+      "learning_rate": 0.00016608084358523726,
+      "loss": 0.9775290489196777,
+      "step": 946
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.48846355080604553,
+      "learning_rate": 0.00016643233743409493,
+      "loss": 0.8957490921020508,
+      "step": 948
+    },
+    {
+      "epoch": 0.4008438818565401,
+      "grad_norm": 0.42990800738334656,
+      "learning_rate": 0.00016678383128295257,
+      "loss": 0.9036174416542053,
+      "step": 950
+    },
+    {
+      "epoch": 0.4016877637130802,
+      "grad_norm": 0.49552062153816223,
+      "learning_rate": 0.0001671353251318102,
+      "loss": 0.991032600402832,
+      "step": 952
+    },
+    {
+      "epoch": 0.40253164556962023,
+      "grad_norm": 0.4565040171146393,
+      "learning_rate": 0.00016748681898066784,
+      "loss": 0.823063313961029,
+      "step": 954
+    },
+    {
+      "epoch": 0.40337552742616034,
+      "grad_norm": 0.4290153682231903,
+      "learning_rate": 0.00016783831282952548,
+      "loss": 0.8785063624382019,
+      "step": 956
+    },
+    {
+      "epoch": 0.40421940928270045,
+      "grad_norm": 0.5419702529907227,
+      "learning_rate": 0.00016818980667838312,
+      "loss": 0.8763971924781799,
+      "step": 958
+    },
+    {
+      "epoch": 0.4050632911392405,
+      "grad_norm": 0.5177501440048218,
+      "learning_rate": 0.00016854130052724079,
+      "loss": 0.9470553398132324,
+      "step": 960
+    },
+    {
+      "epoch": 0.4059071729957806,
+      "grad_norm": 0.539725661277771,
+      "learning_rate": 0.00016889279437609842,
+      "loss": 0.9235025644302368,
+      "step": 962
+    },
+    {
+      "epoch": 0.40675105485232066,
+      "grad_norm": 0.5324983596801758,
+      "learning_rate": 0.0001692442882249561,
+      "loss": 1.0248996019363403,
+      "step": 964
+    },
+    {
+      "epoch": 0.40759493670886077,
+      "grad_norm": 0.4936407506465912,
+      "learning_rate": 0.00016959578207381373,
+      "loss": 0.9076873660087585,
+      "step": 966
+    },
+    {
+      "epoch": 0.4084388185654008,
+      "grad_norm": 0.4960501194000244,
+      "learning_rate": 0.00016994727592267137,
+      "loss": 0.9162673950195312,
+      "step": 968
+    },
+    {
+      "epoch": 0.4092827004219409,
+      "grad_norm": 0.45093682408332825,
+      "learning_rate": 0.000170298769771529,
+      "loss": 0.904100775718689,
+      "step": 970
+    },
+    {
+      "epoch": 0.41012658227848103,
+      "grad_norm": 0.4560275077819824,
+      "learning_rate": 0.00017065026362038664,
+      "loss": 0.8633337020874023,
+      "step": 972
+    },
+    {
+      "epoch": 0.4109704641350211,
+      "grad_norm": 0.44885000586509705,
+      "learning_rate": 0.00017100175746924428,
+      "loss": 0.8454209566116333,
+      "step": 974
+    },
+    {
+      "epoch": 0.4118143459915612,
+      "grad_norm": 0.47251659631729126,
+      "learning_rate": 0.00017135325131810195,
+      "loss": 0.824730396270752,
+      "step": 976
+    },
+    {
+      "epoch": 0.41265822784810124,
+      "grad_norm": 0.6597666144371033,
+      "learning_rate": 0.00017170474516695959,
+      "loss": 0.9496501684188843,
+      "step": 978
+    },
+    {
+      "epoch": 0.41350210970464135,
+      "grad_norm": 0.39806297421455383,
+      "learning_rate": 0.00017205623901581722,
+      "loss": 0.9419087171554565,
+      "step": 980
+    },
+    {
+      "epoch": 0.41434599156118146,
+      "grad_norm": 0.48231109976768494,
+      "learning_rate": 0.0001724077328646749,
+      "loss": 0.9182976484298706,
+      "step": 982
+    },
+    {
+      "epoch": 0.4151898734177215,
+      "grad_norm": 0.5438776612281799,
+      "learning_rate": 0.00017275922671353253,
+      "loss": 0.9386967420578003,
+      "step": 984
+    },
+    {
+      "epoch": 0.4160337552742616,
+      "grad_norm": 0.4959667921066284,
+      "learning_rate": 0.00017311072056239017,
+      "loss": 0.897849440574646,
+      "step": 986
+    },
+    {
+      "epoch": 0.41687763713080167,
+      "grad_norm": 0.43533357977867126,
+      "learning_rate": 0.0001734622144112478,
+      "loss": 0.8776953816413879,
+      "step": 988
+    },
+    {
+      "epoch": 0.4177215189873418,
+      "grad_norm": 0.47513946890830994,
+      "learning_rate": 0.00017381370826010547,
+      "loss": 0.9162989854812622,
+      "step": 990
+    },
+    {
+      "epoch": 0.41856540084388183,
+      "grad_norm": 0.4907188415527344,
+      "learning_rate": 0.0001741652021089631,
+      "loss": 0.9482660889625549,
+      "step": 992
+    },
+    {
+      "epoch": 0.41940928270042194,
+      "grad_norm": 0.44499966502189636,
+      "learning_rate": 0.00017451669595782075,
+      "loss": 0.8812930583953857,
+      "step": 994
+    },
+    {
+      "epoch": 0.42025316455696204,
+      "grad_norm": 0.4535730481147766,
+      "learning_rate": 0.00017486818980667838,
+      "loss": 0.9439874887466431,
+      "step": 996
+    },
+    {
+      "epoch": 0.4210970464135021,
+      "grad_norm": 0.5240745544433594,
+      "learning_rate": 0.00017521968365553602,
+      "loss": 0.8818395137786865,
+      "step": 998
+    },
+    {
+      "epoch": 0.4219409282700422,
+      "grad_norm": 0.5301211476325989,
+      "learning_rate": 0.0001755711775043937,
+      "loss": 0.886186957359314,
+      "step": 1000
+    },
+    {
+      "epoch": 0.4219409282700422,
+      "eval_loss": 0.9487298727035522,
+      "eval_runtime": 689.4288,
+      "eval_samples_per_second": 3.056,
+      "eval_steps_per_second": 3.056,
+      "step": 1000
+    },
+    {
+      "epoch": 0.42278481012658226,
+      "grad_norm": 0.47876957058906555,
+      "learning_rate": 0.00017592267135325133,
+      "loss": 0.8814021348953247,
+      "step": 1002
+    },
+    {
+      "epoch": 0.42362869198312236,
+      "grad_norm": 0.4929780960083008,
+      "learning_rate": 0.00017627416520210896,
+      "loss": 0.8295068740844727,
+      "step": 1004
+    },
+    {
+      "epoch": 0.42447257383966247,
+      "grad_norm": 0.41888436675071716,
+      "learning_rate": 0.00017662565905096663,
+      "loss": 0.8364827036857605,
+      "step": 1006
+    },
+    {
+      "epoch": 0.4253164556962025,
+      "grad_norm": 0.5175151824951172,
+      "learning_rate": 0.00017697715289982427,
+      "loss": 1.0167189836502075,
+      "step": 1008
+    },
+    {
+      "epoch": 0.42616033755274263,
+      "grad_norm": 0.4815356135368347,
+      "learning_rate": 0.0001773286467486819,
+      "loss": 0.8460752964019775,
+      "step": 1010
+    },
+    {
+      "epoch": 0.4270042194092827,
+      "grad_norm": 0.5210875272750854,
+      "learning_rate": 0.00017768014059753954,
+      "loss": 0.8595574498176575,
+      "step": 1012
+    },
+    {
+      "epoch": 0.4278481012658228,
+      "grad_norm": 0.4400486946105957,
+      "learning_rate": 0.00017803163444639718,
+      "loss": 0.8764723539352417,
+      "step": 1014
+    },
+    {
+      "epoch": 0.4286919831223629,
+      "grad_norm": 0.5282127857208252,
+      "learning_rate": 0.00017838312829525482,
+      "loss": 0.9706798791885376,
+      "step": 1016
+    },
+    {
+      "epoch": 0.42953586497890295,
+      "grad_norm": 0.47523441910743713,
+      "learning_rate": 0.00017873462214411249,
+      "loss": 0.8912002444267273,
+      "step": 1018
+    },
+    {
+      "epoch": 0.43037974683544306,
+      "grad_norm": 0.49640706181526184,
+      "learning_rate": 0.00017908611599297012,
+      "loss": 0.8835636973381042,
+      "step": 1020
+    },
+    {
+      "epoch": 0.4312236286919831,
+      "grad_norm": 0.5253039002418518,
+      "learning_rate": 0.0001794376098418278,
+      "loss": 0.8711735606193542,
+      "step": 1022
+    },
+    {
+      "epoch": 0.4320675105485232,
+      "grad_norm": 0.49285009503364563,
+      "learning_rate": 0.00017978910369068543,
+      "loss": 0.9064869284629822,
+      "step": 1024
+    },
+    {
+      "epoch": 0.43291139240506327,
+      "grad_norm": 0.48758041858673096,
+      "learning_rate": 0.00018014059753954307,
+      "loss": 0.8993359208106995,
+      "step": 1026
+    },
+    {
+      "epoch": 0.4337552742616034,
+      "grad_norm": 0.4815461039543152,
+      "learning_rate": 0.0001804920913884007,
+      "loss": 0.8747937679290771,
+      "step": 1028
+    },
+    {
+      "epoch": 0.4345991561181435,
+      "grad_norm": 0.4930349886417389,
+      "learning_rate": 0.00018084358523725834,
+      "loss": 0.8879084587097168,
+      "step": 1030
+    },
+    {
+      "epoch": 0.43544303797468353,
+      "grad_norm": 0.49303027987480164,
+      "learning_rate": 0.00018119507908611598,
+      "loss": 1.0023083686828613,
+      "step": 1032
+    },
+    {
+      "epoch": 0.43628691983122364,
+      "grad_norm": 0.5312249064445496,
+      "learning_rate": 0.00018154657293497365,
+      "loss": 0.938680112361908,
+      "step": 1034
+    },
+    {
+      "epoch": 0.4371308016877637,
+      "grad_norm": 0.5174582004547119,
+      "learning_rate": 0.0001818980667838313,
+      "loss": 0.854195773601532,
+      "step": 1036
+    },
+    {
+      "epoch": 0.4379746835443038,
+      "grad_norm": 0.5452545881271362,
+      "learning_rate": 0.00018224956063268895,
+      "loss": 0.9059375524520874,
+      "step": 1038
+    },
+    {
+      "epoch": 0.4388185654008439,
+      "grad_norm": 0.5480839014053345,
+      "learning_rate": 0.0001826010544815466,
+      "loss": 0.9708920121192932,
+      "step": 1040
+    },
+    {
+      "epoch": 0.43966244725738396,
+      "grad_norm": 0.44693151116371155,
+      "learning_rate": 0.00018295254833040423,
+      "loss": 0.8561046123504639,
+      "step": 1042
+    },
+    {
+      "epoch": 0.44050632911392407,
+      "grad_norm": 0.4024234116077423,
+      "learning_rate": 0.00018330404217926187,
+      "loss": 0.8811968564987183,
+      "step": 1044
+    },
+    {
+      "epoch": 0.4413502109704641,
+      "grad_norm": 0.5586408376693726,
+      "learning_rate": 0.0001836555360281195,
+      "loss": 0.978068470954895,
+      "step": 1046
+    },
+    {
+      "epoch": 0.4421940928270042,
+      "grad_norm": 0.4918624758720398,
+      "learning_rate": 0.00018400702987697714,
+      "loss": 0.8640981316566467,
+      "step": 1048
+    },
+    {
+      "epoch": 0.4430379746835443,
+      "grad_norm": 0.4335230588912964,
+      "learning_rate": 0.0001843585237258348,
+      "loss": 0.9077964425086975,
+      "step": 1050
+    },
+    {
+      "epoch": 0.4438818565400844,
+      "grad_norm": 0.5275123715400696,
+      "learning_rate": 0.00018471001757469247,
+      "loss": 0.960682213306427,
+      "step": 1052
+    },
+    {
+      "epoch": 0.4447257383966245,
+      "grad_norm": 0.5144415497779846,
+      "learning_rate": 0.0001850615114235501,
+      "loss": 0.8549577593803406,
+      "step": 1054
+    },
+    {
+      "epoch": 0.44556962025316454,
+      "grad_norm": 0.5166662931442261,
+      "learning_rate": 0.00018541300527240775,
+      "loss": 0.9320827126502991,
+      "step": 1056
+    },
+    {
+      "epoch": 0.44641350210970465,
+      "grad_norm": 0.5011980533599854,
+      "learning_rate": 0.0001857644991212654,
+      "loss": 0.9632431864738464,
+      "step": 1058
+    },
+    {
+      "epoch": 0.4472573839662447,
+      "grad_norm": 0.49312469363212585,
+      "learning_rate": 0.00018611599297012303,
+      "loss": 0.9442946910858154,
+      "step": 1060
+    },
+    {
+      "epoch": 0.4481012658227848,
+      "grad_norm": 0.49958568811416626,
+      "learning_rate": 0.00018646748681898066,
+      "loss": 0.952802300453186,
+      "step": 1062
+    },
+    {
+      "epoch": 0.4489451476793249,
+      "grad_norm": 0.49827462434768677,
+      "learning_rate": 0.00018681898066783833,
+      "loss": 0.9630650877952576,
+      "step": 1064
+    },
+    {
+      "epoch": 0.44978902953586497,
+      "grad_norm": 0.523980438709259,
+      "learning_rate": 0.00018717047451669597,
+      "loss": 0.9081395268440247,
+      "step": 1066
+    },
+    {
+      "epoch": 0.4506329113924051,
+      "grad_norm": 0.5108568668365479,
+      "learning_rate": 0.0001875219683655536,
+      "loss": 0.9398958683013916,
+      "step": 1068
+    },
+    {
+      "epoch": 0.45147679324894513,
+      "grad_norm": 0.4453965127468109,
+      "learning_rate": 0.00018787346221441127,
+      "loss": 0.93592768907547,
+      "step": 1070
+    },
+    {
+      "epoch": 0.45232067510548524,
+      "grad_norm": 0.4675683081150055,
+      "learning_rate": 0.0001882249560632689,
+      "loss": 0.8879633545875549,
+      "step": 1072
+    },
+    {
+      "epoch": 0.4531645569620253,
+      "grad_norm": 0.47398847341537476,
+      "learning_rate": 0.00018857644991212655,
+      "loss": 0.9966004490852356,
+      "step": 1074
+    },
+    {
+      "epoch": 0.4540084388185654,
+      "grad_norm": 0.46053192019462585,
+      "learning_rate": 0.0001889279437609842,
+      "loss": 0.9184179902076721,
+      "step": 1076
+    },
+    {
+      "epoch": 0.4548523206751055,
+      "grad_norm": 0.5601398348808289,
+      "learning_rate": 0.00018927943760984182,
+      "loss": 0.9434974193572998,
+      "step": 1078
+    },
+    {
+      "epoch": 0.45569620253164556,
+      "grad_norm": 0.48422637581825256,
+      "learning_rate": 0.0001896309314586995,
+      "loss": 0.9522465467453003,
+      "step": 1080
+    },
+    {
+      "epoch": 0.45654008438818566,
+      "grad_norm": 0.46280911564826965,
+      "learning_rate": 0.00018998242530755713,
+      "loss": 0.9315434694290161,
+      "step": 1082
+    },
+    {
+      "epoch": 0.4573839662447257,
+      "grad_norm": 0.4658683240413666,
+      "learning_rate": 0.00019033391915641477,
+      "loss": 0.8165783286094666,
+      "step": 1084
+    },
+    {
+      "epoch": 0.4582278481012658,
+      "grad_norm": 0.4192182123661041,
+      "learning_rate": 0.0001906854130052724,
+      "loss": 0.8024274110794067,
+      "step": 1086
+    },
+    {
+      "epoch": 0.45907172995780593,
+      "grad_norm": 0.4178735613822937,
+      "learning_rate": 0.00019103690685413007,
+      "loss": 0.8620653748512268,
+      "step": 1088
+    },
+    {
+      "epoch": 0.459915611814346,
+      "grad_norm": 0.4488574266433716,
+      "learning_rate": 0.0001913884007029877,
+      "loss": 0.9121530055999756,
+      "step": 1090
+    },
+    {
+      "epoch": 0.4607594936708861,
+      "grad_norm": 0.5164965987205505,
+      "learning_rate": 0.00019173989455184535,
+      "loss": 0.9496700763702393,
+      "step": 1092
+    },
+    {
+      "epoch": 0.46160337552742614,
+      "grad_norm": 0.41563132405281067,
+      "learning_rate": 0.00019209138840070299,
+      "loss": 0.8757708668708801,
+      "step": 1094
+    },
+    {
+      "epoch": 0.46244725738396625,
+      "grad_norm": 2.3222429752349854,
+      "learning_rate": 0.00019244288224956065,
+      "loss": 0.8406533002853394,
+      "step": 1096
+    },
+    {
+      "epoch": 0.46329113924050636,
+      "grad_norm": 0.4813845157623291,
+      "learning_rate": 0.0001927943760984183,
+      "loss": 0.9459465742111206,
+      "step": 1098
+    },
+    {
+      "epoch": 0.4641350210970464,
+      "grad_norm": 0.6233882308006287,
+      "learning_rate": 0.00019314586994727593,
+      "loss": 0.9376904368400574,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4641350210970464,
+      "eval_loss": 0.9357889294624329,
+      "eval_runtime": 676.9573,
+      "eval_samples_per_second": 3.112,
+      "eval_steps_per_second": 3.112,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4649789029535865,
+      "grad_norm": 0.5125579237937927,
+      "learning_rate": 0.00019349736379613357,
+      "loss": 0.8998825550079346,
+      "step": 1102
+    },
+    {
+      "epoch": 0.46582278481012657,
+      "grad_norm": 0.4534320831298828,
+      "learning_rate": 0.00019384885764499123,
+      "loss": 0.9213768839836121,
+      "step": 1104
+    },
+    {
+      "epoch": 0.4666666666666667,
+      "grad_norm": 0.4715143144130707,
+      "learning_rate": 0.00019420035149384887,
+      "loss": 0.9739661812782288,
+      "step": 1106
+    },
+    {
+      "epoch": 0.4675105485232067,
+      "grad_norm": 0.45529672503471375,
+      "learning_rate": 0.0001945518453427065,
+      "loss": 0.9113216400146484,
+      "step": 1108
+    },
+    {
+      "epoch": 0.46835443037974683,
+      "grad_norm": 0.48199015855789185,
+      "learning_rate": 0.00019490333919156417,
+      "loss": 0.92528235912323,
+      "step": 1110
+    },
+    {
+      "epoch": 0.46919831223628694,
+      "grad_norm": 0.4425188899040222,
+      "learning_rate": 0.0001952548330404218,
+      "loss": 0.8612716197967529,
+      "step": 1112
+    },
+    {
+      "epoch": 0.470042194092827,
+      "grad_norm": 0.4311593770980835,
+      "learning_rate": 0.00019560632688927945,
+      "loss": 0.8901699185371399,
+      "step": 1114
+    },
+    {
+      "epoch": 0.4708860759493671,
+      "grad_norm": 0.497806578874588,
+      "learning_rate": 0.0001959578207381371,
+      "loss": 0.9271994829177856,
+      "step": 1116
+    },
+    {
+      "epoch": 0.47172995780590715,
+      "grad_norm": 0.47149473428726196,
+      "learning_rate": 0.00019630931458699473,
+      "loss": 0.8740925788879395,
+      "step": 1118
+    },
+    {
+      "epoch": 0.47257383966244726,
+      "grad_norm": 0.5029966235160828,
+      "learning_rate": 0.00019666080843585236,
+      "loss": 0.9190115928649902,
+      "step": 1120
+    },
+    {
+      "epoch": 0.47341772151898737,
+      "grad_norm": 0.47972601652145386,
+      "learning_rate": 0.00019701230228471003,
+      "loss": 1.0043057203292847,
+      "step": 1122
+    },
+    {
+      "epoch": 0.4742616033755274,
+      "grad_norm": 0.45096471905708313,
+      "learning_rate": 0.00019736379613356767,
+      "loss": 0.9472925066947937,
+      "step": 1124
+    },
+    {
+      "epoch": 0.4751054852320675,
+      "grad_norm": 0.459852933883667,
+      "learning_rate": 0.00019771528998242533,
+      "loss": 0.8043124079704285,
+      "step": 1126
+    },
+    {
+      "epoch": 0.4759493670886076,
+      "grad_norm": 0.5313422679901123,
+      "learning_rate": 0.00019806678383128297,
+      "loss": 0.9662142395973206,
+      "step": 1128
+    },
+    {
+      "epoch": 0.4767932489451477,
+      "grad_norm": 0.43474531173706055,
+      "learning_rate": 0.0001984182776801406,
+      "loss": 0.8691151738166809,
+      "step": 1130
+    },
+    {
+      "epoch": 0.47763713080168774,
+      "grad_norm": 0.5232312083244324,
+      "learning_rate": 0.00019876977152899825,
+      "loss": 0.9024254083633423,
+      "step": 1132
+    },
+    {
+      "epoch": 0.47848101265822784,
+      "grad_norm": 0.47671905159950256,
+      "learning_rate": 0.0001991212653778559,
+      "loss": 0.8996873497962952,
+      "step": 1134
+    },
+    {
+      "epoch": 0.47932489451476795,
+      "grad_norm": 0.42289480566978455,
+      "learning_rate": 0.00019947275922671353,
+      "loss": 0.8336917757987976,
+      "step": 1136
+    },
+    {
+      "epoch": 0.480168776371308,
+      "grad_norm": 0.4700844883918762,
+      "learning_rate": 0.0001998242530755712,
+      "loss": 0.8943206071853638,
+      "step": 1138
+    },
+    {
+      "epoch": 0.4810126582278481,
+      "grad_norm": 0.5341399312019348,
+      "learning_rate": 0.00019999999711649004,
+      "loss": 0.9340365529060364,
+      "step": 1140
+    },
+    {
+      "epoch": 0.48185654008438816,
+      "grad_norm": 0.46169522404670715,
+      "learning_rate": 0.00019999997404841123,
+      "loss": 0.8241778016090393,
+      "step": 1142
+    },
+    {
+      "epoch": 0.48270042194092827,
+      "grad_norm": 0.5475223064422607,
+      "learning_rate": 0.00019999992791225896,
+      "loss": 1.0096158981323242,
+      "step": 1144
+    },
+    {
+      "epoch": 0.4835443037974684,
+      "grad_norm": 0.524641215801239,
+      "learning_rate": 0.00019999985870804385,
+      "loss": 0.9650378227233887,
+      "step": 1146
+    },
+    {
+      "epoch": 0.48438818565400843,
+      "grad_norm": 0.4326174855232239,
+      "learning_rate": 0.00019999976643578186,
+      "loss": 0.9003009796142578,
+      "step": 1148
+    },
+    {
+      "epoch": 0.48523206751054854,
+      "grad_norm": 0.49034059047698975,
+      "learning_rate": 0.0001999996510954943,
+      "loss": 0.9793432950973511,
+      "step": 1150
+    },
+    {
+      "epoch": 0.4860759493670886,
+      "grad_norm": 0.4760092496871948,
+      "learning_rate": 0.00019999951268720776,
+      "loss": 0.8793007731437683,
+      "step": 1152
+    },
+    {
+      "epoch": 0.4869198312236287,
+      "grad_norm": 0.44606879353523254,
+      "learning_rate": 0.00019999935121095417,
+      "loss": 0.888630747795105,
+      "step": 1154
+    },
+    {
+      "epoch": 0.4877637130801688,
+      "grad_norm": 0.5030332207679749,
+      "learning_rate": 0.0001999991666667708,
+      "loss": 0.9660000801086426,
+      "step": 1156
+    },
+    {
+      "epoch": 0.48860759493670886,
+      "grad_norm": 0.4295555055141449,
+      "learning_rate": 0.00019999895905470014,
+      "loss": 0.824654757976532,
+      "step": 1158
+    },
+    {
+      "epoch": 0.48945147679324896,
+      "grad_norm": 0.392167866230011,
+      "learning_rate": 0.0001999987283747902,
+      "loss": 0.8159562349319458,
+      "step": 1160
+    },
+    {
+      "epoch": 0.490295358649789,
+      "grad_norm": 0.484611839056015,
+      "learning_rate": 0.00019999847462709412,
+      "loss": 0.9630686044692993,
+      "step": 1162
+    },
+    {
+      "epoch": 0.4911392405063291,
+      "grad_norm": 0.45891445875167847,
+      "learning_rate": 0.00019999819781167042,
+      "loss": 0.8396129608154297,
+      "step": 1164
+    },
+    {
+      "epoch": 0.4919831223628692,
+      "grad_norm": 0.5553452968597412,
+      "learning_rate": 0.00019999789792858304,
+      "loss": 0.8194513916969299,
+      "step": 1166
+    },
+    {
+      "epoch": 0.4928270042194093,
+      "grad_norm": 0.38998672366142273,
+      "learning_rate": 0.00019999757497790106,
+      "loss": 0.9268721342086792,
+      "step": 1168
+    },
+    {
+      "epoch": 0.4936708860759494,
+      "grad_norm": 0.45445355772972107,
+      "learning_rate": 0.00019999722895969904,
+      "loss": 0.9204684495925903,
+      "step": 1170
+    },
+    {
+      "epoch": 0.49451476793248944,
+      "grad_norm": 0.484225332736969,
+      "learning_rate": 0.00019999685987405678,
+      "loss": 0.9270301461219788,
+      "step": 1172
+    },
+    {
+      "epoch": 0.49535864978902955,
+      "grad_norm": 0.45215415954589844,
+      "learning_rate": 0.00019999646772105942,
+      "loss": 0.8782645463943481,
+      "step": 1174
+    },
+    {
+      "epoch": 0.4962025316455696,
+      "grad_norm": 0.41578832268714905,
+      "learning_rate": 0.00019999605250079744,
+      "loss": 0.873112678527832,
+      "step": 1176
+    },
+    {
+      "epoch": 0.4970464135021097,
+      "grad_norm": 0.4632788300514221,
+      "learning_rate": 0.0001999956142133666,
+      "loss": 0.9593189358711243,
+      "step": 1178
+    },
+    {
+      "epoch": 0.4978902953586498,
+      "grad_norm": 0.5007622838020325,
+      "learning_rate": 0.000199995152858868,
+      "loss": 0.9215621948242188,
+      "step": 1180
+    },
+    {
+      "epoch": 0.49873417721518987,
+      "grad_norm": 0.5201935768127441,
+      "learning_rate": 0.0001999946684374081,
+      "loss": 0.8964219689369202,
+      "step": 1182
+    },
+    {
+      "epoch": 0.49957805907173,
+      "grad_norm": 0.5053967237472534,
+      "learning_rate": 0.0001999941609490986,
+      "loss": 0.8528663516044617,
+      "step": 1184
+    },
+    {
+      "epoch": 0.5004219409282701,
+      "grad_norm": 0.44397974014282227,
+      "learning_rate": 0.0001999936303940566,
+      "loss": 0.8214734792709351,
+      "step": 1186
+    },
+    {
+      "epoch": 0.5012658227848101,
+      "grad_norm": 0.4563101530075073,
+      "learning_rate": 0.0001999930767724045,
+      "loss": 0.8882166743278503,
+      "step": 1188
+    },
+    {
+      "epoch": 0.5021097046413502,
+      "grad_norm": 0.4666728675365448,
+      "learning_rate": 0.00019999250008426997,
+      "loss": 0.8882588148117065,
+      "step": 1190
+    },
+    {
+      "epoch": 0.5029535864978903,
+      "grad_norm": 0.4423629939556122,
+      "learning_rate": 0.00019999190032978607,
+      "loss": 0.8353691697120667,
+      "step": 1192
+    },
+    {
+      "epoch": 0.5037974683544304,
+      "grad_norm": 0.4514595866203308,
+      "learning_rate": 0.00019999127750909118,
+      "loss": 0.8721219301223755,
+      "step": 1194
+    },
+    {
+      "epoch": 0.5046413502109705,
+      "grad_norm": 0.48469996452331543,
+      "learning_rate": 0.00019999063162232886,
+      "loss": 0.902795135974884,
+      "step": 1196
+    },
+    {
+      "epoch": 0.5054852320675105,
+      "grad_norm": 0.4931983947753906,
+      "learning_rate": 0.00019998996266964823,
+      "loss": 0.8843175768852234,
+      "step": 1198
+    },
+    {
+      "epoch": 0.5063291139240507,
+      "grad_norm": 0.507408082485199,
+      "learning_rate": 0.00019998927065120357,
+      "loss": 0.9179208278656006,
+      "step": 1200
+    },
+    {
+      "epoch": 0.5063291139240507,
+      "eval_loss": 0.9224098324775696,
+      "eval_runtime": 669.7542,
+      "eval_samples_per_second": 3.146,
+      "eval_steps_per_second": 3.146,
+      "step": 1200
+    },
+    {
+      "epoch": 0.5071729957805907,
+      "grad_norm": 0.4024188220500946,
+      "learning_rate": 0.00019998855556715447,
+      "loss": 0.9078981876373291,
+      "step": 1202
+    },
+    {
+      "epoch": 0.5080168776371308,
+      "grad_norm": 0.6558667421340942,
+      "learning_rate": 0.00019998781741766594,
+      "loss": 0.8089252710342407,
+      "step": 1204
+    },
+    {
+      "epoch": 0.5088607594936709,
+      "grad_norm": 0.45128145813941956,
+      "learning_rate": 0.00019998705620290823,
+      "loss": 0.9316248297691345,
+      "step": 1206
+    },
+    {
+      "epoch": 0.509704641350211,
+      "grad_norm": 0.4733511507511139,
+      "learning_rate": 0.00019998627192305694,
+      "loss": 0.9047867655754089,
+      "step": 1208
+    },
+    {
+      "epoch": 0.510548523206751,
+      "grad_norm": 0.4026021659374237,
+      "learning_rate": 0.00019998546457829298,
+      "loss": 0.8507166504859924,
+      "step": 1210
+    },
+    {
+      "epoch": 0.5113924050632911,
+      "grad_norm": 0.43416184186935425,
+      "learning_rate": 0.00019998463416880262,
+      "loss": 0.8490090370178223,
+      "step": 1212
+    },
+    {
+      "epoch": 0.5122362869198313,
+      "grad_norm": 0.3678364157676697,
+      "learning_rate": 0.0001999837806947774,
+      "loss": 0.7730492949485779,
+      "step": 1214
+    },
+    {
+      "epoch": 0.5130801687763713,
+      "grad_norm": 0.49042677879333496,
+      "learning_rate": 0.00019998290415641415,
+      "loss": 0.9625269174575806,
+      "step": 1216
+    },
+    {
+      "epoch": 0.5139240506329114,
+      "grad_norm": 0.4576701819896698,
+      "learning_rate": 0.00019998200455391516,
+      "loss": 0.8961732983589172,
+      "step": 1218
+    },
+    {
+      "epoch": 0.5147679324894515,
+      "grad_norm": 0.4465518593788147,
+      "learning_rate": 0.00019998108188748793,
+      "loss": 0.8508996367454529,
+      "step": 1220
+    },
+    {
+      "epoch": 0.5156118143459916,
+      "grad_norm": 0.4267960488796234,
+      "learning_rate": 0.00019998013615734524,
+      "loss": 0.8664930462837219,
+      "step": 1222
+    },
+    {
+      "epoch": 0.5164556962025316,
+      "grad_norm": 0.39936602115631104,
+      "learning_rate": 0.0001999791673637053,
+      "loss": 0.8841317892074585,
+      "step": 1224
+    },
+    {
+      "epoch": 0.5172995780590718,
+      "grad_norm": 0.4442414343357086,
+      "learning_rate": 0.0001999781755067916,
+      "loss": 0.8601276874542236,
+      "step": 1226
+    },
+    {
+      "epoch": 0.5181434599156118,
+      "grad_norm": 0.4397028088569641,
+      "learning_rate": 0.00019997716058683292,
+      "loss": 0.8377046585083008,
+      "step": 1228
+    },
+    {
+      "epoch": 0.5189873417721519,
+      "grad_norm": 0.5327648520469666,
+      "learning_rate": 0.0001999761226040634,
+      "loss": 0.9853615164756775,
+      "step": 1230
+    },
+    {
+      "epoch": 0.5198312236286919,
+      "grad_norm": 0.40931421518325806,
+      "learning_rate": 0.00019997506155872244,
+      "loss": 0.9235715866088867,
+      "step": 1232
+    },
+    {
+      "epoch": 0.5206751054852321,
+      "grad_norm": 0.4599161148071289,
+      "learning_rate": 0.00019997397745105487,
+      "loss": 0.868396520614624,
+      "step": 1234
+    },
+    {
+      "epoch": 0.5215189873417722,
+      "grad_norm": 0.4813833236694336,
+      "learning_rate": 0.0001999728702813107,
+      "loss": 0.9946733117103577,
+      "step": 1236
+    },
+    {
+      "epoch": 0.5223628691983122,
+      "grad_norm": 0.43518269062042236,
+      "learning_rate": 0.00019997174004974543,
+      "loss": 0.8608635067939758,
+      "step": 1238
+    },
+    {
+      "epoch": 0.5232067510548524,
+      "grad_norm": 0.45646214485168457,
+      "learning_rate": 0.0001999705867566197,
+      "loss": 0.819354772567749,
+      "step": 1240
+    },
+    {
+      "epoch": 0.5240506329113924,
+      "grad_norm": 0.46834197640419006,
+      "learning_rate": 0.00019996941040219954,
+      "loss": 0.8934658765792847,
+      "step": 1242
+    },
+    {
+      "epoch": 0.5248945147679325,
+      "grad_norm": 0.46065032482147217,
+      "learning_rate": 0.0001999682109867564,
+      "loss": 0.8624778985977173,
+      "step": 1244
+    },
+    {
+      "epoch": 0.5257383966244725,
+      "grad_norm": 0.4506741166114807,
+      "learning_rate": 0.00019996698851056688,
+      "loss": 0.8540882468223572,
+      "step": 1246
+    },
+    {
+      "epoch": 0.5265822784810127,
+      "grad_norm": 0.4550519287586212,
+      "learning_rate": 0.00019996574297391302,
+      "loss": 0.9003006219863892,
+      "step": 1248
+    },
+    {
+      "epoch": 0.5274261603375527,
+      "grad_norm": 0.4097813367843628,
+      "learning_rate": 0.00019996447437708214,
+      "loss": 0.8490248918533325,
+      "step": 1250
+    },
+    {
+      "epoch": 0.5282700421940928,
+      "grad_norm": 0.4375711679458618,
+      "learning_rate": 0.00019996318272036688,
+      "loss": 0.9204569458961487,
+      "step": 1252
+    },
+    {
+      "epoch": 0.529113924050633,
+      "grad_norm": 0.4335198998451233,
+      "learning_rate": 0.00019996186800406518,
+      "loss": 0.8831789493560791,
+      "step": 1254
+    },
+    {
+      "epoch": 0.529957805907173,
+      "grad_norm": 0.4840933084487915,
+      "learning_rate": 0.00019996053022848035,
+      "loss": 0.8578064441680908,
+      "step": 1256
+    },
+    {
+      "epoch": 0.5308016877637131,
+      "grad_norm": 0.48911216855049133,
+      "learning_rate": 0.00019995916939392097,
+      "loss": 0.9235416650772095,
+      "step": 1258
+    },
+    {
+      "epoch": 0.5316455696202531,
+      "grad_norm": 0.41106730699539185,
+      "learning_rate": 0.00019995778550070098,
+      "loss": 0.7897764444351196,
+      "step": 1260
+    },
+    {
+      "epoch": 0.5324894514767933,
+      "grad_norm": 0.42233356833457947,
+      "learning_rate": 0.00019995637854913957,
+      "loss": 0.9392989873886108,
+      "step": 1262
+    },
+    {
+      "epoch": 0.5333333333333333,
+      "grad_norm": 0.39145180583000183,
+      "learning_rate": 0.0001999549485395613,
+      "loss": 0.9130091667175293,
+      "step": 1264
+    },
+    {
+      "epoch": 0.5341772151898734,
+      "grad_norm": 0.42780739068984985,
+      "learning_rate": 0.00019995349547229614,
+      "loss": 0.9690561890602112,
+      "step": 1266
+    },
+    {
+      "epoch": 0.5350210970464135,
+      "grad_norm": 0.4823327362537384,
+      "learning_rate": 0.0001999520193476792,
+      "loss": 0.9990131258964539,
+      "step": 1268
+    },
+    {
+      "epoch": 0.5358649789029536,
+      "grad_norm": 0.4669715166091919,
+      "learning_rate": 0.00019995052016605097,
+      "loss": 0.8343052864074707,
+      "step": 1270
+    },
+    {
+      "epoch": 0.5367088607594936,
+      "grad_norm": 0.4024743437767029,
+      "learning_rate": 0.0001999489979277574,
+      "loss": 0.9605970978736877,
+      "step": 1272
+    },
+    {
+      "epoch": 0.5375527426160338,
+      "grad_norm": 0.42063432931900024,
+      "learning_rate": 0.0001999474526331495,
+      "loss": 0.8164438009262085,
+      "step": 1274
+    },
+    {
+      "epoch": 0.5383966244725739,
+      "grad_norm": 0.47953495383262634,
+      "learning_rate": 0.00019994588428258383,
+      "loss": 0.879767656326294,
+      "step": 1276
+    },
+    {
+      "epoch": 0.5392405063291139,
+      "grad_norm": 0.39668557047843933,
+      "learning_rate": 0.00019994429287642214,
+      "loss": 0.7658726572990417,
+      "step": 1278
+    },
+    {
+      "epoch": 0.540084388185654,
+      "grad_norm": 0.5136121511459351,
+      "learning_rate": 0.00019994267841503155,
+      "loss": 0.9477384686470032,
+      "step": 1280
+    },
+    {
+      "epoch": 0.5409282700421941,
+      "grad_norm": 0.4039286673069,
+      "learning_rate": 0.00019994104089878452,
+      "loss": 0.7883896827697754,
+      "step": 1282
+    },
+    {
+      "epoch": 0.5417721518987342,
+      "grad_norm": 0.4425487816333771,
+      "learning_rate": 0.0001999393803280587,
+      "loss": 0.8526129126548767,
+      "step": 1284
+    },
+    {
+      "epoch": 0.5426160337552742,
+      "grad_norm": 0.4544559121131897,
+      "learning_rate": 0.00019993769670323725,
+      "loss": 0.8694683909416199,
+      "step": 1286
+    },
+    {
+      "epoch": 0.5434599156118144,
+      "grad_norm": 0.42872917652130127,
+      "learning_rate": 0.0001999359900247085,
+      "loss": 0.8211527466773987,
+      "step": 1288
+    },
+    {
+      "epoch": 0.5443037974683544,
+      "grad_norm": 0.41549673676490784,
+      "learning_rate": 0.00019993426029286617,
+      "loss": 0.7967932224273682,
+      "step": 1290
+    },
+    {
+      "epoch": 0.5451476793248945,
+      "grad_norm": 0.42888355255126953,
+      "learning_rate": 0.00019993250750810926,
+      "loss": 0.8291563987731934,
+      "step": 1292
+    },
+    {
+      "epoch": 0.5459915611814345,
+      "grad_norm": 0.44985485076904297,
+      "learning_rate": 0.0001999307316708421,
+      "loss": 0.8787803649902344,
+      "step": 1294
+    },
+    {
+      "epoch": 0.5468354430379747,
+      "grad_norm": 0.43326008319854736,
+      "learning_rate": 0.00019992893278147436,
+      "loss": 0.8375519514083862,
+      "step": 1296
+    },
+    {
+      "epoch": 0.5476793248945148,
+      "grad_norm": 0.406556099653244,
+      "learning_rate": 0.00019992711084042096,
+      "loss": 0.8538051247596741,
+      "step": 1298
+    },
+    {
+      "epoch": 0.5485232067510548,
+      "grad_norm": 0.46090877056121826,
+      "learning_rate": 0.00019992526584810226,
+      "loss": 0.9722132086753845,
+      "step": 1300
+    },
+    {
+      "epoch": 0.5485232067510548,
+      "eval_loss": 0.9068717360496521,
+      "eval_runtime": 680.7718,
+      "eval_samples_per_second": 3.095,
+      "eval_steps_per_second": 3.095,
+      "step": 1300
+    },
+    {
+      "epoch": 0.549367088607595,
+      "grad_norm": 0.44932207465171814,
+      "learning_rate": 0.00019992339780494382,
+      "loss": 0.8474152684211731,
+      "step": 1302
+    },
+    {
+      "epoch": 0.550210970464135,
+      "grad_norm": 0.5185276865959167,
+      "learning_rate": 0.0001999215067113766,
+      "loss": 0.8440775871276855,
+      "step": 1304
+    },
+    {
+      "epoch": 0.5510548523206751,
+      "grad_norm": 0.43811365962028503,
+      "learning_rate": 0.0001999195925678368,
+      "loss": 0.8848010897636414,
+      "step": 1306
+    },
+    {
+      "epoch": 0.5518987341772152,
+      "grad_norm": 0.4861660301685333,
+      "learning_rate": 0.00019991765537476596,
+      "loss": 0.8265100121498108,
+      "step": 1308
+    },
+    {
+      "epoch": 0.5527426160337553,
+      "grad_norm": 0.5068721771240234,
+      "learning_rate": 0.000199915695132611,
+      "loss": 0.9327963590621948,
+      "step": 1310
+    },
+    {
+      "epoch": 0.5535864978902953,
+      "grad_norm": 0.4902805685997009,
+      "learning_rate": 0.0001999137118418241,
+      "loss": 0.9315155744552612,
+      "step": 1312
+    },
+    {
+      "epoch": 0.5544303797468354,
+      "grad_norm": 0.4530964493751526,
+      "learning_rate": 0.00019991170550286272,
+      "loss": 0.9344852566719055,
+      "step": 1314
+    },
+    {
+      "epoch": 0.5552742616033756,
+      "grad_norm": 0.4106673300266266,
+      "learning_rate": 0.00019990967611618974,
+      "loss": 0.8326173424720764,
+      "step": 1316
+    },
+    {
+      "epoch": 0.5561181434599156,
+      "grad_norm": 0.4439711570739746,
+      "learning_rate": 0.0001999076236822733,
+      "loss": 0.8934909105300903,
+      "step": 1318
+    },
+    {
+      "epoch": 0.5569620253164557,
+      "grad_norm": 0.40245553851127625,
+      "learning_rate": 0.0001999055482015868,
+      "loss": 0.6691107153892517,
+      "step": 1320
+    },
+    {
+      "epoch": 0.5578059071729958,
+      "grad_norm": 0.43904435634613037,
+      "learning_rate": 0.0001999034496746091,
+      "loss": 0.8560307621955872,
+      "step": 1322
+    },
+    {
+      "epoch": 0.5586497890295359,
+      "grad_norm": 0.5131705403327942,
+      "learning_rate": 0.00019990132810182422,
+      "loss": 0.9024442434310913,
+      "step": 1324
+    },
+    {
+      "epoch": 0.5594936708860759,
+      "grad_norm": 0.4539531171321869,
+      "learning_rate": 0.00019989918348372156,
+      "loss": 0.8675428628921509,
+      "step": 1326
+    },
+    {
+      "epoch": 0.560337552742616,
+      "grad_norm": 0.44469985365867615,
+      "learning_rate": 0.00019989701582079591,
+      "loss": 0.8853429555892944,
+      "step": 1328
+    },
+    {
+      "epoch": 0.5611814345991561,
+      "grad_norm": 0.4485546946525574,
+      "learning_rate": 0.00019989482511354725,
+      "loss": 0.9675378799438477,
+      "step": 1330
+    },
+    {
+      "epoch": 0.5620253164556962,
+      "grad_norm": 0.43316468596458435,
+      "learning_rate": 0.00019989261136248097,
+      "loss": 0.8256655335426331,
+      "step": 1332
+    },
+    {
+      "epoch": 0.5628691983122363,
+      "grad_norm": 0.4315780699253082,
+      "learning_rate": 0.00019989037456810772,
+      "loss": 0.7727690935134888,
+      "step": 1334
+    },
+    {
+      "epoch": 0.5637130801687764,
+      "grad_norm": 0.45147502422332764,
+      "learning_rate": 0.00019988811473094348,
+      "loss": 0.9240618348121643,
+      "step": 1336
+    },
+    {
+      "epoch": 0.5645569620253165,
+      "grad_norm": 0.4606908857822418,
+      "learning_rate": 0.00019988583185150957,
+      "loss": 0.8005949258804321,
+      "step": 1338
+    },
+    {
+      "epoch": 0.5654008438818565,
+      "grad_norm": 0.48932451009750366,
+      "learning_rate": 0.00019988352593033255,
+      "loss": 0.9042718410491943,
+      "step": 1340
+    },
+    {
+      "epoch": 0.5662447257383966,
+      "grad_norm": 0.46022218465805054,
+      "learning_rate": 0.00019988119696794443,
+      "loss": 0.8724613785743713,
+      "step": 1342
+    },
+    {
+      "epoch": 0.5670886075949367,
+      "grad_norm": 0.4414621889591217,
+      "learning_rate": 0.00019987884496488242,
+      "loss": 0.8653066158294678,
+      "step": 1344
+    },
+    {
+      "epoch": 0.5679324894514768,
+      "grad_norm": 0.4490342140197754,
+      "learning_rate": 0.0001998764699216891,
+      "loss": 0.8843849301338196,
+      "step": 1346
+    },
+    {
+      "epoch": 0.5687763713080168,
+      "grad_norm": 0.47690069675445557,
+      "learning_rate": 0.00019987407183891232,
+      "loss": 0.8385393023490906,
+      "step": 1348
+    },
+    {
+      "epoch": 0.569620253164557,
+      "grad_norm": 0.41409361362457275,
+      "learning_rate": 0.00019987165071710527,
+      "loss": 0.77492755651474,
+      "step": 1350
+    },
+    {
+      "epoch": 0.570464135021097,
+      "grad_norm": 0.4151647984981537,
+      "learning_rate": 0.0001998692065568265,
+      "loss": 0.8368680477142334,
+      "step": 1352
+    },
+    {
+      "epoch": 0.5713080168776371,
+      "grad_norm": 0.4105067849159241,
+      "learning_rate": 0.0001998667393586398,
+      "loss": 0.833220362663269,
+      "step": 1354
+    },
+    {
+      "epoch": 0.5721518987341773,
+      "grad_norm": 0.44701820611953735,
+      "learning_rate": 0.0001998642491231143,
+      "loss": 0.8041552305221558,
+      "step": 1356
+    },
+    {
+      "epoch": 0.5729957805907173,
+      "grad_norm": 0.4621582329273224,
+      "learning_rate": 0.00019986173585082444,
+      "loss": 0.9052709937095642,
+      "step": 1358
+    },
+    {
+      "epoch": 0.5738396624472574,
+      "grad_norm": 0.4934713840484619,
+      "learning_rate": 0.00019985919954235005,
+      "loss": 0.8036054372787476,
+      "step": 1360
+    },
+    {
+      "epoch": 0.5746835443037974,
+      "grad_norm": 0.4557384252548218,
+      "learning_rate": 0.0001998566401982761,
+      "loss": 0.8340095281600952,
+      "step": 1362
+    },
+    {
+      "epoch": 0.5755274261603376,
+      "grad_norm": 0.5048403739929199,
+      "learning_rate": 0.00019985405781919305,
+      "loss": 0.8703633546829224,
+      "step": 1364
+    },
+    {
+      "epoch": 0.5763713080168776,
+      "grad_norm": 0.4707714319229126,
+      "learning_rate": 0.0001998514524056966,
+      "loss": 0.9115830659866333,
+      "step": 1366
+    },
+    {
+      "epoch": 0.5772151898734177,
+      "grad_norm": 0.4502186179161072,
+      "learning_rate": 0.00019984882395838777,
+      "loss": 0.8603149056434631,
+      "step": 1368
+    },
+    {
+      "epoch": 0.5780590717299579,
+      "grad_norm": 0.4229136109352112,
+      "learning_rate": 0.00019984617247787288,
+      "loss": 0.7482197284698486,
+      "step": 1370
+    },
+    {
+      "epoch": 0.5789029535864979,
+      "grad_norm": 0.49208030104637146,
+      "learning_rate": 0.0001998434979647636,
+      "loss": 0.9071239233016968,
+      "step": 1372
+    },
+    {
+      "epoch": 0.579746835443038,
+      "grad_norm": 0.47393161058425903,
+      "learning_rate": 0.00019984080041967687,
+      "loss": 0.9370644092559814,
+      "step": 1374
+    },
+    {
+      "epoch": 0.580590717299578,
+      "grad_norm": 0.4483324885368347,
+      "learning_rate": 0.00019983807984323492,
+      "loss": 0.8536180257797241,
+      "step": 1376
+    },
+    {
+      "epoch": 0.5814345991561182,
+      "grad_norm": 0.4163796007633209,
+      "learning_rate": 0.00019983533623606543,
+      "loss": 0.7360405325889587,
+      "step": 1378
+    },
+    {
+      "epoch": 0.5822784810126582,
+      "grad_norm": 0.44147446751594543,
+      "learning_rate": 0.00019983256959880122,
+      "loss": 0.8162824511528015,
+      "step": 1380
+    },
+    {
+      "epoch": 0.5831223628691983,
+      "grad_norm": 0.44443491101264954,
+      "learning_rate": 0.0001998297799320805,
+      "loss": 0.9169327616691589,
+      "step": 1382
+    },
+    {
+      "epoch": 0.5839662447257384,
+      "grad_norm": 0.4641949236392975,
+      "learning_rate": 0.00019982696723654686,
+      "loss": 0.8875693678855896,
+      "step": 1384
+    },
+    {
+      "epoch": 0.5848101265822785,
+      "grad_norm": 0.4327974319458008,
+      "learning_rate": 0.00019982413151284906,
+      "loss": 0.7747344970703125,
+      "step": 1386
+    },
+    {
+      "epoch": 0.5856540084388185,
+      "grad_norm": 0.44200772047042847,
+      "learning_rate": 0.0001998212727616413,
+      "loss": 0.8490481972694397,
+      "step": 1388
+    },
+    {
+      "epoch": 0.5864978902953587,
+      "grad_norm": 0.39084959030151367,
+      "learning_rate": 0.000199818390983583,
+      "loss": 0.8014808297157288,
+      "step": 1390
+    },
+    {
+      "epoch": 0.5873417721518988,
+      "grad_norm": 0.3691277801990509,
+      "learning_rate": 0.00019981548617933897,
+      "loss": 0.8650928139686584,
+      "step": 1392
+    },
+    {
+      "epoch": 0.5881856540084388,
+      "grad_norm": 0.480002760887146,
+      "learning_rate": 0.00019981255834957925,
+      "loss": 0.7713267803192139,
+      "step": 1394
+    },
+    {
+      "epoch": 0.5890295358649789,
+      "grad_norm": 0.4397226572036743,
+      "learning_rate": 0.00019980960749497927,
+      "loss": 0.9053656458854675,
+      "step": 1396
+    },
+    {
+      "epoch": 0.589873417721519,
+      "grad_norm": 0.4263727366924286,
+      "learning_rate": 0.00019980663361621973,
+      "loss": 0.8737669587135315,
+      "step": 1398
+    },
+    {
+      "epoch": 0.5907172995780591,
+      "grad_norm": 0.4468817710876465,
+      "learning_rate": 0.0001998036367139866,
+      "loss": 0.9396650791168213,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5907172995780591,
+      "eval_loss": 0.8971880674362183,
+      "eval_runtime": 692.8046,
+      "eval_samples_per_second": 3.041,
+      "eval_steps_per_second": 3.041,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5915611814345991,
+      "grad_norm": 0.4128144383430481,
+      "learning_rate": 0.00019980061678897127,
+      "loss": 0.926134467124939,
+      "step": 1402
+    },
+    {
+      "epoch": 0.5924050632911393,
+      "grad_norm": 0.45524629950523376,
+      "learning_rate": 0.00019979757384187035,
+      "loss": 0.826360285282135,
+      "step": 1404
+    },
+    {
+      "epoch": 0.5932489451476793,
+      "grad_norm": 0.42521336674690247,
+      "learning_rate": 0.0001997945078733858,
+      "loss": 0.8147702813148499,
+      "step": 1406
+    },
+    {
+      "epoch": 0.5940928270042194,
+      "grad_norm": 0.4172525107860565,
+      "learning_rate": 0.00019979141888422489,
+      "loss": 0.7711596488952637,
+      "step": 1408
+    },
+    {
+      "epoch": 0.5949367088607594,
+      "grad_norm": 0.46193623542785645,
+      "learning_rate": 0.00019978830687510013,
+      "loss": 0.7875980138778687,
+      "step": 1410
+    },
+    {
+      "epoch": 0.5957805907172996,
+      "grad_norm": 0.4455825686454773,
+      "learning_rate": 0.00019978517184672946,
+      "loss": 0.8590195178985596,
+      "step": 1412
+    },
+    {
+      "epoch": 0.5966244725738397,
+      "grad_norm": 0.4266505539417267,
+      "learning_rate": 0.00019978201379983608,
+      "loss": 0.9318227767944336,
+      "step": 1414
+    },
+    {
+      "epoch": 0.5974683544303797,
+      "grad_norm": 0.42627251148223877,
+      "learning_rate": 0.00019977883273514843,
+      "loss": 0.8644474744796753,
+      "step": 1416
+    },
+    {
+      "epoch": 0.5983122362869199,
+      "grad_norm": 0.393873393535614,
+      "learning_rate": 0.00019977562865340038,
+      "loss": 0.7760446071624756,
+      "step": 1418
+    },
+    {
+      "epoch": 0.5991561181434599,
+      "grad_norm": 0.44334208965301514,
+      "learning_rate": 0.000199772401555331,
+      "loss": 0.8071584701538086,
+      "step": 1420
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.39390063285827637,
+      "learning_rate": 0.0001997691514416848,
+      "loss": 0.9076889753341675,
+      "step": 1422
+    },
+    {
+      "epoch": 0.60084388185654,
+      "grad_norm": 0.46753421425819397,
+      "learning_rate": 0.00019976587831321144,
+      "loss": 0.8534318804740906,
+      "step": 1424
+    },
+    {
+      "epoch": 0.6016877637130802,
+      "grad_norm": 0.4840681254863739,
+      "learning_rate": 0.00019976258217066598,
+      "loss": 0.9049596190452576,
+      "step": 1426
+    },
+    {
+      "epoch": 0.6025316455696202,
+      "grad_norm": 0.4740557372570038,
+      "learning_rate": 0.00019975926301480878,
+      "loss": 0.8520918488502502,
+      "step": 1428
+    },
+    {
+      "epoch": 0.6033755274261603,
+      "grad_norm": 0.46498584747314453,
+      "learning_rate": 0.00019975592084640553,
+      "loss": 0.8221305012702942,
+      "step": 1430
+    },
+    {
+      "epoch": 0.6042194092827005,
+      "grad_norm": 0.40885692834854126,
+      "learning_rate": 0.0001997525556662272,
+      "loss": 0.8132198452949524,
+      "step": 1432
+    },
+    {
+      "epoch": 0.6050632911392405,
+      "grad_norm": 0.4375821053981781,
+      "learning_rate": 0.00019974916747505006,
+      "loss": 0.870534360408783,
+      "step": 1434
+    },
+    {
+      "epoch": 0.6059071729957806,
+      "grad_norm": 0.4780935049057007,
+      "learning_rate": 0.00019974575627365569,
+      "loss": 0.8825759291648865,
+      "step": 1436
+    },
+    {
+      "epoch": 0.6067510548523207,
+      "grad_norm": 0.4623970687389374,
+      "learning_rate": 0.00019974232206283098,
+      "loss": 0.8187641501426697,
+      "step": 1438
+    },
+    {
+      "epoch": 0.6075949367088608,
+      "grad_norm": 0.42172595858573914,
+      "learning_rate": 0.0001997388648433682,
+      "loss": 0.8569821119308472,
+      "step": 1440
+    },
+    {
+      "epoch": 0.6084388185654008,
+      "grad_norm": 0.49386659264564514,
+      "learning_rate": 0.0001997353846160648,
+      "loss": 0.9173614382743835,
+      "step": 1442
+    },
+    {
+      "epoch": 0.6092827004219409,
+      "grad_norm": 0.47122615575790405,
+      "learning_rate": 0.00019973188138172363,
+      "loss": 0.9444975256919861,
+      "step": 1444
+    },
+    {
+      "epoch": 0.610126582278481,
+      "grad_norm": 0.4831530451774597,
+      "learning_rate": 0.0001997283551411528,
+      "loss": 0.9537245631217957,
+      "step": 1446
+    },
+    {
+      "epoch": 0.6109704641350211,
+      "grad_norm": 0.44460317492485046,
+      "learning_rate": 0.0001997248058951658,
+      "loss": 0.8481367230415344,
+      "step": 1448
+    },
+    {
+      "epoch": 0.6118143459915611,
+      "grad_norm": 0.4315263628959656,
+      "learning_rate": 0.00019972123364458128,
+      "loss": 0.8572371006011963,
+      "step": 1450
+    },
+    {
+      "epoch": 0.6126582278481013,
+      "grad_norm": 0.38559582829475403,
+      "learning_rate": 0.00019971763839022336,
+      "loss": 0.762590765953064,
+      "step": 1452
+    },
+    {
+      "epoch": 0.6135021097046414,
+      "grad_norm": 0.3820290267467499,
+      "learning_rate": 0.00019971402013292138,
+      "loss": 0.7879500389099121,
+      "step": 1454
+    },
+    {
+      "epoch": 0.6143459915611814,
+      "grad_norm": 0.42509570717811584,
+      "learning_rate": 0.00019971037887351,
+      "loss": 0.8696863055229187,
+      "step": 1456
+    },
+    {
+      "epoch": 0.6151898734177215,
+      "grad_norm": 0.46750548481941223,
+      "learning_rate": 0.00019970671461282916,
+      "loss": 0.8425862789154053,
+      "step": 1458
+    },
+    {
+      "epoch": 0.6160337552742616,
+      "grad_norm": 0.3937121033668518,
+      "learning_rate": 0.0001997030273517242,
+      "loss": 0.8028931021690369,
+      "step": 1460
+    },
+    {
+      "epoch": 0.6168776371308017,
+      "grad_norm": 0.4361220896244049,
+      "learning_rate": 0.00019969931709104565,
+      "loss": 0.8662509918212891,
+      "step": 1462
+    },
+    {
+      "epoch": 0.6177215189873417,
+      "grad_norm": 0.4737963378429413,
+      "learning_rate": 0.00019969558383164943,
+      "loss": 0.8962596654891968,
+      "step": 1464
+    },
+    {
+      "epoch": 0.6185654008438819,
+      "grad_norm": 0.3952867388725281,
+      "learning_rate": 0.00019969182757439668,
+      "loss": 0.8630867600440979,
+      "step": 1466
+    },
+    {
+      "epoch": 0.619409282700422,
+      "grad_norm": 0.41688743233680725,
+      "learning_rate": 0.00019968804832015393,
+      "loss": 0.8759240508079529,
+      "step": 1468
+    },
+    {
+      "epoch": 0.620253164556962,
+      "grad_norm": 0.4599224627017975,
+      "learning_rate": 0.00019968424606979298,
+      "loss": 0.8459385633468628,
+      "step": 1470
+    },
+    {
+      "epoch": 0.6210970464135022,
+      "grad_norm": 0.5476765632629395,
+      "learning_rate": 0.00019968042082419094,
+      "loss": 0.8844659328460693,
+      "step": 1472
+    },
+    {
+      "epoch": 0.6219409282700422,
+      "grad_norm": 0.46202269196510315,
+      "learning_rate": 0.00019967657258423022,
+      "loss": 0.860946536064148,
+      "step": 1474
+    },
+    {
+      "epoch": 0.6227848101265823,
+      "grad_norm": 0.39446666836738586,
+      "learning_rate": 0.00019967270135079853,
+      "loss": 0.7587860822677612,
+      "step": 1476
+    },
+    {
+      "epoch": 0.6236286919831223,
+      "grad_norm": 0.4500375986099243,
+      "learning_rate": 0.0001996688071247889,
+      "loss": 0.9143000245094299,
+      "step": 1478
+    },
+    {
+      "epoch": 0.6244725738396625,
+      "grad_norm": 0.4203545153141022,
+      "learning_rate": 0.0001996648899070996,
+      "loss": 0.7865519523620605,
+      "step": 1480
+    },
+    {
+      "epoch": 0.6253164556962025,
+      "grad_norm": 0.45623889565467834,
+      "learning_rate": 0.00019966094969863432,
+      "loss": 0.9192912578582764,
+      "step": 1482
+    },
+    {
+      "epoch": 0.6261603375527426,
+      "grad_norm": 0.4139637053012848,
+      "learning_rate": 0.00019965698650030195,
+      "loss": 0.8459161520004272,
+      "step": 1484
+    },
+    {
+      "epoch": 0.6270042194092827,
+      "grad_norm": 0.4410824179649353,
+      "learning_rate": 0.00019965300031301678,
+      "loss": 0.8159077763557434,
+      "step": 1486
+    },
+    {
+      "epoch": 0.6278481012658228,
+      "grad_norm": 0.43821197748184204,
+      "learning_rate": 0.00019964899113769828,
+      "loss": 0.884467363357544,
+      "step": 1488
+    },
+    {
+      "epoch": 0.6286919831223629,
+      "grad_norm": 0.4692091643810272,
+      "learning_rate": 0.00019964495897527133,
+      "loss": 0.8693601489067078,
+      "step": 1490
+    },
+    {
+      "epoch": 0.6295358649789029,
+      "grad_norm": 0.40715447068214417,
+      "learning_rate": 0.00019964090382666608,
+      "loss": 0.8897743225097656,
+      "step": 1492
+    },
+    {
+      "epoch": 0.6303797468354431,
+      "grad_norm": 0.4123334586620331,
+      "learning_rate": 0.0001996368256928179,
+      "loss": 0.8274733424186707,
+      "step": 1494
+    },
+    {
+      "epoch": 0.6312236286919831,
+      "grad_norm": 0.44347211718559265,
+      "learning_rate": 0.00019963272457466767,
+      "loss": 0.8158749938011169,
+      "step": 1496
+    },
+    {
+      "epoch": 0.6320675105485232,
+      "grad_norm": 0.4700329601764679,
+      "learning_rate": 0.00019962860047316135,
+      "loss": 0.8439569473266602,
+      "step": 1498
+    },
+    {
+      "epoch": 0.6329113924050633,
+      "grad_norm": 0.46415844559669495,
+      "learning_rate": 0.00019962445338925027,
+      "loss": 0.8316822648048401,
+      "step": 1500
+    },
+    {
+      "epoch": 0.6329113924050633,
+      "eval_loss": 0.887488842010498,
+      "eval_runtime": 686.2804,
+      "eval_samples_per_second": 3.07,
+      "eval_steps_per_second": 3.07,
+      "step": 1500
+    },
+    {
+      "epoch": 0.6337552742616034,
+      "grad_norm": 0.44183167815208435,
+      "learning_rate": 0.0001996202833238911,
+      "loss": 0.9121994972229004,
+      "step": 1502
+    },
+    {
+      "epoch": 0.6345991561181434,
+      "grad_norm": 0.4279208779335022,
+      "learning_rate": 0.00019961609027804587,
+      "loss": 0.9160211086273193,
+      "step": 1504
+    },
+    {
+      "epoch": 0.6354430379746835,
+      "grad_norm": 0.3967169523239136,
+      "learning_rate": 0.00019961187425268176,
+      "loss": 0.861677348613739,
+      "step": 1506
+    },
+    {
+      "epoch": 0.6362869198312237,
+      "grad_norm": 0.4796451926231384,
+      "learning_rate": 0.0001996076352487713,
+      "loss": 0.8478423357009888,
+      "step": 1508
+    },
+    {
+      "epoch": 0.6371308016877637,
+      "grad_norm": 0.3861426115036011,
+      "learning_rate": 0.00019960337326729245,
+      "loss": 0.8739159107208252,
+      "step": 1510
+    },
+    {
+      "epoch": 0.6379746835443038,
+      "grad_norm": 0.4567820429801941,
+      "learning_rate": 0.00019959908830922824,
+      "loss": 0.7693920731544495,
+      "step": 1512
+    },
+    {
+      "epoch": 0.6388185654008439,
+      "grad_norm": 0.4457108676433563,
+      "learning_rate": 0.00019959478037556724,
+      "loss": 0.8482301831245422,
+      "step": 1514
+    },
+    {
+      "epoch": 0.639662447257384,
+      "grad_norm": 0.43918928503990173,
+      "learning_rate": 0.00019959044946730314,
+      "loss": 0.8626812696456909,
+      "step": 1516
+    },
+    {
+      "epoch": 0.640506329113924,
+      "grad_norm": 0.4095900058746338,
+      "learning_rate": 0.00019958609558543504,
+      "loss": 0.8342230916023254,
+      "step": 1518
+    },
+    {
+      "epoch": 0.6413502109704642,
+      "grad_norm": 0.4181270897388458,
+      "learning_rate": 0.00019958171873096724,
+      "loss": 0.8794118165969849,
+      "step": 1520
+    },
+    {
+      "epoch": 0.6421940928270042,
+      "grad_norm": 0.4750959277153015,
+      "learning_rate": 0.00019957731890490947,
+      "loss": 0.7787677049636841,
+      "step": 1522
+    },
+    {
+      "epoch": 0.6430379746835443,
+      "grad_norm": 0.5129296779632568,
+      "learning_rate": 0.00019957289610827663,
+      "loss": 0.9084368348121643,
+      "step": 1524
+    },
+    {
+      "epoch": 0.6438818565400843,
+      "grad_norm": 0.4829029142856598,
+      "learning_rate": 0.000199568450342089,
+      "loss": 0.9066952466964722,
+      "step": 1526
+    },
+    {
+      "epoch": 0.6447257383966245,
+      "grad_norm": 0.4225057363510132,
+      "learning_rate": 0.00019956398160737214,
+      "loss": 0.8127874732017517,
+      "step": 1528
+    },
+    {
+      "epoch": 0.6455696202531646,
+      "grad_norm": 0.4231826364994049,
+      "learning_rate": 0.00019955948990515684,
+      "loss": 0.8781921863555908,
+      "step": 1530
+    },
+    {
+      "epoch": 0.6464135021097046,
+      "grad_norm": 0.4436216950416565,
+      "learning_rate": 0.00019955497523647933,
+      "loss": 0.8136996626853943,
+      "step": 1532
+    },
+    {
+      "epoch": 0.6472573839662448,
+      "grad_norm": 0.4018244743347168,
+      "learning_rate": 0.00019955043760238098,
+      "loss": 0.8003877401351929,
+      "step": 1534
+    },
+    {
+      "epoch": 0.6481012658227848,
+      "grad_norm": 0.40024611353874207,
+      "learning_rate": 0.0001995458770039086,
+      "loss": 0.8996267914772034,
+      "step": 1536
+    },
+    {
+      "epoch": 0.6489451476793249,
+      "grad_norm": 0.42845702171325684,
+      "learning_rate": 0.00019954129344211424,
+      "loss": 0.8610522150993347,
+      "step": 1538
+    },
+    {
+      "epoch": 0.6497890295358649,
+      "grad_norm": 0.4258182644844055,
+      "learning_rate": 0.00019953668691805517,
+      "loss": 0.8991837501525879,
+      "step": 1540
+    },
+    {
+      "epoch": 0.6506329113924051,
+      "grad_norm": 0.4483257830142975,
+      "learning_rate": 0.0001995320574327941,
+      "loss": 0.8150189518928528,
+      "step": 1542
+    },
+    {
+      "epoch": 0.6514767932489451,
+      "grad_norm": 0.4320627748966217,
+      "learning_rate": 0.0001995274049873989,
+      "loss": 0.8572840690612793,
+      "step": 1544
+    },
+    {
+      "epoch": 0.6523206751054852,
+      "grad_norm": 0.39950594305992126,
+      "learning_rate": 0.00019952272958294288,
+      "loss": 0.7766129374504089,
+      "step": 1546
+    },
+    {
+      "epoch": 0.6531645569620254,
+      "grad_norm": 0.4130117893218994,
+      "learning_rate": 0.00019951803122050446,
+      "loss": 0.9032199382781982,
+      "step": 1548
+    },
+    {
+      "epoch": 0.6540084388185654,
+      "grad_norm": 0.4436919391155243,
+      "learning_rate": 0.00019951330990116754,
+      "loss": 0.8361048102378845,
+      "step": 1550
+    },
+    {
+      "epoch": 0.6548523206751055,
+      "grad_norm": 0.4229227900505066,
+      "learning_rate": 0.00019950856562602121,
+      "loss": 0.8661768436431885,
+      "step": 1552
+    },
+    {
+      "epoch": 0.6556962025316456,
+      "grad_norm": 0.4839727580547333,
+      "learning_rate": 0.00019950379839615988,
+      "loss": 0.9737826585769653,
+      "step": 1554
+    },
+    {
+      "epoch": 0.6565400843881857,
+      "grad_norm": 0.3859386742115021,
+      "learning_rate": 0.00019949900821268328,
+      "loss": 0.7542453408241272,
+      "step": 1556
+    },
+    {
+      "epoch": 0.6573839662447257,
+      "grad_norm": 0.39643988013267517,
+      "learning_rate": 0.0001994941950766964,
+      "loss": 0.9226290583610535,
+      "step": 1558
+    },
+    {
+      "epoch": 0.6582278481012658,
+      "grad_norm": 0.4261355400085449,
+      "learning_rate": 0.00019948935898930954,
+      "loss": 0.8375602960586548,
+      "step": 1560
+    },
+    {
+      "epoch": 0.6590717299578059,
+      "grad_norm": 0.445931077003479,
+      "learning_rate": 0.00019948449995163826,
+      "loss": 0.7847310304641724,
+      "step": 1562
+    },
+    {
+      "epoch": 0.659915611814346,
+      "grad_norm": 0.4680459797382355,
+      "learning_rate": 0.00019947961796480353,
+      "loss": 0.851150631904602,
+      "step": 1564
+    },
+    {
+      "epoch": 0.660759493670886,
+      "grad_norm": 0.6826074719429016,
+      "learning_rate": 0.00019947471302993143,
+      "loss": 0.8259562253952026,
+      "step": 1566
+    },
+    {
+      "epoch": 0.6616033755274262,
+      "grad_norm": 0.47369205951690674,
+      "learning_rate": 0.00019946978514815352,
+      "loss": 0.879021406173706,
+      "step": 1568
+    },
+    {
+      "epoch": 0.6624472573839663,
+      "grad_norm": 0.41111937165260315,
+      "learning_rate": 0.0001994648343206065,
+      "loss": 0.9296056032180786,
+      "step": 1570
+    },
+    {
+      "epoch": 0.6632911392405063,
+      "grad_norm": 0.4297783076763153,
+      "learning_rate": 0.00019945986054843248,
+      "loss": 0.873822808265686,
+      "step": 1572
+    },
+    {
+      "epoch": 0.6641350210970464,
+      "grad_norm": 0.48124316334724426,
+      "learning_rate": 0.00019945486383277878,
+      "loss": 0.937151312828064,
+      "step": 1574
+    },
+    {
+      "epoch": 0.6649789029535865,
+      "grad_norm": 0.4048190116882324,
+      "learning_rate": 0.00019944984417479805,
+      "loss": 0.8398929834365845,
+      "step": 1576
+    },
+    {
+      "epoch": 0.6658227848101266,
+      "grad_norm": 0.4023774266242981,
+      "learning_rate": 0.00019944480157564826,
+      "loss": 0.8269520998001099,
+      "step": 1578
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.3784184157848358,
+      "learning_rate": 0.00019943973603649262,
+      "loss": 0.8237608671188354,
+      "step": 1580
+    },
+    {
+      "epoch": 0.6675105485232068,
+      "grad_norm": 0.4448573887348175,
+      "learning_rate": 0.00019943464755849965,
+      "loss": 0.8469099998474121,
+      "step": 1582
+    },
+    {
+      "epoch": 0.6683544303797468,
+      "grad_norm": 0.4690774083137512,
+      "learning_rate": 0.00019942953614284317,
+      "loss": 0.8630822896957397,
+      "step": 1584
+    },
+    {
+      "epoch": 0.6691983122362869,
+      "grad_norm": 0.43670853972435,
+      "learning_rate": 0.00019942440179070227,
+      "loss": 0.8904082179069519,
+      "step": 1586
+    },
+    {
+      "epoch": 0.6700421940928271,
+      "grad_norm": 0.4236849844455719,
+      "learning_rate": 0.00019941924450326138,
+      "loss": 0.8355640172958374,
+      "step": 1588
+    },
+    {
+      "epoch": 0.6708860759493671,
+      "grad_norm": 0.48624539375305176,
+      "learning_rate": 0.00019941406428171013,
+      "loss": 0.8898435831069946,
+      "step": 1590
+    },
+    {
+      "epoch": 0.6717299578059072,
+      "grad_norm": 0.4453780949115753,
+      "learning_rate": 0.00019940886112724355,
+      "loss": 0.9000012278556824,
+      "step": 1592
+    },
+    {
+      "epoch": 0.6725738396624472,
+      "grad_norm": 0.4417199194431305,
+      "learning_rate": 0.00019940363504106193,
+      "loss": 0.8171058893203735,
+      "step": 1594
+    },
+    {
+      "epoch": 0.6734177215189874,
+      "grad_norm": 0.45055362582206726,
+      "learning_rate": 0.00019939838602437074,
+      "loss": 0.8703798055648804,
+      "step": 1596
+    },
+    {
+      "epoch": 0.6742616033755274,
+      "grad_norm": 0.42316532135009766,
+      "learning_rate": 0.00019939311407838088,
+      "loss": 0.8677281737327576,
+      "step": 1598
+    },
+    {
+      "epoch": 0.6751054852320675,
+      "grad_norm": 0.3935781717300415,
+      "learning_rate": 0.0001993878192043085,
+      "loss": 0.7314013242721558,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6751054852320675,
+      "eval_loss": 0.8769772052764893,
+      "eval_runtime": 677.9338,
+      "eval_samples_per_second": 3.108,
+      "eval_steps_per_second": 3.108,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6759493670886076,
+      "grad_norm": 0.4197012186050415,
+      "learning_rate": 0.000199382501403375,
+      "loss": 0.8075380325317383,
+      "step": 1602
+    },
+    {
+      "epoch": 0.6767932489451477,
+      "grad_norm": 0.45227327942848206,
+      "learning_rate": 0.00019937716067680713,
+      "loss": 0.7885794639587402,
+      "step": 1604
+    },
+    {
+      "epoch": 0.6776371308016877,
+      "grad_norm": 0.45660385489463806,
+      "learning_rate": 0.00019937179702583682,
+      "loss": 0.868261992931366,
+      "step": 1606
+    },
+    {
+      "epoch": 0.6784810126582278,
+      "grad_norm": 0.3955288231372833,
+      "learning_rate": 0.00019936641045170144,
+      "loss": 0.8306655287742615,
+      "step": 1608
+    },
+    {
+      "epoch": 0.679324894514768,
+      "grad_norm": 0.3651324510574341,
+      "learning_rate": 0.00019936100095564353,
+      "loss": 0.7505315542221069,
+      "step": 1610
+    },
+    {
+      "epoch": 0.680168776371308,
+      "grad_norm": 0.4240955412387848,
+      "learning_rate": 0.00019935556853891096,
+      "loss": 0.8792756795883179,
+      "step": 1612
+    },
+    {
+      "epoch": 0.6810126582278481,
+      "grad_norm": 0.4142507314682007,
+      "learning_rate": 0.00019935011320275687,
+      "loss": 0.7328272461891174,
+      "step": 1614
+    },
+    {
+      "epoch": 0.6818565400843882,
+      "grad_norm": 0.48339322209358215,
+      "learning_rate": 0.00019934463494843975,
+      "loss": 0.8414849042892456,
+      "step": 1616
+    },
+    {
+      "epoch": 0.6827004219409283,
+      "grad_norm": 0.4312443435192108,
+      "learning_rate": 0.00019933913377722328,
+      "loss": 0.8588716983795166,
+      "step": 1618
+    },
+    {
+      "epoch": 0.6835443037974683,
+      "grad_norm": 0.43849512934684753,
+      "learning_rate": 0.0001993336096903765,
+      "loss": 0.846511960029602,
+      "step": 1620
+    },
+    {
+      "epoch": 0.6843881856540084,
+      "grad_norm": 0.45767566561698914,
+      "learning_rate": 0.0001993280626891737,
+      "loss": 0.9352323412895203,
+      "step": 1622
+    },
+    {
+      "epoch": 0.6852320675105485,
+      "grad_norm": 0.438412606716156,
+      "learning_rate": 0.00019932249277489444,
+      "loss": 0.8884757161140442,
+      "step": 1624
+    },
+    {
+      "epoch": 0.6860759493670886,
+      "grad_norm": 0.45893922448158264,
+      "learning_rate": 0.0001993168999488237,
+      "loss": 0.7739649415016174,
+      "step": 1626
+    },
+    {
+      "epoch": 0.6869198312236287,
+      "grad_norm": 0.38631772994995117,
+      "learning_rate": 0.0001993112842122515,
+      "loss": 0.7948258519172668,
+      "step": 1628
+    },
+    {
+      "epoch": 0.6877637130801688,
+      "grad_norm": 0.3824027180671692,
+      "learning_rate": 0.00019930564556647334,
+      "loss": 0.750511884689331,
+      "step": 1630
+    },
+    {
+      "epoch": 0.6886075949367089,
+      "grad_norm": 0.3980776071548462,
+      "learning_rate": 0.00019929998401278996,
+      "loss": 0.9006738662719727,
+      "step": 1632
+    },
+    {
+      "epoch": 0.6894514767932489,
+      "grad_norm": 0.41419896483421326,
+      "learning_rate": 0.00019929429955250734,
+      "loss": 0.6996869444847107,
+      "step": 1634
+    },
+    {
+      "epoch": 0.6902953586497891,
+      "grad_norm": 0.43783676624298096,
+      "learning_rate": 0.00019928859218693682,
+      "loss": 0.8523393273353577,
+      "step": 1636
+    },
+    {
+      "epoch": 0.6911392405063291,
+      "grad_norm": 0.4378126561641693,
+      "learning_rate": 0.00019928286191739497,
+      "loss": 0.8885987401008606,
+      "step": 1638
+    },
+    {
+      "epoch": 0.6919831223628692,
+      "grad_norm": 0.4276133179664612,
+      "learning_rate": 0.00019927710874520363,
+      "loss": 0.8563809990882874,
+      "step": 1640
+    },
+    {
+      "epoch": 0.6928270042194092,
+      "grad_norm": 0.4600282311439514,
+      "learning_rate": 0.00019927133267168992,
+      "loss": 0.854114294052124,
+      "step": 1642
+    },
+    {
+      "epoch": 0.6936708860759494,
+      "grad_norm": 0.5058461427688599,
+      "learning_rate": 0.00019926553369818635,
+      "loss": 0.8557889461517334,
+      "step": 1644
+    },
+    {
+      "epoch": 0.6945147679324895,
+      "grad_norm": 0.4947412312030792,
+      "learning_rate": 0.00019925971182603057,
+      "loss": 0.8622767925262451,
+      "step": 1646
+    },
+    {
+      "epoch": 0.6953586497890295,
+      "grad_norm": 0.42555496096611023,
+      "learning_rate": 0.00019925386705656558,
+      "loss": 0.7836558818817139,
+      "step": 1648
+    },
+    {
+      "epoch": 0.6962025316455697,
+      "grad_norm": 0.47560831904411316,
+      "learning_rate": 0.00019924799939113967,
+      "loss": 0.7877846956253052,
+      "step": 1650
+    },
+    {
+      "epoch": 0.6970464135021097,
+      "grad_norm": 0.391702264547348,
+      "learning_rate": 0.0001992421088311064,
+      "loss": 0.775681734085083,
+      "step": 1652
+    },
+    {
+      "epoch": 0.6978902953586498,
+      "grad_norm": 0.463679701089859,
+      "learning_rate": 0.0001992361953778246,
+      "loss": 0.8809882402420044,
+      "step": 1654
+    },
+    {
+      "epoch": 0.6987341772151898,
+      "grad_norm": 0.44096484780311584,
+      "learning_rate": 0.00019923025903265836,
+      "loss": 0.8513299226760864,
+      "step": 1656
+    },
+    {
+      "epoch": 0.69957805907173,
+      "grad_norm": 0.47818854451179504,
+      "learning_rate": 0.0001992242997969772,
+      "loss": 0.8309667706489563,
+      "step": 1658
+    },
+    {
+      "epoch": 0.70042194092827,
+      "grad_norm": 0.4192790985107422,
+      "learning_rate": 0.00019921831767215565,
+      "loss": 0.8817558884620667,
+      "step": 1660
+    },
+    {
+      "epoch": 0.7012658227848101,
+      "grad_norm": 0.45285239815711975,
+      "learning_rate": 0.00019921231265957373,
+      "loss": 0.8315755724906921,
+      "step": 1662
+    },
+    {
+      "epoch": 0.7021097046413503,
+      "grad_norm": 0.47118625044822693,
+      "learning_rate": 0.00019920628476061673,
+      "loss": 0.8698058724403381,
+      "step": 1664
+    },
+    {
+      "epoch": 0.7029535864978903,
+      "grad_norm": 0.4769238829612732,
+      "learning_rate": 0.0001992002339766751,
+      "loss": 0.953620970249176,
+      "step": 1666
+    },
+    {
+      "epoch": 0.7037974683544304,
+      "grad_norm": 0.39397355914115906,
+      "learning_rate": 0.00019919416030914468,
+      "loss": 0.8536615371704102,
+      "step": 1668
+    },
+    {
+      "epoch": 0.7046413502109705,
+      "grad_norm": 0.3659648299217224,
+      "learning_rate": 0.00019918806375942655,
+      "loss": 0.8159828782081604,
+      "step": 1670
+    },
+    {
+      "epoch": 0.7054852320675106,
+      "grad_norm": 0.4381789565086365,
+      "learning_rate": 0.00019918194432892703,
+      "loss": 0.782463550567627,
+      "step": 1672
+    },
+    {
+      "epoch": 0.7063291139240506,
+      "grad_norm": 0.44745752215385437,
+      "learning_rate": 0.00019917580201905778,
+      "loss": 0.9314272999763489,
+      "step": 1674
+    },
+    {
+      "epoch": 0.7071729957805907,
+      "grad_norm": 0.4515164792537689,
+      "learning_rate": 0.00019916963683123574,
+      "loss": 0.860816240310669,
+      "step": 1676
+    },
+    {
+      "epoch": 0.7080168776371308,
+      "grad_norm": 0.42735886573791504,
+      "learning_rate": 0.00019916344876688303,
+      "loss": 0.8008456826210022,
+      "step": 1678
+    },
+    {
+      "epoch": 0.7088607594936709,
+      "grad_norm": 0.418275386095047,
+      "learning_rate": 0.00019915723782742718,
+      "loss": 0.7796595096588135,
+      "step": 1680
+    },
+    {
+      "epoch": 0.7097046413502109,
+      "grad_norm": 0.44347989559173584,
+      "learning_rate": 0.00019915100401430089,
+      "loss": 0.9006506204605103,
+      "step": 1682
+    },
+    {
+      "epoch": 0.7105485232067511,
+      "grad_norm": 0.3932702839374542,
+      "learning_rate": 0.00019914474732894225,
+      "loss": 0.7750146985054016,
+      "step": 1684
+    },
+    {
+      "epoch": 0.7113924050632912,
+      "grad_norm": 0.4810502231121063,
+      "learning_rate": 0.00019913846777279447,
+      "loss": 0.7508097290992737,
+      "step": 1686
+    },
+    {
+      "epoch": 0.7122362869198312,
+      "grad_norm": 0.4166778028011322,
+      "learning_rate": 0.00019913216534730622,
+      "loss": 0.7926796078681946,
+      "step": 1688
+    },
+    {
+      "epoch": 0.7130801687763713,
+      "grad_norm": 0.47900474071502686,
+      "learning_rate": 0.00019912584005393123,
+      "loss": 0.8363928198814392,
+      "step": 1690
+    },
+    {
+      "epoch": 0.7139240506329114,
+      "grad_norm": 0.435143381357193,
+      "learning_rate": 0.00019911949189412872,
+      "loss": 0.8649623990058899,
+      "step": 1692
+    },
+    {
+      "epoch": 0.7147679324894515,
+      "grad_norm": 0.4070208966732025,
+      "learning_rate": 0.00019911312086936305,
+      "loss": 0.8196499943733215,
+      "step": 1694
+    },
+    {
+      "epoch": 0.7156118143459915,
+      "grad_norm": 0.4717724621295929,
+      "learning_rate": 0.0001991067269811039,
+      "loss": 0.9157413840293884,
+      "step": 1696
+    },
+    {
+      "epoch": 0.7164556962025317,
+      "grad_norm": 0.3942880928516388,
+      "learning_rate": 0.0001991003102308262,
+      "loss": 0.7880456447601318,
+      "step": 1698
+    },
+    {
+      "epoch": 0.7172995780590717,
+      "grad_norm": 0.4037776589393616,
+      "learning_rate": 0.0001990938706200102,
+      "loss": 0.8088646531105042,
+      "step": 1700
+    },
+    {
+      "epoch": 0.7172995780590717,
+      "eval_loss": 0.8708170056343079,
+      "eval_runtime": 670.3019,
+      "eval_samples_per_second": 3.143,
+      "eval_steps_per_second": 3.143,
+      "step": 1700
+    },
+    {
+      "epoch": 0.7181434599156118,
+      "grad_norm": 0.4443536400794983,
+      "learning_rate": 0.0001990874081501414,
+      "loss": 0.8740324378013611,
+      "step": 1702
+    },
+    {
+      "epoch": 0.7189873417721518,
+      "grad_norm": 0.45089733600616455,
+      "learning_rate": 0.00019908092282271053,
+      "loss": 0.8305023908615112,
+      "step": 1704
+    },
+    {
+      "epoch": 0.719831223628692,
+      "grad_norm": 0.3903036415576935,
+      "learning_rate": 0.00019907441463921366,
+      "loss": 0.7718394994735718,
+      "step": 1706
+    },
+    {
+      "epoch": 0.7206751054852321,
+      "grad_norm": 0.42990729212760925,
+      "learning_rate": 0.00019906788360115208,
+      "loss": 0.843289315700531,
+      "step": 1708
+    },
+    {
+      "epoch": 0.7215189873417721,
+      "grad_norm": 0.4287833571434021,
+      "learning_rate": 0.0001990613297100324,
+      "loss": 0.8332282900810242,
+      "step": 1710
+    },
+    {
+      "epoch": 0.7223628691983123,
+      "grad_norm": 0.4557269513607025,
+      "learning_rate": 0.00019905475296736646,
+      "loss": 0.784180223941803,
+      "step": 1712
+    },
+    {
+      "epoch": 0.7232067510548523,
+      "grad_norm": 0.4001525640487671,
+      "learning_rate": 0.0001990481533746714,
+      "loss": 0.7807716131210327,
+      "step": 1714
+    },
+    {
+      "epoch": 0.7240506329113924,
+      "grad_norm": 0.5208709836006165,
+      "learning_rate": 0.0001990415309334696,
+      "loss": 0.7936414480209351,
+      "step": 1716
+    },
+    {
+      "epoch": 0.7248945147679325,
+      "grad_norm": 0.46469953656196594,
+      "learning_rate": 0.00019903488564528875,
+      "loss": 0.7964845895767212,
+      "step": 1718
+    },
+    {
+      "epoch": 0.7257383966244726,
+      "grad_norm": 0.405773401260376,
+      "learning_rate": 0.00019902821751166178,
+      "loss": 0.8302215337753296,
+      "step": 1720
+    },
+    {
+      "epoch": 0.7265822784810126,
+      "grad_norm": 0.3874973654747009,
+      "learning_rate": 0.00019902152653412692,
+      "loss": 0.8486512899398804,
+      "step": 1722
+    },
+    {
+      "epoch": 0.7274261603375527,
+      "grad_norm": 0.46865755319595337,
+      "learning_rate": 0.00019901481271422758,
+      "loss": 0.837841272354126,
+      "step": 1724
+    },
+    {
+      "epoch": 0.7282700421940929,
+      "grad_norm": 0.4139123558998108,
+      "learning_rate": 0.0001990080760535126,
+      "loss": 0.7640942335128784,
+      "step": 1726
+    },
+    {
+      "epoch": 0.7291139240506329,
+      "grad_norm": 0.5003024935722351,
+      "learning_rate": 0.000199001316553536,
+      "loss": 0.8236244320869446,
+      "step": 1728
+    },
+    {
+      "epoch": 0.729957805907173,
+      "grad_norm": 0.38495776057243347,
+      "learning_rate": 0.00019899453421585696,
+      "loss": 0.7792612314224243,
+      "step": 1730
+    },
+    {
+      "epoch": 0.7308016877637131,
+      "grad_norm": 0.4462733268737793,
+      "learning_rate": 0.00019898772904204014,
+      "loss": 0.7733872532844543,
+      "step": 1732
+    },
+    {
+      "epoch": 0.7316455696202532,
+      "grad_norm": 0.398248553276062,
+      "learning_rate": 0.00019898090103365532,
+      "loss": 0.8933147192001343,
+      "step": 1734
+    },
+    {
+      "epoch": 0.7324894514767932,
+      "grad_norm": 0.4231175482273102,
+      "learning_rate": 0.00019897405019227756,
+      "loss": 0.8198356628417969,
+      "step": 1736
+    },
+    {
+      "epoch": 0.7333333333333333,
+      "grad_norm": 0.43639084696769714,
+      "learning_rate": 0.0001989671765194873,
+      "loss": 0.8266391158103943,
+      "step": 1738
+    },
+    {
+      "epoch": 0.7341772151898734,
+      "grad_norm": 0.4235835671424866,
+      "learning_rate": 0.00019896028001687008,
+      "loss": 0.7839997410774231,
+      "step": 1740
+    },
+    {
+      "epoch": 0.7350210970464135,
+      "grad_norm": 0.4033111035823822,
+      "learning_rate": 0.00019895336068601687,
+      "loss": 0.7153518199920654,
+      "step": 1742
+    },
+    {
+      "epoch": 0.7358649789029535,
+      "grad_norm": 0.44281265139579773,
+      "learning_rate": 0.00019894641852852377,
+      "loss": 0.9079289436340332,
+      "step": 1744
+    },
+    {
+      "epoch": 0.7367088607594937,
+      "grad_norm": 0.4882747232913971,
+      "learning_rate": 0.0001989394535459922,
+      "loss": 0.8645254373550415,
+      "step": 1746
+    },
+    {
+      "epoch": 0.7375527426160338,
+      "grad_norm": 0.34858641028404236,
+      "learning_rate": 0.00019893246574002887,
+      "loss": 0.8134214878082275,
+      "step": 1748
+    },
+    {
+      "epoch": 0.7383966244725738,
+      "grad_norm": 0.4273303747177124,
+      "learning_rate": 0.00019892545511224574,
+      "loss": 0.8142663240432739,
+      "step": 1750
+    },
+    {
+      "epoch": 0.739240506329114,
+      "grad_norm": 0.4289957284927368,
+      "learning_rate": 0.00019891842166426004,
+      "loss": 0.7957643866539001,
+      "step": 1752
+    },
+    {
+      "epoch": 0.740084388185654,
+      "grad_norm": 0.4812224805355072,
+      "learning_rate": 0.00019891136539769416,
+      "loss": 0.9000363945960999,
+      "step": 1754
+    },
+    {
+      "epoch": 0.7409282700421941,
+      "grad_norm": 0.4078022837638855,
+      "learning_rate": 0.00019890428631417595,
+      "loss": 0.7719792723655701,
+      "step": 1756
+    },
+    {
+      "epoch": 0.7417721518987341,
+      "grad_norm": 0.40169858932495117,
+      "learning_rate": 0.00019889718441533838,
+      "loss": 0.7681847214698792,
+      "step": 1758
+    },
+    {
+      "epoch": 0.7426160337552743,
+      "grad_norm": 0.3713594377040863,
+      "learning_rate": 0.00019889005970281972,
+      "loss": 0.7536827325820923,
+      "step": 1760
+    },
+    {
+      "epoch": 0.7434599156118143,
+      "grad_norm": 0.4183627665042877,
+      "learning_rate": 0.0001988829121782635,
+      "loss": 0.9000160694122314,
+      "step": 1762
+    },
+    {
+      "epoch": 0.7443037974683544,
+      "grad_norm": 0.4241594672203064,
+      "learning_rate": 0.00019887574184331851,
+      "loss": 0.8047307133674622,
+      "step": 1764
+    },
+    {
+      "epoch": 0.7451476793248946,
+      "grad_norm": 0.4249233603477478,
+      "learning_rate": 0.00019886854869963883,
+      "loss": 0.8932583928108215,
+      "step": 1766
+    },
+    {
+      "epoch": 0.7459915611814346,
+      "grad_norm": 0.35815340280532837,
+      "learning_rate": 0.00019886133274888378,
+      "loss": 0.7684977054595947,
+      "step": 1768
+    },
+    {
+      "epoch": 0.7468354430379747,
+      "grad_norm": 0.41035401821136475,
+      "learning_rate": 0.00019885409399271795,
+      "loss": 0.8644338250160217,
+      "step": 1770
+    },
+    {
+      "epoch": 0.7476793248945147,
+      "grad_norm": 0.5450286269187927,
+      "learning_rate": 0.00019884683243281116,
+      "loss": 0.8035860657691956,
+      "step": 1772
+    },
+    {
+      "epoch": 0.7485232067510549,
+      "grad_norm": 0.5280999541282654,
+      "learning_rate": 0.0001988395480708385,
+      "loss": 0.8861207962036133,
+      "step": 1774
+    },
+    {
+      "epoch": 0.7493670886075949,
+      "grad_norm": 0.42665231227874756,
+      "learning_rate": 0.00019883224090848036,
+      "loss": 0.7506847977638245,
+      "step": 1776
+    },
+    {
+      "epoch": 0.750210970464135,
+      "grad_norm": 0.39029282331466675,
+      "learning_rate": 0.00019882491094742232,
+      "loss": 0.7662046551704407,
+      "step": 1778
+    },
+    {
+      "epoch": 0.7510548523206751,
+      "grad_norm": 0.43870019912719727,
+      "learning_rate": 0.00019881755818935534,
+      "loss": 0.7590143084526062,
+      "step": 1780
+    },
+    {
+      "epoch": 0.7518987341772152,
+      "grad_norm": 0.4195050001144409,
+      "learning_rate": 0.0001988101826359755,
+      "loss": 0.9340365529060364,
+      "step": 1782
+    },
+    {
+      "epoch": 0.7527426160337553,
+      "grad_norm": 0.4236123263835907,
+      "learning_rate": 0.0001988027842889842,
+      "loss": 0.7969209551811218,
+      "step": 1784
+    },
+    {
+      "epoch": 0.7535864978902953,
+      "grad_norm": 0.4274357259273529,
+      "learning_rate": 0.00019879536315008808,
+      "loss": 0.7892382144927979,
+      "step": 1786
+    },
+    {
+      "epoch": 0.7544303797468355,
+      "grad_norm": 0.4622916579246521,
+      "learning_rate": 0.00019878791922099912,
+      "loss": 0.8752562403678894,
+      "step": 1788
+    },
+    {
+      "epoch": 0.7552742616033755,
+      "grad_norm": 0.3648734986782074,
+      "learning_rate": 0.00019878045250343445,
+      "loss": 0.813011884689331,
+      "step": 1790
+    },
+    {
+      "epoch": 0.7561181434599156,
+      "grad_norm": 0.41361239552497864,
+      "learning_rate": 0.00019877296299911648,
+      "loss": 0.7310198545455933,
+      "step": 1792
+    },
+    {
+      "epoch": 0.7569620253164557,
+      "grad_norm": 0.4113222062587738,
+      "learning_rate": 0.00019876545070977294,
+      "loss": 0.858386754989624,
+      "step": 1794
+    },
+    {
+      "epoch": 0.7578059071729958,
+      "grad_norm": 0.4295370876789093,
+      "learning_rate": 0.00019875791563713676,
+      "loss": 0.8377325534820557,
+      "step": 1796
+    },
+    {
+      "epoch": 0.7586497890295358,
+      "grad_norm": 0.4227522015571594,
+      "learning_rate": 0.0001987503577829461,
+      "loss": 0.9139418005943298,
+      "step": 1798
+    },
+    {
+      "epoch": 0.759493670886076,
+      "grad_norm": 0.3638151288032532,
+      "learning_rate": 0.00019874277714894442,
+      "loss": 0.6994872689247131,
+      "step": 1800
+    },
+    {
+      "epoch": 0.759493670886076,
+      "eval_loss": 0.8625519275665283,
+      "eval_runtime": 686.4271,
+      "eval_samples_per_second": 3.07,
+      "eval_steps_per_second": 3.07,
+      "step": 1800
+    },
+    {
+      "epoch": 0.760337552742616,
+      "grad_norm": 0.46954411268234253,
+      "learning_rate": 0.00019873517373688047,
+      "loss": 0.8681167960166931,
+      "step": 1802
+    },
+    {
+      "epoch": 0.7611814345991561,
+      "grad_norm": 0.41958674788475037,
+      "learning_rate": 0.00019872754754850819,
+      "loss": 0.7420852184295654,
+      "step": 1804
+    },
+    {
+      "epoch": 0.7620253164556962,
+      "grad_norm": 0.431226521730423,
+      "learning_rate": 0.00019871989858558678,
+      "loss": 0.8018608093261719,
+      "step": 1806
+    },
+    {
+      "epoch": 0.7628691983122363,
+      "grad_norm": 0.45129454135894775,
+      "learning_rate": 0.0001987122268498807,
+      "loss": 0.8793904781341553,
+      "step": 1808
+    },
+    {
+      "epoch": 0.7637130801687764,
+      "grad_norm": 0.44278961420059204,
+      "learning_rate": 0.00019870453234315972,
+      "loss": 0.8416730761528015,
+      "step": 1810
+    },
+    {
+      "epoch": 0.7645569620253164,
+      "grad_norm": 0.4689098000526428,
+      "learning_rate": 0.00019869681506719876,
+      "loss": 0.8225743174552917,
+      "step": 1812
+    },
+    {
+      "epoch": 0.7654008438818566,
+      "grad_norm": 0.4413386881351471,
+      "learning_rate": 0.00019868907502377806,
+      "loss": 0.7871913313865662,
+      "step": 1814
+    },
+    {
+      "epoch": 0.7662447257383966,
+      "grad_norm": 0.48395273089408875,
+      "learning_rate": 0.00019868131221468315,
+      "loss": 0.8616237044334412,
+      "step": 1816
+    },
+    {
+      "epoch": 0.7670886075949367,
+      "grad_norm": 0.370263934135437,
+      "learning_rate": 0.0001986735266417047,
+      "loss": 0.7243452668190002,
+      "step": 1818
+    },
+    {
+      "epoch": 0.7679324894514767,
+      "grad_norm": 0.4522445797920227,
+      "learning_rate": 0.00019866571830663875,
+      "loss": 0.8623812198638916,
+      "step": 1820
+    },
+    {
+      "epoch": 0.7687763713080169,
+      "grad_norm": 0.43821004033088684,
+      "learning_rate": 0.00019865788721128643,
+      "loss": 0.776618480682373,
+      "step": 1822
+    },
+    {
+      "epoch": 0.769620253164557,
+      "grad_norm": 0.3825650215148926,
+      "learning_rate": 0.00019865003335745436,
+      "loss": 0.8722774982452393,
+      "step": 1824
+    },
+    {
+      "epoch": 0.770464135021097,
+      "grad_norm": 0.43684205412864685,
+      "learning_rate": 0.00019864215674695416,
+      "loss": 0.907802939414978,
+      "step": 1826
+    },
+    {
+      "epoch": 0.7713080168776372,
+      "grad_norm": 0.4128975570201874,
+      "learning_rate": 0.00019863425738160287,
+      "loss": 0.8505539298057556,
+      "step": 1828
+    },
+    {
+      "epoch": 0.7721518987341772,
+      "grad_norm": 0.4254065454006195,
+      "learning_rate": 0.0001986263352632227,
+      "loss": 0.83274906873703,
+      "step": 1830
+    },
+    {
+      "epoch": 0.7729957805907173,
+      "grad_norm": 0.4137375056743622,
+      "learning_rate": 0.00019861839039364118,
+      "loss": 0.8172078728675842,
+      "step": 1832
+    },
+    {
+      "epoch": 0.7738396624472574,
+      "grad_norm": 0.4387703239917755,
+      "learning_rate": 0.00019861042277469096,
+      "loss": 0.8610842227935791,
+      "step": 1834
+    },
+    {
+      "epoch": 0.7746835443037975,
+      "grad_norm": 0.398952454328537,
+      "learning_rate": 0.00019860243240821005,
+      "loss": 0.7961188554763794,
+      "step": 1836
+    },
+    {
+      "epoch": 0.7755274261603375,
+      "grad_norm": 0.4079921543598175,
+      "learning_rate": 0.0001985944192960417,
+      "loss": 0.8078321218490601,
+      "step": 1838
+    },
+    {
+      "epoch": 0.7763713080168776,
+      "grad_norm": 0.4701489508152008,
+      "learning_rate": 0.00019858638344003433,
+      "loss": 0.8743909001350403,
+      "step": 1840
+    },
+    {
+      "epoch": 0.7772151898734178,
+      "grad_norm": 0.5391269326210022,
+      "learning_rate": 0.00019857832484204173,
+      "loss": 0.7681707143783569,
+      "step": 1842
+    },
+    {
+      "epoch": 0.7780590717299578,
+      "grad_norm": 0.36250734329223633,
+      "learning_rate": 0.00019857024350392277,
+      "loss": 0.7817525863647461,
+      "step": 1844
+    },
+    {
+      "epoch": 0.7789029535864979,
+      "grad_norm": 0.41237136721611023,
+      "learning_rate": 0.00019856213942754175,
+      "loss": 0.7614551782608032,
+      "step": 1846
+    },
+    {
+      "epoch": 0.779746835443038,
+      "grad_norm": 0.3968869745731354,
+      "learning_rate": 0.00019855401261476807,
+      "loss": 0.7492313981056213,
+      "step": 1848
+    },
+    {
+      "epoch": 0.7805907172995781,
+      "grad_norm": 0.441476434469223,
+      "learning_rate": 0.00019854586306747646,
+      "loss": 0.8615695238113403,
+      "step": 1850
+    },
+    {
+      "epoch": 0.7814345991561181,
+      "grad_norm": 0.4486154317855835,
+      "learning_rate": 0.00019853769078754686,
+      "loss": 0.8416724801063538,
+      "step": 1852
+    },
+    {
+      "epoch": 0.7822784810126582,
+      "grad_norm": 0.45148879289627075,
+      "learning_rate": 0.0001985294957768644,
+      "loss": 0.8987806439399719,
+      "step": 1854
+    },
+    {
+      "epoch": 0.7831223628691983,
+      "grad_norm": 0.39330095052719116,
+      "learning_rate": 0.0001985212780373196,
+      "loss": 0.7597590088844299,
+      "step": 1856
+    },
+    {
+      "epoch": 0.7839662447257384,
+      "grad_norm": 0.4260408580303192,
+      "learning_rate": 0.00019851303757080808,
+      "loss": 0.8429927229881287,
+      "step": 1858
+    },
+    {
+      "epoch": 0.7848101265822784,
+      "grad_norm": 0.37519171833992004,
+      "learning_rate": 0.00019850477437923075,
+      "loss": 0.7443564534187317,
+      "step": 1860
+    },
+    {
+      "epoch": 0.7856540084388186,
+      "grad_norm": 0.4991084337234497,
+      "learning_rate": 0.00019849648846449383,
+      "loss": 0.934238612651825,
+      "step": 1862
+    },
+    {
+      "epoch": 0.7864978902953587,
+      "grad_norm": 0.4711335003376007,
+      "learning_rate": 0.00019848817982850867,
+      "loss": 0.8938905000686646,
+      "step": 1864
+    },
+    {
+      "epoch": 0.7873417721518987,
+      "grad_norm": 0.4145370423793793,
+      "learning_rate": 0.00019847984847319193,
+      "loss": 0.8745643496513367,
+      "step": 1866
+    },
+    {
+      "epoch": 0.7881856540084389,
+      "grad_norm": 0.3444044888019562,
+      "learning_rate": 0.00019847149440046548,
+      "loss": 0.7178786396980286,
+      "step": 1868
+    },
+    {
+      "epoch": 0.7890295358649789,
+      "grad_norm": 0.34632962942123413,
+      "learning_rate": 0.00019846311761225646,
+      "loss": 0.9114303588867188,
+      "step": 1870
+    },
+    {
+      "epoch": 0.789873417721519,
+      "grad_norm": 0.38630741834640503,
+      "learning_rate": 0.0001984547181104972,
+      "loss": 0.836281955242157,
+      "step": 1872
+    },
+    {
+      "epoch": 0.790717299578059,
+      "grad_norm": 0.40947434306144714,
+      "learning_rate": 0.00019844629589712534,
+      "loss": 0.8247858881950378,
+      "step": 1874
+    },
+    {
+      "epoch": 0.7915611814345992,
+      "grad_norm": 0.45878684520721436,
+      "learning_rate": 0.00019843785097408372,
+      "loss": 0.8446351289749146,
+      "step": 1876
+    },
+    {
+      "epoch": 0.7924050632911392,
+      "grad_norm": 0.37202852964401245,
+      "learning_rate": 0.0001984293833433204,
+      "loss": 0.8567686676979065,
+      "step": 1878
+    },
+    {
+      "epoch": 0.7932489451476793,
+      "grad_norm": 0.3839069902896881,
+      "learning_rate": 0.00019842089300678876,
+      "loss": 0.8134047985076904,
+      "step": 1880
+    },
+    {
+      "epoch": 0.7940928270042195,
+      "grad_norm": 0.4018687605857849,
+      "learning_rate": 0.00019841237996644727,
+      "loss": 0.8670110106468201,
+      "step": 1882
+    },
+    {
+      "epoch": 0.7949367088607595,
+      "grad_norm": 0.372089684009552,
+      "learning_rate": 0.0001984038442242598,
+      "loss": 0.7975355386734009,
+      "step": 1884
+    },
+    {
+      "epoch": 0.7957805907172996,
+      "grad_norm": 0.4492705166339874,
+      "learning_rate": 0.0001983952857821953,
+      "loss": 0.8755611777305603,
+      "step": 1886
+    },
+    {
+      "epoch": 0.7966244725738396,
+      "grad_norm": 0.451054185628891,
+      "learning_rate": 0.00019838670464222813,
+      "loss": 0.8680241107940674,
+      "step": 1888
+    },
+    {
+      "epoch": 0.7974683544303798,
+      "grad_norm": 0.4036900997161865,
+      "learning_rate": 0.00019837810080633773,
+      "loss": 0.7990990877151489,
+      "step": 1890
+    },
+    {
+      "epoch": 0.7983122362869198,
+      "grad_norm": 0.46125802397727966,
+      "learning_rate": 0.0001983694742765089,
+      "loss": 0.8598953485488892,
+      "step": 1892
+    },
+    {
+      "epoch": 0.7991561181434599,
+      "grad_norm": 0.4458785951137543,
+      "learning_rate": 0.00019836082505473153,
+      "loss": 0.7707474827766418,
+      "step": 1894
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.44501590728759766,
+      "learning_rate": 0.0001983521531430009,
+      "loss": 0.8069534301757812,
+      "step": 1896
+    },
+    {
+      "epoch": 0.8008438818565401,
+      "grad_norm": 0.400771826505661,
+      "learning_rate": 0.0001983434585433174,
+      "loss": 0.862334668636322,
+      "step": 1898
+    },
+    {
+      "epoch": 0.8016877637130801,
+      "grad_norm": 0.39193347096443176,
+      "learning_rate": 0.00019833474125768676,
+      "loss": 0.8125433325767517,
+      "step": 1900
+    },
+    {
+      "epoch": 0.8016877637130801,
+      "eval_loss": 0.8546335697174072,
+      "eval_runtime": 688.5301,
+      "eval_samples_per_second": 3.06,
+      "eval_steps_per_second": 3.06,
+      "step": 1900
+    },
+    {
+      "epoch": 0.8025316455696202,
+      "grad_norm": 0.441129207611084,
+      "learning_rate": 0.00019832600128811986,
+      "loss": 0.8246012330055237,
+      "step": 1902
+    },
+    {
+      "epoch": 0.8033755274261604,
+      "grad_norm": 0.4089467227458954,
+      "learning_rate": 0.00019831723863663285,
+      "loss": 0.8387641310691833,
+      "step": 1904
+    },
+    {
+      "epoch": 0.8042194092827004,
+      "grad_norm": 0.4871654510498047,
+      "learning_rate": 0.0001983084533052471,
+      "loss": 0.7891429662704468,
+      "step": 1906
+    },
+    {
+      "epoch": 0.8050632911392405,
+      "grad_norm": 0.40085604786872864,
+      "learning_rate": 0.00019829964529598923,
+      "loss": 0.8008774518966675,
+      "step": 1908
+    },
+    {
+      "epoch": 0.8059071729957806,
+      "grad_norm": 0.38775014877319336,
+      "learning_rate": 0.00019829081461089109,
+      "loss": 0.7921834588050842,
+      "step": 1910
+    },
+    {
+      "epoch": 0.8067510548523207,
+      "grad_norm": 0.37830594182014465,
+      "learning_rate": 0.0001982819612519897,
+      "loss": 0.7752519845962524,
+      "step": 1912
+    },
+    {
+      "epoch": 0.8075949367088607,
+      "grad_norm": 0.41137194633483887,
+      "learning_rate": 0.0001982730852213274,
+      "loss": 0.7958255410194397,
+      "step": 1914
+    },
+    {
+      "epoch": 0.8084388185654009,
+      "grad_norm": 0.4871830940246582,
+      "learning_rate": 0.0001982641865209517,
+      "loss": 0.8177281618118286,
+      "step": 1916
+    },
+    {
+      "epoch": 0.809282700421941,
+      "grad_norm": 0.4157388210296631,
+      "learning_rate": 0.0001982552651529154,
+      "loss": 0.7726616859436035,
+      "step": 1918
+    },
+    {
+      "epoch": 0.810126582278481,
+      "grad_norm": 0.32338014245033264,
+      "learning_rate": 0.00019824632111927645,
+      "loss": 0.6601400375366211,
+      "step": 1920
+    },
+    {
+      "epoch": 0.810970464135021,
+      "grad_norm": 0.4642949104309082,
+      "learning_rate": 0.00019823735442209804,
+      "loss": 0.7866622805595398,
+      "step": 1922
+    },
+    {
+      "epoch": 0.8118143459915612,
+      "grad_norm": 0.43558797240257263,
+      "learning_rate": 0.00019822836506344865,
+      "loss": 0.7857986092567444,
+      "step": 1924
+    },
+    {
+      "epoch": 0.8126582278481013,
+      "grad_norm": 0.3996953070163727,
+      "learning_rate": 0.000198219353045402,
+      "loss": 0.8407763838768005,
+      "step": 1926
+    },
+    {
+      "epoch": 0.8135021097046413,
+      "grad_norm": 0.4415414333343506,
+      "learning_rate": 0.00019821031837003686,
+      "loss": 0.8433752655982971,
+      "step": 1928
+    },
+    {
+      "epoch": 0.8143459915611815,
+      "grad_norm": 0.37329310178756714,
+      "learning_rate": 0.00019820126103943752,
+      "loss": 0.6941158175468445,
+      "step": 1930
+    },
+    {
+      "epoch": 0.8151898734177215,
+      "grad_norm": 0.4482601284980774,
+      "learning_rate": 0.00019819218105569323,
+      "loss": 0.775604248046875,
+      "step": 1932
+    },
+    {
+      "epoch": 0.8160337552742616,
+      "grad_norm": 0.426194429397583,
+      "learning_rate": 0.00019818307842089854,
+      "loss": 0.8496418595314026,
+      "step": 1934
+    },
+    {
+      "epoch": 0.8168776371308016,
+      "grad_norm": 0.42379093170166016,
+      "learning_rate": 0.00019817395313715335,
+      "loss": 0.7767958045005798,
+      "step": 1936
+    },
+    {
+      "epoch": 0.8177215189873418,
+      "grad_norm": 0.4149782061576843,
+      "learning_rate": 0.0001981648052065626,
+      "loss": 0.7347666025161743,
+      "step": 1938
+    },
+    {
+      "epoch": 0.8185654008438819,
+      "grad_norm": 0.3927431106567383,
+      "learning_rate": 0.0001981556346312366,
+      "loss": 0.7839647531509399,
+      "step": 1940
+    },
+    {
+      "epoch": 0.8194092827004219,
+      "grad_norm": 0.4483136236667633,
+      "learning_rate": 0.00019814644141329083,
+      "loss": 0.8870531916618347,
+      "step": 1942
+    },
+    {
+      "epoch": 0.8202531645569621,
+      "grad_norm": 0.4400147497653961,
+      "learning_rate": 0.00019813722555484594,
+      "loss": 0.8646620512008667,
+      "step": 1944
+    },
+    {
+      "epoch": 0.8210970464135021,
+      "grad_norm": 0.469971626996994,
+      "learning_rate": 0.00019812798705802785,
+      "loss": 0.9576541781425476,
+      "step": 1946
+    },
+    {
+      "epoch": 0.8219409282700422,
+      "grad_norm": 0.388113409280777,
+      "learning_rate": 0.00019811872592496776,
+      "loss": 0.7956477403640747,
+      "step": 1948
+    },
+    {
+      "epoch": 0.8227848101265823,
+      "grad_norm": 0.4271424114704132,
+      "learning_rate": 0.00019810944215780201,
+      "loss": 0.8440352082252502,
+      "step": 1950
+    },
+    {
+      "epoch": 0.8236286919831224,
+      "grad_norm": 0.4097966253757477,
+      "learning_rate": 0.00019810013575867217,
+      "loss": 0.8589065074920654,
+      "step": 1952
+    },
+    {
+      "epoch": 0.8244725738396624,
+      "grad_norm": 0.41550710797309875,
+      "learning_rate": 0.00019809080672972504,
+      "loss": 0.8606626987457275,
+      "step": 1954
+    },
+    {
+      "epoch": 0.8253164556962025,
+      "grad_norm": 0.3680916130542755,
+      "learning_rate": 0.0001980814550731127,
+      "loss": 0.7582436800003052,
+      "step": 1956
+    },
+    {
+      "epoch": 0.8261603375527427,
+      "grad_norm": 0.573297917842865,
+      "learning_rate": 0.00019807208079099234,
+      "loss": 0.8345186710357666,
+      "step": 1958
+    },
+    {
+      "epoch": 0.8270042194092827,
+      "grad_norm": 0.43631303310394287,
+      "learning_rate": 0.00019806268388552646,
+      "loss": 0.889683723449707,
+      "step": 1960
+    },
+    {
+      "epoch": 0.8278481012658228,
+      "grad_norm": 0.442531019449234,
+      "learning_rate": 0.00019805326435888275,
+      "loss": 0.8092973232269287,
+      "step": 1962
+    },
+    {
+      "epoch": 0.8286919831223629,
+      "grad_norm": 0.45020824670791626,
+      "learning_rate": 0.0001980438222132341,
+      "loss": 0.8919535279273987,
+      "step": 1964
+    },
+    {
+      "epoch": 0.829535864978903,
+      "grad_norm": 0.43131786584854126,
+      "learning_rate": 0.0001980343574507586,
+      "loss": 0.7652381658554077,
+      "step": 1966
+    },
+    {
+      "epoch": 0.830379746835443,
+      "grad_norm": 0.4461326599121094,
+      "learning_rate": 0.00019802487007363967,
+      "loss": 0.8512389659881592,
+      "step": 1968
+    },
+    {
+      "epoch": 0.8312236286919831,
+      "grad_norm": 0.4676894247531891,
+      "learning_rate": 0.0001980153600840658,
+      "loss": 0.8681327700614929,
+      "step": 1970
+    },
+    {
+      "epoch": 0.8320675105485232,
+      "grad_norm": 0.38376888632774353,
+      "learning_rate": 0.00019800582748423075,
+      "loss": 0.7476955652236938,
+      "step": 1972
+    },
+    {
+      "epoch": 0.8329113924050633,
+      "grad_norm": 0.37173229455947876,
+      "learning_rate": 0.00019799627227633354,
+      "loss": 0.814192533493042,
+      "step": 1974
+    },
+    {
+      "epoch": 0.8337552742616033,
+      "grad_norm": 0.4345237910747528,
+      "learning_rate": 0.00019798669446257844,
+      "loss": 0.7668994069099426,
+      "step": 1976
+    },
+    {
+      "epoch": 0.8345991561181435,
+      "grad_norm": 0.45872583985328674,
+      "learning_rate": 0.0001979770940451747,
+      "loss": 0.8321775197982788,
+      "step": 1978
+    },
+    {
+      "epoch": 0.8354430379746836,
+      "grad_norm": 0.40618205070495605,
+      "learning_rate": 0.0001979674710263371,
+      "loss": 0.7824342250823975,
+      "step": 1980
+    },
+    {
+      "epoch": 0.8362869198312236,
+      "grad_norm": 0.38377392292022705,
+      "learning_rate": 0.00019795782540828544,
+      "loss": 0.8752480745315552,
+      "step": 1982
+    },
+    {
+      "epoch": 0.8371308016877637,
+      "grad_norm": 0.36340072751045227,
+      "learning_rate": 0.00019794815719324476,
+      "loss": 0.66700679063797,
+      "step": 1984
+    },
+    {
+      "epoch": 0.8379746835443038,
+      "grad_norm": 0.4638384282588959,
+      "learning_rate": 0.00019793846638344536,
+      "loss": 0.8582209944725037,
+      "step": 1986
+    },
+    {
+      "epoch": 0.8388185654008439,
+      "grad_norm": 0.38364699482917786,
+      "learning_rate": 0.00019792875298112268,
+      "loss": 0.7724968791007996,
+      "step": 1988
+    },
+    {
+      "epoch": 0.8396624472573839,
+      "grad_norm": 0.439730703830719,
+      "learning_rate": 0.00019791901698851749,
+      "loss": 0.8542404174804688,
+      "step": 1990
+    },
+    {
+      "epoch": 0.8405063291139241,
+      "grad_norm": 0.4622722566127777,
+      "learning_rate": 0.00019790925840787563,
+      "loss": 0.8424296975135803,
+      "step": 1992
+    },
+    {
+      "epoch": 0.8413502109704641,
+      "grad_norm": 0.4491989314556122,
+      "learning_rate": 0.0001978994772414482,
+      "loss": 0.8400710225105286,
+      "step": 1994
+    },
+    {
+      "epoch": 0.8421940928270042,
+      "grad_norm": 0.372295081615448,
+      "learning_rate": 0.0001978896734914916,
+      "loss": 0.7265452146530151,
+      "step": 1996
+    },
+    {
+      "epoch": 0.8430379746835444,
+      "grad_norm": 0.4328666925430298,
+      "learning_rate": 0.00019787984716026732,
+      "loss": 0.8458228707313538,
+      "step": 1998
+    },
+    {
+      "epoch": 0.8438818565400844,
+      "grad_norm": 0.42246317863464355,
+      "learning_rate": 0.00019786999825004216,
+      "loss": 0.8517491817474365,
+      "step": 2000
+    },
+    {
+      "epoch": 0.8438818565400844,
+      "eval_loss": 0.8460908532142639,
+      "eval_runtime": 685.2518,
+      "eval_samples_per_second": 3.075,
+      "eval_steps_per_second": 3.075,
+      "step": 2000
+    },
+    {
+      "epoch": 0.8447257383966245,
+      "grad_norm": 0.3927323818206787,
+      "learning_rate": 0.00019786012676308802,
+      "loss": 0.8045481443405151,
+      "step": 2002
+    },
+    {
+      "epoch": 0.8455696202531645,
+      "grad_norm": 0.38189369440078735,
+      "learning_rate": 0.00019785023270168202,
+      "loss": 0.7509838938713074,
+      "step": 2004
+    },
+    {
+      "epoch": 0.8464135021097047,
+      "grad_norm": 0.40383535623550415,
+      "learning_rate": 0.00019784031606810663,
+      "loss": 0.7730163335800171,
+      "step": 2006
+    },
+    {
+      "epoch": 0.8472573839662447,
+      "grad_norm": 0.3665419816970825,
+      "learning_rate": 0.00019783037686464935,
+      "loss": 0.7096828818321228,
+      "step": 2008
+    },
+    {
+      "epoch": 0.8481012658227848,
+      "grad_norm": 0.36623379588127136,
+      "learning_rate": 0.00019782041509360302,
+      "loss": 0.8186401128768921,
+      "step": 2010
+    },
+    {
+      "epoch": 0.8489451476793249,
+      "grad_norm": 0.477255642414093,
+      "learning_rate": 0.0001978104307572656,
+      "loss": 0.8502879738807678,
+      "step": 2012
+    },
+    {
+      "epoch": 0.849789029535865,
+      "grad_norm": 0.44456636905670166,
+      "learning_rate": 0.00019780042385794032,
+      "loss": 0.9010884761810303,
+      "step": 2014
+    },
+    {
+      "epoch": 0.850632911392405,
+      "grad_norm": 0.42378175258636475,
+      "learning_rate": 0.00019779039439793552,
+      "loss": 0.774183452129364,
+      "step": 2016
+    },
+    {
+      "epoch": 0.8514767932489451,
+      "grad_norm": 0.43021997809410095,
+      "learning_rate": 0.00019778034237956482,
+      "loss": 0.8487694263458252,
+      "step": 2018
+    },
+    {
+      "epoch": 0.8523206751054853,
+      "grad_norm": 0.38188812136650085,
+      "learning_rate": 0.00019777026780514704,
+      "loss": 0.8004868030548096,
+      "step": 2020
+    },
+    {
+      "epoch": 0.8531645569620253,
+      "grad_norm": 0.3684130609035492,
+      "learning_rate": 0.0001977601706770062,
+      "loss": 0.7680953145027161,
+      "step": 2022
+    },
+    {
+      "epoch": 0.8540084388185654,
+      "grad_norm": 0.444844514131546,
+      "learning_rate": 0.00019775005099747153,
+      "loss": 0.8990264534950256,
+      "step": 2024
+    },
+    {
+      "epoch": 0.8548523206751055,
+      "grad_norm": 0.3432365655899048,
+      "learning_rate": 0.00019773990876887737,
+      "loss": 0.7676048874855042,
+      "step": 2026
+    },
+    {
+      "epoch": 0.8556962025316456,
+      "grad_norm": 0.3728618025779724,
+      "learning_rate": 0.00019772974399356343,
+      "loss": 0.8546769618988037,
+      "step": 2028
+    },
+    {
+      "epoch": 0.8565400843881856,
+      "grad_norm": 0.42767980694770813,
+      "learning_rate": 0.00019771955667387449,
+      "loss": 0.8009977340698242,
+      "step": 2030
+    },
+    {
+      "epoch": 0.8573839662447258,
+      "grad_norm": 0.4125710129737854,
+      "learning_rate": 0.00019770934681216055,
+      "loss": 0.8424299955368042,
+      "step": 2032
+    },
+    {
+      "epoch": 0.8582278481012658,
+      "grad_norm": 0.4399266839027405,
+      "learning_rate": 0.00019769911441077685,
+      "loss": 0.7940502762794495,
+      "step": 2034
+    },
+    {
+      "epoch": 0.8590717299578059,
+      "grad_norm": 0.41797709465026855,
+      "learning_rate": 0.0001976888594720838,
+      "loss": 0.9424314498901367,
+      "step": 2036
+    },
+    {
+      "epoch": 0.859915611814346,
+      "grad_norm": 0.5245009660720825,
+      "learning_rate": 0.00019767858199844698,
+      "loss": 0.9102896451950073,
+      "step": 2038
+    },
+    {
+      "epoch": 0.8607594936708861,
+      "grad_norm": 0.4092593193054199,
+      "learning_rate": 0.00019766828199223728,
+      "loss": 0.7434720396995544,
+      "step": 2040
+    },
+    {
+      "epoch": 0.8616033755274262,
+      "grad_norm": 0.40787473320961,
+      "learning_rate": 0.00019765795945583068,
+      "loss": 0.7441987991333008,
+      "step": 2042
+    },
+    {
+      "epoch": 0.8624472573839662,
+      "grad_norm": 0.42438021302223206,
+      "learning_rate": 0.0001976476143916084,
+      "loss": 0.8317872881889343,
+      "step": 2044
+    },
+    {
+      "epoch": 0.8632911392405064,
+      "grad_norm": 0.38261914253234863,
+      "learning_rate": 0.00019763724680195682,
+      "loss": 0.764509379863739,
+      "step": 2046
+    },
+    {
+      "epoch": 0.8641350210970464,
+      "grad_norm": 0.43295741081237793,
+      "learning_rate": 0.00019762685668926758,
+      "loss": 0.8666167855262756,
+      "step": 2048
+    },
+    {
+      "epoch": 0.8649789029535865,
+      "grad_norm": 0.36384403705596924,
+      "learning_rate": 0.00019761644405593743,
+      "loss": 0.7606754302978516,
+      "step": 2050
+    },
+    {
+      "epoch": 0.8658227848101265,
+      "grad_norm": 0.4296814501285553,
+      "learning_rate": 0.0001976060089043684,
+      "loss": 0.827495813369751,
+      "step": 2052
+    },
+    {
+      "epoch": 0.8666666666666667,
+      "grad_norm": 0.3831455111503601,
+      "learning_rate": 0.0001975955512369677,
+      "loss": 0.7418577075004578,
+      "step": 2054
+    },
+    {
+      "epoch": 0.8675105485232067,
+      "grad_norm": 0.44710344076156616,
+      "learning_rate": 0.0001975850710561477,
+      "loss": 0.8603045344352722,
+      "step": 2056
+    },
+    {
+      "epoch": 0.8683544303797468,
+      "grad_norm": 0.44012558460235596,
+      "learning_rate": 0.00019757456836432593,
+      "loss": 0.900658905506134,
+      "step": 2058
+    },
+    {
+      "epoch": 0.869198312236287,
+      "grad_norm": 0.42166298627853394,
+      "learning_rate": 0.0001975640431639252,
+      "loss": 0.8416497707366943,
+      "step": 2060
+    },
+    {
+      "epoch": 0.870042194092827,
+      "grad_norm": 0.40824490785598755,
+      "learning_rate": 0.00019755349545737345,
+      "loss": 0.8281066417694092,
+      "step": 2062
+    },
+    {
+      "epoch": 0.8708860759493671,
+      "grad_norm": 0.4272223711013794,
+      "learning_rate": 0.00019754292524710388,
+      "loss": 0.83869469165802,
+      "step": 2064
+    },
+    {
+      "epoch": 0.8717299578059071,
+      "grad_norm": 0.3732798397541046,
+      "learning_rate": 0.00019753233253555478,
+      "loss": 0.8337644934654236,
+      "step": 2066
+    },
+    {
+      "epoch": 0.8725738396624473,
+      "grad_norm": 0.4048069417476654,
+      "learning_rate": 0.00019752171732516973,
+      "loss": 0.7677452564239502,
+      "step": 2068
+    },
+    {
+      "epoch": 0.8734177215189873,
+      "grad_norm": 0.43628430366516113,
+      "learning_rate": 0.00019751107961839742,
+      "loss": 0.8320774435997009,
+      "step": 2070
+    },
+    {
+      "epoch": 0.8742616033755274,
+      "grad_norm": 0.43674105405807495,
+      "learning_rate": 0.00019750041941769174,
+      "loss": 0.8480262160301208,
+      "step": 2072
+    },
+    {
+      "epoch": 0.8751054852320675,
+      "grad_norm": 0.3678584098815918,
+      "learning_rate": 0.0001974897367255119,
+      "loss": 0.7598503828048706,
+      "step": 2074
+    },
+    {
+      "epoch": 0.8759493670886076,
+      "grad_norm": 0.4316411316394806,
+      "learning_rate": 0.0001974790315443221,
+      "loss": 0.7907084822654724,
+      "step": 2076
+    },
+    {
+      "epoch": 0.8767932489451477,
+      "grad_norm": 0.426465779542923,
+      "learning_rate": 0.00019746830387659182,
+      "loss": 0.8287727236747742,
+      "step": 2078
+    },
+    {
+      "epoch": 0.8776371308016878,
+      "grad_norm": 0.447337806224823,
+      "learning_rate": 0.00019745755372479578,
+      "loss": 0.8464577794075012,
+      "step": 2080
+    },
+    {
+      "epoch": 0.8784810126582279,
+      "grad_norm": 0.37877580523490906,
+      "learning_rate": 0.0001974467810914138,
+      "loss": 0.7383086681365967,
+      "step": 2082
+    },
+    {
+      "epoch": 0.8793248945147679,
+      "grad_norm": 0.3914664089679718,
+      "learning_rate": 0.0001974359859789309,
+      "loss": 0.729221522808075,
+      "step": 2084
+    },
+    {
+      "epoch": 0.880168776371308,
+      "grad_norm": 0.39767366647720337,
+      "learning_rate": 0.00019742516838983733,
+      "loss": 0.692332923412323,
+      "step": 2086
+    },
+    {
+      "epoch": 0.8810126582278481,
+      "grad_norm": 0.4396754205226898,
+      "learning_rate": 0.0001974143283266285,
+      "loss": 0.8540068864822388,
+      "step": 2088
+    },
+    {
+      "epoch": 0.8818565400843882,
+      "grad_norm": 0.4968172609806061,
+      "learning_rate": 0.00019740346579180502,
+      "loss": 0.8570998907089233,
+      "step": 2090
+    },
+    {
+      "epoch": 0.8827004219409282,
+      "grad_norm": 0.41686809062957764,
+      "learning_rate": 0.00019739258078787264,
+      "loss": 0.784139096736908,
+      "step": 2092
+    },
+    {
+      "epoch": 0.8835443037974684,
+      "grad_norm": 0.38788047432899475,
+      "learning_rate": 0.0001973816733173423,
+      "loss": 0.8521397113800049,
+      "step": 2094
+    },
+    {
+      "epoch": 0.8843881856540085,
+      "grad_norm": 0.38369178771972656,
+      "learning_rate": 0.00019737074338273023,
+      "loss": 0.8457935452461243,
+      "step": 2096
+    },
+    {
+      "epoch": 0.8852320675105485,
+      "grad_norm": 0.38869398832321167,
+      "learning_rate": 0.0001973597909865577,
+      "loss": 0.7425959706306458,
+      "step": 2098
+    },
+    {
+      "epoch": 0.8860759493670886,
+      "grad_norm": 0.4067917466163635,
+      "learning_rate": 0.00019734881613135122,
+      "loss": 0.7742393016815186,
+      "step": 2100
+    },
+    {
+      "epoch": 0.8860759493670886,
+      "eval_loss": 0.8401098847389221,
+      "eval_runtime": 669.1149,
+      "eval_samples_per_second": 3.149,
+      "eval_steps_per_second": 3.149,
+      "step": 2100
+    },
+    {
+      "epoch": 0.8869198312236287,
+      "grad_norm": 0.4282098412513733,
+      "learning_rate": 0.00019733781881964242,
+      "loss": 0.9099814295768738,
+      "step": 2102
+    },
+    {
+      "epoch": 0.8877637130801688,
+      "grad_norm": 0.33882173895835876,
+      "learning_rate": 0.00019732679905396826,
+      "loss": 0.7458856105804443,
+      "step": 2104
+    },
+    {
+      "epoch": 0.8886075949367088,
+      "grad_norm": 0.43959516286849976,
+      "learning_rate": 0.00019731575683687078,
+      "loss": 0.7829693555831909,
+      "step": 2106
+    },
+    {
+      "epoch": 0.889451476793249,
+      "grad_norm": 0.40046483278274536,
+      "learning_rate": 0.00019730469217089716,
+      "loss": 0.7898997068405151,
+      "step": 2108
+    },
+    {
+      "epoch": 0.890295358649789,
+      "grad_norm": 0.40572553873062134,
+      "learning_rate": 0.00019729360505859984,
+      "loss": 0.7874186635017395,
+      "step": 2110
+    },
+    {
+      "epoch": 0.8911392405063291,
+      "grad_norm": 0.42086347937583923,
+      "learning_rate": 0.0001972824955025364,
+      "loss": 0.8499898910522461,
+      "step": 2112
+    },
+    {
+      "epoch": 0.8919831223628693,
+      "grad_norm": 0.38845229148864746,
+      "learning_rate": 0.00019727136350526956,
+      "loss": 0.7940409779548645,
+      "step": 2114
+    },
+    {
+      "epoch": 0.8928270042194093,
+      "grad_norm": 0.41575267910957336,
+      "learning_rate": 0.00019726020906936728,
+      "loss": 0.8279082775115967,
+      "step": 2116
+    },
+    {
+      "epoch": 0.8936708860759494,
+      "grad_norm": 0.37615057826042175,
+      "learning_rate": 0.0001972490321974027,
+      "loss": 0.8217757940292358,
+      "step": 2118
+    },
+    {
+      "epoch": 0.8945147679324894,
+      "grad_norm": 0.9318385124206543,
+      "learning_rate": 0.00019723783289195415,
+      "loss": 0.7637750506401062,
+      "step": 2120
+    },
+    {
+      "epoch": 0.8953586497890296,
+      "grad_norm": 0.41948962211608887,
+      "learning_rate": 0.000197226611155605,
+      "loss": 0.8702366948127747,
+      "step": 2122
+    },
+    {
+      "epoch": 0.8962025316455696,
+      "grad_norm": 0.4718417525291443,
+      "learning_rate": 0.00019721536699094392,
+      "loss": 0.8146984577178955,
+      "step": 2124
+    },
+    {
+      "epoch": 0.8970464135021097,
+      "grad_norm": 0.4470657408237457,
+      "learning_rate": 0.00019720410040056474,
+      "loss": 0.8646075129508972,
+      "step": 2126
+    },
+    {
+      "epoch": 0.8978902953586498,
+      "grad_norm": 0.4176540970802307,
+      "learning_rate": 0.00019719281138706643,
+      "loss": 0.7925992012023926,
+      "step": 2128
+    },
+    {
+      "epoch": 0.8987341772151899,
+      "grad_norm": 0.4854719042778015,
+      "learning_rate": 0.00019718149995305315,
+      "loss": 0.8142143487930298,
+      "step": 2130
+    },
+    {
+      "epoch": 0.8995780590717299,
+      "grad_norm": 0.4270440340042114,
+      "learning_rate": 0.00019717016610113427,
+      "loss": 0.8050890564918518,
+      "step": 2132
+    },
+    {
+      "epoch": 0.90042194092827,
+      "grad_norm": 0.37710970640182495,
+      "learning_rate": 0.00019715880983392422,
+      "loss": 0.7888354659080505,
+      "step": 2134
+    },
+    {
+      "epoch": 0.9012658227848102,
+      "grad_norm": 0.4181252717971802,
+      "learning_rate": 0.00019714743115404273,
+      "loss": 0.7348225712776184,
+      "step": 2136
+    },
+    {
+      "epoch": 0.9021097046413502,
+      "grad_norm": 0.4475359618663788,
+      "learning_rate": 0.0001971360300641146,
+      "loss": 0.7729623913764954,
+      "step": 2138
+    },
+    {
+      "epoch": 0.9029535864978903,
+      "grad_norm": 0.4052332937717438,
+      "learning_rate": 0.00019712460656676992,
+      "loss": 0.8066384792327881,
+      "step": 2140
+    },
+    {
+      "epoch": 0.9037974683544304,
+      "grad_norm": 0.4567393362522125,
+      "learning_rate": 0.0001971131606646438,
+      "loss": 0.8053154349327087,
+      "step": 2142
+    },
+    {
+      "epoch": 0.9046413502109705,
+      "grad_norm": 0.4304409325122833,
+      "learning_rate": 0.00019710169236037656,
+      "loss": 0.7242560386657715,
+      "step": 2144
+    },
+    {
+      "epoch": 0.9054852320675105,
+      "grad_norm": 0.4785008132457733,
+      "learning_rate": 0.0001970902016566138,
+      "loss": 0.8475301265716553,
+      "step": 2146
+    },
+    {
+      "epoch": 0.9063291139240506,
+      "grad_norm": 0.45226508378982544,
+      "learning_rate": 0.00019707868855600618,
+      "loss": 0.928544282913208,
+      "step": 2148
+    },
+    {
+      "epoch": 0.9071729957805907,
+      "grad_norm": 0.3794468641281128,
+      "learning_rate": 0.00019706715306120956,
+      "loss": 0.8135722875595093,
+      "step": 2150
+    },
+    {
+      "epoch": 0.9080168776371308,
+      "grad_norm": 0.4005141258239746,
+      "learning_rate": 0.0001970555951748849,
+      "loss": 0.7808290719985962,
+      "step": 2152
+    },
+    {
+      "epoch": 0.9088607594936708,
+      "grad_norm": 0.37929919362068176,
+      "learning_rate": 0.00019704401489969842,
+      "loss": 0.8292917013168335,
+      "step": 2154
+    },
+    {
+      "epoch": 0.909704641350211,
+      "grad_norm": 0.4458678066730499,
+      "learning_rate": 0.0001970324122383215,
+      "loss": 0.772884726524353,
+      "step": 2156
+    },
+    {
+      "epoch": 0.9105485232067511,
+      "grad_norm": 0.41907885670661926,
+      "learning_rate": 0.0001970207871934306,
+      "loss": 0.8204470872879028,
+      "step": 2158
+    },
+    {
+      "epoch": 0.9113924050632911,
+      "grad_norm": 0.40610212087631226,
+      "learning_rate": 0.00019700913976770741,
+      "loss": 0.8361451625823975,
+      "step": 2160
+    },
+    {
+      "epoch": 0.9122362869198313,
+      "grad_norm": 0.4213273227214813,
+      "learning_rate": 0.00019699746996383878,
+      "loss": 0.8456075191497803,
+      "step": 2162
+    },
+    {
+      "epoch": 0.9130801687763713,
+      "grad_norm": 0.49264612793922424,
+      "learning_rate": 0.0001969857777845167,
+      "loss": 0.8731603622436523,
+      "step": 2164
+    },
+    {
+      "epoch": 0.9139240506329114,
+      "grad_norm": 0.42002132534980774,
+      "learning_rate": 0.00019697406323243834,
+      "loss": 0.8149666786193848,
+      "step": 2166
+    },
+    {
+      "epoch": 0.9147679324894514,
+      "grad_norm": 0.37501421570777893,
+      "learning_rate": 0.000196962326310306,
+      "loss": 0.7508546710014343,
+      "step": 2168
+    },
+    {
+      "epoch": 0.9156118143459916,
+      "grad_norm": 0.3706159293651581,
+      "learning_rate": 0.00019695056702082717,
+      "loss": 0.830280601978302,
+      "step": 2170
+    },
+    {
+      "epoch": 0.9164556962025316,
+      "grad_norm": 0.3873467445373535,
+      "learning_rate": 0.00019693878536671452,
+      "loss": 0.7297167778015137,
+      "step": 2172
+    },
+    {
+      "epoch": 0.9172995780590717,
+      "grad_norm": 0.42876651883125305,
+      "learning_rate": 0.00019692698135068582,
+      "loss": 0.7924292087554932,
+      "step": 2174
+    },
+    {
+      "epoch": 0.9181434599156119,
+      "grad_norm": 0.4259159564971924,
+      "learning_rate": 0.00019691515497546406,
+      "loss": 0.8098247051239014,
+      "step": 2176
+    },
+    {
+      "epoch": 0.9189873417721519,
+      "grad_norm": 0.38152721524238586,
+      "learning_rate": 0.00019690330624377728,
+      "loss": 0.6789063215255737,
+      "step": 2178
+    },
+    {
+      "epoch": 0.919831223628692,
+      "grad_norm": 0.4135437607765198,
+      "learning_rate": 0.00019689143515835887,
+      "loss": 0.8582620620727539,
+      "step": 2180
+    },
+    {
+      "epoch": 0.920675105485232,
+      "grad_norm": 0.3818798065185547,
+      "learning_rate": 0.0001968795417219472,
+      "loss": 0.8059743642807007,
+      "step": 2182
+    },
+    {
+      "epoch": 0.9215189873417722,
+      "grad_norm": 0.39021286368370056,
+      "learning_rate": 0.00019686762593728582,
+      "loss": 0.7432259917259216,
+      "step": 2184
+    },
+    {
+      "epoch": 0.9223628691983122,
+      "grad_norm": 0.37864068150520325,
+      "learning_rate": 0.00019685568780712355,
+      "loss": 0.7561086416244507,
+      "step": 2186
+    },
+    {
+      "epoch": 0.9232067510548523,
+      "grad_norm": 0.450847864151001,
+      "learning_rate": 0.00019684372733421425,
+      "loss": 0.7552722692489624,
+      "step": 2188
+    },
+    {
+      "epoch": 0.9240506329113924,
+      "grad_norm": 0.37909507751464844,
+      "learning_rate": 0.00019683174452131695,
+      "loss": 0.6905336380004883,
+      "step": 2190
+    },
+    {
+      "epoch": 0.9248945147679325,
+      "grad_norm": 0.4439173638820648,
+      "learning_rate": 0.0001968197393711959,
+      "loss": 0.7710593938827515,
+      "step": 2192
+    },
+    {
+      "epoch": 0.9257383966244725,
+      "grad_norm": 0.3908376395702362,
+      "learning_rate": 0.00019680771188662044,
+      "loss": 0.8212814331054688,
+      "step": 2194
+    },
+    {
+      "epoch": 0.9265822784810127,
+      "grad_norm": 0.3528299033641815,
+      "learning_rate": 0.00019679566207036507,
+      "loss": 0.751355767250061,
+      "step": 2196
+    },
+    {
+      "epoch": 0.9274261603375528,
+      "grad_norm": 0.4351397156715393,
+      "learning_rate": 0.00019678358992520945,
+      "loss": 0.8204057812690735,
+      "step": 2198
+    },
+    {
+      "epoch": 0.9282700421940928,
+      "grad_norm": 0.45566943287849426,
+      "learning_rate": 0.0001967714954539384,
+      "loss": 0.763841450214386,
+      "step": 2200
+    },
+    {
+      "epoch": 0.9282700421940928,
+      "eval_loss": 0.8336610198020935,
+      "eval_runtime": 674.5134,
+      "eval_samples_per_second": 3.124,
+      "eval_steps_per_second": 3.124,
+      "step": 2200
+    },
+    {
+      "epoch": 0.9291139240506329,
+      "grad_norm": 0.4516221284866333,
+      "learning_rate": 0.00019675937865934188,
+      "loss": 0.861711323261261,
+      "step": 2202
+    },
+    {
+      "epoch": 0.929957805907173,
+      "grad_norm": 0.43421855568885803,
+      "learning_rate": 0.00019674723954421502,
+      "loss": 0.8062718510627747,
+      "step": 2204
+    },
+    {
+      "epoch": 0.9308016877637131,
+      "grad_norm": 0.42442476749420166,
+      "learning_rate": 0.00019673507811135803,
+      "loss": 0.8069695234298706,
+      "step": 2206
+    },
+    {
+      "epoch": 0.9316455696202531,
+      "grad_norm": 0.3976045250892639,
+      "learning_rate": 0.00019672289436357637,
+      "loss": 0.7336108684539795,
+      "step": 2208
+    },
+    {
+      "epoch": 0.9324894514767933,
+      "grad_norm": 0.4131270945072174,
+      "learning_rate": 0.00019671068830368058,
+      "loss": 0.7559890151023865,
+      "step": 2210
+    },
+    {
+      "epoch": 0.9333333333333333,
+      "grad_norm": 0.4599294364452362,
+      "learning_rate": 0.00019669845993448634,
+      "loss": 0.8447075486183167,
+      "step": 2212
+    },
+    {
+      "epoch": 0.9341772151898734,
+      "grad_norm": 0.4291204810142517,
+      "learning_rate": 0.00019668620925881455,
+      "loss": 0.8329233527183533,
+      "step": 2214
+    },
+    {
+      "epoch": 0.9350210970464135,
+      "grad_norm": 0.4226539731025696,
+      "learning_rate": 0.00019667393627949116,
+      "loss": 0.749232292175293,
+      "step": 2216
+    },
+    {
+      "epoch": 0.9358649789029536,
+      "grad_norm": 0.4159580171108246,
+      "learning_rate": 0.00019666164099934732,
+      "loss": 0.7470011115074158,
+      "step": 2218
+    },
+    {
+      "epoch": 0.9367088607594937,
+      "grad_norm": 0.4160766303539276,
+      "learning_rate": 0.00019664932342121935,
+      "loss": 0.8094202280044556,
+      "step": 2220
+    },
+    {
+      "epoch": 0.9375527426160337,
+      "grad_norm": 0.45284298062324524,
+      "learning_rate": 0.00019663698354794863,
+      "loss": 0.8398169875144958,
+      "step": 2222
+    },
+    {
+      "epoch": 0.9383966244725739,
+      "grad_norm": 0.3960540294647217,
+      "learning_rate": 0.00019662462138238178,
+      "loss": 0.7652443051338196,
+      "step": 2224
+    },
+    {
+      "epoch": 0.9392405063291139,
+      "grad_norm": 0.4269275367259979,
+      "learning_rate": 0.00019661223692737047,
+      "loss": 0.7909489870071411,
+      "step": 2226
+    },
+    {
+      "epoch": 0.940084388185654,
+      "grad_norm": 0.38187873363494873,
+      "learning_rate": 0.00019659983018577157,
+      "loss": 0.7821935415267944,
+      "step": 2228
+    },
+    {
+      "epoch": 0.9409282700421941,
+      "grad_norm": 0.45898866653442383,
+      "learning_rate": 0.0001965874011604471,
+      "loss": 0.7792721390724182,
+      "step": 2230
+    },
+    {
+      "epoch": 0.9417721518987342,
+      "grad_norm": 0.4740699827671051,
+      "learning_rate": 0.00019657494985426418,
+      "loss": 0.885725200176239,
+      "step": 2232
+    },
+    {
+      "epoch": 0.9426160337552743,
+      "grad_norm": 0.3912813365459442,
+      "learning_rate": 0.00019656247627009506,
+      "loss": 0.7898101806640625,
+      "step": 2234
+    },
+    {
+      "epoch": 0.9434599156118143,
+      "grad_norm": 0.4552418291568756,
+      "learning_rate": 0.0001965499804108172,
+      "loss": 0.8646330833435059,
+      "step": 2236
+    },
+    {
+      "epoch": 0.9443037974683545,
+      "grad_norm": 0.4178548753261566,
+      "learning_rate": 0.0001965374622793131,
+      "loss": 0.8068296313285828,
+      "step": 2238
+    },
+    {
+      "epoch": 0.9451476793248945,
+      "grad_norm": 0.38649076223373413,
+      "learning_rate": 0.00019652492187847057,
+      "loss": 0.7953271865844727,
+      "step": 2240
+    },
+    {
+      "epoch": 0.9459915611814346,
+      "grad_norm": 0.39785075187683105,
+      "learning_rate": 0.0001965123592111823,
+      "loss": 0.8423402309417725,
+      "step": 2242
+    },
+    {
+      "epoch": 0.9468354430379747,
+      "grad_norm": 0.47455263137817383,
+      "learning_rate": 0.00019649977428034632,
+      "loss": 0.8883589506149292,
+      "step": 2244
+    },
+    {
+      "epoch": 0.9476793248945148,
+      "grad_norm": 0.3866841197013855,
+      "learning_rate": 0.0001964871670888657,
+      "loss": 0.7232863903045654,
+      "step": 2246
+    },
+    {
+      "epoch": 0.9485232067510548,
+      "grad_norm": 0.3950609564781189,
+      "learning_rate": 0.00019647453763964876,
+      "loss": 0.7265347838401794,
+      "step": 2248
+    },
+    {
+      "epoch": 0.9493670886075949,
+      "grad_norm": 0.4315829575061798,
+      "learning_rate": 0.0001964618859356088,
+      "loss": 0.8000131249427795,
+      "step": 2250
+    },
+    {
+      "epoch": 0.950210970464135,
+      "grad_norm": 0.403984934091568,
+      "learning_rate": 0.00019644921197966433,
+      "loss": 0.823086678981781,
+      "step": 2252
+    },
+    {
+      "epoch": 0.9510548523206751,
+      "grad_norm": 0.4311080574989319,
+      "learning_rate": 0.00019643651577473897,
+      "loss": 0.8457553386688232,
+      "step": 2254
+    },
+    {
+      "epoch": 0.9518987341772152,
+      "grad_norm": 0.3884856104850769,
+      "learning_rate": 0.00019642379732376158,
+      "loss": 0.7866189479827881,
+      "step": 2256
+    },
+    {
+      "epoch": 0.9527426160337553,
+      "grad_norm": 0.3384067714214325,
+      "learning_rate": 0.00019641105662966595,
+      "loss": 0.7095813751220703,
+      "step": 2258
+    },
+    {
+      "epoch": 0.9535864978902954,
+      "grad_norm": 0.40426746010780334,
+      "learning_rate": 0.00019639829369539117,
+      "loss": 0.8271632194519043,
+      "step": 2260
+    },
+    {
+      "epoch": 0.9544303797468354,
+      "grad_norm": 0.45495131611824036,
+      "learning_rate": 0.00019638550852388144,
+      "loss": 0.8204861283302307,
+      "step": 2262
+    },
+    {
+      "epoch": 0.9552742616033755,
+      "grad_norm": 0.47201067209243774,
+      "learning_rate": 0.00019637270111808598,
+      "loss": 0.884763777256012,
+      "step": 2264
+    },
+    {
+      "epoch": 0.9561181434599156,
+      "grad_norm": 0.3959396779537201,
+      "learning_rate": 0.0001963598714809592,
+      "loss": 0.7630926370620728,
+      "step": 2266
+    },
+    {
+      "epoch": 0.9569620253164557,
+      "grad_norm": 0.4197136461734772,
+      "learning_rate": 0.00019634701961546077,
+      "loss": 0.797870397567749,
+      "step": 2268
+    },
+    {
+      "epoch": 0.9578059071729957,
+      "grad_norm": 0.381290078163147,
+      "learning_rate": 0.00019633414552455526,
+      "loss": 0.7755846977233887,
+      "step": 2270
+    },
+    {
+      "epoch": 0.9586497890295359,
+      "grad_norm": 0.3868124783039093,
+      "learning_rate": 0.00019632124921121255,
+      "loss": 0.7553264498710632,
+      "step": 2272
+    },
+    {
+      "epoch": 0.959493670886076,
+      "grad_norm": 0.4350631833076477,
+      "learning_rate": 0.00019630833067840747,
+      "loss": 0.7792160511016846,
+      "step": 2274
+    },
+    {
+      "epoch": 0.960337552742616,
+      "grad_norm": 0.4717157781124115,
+      "learning_rate": 0.00019629538992912016,
+      "loss": 0.878220796585083,
+      "step": 2276
+    },
+    {
+      "epoch": 0.9611814345991562,
+      "grad_norm": 0.45545122027397156,
+      "learning_rate": 0.00019628242696633584,
+      "loss": 0.8891014456748962,
+      "step": 2278
+    },
+    {
+      "epoch": 0.9620253164556962,
+      "grad_norm": 0.4421025514602661,
+      "learning_rate": 0.00019626944179304467,
+      "loss": 0.8024723529815674,
+      "step": 2280
+    },
+    {
+      "epoch": 0.9628691983122363,
+      "grad_norm": 0.40385812520980835,
+      "learning_rate": 0.00019625643441224222,
+      "loss": 0.8036280274391174,
+      "step": 2282
+    },
+    {
+      "epoch": 0.9637130801687763,
+      "grad_norm": 0.3866552412509918,
+      "learning_rate": 0.00019624340482692898,
+      "loss": 0.8072395324707031,
+      "step": 2284
+    },
+    {
+      "epoch": 0.9645569620253165,
+      "grad_norm": 0.4113233685493469,
+      "learning_rate": 0.0001962303530401106,
+      "loss": 0.8046114444732666,
+      "step": 2286
+    },
+    {
+      "epoch": 0.9654008438818565,
+      "grad_norm": 0.40720227360725403,
+      "learning_rate": 0.00019621727905479798,
+      "loss": 0.8434037566184998,
+      "step": 2288
+    },
+    {
+      "epoch": 0.9662447257383966,
+      "grad_norm": 0.35759273171424866,
+      "learning_rate": 0.00019620418287400693,
+      "loss": 0.7832419276237488,
+      "step": 2290
+    },
+    {
+      "epoch": 0.9670886075949368,
+      "grad_norm": 0.4170442223548889,
+      "learning_rate": 0.0001961910645007585,
+      "loss": 0.7975693345069885,
+      "step": 2292
+    },
+    {
+      "epoch": 0.9679324894514768,
+      "grad_norm": 0.4126700162887573,
+      "learning_rate": 0.00019617792393807895,
+      "loss": 0.8554494976997375,
+      "step": 2294
+    },
+    {
+      "epoch": 0.9687763713080169,
+      "grad_norm": 0.4231695830821991,
+      "learning_rate": 0.00019616476118899942,
+      "loss": 0.824539065361023,
+      "step": 2296
+    },
+    {
+      "epoch": 0.9696202531645569,
+      "grad_norm": 0.3556209206581116,
+      "learning_rate": 0.0001961515762565564,
+      "loss": 0.7008040547370911,
+      "step": 2298
+    },
+    {
+      "epoch": 0.9704641350210971,
+      "grad_norm": 0.4370400309562683,
+      "learning_rate": 0.00019613836914379132,
+      "loss": 0.8192558288574219,
+      "step": 2300
+    },
+    {
+      "epoch": 0.9704641350210971,
+      "eval_loss": 0.8281980156898499,
+      "eval_runtime": 688.6136,
+      "eval_samples_per_second": 3.06,
+      "eval_steps_per_second": 3.06,
+      "step": 2300
+    },
+    {
+      "epoch": 0.9713080168776371,
+      "grad_norm": 0.4207614064216614,
+      "learning_rate": 0.00019612513985375088,
+      "loss": 0.878928005695343,
+      "step": 2302
+    },
+    {
+      "epoch": 0.9721518987341772,
+      "grad_norm": 0.37607815861701965,
+      "learning_rate": 0.0001961118883894868,
+      "loss": 0.736182689666748,
+      "step": 2304
+    },
+    {
+      "epoch": 0.9729957805907173,
+      "grad_norm": 0.43311476707458496,
+      "learning_rate": 0.00019609861475405592,
+      "loss": 0.8403618335723877,
+      "step": 2306
+    },
+    {
+      "epoch": 0.9738396624472574,
+      "grad_norm": 0.4533176124095917,
+      "learning_rate": 0.00019608531895052021,
+      "loss": 0.8414564728736877,
+      "step": 2308
+    },
+    {
+      "epoch": 0.9746835443037974,
+      "grad_norm": 0.4025697410106659,
+      "learning_rate": 0.00019607200098194677,
+      "loss": 0.8105775117874146,
+      "step": 2310
+    },
+    {
+      "epoch": 0.9755274261603376,
+      "grad_norm": 0.37894198298454285,
+      "learning_rate": 0.00019605866085140784,
+      "loss": 0.8053821325302124,
+      "step": 2312
+    },
+    {
+      "epoch": 0.9763713080168777,
+      "grad_norm": 0.46265891194343567,
+      "learning_rate": 0.0001960452985619807,
+      "loss": 0.8092884421348572,
+      "step": 2314
+    },
+    {
+      "epoch": 0.9772151898734177,
+      "grad_norm": 0.3770054280757904,
+      "learning_rate": 0.0001960319141167477,
+      "loss": 0.7456095218658447,
+      "step": 2316
+    },
+    {
+      "epoch": 0.9780590717299578,
+      "grad_norm": 0.44084450602531433,
+      "learning_rate": 0.0001960185075187965,
+      "loss": 0.7739883661270142,
+      "step": 2318
+    },
+    {
+      "epoch": 0.9789029535864979,
+      "grad_norm": 0.3880341351032257,
+      "learning_rate": 0.00019600507877121965,
+      "loss": 0.6961302757263184,
+      "step": 2320
+    },
+    {
+      "epoch": 0.979746835443038,
+      "grad_norm": 0.468037486076355,
+      "learning_rate": 0.00019599162787711495,
+      "loss": 0.8447984457015991,
+      "step": 2322
+    },
+    {
+      "epoch": 0.980590717299578,
+      "grad_norm": 0.4123631417751312,
+      "learning_rate": 0.00019597815483958527,
+      "loss": 0.8155615329742432,
+      "step": 2324
+    },
+    {
+      "epoch": 0.9814345991561182,
+      "grad_norm": 0.3827875554561615,
+      "learning_rate": 0.00019596465966173853,
+      "loss": 0.7990952730178833,
+      "step": 2326
+    },
+    {
+      "epoch": 0.9822784810126582,
+      "grad_norm": 0.4644222557544708,
+      "learning_rate": 0.00019595114234668788,
+      "loss": 0.8899750709533691,
+      "step": 2328
+    },
+    {
+      "epoch": 0.9831223628691983,
+      "grad_norm": 0.43699249625205994,
+      "learning_rate": 0.00019593760289755144,
+      "loss": 0.7863160371780396,
+      "step": 2330
+    },
+    {
+      "epoch": 0.9839662447257383,
+      "grad_norm": 0.40713244676589966,
+      "learning_rate": 0.00019592404131745254,
+      "loss": 0.7960637807846069,
+      "step": 2332
+    },
+    {
+      "epoch": 0.9848101265822785,
+      "grad_norm": 0.37364068627357483,
+      "learning_rate": 0.00019591045760951957,
+      "loss": 0.714144229888916,
+      "step": 2334
+    },
+    {
+      "epoch": 0.9856540084388186,
+      "grad_norm": 0.4094736576080322,
+      "learning_rate": 0.00019589685177688599,
+      "loss": 0.7311791777610779,
+      "step": 2336
+    },
+    {
+      "epoch": 0.9864978902953586,
+      "grad_norm": 0.4275180697441101,
+      "learning_rate": 0.0001958832238226905,
+      "loss": 0.8191342353820801,
+      "step": 2338
+    },
+    {
+      "epoch": 0.9873417721518988,
+      "grad_norm": 0.4627651274204254,
+      "learning_rate": 0.00019586957375007668,
+      "loss": 0.7476106286048889,
+      "step": 2340
+    },
+    {
+      "epoch": 0.9881856540084388,
+      "grad_norm": 0.44676414132118225,
+      "learning_rate": 0.00019585590156219346,
+      "loss": 0.8257967233657837,
+      "step": 2342
+    },
+    {
+      "epoch": 0.9890295358649789,
+      "grad_norm": 0.40667134523391724,
+      "learning_rate": 0.00019584220726219466,
+      "loss": 0.7922079563140869,
+      "step": 2344
+    },
+    {
+      "epoch": 0.9898734177215189,
+      "grad_norm": 0.40865346789360046,
+      "learning_rate": 0.00019582849085323934,
+      "loss": 0.7892434597015381,
+      "step": 2346
+    },
+    {
+      "epoch": 0.9907172995780591,
+      "grad_norm": 0.43309709429740906,
+      "learning_rate": 0.00019581475233849163,
+      "loss": 0.8766644597053528,
+      "step": 2348
+    },
+    {
+      "epoch": 0.9915611814345991,
+      "grad_norm": 0.41558781266212463,
+      "learning_rate": 0.00019580099172112069,
+      "loss": 0.9183312058448792,
+      "step": 2350
+    },
+    {
+      "epoch": 0.9924050632911392,
+      "grad_norm": 0.432748019695282,
+      "learning_rate": 0.00019578720900430082,
+      "loss": 0.7905668020248413,
+      "step": 2352
+    },
+    {
+      "epoch": 0.9932489451476794,
+      "grad_norm": 0.4116985499858856,
+      "learning_rate": 0.0001957734041912115,
+      "loss": 0.8279712200164795,
+      "step": 2354
+    },
+    {
+      "epoch": 0.9940928270042194,
+      "grad_norm": 0.3927275240421295,
+      "learning_rate": 0.00019575957728503718,
+      "loss": 0.7451362609863281,
+      "step": 2356
+    },
+    {
+      "epoch": 0.9949367088607595,
+      "grad_norm": 0.3389509320259094,
+      "learning_rate": 0.0001957457282889675,
+      "loss": 0.6769390106201172,
+      "step": 2358
+    },
+    {
+      "epoch": 0.9957805907172996,
+      "grad_norm": 0.4220774471759796,
+      "learning_rate": 0.00019573185720619712,
+      "loss": 0.7590897083282471,
+      "step": 2360
+    },
+    {
+      "epoch": 0.9966244725738397,
+      "grad_norm": 0.4299897849559784,
+      "learning_rate": 0.00019571796403992583,
+      "loss": 0.8301284313201904,
+      "step": 2362
+    },
+    {
+      "epoch": 0.9974683544303797,
+      "grad_norm": 0.4031789302825928,
+      "learning_rate": 0.00019570404879335856,
+      "loss": 0.8019471764564514,
+      "step": 2364
+    },
+    {
+      "epoch": 0.9983122362869198,
+      "grad_norm": 0.3957567811012268,
+      "learning_rate": 0.00019569011146970526,
+      "loss": 0.7427027821540833,
+      "step": 2366
+    },
+    {
+      "epoch": 0.99915611814346,
+      "grad_norm": 0.41741347312927246,
+      "learning_rate": 0.000195676152072181,
+      "loss": 0.8221617341041565,
+      "step": 2368
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.47121551632881165,
+      "learning_rate": 0.00019566217060400595,
+      "loss": 0.8143455982208252,
+      "step": 2370
+    },
+    {
+      "epoch": 1.0008438818565402,
+      "grad_norm": 0.3854716122150421,
+      "learning_rate": 0.0001956481670684054,
+      "loss": 0.6971694231033325,
+      "step": 2372
+    },
+    {
+      "epoch": 1.00168776371308,
+      "grad_norm": 0.4029138386249542,
+      "learning_rate": 0.00019563414146860964,
+      "loss": 0.6719155311584473,
+      "step": 2374
+    },
+    {
+      "epoch": 1.0025316455696203,
+      "grad_norm": 0.40136656165122986,
+      "learning_rate": 0.00019562009380785412,
+      "loss": 0.7161872386932373,
+      "step": 2376
+    },
+    {
+      "epoch": 1.0033755274261604,
+      "grad_norm": 0.40685850381851196,
+      "learning_rate": 0.0001956060240893794,
+      "loss": 0.71699059009552,
+      "step": 2378
+    },
+    {
+      "epoch": 1.0042194092827004,
+      "grad_norm": 0.46153366565704346,
+      "learning_rate": 0.00019559193231643104,
+      "loss": 0.8030059337615967,
+      "step": 2380
+    },
+    {
+      "epoch": 1.0050632911392405,
+      "grad_norm": 0.45048823952674866,
+      "learning_rate": 0.0001955778184922598,
+      "loss": 0.7666479349136353,
+      "step": 2382
+    },
+    {
+      "epoch": 1.0059071729957807,
+      "grad_norm": 0.4260413348674774,
+      "learning_rate": 0.00019556368262012142,
+      "loss": 0.7481234073638916,
+      "step": 2384
+    },
+    {
+      "epoch": 1.0067510548523206,
+      "grad_norm": 0.4819575548171997,
+      "learning_rate": 0.0001955495247032768,
+      "loss": 0.7943820953369141,
+      "step": 2386
+    },
+    {
+      "epoch": 1.0075949367088608,
+      "grad_norm": 0.45789605379104614,
+      "learning_rate": 0.0001955353447449919,
+      "loss": 0.6401634812355042,
+      "step": 2388
+    },
+    {
+      "epoch": 1.0084388185654007,
+      "grad_norm": 0.37019097805023193,
+      "learning_rate": 0.00019552114274853778,
+      "loss": 0.7058886885643005,
+      "step": 2390
+    },
+    {
+      "epoch": 1.009282700421941,
+      "grad_norm": 0.44416263699531555,
+      "learning_rate": 0.0001955069187171905,
+      "loss": 0.7631242871284485,
+      "step": 2392
+    },
+    {
+      "epoch": 1.010126582278481,
+      "grad_norm": 0.4478379786014557,
+      "learning_rate": 0.00019549267265423138,
+      "loss": 0.7873892784118652,
+      "step": 2394
+    },
+    {
+      "epoch": 1.010970464135021,
+      "grad_norm": 0.4596130847930908,
+      "learning_rate": 0.00019547840456294658,
+      "loss": 0.7882059216499329,
+      "step": 2396
+    },
+    {
+      "epoch": 1.0118143459915612,
+      "grad_norm": 0.4062999188899994,
+      "learning_rate": 0.0001954641144466276,
+      "loss": 0.6358574628829956,
+      "step": 2398
+    },
+    {
+      "epoch": 1.0126582278481013,
+      "grad_norm": 0.47627049684524536,
+      "learning_rate": 0.0001954498023085708,
+      "loss": 0.7803886532783508,
+      "step": 2400
+    },
+    {
+      "epoch": 1.0126582278481013,
+      "eval_loss": 0.8250564932823181,
+      "eval_runtime": 691.5833,
+      "eval_samples_per_second": 3.047,
+      "eval_steps_per_second": 3.047,
+      "step": 2400
+    },
+    {
+      "epoch": 1.0135021097046413,
+      "grad_norm": 0.40690696239471436,
+      "learning_rate": 0.00019543546815207777,
+      "loss": 0.7394700050354004,
+      "step": 2402
+    },
+    {
+      "epoch": 1.0143459915611814,
+      "grad_norm": 0.45543980598449707,
+      "learning_rate": 0.0001954211119804551,
+      "loss": 0.7096555829048157,
+      "step": 2404
+    },
+    {
+      "epoch": 1.0151898734177216,
+      "grad_norm": 0.4367314279079437,
+      "learning_rate": 0.00019540673379701454,
+      "loss": 0.7961843013763428,
+      "step": 2406
+    },
+    {
+      "epoch": 1.0160337552742615,
+      "grad_norm": 0.4458340108394623,
+      "learning_rate": 0.00019539233360507278,
+      "loss": 0.743843138217926,
+      "step": 2408
+    },
+    {
+      "epoch": 1.0168776371308017,
+      "grad_norm": 0.3741893470287323,
+      "learning_rate": 0.0001953779114079517,
+      "loss": 0.7273756265640259,
+      "step": 2410
+    },
+    {
+      "epoch": 1.0177215189873419,
+      "grad_norm": 0.4340079426765442,
+      "learning_rate": 0.00019536346720897823,
+      "loss": 0.7922506928443909,
+      "step": 2412
+    },
+    {
+      "epoch": 1.0185654008438818,
+      "grad_norm": 0.4518018662929535,
+      "learning_rate": 0.00019534900101148437,
+      "loss": 0.6492037773132324,
+      "step": 2414
+    },
+    {
+      "epoch": 1.019409282700422,
+      "grad_norm": 0.4160727560520172,
+      "learning_rate": 0.00019533451281880716,
+      "loss": 0.7422950863838196,
+      "step": 2416
+    },
+    {
+      "epoch": 1.0202531645569621,
+      "grad_norm": 0.35953211784362793,
+      "learning_rate": 0.00019532000263428883,
+      "loss": 0.6489207148551941,
+      "step": 2418
+    },
+    {
+      "epoch": 1.021097046413502,
+      "grad_norm": 0.38774433732032776,
+      "learning_rate": 0.0001953054704612765,
+      "loss": 0.7449417114257812,
+      "step": 2420
+    },
+    {
+      "epoch": 1.0219409282700422,
+      "grad_norm": 0.48185858130455017,
+      "learning_rate": 0.0001952909163031225,
+      "loss": 0.8481494188308716,
+      "step": 2422
+    },
+    {
+      "epoch": 1.0227848101265822,
+      "grad_norm": 0.4098491072654724,
+      "learning_rate": 0.00019527634016318425,
+      "loss": 0.7442291378974915,
+      "step": 2424
+    },
+    {
+      "epoch": 1.0236286919831223,
+      "grad_norm": 0.41284480690956116,
+      "learning_rate": 0.00019526174204482412,
+      "loss": 0.7564480900764465,
+      "step": 2426
+    },
+    {
+      "epoch": 1.0244725738396625,
+      "grad_norm": 0.44948479533195496,
+      "learning_rate": 0.00019524712195140965,
+      "loss": 0.7234854102134705,
+      "step": 2428
+    },
+    {
+      "epoch": 1.0253164556962024,
+      "grad_norm": 0.39939233660697937,
+      "learning_rate": 0.0001952324798863134,
+      "loss": 0.7638729810714722,
+      "step": 2430
+    },
+    {
+      "epoch": 1.0261603375527426,
+      "grad_norm": 0.40807482600212097,
+      "learning_rate": 0.00019521781585291298,
+      "loss": 0.6922630071640015,
+      "step": 2432
+    },
+    {
+      "epoch": 1.0270042194092828,
+      "grad_norm": 0.45735791325569153,
+      "learning_rate": 0.00019520312985459116,
+      "loss": 0.762996256351471,
+      "step": 2434
+    },
+    {
+      "epoch": 1.0278481012658227,
+      "grad_norm": 0.4280409812927246,
+      "learning_rate": 0.0001951884218947357,
+      "loss": 0.7811596989631653,
+      "step": 2436
+    },
+    {
+      "epoch": 1.0286919831223629,
+      "grad_norm": 0.41957560181617737,
+      "learning_rate": 0.00019517369197673943,
+      "loss": 0.7278305292129517,
+      "step": 2438
+    },
+    {
+      "epoch": 1.029535864978903,
+      "grad_norm": 0.4832586646080017,
+      "learning_rate": 0.00019515894010400028,
+      "loss": 0.807456910610199,
+      "step": 2440
+    },
+    {
+      "epoch": 1.030379746835443,
+      "grad_norm": 0.46448788046836853,
+      "learning_rate": 0.00019514416627992122,
+      "loss": 0.7421302795410156,
+      "step": 2442
+    },
+    {
+      "epoch": 1.0312236286919831,
+      "grad_norm": 0.41362619400024414,
+      "learning_rate": 0.00019512937050791024,
+      "loss": 0.6892213225364685,
+      "step": 2444
+    },
+    {
+      "epoch": 1.0320675105485233,
+      "grad_norm": 0.4158109128475189,
+      "learning_rate": 0.0001951145527913805,
+      "loss": 0.7305331826210022,
+      "step": 2446
+    },
+    {
+      "epoch": 1.0329113924050632,
+      "grad_norm": 0.4725087881088257,
+      "learning_rate": 0.00019509971313375011,
+      "loss": 0.7682753205299377,
+      "step": 2448
+    },
+    {
+      "epoch": 1.0337552742616034,
+      "grad_norm": 0.4341438114643097,
+      "learning_rate": 0.00019508485153844237,
+      "loss": 0.747041642665863,
+      "step": 2450
+    },
+    {
+      "epoch": 1.0345991561181433,
+      "grad_norm": 0.44515377283096313,
+      "learning_rate": 0.00019506996800888548,
+      "loss": 0.7312062382698059,
+      "step": 2452
+    },
+    {
+      "epoch": 1.0354430379746835,
+      "grad_norm": 0.5313915014266968,
+      "learning_rate": 0.00019505506254851283,
+      "loss": 0.7861107587814331,
+      "step": 2454
+    },
+    {
+      "epoch": 1.0362869198312237,
+      "grad_norm": 0.6094246506690979,
+      "learning_rate": 0.00019504013516076285,
+      "loss": 0.8458115458488464,
+      "step": 2456
+    },
+    {
+      "epoch": 1.0371308016877636,
+      "grad_norm": 0.36419153213500977,
+      "learning_rate": 0.00019502518584907895,
+      "loss": 0.6091256141662598,
+      "step": 2458
+    },
+    {
+      "epoch": 1.0379746835443038,
+      "grad_norm": 0.399271696805954,
+      "learning_rate": 0.00019501021461690966,
+      "loss": 0.6703370809555054,
+      "step": 2460
+    },
+    {
+      "epoch": 1.038818565400844,
+      "grad_norm": 0.38560932874679565,
+      "learning_rate": 0.00019499522146770856,
+      "loss": 0.7220943570137024,
+      "step": 2462
+    },
+    {
+      "epoch": 1.0396624472573839,
+      "grad_norm": 0.44199803471565247,
+      "learning_rate": 0.00019498020640493428,
+      "loss": 0.7501406073570251,
+      "step": 2464
+    },
+    {
+      "epoch": 1.040506329113924,
+      "grad_norm": 0.47394421696662903,
+      "learning_rate": 0.00019496516943205054,
+      "loss": 0.7957925796508789,
+      "step": 2466
+    },
+    {
+      "epoch": 1.0413502109704642,
+      "grad_norm": 0.5274425745010376,
+      "learning_rate": 0.00019495011055252603,
+      "loss": 0.7964824438095093,
+      "step": 2468
+    },
+    {
+      "epoch": 1.0421940928270041,
+      "grad_norm": 0.44004881381988525,
+      "learning_rate": 0.0001949350297698346,
+      "loss": 0.7478582262992859,
+      "step": 2470
+    },
+    {
+      "epoch": 1.0430379746835443,
+      "grad_norm": 0.40662670135498047,
+      "learning_rate": 0.000194919927087455,
+      "loss": 0.738339900970459,
+      "step": 2472
+    },
+    {
+      "epoch": 1.0438818565400845,
+      "grad_norm": 0.43934181332588196,
+      "learning_rate": 0.00019490480250887123,
+      "loss": 0.817171037197113,
+      "step": 2474
+    },
+    {
+      "epoch": 1.0447257383966244,
+      "grad_norm": 0.4111221432685852,
+      "learning_rate": 0.00019488965603757223,
+      "loss": 0.7413169145584106,
+      "step": 2476
+    },
+    {
+      "epoch": 1.0455696202531646,
+      "grad_norm": 0.3770996630191803,
+      "learning_rate": 0.00019487448767705194,
+      "loss": 0.7893102169036865,
+      "step": 2478
+    },
+    {
+      "epoch": 1.0464135021097047,
+      "grad_norm": 0.4524461627006531,
+      "learning_rate": 0.00019485929743080945,
+      "loss": 0.7037841081619263,
+      "step": 2480
+    },
+    {
+      "epoch": 1.0472573839662447,
+      "grad_norm": 0.4232378900051117,
+      "learning_rate": 0.00019484408530234885,
+      "loss": 0.7456525564193726,
+      "step": 2482
+    },
+    {
+      "epoch": 1.0481012658227848,
+      "grad_norm": 0.3842810392379761,
+      "learning_rate": 0.00019482885129517929,
+      "loss": 0.7263808250427246,
+      "step": 2484
+    },
+    {
+      "epoch": 1.048945147679325,
+      "grad_norm": 0.43203726410865784,
+      "learning_rate": 0.00019481359541281495,
+      "loss": 0.7101474404335022,
+      "step": 2486
+    },
+    {
+      "epoch": 1.049789029535865,
+      "grad_norm": 0.46944454312324524,
+      "learning_rate": 0.00019479831765877509,
+      "loss": 0.8561921715736389,
+      "step": 2488
+    },
+    {
+      "epoch": 1.0506329113924051,
+      "grad_norm": 0.4455019533634186,
+      "learning_rate": 0.00019478301803658397,
+      "loss": 0.7206679582595825,
+      "step": 2490
+    },
+    {
+      "epoch": 1.051476793248945,
+      "grad_norm": 0.4253813326358795,
+      "learning_rate": 0.00019476769654977095,
+      "loss": 0.7271745800971985,
+      "step": 2492
+    },
+    {
+      "epoch": 1.0523206751054852,
+      "grad_norm": 0.42758429050445557,
+      "learning_rate": 0.00019475235320187033,
+      "loss": 0.7358576059341431,
+      "step": 2494
+    },
+    {
+      "epoch": 1.0531645569620254,
+      "grad_norm": 0.400634765625,
+      "learning_rate": 0.00019473698799642164,
+      "loss": 0.6617647409439087,
+      "step": 2496
+    },
+    {
+      "epoch": 1.0540084388185653,
+      "grad_norm": 0.4132104814052582,
+      "learning_rate": 0.00019472160093696926,
+      "loss": 0.6900970935821533,
+      "step": 2498
+    },
+    {
+      "epoch": 1.0548523206751055,
+      "grad_norm": 0.5116495490074158,
+      "learning_rate": 0.00019470619202706268,
+      "loss": 0.7352861762046814,
+      "step": 2500
+    },
+    {
+      "epoch": 1.0548523206751055,
+      "eval_loss": 0.8249453902244568,
+      "eval_runtime": 679.4446,
+      "eval_samples_per_second": 3.101,
+      "eval_steps_per_second": 3.101,
+      "step": 2500
+    },
+    {
+      "epoch": 1.0556962025316456,
+      "grad_norm": 0.42822444438934326,
+      "learning_rate": 0.0001946907612702565,
+      "loss": 0.7583479881286621,
+      "step": 2502
+    },
+    {
+      "epoch": 1.0565400843881856,
+      "grad_norm": 0.4788980782032013,
+      "learning_rate": 0.00019467530867011025,
+      "loss": 0.777511477470398,
+      "step": 2504
+    },
+    {
+      "epoch": 1.0573839662447257,
+      "grad_norm": 0.4210498034954071,
+      "learning_rate": 0.00019465983423018856,
+      "loss": 0.7452852129936218,
+      "step": 2506
+    },
+    {
+      "epoch": 1.058227848101266,
+      "grad_norm": 0.43219906091690063,
+      "learning_rate": 0.0001946443379540611,
+      "loss": 0.7540958523750305,
+      "step": 2508
+    },
+    {
+      "epoch": 1.0590717299578059,
+      "grad_norm": 0.4361002445220947,
+      "learning_rate": 0.00019462881984530253,
+      "loss": 0.7175624966621399,
+      "step": 2510
+    },
+    {
+      "epoch": 1.059915611814346,
+      "grad_norm": 0.4575091600418091,
+      "learning_rate": 0.00019461327990749264,
+      "loss": 0.7342076301574707,
+      "step": 2512
+    },
+    {
+      "epoch": 1.0607594936708862,
+      "grad_norm": 0.4823324978351593,
+      "learning_rate": 0.00019459771814421615,
+      "loss": 0.8255488872528076,
+      "step": 2514
+    },
+    {
+      "epoch": 1.0616033755274261,
+      "grad_norm": 0.41693785786628723,
+      "learning_rate": 0.00019458213455906284,
+      "loss": 0.6624729633331299,
+      "step": 2516
+    },
+    {
+      "epoch": 1.0624472573839663,
+      "grad_norm": 0.41319167613983154,
+      "learning_rate": 0.00019456652915562762,
+      "loss": 0.8224266767501831,
+      "step": 2518
+    },
+    {
+      "epoch": 1.0632911392405062,
+      "grad_norm": 0.4405190050601959,
+      "learning_rate": 0.00019455090193751023,
+      "loss": 0.7582900524139404,
+      "step": 2520
+    },
+    {
+      "epoch": 1.0641350210970464,
+      "grad_norm": 0.4564387798309326,
+      "learning_rate": 0.0001945352529083157,
+      "loss": 0.7948101162910461,
+      "step": 2522
+    },
+    {
+      "epoch": 1.0649789029535865,
+      "grad_norm": 0.4247225224971771,
+      "learning_rate": 0.00019451958207165394,
+      "loss": 0.7670508027076721,
+      "step": 2524
+    },
+    {
+      "epoch": 1.0658227848101265,
+      "grad_norm": 0.42070862650871277,
+      "learning_rate": 0.00019450388943113983,
+      "loss": 0.7753698229789734,
+      "step": 2526
+    },
+    {
+      "epoch": 1.0666666666666667,
+      "grad_norm": 0.3729179799556732,
+      "learning_rate": 0.00019448817499039342,
+      "loss": 0.710904598236084,
+      "step": 2528
+    },
+    {
+      "epoch": 1.0675105485232068,
+      "grad_norm": 0.40901753306388855,
+      "learning_rate": 0.0001944724387530397,
+      "loss": 0.7000213265419006,
+      "step": 2530
+    },
+    {
+      "epoch": 1.0683544303797468,
+      "grad_norm": 0.4481421113014221,
+      "learning_rate": 0.00019445668072270877,
+      "loss": 0.7374410629272461,
+      "step": 2532
+    },
+    {
+      "epoch": 1.069198312236287,
+      "grad_norm": 0.4839915335178375,
+      "learning_rate": 0.00019444090090303567,
+      "loss": 0.746309220790863,
+      "step": 2534
+    },
+    {
+      "epoch": 1.070042194092827,
+      "grad_norm": 0.4982132613658905,
+      "learning_rate": 0.00019442509929766048,
+      "loss": 0.7448083162307739,
+      "step": 2536
+    },
+    {
+      "epoch": 1.070886075949367,
+      "grad_norm": 0.3987475037574768,
+      "learning_rate": 0.00019440927591022838,
+      "loss": 0.7052555680274963,
+      "step": 2538
+    },
+    {
+      "epoch": 1.0717299578059072,
+      "grad_norm": 0.4089896082878113,
+      "learning_rate": 0.00019439343074438944,
+      "loss": 0.7090281248092651,
+      "step": 2540
+    },
+    {
+      "epoch": 1.0725738396624473,
+      "grad_norm": 0.3703136146068573,
+      "learning_rate": 0.00019437756380379895,
+      "loss": 0.7530601024627686,
+      "step": 2542
+    },
+    {
+      "epoch": 1.0734177215189873,
+      "grad_norm": 0.4015701413154602,
+      "learning_rate": 0.00019436167509211702,
+      "loss": 0.7597941160202026,
+      "step": 2544
+    },
+    {
+      "epoch": 1.0742616033755275,
+      "grad_norm": 0.3915148675441742,
+      "learning_rate": 0.00019434576461300888,
+      "loss": 0.7168124318122864,
+      "step": 2546
+    },
+    {
+      "epoch": 1.0751054852320676,
+      "grad_norm": 0.44643113017082214,
+      "learning_rate": 0.00019432983237014482,
+      "loss": 0.7762725949287415,
+      "step": 2548
+    },
+    {
+      "epoch": 1.0759493670886076,
+      "grad_norm": 0.46874719858169556,
+      "learning_rate": 0.00019431387836720004,
+      "loss": 0.7879725098609924,
+      "step": 2550
+    },
+    {
+      "epoch": 1.0767932489451477,
+      "grad_norm": 0.4143001437187195,
+      "learning_rate": 0.00019429790260785487,
+      "loss": 0.7351391315460205,
+      "step": 2552
+    },
+    {
+      "epoch": 1.0776371308016879,
+      "grad_norm": 0.4569888412952423,
+      "learning_rate": 0.0001942819050957946,
+      "loss": 0.86613929271698,
+      "step": 2554
+    },
+    {
+      "epoch": 1.0784810126582278,
+      "grad_norm": 0.41729897260665894,
+      "learning_rate": 0.00019426588583470954,
+      "loss": 0.7122719287872314,
+      "step": 2556
+    },
+    {
+      "epoch": 1.079324894514768,
+      "grad_norm": 0.42295604944229126,
+      "learning_rate": 0.000194249844828295,
+      "loss": 0.6605332493782043,
+      "step": 2558
+    },
+    {
+      "epoch": 1.080168776371308,
+      "grad_norm": 0.45787641406059265,
+      "learning_rate": 0.00019423378208025137,
+      "loss": 0.7839207649230957,
+      "step": 2560
+    },
+    {
+      "epoch": 1.081012658227848,
+      "grad_norm": 0.43370622396469116,
+      "learning_rate": 0.00019421769759428404,
+      "loss": 0.7557987570762634,
+      "step": 2562
+    },
+    {
+      "epoch": 1.0818565400843883,
+      "grad_norm": 0.4371975362300873,
+      "learning_rate": 0.00019420159137410335,
+      "loss": 0.7349345088005066,
+      "step": 2564
+    },
+    {
+      "epoch": 1.0827004219409282,
+      "grad_norm": 0.468730628490448,
+      "learning_rate": 0.0001941854634234247,
+      "loss": 0.7024962902069092,
+      "step": 2566
+    },
+    {
+      "epoch": 1.0835443037974684,
+      "grad_norm": 0.4257284998893738,
+      "learning_rate": 0.0001941693137459685,
+      "loss": 0.6981064081192017,
+      "step": 2568
+    },
+    {
+      "epoch": 1.0843881856540085,
+      "grad_norm": 0.4322902262210846,
+      "learning_rate": 0.00019415314234546015,
+      "loss": 0.7899637818336487,
+      "step": 2570
+    },
+    {
+      "epoch": 1.0852320675105485,
+      "grad_norm": 0.44384217262268066,
+      "learning_rate": 0.00019413694922563014,
+      "loss": 0.7964152693748474,
+      "step": 2572
+    },
+    {
+      "epoch": 1.0860759493670886,
+      "grad_norm": 0.47391360998153687,
+      "learning_rate": 0.0001941207343902139,
+      "loss": 0.7902941107749939,
+      "step": 2574
+    },
+    {
+      "epoch": 1.0869198312236288,
+      "grad_norm": 0.4575883746147156,
+      "learning_rate": 0.00019410449784295182,
+      "loss": 0.7818595170974731,
+      "step": 2576
+    },
+    {
+      "epoch": 1.0877637130801687,
+      "grad_norm": 0.41236087679862976,
+      "learning_rate": 0.0001940882395875894,
+      "loss": 0.7516021728515625,
+      "step": 2578
+    },
+    {
+      "epoch": 1.0886075949367089,
+      "grad_norm": 0.3817494213581085,
+      "learning_rate": 0.0001940719596278771,
+      "loss": 0.6870401501655579,
+      "step": 2580
+    },
+    {
+      "epoch": 1.0894514767932488,
+      "grad_norm": 0.4630168378353119,
+      "learning_rate": 0.00019405565796757043,
+      "loss": 0.7437341809272766,
+      "step": 2582
+    },
+    {
+      "epoch": 1.090295358649789,
+      "grad_norm": 0.4063820540904999,
+      "learning_rate": 0.00019403933461042982,
+      "loss": 0.7101680040359497,
+      "step": 2584
+    },
+    {
+      "epoch": 1.0911392405063292,
+      "grad_norm": 0.44423359632492065,
+      "learning_rate": 0.0001940229895602208,
+      "loss": 0.6894780397415161,
+      "step": 2586
+    },
+    {
+      "epoch": 1.091983122362869,
+      "grad_norm": 0.4267353117465973,
+      "learning_rate": 0.0001940066228207138,
+      "loss": 0.7277120351791382,
+      "step": 2588
+    },
+    {
+      "epoch": 1.0928270042194093,
+      "grad_norm": 0.44747647643089294,
+      "learning_rate": 0.00019399023439568434,
+      "loss": 0.6512353420257568,
+      "step": 2590
+    },
+    {
+      "epoch": 1.0936708860759494,
+      "grad_norm": 0.4557100832462311,
+      "learning_rate": 0.00019397382428891296,
+      "loss": 0.7850068807601929,
+      "step": 2592
+    },
+    {
+      "epoch": 1.0945147679324894,
+      "grad_norm": 0.44980114698410034,
+      "learning_rate": 0.0001939573925041851,
+      "loss": 0.786868691444397,
+      "step": 2594
+    },
+    {
+      "epoch": 1.0953586497890295,
+      "grad_norm": 0.45374399423599243,
+      "learning_rate": 0.00019394093904529126,
+      "loss": 0.8260899186134338,
+      "step": 2596
+    },
+    {
+      "epoch": 1.0962025316455697,
+      "grad_norm": 0.40681353211402893,
+      "learning_rate": 0.00019392446391602695,
+      "loss": 0.7056689262390137,
+      "step": 2598
+    },
+    {
+      "epoch": 1.0970464135021096,
+      "grad_norm": 0.39278075098991394,
+      "learning_rate": 0.0001939079671201927,
+      "loss": 0.7557252645492554,
+      "step": 2600
+    },
+    {
+      "epoch": 1.0970464135021096,
+      "eval_loss": 0.8211485743522644,
+      "eval_runtime": 670.2276,
+      "eval_samples_per_second": 3.144,
+      "eval_steps_per_second": 3.144,
+      "step": 2600
+    },
+    {
+      "epoch": 1.0978902953586498,
+      "grad_norm": 0.4261230230331421,
+      "learning_rate": 0.00019389144866159398,
+      "loss": 0.7139819264411926,
+      "step": 2602
+    },
+    {
+      "epoch": 1.09873417721519,
+      "grad_norm": 0.4664439558982849,
+      "learning_rate": 0.0001938749085440412,
+      "loss": 0.7274705767631531,
+      "step": 2604
+    },
+    {
+      "epoch": 1.09957805907173,
+      "grad_norm": 0.46306154131889343,
+      "learning_rate": 0.00019385834677134998,
+      "loss": 0.7547051906585693,
+      "step": 2606
+    },
+    {
+      "epoch": 1.10042194092827,
+      "grad_norm": 0.39753687381744385,
+      "learning_rate": 0.00019384176334734073,
+      "loss": 0.6942694187164307,
+      "step": 2608
+    },
+    {
+      "epoch": 1.1012658227848102,
+      "grad_norm": 0.4556522071361542,
+      "learning_rate": 0.00019382515827583898,
+      "loss": 0.7161701321601868,
+      "step": 2610
+    },
+    {
+      "epoch": 1.1021097046413502,
+      "grad_norm": 0.3863290250301361,
+      "learning_rate": 0.0001938085315606751,
+      "loss": 0.7165710926055908,
+      "step": 2612
+    },
+    {
+      "epoch": 1.1029535864978903,
+      "grad_norm": 0.49601292610168457,
+      "learning_rate": 0.00019379188320568466,
+      "loss": 0.8080679774284363,
+      "step": 2614
+    },
+    {
+      "epoch": 1.1037974683544305,
+      "grad_norm": 0.4498465955257416,
+      "learning_rate": 0.00019377521321470805,
+      "loss": 0.7032502889633179,
+      "step": 2616
+    },
+    {
+      "epoch": 1.1046413502109704,
+      "grad_norm": 0.4537963569164276,
+      "learning_rate": 0.00019375852159159073,
+      "loss": 0.7076731324195862,
+      "step": 2618
+    },
+    {
+      "epoch": 1.1054852320675106,
+      "grad_norm": 0.4042968153953552,
+      "learning_rate": 0.0001937418083401832,
+      "loss": 0.7020825743675232,
+      "step": 2620
+    },
+    {
+      "epoch": 1.1063291139240505,
+      "grad_norm": 0.44306516647338867,
+      "learning_rate": 0.00019372507346434076,
+      "loss": 0.7222874760627747,
+      "step": 2622
+    },
+    {
+      "epoch": 1.1071729957805907,
+      "grad_norm": 0.3792516887187958,
+      "learning_rate": 0.00019370831696792392,
+      "loss": 0.6136211156845093,
+      "step": 2624
+    },
+    {
+      "epoch": 1.1080168776371309,
+      "grad_norm": 0.46655353903770447,
+      "learning_rate": 0.00019369153885479806,
+      "loss": 0.7358977198600769,
+      "step": 2626
+    },
+    {
+      "epoch": 1.1088607594936708,
+      "grad_norm": 0.4416999816894531,
+      "learning_rate": 0.0001936747391288336,
+      "loss": 0.7484551072120667,
+      "step": 2628
+    },
+    {
+      "epoch": 1.109704641350211,
+      "grad_norm": 0.4620150327682495,
+      "learning_rate": 0.00019365791779390584,
+      "loss": 0.6874802708625793,
+      "step": 2630
+    },
+    {
+      "epoch": 1.1105485232067511,
+      "grad_norm": 0.4357883334159851,
+      "learning_rate": 0.00019364107485389518,
+      "loss": 0.7168865203857422,
+      "step": 2632
+    },
+    {
+      "epoch": 1.111392405063291,
+      "grad_norm": 0.4581143856048584,
+      "learning_rate": 0.00019362421031268699,
+      "loss": 0.7533737421035767,
+      "step": 2634
+    },
+    {
+      "epoch": 1.1122362869198312,
+      "grad_norm": 0.46408653259277344,
+      "learning_rate": 0.00019360732417417152,
+      "loss": 0.8245398998260498,
+      "step": 2636
+    },
+    {
+      "epoch": 1.1130801687763714,
+      "grad_norm": 0.4639282524585724,
+      "learning_rate": 0.00019359041644224415,
+      "loss": 0.7095663547515869,
+      "step": 2638
+    },
+    {
+      "epoch": 1.1139240506329113,
+      "grad_norm": 0.4598273038864136,
+      "learning_rate": 0.00019357348712080517,
+      "loss": 0.7030285596847534,
+      "step": 2640
+    },
+    {
+      "epoch": 1.1147679324894515,
+      "grad_norm": 0.42993804812431335,
+      "learning_rate": 0.00019355653621375978,
+      "loss": 0.7158969640731812,
+      "step": 2642
+    },
+    {
+      "epoch": 1.1156118143459917,
+      "grad_norm": 0.36132168769836426,
+      "learning_rate": 0.0001935395637250183,
+      "loss": 0.6800820231437683,
+      "step": 2644
+    },
+    {
+      "epoch": 1.1164556962025316,
+      "grad_norm": 0.5004428029060364,
+      "learning_rate": 0.00019352256965849592,
+      "loss": 0.694972813129425,
+      "step": 2646
+    },
+    {
+      "epoch": 1.1172995780590718,
+      "grad_norm": 0.43715354800224304,
+      "learning_rate": 0.00019350555401811286,
+      "loss": 0.7164262533187866,
+      "step": 2648
+    },
+    {
+      "epoch": 1.1181434599156117,
+      "grad_norm": 0.42316755652427673,
+      "learning_rate": 0.00019348851680779433,
+      "loss": 0.7026737928390503,
+      "step": 2650
+    },
+    {
+      "epoch": 1.1189873417721519,
+      "grad_norm": 0.35848063230514526,
+      "learning_rate": 0.0001934714580314704,
+      "loss": 0.6774232387542725,
+      "step": 2652
+    },
+    {
+      "epoch": 1.119831223628692,
+      "grad_norm": 0.46613892912864685,
+      "learning_rate": 0.00019345437769307627,
+      "loss": 0.7789244055747986,
+      "step": 2654
+    },
+    {
+      "epoch": 1.120675105485232,
+      "grad_norm": 0.4298146367073059,
+      "learning_rate": 0.00019343727579655204,
+      "loss": 0.8647177815437317,
+      "step": 2656
+    },
+    {
+      "epoch": 1.1215189873417721,
+      "grad_norm": 0.4638161361217499,
+      "learning_rate": 0.00019342015234584276,
+      "loss": 0.6949124336242676,
+      "step": 2658
+    },
+    {
+      "epoch": 1.1223628691983123,
+      "grad_norm": 0.4079605042934418,
+      "learning_rate": 0.00019340300734489854,
+      "loss": 0.6677125096321106,
+      "step": 2660
+    },
+    {
+      "epoch": 1.1232067510548522,
+      "grad_norm": 0.46079063415527344,
+      "learning_rate": 0.00019338584079767434,
+      "loss": 0.7521067261695862,
+      "step": 2662
+    },
+    {
+      "epoch": 1.1240506329113924,
+      "grad_norm": 0.3935202956199646,
+      "learning_rate": 0.0001933686527081302,
+      "loss": 0.6829184889793396,
+      "step": 2664
+    },
+    {
+      "epoch": 1.1248945147679326,
+      "grad_norm": 0.45696479082107544,
+      "learning_rate": 0.000193351443080231,
+      "loss": 0.7122842669487,
+      "step": 2666
+    },
+    {
+      "epoch": 1.1257383966244725,
+      "grad_norm": 0.4078282415866852,
+      "learning_rate": 0.0001933342119179468,
+      "loss": 0.7316977977752686,
+      "step": 2668
+    },
+    {
+      "epoch": 1.1265822784810127,
+      "grad_norm": 0.4282447099685669,
+      "learning_rate": 0.0001933169592252524,
+      "loss": 0.7323395013809204,
+      "step": 2670
+    },
+    {
+      "epoch": 1.1274261603375528,
+      "grad_norm": 0.46398648619651794,
+      "learning_rate": 0.00019329968500612766,
+      "loss": 0.7177348732948303,
+      "step": 2672
+    },
+    {
+      "epoch": 1.1282700421940928,
+      "grad_norm": 0.4560202658176422,
+      "learning_rate": 0.0001932823892645575,
+      "loss": 0.7957183122634888,
+      "step": 2674
+    },
+    {
+      "epoch": 1.129113924050633,
+      "grad_norm": 0.4790743887424469,
+      "learning_rate": 0.00019326507200453162,
+      "loss": 0.762060284614563,
+      "step": 2676
+    },
+    {
+      "epoch": 1.129957805907173,
+      "grad_norm": 0.46524202823638916,
+      "learning_rate": 0.00019324773323004484,
+      "loss": 0.7227895259857178,
+      "step": 2678
+    },
+    {
+      "epoch": 1.130801687763713,
+      "grad_norm": 0.47912222146987915,
+      "learning_rate": 0.00019323037294509683,
+      "loss": 0.6968042254447937,
+      "step": 2680
+    },
+    {
+      "epoch": 1.1316455696202532,
+      "grad_norm": 0.4412849247455597,
+      "learning_rate": 0.00019321299115369233,
+      "loss": 0.6682016253471375,
+      "step": 2682
+    },
+    {
+      "epoch": 1.1324894514767934,
+      "grad_norm": 0.4138370752334595,
+      "learning_rate": 0.00019319558785984095,
+      "loss": 0.7287812829017639,
+      "step": 2684
+    },
+    {
+      "epoch": 1.1333333333333333,
+      "grad_norm": 0.391170859336853,
+      "learning_rate": 0.00019317816306755733,
+      "loss": 0.6547526121139526,
+      "step": 2686
+    },
+    {
+      "epoch": 1.1341772151898735,
+      "grad_norm": 0.45472806692123413,
+      "learning_rate": 0.00019316071678086098,
+      "loss": 0.6942042708396912,
+      "step": 2688
+    },
+    {
+      "epoch": 1.1350210970464134,
+      "grad_norm": 0.40321481227874756,
+      "learning_rate": 0.00019314324900377648,
+      "loss": 0.7417841553688049,
+      "step": 2690
+    },
+    {
+      "epoch": 1.1358649789029536,
+      "grad_norm": 0.3806784152984619,
+      "learning_rate": 0.00019312575974033327,
+      "loss": 0.7780525088310242,
+      "step": 2692
+    },
+    {
+      "epoch": 1.1367088607594937,
+      "grad_norm": 0.43763765692710876,
+      "learning_rate": 0.0001931082489945658,
+      "loss": 0.661472499370575,
+      "step": 2694
+    },
+    {
+      "epoch": 1.1375527426160337,
+      "grad_norm": 0.45410826802253723,
+      "learning_rate": 0.00019309071677051348,
+      "loss": 0.7443127632141113,
+      "step": 2696
+    },
+    {
+      "epoch": 1.1383966244725738,
+      "grad_norm": 0.47907447814941406,
+      "learning_rate": 0.00019307316307222062,
+      "loss": 0.744393527507782,
+      "step": 2698
+    },
+    {
+      "epoch": 1.139240506329114,
+      "grad_norm": 0.41723814606666565,
+      "learning_rate": 0.00019305558790373654,
+      "loss": 0.7042794227600098,
+      "step": 2700
+    },
+    {
+      "epoch": 1.139240506329114,
+      "eval_loss": 0.8155058026313782,
+      "eval_runtime": 678.284,
+      "eval_samples_per_second": 3.106,
+      "eval_steps_per_second": 3.106,
+      "step": 2700
+    },
+    {
+      "epoch": 1.140084388185654,
+      "grad_norm": 0.3863268196582794,
+      "learning_rate": 0.0001930379912691155,
+      "loss": 0.6715745329856873,
+      "step": 2702
+    },
+    {
+      "epoch": 1.140928270042194,
+      "grad_norm": 0.49994897842407227,
+      "learning_rate": 0.00019302037317241673,
+      "loss": 0.6959030628204346,
+      "step": 2704
+    },
+    {
+      "epoch": 1.1417721518987343,
+      "grad_norm": 0.46377572417259216,
+      "learning_rate": 0.00019300273361770432,
+      "loss": 0.6534222364425659,
+      "step": 2706
+    },
+    {
+      "epoch": 1.1426160337552742,
+      "grad_norm": 0.3841579556465149,
+      "learning_rate": 0.00019298507260904745,
+      "loss": 0.6750475764274597,
+      "step": 2708
+    },
+    {
+      "epoch": 1.1434599156118144,
+      "grad_norm": 0.4171544313430786,
+      "learning_rate": 0.00019296739015052015,
+      "loss": 0.7714900374412537,
+      "step": 2710
+    },
+    {
+      "epoch": 1.1443037974683543,
+      "grad_norm": 0.44384199380874634,
+      "learning_rate": 0.00019294968624620137,
+      "loss": 0.7281306982040405,
+      "step": 2712
+    },
+    {
+      "epoch": 1.1451476793248945,
+      "grad_norm": 0.43345192074775696,
+      "learning_rate": 0.00019293196090017513,
+      "loss": 0.7407262325286865,
+      "step": 2714
+    },
+    {
+      "epoch": 1.1459915611814346,
+      "grad_norm": 0.3897020220756531,
+      "learning_rate": 0.00019291421411653034,
+      "loss": 0.6794347167015076,
+      "step": 2716
+    },
+    {
+      "epoch": 1.1468354430379746,
+      "grad_norm": 0.4289143681526184,
+      "learning_rate": 0.00019289644589936075,
+      "loss": 0.7026289105415344,
+      "step": 2718
+    },
+    {
+      "epoch": 1.1476793248945147,
+      "grad_norm": 0.4783557057380676,
+      "learning_rate": 0.0001928786562527652,
+      "loss": 0.7613762617111206,
+      "step": 2720
+    },
+    {
+      "epoch": 1.148523206751055,
+      "grad_norm": 0.4152112901210785,
+      "learning_rate": 0.00019286084518084745,
+      "loss": 0.7123059630393982,
+      "step": 2722
+    },
+    {
+      "epoch": 1.1493670886075948,
+      "grad_norm": 0.4204908013343811,
+      "learning_rate": 0.0001928430126877161,
+      "loss": 0.7198041677474976,
+      "step": 2724
+    },
+    {
+      "epoch": 1.150210970464135,
+      "grad_norm": 0.41333863139152527,
+      "learning_rate": 0.00019282515877748485,
+      "loss": 0.6859839558601379,
+      "step": 2726
+    },
+    {
+      "epoch": 1.1510548523206752,
+      "grad_norm": 0.4041498005390167,
+      "learning_rate": 0.00019280728345427218,
+      "loss": 0.7185758948326111,
+      "step": 2728
+    },
+    {
+      "epoch": 1.1518987341772151,
+      "grad_norm": 0.4487358629703522,
+      "learning_rate": 0.00019278938672220167,
+      "loss": 0.7899971008300781,
+      "step": 2730
+    },
+    {
+      "epoch": 1.1527426160337553,
+      "grad_norm": 0.3510114252567291,
+      "learning_rate": 0.00019277146858540166,
+      "loss": 0.6266586780548096,
+      "step": 2732
+    },
+    {
+      "epoch": 1.1535864978902954,
+      "grad_norm": 0.4295555055141449,
+      "learning_rate": 0.00019275352904800553,
+      "loss": 0.7450110912322998,
+      "step": 2734
+    },
+    {
+      "epoch": 1.1544303797468354,
+      "grad_norm": 0.40629687905311584,
+      "learning_rate": 0.00019273556811415165,
+      "loss": 0.6875549554824829,
+      "step": 2736
+    },
+    {
+      "epoch": 1.1552742616033755,
+      "grad_norm": 0.4243362247943878,
+      "learning_rate": 0.00019271758578798322,
+      "loss": 0.8141522407531738,
+      "step": 2738
+    },
+    {
+      "epoch": 1.1561181434599157,
+      "grad_norm": 0.483306884765625,
+      "learning_rate": 0.00019269958207364843,
+      "loss": 0.6905186176300049,
+      "step": 2740
+    },
+    {
+      "epoch": 1.1569620253164556,
+      "grad_norm": 0.4303925335407257,
+      "learning_rate": 0.00019268155697530037,
+      "loss": 0.6709047555923462,
+      "step": 2742
+    },
+    {
+      "epoch": 1.1578059071729958,
+      "grad_norm": 0.4533381760120392,
+      "learning_rate": 0.0001926635104970971,
+      "loss": 0.7107465863227844,
+      "step": 2744
+    },
+    {
+      "epoch": 1.158649789029536,
+      "grad_norm": 0.4823286533355713,
+      "learning_rate": 0.0001926454426432016,
+      "loss": 0.8043986558914185,
+      "step": 2746
+    },
+    {
+      "epoch": 1.159493670886076,
+      "grad_norm": 0.3892867863178253,
+      "learning_rate": 0.00019262735341778179,
+      "loss": 0.7812165021896362,
+      "step": 2748
+    },
+    {
+      "epoch": 1.160337552742616,
+      "grad_norm": 0.46737003326416016,
+      "learning_rate": 0.00019260924282501045,
+      "loss": 0.7141886353492737,
+      "step": 2750
+    },
+    {
+      "epoch": 1.1611814345991562,
+      "grad_norm": 0.4335328936576843,
+      "learning_rate": 0.0001925911108690654,
+      "loss": 0.7528460621833801,
+      "step": 2752
+    },
+    {
+      "epoch": 1.1620253164556962,
+      "grad_norm": 0.3792484700679779,
+      "learning_rate": 0.00019257295755412932,
+      "loss": 0.7817611694335938,
+      "step": 2754
+    },
+    {
+      "epoch": 1.1628691983122363,
+      "grad_norm": 0.41246429085731506,
+      "learning_rate": 0.00019255478288438983,
+      "loss": 0.6904958486557007,
+      "step": 2756
+    },
+    {
+      "epoch": 1.1637130801687763,
+      "grad_norm": 0.44368481636047363,
+      "learning_rate": 0.00019253658686403948,
+      "loss": 0.7571630477905273,
+      "step": 2758
+    },
+    {
+      "epoch": 1.1645569620253164,
+      "grad_norm": 0.40121781826019287,
+      "learning_rate": 0.00019251836949727574,
+      "loss": 0.6747061014175415,
+      "step": 2760
+    },
+    {
+      "epoch": 1.1654008438818566,
+      "grad_norm": 0.4660188853740692,
+      "learning_rate": 0.000192500130788301,
+      "loss": 0.7722339034080505,
+      "step": 2762
+    },
+    {
+      "epoch": 1.1662447257383965,
+      "grad_norm": 0.41693782806396484,
+      "learning_rate": 0.0001924818707413226,
+      "loss": 0.8081457018852234,
+      "step": 2764
+    },
+    {
+      "epoch": 1.1670886075949367,
+      "grad_norm": 0.4274570345878601,
+      "learning_rate": 0.00019246358936055274,
+      "loss": 0.7745853066444397,
+      "step": 2766
+    },
+    {
+      "epoch": 1.1679324894514769,
+      "grad_norm": 0.4385969638824463,
+      "learning_rate": 0.00019244528665020863,
+      "loss": 0.7918146252632141,
+      "step": 2768
+    },
+    {
+      "epoch": 1.1687763713080168,
+      "grad_norm": 0.42832866311073303,
+      "learning_rate": 0.00019242696261451234,
+      "loss": 0.7382830381393433,
+      "step": 2770
+    },
+    {
+      "epoch": 1.169620253164557,
+      "grad_norm": 0.44177982211112976,
+      "learning_rate": 0.00019240861725769085,
+      "loss": 0.7501721978187561,
+      "step": 2772
+    },
+    {
+      "epoch": 1.1704641350210971,
+      "grad_norm": 0.4193457365036011,
+      "learning_rate": 0.00019239025058397611,
+      "loss": 0.6495388150215149,
+      "step": 2774
+    },
+    {
+      "epoch": 1.171308016877637,
+      "grad_norm": 0.5145925283432007,
+      "learning_rate": 0.000192371862597605,
+      "loss": 0.8291991353034973,
+      "step": 2776
+    },
+    {
+      "epoch": 1.1721518987341772,
+      "grad_norm": 0.40911251306533813,
+      "learning_rate": 0.00019235345330281914,
+      "loss": 0.7491340637207031,
+      "step": 2778
+    },
+    {
+      "epoch": 1.1729957805907172,
+      "grad_norm": 0.3920276165008545,
+      "learning_rate": 0.0001923350227038653,
+      "loss": 0.6315004825592041,
+      "step": 2780
+    },
+    {
+      "epoch": 1.1738396624472573,
+      "grad_norm": 0.442303329706192,
+      "learning_rate": 0.0001923165708049951,
+      "loss": 0.7538990378379822,
+      "step": 2782
+    },
+    {
+      "epoch": 1.1746835443037975,
+      "grad_norm": 0.4693523049354553,
+      "learning_rate": 0.00019229809761046497,
+      "loss": 0.7440794110298157,
+      "step": 2784
+    },
+    {
+      "epoch": 1.1755274261603375,
+      "grad_norm": 0.4948059618473053,
+      "learning_rate": 0.0001922796031245363,
+      "loss": 0.7342568635940552,
+      "step": 2786
+    },
+    {
+      "epoch": 1.1763713080168776,
+      "grad_norm": 0.3900465965270996,
+      "learning_rate": 0.00019226108735147547,
+      "loss": 0.6969661116600037,
+      "step": 2788
+    },
+    {
+      "epoch": 1.1772151898734178,
+      "grad_norm": 0.36153602600097656,
+      "learning_rate": 0.00019224255029555372,
+      "loss": 0.7097541689872742,
+      "step": 2790
+    },
+    {
+      "epoch": 1.1780590717299577,
+      "grad_norm": 0.41168832778930664,
+      "learning_rate": 0.00019222399196104715,
+      "loss": 0.8141831755638123,
+      "step": 2792
+    },
+    {
+      "epoch": 1.1789029535864979,
+      "grad_norm": 0.4542821943759918,
+      "learning_rate": 0.0001922054123522368,
+      "loss": 0.7458863258361816,
+      "step": 2794
+    },
+    {
+      "epoch": 1.179746835443038,
+      "grad_norm": 0.4026047885417938,
+      "learning_rate": 0.0001921868114734087,
+      "loss": 0.6535642147064209,
+      "step": 2796
+    },
+    {
+      "epoch": 1.180590717299578,
+      "grad_norm": 0.4417034387588501,
+      "learning_rate": 0.0001921681893288537,
+      "loss": 0.6248135566711426,
+      "step": 2798
+    },
+    {
+      "epoch": 1.1814345991561181,
+      "grad_norm": 0.45027074217796326,
+      "learning_rate": 0.00019214954592286748,
+      "loss": 0.6777785420417786,
+      "step": 2800
+    },
+    {
+      "epoch": 1.1814345991561181,
+      "eval_loss": 0.8124309182167053,
+      "eval_runtime": 688.4759,
+      "eval_samples_per_second": 3.06,
+      "eval_steps_per_second": 3.06,
+      "step": 2800
+    },
+    {
+      "epoch": 1.1822784810126583,
+      "grad_norm": 0.406865656375885,
+      "learning_rate": 0.00019213088125975082,
+      "loss": 0.6460301876068115,
+      "step": 2802
+    },
+    {
+      "epoch": 1.1831223628691983,
+      "grad_norm": 0.5052749514579773,
+      "learning_rate": 0.00019211219534380928,
+      "loss": 0.7292830944061279,
+      "step": 2804
+    },
+    {
+      "epoch": 1.1839662447257384,
+      "grad_norm": 0.4028380513191223,
+      "learning_rate": 0.00019209348817935328,
+      "loss": 0.7399409413337708,
+      "step": 2806
+    },
+    {
+      "epoch": 1.1848101265822786,
+      "grad_norm": 0.4522664248943329,
+      "learning_rate": 0.00019207475977069826,
+      "loss": 0.6885398030281067,
+      "step": 2808
+    },
+    {
+      "epoch": 1.1856540084388185,
+      "grad_norm": 0.419546514749527,
+      "learning_rate": 0.0001920560101221645,
+      "loss": 0.6962382197380066,
+      "step": 2810
+    },
+    {
+      "epoch": 1.1864978902953587,
+      "grad_norm": 0.48912352323532104,
+      "learning_rate": 0.00019203723923807715,
+      "loss": 0.7380965352058411,
+      "step": 2812
+    },
+    {
+      "epoch": 1.1873417721518988,
+      "grad_norm": 0.40905487537384033,
+      "learning_rate": 0.00019201844712276635,
+      "loss": 0.755382776260376,
+      "step": 2814
+    },
+    {
+      "epoch": 1.1881856540084388,
+      "grad_norm": 0.385623037815094,
+      "learning_rate": 0.000191999633780567,
+      "loss": 0.7066168189048767,
+      "step": 2816
+    },
+    {
+      "epoch": 1.189029535864979,
+      "grad_norm": 0.4488723576068878,
+      "learning_rate": 0.00019198079921581908,
+      "loss": 0.7172947525978088,
+      "step": 2818
+    },
+    {
+      "epoch": 1.189873417721519,
+      "grad_norm": 0.385277658700943,
+      "learning_rate": 0.00019196194343286727,
+      "loss": 0.6714195013046265,
+      "step": 2820
+    },
+    {
+      "epoch": 1.190717299578059,
+      "grad_norm": 0.4280855655670166,
+      "learning_rate": 0.00019194306643606127,
+      "loss": 0.7076966166496277,
+      "step": 2822
+    },
+    {
+      "epoch": 1.1915611814345992,
+      "grad_norm": 0.4626067578792572,
+      "learning_rate": 0.00019192416822975567,
+      "loss": 0.7926747798919678,
+      "step": 2824
+    },
+    {
+      "epoch": 1.1924050632911392,
+      "grad_norm": 0.3994208574295044,
+      "learning_rate": 0.00019190524881830987,
+      "loss": 0.7654290199279785,
+      "step": 2826
+    },
+    {
+      "epoch": 1.1932489451476793,
+      "grad_norm": 0.47346508502960205,
+      "learning_rate": 0.00019188630820608825,
+      "loss": 0.7857998609542847,
+      "step": 2828
+    },
+    {
+      "epoch": 1.1940928270042195,
+      "grad_norm": 0.446077823638916,
+      "learning_rate": 0.00019186734639746003,
+      "loss": 0.7603738903999329,
+      "step": 2830
+    },
+    {
+      "epoch": 1.1949367088607594,
+      "grad_norm": 0.41503727436065674,
+      "learning_rate": 0.00019184836339679936,
+      "loss": 0.790582537651062,
+      "step": 2832
+    },
+    {
+      "epoch": 1.1957805907172996,
+      "grad_norm": 0.43651700019836426,
+      "learning_rate": 0.00019182935920848522,
+      "loss": 0.7229864597320557,
+      "step": 2834
+    },
+    {
+      "epoch": 1.1966244725738397,
+      "grad_norm": 0.4409031867980957,
+      "learning_rate": 0.00019181033383690153,
+      "loss": 0.7817287445068359,
+      "step": 2836
+    },
+    {
+      "epoch": 1.1974683544303797,
+      "grad_norm": 0.3562015891075134,
+      "learning_rate": 0.0001917912872864371,
+      "loss": 0.6342297792434692,
+      "step": 2838
+    },
+    {
+      "epoch": 1.1983122362869199,
+      "grad_norm": 0.4767192602157593,
+      "learning_rate": 0.00019177221956148555,
+      "loss": 0.7065272927284241,
+      "step": 2840
+    },
+    {
+      "epoch": 1.1991561181434598,
+      "grad_norm": 0.45131945610046387,
+      "learning_rate": 0.00019175313066644546,
+      "loss": 0.777110755443573,
+      "step": 2842
+    },
+    {
+      "epoch": 1.2,
+      "grad_norm": 0.47566652297973633,
+      "learning_rate": 0.00019173402060572027,
+      "loss": 0.7456457018852234,
+      "step": 2844
+    },
+    {
+      "epoch": 1.2008438818565401,
+      "grad_norm": 0.48909857869148254,
+      "learning_rate": 0.00019171488938371834,
+      "loss": 0.8041547536849976,
+      "step": 2846
+    },
+    {
+      "epoch": 1.20168776371308,
+      "grad_norm": 0.436591774225235,
+      "learning_rate": 0.00019169573700485284,
+      "loss": 0.6786676049232483,
+      "step": 2848
+    },
+    {
+      "epoch": 1.2025316455696202,
+      "grad_norm": 0.44230160117149353,
+      "learning_rate": 0.00019167656347354183,
+      "loss": 0.7661880850791931,
+      "step": 2850
+    },
+    {
+      "epoch": 1.2033755274261604,
+      "grad_norm": 0.4844048321247101,
+      "learning_rate": 0.00019165736879420836,
+      "loss": 0.7369645833969116,
+      "step": 2852
+    },
+    {
+      "epoch": 1.2042194092827003,
+      "grad_norm": 0.4272955060005188,
+      "learning_rate": 0.0001916381529712802,
+      "loss": 0.7776681184768677,
+      "step": 2854
+    },
+    {
+      "epoch": 1.2050632911392405,
+      "grad_norm": 0.42392730712890625,
+      "learning_rate": 0.00019161891600919008,
+      "loss": 0.8019667267799377,
+      "step": 2856
+    },
+    {
+      "epoch": 1.2059071729957807,
+      "grad_norm": 0.4528326988220215,
+      "learning_rate": 0.00019159965791237565,
+      "loss": 0.7811952233314514,
+      "step": 2858
+    },
+    {
+      "epoch": 1.2067510548523206,
+      "grad_norm": 0.42840948700904846,
+      "learning_rate": 0.00019158037868527932,
+      "loss": 0.7244279980659485,
+      "step": 2860
+    },
+    {
+      "epoch": 1.2075949367088608,
+      "grad_norm": 0.4291226863861084,
+      "learning_rate": 0.00019156107833234845,
+      "loss": 0.7675760984420776,
+      "step": 2862
+    },
+    {
+      "epoch": 1.208438818565401,
+      "grad_norm": 0.4455093443393707,
+      "learning_rate": 0.00019154175685803527,
+      "loss": 0.7071483135223389,
+      "step": 2864
+    },
+    {
+      "epoch": 1.2092827004219409,
+      "grad_norm": 0.41356268525123596,
+      "learning_rate": 0.00019152241426679688,
+      "loss": 0.7554603815078735,
+      "step": 2866
+    },
+    {
+      "epoch": 1.210126582278481,
+      "grad_norm": 0.43875646591186523,
+      "learning_rate": 0.00019150305056309525,
+      "loss": 0.7198356986045837,
+      "step": 2868
+    },
+    {
+      "epoch": 1.2109704641350212,
+      "grad_norm": 0.454356849193573,
+      "learning_rate": 0.0001914836657513972,
+      "loss": 0.7393195629119873,
+      "step": 2870
+    },
+    {
+      "epoch": 1.2118143459915611,
+      "grad_norm": 0.37668535113334656,
+      "learning_rate": 0.00019146425983617444,
+      "loss": 0.7277671694755554,
+      "step": 2872
+    },
+    {
+      "epoch": 1.2126582278481013,
+      "grad_norm": 0.37869662046432495,
+      "learning_rate": 0.00019144483282190354,
+      "loss": 0.6783158183097839,
+      "step": 2874
+    },
+    {
+      "epoch": 1.2135021097046415,
+      "grad_norm": 0.4333525002002716,
+      "learning_rate": 0.00019142538471306596,
+      "loss": 0.7024772763252258,
+      "step": 2876
+    },
+    {
+      "epoch": 1.2143459915611814,
+      "grad_norm": 0.4693842828273773,
+      "learning_rate": 0.00019140591551414796,
+      "loss": 0.7387984395027161,
+      "step": 2878
+    },
+    {
+      "epoch": 1.2151898734177216,
+      "grad_norm": 0.432640939950943,
+      "learning_rate": 0.00019138642522964074,
+      "loss": 0.7597174644470215,
+      "step": 2880
+    },
+    {
+      "epoch": 1.2160337552742617,
+      "grad_norm": 0.432477205991745,
+      "learning_rate": 0.00019136691386404032,
+      "loss": 0.7884587049484253,
+      "step": 2882
+    },
+    {
+      "epoch": 1.2168776371308017,
+      "grad_norm": 0.39305904507637024,
+      "learning_rate": 0.00019134738142184763,
+      "loss": 0.6873269081115723,
+      "step": 2884
+    },
+    {
+      "epoch": 1.2177215189873418,
+      "grad_norm": 0.4485352635383606,
+      "learning_rate": 0.00019132782790756838,
+      "loss": 0.6523491144180298,
+      "step": 2886
+    },
+    {
+      "epoch": 1.2185654008438818,
+      "grad_norm": 0.4480868875980377,
+      "learning_rate": 0.00019130825332571322,
+      "loss": 0.7383941411972046,
+      "step": 2888
+    },
+    {
+      "epoch": 1.219409282700422,
+      "grad_norm": 0.44085049629211426,
+      "learning_rate": 0.00019128865768079764,
+      "loss": 0.7391335964202881,
+      "step": 2890
+    },
+    {
+      "epoch": 1.220253164556962,
+      "grad_norm": 0.4621105492115021,
+      "learning_rate": 0.00019126904097734198,
+      "loss": 0.7282017469406128,
+      "step": 2892
+    },
+    {
+      "epoch": 1.221097046413502,
+      "grad_norm": 0.41478198766708374,
+      "learning_rate": 0.00019124940321987136,
+      "loss": 0.7554038763046265,
+      "step": 2894
+    },
+    {
+      "epoch": 1.2219409282700422,
+      "grad_norm": 0.4143751263618469,
+      "learning_rate": 0.00019122974441291595,
+      "loss": 0.6826799511909485,
+      "step": 2896
+    },
+    {
+      "epoch": 1.2227848101265824,
+      "grad_norm": 0.45392894744873047,
+      "learning_rate": 0.00019121006456101057,
+      "loss": 0.751000702381134,
+      "step": 2898
+    },
+    {
+      "epoch": 1.2236286919831223,
+      "grad_norm": 0.46946296095848083,
+      "learning_rate": 0.00019119036366869506,
+      "loss": 0.7598516941070557,
+      "step": 2900
+    },
+    {
+      "epoch": 1.2236286919831223,
+      "eval_loss": 0.8077136278152466,
+      "eval_runtime": 685.6042,
+      "eval_samples_per_second": 3.073,
+      "eval_steps_per_second": 3.073,
+      "step": 2900
+    },
+    {
+      "epoch": 1.2244725738396625,
+      "grad_norm": 0.4316656291484833,
+      "learning_rate": 0.00019117064174051395,
+      "loss": 0.7518358826637268,
+      "step": 2902
+    },
+    {
+      "epoch": 1.2253164556962026,
+      "grad_norm": 0.47737377882003784,
+      "learning_rate": 0.0001911508987810168,
+      "loss": 0.7308215498924255,
+      "step": 2904
+    },
+    {
+      "epoch": 1.2261603375527426,
+      "grad_norm": 0.40906062722206116,
+      "learning_rate": 0.00019113113479475783,
+      "loss": 0.6996226906776428,
+      "step": 2906
+    },
+    {
+      "epoch": 1.2270042194092827,
+      "grad_norm": 0.517085075378418,
+      "learning_rate": 0.00019111134978629633,
+      "loss": 0.8364415168762207,
+      "step": 2908
+    },
+    {
+      "epoch": 1.2278481012658227,
+      "grad_norm": 0.425581693649292,
+      "learning_rate": 0.00019109154376019624,
+      "loss": 0.722011923789978,
+      "step": 2910
+    },
+    {
+      "epoch": 1.2286919831223628,
+      "grad_norm": 0.4651593565940857,
+      "learning_rate": 0.00019107171672102643,
+      "loss": 0.7586172223091125,
+      "step": 2912
+    },
+    {
+      "epoch": 1.229535864978903,
+      "grad_norm": 0.4578891396522522,
+      "learning_rate": 0.00019105186867336067,
+      "loss": 0.7387955188751221,
+      "step": 2914
+    },
+    {
+      "epoch": 1.230379746835443,
+      "grad_norm": 0.4312961995601654,
+      "learning_rate": 0.00019103199962177748,
+      "loss": 0.8036285638809204,
+      "step": 2916
+    },
+    {
+      "epoch": 1.231223628691983,
+      "grad_norm": 0.4355425536632538,
+      "learning_rate": 0.00019101210957086026,
+      "loss": 0.7356550097465515,
+      "step": 2918
+    },
+    {
+      "epoch": 1.2320675105485233,
+      "grad_norm": 0.424675852060318,
+      "learning_rate": 0.00019099219852519733,
+      "loss": 0.6730139255523682,
+      "step": 2920
+    },
+    {
+      "epoch": 1.2329113924050632,
+      "grad_norm": 0.41102367639541626,
+      "learning_rate": 0.0001909722664893817,
+      "loss": 0.6983892321586609,
+      "step": 2922
+    },
+    {
+      "epoch": 1.2337552742616034,
+      "grad_norm": 0.4088250398635864,
+      "learning_rate": 0.00019095231346801132,
+      "loss": 0.6297144889831543,
+      "step": 2924
+    },
+    {
+      "epoch": 1.2345991561181435,
+      "grad_norm": 0.3685890734195709,
+      "learning_rate": 0.00019093233946568906,
+      "loss": 0.7035690546035767,
+      "step": 2926
+    },
+    {
+      "epoch": 1.2354430379746835,
+      "grad_norm": 0.39438629150390625,
+      "learning_rate": 0.0001909123444870224,
+      "loss": 0.820853054523468,
+      "step": 2928
+    },
+    {
+      "epoch": 1.2362869198312236,
+      "grad_norm": 0.5041676163673401,
+      "learning_rate": 0.00019089232853662392,
+      "loss": 0.7884082794189453,
+      "step": 2930
+    },
+    {
+      "epoch": 1.2371308016877638,
+      "grad_norm": 0.40592801570892334,
+      "learning_rate": 0.00019087229161911086,
+      "loss": 0.773978590965271,
+      "step": 2932
+    },
+    {
+      "epoch": 1.2379746835443037,
+      "grad_norm": 0.41828030347824097,
+      "learning_rate": 0.00019085223373910536,
+      "loss": 0.6742141842842102,
+      "step": 2934
+    },
+    {
+      "epoch": 1.238818565400844,
+      "grad_norm": 0.39984026551246643,
+      "learning_rate": 0.0001908321549012344,
+      "loss": 0.7106125950813293,
+      "step": 2936
+    },
+    {
+      "epoch": 1.239662447257384,
+      "grad_norm": 0.4841296970844269,
+      "learning_rate": 0.00019081205511012973,
+      "loss": 0.7390915751457214,
+      "step": 2938
+    },
+    {
+      "epoch": 1.240506329113924,
+      "grad_norm": 0.39837637543678284,
+      "learning_rate": 0.00019079193437042806,
+      "loss": 0.6527198553085327,
+      "step": 2940
+    },
+    {
+      "epoch": 1.2413502109704642,
+      "grad_norm": 0.40916863083839417,
+      "learning_rate": 0.0001907717926867708,
+      "loss": 0.6939712762832642,
+      "step": 2942
+    },
+    {
+      "epoch": 1.2421940928270043,
+      "grad_norm": 0.4329306483268738,
+      "learning_rate": 0.0001907516300638043,
+      "loss": 0.7566131353378296,
+      "step": 2944
+    },
+    {
+      "epoch": 1.2430379746835443,
+      "grad_norm": 0.47983601689338684,
+      "learning_rate": 0.00019073144650617963,
+      "loss": 0.7893748879432678,
+      "step": 2946
+    },
+    {
+      "epoch": 1.2438818565400844,
+      "grad_norm": 0.4368246793746948,
+      "learning_rate": 0.0001907112420185528,
+      "loss": 0.8259525299072266,
+      "step": 2948
+    },
+    {
+      "epoch": 1.2447257383966246,
+      "grad_norm": 0.41272372007369995,
+      "learning_rate": 0.0001906910166055846,
+      "loss": 0.734692394733429,
+      "step": 2950
+    },
+    {
+      "epoch": 1.2455696202531645,
+      "grad_norm": 0.36929839849472046,
+      "learning_rate": 0.00019067077027194057,
+      "loss": 0.7091750502586365,
+      "step": 2952
+    },
+    {
+      "epoch": 1.2464135021097047,
+      "grad_norm": 0.4395252466201782,
+      "learning_rate": 0.00019065050302229125,
+      "loss": 0.6960210204124451,
+      "step": 2954
+    },
+    {
+      "epoch": 1.2472573839662446,
+      "grad_norm": 0.46784868836402893,
+      "learning_rate": 0.00019063021486131186,
+      "loss": 0.7817115187644958,
+      "step": 2956
+    },
+    {
+      "epoch": 1.2481012658227848,
+      "grad_norm": 0.4191588759422302,
+      "learning_rate": 0.00019060990579368244,
+      "loss": 0.7933676242828369,
+      "step": 2958
+    },
+    {
+      "epoch": 1.248945147679325,
+      "grad_norm": 0.3967524468898773,
+      "learning_rate": 0.000190589575824088,
+      "loss": 0.6856688857078552,
+      "step": 2960
+    },
+    {
+      "epoch": 1.249789029535865,
+      "grad_norm": 0.40346091985702515,
+      "learning_rate": 0.0001905692249572182,
+      "loss": 0.6783818602561951,
+      "step": 2962
+    },
+    {
+      "epoch": 1.250632911392405,
+      "grad_norm": 0.38349980115890503,
+      "learning_rate": 0.0001905488531977676,
+      "loss": 0.6997019052505493,
+      "step": 2964
+    },
+    {
+      "epoch": 1.2514767932489452,
+      "grad_norm": 0.44021397829055786,
+      "learning_rate": 0.0001905284605504356,
+      "loss": 0.6978800892829895,
+      "step": 2966
+    },
+    {
+      "epoch": 1.2523206751054852,
+      "grad_norm": 0.47491583228111267,
+      "learning_rate": 0.0001905080470199264,
+      "loss": 0.7007693648338318,
+      "step": 2968
+    },
+    {
+      "epoch": 1.2531645569620253,
+      "grad_norm": 0.4011451005935669,
+      "learning_rate": 0.00019048761261094898,
+      "loss": 0.7258896231651306,
+      "step": 2970
+    },
+    {
+      "epoch": 1.2540084388185653,
+      "grad_norm": 0.44206640124320984,
+      "learning_rate": 0.0001904671573282172,
+      "loss": 0.6746299266815186,
+      "step": 2972
+    },
+    {
+      "epoch": 1.2548523206751054,
+      "grad_norm": 0.4532456696033478,
+      "learning_rate": 0.00019044668117644965,
+      "loss": 0.7638527154922485,
+      "step": 2974
+    },
+    {
+      "epoch": 1.2556962025316456,
+      "grad_norm": 0.44301095604896545,
+      "learning_rate": 0.00019042618416036984,
+      "loss": 0.659598171710968,
+      "step": 2976
+    },
+    {
+      "epoch": 1.2565400843881855,
+      "grad_norm": 0.37427952885627747,
+      "learning_rate": 0.00019040566628470597,
+      "loss": 0.7132709622383118,
+      "step": 2978
+    },
+    {
+      "epoch": 1.2573839662447257,
+      "grad_norm": 0.4195408821105957,
+      "learning_rate": 0.0001903851275541912,
+      "loss": 0.8163022398948669,
+      "step": 2980
+    },
+    {
+      "epoch": 1.2582278481012659,
+      "grad_norm": 0.3951880633831024,
+      "learning_rate": 0.00019036456797356336,
+      "loss": 0.7481505870819092,
+      "step": 2982
+    },
+    {
+      "epoch": 1.2590717299578058,
+      "grad_norm": 0.4239100217819214,
+      "learning_rate": 0.00019034398754756522,
+      "loss": 0.750126838684082,
+      "step": 2984
+    },
+    {
+      "epoch": 1.259915611814346,
+      "grad_norm": 0.4209354817867279,
+      "learning_rate": 0.0001903233862809442,
+      "loss": 0.746433436870575,
+      "step": 2986
+    },
+    {
+      "epoch": 1.2607594936708861,
+      "grad_norm": 0.4188138544559479,
+      "learning_rate": 0.00019030276417845266,
+      "loss": 0.7104454040527344,
+      "step": 2988
+    },
+    {
+      "epoch": 1.261603375527426,
+      "grad_norm": 0.4531371593475342,
+      "learning_rate": 0.00019028212124484772,
+      "loss": 0.650738000869751,
+      "step": 2990
+    },
+    {
+      "epoch": 1.2624472573839662,
+      "grad_norm": 0.4672204256057739,
+      "learning_rate": 0.00019026145748489133,
+      "loss": 0.6870551109313965,
+      "step": 2992
+    },
+    {
+      "epoch": 1.2632911392405064,
+      "grad_norm": 0.40681278705596924,
+      "learning_rate": 0.00019024077290335018,
+      "loss": 0.6879403591156006,
+      "step": 2994
+    },
+    {
+      "epoch": 1.2641350210970463,
+      "grad_norm": 0.4491691291332245,
+      "learning_rate": 0.00019022006750499585,
+      "loss": 0.6886372566223145,
+      "step": 2996
+    },
+    {
+      "epoch": 1.2649789029535865,
+      "grad_norm": 0.4144807457923889,
+      "learning_rate": 0.0001901993412946046,
+      "loss": 0.7199324369430542,
+      "step": 2998
+    },
+    {
+      "epoch": 1.2658227848101267,
+      "grad_norm": 0.40370482206344604,
+      "learning_rate": 0.0001901785942769577,
+      "loss": 0.7105490565299988,
+      "step": 3000
+    },
+    {
+      "epoch": 1.2658227848101267,
+      "eval_loss": 0.8033165335655212,
+      "eval_runtime": 678.7554,
+      "eval_samples_per_second": 3.104,
+      "eval_steps_per_second": 3.104,
+      "step": 3000
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 14220,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 6,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 5,
+        "early_stopping_threshold": 0.001
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 0
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.7946573294477066e+18,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-3000/training_args.bin b/sft_qwen_14B_v2/checkpoints/checkpoint-3000/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..0bd5f45757bdc6c7fc2d67718b18675b51983e0e
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-3000/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:afae4a8ce1391f149eb00b14eed8f891c715e892ea546bf754d22db2c2bc5969
+size 4792
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-3500/README.md b/sft_qwen_14B_v2/checkpoints/checkpoint-3500/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..134a3a8531b68adda3fed2a98db824f7c1b7dcfd
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-3500/README.md
@@ -0,0 +1,207 @@
+---
+base_model: Models/Qwen2.5-Coder-14B-CPT
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Models/Qwen2.5-Coder-14B-CPT
+- lora
+- transformers
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.0
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-3500/adapter_config.json b/sft_qwen_14B_v2/checkpoints/checkpoint-3500/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..d6316742bc71d30704e2ccdb9080ed0b8062e454
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-3500/adapter_config.json
@@ -0,0 +1,43 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Models/Qwen2.5-Coder-14B-CPT",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.0",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "v_proj",
+    "k_proj",
+    "q_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-3500/adapter_model.safetensors b/sft_qwen_14B_v2/checkpoints/checkpoint-3500/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..fc5fb88e1c43ae3a7ae7ba2a5ddc6ac05e57f5f8
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-3500/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3197e2bedd953dc5715517e86b05e3920fa3d94bd95cca48ae271d35832d38a4
+size 201378736
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-3500/optimizer.pt b/sft_qwen_14B_v2/checkpoints/checkpoint-3500/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..c6e4e5533bce50c38f3e5208adab3fcd6c529fd1
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-3500/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7d76ed9c500b950c64503cd5076327e5834eed723bf3f4de15ff79d8e544b02
+size 402976786
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-3500/rng_state.pth b/sft_qwen_14B_v2/checkpoints/checkpoint-3500/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..996d401ae2ddb28e590144cd2b904af0076d6429
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-3500/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93eebadcd407d6d286a888a8799c348021b397445be0c00df5150b9c54d0e62b
+size 14244
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-3500/scheduler.pt b/sft_qwen_14B_v2/checkpoints/checkpoint-3500/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..6b0f7d26bdea5d4b70097a86362a053edbb7cc83
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-3500/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ca6c2774893895cc7dc1d3185e861ea38b728800e8555122b7afb33495b01da
+size 1064
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-3500/trainer_state.json b/sft_qwen_14B_v2/checkpoints/checkpoint-3500/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..fae568e15a3b7691dc08ce58535b1470cd92326f
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-3500/trainer_state.json
@@ -0,0 +1,12573 @@
+{
+  "best_global_step": 3500,
+  "best_metric": 0.7902651429176331,
+  "best_model_checkpoint": "task2file/sft_qwen_14B_v2/checkpoints/checkpoint-3500",
+  "epoch": 1.4767932489451476,
+  "eval_steps": 100,
+  "global_step": 3500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0008438818565400844,
+      "grad_norm": 0.5386583805084229,
+      "learning_rate": 1.7574692442882248e-07,
+      "loss": 1.6941628456115723,
+      "step": 2
+    },
+    {
+      "epoch": 0.0016877637130801688,
+      "grad_norm": 0.5477277636528015,
+      "learning_rate": 5.272407732864675e-07,
+      "loss": 1.7132279872894287,
+      "step": 4
+    },
+    {
+      "epoch": 0.002531645569620253,
+      "grad_norm": 0.5390765070915222,
+      "learning_rate": 8.787346221441126e-07,
+      "loss": 1.641180396080017,
+      "step": 6
+    },
+    {
+      "epoch": 0.0033755274261603376,
+      "grad_norm": 0.5023683905601501,
+      "learning_rate": 1.2302284710017575e-06,
+      "loss": 1.5616240501403809,
+      "step": 8
+    },
+    {
+      "epoch": 0.004219409282700422,
+      "grad_norm": 0.4899154603481293,
+      "learning_rate": 1.5817223198594026e-06,
+      "loss": 1.572033405303955,
+      "step": 10
+    },
+    {
+      "epoch": 0.005063291139240506,
+      "grad_norm": 0.5239788293838501,
+      "learning_rate": 1.9332161687170474e-06,
+      "loss": 1.6242921352386475,
+      "step": 12
+    },
+    {
+      "epoch": 0.00590717299578059,
+      "grad_norm": 0.5172926783561707,
+      "learning_rate": 2.2847100175746925e-06,
+      "loss": 1.6800041198730469,
+      "step": 14
+    },
+    {
+      "epoch": 0.006751054852320675,
+      "grad_norm": 0.5539224743843079,
+      "learning_rate": 2.6362038664323376e-06,
+      "loss": 1.6450834274291992,
+      "step": 16
+    },
+    {
+      "epoch": 0.007594936708860759,
+      "grad_norm": 0.5255337953567505,
+      "learning_rate": 2.9876977152899827e-06,
+      "loss": 1.6673263311386108,
+      "step": 18
+    },
+    {
+      "epoch": 0.008438818565400843,
+      "grad_norm": 0.5074548721313477,
+      "learning_rate": 3.3391915641476277e-06,
+      "loss": 1.531802773475647,
+      "step": 20
+    },
+    {
+      "epoch": 0.009282700421940928,
+      "grad_norm": 0.4160279333591461,
+      "learning_rate": 3.6906854130052724e-06,
+      "loss": 1.599354863166809,
+      "step": 22
+    },
+    {
+      "epoch": 0.010126582278481013,
+      "grad_norm": 0.5716474652290344,
+      "learning_rate": 4.0421792618629174e-06,
+      "loss": 1.6700962781906128,
+      "step": 24
+    },
+    {
+      "epoch": 0.010970464135021098,
+      "grad_norm": 0.5148899555206299,
+      "learning_rate": 4.3936731107205625e-06,
+      "loss": 1.66217839717865,
+      "step": 26
+    },
+    {
+      "epoch": 0.01181434599156118,
+      "grad_norm": 0.575722336769104,
+      "learning_rate": 4.7451669595782076e-06,
+      "loss": 1.6692266464233398,
+      "step": 28
+    },
+    {
+      "epoch": 0.012658227848101266,
+      "grad_norm": 0.5345953106880188,
+      "learning_rate": 5.096660808435853e-06,
+      "loss": 1.5518689155578613,
+      "step": 30
+    },
+    {
+      "epoch": 0.01350210970464135,
+      "grad_norm": 0.4462043344974518,
+      "learning_rate": 5.448154657293498e-06,
+      "loss": 1.5930007696151733,
+      "step": 32
+    },
+    {
+      "epoch": 0.014345991561181435,
+      "grad_norm": 0.5119605660438538,
+      "learning_rate": 5.799648506151143e-06,
+      "loss": 1.6069684028625488,
+      "step": 34
+    },
+    {
+      "epoch": 0.015189873417721518,
+      "grad_norm": 0.5328608751296997,
+      "learning_rate": 6.151142355008788e-06,
+      "loss": 1.5838109254837036,
+      "step": 36
+    },
+    {
+      "epoch": 0.016033755274261603,
+      "grad_norm": 0.5065920352935791,
+      "learning_rate": 6.502636203866433e-06,
+      "loss": 1.608130931854248,
+      "step": 38
+    },
+    {
+      "epoch": 0.016877637130801686,
+      "grad_norm": 0.4479359984397888,
+      "learning_rate": 6.854130052724078e-06,
+      "loss": 1.5942182540893555,
+      "step": 40
+    },
+    {
+      "epoch": 0.017721518987341773,
+      "grad_norm": 0.42844903469085693,
+      "learning_rate": 7.205623901581722e-06,
+      "loss": 1.6441553831100464,
+      "step": 42
+    },
+    {
+      "epoch": 0.018565400843881856,
+      "grad_norm": 0.476630836725235,
+      "learning_rate": 7.557117750439367e-06,
+      "loss": 1.6068111658096313,
+      "step": 44
+    },
+    {
+      "epoch": 0.019409282700421943,
+      "grad_norm": 0.4532654881477356,
+      "learning_rate": 7.908611599297012e-06,
+      "loss": 1.6618021726608276,
+      "step": 46
+    },
+    {
+      "epoch": 0.020253164556962026,
+      "grad_norm": 0.3701118230819702,
+      "learning_rate": 8.260105448154657e-06,
+      "loss": 1.4730033874511719,
+      "step": 48
+    },
+    {
+      "epoch": 0.02109704641350211,
+      "grad_norm": 0.38471561670303345,
+      "learning_rate": 8.611599297012302e-06,
+      "loss": 1.4828267097473145,
+      "step": 50
+    },
+    {
+      "epoch": 0.021940928270042195,
+      "grad_norm": 0.3602336347103119,
+      "learning_rate": 8.963093145869948e-06,
+      "loss": 1.3877452611923218,
+      "step": 52
+    },
+    {
+      "epoch": 0.02278481012658228,
+      "grad_norm": 0.40318572521209717,
+      "learning_rate": 9.314586994727593e-06,
+      "loss": 1.49052894115448,
+      "step": 54
+    },
+    {
+      "epoch": 0.02362869198312236,
+      "grad_norm": 0.3223826587200165,
+      "learning_rate": 9.666080843585238e-06,
+      "loss": 1.4912524223327637,
+      "step": 56
+    },
+    {
+      "epoch": 0.024472573839662448,
+      "grad_norm": 0.3873065114021301,
+      "learning_rate": 1.0017574692442883e-05,
+      "loss": 1.526674509048462,
+      "step": 58
+    },
+    {
+      "epoch": 0.02531645569620253,
+      "grad_norm": 0.410159707069397,
+      "learning_rate": 1.0369068541300528e-05,
+      "loss": 1.4480271339416504,
+      "step": 60
+    },
+    {
+      "epoch": 0.026160337552742614,
+      "grad_norm": 0.3632003962993622,
+      "learning_rate": 1.0720562390158173e-05,
+      "loss": 1.4222990274429321,
+      "step": 62
+    },
+    {
+      "epoch": 0.0270042194092827,
+      "grad_norm": 0.33118435740470886,
+      "learning_rate": 1.1072056239015818e-05,
+      "loss": 1.387171745300293,
+      "step": 64
+    },
+    {
+      "epoch": 0.027848101265822784,
+      "grad_norm": 0.3301764726638794,
+      "learning_rate": 1.1423550087873463e-05,
+      "loss": 1.3523777723312378,
+      "step": 66
+    },
+    {
+      "epoch": 0.02869198312236287,
+      "grad_norm": 0.34342435002326965,
+      "learning_rate": 1.1775043936731108e-05,
+      "loss": 1.4515162706375122,
+      "step": 68
+    },
+    {
+      "epoch": 0.029535864978902954,
+      "grad_norm": 0.3243122100830078,
+      "learning_rate": 1.2126537785588753e-05,
+      "loss": 1.3509243726730347,
+      "step": 70
+    },
+    {
+      "epoch": 0.030379746835443037,
+      "grad_norm": 0.3450150787830353,
+      "learning_rate": 1.2478031634446398e-05,
+      "loss": 1.4936245679855347,
+      "step": 72
+    },
+    {
+      "epoch": 0.031223628691983123,
+      "grad_norm": 0.38912028074264526,
+      "learning_rate": 1.2829525483304042e-05,
+      "loss": 1.3419109582901,
+      "step": 74
+    },
+    {
+      "epoch": 0.032067510548523206,
+      "grad_norm": 0.3019310235977173,
+      "learning_rate": 1.3181019332161687e-05,
+      "loss": 1.4284154176712036,
+      "step": 76
+    },
+    {
+      "epoch": 0.03291139240506329,
+      "grad_norm": 0.37803682684898376,
+      "learning_rate": 1.3532513181019332e-05,
+      "loss": 1.4256561994552612,
+      "step": 78
+    },
+    {
+      "epoch": 0.03375527426160337,
+      "grad_norm": 0.34191736578941345,
+      "learning_rate": 1.3884007029876977e-05,
+      "loss": 1.3256909847259521,
+      "step": 80
+    },
+    {
+      "epoch": 0.03459915611814346,
+      "grad_norm": 0.35242700576782227,
+      "learning_rate": 1.4235500878734624e-05,
+      "loss": 1.2710685729980469,
+      "step": 82
+    },
+    {
+      "epoch": 0.035443037974683546,
+      "grad_norm": 0.38094228506088257,
+      "learning_rate": 1.4586994727592269e-05,
+      "loss": 1.253411889076233,
+      "step": 84
+    },
+    {
+      "epoch": 0.036286919831223625,
+      "grad_norm": 0.36837366223335266,
+      "learning_rate": 1.4938488576449914e-05,
+      "loss": 1.3064342737197876,
+      "step": 86
+    },
+    {
+      "epoch": 0.03713080168776371,
+      "grad_norm": 0.3443569242954254,
+      "learning_rate": 1.5289982425307557e-05,
+      "loss": 1.293562412261963,
+      "step": 88
+    },
+    {
+      "epoch": 0.0379746835443038,
+      "grad_norm": 0.3799338936805725,
+      "learning_rate": 1.5641476274165202e-05,
+      "loss": 1.3382648229599,
+      "step": 90
+    },
+    {
+      "epoch": 0.038818565400843885,
+      "grad_norm": 0.40501922369003296,
+      "learning_rate": 1.599297012302285e-05,
+      "loss": 1.3925724029541016,
+      "step": 92
+    },
+    {
+      "epoch": 0.039662447257383965,
+      "grad_norm": 0.4419630467891693,
+      "learning_rate": 1.6344463971880492e-05,
+      "loss": 1.357171893119812,
+      "step": 94
+    },
+    {
+      "epoch": 0.04050632911392405,
+      "grad_norm": 0.3619817793369293,
+      "learning_rate": 1.6695957820738137e-05,
+      "loss": 1.3029985427856445,
+      "step": 96
+    },
+    {
+      "epoch": 0.04135021097046414,
+      "grad_norm": 0.4851357340812683,
+      "learning_rate": 1.7047451669595782e-05,
+      "loss": 1.3498191833496094,
+      "step": 98
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "grad_norm": 0.418658584356308,
+      "learning_rate": 1.7398945518453427e-05,
+      "loss": 1.185287356376648,
+      "step": 100
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "eval_loss": 1.2979938983917236,
+      "eval_runtime": 682.1979,
+      "eval_samples_per_second": 3.089,
+      "eval_steps_per_second": 3.089,
+      "step": 100
+    },
+    {
+      "epoch": 0.043037974683544304,
+      "grad_norm": 0.4464418888092041,
+      "learning_rate": 1.7750439367311073e-05,
+      "loss": 1.2217272520065308,
+      "step": 102
+    },
+    {
+      "epoch": 0.04388185654008439,
+      "grad_norm": 0.4706237316131592,
+      "learning_rate": 1.8101933216168718e-05,
+      "loss": 1.2052050828933716,
+      "step": 104
+    },
+    {
+      "epoch": 0.04472573839662447,
+      "grad_norm": 0.46394404768943787,
+      "learning_rate": 1.8453427065026363e-05,
+      "loss": 1.221343994140625,
+      "step": 106
+    },
+    {
+      "epoch": 0.04556962025316456,
+      "grad_norm": 0.4726889431476593,
+      "learning_rate": 1.8804920913884008e-05,
+      "loss": 1.2387475967407227,
+      "step": 108
+    },
+    {
+      "epoch": 0.046413502109704644,
+      "grad_norm": 0.42130985856056213,
+      "learning_rate": 1.9156414762741653e-05,
+      "loss": 1.2851309776306152,
+      "step": 110
+    },
+    {
+      "epoch": 0.04725738396624472,
+      "grad_norm": 0.4504576623439789,
+      "learning_rate": 1.9507908611599298e-05,
+      "loss": 1.2753145694732666,
+      "step": 112
+    },
+    {
+      "epoch": 0.04810126582278481,
+      "grad_norm": 0.396085262298584,
+      "learning_rate": 1.9859402460456943e-05,
+      "loss": 1.2427717447280884,
+      "step": 114
+    },
+    {
+      "epoch": 0.048945147679324896,
+      "grad_norm": 0.5106491446495056,
+      "learning_rate": 2.0210896309314588e-05,
+      "loss": 1.2943825721740723,
+      "step": 116
+    },
+    {
+      "epoch": 0.049789029535864976,
+      "grad_norm": 0.42351317405700684,
+      "learning_rate": 2.0562390158172233e-05,
+      "loss": 1.263301134109497,
+      "step": 118
+    },
+    {
+      "epoch": 0.05063291139240506,
+      "grad_norm": 0.4403539299964905,
+      "learning_rate": 2.0913884007029878e-05,
+      "loss": 1.2647849321365356,
+      "step": 120
+    },
+    {
+      "epoch": 0.05147679324894515,
+      "grad_norm": 0.5260752439498901,
+      "learning_rate": 2.1265377855887523e-05,
+      "loss": 1.2351393699645996,
+      "step": 122
+    },
+    {
+      "epoch": 0.05232067510548523,
+      "grad_norm": 0.44978851079940796,
+      "learning_rate": 2.1616871704745168e-05,
+      "loss": 1.0384471416473389,
+      "step": 124
+    },
+    {
+      "epoch": 0.053164556962025315,
+      "grad_norm": 0.47732362151145935,
+      "learning_rate": 2.1968365553602813e-05,
+      "loss": 1.1518068313598633,
+      "step": 126
+    },
+    {
+      "epoch": 0.0540084388185654,
+      "grad_norm": 0.5473551750183105,
+      "learning_rate": 2.231985940246046e-05,
+      "loss": 1.2264912128448486,
+      "step": 128
+    },
+    {
+      "epoch": 0.05485232067510549,
+      "grad_norm": 0.4473855197429657,
+      "learning_rate": 2.2671353251318103e-05,
+      "loss": 1.1615246534347534,
+      "step": 130
+    },
+    {
+      "epoch": 0.05569620253164557,
+      "grad_norm": 0.5980377197265625,
+      "learning_rate": 2.302284710017575e-05,
+      "loss": 1.1334880590438843,
+      "step": 132
+    },
+    {
+      "epoch": 0.056540084388185655,
+      "grad_norm": 0.5987792015075684,
+      "learning_rate": 2.3374340949033394e-05,
+      "loss": 1.1546804904937744,
+      "step": 134
+    },
+    {
+      "epoch": 0.05738396624472574,
+      "grad_norm": 0.45355498790740967,
+      "learning_rate": 2.372583479789104e-05,
+      "loss": 1.194953441619873,
+      "step": 136
+    },
+    {
+      "epoch": 0.05822784810126582,
+      "grad_norm": 0.5373698472976685,
+      "learning_rate": 2.4077328646748684e-05,
+      "loss": 1.1067466735839844,
+      "step": 138
+    },
+    {
+      "epoch": 0.05907172995780591,
+      "grad_norm": 0.48734328150749207,
+      "learning_rate": 2.442882249560633e-05,
+      "loss": 1.188468098640442,
+      "step": 140
+    },
+    {
+      "epoch": 0.059915611814345994,
+      "grad_norm": 0.4692173898220062,
+      "learning_rate": 2.478031634446397e-05,
+      "loss": 1.1624362468719482,
+      "step": 142
+    },
+    {
+      "epoch": 0.060759493670886074,
+      "grad_norm": 0.532554030418396,
+      "learning_rate": 2.513181019332162e-05,
+      "loss": 1.0978907346725464,
+      "step": 144
+    },
+    {
+      "epoch": 0.06160337552742616,
+      "grad_norm": 0.5853802561759949,
+      "learning_rate": 2.5483304042179264e-05,
+      "loss": 1.2030781507492065,
+      "step": 146
+    },
+    {
+      "epoch": 0.06244725738396625,
+      "grad_norm": 0.5061611533164978,
+      "learning_rate": 2.583479789103691e-05,
+      "loss": 1.082366943359375,
+      "step": 148
+    },
+    {
+      "epoch": 0.06329113924050633,
+      "grad_norm": 0.49426141381263733,
+      "learning_rate": 2.6186291739894554e-05,
+      "loss": 1.10564386844635,
+      "step": 150
+    },
+    {
+      "epoch": 0.06413502109704641,
+      "grad_norm": 0.5846618413925171,
+      "learning_rate": 2.6537785588752196e-05,
+      "loss": 1.1992807388305664,
+      "step": 152
+    },
+    {
+      "epoch": 0.06497890295358649,
+      "grad_norm": 0.5517552495002747,
+      "learning_rate": 2.6889279437609844e-05,
+      "loss": 1.1757566928863525,
+      "step": 154
+    },
+    {
+      "epoch": 0.06582278481012659,
+      "grad_norm": 0.5667305588722229,
+      "learning_rate": 2.7240773286467486e-05,
+      "loss": 1.0548783540725708,
+      "step": 156
+    },
+    {
+      "epoch": 0.06666666666666667,
+      "grad_norm": 0.6760414242744446,
+      "learning_rate": 2.7592267135325134e-05,
+      "loss": 1.184364914894104,
+      "step": 158
+    },
+    {
+      "epoch": 0.06751054852320675,
+      "grad_norm": 0.5261430740356445,
+      "learning_rate": 2.7943760984182776e-05,
+      "loss": 1.1945042610168457,
+      "step": 160
+    },
+    {
+      "epoch": 0.06835443037974684,
+      "grad_norm": 0.6155015230178833,
+      "learning_rate": 2.8295254833040425e-05,
+      "loss": 1.2021973133087158,
+      "step": 162
+    },
+    {
+      "epoch": 0.06919831223628692,
+      "grad_norm": 0.6131619215011597,
+      "learning_rate": 2.8646748681898066e-05,
+      "loss": 1.144123911857605,
+      "step": 164
+    },
+    {
+      "epoch": 0.070042194092827,
+      "grad_norm": 0.5749185681343079,
+      "learning_rate": 2.8998242530755715e-05,
+      "loss": 1.1329256296157837,
+      "step": 166
+    },
+    {
+      "epoch": 0.07088607594936709,
+      "grad_norm": 0.5243118405342102,
+      "learning_rate": 2.9349736379613356e-05,
+      "loss": 1.0892387628555298,
+      "step": 168
+    },
+    {
+      "epoch": 0.07172995780590717,
+      "grad_norm": 0.7190104722976685,
+      "learning_rate": 2.9701230228471005e-05,
+      "loss": 1.163260817527771,
+      "step": 170
+    },
+    {
+      "epoch": 0.07257383966244725,
+      "grad_norm": 0.5486982464790344,
+      "learning_rate": 3.0052724077328647e-05,
+      "loss": 1.0880777835845947,
+      "step": 172
+    },
+    {
+      "epoch": 0.07341772151898734,
+      "grad_norm": 0.5020889043807983,
+      "learning_rate": 3.0404217926186295e-05,
+      "loss": 1.0433368682861328,
+      "step": 174
+    },
+    {
+      "epoch": 0.07426160337552742,
+      "grad_norm": 0.47329774498939514,
+      "learning_rate": 3.075571177504394e-05,
+      "loss": 1.0528991222381592,
+      "step": 176
+    },
+    {
+      "epoch": 0.0751054852320675,
+      "grad_norm": 0.6635547876358032,
+      "learning_rate": 3.110720562390158e-05,
+      "loss": 1.1627811193466187,
+      "step": 178
+    },
+    {
+      "epoch": 0.0759493670886076,
+      "grad_norm": 0.5624618530273438,
+      "learning_rate": 3.145869947275923e-05,
+      "loss": 1.084869384765625,
+      "step": 180
+    },
+    {
+      "epoch": 0.07679324894514768,
+      "grad_norm": 0.6029536724090576,
+      "learning_rate": 3.181019332161687e-05,
+      "loss": 1.2227671146392822,
+      "step": 182
+    },
+    {
+      "epoch": 0.07763713080168777,
+      "grad_norm": 0.930959939956665,
+      "learning_rate": 3.216168717047452e-05,
+      "loss": 1.0955452919006348,
+      "step": 184
+    },
+    {
+      "epoch": 0.07848101265822785,
+      "grad_norm": 0.5326952338218689,
+      "learning_rate": 3.251318101933216e-05,
+      "loss": 1.0640798807144165,
+      "step": 186
+    },
+    {
+      "epoch": 0.07932489451476793,
+      "grad_norm": 0.5484727621078491,
+      "learning_rate": 3.286467486818981e-05,
+      "loss": 1.0700589418411255,
+      "step": 188
+    },
+    {
+      "epoch": 0.08016877637130802,
+      "grad_norm": 0.605273425579071,
+      "learning_rate": 3.3216168717047456e-05,
+      "loss": 1.1593081951141357,
+      "step": 190
+    },
+    {
+      "epoch": 0.0810126582278481,
+      "grad_norm": 0.5704394578933716,
+      "learning_rate": 3.35676625659051e-05,
+      "loss": 1.1617076396942139,
+      "step": 192
+    },
+    {
+      "epoch": 0.08185654008438818,
+      "grad_norm": 0.5929452180862427,
+      "learning_rate": 3.3919156414762746e-05,
+      "loss": 1.1346839666366577,
+      "step": 194
+    },
+    {
+      "epoch": 0.08270042194092828,
+      "grad_norm": 0.5624077916145325,
+      "learning_rate": 3.427065026362039e-05,
+      "loss": 1.0934710502624512,
+      "step": 196
+    },
+    {
+      "epoch": 0.08354430379746836,
+      "grad_norm": 0.6717425584793091,
+      "learning_rate": 3.4622144112478036e-05,
+      "loss": 1.1810534000396729,
+      "step": 198
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "grad_norm": 0.5120199918746948,
+      "learning_rate": 3.4973637961335674e-05,
+      "loss": 1.1525514125823975,
+      "step": 200
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "eval_loss": 1.142486810684204,
+      "eval_runtime": 668.2356,
+      "eval_samples_per_second": 3.153,
+      "eval_steps_per_second": 3.153,
+      "step": 200
+    },
+    {
+      "epoch": 0.08523206751054853,
+      "grad_norm": 0.5144487023353577,
+      "learning_rate": 3.5325131810193326e-05,
+      "loss": 1.0243735313415527,
+      "step": 202
+    },
+    {
+      "epoch": 0.08607594936708861,
+      "grad_norm": 0.6325069069862366,
+      "learning_rate": 3.5676625659050964e-05,
+      "loss": 1.118743896484375,
+      "step": 204
+    },
+    {
+      "epoch": 0.08691983122362869,
+      "grad_norm": 0.5501633882522583,
+      "learning_rate": 3.6028119507908616e-05,
+      "loss": 1.0380504131317139,
+      "step": 206
+    },
+    {
+      "epoch": 0.08776371308016878,
+      "grad_norm": 0.6133899688720703,
+      "learning_rate": 3.6379613356766254e-05,
+      "loss": 0.9837555885314941,
+      "step": 208
+    },
+    {
+      "epoch": 0.08860759493670886,
+      "grad_norm": 0.5799810886383057,
+      "learning_rate": 3.6731107205623906e-05,
+      "loss": 1.090720295906067,
+      "step": 210
+    },
+    {
+      "epoch": 0.08945147679324894,
+      "grad_norm": 0.6039511561393738,
+      "learning_rate": 3.7082601054481544e-05,
+      "loss": 1.120232343673706,
+      "step": 212
+    },
+    {
+      "epoch": 0.09029535864978903,
+      "grad_norm": 0.5983024835586548,
+      "learning_rate": 3.7434094903339196e-05,
+      "loss": 1.096949815750122,
+      "step": 214
+    },
+    {
+      "epoch": 0.09113924050632911,
+      "grad_norm": 0.5641079545021057,
+      "learning_rate": 3.7785588752196835e-05,
+      "loss": 1.1226298809051514,
+      "step": 216
+    },
+    {
+      "epoch": 0.0919831223628692,
+      "grad_norm": 0.655717134475708,
+      "learning_rate": 3.8137082601054486e-05,
+      "loss": 1.1260643005371094,
+      "step": 218
+    },
+    {
+      "epoch": 0.09282700421940929,
+      "grad_norm": 0.6111898422241211,
+      "learning_rate": 3.848857644991213e-05,
+      "loss": 1.0777709484100342,
+      "step": 220
+    },
+    {
+      "epoch": 0.09367088607594937,
+      "grad_norm": 0.6821302771568298,
+      "learning_rate": 3.884007029876977e-05,
+      "loss": 1.10588800907135,
+      "step": 222
+    },
+    {
+      "epoch": 0.09451476793248945,
+      "grad_norm": 0.693175733089447,
+      "learning_rate": 3.919156414762742e-05,
+      "loss": 1.1498671770095825,
+      "step": 224
+    },
+    {
+      "epoch": 0.09535864978902954,
+      "grad_norm": 0.5288166403770447,
+      "learning_rate": 3.954305799648506e-05,
+      "loss": 1.0587562322616577,
+      "step": 226
+    },
+    {
+      "epoch": 0.09620253164556962,
+      "grad_norm": 0.6882867813110352,
+      "learning_rate": 3.989455184534271e-05,
+      "loss": 1.1107512712478638,
+      "step": 228
+    },
+    {
+      "epoch": 0.0970464135021097,
+      "grad_norm": 0.5834154486656189,
+      "learning_rate": 4.024604569420035e-05,
+      "loss": 1.020510196685791,
+      "step": 230
+    },
+    {
+      "epoch": 0.09789029535864979,
+      "grad_norm": 0.7157064080238342,
+      "learning_rate": 4.0597539543058e-05,
+      "loss": 1.0642449855804443,
+      "step": 232
+    },
+    {
+      "epoch": 0.09873417721518987,
+      "grad_norm": 0.6530708074569702,
+      "learning_rate": 4.094903339191564e-05,
+      "loss": 1.0359872579574585,
+      "step": 234
+    },
+    {
+      "epoch": 0.09957805907172995,
+      "grad_norm": 0.6329686045646667,
+      "learning_rate": 4.130052724077329e-05,
+      "loss": 1.050504446029663,
+      "step": 236
+    },
+    {
+      "epoch": 0.10042194092827005,
+      "grad_norm": 0.6597026586532593,
+      "learning_rate": 4.165202108963093e-05,
+      "loss": 1.2621175050735474,
+      "step": 238
+    },
+    {
+      "epoch": 0.10126582278481013,
+      "grad_norm": 0.6195225119590759,
+      "learning_rate": 4.200351493848858e-05,
+      "loss": 1.1218310594558716,
+      "step": 240
+    },
+    {
+      "epoch": 0.1021097046413502,
+      "grad_norm": 0.6764137744903564,
+      "learning_rate": 4.235500878734622e-05,
+      "loss": 1.1250728368759155,
+      "step": 242
+    },
+    {
+      "epoch": 0.1029535864978903,
+      "grad_norm": 0.552363395690918,
+      "learning_rate": 4.270650263620387e-05,
+      "loss": 1.028212308883667,
+      "step": 244
+    },
+    {
+      "epoch": 0.10379746835443038,
+      "grad_norm": 0.5620495676994324,
+      "learning_rate": 4.305799648506151e-05,
+      "loss": 1.0425450801849365,
+      "step": 246
+    },
+    {
+      "epoch": 0.10464135021097046,
+      "grad_norm": 0.6860032081604004,
+      "learning_rate": 4.3409490333919156e-05,
+      "loss": 1.144278883934021,
+      "step": 248
+    },
+    {
+      "epoch": 0.10548523206751055,
+      "grad_norm": 0.6033259034156799,
+      "learning_rate": 4.37609841827768e-05,
+      "loss": 1.1223982572555542,
+      "step": 250
+    },
+    {
+      "epoch": 0.10632911392405063,
+      "grad_norm": 0.6292146444320679,
+      "learning_rate": 4.4112478031634446e-05,
+      "loss": 1.1609960794448853,
+      "step": 252
+    },
+    {
+      "epoch": 0.10717299578059072,
+      "grad_norm": 0.7982883453369141,
+      "learning_rate": 4.44639718804921e-05,
+      "loss": 1.063547968864441,
+      "step": 254
+    },
+    {
+      "epoch": 0.1080168776371308,
+      "grad_norm": 0.7719110250473022,
+      "learning_rate": 4.4815465729349736e-05,
+      "loss": 1.0719804763793945,
+      "step": 256
+    },
+    {
+      "epoch": 0.10886075949367088,
+      "grad_norm": 0.6101011633872986,
+      "learning_rate": 4.516695957820739e-05,
+      "loss": 1.0778400897979736,
+      "step": 258
+    },
+    {
+      "epoch": 0.10970464135021098,
+      "grad_norm": 0.7300994396209717,
+      "learning_rate": 4.5518453427065026e-05,
+      "loss": 1.2129558324813843,
+      "step": 260
+    },
+    {
+      "epoch": 0.11054852320675106,
+      "grad_norm": 0.8348747491836548,
+      "learning_rate": 4.586994727592268e-05,
+      "loss": 1.221714735031128,
+      "step": 262
+    },
+    {
+      "epoch": 0.11139240506329114,
+      "grad_norm": 0.5445612072944641,
+      "learning_rate": 4.6221441124780316e-05,
+      "loss": 1.0187978744506836,
+      "step": 264
+    },
+    {
+      "epoch": 0.11223628691983123,
+      "grad_norm": 0.6230319738388062,
+      "learning_rate": 4.657293497363797e-05,
+      "loss": 1.096561312675476,
+      "step": 266
+    },
+    {
+      "epoch": 0.11308016877637131,
+      "grad_norm": 0.6231237649917603,
+      "learning_rate": 4.6924428822495606e-05,
+      "loss": 1.089842438697815,
+      "step": 268
+    },
+    {
+      "epoch": 0.11392405063291139,
+      "grad_norm": 0.7178627252578735,
+      "learning_rate": 4.727592267135325e-05,
+      "loss": 1.0696645975112915,
+      "step": 270
+    },
+    {
+      "epoch": 0.11476793248945148,
+      "grad_norm": 0.6895854473114014,
+      "learning_rate": 4.7627416520210896e-05,
+      "loss": 1.0511361360549927,
+      "step": 272
+    },
+    {
+      "epoch": 0.11561181434599156,
+      "grad_norm": 0.6046878695487976,
+      "learning_rate": 4.797891036906854e-05,
+      "loss": 1.1373958587646484,
+      "step": 274
+    },
+    {
+      "epoch": 0.11645569620253164,
+      "grad_norm": 0.6524552702903748,
+      "learning_rate": 4.833040421792619e-05,
+      "loss": 1.0734186172485352,
+      "step": 276
+    },
+    {
+      "epoch": 0.11729957805907174,
+      "grad_norm": 0.6331019997596741,
+      "learning_rate": 4.868189806678383e-05,
+      "loss": 1.123913049697876,
+      "step": 278
+    },
+    {
+      "epoch": 0.11814345991561181,
+      "grad_norm": 0.5919018983840942,
+      "learning_rate": 4.903339191564148e-05,
+      "loss": 1.0635710954666138,
+      "step": 280
+    },
+    {
+      "epoch": 0.1189873417721519,
+      "grad_norm": 0.6067633032798767,
+      "learning_rate": 4.938488576449912e-05,
+      "loss": 1.0429247617721558,
+      "step": 282
+    },
+    {
+      "epoch": 0.11983122362869199,
+      "grad_norm": 0.6583750247955322,
+      "learning_rate": 4.9736379613356774e-05,
+      "loss": 1.1397464275360107,
+      "step": 284
+    },
+    {
+      "epoch": 0.12067510548523207,
+      "grad_norm": 0.6200069785118103,
+      "learning_rate": 5.008787346221442e-05,
+      "loss": 1.0590803623199463,
+      "step": 286
+    },
+    {
+      "epoch": 0.12151898734177215,
+      "grad_norm": 0.6798665523529053,
+      "learning_rate": 5.0439367311072064e-05,
+      "loss": 1.1318789720535278,
+      "step": 288
+    },
+    {
+      "epoch": 0.12236286919831224,
+      "grad_norm": 0.7508794069290161,
+      "learning_rate": 5.07908611599297e-05,
+      "loss": 1.0934956073760986,
+      "step": 290
+    },
+    {
+      "epoch": 0.12320675105485232,
+      "grad_norm": 0.6901452541351318,
+      "learning_rate": 5.114235500878735e-05,
+      "loss": 1.163407802581787,
+      "step": 292
+    },
+    {
+      "epoch": 0.1240506329113924,
+      "grad_norm": 0.6423285603523254,
+      "learning_rate": 5.1493848857644985e-05,
+      "loss": 1.09059476852417,
+      "step": 294
+    },
+    {
+      "epoch": 0.1248945147679325,
+      "grad_norm": 0.6839275360107422,
+      "learning_rate": 5.1845342706502644e-05,
+      "loss": 1.0690211057662964,
+      "step": 296
+    },
+    {
+      "epoch": 0.1257383966244726,
+      "grad_norm": 0.6350128054618835,
+      "learning_rate": 5.219683655536028e-05,
+      "loss": 0.982322096824646,
+      "step": 298
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "grad_norm": 0.7136530876159668,
+      "learning_rate": 5.254833040421793e-05,
+      "loss": 1.1132930517196655,
+      "step": 300
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "eval_loss": 1.0952109098434448,
+      "eval_runtime": 677.0652,
+      "eval_samples_per_second": 3.112,
+      "eval_steps_per_second": 3.112,
+      "step": 300
+    },
+    {
+      "epoch": 0.12742616033755275,
+      "grad_norm": 0.7339721322059631,
+      "learning_rate": 5.289982425307557e-05,
+      "loss": 0.973595917224884,
+      "step": 302
+    },
+    {
+      "epoch": 0.12827004219409283,
+      "grad_norm": 0.5941481590270996,
+      "learning_rate": 5.3251318101933224e-05,
+      "loss": 0.9819849729537964,
+      "step": 304
+    },
+    {
+      "epoch": 0.1291139240506329,
+      "grad_norm": 0.7153938412666321,
+      "learning_rate": 5.360281195079086e-05,
+      "loss": 1.0315470695495605,
+      "step": 306
+    },
+    {
+      "epoch": 0.12995780590717299,
+      "grad_norm": 0.5167180299758911,
+      "learning_rate": 5.395430579964851e-05,
+      "loss": 0.9492001533508301,
+      "step": 308
+    },
+    {
+      "epoch": 0.1308016877637131,
+      "grad_norm": 0.6055944561958313,
+      "learning_rate": 5.430579964850615e-05,
+      "loss": 1.0156209468841553,
+      "step": 310
+    },
+    {
+      "epoch": 0.13164556962025317,
+      "grad_norm": 0.7662386298179626,
+      "learning_rate": 5.4657293497363805e-05,
+      "loss": 1.1791651248931885,
+      "step": 312
+    },
+    {
+      "epoch": 0.13248945147679325,
+      "grad_norm": 0.6065546274185181,
+      "learning_rate": 5.500878734622145e-05,
+      "loss": 1.0009297132492065,
+      "step": 314
+    },
+    {
+      "epoch": 0.13333333333333333,
+      "grad_norm": 0.604225754737854,
+      "learning_rate": 5.536028119507909e-05,
+      "loss": 1.0208244323730469,
+      "step": 316
+    },
+    {
+      "epoch": 0.1341772151898734,
+      "grad_norm": 0.6186763048171997,
+      "learning_rate": 5.571177504393673e-05,
+      "loss": 0.9968416690826416,
+      "step": 318
+    },
+    {
+      "epoch": 0.1350210970464135,
+      "grad_norm": 0.7100363969802856,
+      "learning_rate": 5.606326889279437e-05,
+      "loss": 0.9540256857872009,
+      "step": 320
+    },
+    {
+      "epoch": 0.1358649789029536,
+      "grad_norm": 0.6979711055755615,
+      "learning_rate": 5.641476274165203e-05,
+      "loss": 1.0631953477859497,
+      "step": 322
+    },
+    {
+      "epoch": 0.13670886075949368,
+      "grad_norm": 0.6237109303474426,
+      "learning_rate": 5.676625659050967e-05,
+      "loss": 1.0170501470565796,
+      "step": 324
+    },
+    {
+      "epoch": 0.13755274261603376,
+      "grad_norm": 0.6525548696517944,
+      "learning_rate": 5.711775043936731e-05,
+      "loss": 1.0715603828430176,
+      "step": 326
+    },
+    {
+      "epoch": 0.13839662447257384,
+      "grad_norm": 0.6869221329689026,
+      "learning_rate": 5.746924428822495e-05,
+      "loss": 1.0111541748046875,
+      "step": 328
+    },
+    {
+      "epoch": 0.13924050632911392,
+      "grad_norm": 0.553188145160675,
+      "learning_rate": 5.782073813708261e-05,
+      "loss": 1.0311682224273682,
+      "step": 330
+    },
+    {
+      "epoch": 0.140084388185654,
+      "grad_norm": 0.6760852932929993,
+      "learning_rate": 5.817223198594025e-05,
+      "loss": 1.0213634967803955,
+      "step": 332
+    },
+    {
+      "epoch": 0.1409282700421941,
+      "grad_norm": 0.5907419919967651,
+      "learning_rate": 5.8523725834797894e-05,
+      "loss": 0.9748594164848328,
+      "step": 334
+    },
+    {
+      "epoch": 0.14177215189873418,
+      "grad_norm": 0.7044920921325684,
+      "learning_rate": 5.887521968365554e-05,
+      "loss": 1.05863356590271,
+      "step": 336
+    },
+    {
+      "epoch": 0.14261603375527426,
+      "grad_norm": 0.679073691368103,
+      "learning_rate": 5.922671353251318e-05,
+      "loss": 1.1341127157211304,
+      "step": 338
+    },
+    {
+      "epoch": 0.14345991561181434,
+      "grad_norm": 0.7676237225532532,
+      "learning_rate": 5.957820738137083e-05,
+      "loss": 0.9540836215019226,
+      "step": 340
+    },
+    {
+      "epoch": 0.14430379746835442,
+      "grad_norm": 0.6313899755477905,
+      "learning_rate": 5.9929701230228474e-05,
+      "loss": 1.0585911273956299,
+      "step": 342
+    },
+    {
+      "epoch": 0.1451476793248945,
+      "grad_norm": 0.7123099565505981,
+      "learning_rate": 6.028119507908612e-05,
+      "loss": 1.0760118961334229,
+      "step": 344
+    },
+    {
+      "epoch": 0.1459915611814346,
+      "grad_norm": 0.585935652256012,
+      "learning_rate": 6.063268892794376e-05,
+      "loss": 1.036866307258606,
+      "step": 346
+    },
+    {
+      "epoch": 0.1468354430379747,
+      "grad_norm": 0.5643263459205627,
+      "learning_rate": 6.0984182776801416e-05,
+      "loss": 1.0242938995361328,
+      "step": 348
+    },
+    {
+      "epoch": 0.14767932489451477,
+      "grad_norm": 0.626761794090271,
+      "learning_rate": 6.133567662565906e-05,
+      "loss": 1.0497376918792725,
+      "step": 350
+    },
+    {
+      "epoch": 0.14852320675105485,
+      "grad_norm": 0.5106956958770752,
+      "learning_rate": 6.16871704745167e-05,
+      "loss": 0.9811885356903076,
+      "step": 352
+    },
+    {
+      "epoch": 0.14936708860759493,
+      "grad_norm": 0.6948089003562927,
+      "learning_rate": 6.203866432337434e-05,
+      "loss": 1.0715330839157104,
+      "step": 354
+    },
+    {
+      "epoch": 0.150210970464135,
+      "grad_norm": 0.699713945388794,
+      "learning_rate": 6.239015817223199e-05,
+      "loss": 1.0405226945877075,
+      "step": 356
+    },
+    {
+      "epoch": 0.15105485232067511,
+      "grad_norm": 0.6437667012214661,
+      "learning_rate": 6.274165202108964e-05,
+      "loss": 1.0490930080413818,
+      "step": 358
+    },
+    {
+      "epoch": 0.1518987341772152,
+      "grad_norm": 0.6952699422836304,
+      "learning_rate": 6.309314586994728e-05,
+      "loss": 0.9267548322677612,
+      "step": 360
+    },
+    {
+      "epoch": 0.15274261603375527,
+      "grad_norm": 0.6713186502456665,
+      "learning_rate": 6.344463971880492e-05,
+      "loss": 1.0427420139312744,
+      "step": 362
+    },
+    {
+      "epoch": 0.15358649789029535,
+      "grad_norm": 0.6750379800796509,
+      "learning_rate": 6.379613356766257e-05,
+      "loss": 1.048950433731079,
+      "step": 364
+    },
+    {
+      "epoch": 0.15443037974683543,
+      "grad_norm": 0.6053379774093628,
+      "learning_rate": 6.414762741652022e-05,
+      "loss": 1.0156004428863525,
+      "step": 366
+    },
+    {
+      "epoch": 0.15527426160337554,
+      "grad_norm": 0.8063633441925049,
+      "learning_rate": 6.449912126537786e-05,
+      "loss": 1.0020819902420044,
+      "step": 368
+    },
+    {
+      "epoch": 0.15611814345991562,
+      "grad_norm": 0.8027494549751282,
+      "learning_rate": 6.48506151142355e-05,
+      "loss": 1.055633783340454,
+      "step": 370
+    },
+    {
+      "epoch": 0.1569620253164557,
+      "grad_norm": 0.6580121517181396,
+      "learning_rate": 6.520210896309315e-05,
+      "loss": 1.0149940252304077,
+      "step": 372
+    },
+    {
+      "epoch": 0.15780590717299578,
+      "grad_norm": 0.6561233997344971,
+      "learning_rate": 6.55536028119508e-05,
+      "loss": 0.9769611954689026,
+      "step": 374
+    },
+    {
+      "epoch": 0.15864978902953586,
+      "grad_norm": 0.6444346308708191,
+      "learning_rate": 6.590509666080844e-05,
+      "loss": 0.9099349975585938,
+      "step": 376
+    },
+    {
+      "epoch": 0.15949367088607594,
+      "grad_norm": 0.5879359245300293,
+      "learning_rate": 6.625659050966608e-05,
+      "loss": 1.0797548294067383,
+      "step": 378
+    },
+    {
+      "epoch": 0.16033755274261605,
+      "grad_norm": 0.6994144916534424,
+      "learning_rate": 6.660808435852373e-05,
+      "loss": 1.0336791276931763,
+      "step": 380
+    },
+    {
+      "epoch": 0.16118143459915613,
+      "grad_norm": 0.6128669381141663,
+      "learning_rate": 6.695957820738138e-05,
+      "loss": 1.018118143081665,
+      "step": 382
+    },
+    {
+      "epoch": 0.1620253164556962,
+      "grad_norm": 1.0237540006637573,
+      "learning_rate": 6.731107205623902e-05,
+      "loss": 1.1405497789382935,
+      "step": 384
+    },
+    {
+      "epoch": 0.16286919831223629,
+      "grad_norm": 0.6091578006744385,
+      "learning_rate": 6.766256590509666e-05,
+      "loss": 1.0314189195632935,
+      "step": 386
+    },
+    {
+      "epoch": 0.16371308016877636,
+      "grad_norm": 0.5916037559509277,
+      "learning_rate": 6.801405975395431e-05,
+      "loss": 0.9564052820205688,
+      "step": 388
+    },
+    {
+      "epoch": 0.16455696202531644,
+      "grad_norm": 0.771653950214386,
+      "learning_rate": 6.836555360281195e-05,
+      "loss": 1.0023859739303589,
+      "step": 390
+    },
+    {
+      "epoch": 0.16540084388185655,
+      "grad_norm": 0.654658317565918,
+      "learning_rate": 6.87170474516696e-05,
+      "loss": 1.07024085521698,
+      "step": 392
+    },
+    {
+      "epoch": 0.16624472573839663,
+      "grad_norm": 0.6611968874931335,
+      "learning_rate": 6.906854130052724e-05,
+      "loss": 1.0552500486373901,
+      "step": 394
+    },
+    {
+      "epoch": 0.1670886075949367,
+      "grad_norm": 0.6955893039703369,
+      "learning_rate": 6.942003514938489e-05,
+      "loss": 1.0562875270843506,
+      "step": 396
+    },
+    {
+      "epoch": 0.1679324894514768,
+      "grad_norm": 0.6666058301925659,
+      "learning_rate": 6.977152899824253e-05,
+      "loss": 0.9850592017173767,
+      "step": 398
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "grad_norm": 0.6131711006164551,
+      "learning_rate": 7.012302284710018e-05,
+      "loss": 1.0077755451202393,
+      "step": 400
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "eval_loss": 1.0625108480453491,
+      "eval_runtime": 691.0068,
+      "eval_samples_per_second": 3.049,
+      "eval_steps_per_second": 3.049,
+      "step": 400
+    },
+    {
+      "epoch": 0.16962025316455695,
+      "grad_norm": 0.6286499500274658,
+      "learning_rate": 7.047451669595783e-05,
+      "loss": 1.1012427806854248,
+      "step": 402
+    },
+    {
+      "epoch": 0.17046413502109706,
+      "grad_norm": 0.6639351844787598,
+      "learning_rate": 7.082601054481547e-05,
+      "loss": 1.0379719734191895,
+      "step": 404
+    },
+    {
+      "epoch": 0.17130801687763714,
+      "grad_norm": 0.750401496887207,
+      "learning_rate": 7.117750439367311e-05,
+      "loss": 1.031856656074524,
+      "step": 406
+    },
+    {
+      "epoch": 0.17215189873417722,
+      "grad_norm": 0.8084847331047058,
+      "learning_rate": 7.152899824253075e-05,
+      "loss": 1.0493193864822388,
+      "step": 408
+    },
+    {
+      "epoch": 0.1729957805907173,
+      "grad_norm": 0.7448462247848511,
+      "learning_rate": 7.188049209138841e-05,
+      "loss": 1.1012418270111084,
+      "step": 410
+    },
+    {
+      "epoch": 0.17383966244725738,
+      "grad_norm": 0.5841867923736572,
+      "learning_rate": 7.223198594024605e-05,
+      "loss": 0.9926692247390747,
+      "step": 412
+    },
+    {
+      "epoch": 0.17468354430379746,
+      "grad_norm": 0.7125606536865234,
+      "learning_rate": 7.258347978910369e-05,
+      "loss": 1.0588877201080322,
+      "step": 414
+    },
+    {
+      "epoch": 0.17552742616033756,
+      "grad_norm": 0.5750942230224609,
+      "learning_rate": 7.293497363796134e-05,
+      "loss": 1.038270354270935,
+      "step": 416
+    },
+    {
+      "epoch": 0.17637130801687764,
+      "grad_norm": 0.565444827079773,
+      "learning_rate": 7.328646748681899e-05,
+      "loss": 0.9843021035194397,
+      "step": 418
+    },
+    {
+      "epoch": 0.17721518987341772,
+      "grad_norm": 0.5825693011283875,
+      "learning_rate": 7.363796133567663e-05,
+      "loss": 1.0731632709503174,
+      "step": 420
+    },
+    {
+      "epoch": 0.1780590717299578,
+      "grad_norm": 0.6267391443252563,
+      "learning_rate": 7.398945518453427e-05,
+      "loss": 1.0061273574829102,
+      "step": 422
+    },
+    {
+      "epoch": 0.17890295358649788,
+      "grad_norm": 0.6621372103691101,
+      "learning_rate": 7.434094903339192e-05,
+      "loss": 1.0461612939834595,
+      "step": 424
+    },
+    {
+      "epoch": 0.17974683544303796,
+      "grad_norm": 0.6635435223579407,
+      "learning_rate": 7.469244288224957e-05,
+      "loss": 0.9789207577705383,
+      "step": 426
+    },
+    {
+      "epoch": 0.18059071729957807,
+      "grad_norm": 0.6342346668243408,
+      "learning_rate": 7.504393673110721e-05,
+      "loss": 1.0527069568634033,
+      "step": 428
+    },
+    {
+      "epoch": 0.18143459915611815,
+      "grad_norm": 0.6762149930000305,
+      "learning_rate": 7.539543057996485e-05,
+      "loss": 0.9708702564239502,
+      "step": 430
+    },
+    {
+      "epoch": 0.18227848101265823,
+      "grad_norm": 0.7073282599449158,
+      "learning_rate": 7.57469244288225e-05,
+      "loss": 1.0509834289550781,
+      "step": 432
+    },
+    {
+      "epoch": 0.1831223628691983,
+      "grad_norm": 0.6917856931686401,
+      "learning_rate": 7.609841827768014e-05,
+      "loss": 1.0128819942474365,
+      "step": 434
+    },
+    {
+      "epoch": 0.1839662447257384,
+      "grad_norm": 0.5574942231178284,
+      "learning_rate": 7.644991212653779e-05,
+      "loss": 0.989395797252655,
+      "step": 436
+    },
+    {
+      "epoch": 0.1848101265822785,
+      "grad_norm": 0.640765905380249,
+      "learning_rate": 7.680140597539543e-05,
+      "loss": 0.9846042990684509,
+      "step": 438
+    },
+    {
+      "epoch": 0.18565400843881857,
+      "grad_norm": 0.6699127554893494,
+      "learning_rate": 7.715289982425308e-05,
+      "loss": 1.0344442129135132,
+      "step": 440
+    },
+    {
+      "epoch": 0.18649789029535865,
+      "grad_norm": 0.6164930462837219,
+      "learning_rate": 7.750439367311072e-05,
+      "loss": 1.0179373025894165,
+      "step": 442
+    },
+    {
+      "epoch": 0.18734177215189873,
+      "grad_norm": 0.6880720853805542,
+      "learning_rate": 7.785588752196837e-05,
+      "loss": 1.0518895387649536,
+      "step": 444
+    },
+    {
+      "epoch": 0.1881856540084388,
+      "grad_norm": 0.6501413583755493,
+      "learning_rate": 7.820738137082601e-05,
+      "loss": 1.0442606210708618,
+      "step": 446
+    },
+    {
+      "epoch": 0.1890295358649789,
+      "grad_norm": 0.6076085567474365,
+      "learning_rate": 7.855887521968366e-05,
+      "loss": 0.9828442335128784,
+      "step": 448
+    },
+    {
+      "epoch": 0.189873417721519,
+      "grad_norm": 0.6418202519416809,
+      "learning_rate": 7.89103690685413e-05,
+      "loss": 1.0573710203170776,
+      "step": 450
+    },
+    {
+      "epoch": 0.19071729957805908,
+      "grad_norm": 0.7055076360702515,
+      "learning_rate": 7.926186291739895e-05,
+      "loss": 1.0216103792190552,
+      "step": 452
+    },
+    {
+      "epoch": 0.19156118143459916,
+      "grad_norm": 0.5668330192565918,
+      "learning_rate": 7.961335676625659e-05,
+      "loss": 0.9837722778320312,
+      "step": 454
+    },
+    {
+      "epoch": 0.19240506329113924,
+      "grad_norm": 0.6419380307197571,
+      "learning_rate": 7.996485061511424e-05,
+      "loss": 1.0003894567489624,
+      "step": 456
+    },
+    {
+      "epoch": 0.19324894514767932,
+      "grad_norm": 0.5949198007583618,
+      "learning_rate": 8.031634446397188e-05,
+      "loss": 1.0609031915664673,
+      "step": 458
+    },
+    {
+      "epoch": 0.1940928270042194,
+      "grad_norm": 0.7032039761543274,
+      "learning_rate": 8.066783831282952e-05,
+      "loss": 1.0543403625488281,
+      "step": 460
+    },
+    {
+      "epoch": 0.1949367088607595,
+      "grad_norm": 0.5775868892669678,
+      "learning_rate": 8.101933216168718e-05,
+      "loss": 0.9819303154945374,
+      "step": 462
+    },
+    {
+      "epoch": 0.19578059071729959,
+      "grad_norm": 0.9301062226295471,
+      "learning_rate": 8.137082601054482e-05,
+      "loss": 1.0542067289352417,
+      "step": 464
+    },
+    {
+      "epoch": 0.19662447257383966,
+      "grad_norm": 0.6193217039108276,
+      "learning_rate": 8.172231985940246e-05,
+      "loss": 0.9966341257095337,
+      "step": 466
+    },
+    {
+      "epoch": 0.19746835443037974,
+      "grad_norm": 0.6286146640777588,
+      "learning_rate": 8.20738137082601e-05,
+      "loss": 1.0474121570587158,
+      "step": 468
+    },
+    {
+      "epoch": 0.19831223628691982,
+      "grad_norm": 0.7418972253799438,
+      "learning_rate": 8.242530755711776e-05,
+      "loss": 0.9549239277839661,
+      "step": 470
+    },
+    {
+      "epoch": 0.1991561181434599,
+      "grad_norm": 0.6122808456420898,
+      "learning_rate": 8.27768014059754e-05,
+      "loss": 1.0191338062286377,
+      "step": 472
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.6375362277030945,
+      "learning_rate": 8.312829525483304e-05,
+      "loss": 1.0987539291381836,
+      "step": 474
+    },
+    {
+      "epoch": 0.2008438818565401,
+      "grad_norm": 0.6459513306617737,
+      "learning_rate": 8.347978910369068e-05,
+      "loss": 1.0369136333465576,
+      "step": 476
+    },
+    {
+      "epoch": 0.20168776371308017,
+      "grad_norm": 0.7029640674591064,
+      "learning_rate": 8.383128295254833e-05,
+      "loss": 1.0582096576690674,
+      "step": 478
+    },
+    {
+      "epoch": 0.20253164556962025,
+      "grad_norm": 0.6345387697219849,
+      "learning_rate": 8.418277680140598e-05,
+      "loss": 1.022916316986084,
+      "step": 480
+    },
+    {
+      "epoch": 0.20337552742616033,
+      "grad_norm": 0.5764590501785278,
+      "learning_rate": 8.453427065026362e-05,
+      "loss": 0.973024308681488,
+      "step": 482
+    },
+    {
+      "epoch": 0.2042194092827004,
+      "grad_norm": 0.5884482860565186,
+      "learning_rate": 8.488576449912127e-05,
+      "loss": 1.0292812585830688,
+      "step": 484
+    },
+    {
+      "epoch": 0.20506329113924052,
+      "grad_norm": 0.616357147693634,
+      "learning_rate": 8.523725834797891e-05,
+      "loss": 1.0083447694778442,
+      "step": 486
+    },
+    {
+      "epoch": 0.2059071729957806,
+      "grad_norm": 0.7671196460723877,
+      "learning_rate": 8.558875219683656e-05,
+      "loss": 0.9936985373497009,
+      "step": 488
+    },
+    {
+      "epoch": 0.20675105485232068,
+      "grad_norm": 0.6197299957275391,
+      "learning_rate": 8.59402460456942e-05,
+      "loss": 1.051513910293579,
+      "step": 490
+    },
+    {
+      "epoch": 0.20759493670886076,
+      "grad_norm": 0.6912890672683716,
+      "learning_rate": 8.629173989455185e-05,
+      "loss": 0.9474978446960449,
+      "step": 492
+    },
+    {
+      "epoch": 0.20843881856540084,
+      "grad_norm": 0.6941592693328857,
+      "learning_rate": 8.664323374340949e-05,
+      "loss": 1.0671660900115967,
+      "step": 494
+    },
+    {
+      "epoch": 0.20928270042194091,
+      "grad_norm": 0.5889528393745422,
+      "learning_rate": 8.699472759226714e-05,
+      "loss": 1.0020159482955933,
+      "step": 496
+    },
+    {
+      "epoch": 0.21012658227848102,
+      "grad_norm": 0.6478549838066101,
+      "learning_rate": 8.734622144112478e-05,
+      "loss": 1.0165860652923584,
+      "step": 498
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "grad_norm": 0.6018255949020386,
+      "learning_rate": 8.769771528998243e-05,
+      "loss": 0.8798263072967529,
+      "step": 500
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "eval_loss": 1.042096495628357,
+      "eval_runtime": 692.4361,
+      "eval_samples_per_second": 3.043,
+      "eval_steps_per_second": 3.043,
+      "step": 500
+    },
+    {
+      "epoch": 0.21181434599156118,
+      "grad_norm": 0.578990638256073,
+      "learning_rate": 8.804920913884007e-05,
+      "loss": 1.092096209526062,
+      "step": 502
+    },
+    {
+      "epoch": 0.21265822784810126,
+      "grad_norm": 0.6597883701324463,
+      "learning_rate": 8.840070298769771e-05,
+      "loss": 1.0413451194763184,
+      "step": 504
+    },
+    {
+      "epoch": 0.21350210970464134,
+      "grad_norm": 0.6660305261611938,
+      "learning_rate": 8.875219683655536e-05,
+      "loss": 1.0073142051696777,
+      "step": 506
+    },
+    {
+      "epoch": 0.21434599156118145,
+      "grad_norm": 0.6283115148544312,
+      "learning_rate": 8.910369068541301e-05,
+      "loss": 1.0319768190383911,
+      "step": 508
+    },
+    {
+      "epoch": 0.21518987341772153,
+      "grad_norm": 0.6257343292236328,
+      "learning_rate": 8.945518453427065e-05,
+      "loss": 1.0046353340148926,
+      "step": 510
+    },
+    {
+      "epoch": 0.2160337552742616,
+      "grad_norm": 0.5530875325202942,
+      "learning_rate": 8.980667838312829e-05,
+      "loss": 0.9169099926948547,
+      "step": 512
+    },
+    {
+      "epoch": 0.2168776371308017,
+      "grad_norm": 0.5369633436203003,
+      "learning_rate": 9.015817223198594e-05,
+      "loss": 0.9081505537033081,
+      "step": 514
+    },
+    {
+      "epoch": 0.21772151898734177,
+      "grad_norm": 0.6618232131004333,
+      "learning_rate": 9.05096660808436e-05,
+      "loss": 0.9165045022964478,
+      "step": 516
+    },
+    {
+      "epoch": 0.21856540084388185,
+      "grad_norm": 0.600666344165802,
+      "learning_rate": 9.086115992970123e-05,
+      "loss": 0.91348797082901,
+      "step": 518
+    },
+    {
+      "epoch": 0.21940928270042195,
+      "grad_norm": 0.5919831991195679,
+      "learning_rate": 9.121265377855887e-05,
+      "loss": 1.006508469581604,
+      "step": 520
+    },
+    {
+      "epoch": 0.22025316455696203,
+      "grad_norm": 0.688058614730835,
+      "learning_rate": 9.156414762741654e-05,
+      "loss": 1.0013236999511719,
+      "step": 522
+    },
+    {
+      "epoch": 0.2210970464135021,
+      "grad_norm": 0.6721227765083313,
+      "learning_rate": 9.191564147627418e-05,
+      "loss": 1.0909923315048218,
+      "step": 524
+    },
+    {
+      "epoch": 0.2219409282700422,
+      "grad_norm": 0.5987313389778137,
+      "learning_rate": 9.226713532513181e-05,
+      "loss": 1.0117096900939941,
+      "step": 526
+    },
+    {
+      "epoch": 0.22278481012658227,
+      "grad_norm": 0.6191489696502686,
+      "learning_rate": 9.261862917398945e-05,
+      "loss": 1.0153647661209106,
+      "step": 528
+    },
+    {
+      "epoch": 0.22362869198312235,
+      "grad_norm": 0.6821563243865967,
+      "learning_rate": 9.29701230228471e-05,
+      "loss": 0.9649755954742432,
+      "step": 530
+    },
+    {
+      "epoch": 0.22447257383966246,
+      "grad_norm": 1.760398268699646,
+      "learning_rate": 9.332161687170476e-05,
+      "loss": 0.8673232197761536,
+      "step": 532
+    },
+    {
+      "epoch": 0.22531645569620254,
+      "grad_norm": 0.6670058369636536,
+      "learning_rate": 9.36731107205624e-05,
+      "loss": 0.9942440986633301,
+      "step": 534
+    },
+    {
+      "epoch": 0.22616033755274262,
+      "grad_norm": 0.7345916032791138,
+      "learning_rate": 9.402460456942003e-05,
+      "loss": 1.0364389419555664,
+      "step": 536
+    },
+    {
+      "epoch": 0.2270042194092827,
+      "grad_norm": 0.5946128964424133,
+      "learning_rate": 9.437609841827768e-05,
+      "loss": 0.9314924478530884,
+      "step": 538
+    },
+    {
+      "epoch": 0.22784810126582278,
+      "grad_norm": 0.5800848603248596,
+      "learning_rate": 9.472759226713534e-05,
+      "loss": 1.0694862604141235,
+      "step": 540
+    },
+    {
+      "epoch": 0.22869198312236286,
+      "grad_norm": 0.6712192893028259,
+      "learning_rate": 9.507908611599297e-05,
+      "loss": 1.03531014919281,
+      "step": 542
+    },
+    {
+      "epoch": 0.22953586497890296,
+      "grad_norm": 0.5641416311264038,
+      "learning_rate": 9.543057996485063e-05,
+      "loss": 0.9795235991477966,
+      "step": 544
+    },
+    {
+      "epoch": 0.23037974683544304,
+      "grad_norm": 0.50412517786026,
+      "learning_rate": 9.578207381370826e-05,
+      "loss": 0.9641494750976562,
+      "step": 546
+    },
+    {
+      "epoch": 0.23122362869198312,
+      "grad_norm": 0.579118549823761,
+      "learning_rate": 9.61335676625659e-05,
+      "loss": 0.9375281929969788,
+      "step": 548
+    },
+    {
+      "epoch": 0.2320675105485232,
+      "grad_norm": 0.5888341665267944,
+      "learning_rate": 9.648506151142355e-05,
+      "loss": 0.9414046406745911,
+      "step": 550
+    },
+    {
+      "epoch": 0.23291139240506328,
+      "grad_norm": 0.5595056414604187,
+      "learning_rate": 9.68365553602812e-05,
+      "loss": 0.9005617499351501,
+      "step": 552
+    },
+    {
+      "epoch": 0.23375527426160336,
+      "grad_norm": 0.6605326533317566,
+      "learning_rate": 9.718804920913884e-05,
+      "loss": 1.0283968448638916,
+      "step": 554
+    },
+    {
+      "epoch": 0.23459915611814347,
+      "grad_norm": 0.5657313466072083,
+      "learning_rate": 9.753954305799648e-05,
+      "loss": 1.0058249235153198,
+      "step": 556
+    },
+    {
+      "epoch": 0.23544303797468355,
+      "grad_norm": 0.5433364510536194,
+      "learning_rate": 9.789103690685413e-05,
+      "loss": 0.9835494756698608,
+      "step": 558
+    },
+    {
+      "epoch": 0.23628691983122363,
+      "grad_norm": 0.6129802465438843,
+      "learning_rate": 9.824253075571179e-05,
+      "loss": 1.054532527923584,
+      "step": 560
+    },
+    {
+      "epoch": 0.2371308016877637,
+      "grad_norm": 0.6496239304542542,
+      "learning_rate": 9.859402460456942e-05,
+      "loss": 1.0240973234176636,
+      "step": 562
+    },
+    {
+      "epoch": 0.2379746835443038,
+      "grad_norm": 0.6380873918533325,
+      "learning_rate": 9.894551845342706e-05,
+      "loss": 1.0229179859161377,
+      "step": 564
+    },
+    {
+      "epoch": 0.23881856540084387,
+      "grad_norm": 0.6151993870735168,
+      "learning_rate": 9.929701230228471e-05,
+      "loss": 1.0111570358276367,
+      "step": 566
+    },
+    {
+      "epoch": 0.23966244725738398,
+      "grad_norm": 0.5727584958076477,
+      "learning_rate": 9.964850615114237e-05,
+      "loss": 0.9450829029083252,
+      "step": 568
+    },
+    {
+      "epoch": 0.24050632911392406,
+      "grad_norm": 0.6620725989341736,
+      "learning_rate": 0.0001,
+      "loss": 0.9800319075584412,
+      "step": 570
+    },
+    {
+      "epoch": 0.24135021097046414,
+      "grad_norm": 0.6151163578033447,
+      "learning_rate": 0.00010035149384885764,
+      "loss": 0.9757438898086548,
+      "step": 572
+    },
+    {
+      "epoch": 0.24219409282700421,
+      "grad_norm": 0.5672140717506409,
+      "learning_rate": 0.0001007029876977153,
+      "loss": 0.9104921817779541,
+      "step": 574
+    },
+    {
+      "epoch": 0.2430379746835443,
+      "grad_norm": 0.5697256326675415,
+      "learning_rate": 0.00010105448154657293,
+      "loss": 1.1027376651763916,
+      "step": 576
+    },
+    {
+      "epoch": 0.2438818565400844,
+      "grad_norm": 0.5590381622314453,
+      "learning_rate": 0.00010140597539543057,
+      "loss": 1.0055404901504517,
+      "step": 578
+    },
+    {
+      "epoch": 0.24472573839662448,
+      "grad_norm": 0.5518567562103271,
+      "learning_rate": 0.00010175746924428824,
+      "loss": 1.020835518836975,
+      "step": 580
+    },
+    {
+      "epoch": 0.24556962025316456,
+      "grad_norm": 0.6338496208190918,
+      "learning_rate": 0.00010210896309314588,
+      "loss": 0.9528344869613647,
+      "step": 582
+    },
+    {
+      "epoch": 0.24641350210970464,
+      "grad_norm": 0.6497329473495483,
+      "learning_rate": 0.00010246045694200353,
+      "loss": 1.0088670253753662,
+      "step": 584
+    },
+    {
+      "epoch": 0.24725738396624472,
+      "grad_norm": 0.49888095259666443,
+      "learning_rate": 0.00010281195079086117,
+      "loss": 0.9961200952529907,
+      "step": 586
+    },
+    {
+      "epoch": 0.2481012658227848,
+      "grad_norm": 0.5680158734321594,
+      "learning_rate": 0.0001031634446397188,
+      "loss": 0.9635610580444336,
+      "step": 588
+    },
+    {
+      "epoch": 0.2489451476793249,
+      "grad_norm": 0.658168375492096,
+      "learning_rate": 0.00010351493848857646,
+      "loss": 0.9392287135124207,
+      "step": 590
+    },
+    {
+      "epoch": 0.249789029535865,
+      "grad_norm": 0.618262767791748,
+      "learning_rate": 0.0001038664323374341,
+      "loss": 0.9600516557693481,
+      "step": 592
+    },
+    {
+      "epoch": 0.25063291139240507,
+      "grad_norm": 0.6003909111022949,
+      "learning_rate": 0.00010421792618629173,
+      "loss": 1.005476713180542,
+      "step": 594
+    },
+    {
+      "epoch": 0.2514767932489452,
+      "grad_norm": 0.5437078475952148,
+      "learning_rate": 0.00010456942003514938,
+      "loss": 0.9523017406463623,
+      "step": 596
+    },
+    {
+      "epoch": 0.2523206751054852,
+      "grad_norm": 0.5524541735649109,
+      "learning_rate": 0.00010492091388400705,
+      "loss": 0.9526668787002563,
+      "step": 598
+    },
+    {
+      "epoch": 0.25316455696202533,
+      "grad_norm": 0.679504930973053,
+      "learning_rate": 0.00010527240773286469,
+      "loss": 1.019660472869873,
+      "step": 600
+    },
+    {
+      "epoch": 0.25316455696202533,
+      "eval_loss": 1.0193854570388794,
+      "eval_runtime": 677.9523,
+      "eval_samples_per_second": 3.108,
+      "eval_steps_per_second": 3.108,
+      "step": 600
+    },
+    {
+      "epoch": 0.2540084388185654,
+      "grad_norm": 0.5646136999130249,
+      "learning_rate": 0.00010562390158172233,
+      "loss": 0.9910882711410522,
+      "step": 602
+    },
+    {
+      "epoch": 0.2548523206751055,
+      "grad_norm": 0.5238093137741089,
+      "learning_rate": 0.00010597539543057998,
+      "loss": 0.9616432785987854,
+      "step": 604
+    },
+    {
+      "epoch": 0.25569620253164554,
+      "grad_norm": 0.7483857274055481,
+      "learning_rate": 0.00010632688927943762,
+      "loss": 1.0078275203704834,
+      "step": 606
+    },
+    {
+      "epoch": 0.25654008438818565,
+      "grad_norm": 0.578948974609375,
+      "learning_rate": 0.00010667838312829525,
+      "loss": 0.9827103018760681,
+      "step": 608
+    },
+    {
+      "epoch": 0.25738396624472576,
+      "grad_norm": 0.5525906085968018,
+      "learning_rate": 0.00010702987697715289,
+      "loss": 1.0423277616500854,
+      "step": 610
+    },
+    {
+      "epoch": 0.2582278481012658,
+      "grad_norm": 0.6721326112747192,
+      "learning_rate": 0.00010738137082601054,
+      "loss": 0.9561693072319031,
+      "step": 612
+    },
+    {
+      "epoch": 0.2590717299578059,
+      "grad_norm": 0.5701051354408264,
+      "learning_rate": 0.00010773286467486821,
+      "loss": 0.9602992534637451,
+      "step": 614
+    },
+    {
+      "epoch": 0.25991561181434597,
+      "grad_norm": 0.6349860429763794,
+      "learning_rate": 0.00010808435852372585,
+      "loss": 1.1422650814056396,
+      "step": 616
+    },
+    {
+      "epoch": 0.2607594936708861,
+      "grad_norm": 0.5496085286140442,
+      "learning_rate": 0.00010843585237258349,
+      "loss": 0.9762773513793945,
+      "step": 618
+    },
+    {
+      "epoch": 0.2616033755274262,
+      "grad_norm": 0.6080722808837891,
+      "learning_rate": 0.00010878734622144114,
+      "loss": 1.0133616924285889,
+      "step": 620
+    },
+    {
+      "epoch": 0.26244725738396624,
+      "grad_norm": 0.5450218915939331,
+      "learning_rate": 0.00010913884007029878,
+      "loss": 0.9385587573051453,
+      "step": 622
+    },
+    {
+      "epoch": 0.26329113924050634,
+      "grad_norm": 0.592106819152832,
+      "learning_rate": 0.00010949033391915641,
+      "loss": 0.9359989762306213,
+      "step": 624
+    },
+    {
+      "epoch": 0.2641350210970464,
+      "grad_norm": 0.6449427604675293,
+      "learning_rate": 0.00010984182776801407,
+      "loss": 1.0266027450561523,
+      "step": 626
+    },
+    {
+      "epoch": 0.2649789029535865,
+      "grad_norm": 0.538299560546875,
+      "learning_rate": 0.0001101933216168717,
+      "loss": 0.9303187131881714,
+      "step": 628
+    },
+    {
+      "epoch": 0.26582278481012656,
+      "grad_norm": 0.546316921710968,
+      "learning_rate": 0.00011054481546572934,
+      "loss": 0.9368857145309448,
+      "step": 630
+    },
+    {
+      "epoch": 0.26666666666666666,
+      "grad_norm": 0.5818730592727661,
+      "learning_rate": 0.00011089630931458701,
+      "loss": 0.9573145508766174,
+      "step": 632
+    },
+    {
+      "epoch": 0.26751054852320677,
+      "grad_norm": 0.5958262085914612,
+      "learning_rate": 0.00011124780316344465,
+      "loss": 0.9345449805259705,
+      "step": 634
+    },
+    {
+      "epoch": 0.2683544303797468,
+      "grad_norm": 0.6259077787399292,
+      "learning_rate": 0.0001115992970123023,
+      "loss": 1.0906590223312378,
+      "step": 636
+    },
+    {
+      "epoch": 0.26919831223628693,
+      "grad_norm": 0.589672863483429,
+      "learning_rate": 0.00011195079086115994,
+      "loss": 1.0757447481155396,
+      "step": 638
+    },
+    {
+      "epoch": 0.270042194092827,
+      "grad_norm": 0.5714080333709717,
+      "learning_rate": 0.00011230228471001758,
+      "loss": 0.9310855269432068,
+      "step": 640
+    },
+    {
+      "epoch": 0.2708860759493671,
+      "grad_norm": 0.45342639088630676,
+      "learning_rate": 0.00011265377855887523,
+      "loss": 0.9276360273361206,
+      "step": 642
+    },
+    {
+      "epoch": 0.2717299578059072,
+      "grad_norm": 0.6386750340461731,
+      "learning_rate": 0.00011300527240773287,
+      "loss": 1.084719181060791,
+      "step": 644
+    },
+    {
+      "epoch": 0.27257383966244725,
+      "grad_norm": 0.6446163654327393,
+      "learning_rate": 0.0001133567662565905,
+      "loss": 0.9763918519020081,
+      "step": 646
+    },
+    {
+      "epoch": 0.27341772151898736,
+      "grad_norm": 0.5925686359405518,
+      "learning_rate": 0.00011370826010544816,
+      "loss": 0.9517921805381775,
+      "step": 648
+    },
+    {
+      "epoch": 0.2742616033755274,
+      "grad_norm": 0.5399773716926575,
+      "learning_rate": 0.00011405975395430582,
+      "loss": 1.0587927103042603,
+      "step": 650
+    },
+    {
+      "epoch": 0.2751054852320675,
+      "grad_norm": 0.5872456431388855,
+      "learning_rate": 0.00011441124780316346,
+      "loss": 0.883341908454895,
+      "step": 652
+    },
+    {
+      "epoch": 0.2759493670886076,
+      "grad_norm": 0.5574564337730408,
+      "learning_rate": 0.0001147627416520211,
+      "loss": 1.0306891202926636,
+      "step": 654
+    },
+    {
+      "epoch": 0.2767932489451477,
+      "grad_norm": 0.47789013385772705,
+      "learning_rate": 0.00011511423550087874,
+      "loss": 0.8814032077789307,
+      "step": 656
+    },
+    {
+      "epoch": 0.2776371308016878,
+      "grad_norm": 0.5565530061721802,
+      "learning_rate": 0.00011546572934973639,
+      "loss": 0.9460552334785461,
+      "step": 658
+    },
+    {
+      "epoch": 0.27848101265822783,
+      "grad_norm": 0.5299761295318604,
+      "learning_rate": 0.00011581722319859403,
+      "loss": 0.9475110769271851,
+      "step": 660
+    },
+    {
+      "epoch": 0.27932489451476794,
+      "grad_norm": 0.6503344178199768,
+      "learning_rate": 0.00011616871704745166,
+      "loss": 1.0630913972854614,
+      "step": 662
+    },
+    {
+      "epoch": 0.280168776371308,
+      "grad_norm": 0.5794585943222046,
+      "learning_rate": 0.00011652021089630932,
+      "loss": 0.9389138221740723,
+      "step": 664
+    },
+    {
+      "epoch": 0.2810126582278481,
+      "grad_norm": 0.5762867331504822,
+      "learning_rate": 0.00011687170474516695,
+      "loss": 0.8934136033058167,
+      "step": 666
+    },
+    {
+      "epoch": 0.2818565400843882,
+      "grad_norm": 0.6565435528755188,
+      "learning_rate": 0.00011722319859402462,
+      "loss": 1.1072614192962646,
+      "step": 668
+    },
+    {
+      "epoch": 0.28270042194092826,
+      "grad_norm": 0.5819830298423767,
+      "learning_rate": 0.00011757469244288226,
+      "loss": 1.0501434803009033,
+      "step": 670
+    },
+    {
+      "epoch": 0.28354430379746837,
+      "grad_norm": 0.6071487069129944,
+      "learning_rate": 0.00011792618629173991,
+      "loss": 0.9880793690681458,
+      "step": 672
+    },
+    {
+      "epoch": 0.2843881856540084,
+      "grad_norm": 0.5765058398246765,
+      "learning_rate": 0.00011827768014059755,
+      "loss": 0.9670693874359131,
+      "step": 674
+    },
+    {
+      "epoch": 0.2852320675105485,
+      "grad_norm": 0.5245351791381836,
+      "learning_rate": 0.00011862917398945519,
+      "loss": 0.9602360725402832,
+      "step": 676
+    },
+    {
+      "epoch": 0.28607594936708863,
+      "grad_norm": 0.6189922094345093,
+      "learning_rate": 0.00011898066783831282,
+      "loss": 0.9684560894966125,
+      "step": 678
+    },
+    {
+      "epoch": 0.2869198312236287,
+      "grad_norm": 0.6138690710067749,
+      "learning_rate": 0.00011933216168717048,
+      "loss": 0.9465792775154114,
+      "step": 680
+    },
+    {
+      "epoch": 0.2877637130801688,
+      "grad_norm": 0.5371595621109009,
+      "learning_rate": 0.00011968365553602812,
+      "loss": 0.8495944738388062,
+      "step": 682
+    },
+    {
+      "epoch": 0.28860759493670884,
+      "grad_norm": 0.5549944639205933,
+      "learning_rate": 0.00012003514938488578,
+      "loss": 0.9663267135620117,
+      "step": 684
+    },
+    {
+      "epoch": 0.28945147679324895,
+      "grad_norm": 0.6484189033508301,
+      "learning_rate": 0.00012038664323374342,
+      "loss": 0.9736058712005615,
+      "step": 686
+    },
+    {
+      "epoch": 0.290295358649789,
+      "grad_norm": 0.540351390838623,
+      "learning_rate": 0.00012073813708260107,
+      "loss": 1.0591845512390137,
+      "step": 688
+    },
+    {
+      "epoch": 0.2911392405063291,
+      "grad_norm": 0.5657922029495239,
+      "learning_rate": 0.00012108963093145871,
+      "loss": 0.944908618927002,
+      "step": 690
+    },
+    {
+      "epoch": 0.2919831223628692,
+      "grad_norm": 0.6040505170822144,
+      "learning_rate": 0.00012144112478031635,
+      "loss": 1.0018219947814941,
+      "step": 692
+    },
+    {
+      "epoch": 0.29282700421940927,
+      "grad_norm": 0.5435477495193481,
+      "learning_rate": 0.000121792618629174,
+      "loss": 1.0351502895355225,
+      "step": 694
+    },
+    {
+      "epoch": 0.2936708860759494,
+      "grad_norm": 0.5712518692016602,
+      "learning_rate": 0.00012214411247803164,
+      "loss": 0.9935672283172607,
+      "step": 696
+    },
+    {
+      "epoch": 0.29451476793248943,
+      "grad_norm": 0.6138222813606262,
+      "learning_rate": 0.00012249560632688928,
+      "loss": 1.0165108442306519,
+      "step": 698
+    },
+    {
+      "epoch": 0.29535864978902954,
+      "grad_norm": 0.4495212435722351,
+      "learning_rate": 0.00012284710017574691,
+      "loss": 0.9334425926208496,
+      "step": 700
+    },
+    {
+      "epoch": 0.29535864978902954,
+      "eval_loss": 0.996929407119751,
+      "eval_runtime": 668.6398,
+      "eval_samples_per_second": 3.151,
+      "eval_steps_per_second": 3.151,
+      "step": 700
+    },
+    {
+      "epoch": 0.29620253164556964,
+      "grad_norm": 0.5321539044380188,
+      "learning_rate": 0.00012319859402460458,
+      "loss": 1.0516537427902222,
+      "step": 702
+    },
+    {
+      "epoch": 0.2970464135021097,
+      "grad_norm": 0.5716516971588135,
+      "learning_rate": 0.00012355008787346222,
+      "loss": 0.9387198686599731,
+      "step": 704
+    },
+    {
+      "epoch": 0.2978902953586498,
+      "grad_norm": 0.5617920160293579,
+      "learning_rate": 0.00012390158172231988,
+      "loss": 0.99737948179245,
+      "step": 706
+    },
+    {
+      "epoch": 0.29873417721518986,
+      "grad_norm": 0.4922899007797241,
+      "learning_rate": 0.00012425307557117752,
+      "loss": 0.9955025911331177,
+      "step": 708
+    },
+    {
+      "epoch": 0.29957805907172996,
+      "grad_norm": 0.543501615524292,
+      "learning_rate": 0.00012460456942003516,
+      "loss": 0.9124280214309692,
+      "step": 710
+    },
+    {
+      "epoch": 0.30042194092827,
+      "grad_norm": 0.49590054154396057,
+      "learning_rate": 0.0001249560632688928,
+      "loss": 0.9820216298103333,
+      "step": 712
+    },
+    {
+      "epoch": 0.3012658227848101,
+      "grad_norm": 0.5984305739402771,
+      "learning_rate": 0.00012530755711775044,
+      "loss": 1.0152074098587036,
+      "step": 714
+    },
+    {
+      "epoch": 0.30210970464135023,
+      "grad_norm": 0.9343504905700684,
+      "learning_rate": 0.00012565905096660807,
+      "loss": 1.0577725172042847,
+      "step": 716
+    },
+    {
+      "epoch": 0.3029535864978903,
+      "grad_norm": 0.5118702054023743,
+      "learning_rate": 0.0001260105448154657,
+      "loss": 0.9830358028411865,
+      "step": 718
+    },
+    {
+      "epoch": 0.3037974683544304,
+      "grad_norm": 0.4940392076969147,
+      "learning_rate": 0.00012636203866432338,
+      "loss": 0.9466043710708618,
+      "step": 720
+    },
+    {
+      "epoch": 0.30464135021097044,
+      "grad_norm": 0.5965693593025208,
+      "learning_rate": 0.00012671353251318104,
+      "loss": 1.015270709991455,
+      "step": 722
+    },
+    {
+      "epoch": 0.30548523206751055,
+      "grad_norm": 0.5020529627799988,
+      "learning_rate": 0.00012706502636203868,
+      "loss": 0.9703927636146545,
+      "step": 724
+    },
+    {
+      "epoch": 0.30632911392405066,
+      "grad_norm": 0.6067010164260864,
+      "learning_rate": 0.00012741652021089632,
+      "loss": 1.0255526304244995,
+      "step": 726
+    },
+    {
+      "epoch": 0.3071729957805907,
+      "grad_norm": 0.5931884050369263,
+      "learning_rate": 0.00012776801405975396,
+      "loss": 0.9335633516311646,
+      "step": 728
+    },
+    {
+      "epoch": 0.3080168776371308,
+      "grad_norm": 0.5938752293586731,
+      "learning_rate": 0.0001281195079086116,
+      "loss": 1.0921578407287598,
+      "step": 730
+    },
+    {
+      "epoch": 0.30886075949367087,
+      "grad_norm": 0.49728086590766907,
+      "learning_rate": 0.00012847100175746923,
+      "loss": 0.963066041469574,
+      "step": 732
+    },
+    {
+      "epoch": 0.309704641350211,
+      "grad_norm": 0.5452080965042114,
+      "learning_rate": 0.0001288224956063269,
+      "loss": 0.9513075351715088,
+      "step": 734
+    },
+    {
+      "epoch": 0.3105485232067511,
+      "grad_norm": 0.5497731566429138,
+      "learning_rate": 0.00012917398945518454,
+      "loss": 0.8576077222824097,
+      "step": 736
+    },
+    {
+      "epoch": 0.31139240506329113,
+      "grad_norm": 0.5580397248268127,
+      "learning_rate": 0.0001295254833040422,
+      "loss": 0.9542577862739563,
+      "step": 738
+    },
+    {
+      "epoch": 0.31223628691983124,
+      "grad_norm": 0.5890427827835083,
+      "learning_rate": 0.00012987697715289984,
+      "loss": 0.8992732167243958,
+      "step": 740
+    },
+    {
+      "epoch": 0.3130801687763713,
+      "grad_norm": 0.5942965745925903,
+      "learning_rate": 0.00013022847100175748,
+      "loss": 1.0322896242141724,
+      "step": 742
+    },
+    {
+      "epoch": 0.3139240506329114,
+      "grad_norm": 0.6341713070869446,
+      "learning_rate": 0.00013057996485061512,
+      "loss": 0.9217103719711304,
+      "step": 744
+    },
+    {
+      "epoch": 0.31476793248945145,
+      "grad_norm": 0.5294105410575867,
+      "learning_rate": 0.00013093145869947276,
+      "loss": 0.951789915561676,
+      "step": 746
+    },
+    {
+      "epoch": 0.31561181434599156,
+      "grad_norm": 0.6372058391571045,
+      "learning_rate": 0.0001312829525483304,
+      "loss": 0.9459875226020813,
+      "step": 748
+    },
+    {
+      "epoch": 0.31645569620253167,
+      "grad_norm": 0.5979796648025513,
+      "learning_rate": 0.00013163444639718806,
+      "loss": 0.9626097679138184,
+      "step": 750
+    },
+    {
+      "epoch": 0.3172995780590717,
+      "grad_norm": 0.5682399868965149,
+      "learning_rate": 0.0001319859402460457,
+      "loss": 1.0261781215667725,
+      "step": 752
+    },
+    {
+      "epoch": 0.3181434599156118,
+      "grad_norm": 0.5349125266075134,
+      "learning_rate": 0.00013233743409490336,
+      "loss": 0.9319828152656555,
+      "step": 754
+    },
+    {
+      "epoch": 0.3189873417721519,
+      "grad_norm": 0.6093934178352356,
+      "learning_rate": 0.000132688927943761,
+      "loss": 0.9216550588607788,
+      "step": 756
+    },
+    {
+      "epoch": 0.319831223628692,
+      "grad_norm": 0.5188612341880798,
+      "learning_rate": 0.00013304042179261864,
+      "loss": 0.901739776134491,
+      "step": 758
+    },
+    {
+      "epoch": 0.3206751054852321,
+      "grad_norm": 0.5877130627632141,
+      "learning_rate": 0.00013339191564147628,
+      "loss": 1.0362589359283447,
+      "step": 760
+    },
+    {
+      "epoch": 0.32151898734177214,
+      "grad_norm": 0.5542771816253662,
+      "learning_rate": 0.00013374340949033392,
+      "loss": 0.8787116408348083,
+      "step": 762
+    },
+    {
+      "epoch": 0.32236286919831225,
+      "grad_norm": 0.5084902048110962,
+      "learning_rate": 0.00013409490333919156,
+      "loss": 0.9237037301063538,
+      "step": 764
+    },
+    {
+      "epoch": 0.3232067510548523,
+      "grad_norm": 0.5461528301239014,
+      "learning_rate": 0.00013444639718804922,
+      "loss": 1.0150731801986694,
+      "step": 766
+    },
+    {
+      "epoch": 0.3240506329113924,
+      "grad_norm": 0.53483647108078,
+      "learning_rate": 0.00013479789103690686,
+      "loss": 0.8985214829444885,
+      "step": 768
+    },
+    {
+      "epoch": 0.32489451476793246,
+      "grad_norm": 0.5580531358718872,
+      "learning_rate": 0.0001351493848857645,
+      "loss": 1.0225775241851807,
+      "step": 770
+    },
+    {
+      "epoch": 0.32573839662447257,
+      "grad_norm": 0.5203377604484558,
+      "learning_rate": 0.00013550087873462216,
+      "loss": 0.9571293592453003,
+      "step": 772
+    },
+    {
+      "epoch": 0.3265822784810127,
+      "grad_norm": 0.5049671530723572,
+      "learning_rate": 0.0001358523725834798,
+      "loss": 1.0468909740447998,
+      "step": 774
+    },
+    {
+      "epoch": 0.32742616033755273,
+      "grad_norm": 0.4723063111305237,
+      "learning_rate": 0.00013620386643233744,
+      "loss": 0.7743215560913086,
+      "step": 776
+    },
+    {
+      "epoch": 0.32827004219409284,
+      "grad_norm": 0.6310980916023254,
+      "learning_rate": 0.00013655536028119508,
+      "loss": 1.021510362625122,
+      "step": 778
+    },
+    {
+      "epoch": 0.3291139240506329,
+      "grad_norm": 0.47066664695739746,
+      "learning_rate": 0.00013690685413005274,
+      "loss": 0.9134382605552673,
+      "step": 780
+    },
+    {
+      "epoch": 0.329957805907173,
+      "grad_norm": 0.5725092887878418,
+      "learning_rate": 0.00013725834797891038,
+      "loss": 0.9797834753990173,
+      "step": 782
+    },
+    {
+      "epoch": 0.3308016877637131,
+      "grad_norm": 0.5139563083648682,
+      "learning_rate": 0.00013760984182776802,
+      "loss": 0.9372621178627014,
+      "step": 784
+    },
+    {
+      "epoch": 0.33164556962025316,
+      "grad_norm": 0.5275821685791016,
+      "learning_rate": 0.00013796133567662566,
+      "loss": 0.9528245329856873,
+      "step": 786
+    },
+    {
+      "epoch": 0.33248945147679326,
+      "grad_norm": 0.5702582001686096,
+      "learning_rate": 0.0001383128295254833,
+      "loss": 0.9750176072120667,
+      "step": 788
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.5281293392181396,
+      "learning_rate": 0.00013866432337434096,
+      "loss": 0.9412306547164917,
+      "step": 790
+    },
+    {
+      "epoch": 0.3341772151898734,
+      "grad_norm": 0.5578986406326294,
+      "learning_rate": 0.0001390158172231986,
+      "loss": 0.997580885887146,
+      "step": 792
+    },
+    {
+      "epoch": 0.33502109704641353,
+      "grad_norm": 0.50461345911026,
+      "learning_rate": 0.00013936731107205624,
+      "loss": 0.9082320928573608,
+      "step": 794
+    },
+    {
+      "epoch": 0.3358649789029536,
+      "grad_norm": 0.5258530378341675,
+      "learning_rate": 0.0001397188049209139,
+      "loss": 1.0082844495773315,
+      "step": 796
+    },
+    {
+      "epoch": 0.3367088607594937,
+      "grad_norm": 0.5548169016838074,
+      "learning_rate": 0.00014007029876977154,
+      "loss": 0.9729003310203552,
+      "step": 798
+    },
+    {
+      "epoch": 0.33755274261603374,
+      "grad_norm": 0.48601076006889343,
+      "learning_rate": 0.00014042179261862918,
+      "loss": 0.9099526405334473,
+      "step": 800
+    },
+    {
+      "epoch": 0.33755274261603374,
+      "eval_loss": 0.9800403714179993,
+      "eval_runtime": 678.8306,
+      "eval_samples_per_second": 3.104,
+      "eval_steps_per_second": 3.104,
+      "step": 800
+    },
+    {
+      "epoch": 0.33839662447257385,
+      "grad_norm": 0.5413158535957336,
+      "learning_rate": 0.00014077328646748682,
+      "loss": 0.8610644936561584,
+      "step": 802
+    },
+    {
+      "epoch": 0.3392405063291139,
+      "grad_norm": 0.5147035717964172,
+      "learning_rate": 0.00014112478031634446,
+      "loss": 0.9584825038909912,
+      "step": 804
+    },
+    {
+      "epoch": 0.340084388185654,
+      "grad_norm": 0.5931771397590637,
+      "learning_rate": 0.0001414762741652021,
+      "loss": 1.0142558813095093,
+      "step": 806
+    },
+    {
+      "epoch": 0.3409282700421941,
+      "grad_norm": 0.5178377032279968,
+      "learning_rate": 0.00014182776801405976,
+      "loss": 1.0078763961791992,
+      "step": 808
+    },
+    {
+      "epoch": 0.34177215189873417,
+      "grad_norm": 0.5453237295150757,
+      "learning_rate": 0.0001421792618629174,
+      "loss": 0.9107215404510498,
+      "step": 810
+    },
+    {
+      "epoch": 0.3426160337552743,
+      "grad_norm": 0.5886152982711792,
+      "learning_rate": 0.00014253075571177506,
+      "loss": 0.9981362819671631,
+      "step": 812
+    },
+    {
+      "epoch": 0.3434599156118143,
+      "grad_norm": 0.48040178418159485,
+      "learning_rate": 0.0001428822495606327,
+      "loss": 0.9636131525039673,
+      "step": 814
+    },
+    {
+      "epoch": 0.34430379746835443,
+      "grad_norm": 0.5011753439903259,
+      "learning_rate": 0.00014323374340949034,
+      "loss": 0.9590586423873901,
+      "step": 816
+    },
+    {
+      "epoch": 0.34514767932489454,
+      "grad_norm": 0.57858806848526,
+      "learning_rate": 0.00014358523725834798,
+      "loss": 0.978246808052063,
+      "step": 818
+    },
+    {
+      "epoch": 0.3459915611814346,
+      "grad_norm": 0.46092939376831055,
+      "learning_rate": 0.00014393673110720562,
+      "loss": 0.8549934029579163,
+      "step": 820
+    },
+    {
+      "epoch": 0.3468354430379747,
+      "grad_norm": 0.5756489038467407,
+      "learning_rate": 0.00014428822495606326,
+      "loss": 0.9771265387535095,
+      "step": 822
+    },
+    {
+      "epoch": 0.34767932489451475,
+      "grad_norm": 0.5501731634140015,
+      "learning_rate": 0.00014463971880492092,
+      "loss": 0.8739748001098633,
+      "step": 824
+    },
+    {
+      "epoch": 0.34852320675105486,
+      "grad_norm": 0.5451868176460266,
+      "learning_rate": 0.0001449912126537786,
+      "loss": 0.9129468202590942,
+      "step": 826
+    },
+    {
+      "epoch": 0.3493670886075949,
+      "grad_norm": 0.4624619781970978,
+      "learning_rate": 0.00014534270650263623,
+      "loss": 0.9196704030036926,
+      "step": 828
+    },
+    {
+      "epoch": 0.350210970464135,
+      "grad_norm": 0.520878791809082,
+      "learning_rate": 0.00014569420035149386,
+      "loss": 0.9976527690887451,
+      "step": 830
+    },
+    {
+      "epoch": 0.3510548523206751,
+      "grad_norm": 0.4469171464443207,
+      "learning_rate": 0.0001460456942003515,
+      "loss": 0.7753443717956543,
+      "step": 832
+    },
+    {
+      "epoch": 0.3518987341772152,
+      "grad_norm": 0.5105249881744385,
+      "learning_rate": 0.00014639718804920914,
+      "loss": 0.9584846496582031,
+      "step": 834
+    },
+    {
+      "epoch": 0.3527426160337553,
+      "grad_norm": 0.5043913125991821,
+      "learning_rate": 0.00014674868189806678,
+      "loss": 0.9013500213623047,
+      "step": 836
+    },
+    {
+      "epoch": 0.35358649789029534,
+      "grad_norm": 0.575850784778595,
+      "learning_rate": 0.00014710017574692442,
+      "loss": 0.9775562286376953,
+      "step": 838
+    },
+    {
+      "epoch": 0.35443037974683544,
+      "grad_norm": 0.5128876566886902,
+      "learning_rate": 0.00014745166959578208,
+      "loss": 0.9278940558433533,
+      "step": 840
+    },
+    {
+      "epoch": 0.35527426160337555,
+      "grad_norm": 0.5757885575294495,
+      "learning_rate": 0.00014780316344463975,
+      "loss": 1.0091488361358643,
+      "step": 842
+    },
+    {
+      "epoch": 0.3561181434599156,
+      "grad_norm": 0.500934898853302,
+      "learning_rate": 0.00014815465729349739,
+      "loss": 0.9286836981773376,
+      "step": 844
+    },
+    {
+      "epoch": 0.3569620253164557,
+      "grad_norm": 0.5220686197280884,
+      "learning_rate": 0.00014850615114235502,
+      "loss": 0.9484171867370605,
+      "step": 846
+    },
+    {
+      "epoch": 0.35780590717299576,
+      "grad_norm": 0.5494697690010071,
+      "learning_rate": 0.00014885764499121266,
+      "loss": 1.0556397438049316,
+      "step": 848
+    },
+    {
+      "epoch": 0.35864978902953587,
+      "grad_norm": 0.46633943915367126,
+      "learning_rate": 0.0001492091388400703,
+      "loss": 0.933089017868042,
+      "step": 850
+    },
+    {
+      "epoch": 0.3594936708860759,
+      "grad_norm": 0.4864962697029114,
+      "learning_rate": 0.00014956063268892794,
+      "loss": 1.0016963481903076,
+      "step": 852
+    },
+    {
+      "epoch": 0.36033755274261603,
+      "grad_norm": 0.5032764673233032,
+      "learning_rate": 0.0001499121265377856,
+      "loss": 0.9085348844528198,
+      "step": 854
+    },
+    {
+      "epoch": 0.36118143459915614,
+      "grad_norm": 0.5424998998641968,
+      "learning_rate": 0.00015026362038664324,
+      "loss": 0.923502504825592,
+      "step": 856
+    },
+    {
+      "epoch": 0.3620253164556962,
+      "grad_norm": 0.5181655287742615,
+      "learning_rate": 0.00015061511423550088,
+      "loss": 0.919174313545227,
+      "step": 858
+    },
+    {
+      "epoch": 0.3628691983122363,
+      "grad_norm": 0.5129443407058716,
+      "learning_rate": 0.00015096660808435855,
+      "loss": 0.851981520652771,
+      "step": 860
+    },
+    {
+      "epoch": 0.36371308016877635,
+      "grad_norm": 0.49540698528289795,
+      "learning_rate": 0.00015131810193321618,
+      "loss": 0.8633858561515808,
+      "step": 862
+    },
+    {
+      "epoch": 0.36455696202531646,
+      "grad_norm": 0.4706701934337616,
+      "learning_rate": 0.00015166959578207382,
+      "loss": 0.9473677277565002,
+      "step": 864
+    },
+    {
+      "epoch": 0.36540084388185656,
+      "grad_norm": 0.587704598903656,
+      "learning_rate": 0.00015202108963093146,
+      "loss": 0.942383885383606,
+      "step": 866
+    },
+    {
+      "epoch": 0.3662447257383966,
+      "grad_norm": 0.5851273536682129,
+      "learning_rate": 0.0001523725834797891,
+      "loss": 0.9811251163482666,
+      "step": 868
+    },
+    {
+      "epoch": 0.3670886075949367,
+      "grad_norm": 0.46357613801956177,
+      "learning_rate": 0.00015272407732864676,
+      "loss": 0.8151084184646606,
+      "step": 870
+    },
+    {
+      "epoch": 0.3679324894514768,
+      "grad_norm": 0.47250670194625854,
+      "learning_rate": 0.0001530755711775044,
+      "loss": 0.9023333191871643,
+      "step": 872
+    },
+    {
+      "epoch": 0.3687763713080169,
+      "grad_norm": 0.47510042786598206,
+      "learning_rate": 0.00015342706502636204,
+      "loss": 0.8758499622344971,
+      "step": 874
+    },
+    {
+      "epoch": 0.369620253164557,
+      "grad_norm": 0.5687124729156494,
+      "learning_rate": 0.00015377855887521968,
+      "loss": 0.9837421774864197,
+      "step": 876
+    },
+    {
+      "epoch": 0.37046413502109704,
+      "grad_norm": 0.49064236879348755,
+      "learning_rate": 0.00015413005272407735,
+      "loss": 0.9084216356277466,
+      "step": 878
+    },
+    {
+      "epoch": 0.37130801687763715,
+      "grad_norm": 0.5288164615631104,
+      "learning_rate": 0.00015448154657293498,
+      "loss": 0.8425542712211609,
+      "step": 880
+    },
+    {
+      "epoch": 0.3721518987341772,
+      "grad_norm": 0.5446951985359192,
+      "learning_rate": 0.00015483304042179262,
+      "loss": 1.006197452545166,
+      "step": 882
+    },
+    {
+      "epoch": 0.3729957805907173,
+      "grad_norm": 0.46872642636299133,
+      "learning_rate": 0.00015518453427065026,
+      "loss": 0.9779444932937622,
+      "step": 884
+    },
+    {
+      "epoch": 0.37383966244725736,
+      "grad_norm": 0.4852714240550995,
+      "learning_rate": 0.00015553602811950793,
+      "loss": 0.9099963307380676,
+      "step": 886
+    },
+    {
+      "epoch": 0.37468354430379747,
+      "grad_norm": 0.5219841003417969,
+      "learning_rate": 0.00015588752196836556,
+      "loss": 0.9730570316314697,
+      "step": 888
+    },
+    {
+      "epoch": 0.3755274261603376,
+      "grad_norm": 0.5258626341819763,
+      "learning_rate": 0.0001562390158172232,
+      "loss": 0.9304586052894592,
+      "step": 890
+    },
+    {
+      "epoch": 0.3763713080168776,
+      "grad_norm": 0.4266716241836548,
+      "learning_rate": 0.00015659050966608084,
+      "loss": 0.8680716156959534,
+      "step": 892
+    },
+    {
+      "epoch": 0.37721518987341773,
+      "grad_norm": 0.46361327171325684,
+      "learning_rate": 0.0001569420035149385,
+      "loss": 0.8428018093109131,
+      "step": 894
+    },
+    {
+      "epoch": 0.3780590717299578,
+      "grad_norm": 0.5313687920570374,
+      "learning_rate": 0.00015729349736379614,
+      "loss": 0.8465750217437744,
+      "step": 896
+    },
+    {
+      "epoch": 0.3789029535864979,
+      "grad_norm": 0.47166210412979126,
+      "learning_rate": 0.00015764499121265378,
+      "loss": 0.9120327234268188,
+      "step": 898
+    },
+    {
+      "epoch": 0.379746835443038,
+      "grad_norm": 0.4647318720817566,
+      "learning_rate": 0.00015799648506151145,
+      "loss": 0.8950425982475281,
+      "step": 900
+    },
+    {
+      "epoch": 0.379746835443038,
+      "eval_loss": 0.9643027186393738,
+      "eval_runtime": 691.7929,
+      "eval_samples_per_second": 3.046,
+      "eval_steps_per_second": 3.046,
+      "step": 900
+    },
+    {
+      "epoch": 0.38059071729957805,
+      "grad_norm": 0.5445119738578796,
+      "learning_rate": 0.00015834797891036909,
+      "loss": 0.908163845539093,
+      "step": 902
+    },
+    {
+      "epoch": 0.38143459915611816,
+      "grad_norm": 0.4311858117580414,
+      "learning_rate": 0.00015869947275922672,
+      "loss": 0.8945821523666382,
+      "step": 904
+    },
+    {
+      "epoch": 0.3822784810126582,
+      "grad_norm": 0.5590984225273132,
+      "learning_rate": 0.00015905096660808436,
+      "loss": 0.9478458762168884,
+      "step": 906
+    },
+    {
+      "epoch": 0.3831223628691983,
+      "grad_norm": 0.5470241904258728,
+      "learning_rate": 0.000159402460456942,
+      "loss": 0.9259957671165466,
+      "step": 908
+    },
+    {
+      "epoch": 0.38396624472573837,
+      "grad_norm": 0.5498791337013245,
+      "learning_rate": 0.00015975395430579964,
+      "loss": 0.8824930787086487,
+      "step": 910
+    },
+    {
+      "epoch": 0.3848101265822785,
+      "grad_norm": 0.4779198467731476,
+      "learning_rate": 0.0001601054481546573,
+      "loss": 0.8842340707778931,
+      "step": 912
+    },
+    {
+      "epoch": 0.3856540084388186,
+      "grad_norm": 0.5390620827674866,
+      "learning_rate": 0.00016045694200351494,
+      "loss": 0.92950040102005,
+      "step": 914
+    },
+    {
+      "epoch": 0.38649789029535864,
+      "grad_norm": 0.505519688129425,
+      "learning_rate": 0.0001608084358523726,
+      "loss": 0.8420897126197815,
+      "step": 916
+    },
+    {
+      "epoch": 0.38734177215189874,
+      "grad_norm": 0.4463907778263092,
+      "learning_rate": 0.00016115992970123025,
+      "loss": 0.8256624341011047,
+      "step": 918
+    },
+    {
+      "epoch": 0.3881856540084388,
+      "grad_norm": 0.5321422219276428,
+      "learning_rate": 0.00016151142355008788,
+      "loss": 0.8701168298721313,
+      "step": 920
+    },
+    {
+      "epoch": 0.3890295358649789,
+      "grad_norm": 0.4343073070049286,
+      "learning_rate": 0.00016186291739894552,
+      "loss": 0.9019309878349304,
+      "step": 922
+    },
+    {
+      "epoch": 0.389873417721519,
+      "grad_norm": 0.5311984419822693,
+      "learning_rate": 0.00016221441124780316,
+      "loss": 0.8560551404953003,
+      "step": 924
+    },
+    {
+      "epoch": 0.39071729957805906,
+      "grad_norm": 0.588691771030426,
+      "learning_rate": 0.0001625659050966608,
+      "loss": 0.9700050354003906,
+      "step": 926
+    },
+    {
+      "epoch": 0.39156118143459917,
+      "grad_norm": 0.5425586104393005,
+      "learning_rate": 0.00016291739894551844,
+      "loss": 1.043768048286438,
+      "step": 928
+    },
+    {
+      "epoch": 0.3924050632911392,
+      "grad_norm": 0.5228736996650696,
+      "learning_rate": 0.0001632688927943761,
+      "loss": 0.9501712918281555,
+      "step": 930
+    },
+    {
+      "epoch": 0.39324894514767933,
+      "grad_norm": 0.48960360884666443,
+      "learning_rate": 0.00016362038664323377,
+      "loss": 0.9223058223724365,
+      "step": 932
+    },
+    {
+      "epoch": 0.39409282700421944,
+      "grad_norm": 0.45204755663871765,
+      "learning_rate": 0.0001639718804920914,
+      "loss": 0.9692960977554321,
+      "step": 934
+    },
+    {
+      "epoch": 0.3949367088607595,
+      "grad_norm": 0.5299274921417236,
+      "learning_rate": 0.00016432337434094905,
+      "loss": 0.9467466473579407,
+      "step": 936
+    },
+    {
+      "epoch": 0.3957805907172996,
+      "grad_norm": 0.5607715249061584,
+      "learning_rate": 0.00016467486818980668,
+      "loss": 0.9118053317070007,
+      "step": 938
+    },
+    {
+      "epoch": 0.39662447257383965,
+      "grad_norm": 0.5271831154823303,
+      "learning_rate": 0.00016502636203866432,
+      "loss": 0.9131460189819336,
+      "step": 940
+    },
+    {
+      "epoch": 0.39746835443037976,
+      "grad_norm": 0.5075286030769348,
+      "learning_rate": 0.00016537785588752196,
+      "loss": 0.9358300566673279,
+      "step": 942
+    },
+    {
+      "epoch": 0.3983122362869198,
+      "grad_norm": 0.515731155872345,
+      "learning_rate": 0.00016572934973637963,
+      "loss": 0.8908210396766663,
+      "step": 944
+    },
+    {
+      "epoch": 0.3991561181434599,
+      "grad_norm": 0.4856977164745331,
+      "learning_rate": 0.00016608084358523726,
+      "loss": 0.9775290489196777,
+      "step": 946
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.48846355080604553,
+      "learning_rate": 0.00016643233743409493,
+      "loss": 0.8957490921020508,
+      "step": 948
+    },
+    {
+      "epoch": 0.4008438818565401,
+      "grad_norm": 0.42990800738334656,
+      "learning_rate": 0.00016678383128295257,
+      "loss": 0.9036174416542053,
+      "step": 950
+    },
+    {
+      "epoch": 0.4016877637130802,
+      "grad_norm": 0.49552062153816223,
+      "learning_rate": 0.0001671353251318102,
+      "loss": 0.991032600402832,
+      "step": 952
+    },
+    {
+      "epoch": 0.40253164556962023,
+      "grad_norm": 0.4565040171146393,
+      "learning_rate": 0.00016748681898066784,
+      "loss": 0.823063313961029,
+      "step": 954
+    },
+    {
+      "epoch": 0.40337552742616034,
+      "grad_norm": 0.4290153682231903,
+      "learning_rate": 0.00016783831282952548,
+      "loss": 0.8785063624382019,
+      "step": 956
+    },
+    {
+      "epoch": 0.40421940928270045,
+      "grad_norm": 0.5419702529907227,
+      "learning_rate": 0.00016818980667838312,
+      "loss": 0.8763971924781799,
+      "step": 958
+    },
+    {
+      "epoch": 0.4050632911392405,
+      "grad_norm": 0.5177501440048218,
+      "learning_rate": 0.00016854130052724079,
+      "loss": 0.9470553398132324,
+      "step": 960
+    },
+    {
+      "epoch": 0.4059071729957806,
+      "grad_norm": 0.539725661277771,
+      "learning_rate": 0.00016889279437609842,
+      "loss": 0.9235025644302368,
+      "step": 962
+    },
+    {
+      "epoch": 0.40675105485232066,
+      "grad_norm": 0.5324983596801758,
+      "learning_rate": 0.0001692442882249561,
+      "loss": 1.0248996019363403,
+      "step": 964
+    },
+    {
+      "epoch": 0.40759493670886077,
+      "grad_norm": 0.4936407506465912,
+      "learning_rate": 0.00016959578207381373,
+      "loss": 0.9076873660087585,
+      "step": 966
+    },
+    {
+      "epoch": 0.4084388185654008,
+      "grad_norm": 0.4960501194000244,
+      "learning_rate": 0.00016994727592267137,
+      "loss": 0.9162673950195312,
+      "step": 968
+    },
+    {
+      "epoch": 0.4092827004219409,
+      "grad_norm": 0.45093682408332825,
+      "learning_rate": 0.000170298769771529,
+      "loss": 0.904100775718689,
+      "step": 970
+    },
+    {
+      "epoch": 0.41012658227848103,
+      "grad_norm": 0.4560275077819824,
+      "learning_rate": 0.00017065026362038664,
+      "loss": 0.8633337020874023,
+      "step": 972
+    },
+    {
+      "epoch": 0.4109704641350211,
+      "grad_norm": 0.44885000586509705,
+      "learning_rate": 0.00017100175746924428,
+      "loss": 0.8454209566116333,
+      "step": 974
+    },
+    {
+      "epoch": 0.4118143459915612,
+      "grad_norm": 0.47251659631729126,
+      "learning_rate": 0.00017135325131810195,
+      "loss": 0.824730396270752,
+      "step": 976
+    },
+    {
+      "epoch": 0.41265822784810124,
+      "grad_norm": 0.6597666144371033,
+      "learning_rate": 0.00017170474516695959,
+      "loss": 0.9496501684188843,
+      "step": 978
+    },
+    {
+      "epoch": 0.41350210970464135,
+      "grad_norm": 0.39806297421455383,
+      "learning_rate": 0.00017205623901581722,
+      "loss": 0.9419087171554565,
+      "step": 980
+    },
+    {
+      "epoch": 0.41434599156118146,
+      "grad_norm": 0.48231109976768494,
+      "learning_rate": 0.0001724077328646749,
+      "loss": 0.9182976484298706,
+      "step": 982
+    },
+    {
+      "epoch": 0.4151898734177215,
+      "grad_norm": 0.5438776612281799,
+      "learning_rate": 0.00017275922671353253,
+      "loss": 0.9386967420578003,
+      "step": 984
+    },
+    {
+      "epoch": 0.4160337552742616,
+      "grad_norm": 0.4959667921066284,
+      "learning_rate": 0.00017311072056239017,
+      "loss": 0.897849440574646,
+      "step": 986
+    },
+    {
+      "epoch": 0.41687763713080167,
+      "grad_norm": 0.43533357977867126,
+      "learning_rate": 0.0001734622144112478,
+      "loss": 0.8776953816413879,
+      "step": 988
+    },
+    {
+      "epoch": 0.4177215189873418,
+      "grad_norm": 0.47513946890830994,
+      "learning_rate": 0.00017381370826010547,
+      "loss": 0.9162989854812622,
+      "step": 990
+    },
+    {
+      "epoch": 0.41856540084388183,
+      "grad_norm": 0.4907188415527344,
+      "learning_rate": 0.0001741652021089631,
+      "loss": 0.9482660889625549,
+      "step": 992
+    },
+    {
+      "epoch": 0.41940928270042194,
+      "grad_norm": 0.44499966502189636,
+      "learning_rate": 0.00017451669595782075,
+      "loss": 0.8812930583953857,
+      "step": 994
+    },
+    {
+      "epoch": 0.42025316455696204,
+      "grad_norm": 0.4535730481147766,
+      "learning_rate": 0.00017486818980667838,
+      "loss": 0.9439874887466431,
+      "step": 996
+    },
+    {
+      "epoch": 0.4210970464135021,
+      "grad_norm": 0.5240745544433594,
+      "learning_rate": 0.00017521968365553602,
+      "loss": 0.8818395137786865,
+      "step": 998
+    },
+    {
+      "epoch": 0.4219409282700422,
+      "grad_norm": 0.5301211476325989,
+      "learning_rate": 0.0001755711775043937,
+      "loss": 0.886186957359314,
+      "step": 1000
+    },
+    {
+      "epoch": 0.4219409282700422,
+      "eval_loss": 0.9487298727035522,
+      "eval_runtime": 689.4288,
+      "eval_samples_per_second": 3.056,
+      "eval_steps_per_second": 3.056,
+      "step": 1000
+    },
+    {
+      "epoch": 0.42278481012658226,
+      "grad_norm": 0.47876957058906555,
+      "learning_rate": 0.00017592267135325133,
+      "loss": 0.8814021348953247,
+      "step": 1002
+    },
+    {
+      "epoch": 0.42362869198312236,
+      "grad_norm": 0.4929780960083008,
+      "learning_rate": 0.00017627416520210896,
+      "loss": 0.8295068740844727,
+      "step": 1004
+    },
+    {
+      "epoch": 0.42447257383966247,
+      "grad_norm": 0.41888436675071716,
+      "learning_rate": 0.00017662565905096663,
+      "loss": 0.8364827036857605,
+      "step": 1006
+    },
+    {
+      "epoch": 0.4253164556962025,
+      "grad_norm": 0.5175151824951172,
+      "learning_rate": 0.00017697715289982427,
+      "loss": 1.0167189836502075,
+      "step": 1008
+    },
+    {
+      "epoch": 0.42616033755274263,
+      "grad_norm": 0.4815356135368347,
+      "learning_rate": 0.0001773286467486819,
+      "loss": 0.8460752964019775,
+      "step": 1010
+    },
+    {
+      "epoch": 0.4270042194092827,
+      "grad_norm": 0.5210875272750854,
+      "learning_rate": 0.00017768014059753954,
+      "loss": 0.8595574498176575,
+      "step": 1012
+    },
+    {
+      "epoch": 0.4278481012658228,
+      "grad_norm": 0.4400486946105957,
+      "learning_rate": 0.00017803163444639718,
+      "loss": 0.8764723539352417,
+      "step": 1014
+    },
+    {
+      "epoch": 0.4286919831223629,
+      "grad_norm": 0.5282127857208252,
+      "learning_rate": 0.00017838312829525482,
+      "loss": 0.9706798791885376,
+      "step": 1016
+    },
+    {
+      "epoch": 0.42953586497890295,
+      "grad_norm": 0.47523441910743713,
+      "learning_rate": 0.00017873462214411249,
+      "loss": 0.8912002444267273,
+      "step": 1018
+    },
+    {
+      "epoch": 0.43037974683544306,
+      "grad_norm": 0.49640706181526184,
+      "learning_rate": 0.00017908611599297012,
+      "loss": 0.8835636973381042,
+      "step": 1020
+    },
+    {
+      "epoch": 0.4312236286919831,
+      "grad_norm": 0.5253039002418518,
+      "learning_rate": 0.0001794376098418278,
+      "loss": 0.8711735606193542,
+      "step": 1022
+    },
+    {
+      "epoch": 0.4320675105485232,
+      "grad_norm": 0.49285009503364563,
+      "learning_rate": 0.00017978910369068543,
+      "loss": 0.9064869284629822,
+      "step": 1024
+    },
+    {
+      "epoch": 0.43291139240506327,
+      "grad_norm": 0.48758041858673096,
+      "learning_rate": 0.00018014059753954307,
+      "loss": 0.8993359208106995,
+      "step": 1026
+    },
+    {
+      "epoch": 0.4337552742616034,
+      "grad_norm": 0.4815461039543152,
+      "learning_rate": 0.0001804920913884007,
+      "loss": 0.8747937679290771,
+      "step": 1028
+    },
+    {
+      "epoch": 0.4345991561181435,
+      "grad_norm": 0.4930349886417389,
+      "learning_rate": 0.00018084358523725834,
+      "loss": 0.8879084587097168,
+      "step": 1030
+    },
+    {
+      "epoch": 0.43544303797468353,
+      "grad_norm": 0.49303027987480164,
+      "learning_rate": 0.00018119507908611598,
+      "loss": 1.0023083686828613,
+      "step": 1032
+    },
+    {
+      "epoch": 0.43628691983122364,
+      "grad_norm": 0.5312249064445496,
+      "learning_rate": 0.00018154657293497365,
+      "loss": 0.938680112361908,
+      "step": 1034
+    },
+    {
+      "epoch": 0.4371308016877637,
+      "grad_norm": 0.5174582004547119,
+      "learning_rate": 0.0001818980667838313,
+      "loss": 0.854195773601532,
+      "step": 1036
+    },
+    {
+      "epoch": 0.4379746835443038,
+      "grad_norm": 0.5452545881271362,
+      "learning_rate": 0.00018224956063268895,
+      "loss": 0.9059375524520874,
+      "step": 1038
+    },
+    {
+      "epoch": 0.4388185654008439,
+      "grad_norm": 0.5480839014053345,
+      "learning_rate": 0.0001826010544815466,
+      "loss": 0.9708920121192932,
+      "step": 1040
+    },
+    {
+      "epoch": 0.43966244725738396,
+      "grad_norm": 0.44693151116371155,
+      "learning_rate": 0.00018295254833040423,
+      "loss": 0.8561046123504639,
+      "step": 1042
+    },
+    {
+      "epoch": 0.44050632911392407,
+      "grad_norm": 0.4024234116077423,
+      "learning_rate": 0.00018330404217926187,
+      "loss": 0.8811968564987183,
+      "step": 1044
+    },
+    {
+      "epoch": 0.4413502109704641,
+      "grad_norm": 0.5586408376693726,
+      "learning_rate": 0.0001836555360281195,
+      "loss": 0.978068470954895,
+      "step": 1046
+    },
+    {
+      "epoch": 0.4421940928270042,
+      "grad_norm": 0.4918624758720398,
+      "learning_rate": 0.00018400702987697714,
+      "loss": 0.8640981316566467,
+      "step": 1048
+    },
+    {
+      "epoch": 0.4430379746835443,
+      "grad_norm": 0.4335230588912964,
+      "learning_rate": 0.0001843585237258348,
+      "loss": 0.9077964425086975,
+      "step": 1050
+    },
+    {
+      "epoch": 0.4438818565400844,
+      "grad_norm": 0.5275123715400696,
+      "learning_rate": 0.00018471001757469247,
+      "loss": 0.960682213306427,
+      "step": 1052
+    },
+    {
+      "epoch": 0.4447257383966245,
+      "grad_norm": 0.5144415497779846,
+      "learning_rate": 0.0001850615114235501,
+      "loss": 0.8549577593803406,
+      "step": 1054
+    },
+    {
+      "epoch": 0.44556962025316454,
+      "grad_norm": 0.5166662931442261,
+      "learning_rate": 0.00018541300527240775,
+      "loss": 0.9320827126502991,
+      "step": 1056
+    },
+    {
+      "epoch": 0.44641350210970465,
+      "grad_norm": 0.5011980533599854,
+      "learning_rate": 0.0001857644991212654,
+      "loss": 0.9632431864738464,
+      "step": 1058
+    },
+    {
+      "epoch": 0.4472573839662447,
+      "grad_norm": 0.49312469363212585,
+      "learning_rate": 0.00018611599297012303,
+      "loss": 0.9442946910858154,
+      "step": 1060
+    },
+    {
+      "epoch": 0.4481012658227848,
+      "grad_norm": 0.49958568811416626,
+      "learning_rate": 0.00018646748681898066,
+      "loss": 0.952802300453186,
+      "step": 1062
+    },
+    {
+      "epoch": 0.4489451476793249,
+      "grad_norm": 0.49827462434768677,
+      "learning_rate": 0.00018681898066783833,
+      "loss": 0.9630650877952576,
+      "step": 1064
+    },
+    {
+      "epoch": 0.44978902953586497,
+      "grad_norm": 0.523980438709259,
+      "learning_rate": 0.00018717047451669597,
+      "loss": 0.9081395268440247,
+      "step": 1066
+    },
+    {
+      "epoch": 0.4506329113924051,
+      "grad_norm": 0.5108568668365479,
+      "learning_rate": 0.0001875219683655536,
+      "loss": 0.9398958683013916,
+      "step": 1068
+    },
+    {
+      "epoch": 0.45147679324894513,
+      "grad_norm": 0.4453965127468109,
+      "learning_rate": 0.00018787346221441127,
+      "loss": 0.93592768907547,
+      "step": 1070
+    },
+    {
+      "epoch": 0.45232067510548524,
+      "grad_norm": 0.4675683081150055,
+      "learning_rate": 0.0001882249560632689,
+      "loss": 0.8879633545875549,
+      "step": 1072
+    },
+    {
+      "epoch": 0.4531645569620253,
+      "grad_norm": 0.47398847341537476,
+      "learning_rate": 0.00018857644991212655,
+      "loss": 0.9966004490852356,
+      "step": 1074
+    },
+    {
+      "epoch": 0.4540084388185654,
+      "grad_norm": 0.46053192019462585,
+      "learning_rate": 0.0001889279437609842,
+      "loss": 0.9184179902076721,
+      "step": 1076
+    },
+    {
+      "epoch": 0.4548523206751055,
+      "grad_norm": 0.5601398348808289,
+      "learning_rate": 0.00018927943760984182,
+      "loss": 0.9434974193572998,
+      "step": 1078
+    },
+    {
+      "epoch": 0.45569620253164556,
+      "grad_norm": 0.48422637581825256,
+      "learning_rate": 0.0001896309314586995,
+      "loss": 0.9522465467453003,
+      "step": 1080
+    },
+    {
+      "epoch": 0.45654008438818566,
+      "grad_norm": 0.46280911564826965,
+      "learning_rate": 0.00018998242530755713,
+      "loss": 0.9315434694290161,
+      "step": 1082
+    },
+    {
+      "epoch": 0.4573839662447257,
+      "grad_norm": 0.4658683240413666,
+      "learning_rate": 0.00019033391915641477,
+      "loss": 0.8165783286094666,
+      "step": 1084
+    },
+    {
+      "epoch": 0.4582278481012658,
+      "grad_norm": 0.4192182123661041,
+      "learning_rate": 0.0001906854130052724,
+      "loss": 0.8024274110794067,
+      "step": 1086
+    },
+    {
+      "epoch": 0.45907172995780593,
+      "grad_norm": 0.4178735613822937,
+      "learning_rate": 0.00019103690685413007,
+      "loss": 0.8620653748512268,
+      "step": 1088
+    },
+    {
+      "epoch": 0.459915611814346,
+      "grad_norm": 0.4488574266433716,
+      "learning_rate": 0.0001913884007029877,
+      "loss": 0.9121530055999756,
+      "step": 1090
+    },
+    {
+      "epoch": 0.4607594936708861,
+      "grad_norm": 0.5164965987205505,
+      "learning_rate": 0.00019173989455184535,
+      "loss": 0.9496700763702393,
+      "step": 1092
+    },
+    {
+      "epoch": 0.46160337552742614,
+      "grad_norm": 0.41563132405281067,
+      "learning_rate": 0.00019209138840070299,
+      "loss": 0.8757708668708801,
+      "step": 1094
+    },
+    {
+      "epoch": 0.46244725738396625,
+      "grad_norm": 2.3222429752349854,
+      "learning_rate": 0.00019244288224956065,
+      "loss": 0.8406533002853394,
+      "step": 1096
+    },
+    {
+      "epoch": 0.46329113924050636,
+      "grad_norm": 0.4813845157623291,
+      "learning_rate": 0.0001927943760984183,
+      "loss": 0.9459465742111206,
+      "step": 1098
+    },
+    {
+      "epoch": 0.4641350210970464,
+      "grad_norm": 0.6233882308006287,
+      "learning_rate": 0.00019314586994727593,
+      "loss": 0.9376904368400574,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4641350210970464,
+      "eval_loss": 0.9357889294624329,
+      "eval_runtime": 676.9573,
+      "eval_samples_per_second": 3.112,
+      "eval_steps_per_second": 3.112,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4649789029535865,
+      "grad_norm": 0.5125579237937927,
+      "learning_rate": 0.00019349736379613357,
+      "loss": 0.8998825550079346,
+      "step": 1102
+    },
+    {
+      "epoch": 0.46582278481012657,
+      "grad_norm": 0.4534320831298828,
+      "learning_rate": 0.00019384885764499123,
+      "loss": 0.9213768839836121,
+      "step": 1104
+    },
+    {
+      "epoch": 0.4666666666666667,
+      "grad_norm": 0.4715143144130707,
+      "learning_rate": 0.00019420035149384887,
+      "loss": 0.9739661812782288,
+      "step": 1106
+    },
+    {
+      "epoch": 0.4675105485232067,
+      "grad_norm": 0.45529672503471375,
+      "learning_rate": 0.0001945518453427065,
+      "loss": 0.9113216400146484,
+      "step": 1108
+    },
+    {
+      "epoch": 0.46835443037974683,
+      "grad_norm": 0.48199015855789185,
+      "learning_rate": 0.00019490333919156417,
+      "loss": 0.92528235912323,
+      "step": 1110
+    },
+    {
+      "epoch": 0.46919831223628694,
+      "grad_norm": 0.4425188899040222,
+      "learning_rate": 0.0001952548330404218,
+      "loss": 0.8612716197967529,
+      "step": 1112
+    },
+    {
+      "epoch": 0.470042194092827,
+      "grad_norm": 0.4311593770980835,
+      "learning_rate": 0.00019560632688927945,
+      "loss": 0.8901699185371399,
+      "step": 1114
+    },
+    {
+      "epoch": 0.4708860759493671,
+      "grad_norm": 0.497806578874588,
+      "learning_rate": 0.0001959578207381371,
+      "loss": 0.9271994829177856,
+      "step": 1116
+    },
+    {
+      "epoch": 0.47172995780590715,
+      "grad_norm": 0.47149473428726196,
+      "learning_rate": 0.00019630931458699473,
+      "loss": 0.8740925788879395,
+      "step": 1118
+    },
+    {
+      "epoch": 0.47257383966244726,
+      "grad_norm": 0.5029966235160828,
+      "learning_rate": 0.00019666080843585236,
+      "loss": 0.9190115928649902,
+      "step": 1120
+    },
+    {
+      "epoch": 0.47341772151898737,
+      "grad_norm": 0.47972601652145386,
+      "learning_rate": 0.00019701230228471003,
+      "loss": 1.0043057203292847,
+      "step": 1122
+    },
+    {
+      "epoch": 0.4742616033755274,
+      "grad_norm": 0.45096471905708313,
+      "learning_rate": 0.00019736379613356767,
+      "loss": 0.9472925066947937,
+      "step": 1124
+    },
+    {
+      "epoch": 0.4751054852320675,
+      "grad_norm": 0.459852933883667,
+      "learning_rate": 0.00019771528998242533,
+      "loss": 0.8043124079704285,
+      "step": 1126
+    },
+    {
+      "epoch": 0.4759493670886076,
+      "grad_norm": 0.5313422679901123,
+      "learning_rate": 0.00019806678383128297,
+      "loss": 0.9662142395973206,
+      "step": 1128
+    },
+    {
+      "epoch": 0.4767932489451477,
+      "grad_norm": 0.43474531173706055,
+      "learning_rate": 0.0001984182776801406,
+      "loss": 0.8691151738166809,
+      "step": 1130
+    },
+    {
+      "epoch": 0.47763713080168774,
+      "grad_norm": 0.5232312083244324,
+      "learning_rate": 0.00019876977152899825,
+      "loss": 0.9024254083633423,
+      "step": 1132
+    },
+    {
+      "epoch": 0.47848101265822784,
+      "grad_norm": 0.47671905159950256,
+      "learning_rate": 0.0001991212653778559,
+      "loss": 0.8996873497962952,
+      "step": 1134
+    },
+    {
+      "epoch": 0.47932489451476795,
+      "grad_norm": 0.42289480566978455,
+      "learning_rate": 0.00019947275922671353,
+      "loss": 0.8336917757987976,
+      "step": 1136
+    },
+    {
+      "epoch": 0.480168776371308,
+      "grad_norm": 0.4700844883918762,
+      "learning_rate": 0.0001998242530755712,
+      "loss": 0.8943206071853638,
+      "step": 1138
+    },
+    {
+      "epoch": 0.4810126582278481,
+      "grad_norm": 0.5341399312019348,
+      "learning_rate": 0.00019999999711649004,
+      "loss": 0.9340365529060364,
+      "step": 1140
+    },
+    {
+      "epoch": 0.48185654008438816,
+      "grad_norm": 0.46169522404670715,
+      "learning_rate": 0.00019999997404841123,
+      "loss": 0.8241778016090393,
+      "step": 1142
+    },
+    {
+      "epoch": 0.48270042194092827,
+      "grad_norm": 0.5475223064422607,
+      "learning_rate": 0.00019999992791225896,
+      "loss": 1.0096158981323242,
+      "step": 1144
+    },
+    {
+      "epoch": 0.4835443037974684,
+      "grad_norm": 0.524641215801239,
+      "learning_rate": 0.00019999985870804385,
+      "loss": 0.9650378227233887,
+      "step": 1146
+    },
+    {
+      "epoch": 0.48438818565400843,
+      "grad_norm": 0.4326174855232239,
+      "learning_rate": 0.00019999976643578186,
+      "loss": 0.9003009796142578,
+      "step": 1148
+    },
+    {
+      "epoch": 0.48523206751054854,
+      "grad_norm": 0.49034059047698975,
+      "learning_rate": 0.0001999996510954943,
+      "loss": 0.9793432950973511,
+      "step": 1150
+    },
+    {
+      "epoch": 0.4860759493670886,
+      "grad_norm": 0.4760092496871948,
+      "learning_rate": 0.00019999951268720776,
+      "loss": 0.8793007731437683,
+      "step": 1152
+    },
+    {
+      "epoch": 0.4869198312236287,
+      "grad_norm": 0.44606879353523254,
+      "learning_rate": 0.00019999935121095417,
+      "loss": 0.888630747795105,
+      "step": 1154
+    },
+    {
+      "epoch": 0.4877637130801688,
+      "grad_norm": 0.5030332207679749,
+      "learning_rate": 0.0001999991666667708,
+      "loss": 0.9660000801086426,
+      "step": 1156
+    },
+    {
+      "epoch": 0.48860759493670886,
+      "grad_norm": 0.4295555055141449,
+      "learning_rate": 0.00019999895905470014,
+      "loss": 0.824654757976532,
+      "step": 1158
+    },
+    {
+      "epoch": 0.48945147679324896,
+      "grad_norm": 0.392167866230011,
+      "learning_rate": 0.0001999987283747902,
+      "loss": 0.8159562349319458,
+      "step": 1160
+    },
+    {
+      "epoch": 0.490295358649789,
+      "grad_norm": 0.484611839056015,
+      "learning_rate": 0.00019999847462709412,
+      "loss": 0.9630686044692993,
+      "step": 1162
+    },
+    {
+      "epoch": 0.4911392405063291,
+      "grad_norm": 0.45891445875167847,
+      "learning_rate": 0.00019999819781167042,
+      "loss": 0.8396129608154297,
+      "step": 1164
+    },
+    {
+      "epoch": 0.4919831223628692,
+      "grad_norm": 0.5553452968597412,
+      "learning_rate": 0.00019999789792858304,
+      "loss": 0.8194513916969299,
+      "step": 1166
+    },
+    {
+      "epoch": 0.4928270042194093,
+      "grad_norm": 0.38998672366142273,
+      "learning_rate": 0.00019999757497790106,
+      "loss": 0.9268721342086792,
+      "step": 1168
+    },
+    {
+      "epoch": 0.4936708860759494,
+      "grad_norm": 0.45445355772972107,
+      "learning_rate": 0.00019999722895969904,
+      "loss": 0.9204684495925903,
+      "step": 1170
+    },
+    {
+      "epoch": 0.49451476793248944,
+      "grad_norm": 0.484225332736969,
+      "learning_rate": 0.00019999685987405678,
+      "loss": 0.9270301461219788,
+      "step": 1172
+    },
+    {
+      "epoch": 0.49535864978902955,
+      "grad_norm": 0.45215415954589844,
+      "learning_rate": 0.00019999646772105942,
+      "loss": 0.8782645463943481,
+      "step": 1174
+    },
+    {
+      "epoch": 0.4962025316455696,
+      "grad_norm": 0.41578832268714905,
+      "learning_rate": 0.00019999605250079744,
+      "loss": 0.873112678527832,
+      "step": 1176
+    },
+    {
+      "epoch": 0.4970464135021097,
+      "grad_norm": 0.4632788300514221,
+      "learning_rate": 0.0001999956142133666,
+      "loss": 0.9593189358711243,
+      "step": 1178
+    },
+    {
+      "epoch": 0.4978902953586498,
+      "grad_norm": 0.5007622838020325,
+      "learning_rate": 0.000199995152858868,
+      "loss": 0.9215621948242188,
+      "step": 1180
+    },
+    {
+      "epoch": 0.49873417721518987,
+      "grad_norm": 0.5201935768127441,
+      "learning_rate": 0.0001999946684374081,
+      "loss": 0.8964219689369202,
+      "step": 1182
+    },
+    {
+      "epoch": 0.49957805907173,
+      "grad_norm": 0.5053967237472534,
+      "learning_rate": 0.0001999941609490986,
+      "loss": 0.8528663516044617,
+      "step": 1184
+    },
+    {
+      "epoch": 0.5004219409282701,
+      "grad_norm": 0.44397974014282227,
+      "learning_rate": 0.0001999936303940566,
+      "loss": 0.8214734792709351,
+      "step": 1186
+    },
+    {
+      "epoch": 0.5012658227848101,
+      "grad_norm": 0.4563101530075073,
+      "learning_rate": 0.0001999930767724045,
+      "loss": 0.8882166743278503,
+      "step": 1188
+    },
+    {
+      "epoch": 0.5021097046413502,
+      "grad_norm": 0.4666728675365448,
+      "learning_rate": 0.00019999250008426997,
+      "loss": 0.8882588148117065,
+      "step": 1190
+    },
+    {
+      "epoch": 0.5029535864978903,
+      "grad_norm": 0.4423629939556122,
+      "learning_rate": 0.00019999190032978607,
+      "loss": 0.8353691697120667,
+      "step": 1192
+    },
+    {
+      "epoch": 0.5037974683544304,
+      "grad_norm": 0.4514595866203308,
+      "learning_rate": 0.00019999127750909118,
+      "loss": 0.8721219301223755,
+      "step": 1194
+    },
+    {
+      "epoch": 0.5046413502109705,
+      "grad_norm": 0.48469996452331543,
+      "learning_rate": 0.00019999063162232886,
+      "loss": 0.902795135974884,
+      "step": 1196
+    },
+    {
+      "epoch": 0.5054852320675105,
+      "grad_norm": 0.4931983947753906,
+      "learning_rate": 0.00019998996266964823,
+      "loss": 0.8843175768852234,
+      "step": 1198
+    },
+    {
+      "epoch": 0.5063291139240507,
+      "grad_norm": 0.507408082485199,
+      "learning_rate": 0.00019998927065120357,
+      "loss": 0.9179208278656006,
+      "step": 1200
+    },
+    {
+      "epoch": 0.5063291139240507,
+      "eval_loss": 0.9224098324775696,
+      "eval_runtime": 669.7542,
+      "eval_samples_per_second": 3.146,
+      "eval_steps_per_second": 3.146,
+      "step": 1200
+    },
+    {
+      "epoch": 0.5071729957805907,
+      "grad_norm": 0.4024188220500946,
+      "learning_rate": 0.00019998855556715447,
+      "loss": 0.9078981876373291,
+      "step": 1202
+    },
+    {
+      "epoch": 0.5080168776371308,
+      "grad_norm": 0.6558667421340942,
+      "learning_rate": 0.00019998781741766594,
+      "loss": 0.8089252710342407,
+      "step": 1204
+    },
+    {
+      "epoch": 0.5088607594936709,
+      "grad_norm": 0.45128145813941956,
+      "learning_rate": 0.00019998705620290823,
+      "loss": 0.9316248297691345,
+      "step": 1206
+    },
+    {
+      "epoch": 0.509704641350211,
+      "grad_norm": 0.4733511507511139,
+      "learning_rate": 0.00019998627192305694,
+      "loss": 0.9047867655754089,
+      "step": 1208
+    },
+    {
+      "epoch": 0.510548523206751,
+      "grad_norm": 0.4026021659374237,
+      "learning_rate": 0.00019998546457829298,
+      "loss": 0.8507166504859924,
+      "step": 1210
+    },
+    {
+      "epoch": 0.5113924050632911,
+      "grad_norm": 0.43416184186935425,
+      "learning_rate": 0.00019998463416880262,
+      "loss": 0.8490090370178223,
+      "step": 1212
+    },
+    {
+      "epoch": 0.5122362869198313,
+      "grad_norm": 0.3678364157676697,
+      "learning_rate": 0.0001999837806947774,
+      "loss": 0.7730492949485779,
+      "step": 1214
+    },
+    {
+      "epoch": 0.5130801687763713,
+      "grad_norm": 0.49042677879333496,
+      "learning_rate": 0.00019998290415641415,
+      "loss": 0.9625269174575806,
+      "step": 1216
+    },
+    {
+      "epoch": 0.5139240506329114,
+      "grad_norm": 0.4576701819896698,
+      "learning_rate": 0.00019998200455391516,
+      "loss": 0.8961732983589172,
+      "step": 1218
+    },
+    {
+      "epoch": 0.5147679324894515,
+      "grad_norm": 0.4465518593788147,
+      "learning_rate": 0.00019998108188748793,
+      "loss": 0.8508996367454529,
+      "step": 1220
+    },
+    {
+      "epoch": 0.5156118143459916,
+      "grad_norm": 0.4267960488796234,
+      "learning_rate": 0.00019998013615734524,
+      "loss": 0.8664930462837219,
+      "step": 1222
+    },
+    {
+      "epoch": 0.5164556962025316,
+      "grad_norm": 0.39936602115631104,
+      "learning_rate": 0.0001999791673637053,
+      "loss": 0.8841317892074585,
+      "step": 1224
+    },
+    {
+      "epoch": 0.5172995780590718,
+      "grad_norm": 0.4442414343357086,
+      "learning_rate": 0.0001999781755067916,
+      "loss": 0.8601276874542236,
+      "step": 1226
+    },
+    {
+      "epoch": 0.5181434599156118,
+      "grad_norm": 0.4397028088569641,
+      "learning_rate": 0.00019997716058683292,
+      "loss": 0.8377046585083008,
+      "step": 1228
+    },
+    {
+      "epoch": 0.5189873417721519,
+      "grad_norm": 0.5327648520469666,
+      "learning_rate": 0.0001999761226040634,
+      "loss": 0.9853615164756775,
+      "step": 1230
+    },
+    {
+      "epoch": 0.5198312236286919,
+      "grad_norm": 0.40931421518325806,
+      "learning_rate": 0.00019997506155872244,
+      "loss": 0.9235715866088867,
+      "step": 1232
+    },
+    {
+      "epoch": 0.5206751054852321,
+      "grad_norm": 0.4599161148071289,
+      "learning_rate": 0.00019997397745105487,
+      "loss": 0.868396520614624,
+      "step": 1234
+    },
+    {
+      "epoch": 0.5215189873417722,
+      "grad_norm": 0.4813833236694336,
+      "learning_rate": 0.0001999728702813107,
+      "loss": 0.9946733117103577,
+      "step": 1236
+    },
+    {
+      "epoch": 0.5223628691983122,
+      "grad_norm": 0.43518269062042236,
+      "learning_rate": 0.00019997174004974543,
+      "loss": 0.8608635067939758,
+      "step": 1238
+    },
+    {
+      "epoch": 0.5232067510548524,
+      "grad_norm": 0.45646214485168457,
+      "learning_rate": 0.0001999705867566197,
+      "loss": 0.819354772567749,
+      "step": 1240
+    },
+    {
+      "epoch": 0.5240506329113924,
+      "grad_norm": 0.46834197640419006,
+      "learning_rate": 0.00019996941040219954,
+      "loss": 0.8934658765792847,
+      "step": 1242
+    },
+    {
+      "epoch": 0.5248945147679325,
+      "grad_norm": 0.46065032482147217,
+      "learning_rate": 0.0001999682109867564,
+      "loss": 0.8624778985977173,
+      "step": 1244
+    },
+    {
+      "epoch": 0.5257383966244725,
+      "grad_norm": 0.4506741166114807,
+      "learning_rate": 0.00019996698851056688,
+      "loss": 0.8540882468223572,
+      "step": 1246
+    },
+    {
+      "epoch": 0.5265822784810127,
+      "grad_norm": 0.4550519287586212,
+      "learning_rate": 0.00019996574297391302,
+      "loss": 0.9003006219863892,
+      "step": 1248
+    },
+    {
+      "epoch": 0.5274261603375527,
+      "grad_norm": 0.4097813367843628,
+      "learning_rate": 0.00019996447437708214,
+      "loss": 0.8490248918533325,
+      "step": 1250
+    },
+    {
+      "epoch": 0.5282700421940928,
+      "grad_norm": 0.4375711679458618,
+      "learning_rate": 0.00019996318272036688,
+      "loss": 0.9204569458961487,
+      "step": 1252
+    },
+    {
+      "epoch": 0.529113924050633,
+      "grad_norm": 0.4335198998451233,
+      "learning_rate": 0.00019996186800406518,
+      "loss": 0.8831789493560791,
+      "step": 1254
+    },
+    {
+      "epoch": 0.529957805907173,
+      "grad_norm": 0.4840933084487915,
+      "learning_rate": 0.00019996053022848035,
+      "loss": 0.8578064441680908,
+      "step": 1256
+    },
+    {
+      "epoch": 0.5308016877637131,
+      "grad_norm": 0.48911216855049133,
+      "learning_rate": 0.00019995916939392097,
+      "loss": 0.9235416650772095,
+      "step": 1258
+    },
+    {
+      "epoch": 0.5316455696202531,
+      "grad_norm": 0.41106730699539185,
+      "learning_rate": 0.00019995778550070098,
+      "loss": 0.7897764444351196,
+      "step": 1260
+    },
+    {
+      "epoch": 0.5324894514767933,
+      "grad_norm": 0.42233356833457947,
+      "learning_rate": 0.00019995637854913957,
+      "loss": 0.9392989873886108,
+      "step": 1262
+    },
+    {
+      "epoch": 0.5333333333333333,
+      "grad_norm": 0.39145180583000183,
+      "learning_rate": 0.0001999549485395613,
+      "loss": 0.9130091667175293,
+      "step": 1264
+    },
+    {
+      "epoch": 0.5341772151898734,
+      "grad_norm": 0.42780739068984985,
+      "learning_rate": 0.00019995349547229614,
+      "loss": 0.9690561890602112,
+      "step": 1266
+    },
+    {
+      "epoch": 0.5350210970464135,
+      "grad_norm": 0.4823327362537384,
+      "learning_rate": 0.0001999520193476792,
+      "loss": 0.9990131258964539,
+      "step": 1268
+    },
+    {
+      "epoch": 0.5358649789029536,
+      "grad_norm": 0.4669715166091919,
+      "learning_rate": 0.00019995052016605097,
+      "loss": 0.8343052864074707,
+      "step": 1270
+    },
+    {
+      "epoch": 0.5367088607594936,
+      "grad_norm": 0.4024743437767029,
+      "learning_rate": 0.0001999489979277574,
+      "loss": 0.9605970978736877,
+      "step": 1272
+    },
+    {
+      "epoch": 0.5375527426160338,
+      "grad_norm": 0.42063432931900024,
+      "learning_rate": 0.0001999474526331495,
+      "loss": 0.8164438009262085,
+      "step": 1274
+    },
+    {
+      "epoch": 0.5383966244725739,
+      "grad_norm": 0.47953495383262634,
+      "learning_rate": 0.00019994588428258383,
+      "loss": 0.879767656326294,
+      "step": 1276
+    },
+    {
+      "epoch": 0.5392405063291139,
+      "grad_norm": 0.39668557047843933,
+      "learning_rate": 0.00019994429287642214,
+      "loss": 0.7658726572990417,
+      "step": 1278
+    },
+    {
+      "epoch": 0.540084388185654,
+      "grad_norm": 0.5136121511459351,
+      "learning_rate": 0.00019994267841503155,
+      "loss": 0.9477384686470032,
+      "step": 1280
+    },
+    {
+      "epoch": 0.5409282700421941,
+      "grad_norm": 0.4039286673069,
+      "learning_rate": 0.00019994104089878452,
+      "loss": 0.7883896827697754,
+      "step": 1282
+    },
+    {
+      "epoch": 0.5417721518987342,
+      "grad_norm": 0.4425487816333771,
+      "learning_rate": 0.0001999393803280587,
+      "loss": 0.8526129126548767,
+      "step": 1284
+    },
+    {
+      "epoch": 0.5426160337552742,
+      "grad_norm": 0.4544559121131897,
+      "learning_rate": 0.00019993769670323725,
+      "loss": 0.8694683909416199,
+      "step": 1286
+    },
+    {
+      "epoch": 0.5434599156118144,
+      "grad_norm": 0.42872917652130127,
+      "learning_rate": 0.0001999359900247085,
+      "loss": 0.8211527466773987,
+      "step": 1288
+    },
+    {
+      "epoch": 0.5443037974683544,
+      "grad_norm": 0.41549673676490784,
+      "learning_rate": 0.00019993426029286617,
+      "loss": 0.7967932224273682,
+      "step": 1290
+    },
+    {
+      "epoch": 0.5451476793248945,
+      "grad_norm": 0.42888355255126953,
+      "learning_rate": 0.00019993250750810926,
+      "loss": 0.8291563987731934,
+      "step": 1292
+    },
+    {
+      "epoch": 0.5459915611814345,
+      "grad_norm": 0.44985485076904297,
+      "learning_rate": 0.0001999307316708421,
+      "loss": 0.8787803649902344,
+      "step": 1294
+    },
+    {
+      "epoch": 0.5468354430379747,
+      "grad_norm": 0.43326008319854736,
+      "learning_rate": 0.00019992893278147436,
+      "loss": 0.8375519514083862,
+      "step": 1296
+    },
+    {
+      "epoch": 0.5476793248945148,
+      "grad_norm": 0.406556099653244,
+      "learning_rate": 0.00019992711084042096,
+      "loss": 0.8538051247596741,
+      "step": 1298
+    },
+    {
+      "epoch": 0.5485232067510548,
+      "grad_norm": 0.46090877056121826,
+      "learning_rate": 0.00019992526584810226,
+      "loss": 0.9722132086753845,
+      "step": 1300
+    },
+    {
+      "epoch": 0.5485232067510548,
+      "eval_loss": 0.9068717360496521,
+      "eval_runtime": 680.7718,
+      "eval_samples_per_second": 3.095,
+      "eval_steps_per_second": 3.095,
+      "step": 1300
+    },
+    {
+      "epoch": 0.549367088607595,
+      "grad_norm": 0.44932207465171814,
+      "learning_rate": 0.00019992339780494382,
+      "loss": 0.8474152684211731,
+      "step": 1302
+    },
+    {
+      "epoch": 0.550210970464135,
+      "grad_norm": 0.5185276865959167,
+      "learning_rate": 0.0001999215067113766,
+      "loss": 0.8440775871276855,
+      "step": 1304
+    },
+    {
+      "epoch": 0.5510548523206751,
+      "grad_norm": 0.43811365962028503,
+      "learning_rate": 0.0001999195925678368,
+      "loss": 0.8848010897636414,
+      "step": 1306
+    },
+    {
+      "epoch": 0.5518987341772152,
+      "grad_norm": 0.4861660301685333,
+      "learning_rate": 0.00019991765537476596,
+      "loss": 0.8265100121498108,
+      "step": 1308
+    },
+    {
+      "epoch": 0.5527426160337553,
+      "grad_norm": 0.5068721771240234,
+      "learning_rate": 0.000199915695132611,
+      "loss": 0.9327963590621948,
+      "step": 1310
+    },
+    {
+      "epoch": 0.5535864978902953,
+      "grad_norm": 0.4902805685997009,
+      "learning_rate": 0.0001999137118418241,
+      "loss": 0.9315155744552612,
+      "step": 1312
+    },
+    {
+      "epoch": 0.5544303797468354,
+      "grad_norm": 0.4530964493751526,
+      "learning_rate": 0.00019991170550286272,
+      "loss": 0.9344852566719055,
+      "step": 1314
+    },
+    {
+      "epoch": 0.5552742616033756,
+      "grad_norm": 0.4106673300266266,
+      "learning_rate": 0.00019990967611618974,
+      "loss": 0.8326173424720764,
+      "step": 1316
+    },
+    {
+      "epoch": 0.5561181434599156,
+      "grad_norm": 0.4439711570739746,
+      "learning_rate": 0.0001999076236822733,
+      "loss": 0.8934909105300903,
+      "step": 1318
+    },
+    {
+      "epoch": 0.5569620253164557,
+      "grad_norm": 0.40245553851127625,
+      "learning_rate": 0.0001999055482015868,
+      "loss": 0.6691107153892517,
+      "step": 1320
+    },
+    {
+      "epoch": 0.5578059071729958,
+      "grad_norm": 0.43904435634613037,
+      "learning_rate": 0.0001999034496746091,
+      "loss": 0.8560307621955872,
+      "step": 1322
+    },
+    {
+      "epoch": 0.5586497890295359,
+      "grad_norm": 0.5131705403327942,
+      "learning_rate": 0.00019990132810182422,
+      "loss": 0.9024442434310913,
+      "step": 1324
+    },
+    {
+      "epoch": 0.5594936708860759,
+      "grad_norm": 0.4539531171321869,
+      "learning_rate": 0.00019989918348372156,
+      "loss": 0.8675428628921509,
+      "step": 1326
+    },
+    {
+      "epoch": 0.560337552742616,
+      "grad_norm": 0.44469985365867615,
+      "learning_rate": 0.00019989701582079591,
+      "loss": 0.8853429555892944,
+      "step": 1328
+    },
+    {
+      "epoch": 0.5611814345991561,
+      "grad_norm": 0.4485546946525574,
+      "learning_rate": 0.00019989482511354725,
+      "loss": 0.9675378799438477,
+      "step": 1330
+    },
+    {
+      "epoch": 0.5620253164556962,
+      "grad_norm": 0.43316468596458435,
+      "learning_rate": 0.00019989261136248097,
+      "loss": 0.8256655335426331,
+      "step": 1332
+    },
+    {
+      "epoch": 0.5628691983122363,
+      "grad_norm": 0.4315780699253082,
+      "learning_rate": 0.00019989037456810772,
+      "loss": 0.7727690935134888,
+      "step": 1334
+    },
+    {
+      "epoch": 0.5637130801687764,
+      "grad_norm": 0.45147502422332764,
+      "learning_rate": 0.00019988811473094348,
+      "loss": 0.9240618348121643,
+      "step": 1336
+    },
+    {
+      "epoch": 0.5645569620253165,
+      "grad_norm": 0.4606908857822418,
+      "learning_rate": 0.00019988583185150957,
+      "loss": 0.8005949258804321,
+      "step": 1338
+    },
+    {
+      "epoch": 0.5654008438818565,
+      "grad_norm": 0.48932451009750366,
+      "learning_rate": 0.00019988352593033255,
+      "loss": 0.9042718410491943,
+      "step": 1340
+    },
+    {
+      "epoch": 0.5662447257383966,
+      "grad_norm": 0.46022218465805054,
+      "learning_rate": 0.00019988119696794443,
+      "loss": 0.8724613785743713,
+      "step": 1342
+    },
+    {
+      "epoch": 0.5670886075949367,
+      "grad_norm": 0.4414621889591217,
+      "learning_rate": 0.00019987884496488242,
+      "loss": 0.8653066158294678,
+      "step": 1344
+    },
+    {
+      "epoch": 0.5679324894514768,
+      "grad_norm": 0.4490342140197754,
+      "learning_rate": 0.0001998764699216891,
+      "loss": 0.8843849301338196,
+      "step": 1346
+    },
+    {
+      "epoch": 0.5687763713080168,
+      "grad_norm": 0.47690069675445557,
+      "learning_rate": 0.00019987407183891232,
+      "loss": 0.8385393023490906,
+      "step": 1348
+    },
+    {
+      "epoch": 0.569620253164557,
+      "grad_norm": 0.41409361362457275,
+      "learning_rate": 0.00019987165071710527,
+      "loss": 0.77492755651474,
+      "step": 1350
+    },
+    {
+      "epoch": 0.570464135021097,
+      "grad_norm": 0.4151647984981537,
+      "learning_rate": 0.0001998692065568265,
+      "loss": 0.8368680477142334,
+      "step": 1352
+    },
+    {
+      "epoch": 0.5713080168776371,
+      "grad_norm": 0.4105067849159241,
+      "learning_rate": 0.0001998667393586398,
+      "loss": 0.833220362663269,
+      "step": 1354
+    },
+    {
+      "epoch": 0.5721518987341773,
+      "grad_norm": 0.44701820611953735,
+      "learning_rate": 0.0001998642491231143,
+      "loss": 0.8041552305221558,
+      "step": 1356
+    },
+    {
+      "epoch": 0.5729957805907173,
+      "grad_norm": 0.4621582329273224,
+      "learning_rate": 0.00019986173585082444,
+      "loss": 0.9052709937095642,
+      "step": 1358
+    },
+    {
+      "epoch": 0.5738396624472574,
+      "grad_norm": 0.4934713840484619,
+      "learning_rate": 0.00019985919954235005,
+      "loss": 0.8036054372787476,
+      "step": 1360
+    },
+    {
+      "epoch": 0.5746835443037974,
+      "grad_norm": 0.4557384252548218,
+      "learning_rate": 0.0001998566401982761,
+      "loss": 0.8340095281600952,
+      "step": 1362
+    },
+    {
+      "epoch": 0.5755274261603376,
+      "grad_norm": 0.5048403739929199,
+      "learning_rate": 0.00019985405781919305,
+      "loss": 0.8703633546829224,
+      "step": 1364
+    },
+    {
+      "epoch": 0.5763713080168776,
+      "grad_norm": 0.4707714319229126,
+      "learning_rate": 0.0001998514524056966,
+      "loss": 0.9115830659866333,
+      "step": 1366
+    },
+    {
+      "epoch": 0.5772151898734177,
+      "grad_norm": 0.4502186179161072,
+      "learning_rate": 0.00019984882395838777,
+      "loss": 0.8603149056434631,
+      "step": 1368
+    },
+    {
+      "epoch": 0.5780590717299579,
+      "grad_norm": 0.4229136109352112,
+      "learning_rate": 0.00019984617247787288,
+      "loss": 0.7482197284698486,
+      "step": 1370
+    },
+    {
+      "epoch": 0.5789029535864979,
+      "grad_norm": 0.49208030104637146,
+      "learning_rate": 0.0001998434979647636,
+      "loss": 0.9071239233016968,
+      "step": 1372
+    },
+    {
+      "epoch": 0.579746835443038,
+      "grad_norm": 0.47393161058425903,
+      "learning_rate": 0.00019984080041967687,
+      "loss": 0.9370644092559814,
+      "step": 1374
+    },
+    {
+      "epoch": 0.580590717299578,
+      "grad_norm": 0.4483324885368347,
+      "learning_rate": 0.00019983807984323492,
+      "loss": 0.8536180257797241,
+      "step": 1376
+    },
+    {
+      "epoch": 0.5814345991561182,
+      "grad_norm": 0.4163796007633209,
+      "learning_rate": 0.00019983533623606543,
+      "loss": 0.7360405325889587,
+      "step": 1378
+    },
+    {
+      "epoch": 0.5822784810126582,
+      "grad_norm": 0.44147446751594543,
+      "learning_rate": 0.00019983256959880122,
+      "loss": 0.8162824511528015,
+      "step": 1380
+    },
+    {
+      "epoch": 0.5831223628691983,
+      "grad_norm": 0.44443491101264954,
+      "learning_rate": 0.0001998297799320805,
+      "loss": 0.9169327616691589,
+      "step": 1382
+    },
+    {
+      "epoch": 0.5839662447257384,
+      "grad_norm": 0.4641949236392975,
+      "learning_rate": 0.00019982696723654686,
+      "loss": 0.8875693678855896,
+      "step": 1384
+    },
+    {
+      "epoch": 0.5848101265822785,
+      "grad_norm": 0.4327974319458008,
+      "learning_rate": 0.00019982413151284906,
+      "loss": 0.7747344970703125,
+      "step": 1386
+    },
+    {
+      "epoch": 0.5856540084388185,
+      "grad_norm": 0.44200772047042847,
+      "learning_rate": 0.0001998212727616413,
+      "loss": 0.8490481972694397,
+      "step": 1388
+    },
+    {
+      "epoch": 0.5864978902953587,
+      "grad_norm": 0.39084959030151367,
+      "learning_rate": 0.000199818390983583,
+      "loss": 0.8014808297157288,
+      "step": 1390
+    },
+    {
+      "epoch": 0.5873417721518988,
+      "grad_norm": 0.3691277801990509,
+      "learning_rate": 0.00019981548617933897,
+      "loss": 0.8650928139686584,
+      "step": 1392
+    },
+    {
+      "epoch": 0.5881856540084388,
+      "grad_norm": 0.480002760887146,
+      "learning_rate": 0.00019981255834957925,
+      "loss": 0.7713267803192139,
+      "step": 1394
+    },
+    {
+      "epoch": 0.5890295358649789,
+      "grad_norm": 0.4397226572036743,
+      "learning_rate": 0.00019980960749497927,
+      "loss": 0.9053656458854675,
+      "step": 1396
+    },
+    {
+      "epoch": 0.589873417721519,
+      "grad_norm": 0.4263727366924286,
+      "learning_rate": 0.00019980663361621973,
+      "loss": 0.8737669587135315,
+      "step": 1398
+    },
+    {
+      "epoch": 0.5907172995780591,
+      "grad_norm": 0.4468817710876465,
+      "learning_rate": 0.0001998036367139866,
+      "loss": 0.9396650791168213,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5907172995780591,
+      "eval_loss": 0.8971880674362183,
+      "eval_runtime": 692.8046,
+      "eval_samples_per_second": 3.041,
+      "eval_steps_per_second": 3.041,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5915611814345991,
+      "grad_norm": 0.4128144383430481,
+      "learning_rate": 0.00019980061678897127,
+      "loss": 0.926134467124939,
+      "step": 1402
+    },
+    {
+      "epoch": 0.5924050632911393,
+      "grad_norm": 0.45524629950523376,
+      "learning_rate": 0.00019979757384187035,
+      "loss": 0.826360285282135,
+      "step": 1404
+    },
+    {
+      "epoch": 0.5932489451476793,
+      "grad_norm": 0.42521336674690247,
+      "learning_rate": 0.0001997945078733858,
+      "loss": 0.8147702813148499,
+      "step": 1406
+    },
+    {
+      "epoch": 0.5940928270042194,
+      "grad_norm": 0.4172525107860565,
+      "learning_rate": 0.00019979141888422489,
+      "loss": 0.7711596488952637,
+      "step": 1408
+    },
+    {
+      "epoch": 0.5949367088607594,
+      "grad_norm": 0.46193623542785645,
+      "learning_rate": 0.00019978830687510013,
+      "loss": 0.7875980138778687,
+      "step": 1410
+    },
+    {
+      "epoch": 0.5957805907172996,
+      "grad_norm": 0.4455825686454773,
+      "learning_rate": 0.00019978517184672946,
+      "loss": 0.8590195178985596,
+      "step": 1412
+    },
+    {
+      "epoch": 0.5966244725738397,
+      "grad_norm": 0.4266505539417267,
+      "learning_rate": 0.00019978201379983608,
+      "loss": 0.9318227767944336,
+      "step": 1414
+    },
+    {
+      "epoch": 0.5974683544303797,
+      "grad_norm": 0.42627251148223877,
+      "learning_rate": 0.00019977883273514843,
+      "loss": 0.8644474744796753,
+      "step": 1416
+    },
+    {
+      "epoch": 0.5983122362869199,
+      "grad_norm": 0.393873393535614,
+      "learning_rate": 0.00019977562865340038,
+      "loss": 0.7760446071624756,
+      "step": 1418
+    },
+    {
+      "epoch": 0.5991561181434599,
+      "grad_norm": 0.44334208965301514,
+      "learning_rate": 0.000199772401555331,
+      "loss": 0.8071584701538086,
+      "step": 1420
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.39390063285827637,
+      "learning_rate": 0.0001997691514416848,
+      "loss": 0.9076889753341675,
+      "step": 1422
+    },
+    {
+      "epoch": 0.60084388185654,
+      "grad_norm": 0.46753421425819397,
+      "learning_rate": 0.00019976587831321144,
+      "loss": 0.8534318804740906,
+      "step": 1424
+    },
+    {
+      "epoch": 0.6016877637130802,
+      "grad_norm": 0.4840681254863739,
+      "learning_rate": 0.00019976258217066598,
+      "loss": 0.9049596190452576,
+      "step": 1426
+    },
+    {
+      "epoch": 0.6025316455696202,
+      "grad_norm": 0.4740557372570038,
+      "learning_rate": 0.00019975926301480878,
+      "loss": 0.8520918488502502,
+      "step": 1428
+    },
+    {
+      "epoch": 0.6033755274261603,
+      "grad_norm": 0.46498584747314453,
+      "learning_rate": 0.00019975592084640553,
+      "loss": 0.8221305012702942,
+      "step": 1430
+    },
+    {
+      "epoch": 0.6042194092827005,
+      "grad_norm": 0.40885692834854126,
+      "learning_rate": 0.0001997525556662272,
+      "loss": 0.8132198452949524,
+      "step": 1432
+    },
+    {
+      "epoch": 0.6050632911392405,
+      "grad_norm": 0.4375821053981781,
+      "learning_rate": 0.00019974916747505006,
+      "loss": 0.870534360408783,
+      "step": 1434
+    },
+    {
+      "epoch": 0.6059071729957806,
+      "grad_norm": 0.4780935049057007,
+      "learning_rate": 0.00019974575627365569,
+      "loss": 0.8825759291648865,
+      "step": 1436
+    },
+    {
+      "epoch": 0.6067510548523207,
+      "grad_norm": 0.4623970687389374,
+      "learning_rate": 0.00019974232206283098,
+      "loss": 0.8187641501426697,
+      "step": 1438
+    },
+    {
+      "epoch": 0.6075949367088608,
+      "grad_norm": 0.42172595858573914,
+      "learning_rate": 0.0001997388648433682,
+      "loss": 0.8569821119308472,
+      "step": 1440
+    },
+    {
+      "epoch": 0.6084388185654008,
+      "grad_norm": 0.49386659264564514,
+      "learning_rate": 0.0001997353846160648,
+      "loss": 0.9173614382743835,
+      "step": 1442
+    },
+    {
+      "epoch": 0.6092827004219409,
+      "grad_norm": 0.47122615575790405,
+      "learning_rate": 0.00019973188138172363,
+      "loss": 0.9444975256919861,
+      "step": 1444
+    },
+    {
+      "epoch": 0.610126582278481,
+      "grad_norm": 0.4831530451774597,
+      "learning_rate": 0.0001997283551411528,
+      "loss": 0.9537245631217957,
+      "step": 1446
+    },
+    {
+      "epoch": 0.6109704641350211,
+      "grad_norm": 0.44460317492485046,
+      "learning_rate": 0.0001997248058951658,
+      "loss": 0.8481367230415344,
+      "step": 1448
+    },
+    {
+      "epoch": 0.6118143459915611,
+      "grad_norm": 0.4315263628959656,
+      "learning_rate": 0.00019972123364458128,
+      "loss": 0.8572371006011963,
+      "step": 1450
+    },
+    {
+      "epoch": 0.6126582278481013,
+      "grad_norm": 0.38559582829475403,
+      "learning_rate": 0.00019971763839022336,
+      "loss": 0.762590765953064,
+      "step": 1452
+    },
+    {
+      "epoch": 0.6135021097046414,
+      "grad_norm": 0.3820290267467499,
+      "learning_rate": 0.00019971402013292138,
+      "loss": 0.7879500389099121,
+      "step": 1454
+    },
+    {
+      "epoch": 0.6143459915611814,
+      "grad_norm": 0.42509570717811584,
+      "learning_rate": 0.00019971037887351,
+      "loss": 0.8696863055229187,
+      "step": 1456
+    },
+    {
+      "epoch": 0.6151898734177215,
+      "grad_norm": 0.46750548481941223,
+      "learning_rate": 0.00019970671461282916,
+      "loss": 0.8425862789154053,
+      "step": 1458
+    },
+    {
+      "epoch": 0.6160337552742616,
+      "grad_norm": 0.3937121033668518,
+      "learning_rate": 0.0001997030273517242,
+      "loss": 0.8028931021690369,
+      "step": 1460
+    },
+    {
+      "epoch": 0.6168776371308017,
+      "grad_norm": 0.4361220896244049,
+      "learning_rate": 0.00019969931709104565,
+      "loss": 0.8662509918212891,
+      "step": 1462
+    },
+    {
+      "epoch": 0.6177215189873417,
+      "grad_norm": 0.4737963378429413,
+      "learning_rate": 0.00019969558383164943,
+      "loss": 0.8962596654891968,
+      "step": 1464
+    },
+    {
+      "epoch": 0.6185654008438819,
+      "grad_norm": 0.3952867388725281,
+      "learning_rate": 0.00019969182757439668,
+      "loss": 0.8630867600440979,
+      "step": 1466
+    },
+    {
+      "epoch": 0.619409282700422,
+      "grad_norm": 0.41688743233680725,
+      "learning_rate": 0.00019968804832015393,
+      "loss": 0.8759240508079529,
+      "step": 1468
+    },
+    {
+      "epoch": 0.620253164556962,
+      "grad_norm": 0.4599224627017975,
+      "learning_rate": 0.00019968424606979298,
+      "loss": 0.8459385633468628,
+      "step": 1470
+    },
+    {
+      "epoch": 0.6210970464135022,
+      "grad_norm": 0.5476765632629395,
+      "learning_rate": 0.00019968042082419094,
+      "loss": 0.8844659328460693,
+      "step": 1472
+    },
+    {
+      "epoch": 0.6219409282700422,
+      "grad_norm": 0.46202269196510315,
+      "learning_rate": 0.00019967657258423022,
+      "loss": 0.860946536064148,
+      "step": 1474
+    },
+    {
+      "epoch": 0.6227848101265823,
+      "grad_norm": 0.39446666836738586,
+      "learning_rate": 0.00019967270135079853,
+      "loss": 0.7587860822677612,
+      "step": 1476
+    },
+    {
+      "epoch": 0.6236286919831223,
+      "grad_norm": 0.4500375986099243,
+      "learning_rate": 0.0001996688071247889,
+      "loss": 0.9143000245094299,
+      "step": 1478
+    },
+    {
+      "epoch": 0.6244725738396625,
+      "grad_norm": 0.4203545153141022,
+      "learning_rate": 0.0001996648899070996,
+      "loss": 0.7865519523620605,
+      "step": 1480
+    },
+    {
+      "epoch": 0.6253164556962025,
+      "grad_norm": 0.45623889565467834,
+      "learning_rate": 0.00019966094969863432,
+      "loss": 0.9192912578582764,
+      "step": 1482
+    },
+    {
+      "epoch": 0.6261603375527426,
+      "grad_norm": 0.4139637053012848,
+      "learning_rate": 0.00019965698650030195,
+      "loss": 0.8459161520004272,
+      "step": 1484
+    },
+    {
+      "epoch": 0.6270042194092827,
+      "grad_norm": 0.4410824179649353,
+      "learning_rate": 0.00019965300031301678,
+      "loss": 0.8159077763557434,
+      "step": 1486
+    },
+    {
+      "epoch": 0.6278481012658228,
+      "grad_norm": 0.43821197748184204,
+      "learning_rate": 0.00019964899113769828,
+      "loss": 0.884467363357544,
+      "step": 1488
+    },
+    {
+      "epoch": 0.6286919831223629,
+      "grad_norm": 0.4692091643810272,
+      "learning_rate": 0.00019964495897527133,
+      "loss": 0.8693601489067078,
+      "step": 1490
+    },
+    {
+      "epoch": 0.6295358649789029,
+      "grad_norm": 0.40715447068214417,
+      "learning_rate": 0.00019964090382666608,
+      "loss": 0.8897743225097656,
+      "step": 1492
+    },
+    {
+      "epoch": 0.6303797468354431,
+      "grad_norm": 0.4123334586620331,
+      "learning_rate": 0.0001996368256928179,
+      "loss": 0.8274733424186707,
+      "step": 1494
+    },
+    {
+      "epoch": 0.6312236286919831,
+      "grad_norm": 0.44347211718559265,
+      "learning_rate": 0.00019963272457466767,
+      "loss": 0.8158749938011169,
+      "step": 1496
+    },
+    {
+      "epoch": 0.6320675105485232,
+      "grad_norm": 0.4700329601764679,
+      "learning_rate": 0.00019962860047316135,
+      "loss": 0.8439569473266602,
+      "step": 1498
+    },
+    {
+      "epoch": 0.6329113924050633,
+      "grad_norm": 0.46415844559669495,
+      "learning_rate": 0.00019962445338925027,
+      "loss": 0.8316822648048401,
+      "step": 1500
+    },
+    {
+      "epoch": 0.6329113924050633,
+      "eval_loss": 0.887488842010498,
+      "eval_runtime": 686.2804,
+      "eval_samples_per_second": 3.07,
+      "eval_steps_per_second": 3.07,
+      "step": 1500
+    },
+    {
+      "epoch": 0.6337552742616034,
+      "grad_norm": 0.44183167815208435,
+      "learning_rate": 0.0001996202833238911,
+      "loss": 0.9121994972229004,
+      "step": 1502
+    },
+    {
+      "epoch": 0.6345991561181434,
+      "grad_norm": 0.4279208779335022,
+      "learning_rate": 0.00019961609027804587,
+      "loss": 0.9160211086273193,
+      "step": 1504
+    },
+    {
+      "epoch": 0.6354430379746835,
+      "grad_norm": 0.3967169523239136,
+      "learning_rate": 0.00019961187425268176,
+      "loss": 0.861677348613739,
+      "step": 1506
+    },
+    {
+      "epoch": 0.6362869198312237,
+      "grad_norm": 0.4796451926231384,
+      "learning_rate": 0.0001996076352487713,
+      "loss": 0.8478423357009888,
+      "step": 1508
+    },
+    {
+      "epoch": 0.6371308016877637,
+      "grad_norm": 0.3861426115036011,
+      "learning_rate": 0.00019960337326729245,
+      "loss": 0.8739159107208252,
+      "step": 1510
+    },
+    {
+      "epoch": 0.6379746835443038,
+      "grad_norm": 0.4567820429801941,
+      "learning_rate": 0.00019959908830922824,
+      "loss": 0.7693920731544495,
+      "step": 1512
+    },
+    {
+      "epoch": 0.6388185654008439,
+      "grad_norm": 0.4457108676433563,
+      "learning_rate": 0.00019959478037556724,
+      "loss": 0.8482301831245422,
+      "step": 1514
+    },
+    {
+      "epoch": 0.639662447257384,
+      "grad_norm": 0.43918928503990173,
+      "learning_rate": 0.00019959044946730314,
+      "loss": 0.8626812696456909,
+      "step": 1516
+    },
+    {
+      "epoch": 0.640506329113924,
+      "grad_norm": 0.4095900058746338,
+      "learning_rate": 0.00019958609558543504,
+      "loss": 0.8342230916023254,
+      "step": 1518
+    },
+    {
+      "epoch": 0.6413502109704642,
+      "grad_norm": 0.4181270897388458,
+      "learning_rate": 0.00019958171873096724,
+      "loss": 0.8794118165969849,
+      "step": 1520
+    },
+    {
+      "epoch": 0.6421940928270042,
+      "grad_norm": 0.4750959277153015,
+      "learning_rate": 0.00019957731890490947,
+      "loss": 0.7787677049636841,
+      "step": 1522
+    },
+    {
+      "epoch": 0.6430379746835443,
+      "grad_norm": 0.5129296779632568,
+      "learning_rate": 0.00019957289610827663,
+      "loss": 0.9084368348121643,
+      "step": 1524
+    },
+    {
+      "epoch": 0.6438818565400843,
+      "grad_norm": 0.4829029142856598,
+      "learning_rate": 0.000199568450342089,
+      "loss": 0.9066952466964722,
+      "step": 1526
+    },
+    {
+      "epoch": 0.6447257383966245,
+      "grad_norm": 0.4225057363510132,
+      "learning_rate": 0.00019956398160737214,
+      "loss": 0.8127874732017517,
+      "step": 1528
+    },
+    {
+      "epoch": 0.6455696202531646,
+      "grad_norm": 0.4231826364994049,
+      "learning_rate": 0.00019955948990515684,
+      "loss": 0.8781921863555908,
+      "step": 1530
+    },
+    {
+      "epoch": 0.6464135021097046,
+      "grad_norm": 0.4436216950416565,
+      "learning_rate": 0.00019955497523647933,
+      "loss": 0.8136996626853943,
+      "step": 1532
+    },
+    {
+      "epoch": 0.6472573839662448,
+      "grad_norm": 0.4018244743347168,
+      "learning_rate": 0.00019955043760238098,
+      "loss": 0.8003877401351929,
+      "step": 1534
+    },
+    {
+      "epoch": 0.6481012658227848,
+      "grad_norm": 0.40024611353874207,
+      "learning_rate": 0.0001995458770039086,
+      "loss": 0.8996267914772034,
+      "step": 1536
+    },
+    {
+      "epoch": 0.6489451476793249,
+      "grad_norm": 0.42845702171325684,
+      "learning_rate": 0.00019954129344211424,
+      "loss": 0.8610522150993347,
+      "step": 1538
+    },
+    {
+      "epoch": 0.6497890295358649,
+      "grad_norm": 0.4258182644844055,
+      "learning_rate": 0.00019953668691805517,
+      "loss": 0.8991837501525879,
+      "step": 1540
+    },
+    {
+      "epoch": 0.6506329113924051,
+      "grad_norm": 0.4483257830142975,
+      "learning_rate": 0.0001995320574327941,
+      "loss": 0.8150189518928528,
+      "step": 1542
+    },
+    {
+      "epoch": 0.6514767932489451,
+      "grad_norm": 0.4320627748966217,
+      "learning_rate": 0.0001995274049873989,
+      "loss": 0.8572840690612793,
+      "step": 1544
+    },
+    {
+      "epoch": 0.6523206751054852,
+      "grad_norm": 0.39950594305992126,
+      "learning_rate": 0.00019952272958294288,
+      "loss": 0.7766129374504089,
+      "step": 1546
+    },
+    {
+      "epoch": 0.6531645569620254,
+      "grad_norm": 0.4130117893218994,
+      "learning_rate": 0.00019951803122050446,
+      "loss": 0.9032199382781982,
+      "step": 1548
+    },
+    {
+      "epoch": 0.6540084388185654,
+      "grad_norm": 0.4436919391155243,
+      "learning_rate": 0.00019951330990116754,
+      "loss": 0.8361048102378845,
+      "step": 1550
+    },
+    {
+      "epoch": 0.6548523206751055,
+      "grad_norm": 0.4229227900505066,
+      "learning_rate": 0.00019950856562602121,
+      "loss": 0.8661768436431885,
+      "step": 1552
+    },
+    {
+      "epoch": 0.6556962025316456,
+      "grad_norm": 0.4839727580547333,
+      "learning_rate": 0.00019950379839615988,
+      "loss": 0.9737826585769653,
+      "step": 1554
+    },
+    {
+      "epoch": 0.6565400843881857,
+      "grad_norm": 0.3859386742115021,
+      "learning_rate": 0.00019949900821268328,
+      "loss": 0.7542453408241272,
+      "step": 1556
+    },
+    {
+      "epoch": 0.6573839662447257,
+      "grad_norm": 0.39643988013267517,
+      "learning_rate": 0.0001994941950766964,
+      "loss": 0.9226290583610535,
+      "step": 1558
+    },
+    {
+      "epoch": 0.6582278481012658,
+      "grad_norm": 0.4261355400085449,
+      "learning_rate": 0.00019948935898930954,
+      "loss": 0.8375602960586548,
+      "step": 1560
+    },
+    {
+      "epoch": 0.6590717299578059,
+      "grad_norm": 0.445931077003479,
+      "learning_rate": 0.00019948449995163826,
+      "loss": 0.7847310304641724,
+      "step": 1562
+    },
+    {
+      "epoch": 0.659915611814346,
+      "grad_norm": 0.4680459797382355,
+      "learning_rate": 0.00019947961796480353,
+      "loss": 0.851150631904602,
+      "step": 1564
+    },
+    {
+      "epoch": 0.660759493670886,
+      "grad_norm": 0.6826074719429016,
+      "learning_rate": 0.00019947471302993143,
+      "loss": 0.8259562253952026,
+      "step": 1566
+    },
+    {
+      "epoch": 0.6616033755274262,
+      "grad_norm": 0.47369205951690674,
+      "learning_rate": 0.00019946978514815352,
+      "loss": 0.879021406173706,
+      "step": 1568
+    },
+    {
+      "epoch": 0.6624472573839663,
+      "grad_norm": 0.41111937165260315,
+      "learning_rate": 0.0001994648343206065,
+      "loss": 0.9296056032180786,
+      "step": 1570
+    },
+    {
+      "epoch": 0.6632911392405063,
+      "grad_norm": 0.4297783076763153,
+      "learning_rate": 0.00019945986054843248,
+      "loss": 0.873822808265686,
+      "step": 1572
+    },
+    {
+      "epoch": 0.6641350210970464,
+      "grad_norm": 0.48124316334724426,
+      "learning_rate": 0.00019945486383277878,
+      "loss": 0.937151312828064,
+      "step": 1574
+    },
+    {
+      "epoch": 0.6649789029535865,
+      "grad_norm": 0.4048190116882324,
+      "learning_rate": 0.00019944984417479805,
+      "loss": 0.8398929834365845,
+      "step": 1576
+    },
+    {
+      "epoch": 0.6658227848101266,
+      "grad_norm": 0.4023774266242981,
+      "learning_rate": 0.00019944480157564826,
+      "loss": 0.8269520998001099,
+      "step": 1578
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.3784184157848358,
+      "learning_rate": 0.00019943973603649262,
+      "loss": 0.8237608671188354,
+      "step": 1580
+    },
+    {
+      "epoch": 0.6675105485232068,
+      "grad_norm": 0.4448573887348175,
+      "learning_rate": 0.00019943464755849965,
+      "loss": 0.8469099998474121,
+      "step": 1582
+    },
+    {
+      "epoch": 0.6683544303797468,
+      "grad_norm": 0.4690774083137512,
+      "learning_rate": 0.00019942953614284317,
+      "loss": 0.8630822896957397,
+      "step": 1584
+    },
+    {
+      "epoch": 0.6691983122362869,
+      "grad_norm": 0.43670853972435,
+      "learning_rate": 0.00019942440179070227,
+      "loss": 0.8904082179069519,
+      "step": 1586
+    },
+    {
+      "epoch": 0.6700421940928271,
+      "grad_norm": 0.4236849844455719,
+      "learning_rate": 0.00019941924450326138,
+      "loss": 0.8355640172958374,
+      "step": 1588
+    },
+    {
+      "epoch": 0.6708860759493671,
+      "grad_norm": 0.48624539375305176,
+      "learning_rate": 0.00019941406428171013,
+      "loss": 0.8898435831069946,
+      "step": 1590
+    },
+    {
+      "epoch": 0.6717299578059072,
+      "grad_norm": 0.4453780949115753,
+      "learning_rate": 0.00019940886112724355,
+      "loss": 0.9000012278556824,
+      "step": 1592
+    },
+    {
+      "epoch": 0.6725738396624472,
+      "grad_norm": 0.4417199194431305,
+      "learning_rate": 0.00019940363504106193,
+      "loss": 0.8171058893203735,
+      "step": 1594
+    },
+    {
+      "epoch": 0.6734177215189874,
+      "grad_norm": 0.45055362582206726,
+      "learning_rate": 0.00019939838602437074,
+      "loss": 0.8703798055648804,
+      "step": 1596
+    },
+    {
+      "epoch": 0.6742616033755274,
+      "grad_norm": 0.42316532135009766,
+      "learning_rate": 0.00019939311407838088,
+      "loss": 0.8677281737327576,
+      "step": 1598
+    },
+    {
+      "epoch": 0.6751054852320675,
+      "grad_norm": 0.3935781717300415,
+      "learning_rate": 0.0001993878192043085,
+      "loss": 0.7314013242721558,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6751054852320675,
+      "eval_loss": 0.8769772052764893,
+      "eval_runtime": 677.9338,
+      "eval_samples_per_second": 3.108,
+      "eval_steps_per_second": 3.108,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6759493670886076,
+      "grad_norm": 0.4197012186050415,
+      "learning_rate": 0.000199382501403375,
+      "loss": 0.8075380325317383,
+      "step": 1602
+    },
+    {
+      "epoch": 0.6767932489451477,
+      "grad_norm": 0.45227327942848206,
+      "learning_rate": 0.00019937716067680713,
+      "loss": 0.7885794639587402,
+      "step": 1604
+    },
+    {
+      "epoch": 0.6776371308016877,
+      "grad_norm": 0.45660385489463806,
+      "learning_rate": 0.00019937179702583682,
+      "loss": 0.868261992931366,
+      "step": 1606
+    },
+    {
+      "epoch": 0.6784810126582278,
+      "grad_norm": 0.3955288231372833,
+      "learning_rate": 0.00019936641045170144,
+      "loss": 0.8306655287742615,
+      "step": 1608
+    },
+    {
+      "epoch": 0.679324894514768,
+      "grad_norm": 0.3651324510574341,
+      "learning_rate": 0.00019936100095564353,
+      "loss": 0.7505315542221069,
+      "step": 1610
+    },
+    {
+      "epoch": 0.680168776371308,
+      "grad_norm": 0.4240955412387848,
+      "learning_rate": 0.00019935556853891096,
+      "loss": 0.8792756795883179,
+      "step": 1612
+    },
+    {
+      "epoch": 0.6810126582278481,
+      "grad_norm": 0.4142507314682007,
+      "learning_rate": 0.00019935011320275687,
+      "loss": 0.7328272461891174,
+      "step": 1614
+    },
+    {
+      "epoch": 0.6818565400843882,
+      "grad_norm": 0.48339322209358215,
+      "learning_rate": 0.00019934463494843975,
+      "loss": 0.8414849042892456,
+      "step": 1616
+    },
+    {
+      "epoch": 0.6827004219409283,
+      "grad_norm": 0.4312443435192108,
+      "learning_rate": 0.00019933913377722328,
+      "loss": 0.8588716983795166,
+      "step": 1618
+    },
+    {
+      "epoch": 0.6835443037974683,
+      "grad_norm": 0.43849512934684753,
+      "learning_rate": 0.0001993336096903765,
+      "loss": 0.846511960029602,
+      "step": 1620
+    },
+    {
+      "epoch": 0.6843881856540084,
+      "grad_norm": 0.45767566561698914,
+      "learning_rate": 0.0001993280626891737,
+      "loss": 0.9352323412895203,
+      "step": 1622
+    },
+    {
+      "epoch": 0.6852320675105485,
+      "grad_norm": 0.438412606716156,
+      "learning_rate": 0.00019932249277489444,
+      "loss": 0.8884757161140442,
+      "step": 1624
+    },
+    {
+      "epoch": 0.6860759493670886,
+      "grad_norm": 0.45893922448158264,
+      "learning_rate": 0.0001993168999488237,
+      "loss": 0.7739649415016174,
+      "step": 1626
+    },
+    {
+      "epoch": 0.6869198312236287,
+      "grad_norm": 0.38631772994995117,
+      "learning_rate": 0.0001993112842122515,
+      "loss": 0.7948258519172668,
+      "step": 1628
+    },
+    {
+      "epoch": 0.6877637130801688,
+      "grad_norm": 0.3824027180671692,
+      "learning_rate": 0.00019930564556647334,
+      "loss": 0.750511884689331,
+      "step": 1630
+    },
+    {
+      "epoch": 0.6886075949367089,
+      "grad_norm": 0.3980776071548462,
+      "learning_rate": 0.00019929998401278996,
+      "loss": 0.9006738662719727,
+      "step": 1632
+    },
+    {
+      "epoch": 0.6894514767932489,
+      "grad_norm": 0.41419896483421326,
+      "learning_rate": 0.00019929429955250734,
+      "loss": 0.6996869444847107,
+      "step": 1634
+    },
+    {
+      "epoch": 0.6902953586497891,
+      "grad_norm": 0.43783676624298096,
+      "learning_rate": 0.00019928859218693682,
+      "loss": 0.8523393273353577,
+      "step": 1636
+    },
+    {
+      "epoch": 0.6911392405063291,
+      "grad_norm": 0.4378126561641693,
+      "learning_rate": 0.00019928286191739497,
+      "loss": 0.8885987401008606,
+      "step": 1638
+    },
+    {
+      "epoch": 0.6919831223628692,
+      "grad_norm": 0.4276133179664612,
+      "learning_rate": 0.00019927710874520363,
+      "loss": 0.8563809990882874,
+      "step": 1640
+    },
+    {
+      "epoch": 0.6928270042194092,
+      "grad_norm": 0.4600282311439514,
+      "learning_rate": 0.00019927133267168992,
+      "loss": 0.854114294052124,
+      "step": 1642
+    },
+    {
+      "epoch": 0.6936708860759494,
+      "grad_norm": 0.5058461427688599,
+      "learning_rate": 0.00019926553369818635,
+      "loss": 0.8557889461517334,
+      "step": 1644
+    },
+    {
+      "epoch": 0.6945147679324895,
+      "grad_norm": 0.4947412312030792,
+      "learning_rate": 0.00019925971182603057,
+      "loss": 0.8622767925262451,
+      "step": 1646
+    },
+    {
+      "epoch": 0.6953586497890295,
+      "grad_norm": 0.42555496096611023,
+      "learning_rate": 0.00019925386705656558,
+      "loss": 0.7836558818817139,
+      "step": 1648
+    },
+    {
+      "epoch": 0.6962025316455697,
+      "grad_norm": 0.47560831904411316,
+      "learning_rate": 0.00019924799939113967,
+      "loss": 0.7877846956253052,
+      "step": 1650
+    },
+    {
+      "epoch": 0.6970464135021097,
+      "grad_norm": 0.391702264547348,
+      "learning_rate": 0.0001992421088311064,
+      "loss": 0.775681734085083,
+      "step": 1652
+    },
+    {
+      "epoch": 0.6978902953586498,
+      "grad_norm": 0.463679701089859,
+      "learning_rate": 0.0001992361953778246,
+      "loss": 0.8809882402420044,
+      "step": 1654
+    },
+    {
+      "epoch": 0.6987341772151898,
+      "grad_norm": 0.44096484780311584,
+      "learning_rate": 0.00019923025903265836,
+      "loss": 0.8513299226760864,
+      "step": 1656
+    },
+    {
+      "epoch": 0.69957805907173,
+      "grad_norm": 0.47818854451179504,
+      "learning_rate": 0.0001992242997969772,
+      "loss": 0.8309667706489563,
+      "step": 1658
+    },
+    {
+      "epoch": 0.70042194092827,
+      "grad_norm": 0.4192790985107422,
+      "learning_rate": 0.00019921831767215565,
+      "loss": 0.8817558884620667,
+      "step": 1660
+    },
+    {
+      "epoch": 0.7012658227848101,
+      "grad_norm": 0.45285239815711975,
+      "learning_rate": 0.00019921231265957373,
+      "loss": 0.8315755724906921,
+      "step": 1662
+    },
+    {
+      "epoch": 0.7021097046413503,
+      "grad_norm": 0.47118625044822693,
+      "learning_rate": 0.00019920628476061673,
+      "loss": 0.8698058724403381,
+      "step": 1664
+    },
+    {
+      "epoch": 0.7029535864978903,
+      "grad_norm": 0.4769238829612732,
+      "learning_rate": 0.0001992002339766751,
+      "loss": 0.953620970249176,
+      "step": 1666
+    },
+    {
+      "epoch": 0.7037974683544304,
+      "grad_norm": 0.39397355914115906,
+      "learning_rate": 0.00019919416030914468,
+      "loss": 0.8536615371704102,
+      "step": 1668
+    },
+    {
+      "epoch": 0.7046413502109705,
+      "grad_norm": 0.3659648299217224,
+      "learning_rate": 0.00019918806375942655,
+      "loss": 0.8159828782081604,
+      "step": 1670
+    },
+    {
+      "epoch": 0.7054852320675106,
+      "grad_norm": 0.4381789565086365,
+      "learning_rate": 0.00019918194432892703,
+      "loss": 0.782463550567627,
+      "step": 1672
+    },
+    {
+      "epoch": 0.7063291139240506,
+      "grad_norm": 0.44745752215385437,
+      "learning_rate": 0.00019917580201905778,
+      "loss": 0.9314272999763489,
+      "step": 1674
+    },
+    {
+      "epoch": 0.7071729957805907,
+      "grad_norm": 0.4515164792537689,
+      "learning_rate": 0.00019916963683123574,
+      "loss": 0.860816240310669,
+      "step": 1676
+    },
+    {
+      "epoch": 0.7080168776371308,
+      "grad_norm": 0.42735886573791504,
+      "learning_rate": 0.00019916344876688303,
+      "loss": 0.8008456826210022,
+      "step": 1678
+    },
+    {
+      "epoch": 0.7088607594936709,
+      "grad_norm": 0.418275386095047,
+      "learning_rate": 0.00019915723782742718,
+      "loss": 0.7796595096588135,
+      "step": 1680
+    },
+    {
+      "epoch": 0.7097046413502109,
+      "grad_norm": 0.44347989559173584,
+      "learning_rate": 0.00019915100401430089,
+      "loss": 0.9006506204605103,
+      "step": 1682
+    },
+    {
+      "epoch": 0.7105485232067511,
+      "grad_norm": 0.3932702839374542,
+      "learning_rate": 0.00019914474732894225,
+      "loss": 0.7750146985054016,
+      "step": 1684
+    },
+    {
+      "epoch": 0.7113924050632912,
+      "grad_norm": 0.4810502231121063,
+      "learning_rate": 0.00019913846777279447,
+      "loss": 0.7508097290992737,
+      "step": 1686
+    },
+    {
+      "epoch": 0.7122362869198312,
+      "grad_norm": 0.4166778028011322,
+      "learning_rate": 0.00019913216534730622,
+      "loss": 0.7926796078681946,
+      "step": 1688
+    },
+    {
+      "epoch": 0.7130801687763713,
+      "grad_norm": 0.47900474071502686,
+      "learning_rate": 0.00019912584005393123,
+      "loss": 0.8363928198814392,
+      "step": 1690
+    },
+    {
+      "epoch": 0.7139240506329114,
+      "grad_norm": 0.435143381357193,
+      "learning_rate": 0.00019911949189412872,
+      "loss": 0.8649623990058899,
+      "step": 1692
+    },
+    {
+      "epoch": 0.7147679324894515,
+      "grad_norm": 0.4070208966732025,
+      "learning_rate": 0.00019911312086936305,
+      "loss": 0.8196499943733215,
+      "step": 1694
+    },
+    {
+      "epoch": 0.7156118143459915,
+      "grad_norm": 0.4717724621295929,
+      "learning_rate": 0.0001991067269811039,
+      "loss": 0.9157413840293884,
+      "step": 1696
+    },
+    {
+      "epoch": 0.7164556962025317,
+      "grad_norm": 0.3942880928516388,
+      "learning_rate": 0.0001991003102308262,
+      "loss": 0.7880456447601318,
+      "step": 1698
+    },
+    {
+      "epoch": 0.7172995780590717,
+      "grad_norm": 0.4037776589393616,
+      "learning_rate": 0.0001990938706200102,
+      "loss": 0.8088646531105042,
+      "step": 1700
+    },
+    {
+      "epoch": 0.7172995780590717,
+      "eval_loss": 0.8708170056343079,
+      "eval_runtime": 670.3019,
+      "eval_samples_per_second": 3.143,
+      "eval_steps_per_second": 3.143,
+      "step": 1700
+    },
+    {
+      "epoch": 0.7181434599156118,
+      "grad_norm": 0.4443536400794983,
+      "learning_rate": 0.0001990874081501414,
+      "loss": 0.8740324378013611,
+      "step": 1702
+    },
+    {
+      "epoch": 0.7189873417721518,
+      "grad_norm": 0.45089733600616455,
+      "learning_rate": 0.00019908092282271053,
+      "loss": 0.8305023908615112,
+      "step": 1704
+    },
+    {
+      "epoch": 0.719831223628692,
+      "grad_norm": 0.3903036415576935,
+      "learning_rate": 0.00019907441463921366,
+      "loss": 0.7718394994735718,
+      "step": 1706
+    },
+    {
+      "epoch": 0.7206751054852321,
+      "grad_norm": 0.42990729212760925,
+      "learning_rate": 0.00019906788360115208,
+      "loss": 0.843289315700531,
+      "step": 1708
+    },
+    {
+      "epoch": 0.7215189873417721,
+      "grad_norm": 0.4287833571434021,
+      "learning_rate": 0.0001990613297100324,
+      "loss": 0.8332282900810242,
+      "step": 1710
+    },
+    {
+      "epoch": 0.7223628691983123,
+      "grad_norm": 0.4557269513607025,
+      "learning_rate": 0.00019905475296736646,
+      "loss": 0.784180223941803,
+      "step": 1712
+    },
+    {
+      "epoch": 0.7232067510548523,
+      "grad_norm": 0.4001525640487671,
+      "learning_rate": 0.0001990481533746714,
+      "loss": 0.7807716131210327,
+      "step": 1714
+    },
+    {
+      "epoch": 0.7240506329113924,
+      "grad_norm": 0.5208709836006165,
+      "learning_rate": 0.0001990415309334696,
+      "loss": 0.7936414480209351,
+      "step": 1716
+    },
+    {
+      "epoch": 0.7248945147679325,
+      "grad_norm": 0.46469953656196594,
+      "learning_rate": 0.00019903488564528875,
+      "loss": 0.7964845895767212,
+      "step": 1718
+    },
+    {
+      "epoch": 0.7257383966244726,
+      "grad_norm": 0.405773401260376,
+      "learning_rate": 0.00019902821751166178,
+      "loss": 0.8302215337753296,
+      "step": 1720
+    },
+    {
+      "epoch": 0.7265822784810126,
+      "grad_norm": 0.3874973654747009,
+      "learning_rate": 0.00019902152653412692,
+      "loss": 0.8486512899398804,
+      "step": 1722
+    },
+    {
+      "epoch": 0.7274261603375527,
+      "grad_norm": 0.46865755319595337,
+      "learning_rate": 0.00019901481271422758,
+      "loss": 0.837841272354126,
+      "step": 1724
+    },
+    {
+      "epoch": 0.7282700421940929,
+      "grad_norm": 0.4139123558998108,
+      "learning_rate": 0.0001990080760535126,
+      "loss": 0.7640942335128784,
+      "step": 1726
+    },
+    {
+      "epoch": 0.7291139240506329,
+      "grad_norm": 0.5003024935722351,
+      "learning_rate": 0.000199001316553536,
+      "loss": 0.8236244320869446,
+      "step": 1728
+    },
+    {
+      "epoch": 0.729957805907173,
+      "grad_norm": 0.38495776057243347,
+      "learning_rate": 0.00019899453421585696,
+      "loss": 0.7792612314224243,
+      "step": 1730
+    },
+    {
+      "epoch": 0.7308016877637131,
+      "grad_norm": 0.4462733268737793,
+      "learning_rate": 0.00019898772904204014,
+      "loss": 0.7733872532844543,
+      "step": 1732
+    },
+    {
+      "epoch": 0.7316455696202532,
+      "grad_norm": 0.398248553276062,
+      "learning_rate": 0.00019898090103365532,
+      "loss": 0.8933147192001343,
+      "step": 1734
+    },
+    {
+      "epoch": 0.7324894514767932,
+      "grad_norm": 0.4231175482273102,
+      "learning_rate": 0.00019897405019227756,
+      "loss": 0.8198356628417969,
+      "step": 1736
+    },
+    {
+      "epoch": 0.7333333333333333,
+      "grad_norm": 0.43639084696769714,
+      "learning_rate": 0.0001989671765194873,
+      "loss": 0.8266391158103943,
+      "step": 1738
+    },
+    {
+      "epoch": 0.7341772151898734,
+      "grad_norm": 0.4235835671424866,
+      "learning_rate": 0.00019896028001687008,
+      "loss": 0.7839997410774231,
+      "step": 1740
+    },
+    {
+      "epoch": 0.7350210970464135,
+      "grad_norm": 0.4033111035823822,
+      "learning_rate": 0.00019895336068601687,
+      "loss": 0.7153518199920654,
+      "step": 1742
+    },
+    {
+      "epoch": 0.7358649789029535,
+      "grad_norm": 0.44281265139579773,
+      "learning_rate": 0.00019894641852852377,
+      "loss": 0.9079289436340332,
+      "step": 1744
+    },
+    {
+      "epoch": 0.7367088607594937,
+      "grad_norm": 0.4882747232913971,
+      "learning_rate": 0.0001989394535459922,
+      "loss": 0.8645254373550415,
+      "step": 1746
+    },
+    {
+      "epoch": 0.7375527426160338,
+      "grad_norm": 0.34858641028404236,
+      "learning_rate": 0.00019893246574002887,
+      "loss": 0.8134214878082275,
+      "step": 1748
+    },
+    {
+      "epoch": 0.7383966244725738,
+      "grad_norm": 0.4273303747177124,
+      "learning_rate": 0.00019892545511224574,
+      "loss": 0.8142663240432739,
+      "step": 1750
+    },
+    {
+      "epoch": 0.739240506329114,
+      "grad_norm": 0.4289957284927368,
+      "learning_rate": 0.00019891842166426004,
+      "loss": 0.7957643866539001,
+      "step": 1752
+    },
+    {
+      "epoch": 0.740084388185654,
+      "grad_norm": 0.4812224805355072,
+      "learning_rate": 0.00019891136539769416,
+      "loss": 0.9000363945960999,
+      "step": 1754
+    },
+    {
+      "epoch": 0.7409282700421941,
+      "grad_norm": 0.4078022837638855,
+      "learning_rate": 0.00019890428631417595,
+      "loss": 0.7719792723655701,
+      "step": 1756
+    },
+    {
+      "epoch": 0.7417721518987341,
+      "grad_norm": 0.40169858932495117,
+      "learning_rate": 0.00019889718441533838,
+      "loss": 0.7681847214698792,
+      "step": 1758
+    },
+    {
+      "epoch": 0.7426160337552743,
+      "grad_norm": 0.3713594377040863,
+      "learning_rate": 0.00019889005970281972,
+      "loss": 0.7536827325820923,
+      "step": 1760
+    },
+    {
+      "epoch": 0.7434599156118143,
+      "grad_norm": 0.4183627665042877,
+      "learning_rate": 0.0001988829121782635,
+      "loss": 0.9000160694122314,
+      "step": 1762
+    },
+    {
+      "epoch": 0.7443037974683544,
+      "grad_norm": 0.4241594672203064,
+      "learning_rate": 0.00019887574184331851,
+      "loss": 0.8047307133674622,
+      "step": 1764
+    },
+    {
+      "epoch": 0.7451476793248946,
+      "grad_norm": 0.4249233603477478,
+      "learning_rate": 0.00019886854869963883,
+      "loss": 0.8932583928108215,
+      "step": 1766
+    },
+    {
+      "epoch": 0.7459915611814346,
+      "grad_norm": 0.35815340280532837,
+      "learning_rate": 0.00019886133274888378,
+      "loss": 0.7684977054595947,
+      "step": 1768
+    },
+    {
+      "epoch": 0.7468354430379747,
+      "grad_norm": 0.41035401821136475,
+      "learning_rate": 0.00019885409399271795,
+      "loss": 0.8644338250160217,
+      "step": 1770
+    },
+    {
+      "epoch": 0.7476793248945147,
+      "grad_norm": 0.5450286269187927,
+      "learning_rate": 0.00019884683243281116,
+      "loss": 0.8035860657691956,
+      "step": 1772
+    },
+    {
+      "epoch": 0.7485232067510549,
+      "grad_norm": 0.5280999541282654,
+      "learning_rate": 0.0001988395480708385,
+      "loss": 0.8861207962036133,
+      "step": 1774
+    },
+    {
+      "epoch": 0.7493670886075949,
+      "grad_norm": 0.42665231227874756,
+      "learning_rate": 0.00019883224090848036,
+      "loss": 0.7506847977638245,
+      "step": 1776
+    },
+    {
+      "epoch": 0.750210970464135,
+      "grad_norm": 0.39029282331466675,
+      "learning_rate": 0.00019882491094742232,
+      "loss": 0.7662046551704407,
+      "step": 1778
+    },
+    {
+      "epoch": 0.7510548523206751,
+      "grad_norm": 0.43870019912719727,
+      "learning_rate": 0.00019881755818935534,
+      "loss": 0.7590143084526062,
+      "step": 1780
+    },
+    {
+      "epoch": 0.7518987341772152,
+      "grad_norm": 0.4195050001144409,
+      "learning_rate": 0.0001988101826359755,
+      "loss": 0.9340365529060364,
+      "step": 1782
+    },
+    {
+      "epoch": 0.7527426160337553,
+      "grad_norm": 0.4236123263835907,
+      "learning_rate": 0.0001988027842889842,
+      "loss": 0.7969209551811218,
+      "step": 1784
+    },
+    {
+      "epoch": 0.7535864978902953,
+      "grad_norm": 0.4274357259273529,
+      "learning_rate": 0.00019879536315008808,
+      "loss": 0.7892382144927979,
+      "step": 1786
+    },
+    {
+      "epoch": 0.7544303797468355,
+      "grad_norm": 0.4622916579246521,
+      "learning_rate": 0.00019878791922099912,
+      "loss": 0.8752562403678894,
+      "step": 1788
+    },
+    {
+      "epoch": 0.7552742616033755,
+      "grad_norm": 0.3648734986782074,
+      "learning_rate": 0.00019878045250343445,
+      "loss": 0.813011884689331,
+      "step": 1790
+    },
+    {
+      "epoch": 0.7561181434599156,
+      "grad_norm": 0.41361239552497864,
+      "learning_rate": 0.00019877296299911648,
+      "loss": 0.7310198545455933,
+      "step": 1792
+    },
+    {
+      "epoch": 0.7569620253164557,
+      "grad_norm": 0.4113222062587738,
+      "learning_rate": 0.00019876545070977294,
+      "loss": 0.858386754989624,
+      "step": 1794
+    },
+    {
+      "epoch": 0.7578059071729958,
+      "grad_norm": 0.4295370876789093,
+      "learning_rate": 0.00019875791563713676,
+      "loss": 0.8377325534820557,
+      "step": 1796
+    },
+    {
+      "epoch": 0.7586497890295358,
+      "grad_norm": 0.4227522015571594,
+      "learning_rate": 0.0001987503577829461,
+      "loss": 0.9139418005943298,
+      "step": 1798
+    },
+    {
+      "epoch": 0.759493670886076,
+      "grad_norm": 0.3638151288032532,
+      "learning_rate": 0.00019874277714894442,
+      "loss": 0.6994872689247131,
+      "step": 1800
+    },
+    {
+      "epoch": 0.759493670886076,
+      "eval_loss": 0.8625519275665283,
+      "eval_runtime": 686.4271,
+      "eval_samples_per_second": 3.07,
+      "eval_steps_per_second": 3.07,
+      "step": 1800
+    },
+    {
+      "epoch": 0.760337552742616,
+      "grad_norm": 0.46954411268234253,
+      "learning_rate": 0.00019873517373688047,
+      "loss": 0.8681167960166931,
+      "step": 1802
+    },
+    {
+      "epoch": 0.7611814345991561,
+      "grad_norm": 0.41958674788475037,
+      "learning_rate": 0.00019872754754850819,
+      "loss": 0.7420852184295654,
+      "step": 1804
+    },
+    {
+      "epoch": 0.7620253164556962,
+      "grad_norm": 0.431226521730423,
+      "learning_rate": 0.00019871989858558678,
+      "loss": 0.8018608093261719,
+      "step": 1806
+    },
+    {
+      "epoch": 0.7628691983122363,
+      "grad_norm": 0.45129454135894775,
+      "learning_rate": 0.0001987122268498807,
+      "loss": 0.8793904781341553,
+      "step": 1808
+    },
+    {
+      "epoch": 0.7637130801687764,
+      "grad_norm": 0.44278961420059204,
+      "learning_rate": 0.00019870453234315972,
+      "loss": 0.8416730761528015,
+      "step": 1810
+    },
+    {
+      "epoch": 0.7645569620253164,
+      "grad_norm": 0.4689098000526428,
+      "learning_rate": 0.00019869681506719876,
+      "loss": 0.8225743174552917,
+      "step": 1812
+    },
+    {
+      "epoch": 0.7654008438818566,
+      "grad_norm": 0.4413386881351471,
+      "learning_rate": 0.00019868907502377806,
+      "loss": 0.7871913313865662,
+      "step": 1814
+    },
+    {
+      "epoch": 0.7662447257383966,
+      "grad_norm": 0.48395273089408875,
+      "learning_rate": 0.00019868131221468315,
+      "loss": 0.8616237044334412,
+      "step": 1816
+    },
+    {
+      "epoch": 0.7670886075949367,
+      "grad_norm": 0.370263934135437,
+      "learning_rate": 0.0001986735266417047,
+      "loss": 0.7243452668190002,
+      "step": 1818
+    },
+    {
+      "epoch": 0.7679324894514767,
+      "grad_norm": 0.4522445797920227,
+      "learning_rate": 0.00019866571830663875,
+      "loss": 0.8623812198638916,
+      "step": 1820
+    },
+    {
+      "epoch": 0.7687763713080169,
+      "grad_norm": 0.43821004033088684,
+      "learning_rate": 0.00019865788721128643,
+      "loss": 0.776618480682373,
+      "step": 1822
+    },
+    {
+      "epoch": 0.769620253164557,
+      "grad_norm": 0.3825650215148926,
+      "learning_rate": 0.00019865003335745436,
+      "loss": 0.8722774982452393,
+      "step": 1824
+    },
+    {
+      "epoch": 0.770464135021097,
+      "grad_norm": 0.43684205412864685,
+      "learning_rate": 0.00019864215674695416,
+      "loss": 0.907802939414978,
+      "step": 1826
+    },
+    {
+      "epoch": 0.7713080168776372,
+      "grad_norm": 0.4128975570201874,
+      "learning_rate": 0.00019863425738160287,
+      "loss": 0.8505539298057556,
+      "step": 1828
+    },
+    {
+      "epoch": 0.7721518987341772,
+      "grad_norm": 0.4254065454006195,
+      "learning_rate": 0.0001986263352632227,
+      "loss": 0.83274906873703,
+      "step": 1830
+    },
+    {
+      "epoch": 0.7729957805907173,
+      "grad_norm": 0.4137375056743622,
+      "learning_rate": 0.00019861839039364118,
+      "loss": 0.8172078728675842,
+      "step": 1832
+    },
+    {
+      "epoch": 0.7738396624472574,
+      "grad_norm": 0.4387703239917755,
+      "learning_rate": 0.00019861042277469096,
+      "loss": 0.8610842227935791,
+      "step": 1834
+    },
+    {
+      "epoch": 0.7746835443037975,
+      "grad_norm": 0.398952454328537,
+      "learning_rate": 0.00019860243240821005,
+      "loss": 0.7961188554763794,
+      "step": 1836
+    },
+    {
+      "epoch": 0.7755274261603375,
+      "grad_norm": 0.4079921543598175,
+      "learning_rate": 0.0001985944192960417,
+      "loss": 0.8078321218490601,
+      "step": 1838
+    },
+    {
+      "epoch": 0.7763713080168776,
+      "grad_norm": 0.4701489508152008,
+      "learning_rate": 0.00019858638344003433,
+      "loss": 0.8743909001350403,
+      "step": 1840
+    },
+    {
+      "epoch": 0.7772151898734178,
+      "grad_norm": 0.5391269326210022,
+      "learning_rate": 0.00019857832484204173,
+      "loss": 0.7681707143783569,
+      "step": 1842
+    },
+    {
+      "epoch": 0.7780590717299578,
+      "grad_norm": 0.36250734329223633,
+      "learning_rate": 0.00019857024350392277,
+      "loss": 0.7817525863647461,
+      "step": 1844
+    },
+    {
+      "epoch": 0.7789029535864979,
+      "grad_norm": 0.41237136721611023,
+      "learning_rate": 0.00019856213942754175,
+      "loss": 0.7614551782608032,
+      "step": 1846
+    },
+    {
+      "epoch": 0.779746835443038,
+      "grad_norm": 0.3968869745731354,
+      "learning_rate": 0.00019855401261476807,
+      "loss": 0.7492313981056213,
+      "step": 1848
+    },
+    {
+      "epoch": 0.7805907172995781,
+      "grad_norm": 0.441476434469223,
+      "learning_rate": 0.00019854586306747646,
+      "loss": 0.8615695238113403,
+      "step": 1850
+    },
+    {
+      "epoch": 0.7814345991561181,
+      "grad_norm": 0.4486154317855835,
+      "learning_rate": 0.00019853769078754686,
+      "loss": 0.8416724801063538,
+      "step": 1852
+    },
+    {
+      "epoch": 0.7822784810126582,
+      "grad_norm": 0.45148879289627075,
+      "learning_rate": 0.0001985294957768644,
+      "loss": 0.8987806439399719,
+      "step": 1854
+    },
+    {
+      "epoch": 0.7831223628691983,
+      "grad_norm": 0.39330095052719116,
+      "learning_rate": 0.0001985212780373196,
+      "loss": 0.7597590088844299,
+      "step": 1856
+    },
+    {
+      "epoch": 0.7839662447257384,
+      "grad_norm": 0.4260408580303192,
+      "learning_rate": 0.00019851303757080808,
+      "loss": 0.8429927229881287,
+      "step": 1858
+    },
+    {
+      "epoch": 0.7848101265822784,
+      "grad_norm": 0.37519171833992004,
+      "learning_rate": 0.00019850477437923075,
+      "loss": 0.7443564534187317,
+      "step": 1860
+    },
+    {
+      "epoch": 0.7856540084388186,
+      "grad_norm": 0.4991084337234497,
+      "learning_rate": 0.00019849648846449383,
+      "loss": 0.934238612651825,
+      "step": 1862
+    },
+    {
+      "epoch": 0.7864978902953587,
+      "grad_norm": 0.4711335003376007,
+      "learning_rate": 0.00019848817982850867,
+      "loss": 0.8938905000686646,
+      "step": 1864
+    },
+    {
+      "epoch": 0.7873417721518987,
+      "grad_norm": 0.4145370423793793,
+      "learning_rate": 0.00019847984847319193,
+      "loss": 0.8745643496513367,
+      "step": 1866
+    },
+    {
+      "epoch": 0.7881856540084389,
+      "grad_norm": 0.3444044888019562,
+      "learning_rate": 0.00019847149440046548,
+      "loss": 0.7178786396980286,
+      "step": 1868
+    },
+    {
+      "epoch": 0.7890295358649789,
+      "grad_norm": 0.34632962942123413,
+      "learning_rate": 0.00019846311761225646,
+      "loss": 0.9114303588867188,
+      "step": 1870
+    },
+    {
+      "epoch": 0.789873417721519,
+      "grad_norm": 0.38630741834640503,
+      "learning_rate": 0.0001984547181104972,
+      "loss": 0.836281955242157,
+      "step": 1872
+    },
+    {
+      "epoch": 0.790717299578059,
+      "grad_norm": 0.40947434306144714,
+      "learning_rate": 0.00019844629589712534,
+      "loss": 0.8247858881950378,
+      "step": 1874
+    },
+    {
+      "epoch": 0.7915611814345992,
+      "grad_norm": 0.45878684520721436,
+      "learning_rate": 0.00019843785097408372,
+      "loss": 0.8446351289749146,
+      "step": 1876
+    },
+    {
+      "epoch": 0.7924050632911392,
+      "grad_norm": 0.37202852964401245,
+      "learning_rate": 0.0001984293833433204,
+      "loss": 0.8567686676979065,
+      "step": 1878
+    },
+    {
+      "epoch": 0.7932489451476793,
+      "grad_norm": 0.3839069902896881,
+      "learning_rate": 0.00019842089300678876,
+      "loss": 0.8134047985076904,
+      "step": 1880
+    },
+    {
+      "epoch": 0.7940928270042195,
+      "grad_norm": 0.4018687605857849,
+      "learning_rate": 0.00019841237996644727,
+      "loss": 0.8670110106468201,
+      "step": 1882
+    },
+    {
+      "epoch": 0.7949367088607595,
+      "grad_norm": 0.372089684009552,
+      "learning_rate": 0.0001984038442242598,
+      "loss": 0.7975355386734009,
+      "step": 1884
+    },
+    {
+      "epoch": 0.7957805907172996,
+      "grad_norm": 0.4492705166339874,
+      "learning_rate": 0.0001983952857821953,
+      "loss": 0.8755611777305603,
+      "step": 1886
+    },
+    {
+      "epoch": 0.7966244725738396,
+      "grad_norm": 0.451054185628891,
+      "learning_rate": 0.00019838670464222813,
+      "loss": 0.8680241107940674,
+      "step": 1888
+    },
+    {
+      "epoch": 0.7974683544303798,
+      "grad_norm": 0.4036900997161865,
+      "learning_rate": 0.00019837810080633773,
+      "loss": 0.7990990877151489,
+      "step": 1890
+    },
+    {
+      "epoch": 0.7983122362869198,
+      "grad_norm": 0.46125802397727966,
+      "learning_rate": 0.0001983694742765089,
+      "loss": 0.8598953485488892,
+      "step": 1892
+    },
+    {
+      "epoch": 0.7991561181434599,
+      "grad_norm": 0.4458785951137543,
+      "learning_rate": 0.00019836082505473153,
+      "loss": 0.7707474827766418,
+      "step": 1894
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.44501590728759766,
+      "learning_rate": 0.0001983521531430009,
+      "loss": 0.8069534301757812,
+      "step": 1896
+    },
+    {
+      "epoch": 0.8008438818565401,
+      "grad_norm": 0.400771826505661,
+      "learning_rate": 0.0001983434585433174,
+      "loss": 0.862334668636322,
+      "step": 1898
+    },
+    {
+      "epoch": 0.8016877637130801,
+      "grad_norm": 0.39193347096443176,
+      "learning_rate": 0.00019833474125768676,
+      "loss": 0.8125433325767517,
+      "step": 1900
+    },
+    {
+      "epoch": 0.8016877637130801,
+      "eval_loss": 0.8546335697174072,
+      "eval_runtime": 688.5301,
+      "eval_samples_per_second": 3.06,
+      "eval_steps_per_second": 3.06,
+      "step": 1900
+    },
+    {
+      "epoch": 0.8025316455696202,
+      "grad_norm": 0.441129207611084,
+      "learning_rate": 0.00019832600128811986,
+      "loss": 0.8246012330055237,
+      "step": 1902
+    },
+    {
+      "epoch": 0.8033755274261604,
+      "grad_norm": 0.4089467227458954,
+      "learning_rate": 0.00019831723863663285,
+      "loss": 0.8387641310691833,
+      "step": 1904
+    },
+    {
+      "epoch": 0.8042194092827004,
+      "grad_norm": 0.4871654510498047,
+      "learning_rate": 0.0001983084533052471,
+      "loss": 0.7891429662704468,
+      "step": 1906
+    },
+    {
+      "epoch": 0.8050632911392405,
+      "grad_norm": 0.40085604786872864,
+      "learning_rate": 0.00019829964529598923,
+      "loss": 0.8008774518966675,
+      "step": 1908
+    },
+    {
+      "epoch": 0.8059071729957806,
+      "grad_norm": 0.38775014877319336,
+      "learning_rate": 0.00019829081461089109,
+      "loss": 0.7921834588050842,
+      "step": 1910
+    },
+    {
+      "epoch": 0.8067510548523207,
+      "grad_norm": 0.37830594182014465,
+      "learning_rate": 0.0001982819612519897,
+      "loss": 0.7752519845962524,
+      "step": 1912
+    },
+    {
+      "epoch": 0.8075949367088607,
+      "grad_norm": 0.41137194633483887,
+      "learning_rate": 0.0001982730852213274,
+      "loss": 0.7958255410194397,
+      "step": 1914
+    },
+    {
+      "epoch": 0.8084388185654009,
+      "grad_norm": 0.4871830940246582,
+      "learning_rate": 0.0001982641865209517,
+      "loss": 0.8177281618118286,
+      "step": 1916
+    },
+    {
+      "epoch": 0.809282700421941,
+      "grad_norm": 0.4157388210296631,
+      "learning_rate": 0.0001982552651529154,
+      "loss": 0.7726616859436035,
+      "step": 1918
+    },
+    {
+      "epoch": 0.810126582278481,
+      "grad_norm": 0.32338014245033264,
+      "learning_rate": 0.00019824632111927645,
+      "loss": 0.6601400375366211,
+      "step": 1920
+    },
+    {
+      "epoch": 0.810970464135021,
+      "grad_norm": 0.4642949104309082,
+      "learning_rate": 0.00019823735442209804,
+      "loss": 0.7866622805595398,
+      "step": 1922
+    },
+    {
+      "epoch": 0.8118143459915612,
+      "grad_norm": 0.43558797240257263,
+      "learning_rate": 0.00019822836506344865,
+      "loss": 0.7857986092567444,
+      "step": 1924
+    },
+    {
+      "epoch": 0.8126582278481013,
+      "grad_norm": 0.3996953070163727,
+      "learning_rate": 0.000198219353045402,
+      "loss": 0.8407763838768005,
+      "step": 1926
+    },
+    {
+      "epoch": 0.8135021097046413,
+      "grad_norm": 0.4415414333343506,
+      "learning_rate": 0.00019821031837003686,
+      "loss": 0.8433752655982971,
+      "step": 1928
+    },
+    {
+      "epoch": 0.8143459915611815,
+      "grad_norm": 0.37329310178756714,
+      "learning_rate": 0.00019820126103943752,
+      "loss": 0.6941158175468445,
+      "step": 1930
+    },
+    {
+      "epoch": 0.8151898734177215,
+      "grad_norm": 0.4482601284980774,
+      "learning_rate": 0.00019819218105569323,
+      "loss": 0.775604248046875,
+      "step": 1932
+    },
+    {
+      "epoch": 0.8160337552742616,
+      "grad_norm": 0.426194429397583,
+      "learning_rate": 0.00019818307842089854,
+      "loss": 0.8496418595314026,
+      "step": 1934
+    },
+    {
+      "epoch": 0.8168776371308016,
+      "grad_norm": 0.42379093170166016,
+      "learning_rate": 0.00019817395313715335,
+      "loss": 0.7767958045005798,
+      "step": 1936
+    },
+    {
+      "epoch": 0.8177215189873418,
+      "grad_norm": 0.4149782061576843,
+      "learning_rate": 0.0001981648052065626,
+      "loss": 0.7347666025161743,
+      "step": 1938
+    },
+    {
+      "epoch": 0.8185654008438819,
+      "grad_norm": 0.3927431106567383,
+      "learning_rate": 0.0001981556346312366,
+      "loss": 0.7839647531509399,
+      "step": 1940
+    },
+    {
+      "epoch": 0.8194092827004219,
+      "grad_norm": 0.4483136236667633,
+      "learning_rate": 0.00019814644141329083,
+      "loss": 0.8870531916618347,
+      "step": 1942
+    },
+    {
+      "epoch": 0.8202531645569621,
+      "grad_norm": 0.4400147497653961,
+      "learning_rate": 0.00019813722555484594,
+      "loss": 0.8646620512008667,
+      "step": 1944
+    },
+    {
+      "epoch": 0.8210970464135021,
+      "grad_norm": 0.469971626996994,
+      "learning_rate": 0.00019812798705802785,
+      "loss": 0.9576541781425476,
+      "step": 1946
+    },
+    {
+      "epoch": 0.8219409282700422,
+      "grad_norm": 0.388113409280777,
+      "learning_rate": 0.00019811872592496776,
+      "loss": 0.7956477403640747,
+      "step": 1948
+    },
+    {
+      "epoch": 0.8227848101265823,
+      "grad_norm": 0.4271424114704132,
+      "learning_rate": 0.00019810944215780201,
+      "loss": 0.8440352082252502,
+      "step": 1950
+    },
+    {
+      "epoch": 0.8236286919831224,
+      "grad_norm": 0.4097966253757477,
+      "learning_rate": 0.00019810013575867217,
+      "loss": 0.8589065074920654,
+      "step": 1952
+    },
+    {
+      "epoch": 0.8244725738396624,
+      "grad_norm": 0.41550710797309875,
+      "learning_rate": 0.00019809080672972504,
+      "loss": 0.8606626987457275,
+      "step": 1954
+    },
+    {
+      "epoch": 0.8253164556962025,
+      "grad_norm": 0.3680916130542755,
+      "learning_rate": 0.0001980814550731127,
+      "loss": 0.7582436800003052,
+      "step": 1956
+    },
+    {
+      "epoch": 0.8261603375527427,
+      "grad_norm": 0.573297917842865,
+      "learning_rate": 0.00019807208079099234,
+      "loss": 0.8345186710357666,
+      "step": 1958
+    },
+    {
+      "epoch": 0.8270042194092827,
+      "grad_norm": 0.43631303310394287,
+      "learning_rate": 0.00019806268388552646,
+      "loss": 0.889683723449707,
+      "step": 1960
+    },
+    {
+      "epoch": 0.8278481012658228,
+      "grad_norm": 0.442531019449234,
+      "learning_rate": 0.00019805326435888275,
+      "loss": 0.8092973232269287,
+      "step": 1962
+    },
+    {
+      "epoch": 0.8286919831223629,
+      "grad_norm": 0.45020824670791626,
+      "learning_rate": 0.0001980438222132341,
+      "loss": 0.8919535279273987,
+      "step": 1964
+    },
+    {
+      "epoch": 0.829535864978903,
+      "grad_norm": 0.43131786584854126,
+      "learning_rate": 0.0001980343574507586,
+      "loss": 0.7652381658554077,
+      "step": 1966
+    },
+    {
+      "epoch": 0.830379746835443,
+      "grad_norm": 0.4461326599121094,
+      "learning_rate": 0.00019802487007363967,
+      "loss": 0.8512389659881592,
+      "step": 1968
+    },
+    {
+      "epoch": 0.8312236286919831,
+      "grad_norm": 0.4676894247531891,
+      "learning_rate": 0.0001980153600840658,
+      "loss": 0.8681327700614929,
+      "step": 1970
+    },
+    {
+      "epoch": 0.8320675105485232,
+      "grad_norm": 0.38376888632774353,
+      "learning_rate": 0.00019800582748423075,
+      "loss": 0.7476955652236938,
+      "step": 1972
+    },
+    {
+      "epoch": 0.8329113924050633,
+      "grad_norm": 0.37173229455947876,
+      "learning_rate": 0.00019799627227633354,
+      "loss": 0.814192533493042,
+      "step": 1974
+    },
+    {
+      "epoch": 0.8337552742616033,
+      "grad_norm": 0.4345237910747528,
+      "learning_rate": 0.00019798669446257844,
+      "loss": 0.7668994069099426,
+      "step": 1976
+    },
+    {
+      "epoch": 0.8345991561181435,
+      "grad_norm": 0.45872583985328674,
+      "learning_rate": 0.0001979770940451747,
+      "loss": 0.8321775197982788,
+      "step": 1978
+    },
+    {
+      "epoch": 0.8354430379746836,
+      "grad_norm": 0.40618205070495605,
+      "learning_rate": 0.0001979674710263371,
+      "loss": 0.7824342250823975,
+      "step": 1980
+    },
+    {
+      "epoch": 0.8362869198312236,
+      "grad_norm": 0.38377392292022705,
+      "learning_rate": 0.00019795782540828544,
+      "loss": 0.8752480745315552,
+      "step": 1982
+    },
+    {
+      "epoch": 0.8371308016877637,
+      "grad_norm": 0.36340072751045227,
+      "learning_rate": 0.00019794815719324476,
+      "loss": 0.66700679063797,
+      "step": 1984
+    },
+    {
+      "epoch": 0.8379746835443038,
+      "grad_norm": 0.4638384282588959,
+      "learning_rate": 0.00019793846638344536,
+      "loss": 0.8582209944725037,
+      "step": 1986
+    },
+    {
+      "epoch": 0.8388185654008439,
+      "grad_norm": 0.38364699482917786,
+      "learning_rate": 0.00019792875298112268,
+      "loss": 0.7724968791007996,
+      "step": 1988
+    },
+    {
+      "epoch": 0.8396624472573839,
+      "grad_norm": 0.439730703830719,
+      "learning_rate": 0.00019791901698851749,
+      "loss": 0.8542404174804688,
+      "step": 1990
+    },
+    {
+      "epoch": 0.8405063291139241,
+      "grad_norm": 0.4622722566127777,
+      "learning_rate": 0.00019790925840787563,
+      "loss": 0.8424296975135803,
+      "step": 1992
+    },
+    {
+      "epoch": 0.8413502109704641,
+      "grad_norm": 0.4491989314556122,
+      "learning_rate": 0.0001978994772414482,
+      "loss": 0.8400710225105286,
+      "step": 1994
+    },
+    {
+      "epoch": 0.8421940928270042,
+      "grad_norm": 0.372295081615448,
+      "learning_rate": 0.0001978896734914916,
+      "loss": 0.7265452146530151,
+      "step": 1996
+    },
+    {
+      "epoch": 0.8430379746835444,
+      "grad_norm": 0.4328666925430298,
+      "learning_rate": 0.00019787984716026732,
+      "loss": 0.8458228707313538,
+      "step": 1998
+    },
+    {
+      "epoch": 0.8438818565400844,
+      "grad_norm": 0.42246317863464355,
+      "learning_rate": 0.00019786999825004216,
+      "loss": 0.8517491817474365,
+      "step": 2000
+    },
+    {
+      "epoch": 0.8438818565400844,
+      "eval_loss": 0.8460908532142639,
+      "eval_runtime": 685.2518,
+      "eval_samples_per_second": 3.075,
+      "eval_steps_per_second": 3.075,
+      "step": 2000
+    },
+    {
+      "epoch": 0.8447257383966245,
+      "grad_norm": 0.3927323818206787,
+      "learning_rate": 0.00019786012676308802,
+      "loss": 0.8045481443405151,
+      "step": 2002
+    },
+    {
+      "epoch": 0.8455696202531645,
+      "grad_norm": 0.38189369440078735,
+      "learning_rate": 0.00019785023270168202,
+      "loss": 0.7509838938713074,
+      "step": 2004
+    },
+    {
+      "epoch": 0.8464135021097047,
+      "grad_norm": 0.40383535623550415,
+      "learning_rate": 0.00019784031606810663,
+      "loss": 0.7730163335800171,
+      "step": 2006
+    },
+    {
+      "epoch": 0.8472573839662447,
+      "grad_norm": 0.3665419816970825,
+      "learning_rate": 0.00019783037686464935,
+      "loss": 0.7096828818321228,
+      "step": 2008
+    },
+    {
+      "epoch": 0.8481012658227848,
+      "grad_norm": 0.36623379588127136,
+      "learning_rate": 0.00019782041509360302,
+      "loss": 0.8186401128768921,
+      "step": 2010
+    },
+    {
+      "epoch": 0.8489451476793249,
+      "grad_norm": 0.477255642414093,
+      "learning_rate": 0.0001978104307572656,
+      "loss": 0.8502879738807678,
+      "step": 2012
+    },
+    {
+      "epoch": 0.849789029535865,
+      "grad_norm": 0.44456636905670166,
+      "learning_rate": 0.00019780042385794032,
+      "loss": 0.9010884761810303,
+      "step": 2014
+    },
+    {
+      "epoch": 0.850632911392405,
+      "grad_norm": 0.42378175258636475,
+      "learning_rate": 0.00019779039439793552,
+      "loss": 0.774183452129364,
+      "step": 2016
+    },
+    {
+      "epoch": 0.8514767932489451,
+      "grad_norm": 0.43021997809410095,
+      "learning_rate": 0.00019778034237956482,
+      "loss": 0.8487694263458252,
+      "step": 2018
+    },
+    {
+      "epoch": 0.8523206751054853,
+      "grad_norm": 0.38188812136650085,
+      "learning_rate": 0.00019777026780514704,
+      "loss": 0.8004868030548096,
+      "step": 2020
+    },
+    {
+      "epoch": 0.8531645569620253,
+      "grad_norm": 0.3684130609035492,
+      "learning_rate": 0.0001977601706770062,
+      "loss": 0.7680953145027161,
+      "step": 2022
+    },
+    {
+      "epoch": 0.8540084388185654,
+      "grad_norm": 0.444844514131546,
+      "learning_rate": 0.00019775005099747153,
+      "loss": 0.8990264534950256,
+      "step": 2024
+    },
+    {
+      "epoch": 0.8548523206751055,
+      "grad_norm": 0.3432365655899048,
+      "learning_rate": 0.00019773990876887737,
+      "loss": 0.7676048874855042,
+      "step": 2026
+    },
+    {
+      "epoch": 0.8556962025316456,
+      "grad_norm": 0.3728618025779724,
+      "learning_rate": 0.00019772974399356343,
+      "loss": 0.8546769618988037,
+      "step": 2028
+    },
+    {
+      "epoch": 0.8565400843881856,
+      "grad_norm": 0.42767980694770813,
+      "learning_rate": 0.00019771955667387449,
+      "loss": 0.8009977340698242,
+      "step": 2030
+    },
+    {
+      "epoch": 0.8573839662447258,
+      "grad_norm": 0.4125710129737854,
+      "learning_rate": 0.00019770934681216055,
+      "loss": 0.8424299955368042,
+      "step": 2032
+    },
+    {
+      "epoch": 0.8582278481012658,
+      "grad_norm": 0.4399266839027405,
+      "learning_rate": 0.00019769911441077685,
+      "loss": 0.7940502762794495,
+      "step": 2034
+    },
+    {
+      "epoch": 0.8590717299578059,
+      "grad_norm": 0.41797709465026855,
+      "learning_rate": 0.0001976888594720838,
+      "loss": 0.9424314498901367,
+      "step": 2036
+    },
+    {
+      "epoch": 0.859915611814346,
+      "grad_norm": 0.5245009660720825,
+      "learning_rate": 0.00019767858199844698,
+      "loss": 0.9102896451950073,
+      "step": 2038
+    },
+    {
+      "epoch": 0.8607594936708861,
+      "grad_norm": 0.4092593193054199,
+      "learning_rate": 0.00019766828199223728,
+      "loss": 0.7434720396995544,
+      "step": 2040
+    },
+    {
+      "epoch": 0.8616033755274262,
+      "grad_norm": 0.40787473320961,
+      "learning_rate": 0.00019765795945583068,
+      "loss": 0.7441987991333008,
+      "step": 2042
+    },
+    {
+      "epoch": 0.8624472573839662,
+      "grad_norm": 0.42438021302223206,
+      "learning_rate": 0.0001976476143916084,
+      "loss": 0.8317872881889343,
+      "step": 2044
+    },
+    {
+      "epoch": 0.8632911392405064,
+      "grad_norm": 0.38261914253234863,
+      "learning_rate": 0.00019763724680195682,
+      "loss": 0.764509379863739,
+      "step": 2046
+    },
+    {
+      "epoch": 0.8641350210970464,
+      "grad_norm": 0.43295741081237793,
+      "learning_rate": 0.00019762685668926758,
+      "loss": 0.8666167855262756,
+      "step": 2048
+    },
+    {
+      "epoch": 0.8649789029535865,
+      "grad_norm": 0.36384403705596924,
+      "learning_rate": 0.00019761644405593743,
+      "loss": 0.7606754302978516,
+      "step": 2050
+    },
+    {
+      "epoch": 0.8658227848101265,
+      "grad_norm": 0.4296814501285553,
+      "learning_rate": 0.0001976060089043684,
+      "loss": 0.827495813369751,
+      "step": 2052
+    },
+    {
+      "epoch": 0.8666666666666667,
+      "grad_norm": 0.3831455111503601,
+      "learning_rate": 0.0001975955512369677,
+      "loss": 0.7418577075004578,
+      "step": 2054
+    },
+    {
+      "epoch": 0.8675105485232067,
+      "grad_norm": 0.44710344076156616,
+      "learning_rate": 0.0001975850710561477,
+      "loss": 0.8603045344352722,
+      "step": 2056
+    },
+    {
+      "epoch": 0.8683544303797468,
+      "grad_norm": 0.44012558460235596,
+      "learning_rate": 0.00019757456836432593,
+      "loss": 0.900658905506134,
+      "step": 2058
+    },
+    {
+      "epoch": 0.869198312236287,
+      "grad_norm": 0.42166298627853394,
+      "learning_rate": 0.0001975640431639252,
+      "loss": 0.8416497707366943,
+      "step": 2060
+    },
+    {
+      "epoch": 0.870042194092827,
+      "grad_norm": 0.40824490785598755,
+      "learning_rate": 0.00019755349545737345,
+      "loss": 0.8281066417694092,
+      "step": 2062
+    },
+    {
+      "epoch": 0.8708860759493671,
+      "grad_norm": 0.4272223711013794,
+      "learning_rate": 0.00019754292524710388,
+      "loss": 0.83869469165802,
+      "step": 2064
+    },
+    {
+      "epoch": 0.8717299578059071,
+      "grad_norm": 0.3732798397541046,
+      "learning_rate": 0.00019753233253555478,
+      "loss": 0.8337644934654236,
+      "step": 2066
+    },
+    {
+      "epoch": 0.8725738396624473,
+      "grad_norm": 0.4048069417476654,
+      "learning_rate": 0.00019752171732516973,
+      "loss": 0.7677452564239502,
+      "step": 2068
+    },
+    {
+      "epoch": 0.8734177215189873,
+      "grad_norm": 0.43628430366516113,
+      "learning_rate": 0.00019751107961839742,
+      "loss": 0.8320774435997009,
+      "step": 2070
+    },
+    {
+      "epoch": 0.8742616033755274,
+      "grad_norm": 0.43674105405807495,
+      "learning_rate": 0.00019750041941769174,
+      "loss": 0.8480262160301208,
+      "step": 2072
+    },
+    {
+      "epoch": 0.8751054852320675,
+      "grad_norm": 0.3678584098815918,
+      "learning_rate": 0.0001974897367255119,
+      "loss": 0.7598503828048706,
+      "step": 2074
+    },
+    {
+      "epoch": 0.8759493670886076,
+      "grad_norm": 0.4316411316394806,
+      "learning_rate": 0.0001974790315443221,
+      "loss": 0.7907084822654724,
+      "step": 2076
+    },
+    {
+      "epoch": 0.8767932489451477,
+      "grad_norm": 0.426465779542923,
+      "learning_rate": 0.00019746830387659182,
+      "loss": 0.8287727236747742,
+      "step": 2078
+    },
+    {
+      "epoch": 0.8776371308016878,
+      "grad_norm": 0.447337806224823,
+      "learning_rate": 0.00019745755372479578,
+      "loss": 0.8464577794075012,
+      "step": 2080
+    },
+    {
+      "epoch": 0.8784810126582279,
+      "grad_norm": 0.37877580523490906,
+      "learning_rate": 0.0001974467810914138,
+      "loss": 0.7383086681365967,
+      "step": 2082
+    },
+    {
+      "epoch": 0.8793248945147679,
+      "grad_norm": 0.3914664089679718,
+      "learning_rate": 0.0001974359859789309,
+      "loss": 0.729221522808075,
+      "step": 2084
+    },
+    {
+      "epoch": 0.880168776371308,
+      "grad_norm": 0.39767366647720337,
+      "learning_rate": 0.00019742516838983733,
+      "loss": 0.692332923412323,
+      "step": 2086
+    },
+    {
+      "epoch": 0.8810126582278481,
+      "grad_norm": 0.4396754205226898,
+      "learning_rate": 0.0001974143283266285,
+      "loss": 0.8540068864822388,
+      "step": 2088
+    },
+    {
+      "epoch": 0.8818565400843882,
+      "grad_norm": 0.4968172609806061,
+      "learning_rate": 0.00019740346579180502,
+      "loss": 0.8570998907089233,
+      "step": 2090
+    },
+    {
+      "epoch": 0.8827004219409282,
+      "grad_norm": 0.41686809062957764,
+      "learning_rate": 0.00019739258078787264,
+      "loss": 0.784139096736908,
+      "step": 2092
+    },
+    {
+      "epoch": 0.8835443037974684,
+      "grad_norm": 0.38788047432899475,
+      "learning_rate": 0.0001973816733173423,
+      "loss": 0.8521397113800049,
+      "step": 2094
+    },
+    {
+      "epoch": 0.8843881856540085,
+      "grad_norm": 0.38369178771972656,
+      "learning_rate": 0.00019737074338273023,
+      "loss": 0.8457935452461243,
+      "step": 2096
+    },
+    {
+      "epoch": 0.8852320675105485,
+      "grad_norm": 0.38869398832321167,
+      "learning_rate": 0.0001973597909865577,
+      "loss": 0.7425959706306458,
+      "step": 2098
+    },
+    {
+      "epoch": 0.8860759493670886,
+      "grad_norm": 0.4067917466163635,
+      "learning_rate": 0.00019734881613135122,
+      "loss": 0.7742393016815186,
+      "step": 2100
+    },
+    {
+      "epoch": 0.8860759493670886,
+      "eval_loss": 0.8401098847389221,
+      "eval_runtime": 669.1149,
+      "eval_samples_per_second": 3.149,
+      "eval_steps_per_second": 3.149,
+      "step": 2100
+    },
+    {
+      "epoch": 0.8869198312236287,
+      "grad_norm": 0.4282098412513733,
+      "learning_rate": 0.00019733781881964242,
+      "loss": 0.9099814295768738,
+      "step": 2102
+    },
+    {
+      "epoch": 0.8877637130801688,
+      "grad_norm": 0.33882173895835876,
+      "learning_rate": 0.00019732679905396826,
+      "loss": 0.7458856105804443,
+      "step": 2104
+    },
+    {
+      "epoch": 0.8886075949367088,
+      "grad_norm": 0.43959516286849976,
+      "learning_rate": 0.00019731575683687078,
+      "loss": 0.7829693555831909,
+      "step": 2106
+    },
+    {
+      "epoch": 0.889451476793249,
+      "grad_norm": 0.40046483278274536,
+      "learning_rate": 0.00019730469217089716,
+      "loss": 0.7898997068405151,
+      "step": 2108
+    },
+    {
+      "epoch": 0.890295358649789,
+      "grad_norm": 0.40572553873062134,
+      "learning_rate": 0.00019729360505859984,
+      "loss": 0.7874186635017395,
+      "step": 2110
+    },
+    {
+      "epoch": 0.8911392405063291,
+      "grad_norm": 0.42086347937583923,
+      "learning_rate": 0.0001972824955025364,
+      "loss": 0.8499898910522461,
+      "step": 2112
+    },
+    {
+      "epoch": 0.8919831223628693,
+      "grad_norm": 0.38845229148864746,
+      "learning_rate": 0.00019727136350526956,
+      "loss": 0.7940409779548645,
+      "step": 2114
+    },
+    {
+      "epoch": 0.8928270042194093,
+      "grad_norm": 0.41575267910957336,
+      "learning_rate": 0.00019726020906936728,
+      "loss": 0.8279082775115967,
+      "step": 2116
+    },
+    {
+      "epoch": 0.8936708860759494,
+      "grad_norm": 0.37615057826042175,
+      "learning_rate": 0.0001972490321974027,
+      "loss": 0.8217757940292358,
+      "step": 2118
+    },
+    {
+      "epoch": 0.8945147679324894,
+      "grad_norm": 0.9318385124206543,
+      "learning_rate": 0.00019723783289195415,
+      "loss": 0.7637750506401062,
+      "step": 2120
+    },
+    {
+      "epoch": 0.8953586497890296,
+      "grad_norm": 0.41948962211608887,
+      "learning_rate": 0.000197226611155605,
+      "loss": 0.8702366948127747,
+      "step": 2122
+    },
+    {
+      "epoch": 0.8962025316455696,
+      "grad_norm": 0.4718417525291443,
+      "learning_rate": 0.00019721536699094392,
+      "loss": 0.8146984577178955,
+      "step": 2124
+    },
+    {
+      "epoch": 0.8970464135021097,
+      "grad_norm": 0.4470657408237457,
+      "learning_rate": 0.00019720410040056474,
+      "loss": 0.8646075129508972,
+      "step": 2126
+    },
+    {
+      "epoch": 0.8978902953586498,
+      "grad_norm": 0.4176540970802307,
+      "learning_rate": 0.00019719281138706643,
+      "loss": 0.7925992012023926,
+      "step": 2128
+    },
+    {
+      "epoch": 0.8987341772151899,
+      "grad_norm": 0.4854719042778015,
+      "learning_rate": 0.00019718149995305315,
+      "loss": 0.8142143487930298,
+      "step": 2130
+    },
+    {
+      "epoch": 0.8995780590717299,
+      "grad_norm": 0.4270440340042114,
+      "learning_rate": 0.00019717016610113427,
+      "loss": 0.8050890564918518,
+      "step": 2132
+    },
+    {
+      "epoch": 0.90042194092827,
+      "grad_norm": 0.37710970640182495,
+      "learning_rate": 0.00019715880983392422,
+      "loss": 0.7888354659080505,
+      "step": 2134
+    },
+    {
+      "epoch": 0.9012658227848102,
+      "grad_norm": 0.4181252717971802,
+      "learning_rate": 0.00019714743115404273,
+      "loss": 0.7348225712776184,
+      "step": 2136
+    },
+    {
+      "epoch": 0.9021097046413502,
+      "grad_norm": 0.4475359618663788,
+      "learning_rate": 0.0001971360300641146,
+      "loss": 0.7729623913764954,
+      "step": 2138
+    },
+    {
+      "epoch": 0.9029535864978903,
+      "grad_norm": 0.4052332937717438,
+      "learning_rate": 0.00019712460656676992,
+      "loss": 0.8066384792327881,
+      "step": 2140
+    },
+    {
+      "epoch": 0.9037974683544304,
+      "grad_norm": 0.4567393362522125,
+      "learning_rate": 0.0001971131606646438,
+      "loss": 0.8053154349327087,
+      "step": 2142
+    },
+    {
+      "epoch": 0.9046413502109705,
+      "grad_norm": 0.4304409325122833,
+      "learning_rate": 0.00019710169236037656,
+      "loss": 0.7242560386657715,
+      "step": 2144
+    },
+    {
+      "epoch": 0.9054852320675105,
+      "grad_norm": 0.4785008132457733,
+      "learning_rate": 0.0001970902016566138,
+      "loss": 0.8475301265716553,
+      "step": 2146
+    },
+    {
+      "epoch": 0.9063291139240506,
+      "grad_norm": 0.45226508378982544,
+      "learning_rate": 0.00019707868855600618,
+      "loss": 0.928544282913208,
+      "step": 2148
+    },
+    {
+      "epoch": 0.9071729957805907,
+      "grad_norm": 0.3794468641281128,
+      "learning_rate": 0.00019706715306120956,
+      "loss": 0.8135722875595093,
+      "step": 2150
+    },
+    {
+      "epoch": 0.9080168776371308,
+      "grad_norm": 0.4005141258239746,
+      "learning_rate": 0.0001970555951748849,
+      "loss": 0.7808290719985962,
+      "step": 2152
+    },
+    {
+      "epoch": 0.9088607594936708,
+      "grad_norm": 0.37929919362068176,
+      "learning_rate": 0.00019704401489969842,
+      "loss": 0.8292917013168335,
+      "step": 2154
+    },
+    {
+      "epoch": 0.909704641350211,
+      "grad_norm": 0.4458678066730499,
+      "learning_rate": 0.0001970324122383215,
+      "loss": 0.772884726524353,
+      "step": 2156
+    },
+    {
+      "epoch": 0.9105485232067511,
+      "grad_norm": 0.41907885670661926,
+      "learning_rate": 0.0001970207871934306,
+      "loss": 0.8204470872879028,
+      "step": 2158
+    },
+    {
+      "epoch": 0.9113924050632911,
+      "grad_norm": 0.40610212087631226,
+      "learning_rate": 0.00019700913976770741,
+      "loss": 0.8361451625823975,
+      "step": 2160
+    },
+    {
+      "epoch": 0.9122362869198313,
+      "grad_norm": 0.4213273227214813,
+      "learning_rate": 0.00019699746996383878,
+      "loss": 0.8456075191497803,
+      "step": 2162
+    },
+    {
+      "epoch": 0.9130801687763713,
+      "grad_norm": 0.49264612793922424,
+      "learning_rate": 0.0001969857777845167,
+      "loss": 0.8731603622436523,
+      "step": 2164
+    },
+    {
+      "epoch": 0.9139240506329114,
+      "grad_norm": 0.42002132534980774,
+      "learning_rate": 0.00019697406323243834,
+      "loss": 0.8149666786193848,
+      "step": 2166
+    },
+    {
+      "epoch": 0.9147679324894514,
+      "grad_norm": 0.37501421570777893,
+      "learning_rate": 0.000196962326310306,
+      "loss": 0.7508546710014343,
+      "step": 2168
+    },
+    {
+      "epoch": 0.9156118143459916,
+      "grad_norm": 0.3706159293651581,
+      "learning_rate": 0.00019695056702082717,
+      "loss": 0.830280601978302,
+      "step": 2170
+    },
+    {
+      "epoch": 0.9164556962025316,
+      "grad_norm": 0.3873467445373535,
+      "learning_rate": 0.00019693878536671452,
+      "loss": 0.7297167778015137,
+      "step": 2172
+    },
+    {
+      "epoch": 0.9172995780590717,
+      "grad_norm": 0.42876651883125305,
+      "learning_rate": 0.00019692698135068582,
+      "loss": 0.7924292087554932,
+      "step": 2174
+    },
+    {
+      "epoch": 0.9181434599156119,
+      "grad_norm": 0.4259159564971924,
+      "learning_rate": 0.00019691515497546406,
+      "loss": 0.8098247051239014,
+      "step": 2176
+    },
+    {
+      "epoch": 0.9189873417721519,
+      "grad_norm": 0.38152721524238586,
+      "learning_rate": 0.00019690330624377728,
+      "loss": 0.6789063215255737,
+      "step": 2178
+    },
+    {
+      "epoch": 0.919831223628692,
+      "grad_norm": 0.4135437607765198,
+      "learning_rate": 0.00019689143515835887,
+      "loss": 0.8582620620727539,
+      "step": 2180
+    },
+    {
+      "epoch": 0.920675105485232,
+      "grad_norm": 0.3818798065185547,
+      "learning_rate": 0.0001968795417219472,
+      "loss": 0.8059743642807007,
+      "step": 2182
+    },
+    {
+      "epoch": 0.9215189873417722,
+      "grad_norm": 0.39021286368370056,
+      "learning_rate": 0.00019686762593728582,
+      "loss": 0.7432259917259216,
+      "step": 2184
+    },
+    {
+      "epoch": 0.9223628691983122,
+      "grad_norm": 0.37864068150520325,
+      "learning_rate": 0.00019685568780712355,
+      "loss": 0.7561086416244507,
+      "step": 2186
+    },
+    {
+      "epoch": 0.9232067510548523,
+      "grad_norm": 0.450847864151001,
+      "learning_rate": 0.00019684372733421425,
+      "loss": 0.7552722692489624,
+      "step": 2188
+    },
+    {
+      "epoch": 0.9240506329113924,
+      "grad_norm": 0.37909507751464844,
+      "learning_rate": 0.00019683174452131695,
+      "loss": 0.6905336380004883,
+      "step": 2190
+    },
+    {
+      "epoch": 0.9248945147679325,
+      "grad_norm": 0.4439173638820648,
+      "learning_rate": 0.0001968197393711959,
+      "loss": 0.7710593938827515,
+      "step": 2192
+    },
+    {
+      "epoch": 0.9257383966244725,
+      "grad_norm": 0.3908376395702362,
+      "learning_rate": 0.00019680771188662044,
+      "loss": 0.8212814331054688,
+      "step": 2194
+    },
+    {
+      "epoch": 0.9265822784810127,
+      "grad_norm": 0.3528299033641815,
+      "learning_rate": 0.00019679566207036507,
+      "loss": 0.751355767250061,
+      "step": 2196
+    },
+    {
+      "epoch": 0.9274261603375528,
+      "grad_norm": 0.4351397156715393,
+      "learning_rate": 0.00019678358992520945,
+      "loss": 0.8204057812690735,
+      "step": 2198
+    },
+    {
+      "epoch": 0.9282700421940928,
+      "grad_norm": 0.45566943287849426,
+      "learning_rate": 0.0001967714954539384,
+      "loss": 0.763841450214386,
+      "step": 2200
+    },
+    {
+      "epoch": 0.9282700421940928,
+      "eval_loss": 0.8336610198020935,
+      "eval_runtime": 674.5134,
+      "eval_samples_per_second": 3.124,
+      "eval_steps_per_second": 3.124,
+      "step": 2200
+    },
+    {
+      "epoch": 0.9291139240506329,
+      "grad_norm": 0.4516221284866333,
+      "learning_rate": 0.00019675937865934188,
+      "loss": 0.861711323261261,
+      "step": 2202
+    },
+    {
+      "epoch": 0.929957805907173,
+      "grad_norm": 0.43421855568885803,
+      "learning_rate": 0.00019674723954421502,
+      "loss": 0.8062718510627747,
+      "step": 2204
+    },
+    {
+      "epoch": 0.9308016877637131,
+      "grad_norm": 0.42442476749420166,
+      "learning_rate": 0.00019673507811135803,
+      "loss": 0.8069695234298706,
+      "step": 2206
+    },
+    {
+      "epoch": 0.9316455696202531,
+      "grad_norm": 0.3976045250892639,
+      "learning_rate": 0.00019672289436357637,
+      "loss": 0.7336108684539795,
+      "step": 2208
+    },
+    {
+      "epoch": 0.9324894514767933,
+      "grad_norm": 0.4131270945072174,
+      "learning_rate": 0.00019671068830368058,
+      "loss": 0.7559890151023865,
+      "step": 2210
+    },
+    {
+      "epoch": 0.9333333333333333,
+      "grad_norm": 0.4599294364452362,
+      "learning_rate": 0.00019669845993448634,
+      "loss": 0.8447075486183167,
+      "step": 2212
+    },
+    {
+      "epoch": 0.9341772151898734,
+      "grad_norm": 0.4291204810142517,
+      "learning_rate": 0.00019668620925881455,
+      "loss": 0.8329233527183533,
+      "step": 2214
+    },
+    {
+      "epoch": 0.9350210970464135,
+      "grad_norm": 0.4226539731025696,
+      "learning_rate": 0.00019667393627949116,
+      "loss": 0.749232292175293,
+      "step": 2216
+    },
+    {
+      "epoch": 0.9358649789029536,
+      "grad_norm": 0.4159580171108246,
+      "learning_rate": 0.00019666164099934732,
+      "loss": 0.7470011115074158,
+      "step": 2218
+    },
+    {
+      "epoch": 0.9367088607594937,
+      "grad_norm": 0.4160766303539276,
+      "learning_rate": 0.00019664932342121935,
+      "loss": 0.8094202280044556,
+      "step": 2220
+    },
+    {
+      "epoch": 0.9375527426160337,
+      "grad_norm": 0.45284298062324524,
+      "learning_rate": 0.00019663698354794863,
+      "loss": 0.8398169875144958,
+      "step": 2222
+    },
+    {
+      "epoch": 0.9383966244725739,
+      "grad_norm": 0.3960540294647217,
+      "learning_rate": 0.00019662462138238178,
+      "loss": 0.7652443051338196,
+      "step": 2224
+    },
+    {
+      "epoch": 0.9392405063291139,
+      "grad_norm": 0.4269275367259979,
+      "learning_rate": 0.00019661223692737047,
+      "loss": 0.7909489870071411,
+      "step": 2226
+    },
+    {
+      "epoch": 0.940084388185654,
+      "grad_norm": 0.38187873363494873,
+      "learning_rate": 0.00019659983018577157,
+      "loss": 0.7821935415267944,
+      "step": 2228
+    },
+    {
+      "epoch": 0.9409282700421941,
+      "grad_norm": 0.45898866653442383,
+      "learning_rate": 0.0001965874011604471,
+      "loss": 0.7792721390724182,
+      "step": 2230
+    },
+    {
+      "epoch": 0.9417721518987342,
+      "grad_norm": 0.4740699827671051,
+      "learning_rate": 0.00019657494985426418,
+      "loss": 0.885725200176239,
+      "step": 2232
+    },
+    {
+      "epoch": 0.9426160337552743,
+      "grad_norm": 0.3912813365459442,
+      "learning_rate": 0.00019656247627009506,
+      "loss": 0.7898101806640625,
+      "step": 2234
+    },
+    {
+      "epoch": 0.9434599156118143,
+      "grad_norm": 0.4552418291568756,
+      "learning_rate": 0.0001965499804108172,
+      "loss": 0.8646330833435059,
+      "step": 2236
+    },
+    {
+      "epoch": 0.9443037974683545,
+      "grad_norm": 0.4178548753261566,
+      "learning_rate": 0.0001965374622793131,
+      "loss": 0.8068296313285828,
+      "step": 2238
+    },
+    {
+      "epoch": 0.9451476793248945,
+      "grad_norm": 0.38649076223373413,
+      "learning_rate": 0.00019652492187847057,
+      "loss": 0.7953271865844727,
+      "step": 2240
+    },
+    {
+      "epoch": 0.9459915611814346,
+      "grad_norm": 0.39785075187683105,
+      "learning_rate": 0.0001965123592111823,
+      "loss": 0.8423402309417725,
+      "step": 2242
+    },
+    {
+      "epoch": 0.9468354430379747,
+      "grad_norm": 0.47455263137817383,
+      "learning_rate": 0.00019649977428034632,
+      "loss": 0.8883589506149292,
+      "step": 2244
+    },
+    {
+      "epoch": 0.9476793248945148,
+      "grad_norm": 0.3866841197013855,
+      "learning_rate": 0.0001964871670888657,
+      "loss": 0.7232863903045654,
+      "step": 2246
+    },
+    {
+      "epoch": 0.9485232067510548,
+      "grad_norm": 0.3950609564781189,
+      "learning_rate": 0.00019647453763964876,
+      "loss": 0.7265347838401794,
+      "step": 2248
+    },
+    {
+      "epoch": 0.9493670886075949,
+      "grad_norm": 0.4315829575061798,
+      "learning_rate": 0.0001964618859356088,
+      "loss": 0.8000131249427795,
+      "step": 2250
+    },
+    {
+      "epoch": 0.950210970464135,
+      "grad_norm": 0.403984934091568,
+      "learning_rate": 0.00019644921197966433,
+      "loss": 0.823086678981781,
+      "step": 2252
+    },
+    {
+      "epoch": 0.9510548523206751,
+      "grad_norm": 0.4311080574989319,
+      "learning_rate": 0.00019643651577473897,
+      "loss": 0.8457553386688232,
+      "step": 2254
+    },
+    {
+      "epoch": 0.9518987341772152,
+      "grad_norm": 0.3884856104850769,
+      "learning_rate": 0.00019642379732376158,
+      "loss": 0.7866189479827881,
+      "step": 2256
+    },
+    {
+      "epoch": 0.9527426160337553,
+      "grad_norm": 0.3384067714214325,
+      "learning_rate": 0.00019641105662966595,
+      "loss": 0.7095813751220703,
+      "step": 2258
+    },
+    {
+      "epoch": 0.9535864978902954,
+      "grad_norm": 0.40426746010780334,
+      "learning_rate": 0.00019639829369539117,
+      "loss": 0.8271632194519043,
+      "step": 2260
+    },
+    {
+      "epoch": 0.9544303797468354,
+      "grad_norm": 0.45495131611824036,
+      "learning_rate": 0.00019638550852388144,
+      "loss": 0.8204861283302307,
+      "step": 2262
+    },
+    {
+      "epoch": 0.9552742616033755,
+      "grad_norm": 0.47201067209243774,
+      "learning_rate": 0.00019637270111808598,
+      "loss": 0.884763777256012,
+      "step": 2264
+    },
+    {
+      "epoch": 0.9561181434599156,
+      "grad_norm": 0.3959396779537201,
+      "learning_rate": 0.0001963598714809592,
+      "loss": 0.7630926370620728,
+      "step": 2266
+    },
+    {
+      "epoch": 0.9569620253164557,
+      "grad_norm": 0.4197136461734772,
+      "learning_rate": 0.00019634701961546077,
+      "loss": 0.797870397567749,
+      "step": 2268
+    },
+    {
+      "epoch": 0.9578059071729957,
+      "grad_norm": 0.381290078163147,
+      "learning_rate": 0.00019633414552455526,
+      "loss": 0.7755846977233887,
+      "step": 2270
+    },
+    {
+      "epoch": 0.9586497890295359,
+      "grad_norm": 0.3868124783039093,
+      "learning_rate": 0.00019632124921121255,
+      "loss": 0.7553264498710632,
+      "step": 2272
+    },
+    {
+      "epoch": 0.959493670886076,
+      "grad_norm": 0.4350631833076477,
+      "learning_rate": 0.00019630833067840747,
+      "loss": 0.7792160511016846,
+      "step": 2274
+    },
+    {
+      "epoch": 0.960337552742616,
+      "grad_norm": 0.4717157781124115,
+      "learning_rate": 0.00019629538992912016,
+      "loss": 0.878220796585083,
+      "step": 2276
+    },
+    {
+      "epoch": 0.9611814345991562,
+      "grad_norm": 0.45545122027397156,
+      "learning_rate": 0.00019628242696633584,
+      "loss": 0.8891014456748962,
+      "step": 2278
+    },
+    {
+      "epoch": 0.9620253164556962,
+      "grad_norm": 0.4421025514602661,
+      "learning_rate": 0.00019626944179304467,
+      "loss": 0.8024723529815674,
+      "step": 2280
+    },
+    {
+      "epoch": 0.9628691983122363,
+      "grad_norm": 0.40385812520980835,
+      "learning_rate": 0.00019625643441224222,
+      "loss": 0.8036280274391174,
+      "step": 2282
+    },
+    {
+      "epoch": 0.9637130801687763,
+      "grad_norm": 0.3866552412509918,
+      "learning_rate": 0.00019624340482692898,
+      "loss": 0.8072395324707031,
+      "step": 2284
+    },
+    {
+      "epoch": 0.9645569620253165,
+      "grad_norm": 0.4113233685493469,
+      "learning_rate": 0.0001962303530401106,
+      "loss": 0.8046114444732666,
+      "step": 2286
+    },
+    {
+      "epoch": 0.9654008438818565,
+      "grad_norm": 0.40720227360725403,
+      "learning_rate": 0.00019621727905479798,
+      "loss": 0.8434037566184998,
+      "step": 2288
+    },
+    {
+      "epoch": 0.9662447257383966,
+      "grad_norm": 0.35759273171424866,
+      "learning_rate": 0.00019620418287400693,
+      "loss": 0.7832419276237488,
+      "step": 2290
+    },
+    {
+      "epoch": 0.9670886075949368,
+      "grad_norm": 0.4170442223548889,
+      "learning_rate": 0.0001961910645007585,
+      "loss": 0.7975693345069885,
+      "step": 2292
+    },
+    {
+      "epoch": 0.9679324894514768,
+      "grad_norm": 0.4126700162887573,
+      "learning_rate": 0.00019617792393807895,
+      "loss": 0.8554494976997375,
+      "step": 2294
+    },
+    {
+      "epoch": 0.9687763713080169,
+      "grad_norm": 0.4231695830821991,
+      "learning_rate": 0.00019616476118899942,
+      "loss": 0.824539065361023,
+      "step": 2296
+    },
+    {
+      "epoch": 0.9696202531645569,
+      "grad_norm": 0.3556209206581116,
+      "learning_rate": 0.0001961515762565564,
+      "loss": 0.7008040547370911,
+      "step": 2298
+    },
+    {
+      "epoch": 0.9704641350210971,
+      "grad_norm": 0.4370400309562683,
+      "learning_rate": 0.00019613836914379132,
+      "loss": 0.8192558288574219,
+      "step": 2300
+    },
+    {
+      "epoch": 0.9704641350210971,
+      "eval_loss": 0.8281980156898499,
+      "eval_runtime": 688.6136,
+      "eval_samples_per_second": 3.06,
+      "eval_steps_per_second": 3.06,
+      "step": 2300
+    },
+    {
+      "epoch": 0.9713080168776371,
+      "grad_norm": 0.4207614064216614,
+      "learning_rate": 0.00019612513985375088,
+      "loss": 0.878928005695343,
+      "step": 2302
+    },
+    {
+      "epoch": 0.9721518987341772,
+      "grad_norm": 0.37607815861701965,
+      "learning_rate": 0.0001961118883894868,
+      "loss": 0.736182689666748,
+      "step": 2304
+    },
+    {
+      "epoch": 0.9729957805907173,
+      "grad_norm": 0.43311476707458496,
+      "learning_rate": 0.00019609861475405592,
+      "loss": 0.8403618335723877,
+      "step": 2306
+    },
+    {
+      "epoch": 0.9738396624472574,
+      "grad_norm": 0.4533176124095917,
+      "learning_rate": 0.00019608531895052021,
+      "loss": 0.8414564728736877,
+      "step": 2308
+    },
+    {
+      "epoch": 0.9746835443037974,
+      "grad_norm": 0.4025697410106659,
+      "learning_rate": 0.00019607200098194677,
+      "loss": 0.8105775117874146,
+      "step": 2310
+    },
+    {
+      "epoch": 0.9755274261603376,
+      "grad_norm": 0.37894198298454285,
+      "learning_rate": 0.00019605866085140784,
+      "loss": 0.8053821325302124,
+      "step": 2312
+    },
+    {
+      "epoch": 0.9763713080168777,
+      "grad_norm": 0.46265891194343567,
+      "learning_rate": 0.0001960452985619807,
+      "loss": 0.8092884421348572,
+      "step": 2314
+    },
+    {
+      "epoch": 0.9772151898734177,
+      "grad_norm": 0.3770054280757904,
+      "learning_rate": 0.0001960319141167477,
+      "loss": 0.7456095218658447,
+      "step": 2316
+    },
+    {
+      "epoch": 0.9780590717299578,
+      "grad_norm": 0.44084450602531433,
+      "learning_rate": 0.0001960185075187965,
+      "loss": 0.7739883661270142,
+      "step": 2318
+    },
+    {
+      "epoch": 0.9789029535864979,
+      "grad_norm": 0.3880341351032257,
+      "learning_rate": 0.00019600507877121965,
+      "loss": 0.6961302757263184,
+      "step": 2320
+    },
+    {
+      "epoch": 0.979746835443038,
+      "grad_norm": 0.468037486076355,
+      "learning_rate": 0.00019599162787711495,
+      "loss": 0.8447984457015991,
+      "step": 2322
+    },
+    {
+      "epoch": 0.980590717299578,
+      "grad_norm": 0.4123631417751312,
+      "learning_rate": 0.00019597815483958527,
+      "loss": 0.8155615329742432,
+      "step": 2324
+    },
+    {
+      "epoch": 0.9814345991561182,
+      "grad_norm": 0.3827875554561615,
+      "learning_rate": 0.00019596465966173853,
+      "loss": 0.7990952730178833,
+      "step": 2326
+    },
+    {
+      "epoch": 0.9822784810126582,
+      "grad_norm": 0.4644222557544708,
+      "learning_rate": 0.00019595114234668788,
+      "loss": 0.8899750709533691,
+      "step": 2328
+    },
+    {
+      "epoch": 0.9831223628691983,
+      "grad_norm": 0.43699249625205994,
+      "learning_rate": 0.00019593760289755144,
+      "loss": 0.7863160371780396,
+      "step": 2330
+    },
+    {
+      "epoch": 0.9839662447257383,
+      "grad_norm": 0.40713244676589966,
+      "learning_rate": 0.00019592404131745254,
+      "loss": 0.7960637807846069,
+      "step": 2332
+    },
+    {
+      "epoch": 0.9848101265822785,
+      "grad_norm": 0.37364068627357483,
+      "learning_rate": 0.00019591045760951957,
+      "loss": 0.714144229888916,
+      "step": 2334
+    },
+    {
+      "epoch": 0.9856540084388186,
+      "grad_norm": 0.4094736576080322,
+      "learning_rate": 0.00019589685177688599,
+      "loss": 0.7311791777610779,
+      "step": 2336
+    },
+    {
+      "epoch": 0.9864978902953586,
+      "grad_norm": 0.4275180697441101,
+      "learning_rate": 0.0001958832238226905,
+      "loss": 0.8191342353820801,
+      "step": 2338
+    },
+    {
+      "epoch": 0.9873417721518988,
+      "grad_norm": 0.4627651274204254,
+      "learning_rate": 0.00019586957375007668,
+      "loss": 0.7476106286048889,
+      "step": 2340
+    },
+    {
+      "epoch": 0.9881856540084388,
+      "grad_norm": 0.44676414132118225,
+      "learning_rate": 0.00019585590156219346,
+      "loss": 0.8257967233657837,
+      "step": 2342
+    },
+    {
+      "epoch": 0.9890295358649789,
+      "grad_norm": 0.40667134523391724,
+      "learning_rate": 0.00019584220726219466,
+      "loss": 0.7922079563140869,
+      "step": 2344
+    },
+    {
+      "epoch": 0.9898734177215189,
+      "grad_norm": 0.40865346789360046,
+      "learning_rate": 0.00019582849085323934,
+      "loss": 0.7892434597015381,
+      "step": 2346
+    },
+    {
+      "epoch": 0.9907172995780591,
+      "grad_norm": 0.43309709429740906,
+      "learning_rate": 0.00019581475233849163,
+      "loss": 0.8766644597053528,
+      "step": 2348
+    },
+    {
+      "epoch": 0.9915611814345991,
+      "grad_norm": 0.41558781266212463,
+      "learning_rate": 0.00019580099172112069,
+      "loss": 0.9183312058448792,
+      "step": 2350
+    },
+    {
+      "epoch": 0.9924050632911392,
+      "grad_norm": 0.432748019695282,
+      "learning_rate": 0.00019578720900430082,
+      "loss": 0.7905668020248413,
+      "step": 2352
+    },
+    {
+      "epoch": 0.9932489451476794,
+      "grad_norm": 0.4116985499858856,
+      "learning_rate": 0.0001957734041912115,
+      "loss": 0.8279712200164795,
+      "step": 2354
+    },
+    {
+      "epoch": 0.9940928270042194,
+      "grad_norm": 0.3927275240421295,
+      "learning_rate": 0.00019575957728503718,
+      "loss": 0.7451362609863281,
+      "step": 2356
+    },
+    {
+      "epoch": 0.9949367088607595,
+      "grad_norm": 0.3389509320259094,
+      "learning_rate": 0.0001957457282889675,
+      "loss": 0.6769390106201172,
+      "step": 2358
+    },
+    {
+      "epoch": 0.9957805907172996,
+      "grad_norm": 0.4220774471759796,
+      "learning_rate": 0.00019573185720619712,
+      "loss": 0.7590897083282471,
+      "step": 2360
+    },
+    {
+      "epoch": 0.9966244725738397,
+      "grad_norm": 0.4299897849559784,
+      "learning_rate": 0.00019571796403992583,
+      "loss": 0.8301284313201904,
+      "step": 2362
+    },
+    {
+      "epoch": 0.9974683544303797,
+      "grad_norm": 0.4031789302825928,
+      "learning_rate": 0.00019570404879335856,
+      "loss": 0.8019471764564514,
+      "step": 2364
+    },
+    {
+      "epoch": 0.9983122362869198,
+      "grad_norm": 0.3957567811012268,
+      "learning_rate": 0.00019569011146970526,
+      "loss": 0.7427027821540833,
+      "step": 2366
+    },
+    {
+      "epoch": 0.99915611814346,
+      "grad_norm": 0.41741347312927246,
+      "learning_rate": 0.000195676152072181,
+      "loss": 0.8221617341041565,
+      "step": 2368
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.47121551632881165,
+      "learning_rate": 0.00019566217060400595,
+      "loss": 0.8143455982208252,
+      "step": 2370
+    },
+    {
+      "epoch": 1.0008438818565402,
+      "grad_norm": 0.3854716122150421,
+      "learning_rate": 0.0001956481670684054,
+      "loss": 0.6971694231033325,
+      "step": 2372
+    },
+    {
+      "epoch": 1.00168776371308,
+      "grad_norm": 0.4029138386249542,
+      "learning_rate": 0.00019563414146860964,
+      "loss": 0.6719155311584473,
+      "step": 2374
+    },
+    {
+      "epoch": 1.0025316455696203,
+      "grad_norm": 0.40136656165122986,
+      "learning_rate": 0.00019562009380785412,
+      "loss": 0.7161872386932373,
+      "step": 2376
+    },
+    {
+      "epoch": 1.0033755274261604,
+      "grad_norm": 0.40685850381851196,
+      "learning_rate": 0.0001956060240893794,
+      "loss": 0.71699059009552,
+      "step": 2378
+    },
+    {
+      "epoch": 1.0042194092827004,
+      "grad_norm": 0.46153366565704346,
+      "learning_rate": 0.00019559193231643104,
+      "loss": 0.8030059337615967,
+      "step": 2380
+    },
+    {
+      "epoch": 1.0050632911392405,
+      "grad_norm": 0.45048823952674866,
+      "learning_rate": 0.0001955778184922598,
+      "loss": 0.7666479349136353,
+      "step": 2382
+    },
+    {
+      "epoch": 1.0059071729957807,
+      "grad_norm": 0.4260413348674774,
+      "learning_rate": 0.00019556368262012142,
+      "loss": 0.7481234073638916,
+      "step": 2384
+    },
+    {
+      "epoch": 1.0067510548523206,
+      "grad_norm": 0.4819575548171997,
+      "learning_rate": 0.0001955495247032768,
+      "loss": 0.7943820953369141,
+      "step": 2386
+    },
+    {
+      "epoch": 1.0075949367088608,
+      "grad_norm": 0.45789605379104614,
+      "learning_rate": 0.0001955353447449919,
+      "loss": 0.6401634812355042,
+      "step": 2388
+    },
+    {
+      "epoch": 1.0084388185654007,
+      "grad_norm": 0.37019097805023193,
+      "learning_rate": 0.00019552114274853778,
+      "loss": 0.7058886885643005,
+      "step": 2390
+    },
+    {
+      "epoch": 1.009282700421941,
+      "grad_norm": 0.44416263699531555,
+      "learning_rate": 0.0001955069187171905,
+      "loss": 0.7631242871284485,
+      "step": 2392
+    },
+    {
+      "epoch": 1.010126582278481,
+      "grad_norm": 0.4478379786014557,
+      "learning_rate": 0.00019549267265423138,
+      "loss": 0.7873892784118652,
+      "step": 2394
+    },
+    {
+      "epoch": 1.010970464135021,
+      "grad_norm": 0.4596130847930908,
+      "learning_rate": 0.00019547840456294658,
+      "loss": 0.7882059216499329,
+      "step": 2396
+    },
+    {
+      "epoch": 1.0118143459915612,
+      "grad_norm": 0.4062999188899994,
+      "learning_rate": 0.0001954641144466276,
+      "loss": 0.6358574628829956,
+      "step": 2398
+    },
+    {
+      "epoch": 1.0126582278481013,
+      "grad_norm": 0.47627049684524536,
+      "learning_rate": 0.0001954498023085708,
+      "loss": 0.7803886532783508,
+      "step": 2400
+    },
+    {
+      "epoch": 1.0126582278481013,
+      "eval_loss": 0.8250564932823181,
+      "eval_runtime": 691.5833,
+      "eval_samples_per_second": 3.047,
+      "eval_steps_per_second": 3.047,
+      "step": 2400
+    },
+    {
+      "epoch": 1.0135021097046413,
+      "grad_norm": 0.40690696239471436,
+      "learning_rate": 0.00019543546815207777,
+      "loss": 0.7394700050354004,
+      "step": 2402
+    },
+    {
+      "epoch": 1.0143459915611814,
+      "grad_norm": 0.45543980598449707,
+      "learning_rate": 0.0001954211119804551,
+      "loss": 0.7096555829048157,
+      "step": 2404
+    },
+    {
+      "epoch": 1.0151898734177216,
+      "grad_norm": 0.4367314279079437,
+      "learning_rate": 0.00019540673379701454,
+      "loss": 0.7961843013763428,
+      "step": 2406
+    },
+    {
+      "epoch": 1.0160337552742615,
+      "grad_norm": 0.4458340108394623,
+      "learning_rate": 0.00019539233360507278,
+      "loss": 0.743843138217926,
+      "step": 2408
+    },
+    {
+      "epoch": 1.0168776371308017,
+      "grad_norm": 0.3741893470287323,
+      "learning_rate": 0.0001953779114079517,
+      "loss": 0.7273756265640259,
+      "step": 2410
+    },
+    {
+      "epoch": 1.0177215189873419,
+      "grad_norm": 0.4340079426765442,
+      "learning_rate": 0.00019536346720897823,
+      "loss": 0.7922506928443909,
+      "step": 2412
+    },
+    {
+      "epoch": 1.0185654008438818,
+      "grad_norm": 0.4518018662929535,
+      "learning_rate": 0.00019534900101148437,
+      "loss": 0.6492037773132324,
+      "step": 2414
+    },
+    {
+      "epoch": 1.019409282700422,
+      "grad_norm": 0.4160727560520172,
+      "learning_rate": 0.00019533451281880716,
+      "loss": 0.7422950863838196,
+      "step": 2416
+    },
+    {
+      "epoch": 1.0202531645569621,
+      "grad_norm": 0.35953211784362793,
+      "learning_rate": 0.00019532000263428883,
+      "loss": 0.6489207148551941,
+      "step": 2418
+    },
+    {
+      "epoch": 1.021097046413502,
+      "grad_norm": 0.38774433732032776,
+      "learning_rate": 0.0001953054704612765,
+      "loss": 0.7449417114257812,
+      "step": 2420
+    },
+    {
+      "epoch": 1.0219409282700422,
+      "grad_norm": 0.48185858130455017,
+      "learning_rate": 0.0001952909163031225,
+      "loss": 0.8481494188308716,
+      "step": 2422
+    },
+    {
+      "epoch": 1.0227848101265822,
+      "grad_norm": 0.4098491072654724,
+      "learning_rate": 0.00019527634016318425,
+      "loss": 0.7442291378974915,
+      "step": 2424
+    },
+    {
+      "epoch": 1.0236286919831223,
+      "grad_norm": 0.41284480690956116,
+      "learning_rate": 0.00019526174204482412,
+      "loss": 0.7564480900764465,
+      "step": 2426
+    },
+    {
+      "epoch": 1.0244725738396625,
+      "grad_norm": 0.44948479533195496,
+      "learning_rate": 0.00019524712195140965,
+      "loss": 0.7234854102134705,
+      "step": 2428
+    },
+    {
+      "epoch": 1.0253164556962024,
+      "grad_norm": 0.39939233660697937,
+      "learning_rate": 0.0001952324798863134,
+      "loss": 0.7638729810714722,
+      "step": 2430
+    },
+    {
+      "epoch": 1.0261603375527426,
+      "grad_norm": 0.40807482600212097,
+      "learning_rate": 0.00019521781585291298,
+      "loss": 0.6922630071640015,
+      "step": 2432
+    },
+    {
+      "epoch": 1.0270042194092828,
+      "grad_norm": 0.45735791325569153,
+      "learning_rate": 0.00019520312985459116,
+      "loss": 0.762996256351471,
+      "step": 2434
+    },
+    {
+      "epoch": 1.0278481012658227,
+      "grad_norm": 0.4280409812927246,
+      "learning_rate": 0.0001951884218947357,
+      "loss": 0.7811596989631653,
+      "step": 2436
+    },
+    {
+      "epoch": 1.0286919831223629,
+      "grad_norm": 0.41957560181617737,
+      "learning_rate": 0.00019517369197673943,
+      "loss": 0.7278305292129517,
+      "step": 2438
+    },
+    {
+      "epoch": 1.029535864978903,
+      "grad_norm": 0.4832586646080017,
+      "learning_rate": 0.00019515894010400028,
+      "loss": 0.807456910610199,
+      "step": 2440
+    },
+    {
+      "epoch": 1.030379746835443,
+      "grad_norm": 0.46448788046836853,
+      "learning_rate": 0.00019514416627992122,
+      "loss": 0.7421302795410156,
+      "step": 2442
+    },
+    {
+      "epoch": 1.0312236286919831,
+      "grad_norm": 0.41362619400024414,
+      "learning_rate": 0.00019512937050791024,
+      "loss": 0.6892213225364685,
+      "step": 2444
+    },
+    {
+      "epoch": 1.0320675105485233,
+      "grad_norm": 0.4158109128475189,
+      "learning_rate": 0.0001951145527913805,
+      "loss": 0.7305331826210022,
+      "step": 2446
+    },
+    {
+      "epoch": 1.0329113924050632,
+      "grad_norm": 0.4725087881088257,
+      "learning_rate": 0.00019509971313375011,
+      "loss": 0.7682753205299377,
+      "step": 2448
+    },
+    {
+      "epoch": 1.0337552742616034,
+      "grad_norm": 0.4341438114643097,
+      "learning_rate": 0.00019508485153844237,
+      "loss": 0.747041642665863,
+      "step": 2450
+    },
+    {
+      "epoch": 1.0345991561181433,
+      "grad_norm": 0.44515377283096313,
+      "learning_rate": 0.00019506996800888548,
+      "loss": 0.7312062382698059,
+      "step": 2452
+    },
+    {
+      "epoch": 1.0354430379746835,
+      "grad_norm": 0.5313915014266968,
+      "learning_rate": 0.00019505506254851283,
+      "loss": 0.7861107587814331,
+      "step": 2454
+    },
+    {
+      "epoch": 1.0362869198312237,
+      "grad_norm": 0.6094246506690979,
+      "learning_rate": 0.00019504013516076285,
+      "loss": 0.8458115458488464,
+      "step": 2456
+    },
+    {
+      "epoch": 1.0371308016877636,
+      "grad_norm": 0.36419153213500977,
+      "learning_rate": 0.00019502518584907895,
+      "loss": 0.6091256141662598,
+      "step": 2458
+    },
+    {
+      "epoch": 1.0379746835443038,
+      "grad_norm": 0.399271696805954,
+      "learning_rate": 0.00019501021461690966,
+      "loss": 0.6703370809555054,
+      "step": 2460
+    },
+    {
+      "epoch": 1.038818565400844,
+      "grad_norm": 0.38560932874679565,
+      "learning_rate": 0.00019499522146770856,
+      "loss": 0.7220943570137024,
+      "step": 2462
+    },
+    {
+      "epoch": 1.0396624472573839,
+      "grad_norm": 0.44199803471565247,
+      "learning_rate": 0.00019498020640493428,
+      "loss": 0.7501406073570251,
+      "step": 2464
+    },
+    {
+      "epoch": 1.040506329113924,
+      "grad_norm": 0.47394421696662903,
+      "learning_rate": 0.00019496516943205054,
+      "loss": 0.7957925796508789,
+      "step": 2466
+    },
+    {
+      "epoch": 1.0413502109704642,
+      "grad_norm": 0.5274425745010376,
+      "learning_rate": 0.00019495011055252603,
+      "loss": 0.7964824438095093,
+      "step": 2468
+    },
+    {
+      "epoch": 1.0421940928270041,
+      "grad_norm": 0.44004881381988525,
+      "learning_rate": 0.0001949350297698346,
+      "loss": 0.7478582262992859,
+      "step": 2470
+    },
+    {
+      "epoch": 1.0430379746835443,
+      "grad_norm": 0.40662670135498047,
+      "learning_rate": 0.000194919927087455,
+      "loss": 0.738339900970459,
+      "step": 2472
+    },
+    {
+      "epoch": 1.0438818565400845,
+      "grad_norm": 0.43934181332588196,
+      "learning_rate": 0.00019490480250887123,
+      "loss": 0.817171037197113,
+      "step": 2474
+    },
+    {
+      "epoch": 1.0447257383966244,
+      "grad_norm": 0.4111221432685852,
+      "learning_rate": 0.00019488965603757223,
+      "loss": 0.7413169145584106,
+      "step": 2476
+    },
+    {
+      "epoch": 1.0455696202531646,
+      "grad_norm": 0.3770996630191803,
+      "learning_rate": 0.00019487448767705194,
+      "loss": 0.7893102169036865,
+      "step": 2478
+    },
+    {
+      "epoch": 1.0464135021097047,
+      "grad_norm": 0.4524461627006531,
+      "learning_rate": 0.00019485929743080945,
+      "loss": 0.7037841081619263,
+      "step": 2480
+    },
+    {
+      "epoch": 1.0472573839662447,
+      "grad_norm": 0.4232378900051117,
+      "learning_rate": 0.00019484408530234885,
+      "loss": 0.7456525564193726,
+      "step": 2482
+    },
+    {
+      "epoch": 1.0481012658227848,
+      "grad_norm": 0.3842810392379761,
+      "learning_rate": 0.00019482885129517929,
+      "loss": 0.7263808250427246,
+      "step": 2484
+    },
+    {
+      "epoch": 1.048945147679325,
+      "grad_norm": 0.43203726410865784,
+      "learning_rate": 0.00019481359541281495,
+      "loss": 0.7101474404335022,
+      "step": 2486
+    },
+    {
+      "epoch": 1.049789029535865,
+      "grad_norm": 0.46944454312324524,
+      "learning_rate": 0.00019479831765877509,
+      "loss": 0.8561921715736389,
+      "step": 2488
+    },
+    {
+      "epoch": 1.0506329113924051,
+      "grad_norm": 0.4455019533634186,
+      "learning_rate": 0.00019478301803658397,
+      "loss": 0.7206679582595825,
+      "step": 2490
+    },
+    {
+      "epoch": 1.051476793248945,
+      "grad_norm": 0.4253813326358795,
+      "learning_rate": 0.00019476769654977095,
+      "loss": 0.7271745800971985,
+      "step": 2492
+    },
+    {
+      "epoch": 1.0523206751054852,
+      "grad_norm": 0.42758429050445557,
+      "learning_rate": 0.00019475235320187033,
+      "loss": 0.7358576059341431,
+      "step": 2494
+    },
+    {
+      "epoch": 1.0531645569620254,
+      "grad_norm": 0.400634765625,
+      "learning_rate": 0.00019473698799642164,
+      "loss": 0.6617647409439087,
+      "step": 2496
+    },
+    {
+      "epoch": 1.0540084388185653,
+      "grad_norm": 0.4132104814052582,
+      "learning_rate": 0.00019472160093696926,
+      "loss": 0.6900970935821533,
+      "step": 2498
+    },
+    {
+      "epoch": 1.0548523206751055,
+      "grad_norm": 0.5116495490074158,
+      "learning_rate": 0.00019470619202706268,
+      "loss": 0.7352861762046814,
+      "step": 2500
+    },
+    {
+      "epoch": 1.0548523206751055,
+      "eval_loss": 0.8249453902244568,
+      "eval_runtime": 679.4446,
+      "eval_samples_per_second": 3.101,
+      "eval_steps_per_second": 3.101,
+      "step": 2500
+    },
+    {
+      "epoch": 1.0556962025316456,
+      "grad_norm": 0.42822444438934326,
+      "learning_rate": 0.0001946907612702565,
+      "loss": 0.7583479881286621,
+      "step": 2502
+    },
+    {
+      "epoch": 1.0565400843881856,
+      "grad_norm": 0.4788980782032013,
+      "learning_rate": 0.00019467530867011025,
+      "loss": 0.777511477470398,
+      "step": 2504
+    },
+    {
+      "epoch": 1.0573839662447257,
+      "grad_norm": 0.4210498034954071,
+      "learning_rate": 0.00019465983423018856,
+      "loss": 0.7452852129936218,
+      "step": 2506
+    },
+    {
+      "epoch": 1.058227848101266,
+      "grad_norm": 0.43219906091690063,
+      "learning_rate": 0.0001946443379540611,
+      "loss": 0.7540958523750305,
+      "step": 2508
+    },
+    {
+      "epoch": 1.0590717299578059,
+      "grad_norm": 0.4361002445220947,
+      "learning_rate": 0.00019462881984530253,
+      "loss": 0.7175624966621399,
+      "step": 2510
+    },
+    {
+      "epoch": 1.059915611814346,
+      "grad_norm": 0.4575091600418091,
+      "learning_rate": 0.00019461327990749264,
+      "loss": 0.7342076301574707,
+      "step": 2512
+    },
+    {
+      "epoch": 1.0607594936708862,
+      "grad_norm": 0.4823324978351593,
+      "learning_rate": 0.00019459771814421615,
+      "loss": 0.8255488872528076,
+      "step": 2514
+    },
+    {
+      "epoch": 1.0616033755274261,
+      "grad_norm": 0.41693785786628723,
+      "learning_rate": 0.00019458213455906284,
+      "loss": 0.6624729633331299,
+      "step": 2516
+    },
+    {
+      "epoch": 1.0624472573839663,
+      "grad_norm": 0.41319167613983154,
+      "learning_rate": 0.00019456652915562762,
+      "loss": 0.8224266767501831,
+      "step": 2518
+    },
+    {
+      "epoch": 1.0632911392405062,
+      "grad_norm": 0.4405190050601959,
+      "learning_rate": 0.00019455090193751023,
+      "loss": 0.7582900524139404,
+      "step": 2520
+    },
+    {
+      "epoch": 1.0641350210970464,
+      "grad_norm": 0.4564387798309326,
+      "learning_rate": 0.0001945352529083157,
+      "loss": 0.7948101162910461,
+      "step": 2522
+    },
+    {
+      "epoch": 1.0649789029535865,
+      "grad_norm": 0.4247225224971771,
+      "learning_rate": 0.00019451958207165394,
+      "loss": 0.7670508027076721,
+      "step": 2524
+    },
+    {
+      "epoch": 1.0658227848101265,
+      "grad_norm": 0.42070862650871277,
+      "learning_rate": 0.00019450388943113983,
+      "loss": 0.7753698229789734,
+      "step": 2526
+    },
+    {
+      "epoch": 1.0666666666666667,
+      "grad_norm": 0.3729179799556732,
+      "learning_rate": 0.00019448817499039342,
+      "loss": 0.710904598236084,
+      "step": 2528
+    },
+    {
+      "epoch": 1.0675105485232068,
+      "grad_norm": 0.40901753306388855,
+      "learning_rate": 0.0001944724387530397,
+      "loss": 0.7000213265419006,
+      "step": 2530
+    },
+    {
+      "epoch": 1.0683544303797468,
+      "grad_norm": 0.4481421113014221,
+      "learning_rate": 0.00019445668072270877,
+      "loss": 0.7374410629272461,
+      "step": 2532
+    },
+    {
+      "epoch": 1.069198312236287,
+      "grad_norm": 0.4839915335178375,
+      "learning_rate": 0.00019444090090303567,
+      "loss": 0.746309220790863,
+      "step": 2534
+    },
+    {
+      "epoch": 1.070042194092827,
+      "grad_norm": 0.4982132613658905,
+      "learning_rate": 0.00019442509929766048,
+      "loss": 0.7448083162307739,
+      "step": 2536
+    },
+    {
+      "epoch": 1.070886075949367,
+      "grad_norm": 0.3987475037574768,
+      "learning_rate": 0.00019440927591022838,
+      "loss": 0.7052555680274963,
+      "step": 2538
+    },
+    {
+      "epoch": 1.0717299578059072,
+      "grad_norm": 0.4089896082878113,
+      "learning_rate": 0.00019439343074438944,
+      "loss": 0.7090281248092651,
+      "step": 2540
+    },
+    {
+      "epoch": 1.0725738396624473,
+      "grad_norm": 0.3703136146068573,
+      "learning_rate": 0.00019437756380379895,
+      "loss": 0.7530601024627686,
+      "step": 2542
+    },
+    {
+      "epoch": 1.0734177215189873,
+      "grad_norm": 0.4015701413154602,
+      "learning_rate": 0.00019436167509211702,
+      "loss": 0.7597941160202026,
+      "step": 2544
+    },
+    {
+      "epoch": 1.0742616033755275,
+      "grad_norm": 0.3915148675441742,
+      "learning_rate": 0.00019434576461300888,
+      "loss": 0.7168124318122864,
+      "step": 2546
+    },
+    {
+      "epoch": 1.0751054852320676,
+      "grad_norm": 0.44643113017082214,
+      "learning_rate": 0.00019432983237014482,
+      "loss": 0.7762725949287415,
+      "step": 2548
+    },
+    {
+      "epoch": 1.0759493670886076,
+      "grad_norm": 0.46874719858169556,
+      "learning_rate": 0.00019431387836720004,
+      "loss": 0.7879725098609924,
+      "step": 2550
+    },
+    {
+      "epoch": 1.0767932489451477,
+      "grad_norm": 0.4143001437187195,
+      "learning_rate": 0.00019429790260785487,
+      "loss": 0.7351391315460205,
+      "step": 2552
+    },
+    {
+      "epoch": 1.0776371308016879,
+      "grad_norm": 0.4569888412952423,
+      "learning_rate": 0.0001942819050957946,
+      "loss": 0.86613929271698,
+      "step": 2554
+    },
+    {
+      "epoch": 1.0784810126582278,
+      "grad_norm": 0.41729897260665894,
+      "learning_rate": 0.00019426588583470954,
+      "loss": 0.7122719287872314,
+      "step": 2556
+    },
+    {
+      "epoch": 1.079324894514768,
+      "grad_norm": 0.42295604944229126,
+      "learning_rate": 0.000194249844828295,
+      "loss": 0.6605332493782043,
+      "step": 2558
+    },
+    {
+      "epoch": 1.080168776371308,
+      "grad_norm": 0.45787641406059265,
+      "learning_rate": 0.00019423378208025137,
+      "loss": 0.7839207649230957,
+      "step": 2560
+    },
+    {
+      "epoch": 1.081012658227848,
+      "grad_norm": 0.43370622396469116,
+      "learning_rate": 0.00019421769759428404,
+      "loss": 0.7557987570762634,
+      "step": 2562
+    },
+    {
+      "epoch": 1.0818565400843883,
+      "grad_norm": 0.4371975362300873,
+      "learning_rate": 0.00019420159137410335,
+      "loss": 0.7349345088005066,
+      "step": 2564
+    },
+    {
+      "epoch": 1.0827004219409282,
+      "grad_norm": 0.468730628490448,
+      "learning_rate": 0.0001941854634234247,
+      "loss": 0.7024962902069092,
+      "step": 2566
+    },
+    {
+      "epoch": 1.0835443037974684,
+      "grad_norm": 0.4257284998893738,
+      "learning_rate": 0.0001941693137459685,
+      "loss": 0.6981064081192017,
+      "step": 2568
+    },
+    {
+      "epoch": 1.0843881856540085,
+      "grad_norm": 0.4322902262210846,
+      "learning_rate": 0.00019415314234546015,
+      "loss": 0.7899637818336487,
+      "step": 2570
+    },
+    {
+      "epoch": 1.0852320675105485,
+      "grad_norm": 0.44384217262268066,
+      "learning_rate": 0.00019413694922563014,
+      "loss": 0.7964152693748474,
+      "step": 2572
+    },
+    {
+      "epoch": 1.0860759493670886,
+      "grad_norm": 0.47391360998153687,
+      "learning_rate": 0.0001941207343902139,
+      "loss": 0.7902941107749939,
+      "step": 2574
+    },
+    {
+      "epoch": 1.0869198312236288,
+      "grad_norm": 0.4575883746147156,
+      "learning_rate": 0.00019410449784295182,
+      "loss": 0.7818595170974731,
+      "step": 2576
+    },
+    {
+      "epoch": 1.0877637130801687,
+      "grad_norm": 0.41236087679862976,
+      "learning_rate": 0.0001940882395875894,
+      "loss": 0.7516021728515625,
+      "step": 2578
+    },
+    {
+      "epoch": 1.0886075949367089,
+      "grad_norm": 0.3817494213581085,
+      "learning_rate": 0.0001940719596278771,
+      "loss": 0.6870401501655579,
+      "step": 2580
+    },
+    {
+      "epoch": 1.0894514767932488,
+      "grad_norm": 0.4630168378353119,
+      "learning_rate": 0.00019405565796757043,
+      "loss": 0.7437341809272766,
+      "step": 2582
+    },
+    {
+      "epoch": 1.090295358649789,
+      "grad_norm": 0.4063820540904999,
+      "learning_rate": 0.00019403933461042982,
+      "loss": 0.7101680040359497,
+      "step": 2584
+    },
+    {
+      "epoch": 1.0911392405063292,
+      "grad_norm": 0.44423359632492065,
+      "learning_rate": 0.0001940229895602208,
+      "loss": 0.6894780397415161,
+      "step": 2586
+    },
+    {
+      "epoch": 1.091983122362869,
+      "grad_norm": 0.4267353117465973,
+      "learning_rate": 0.0001940066228207138,
+      "loss": 0.7277120351791382,
+      "step": 2588
+    },
+    {
+      "epoch": 1.0928270042194093,
+      "grad_norm": 0.44747647643089294,
+      "learning_rate": 0.00019399023439568434,
+      "loss": 0.6512353420257568,
+      "step": 2590
+    },
+    {
+      "epoch": 1.0936708860759494,
+      "grad_norm": 0.4557100832462311,
+      "learning_rate": 0.00019397382428891296,
+      "loss": 0.7850068807601929,
+      "step": 2592
+    },
+    {
+      "epoch": 1.0945147679324894,
+      "grad_norm": 0.44980114698410034,
+      "learning_rate": 0.0001939573925041851,
+      "loss": 0.786868691444397,
+      "step": 2594
+    },
+    {
+      "epoch": 1.0953586497890295,
+      "grad_norm": 0.45374399423599243,
+      "learning_rate": 0.00019394093904529126,
+      "loss": 0.8260899186134338,
+      "step": 2596
+    },
+    {
+      "epoch": 1.0962025316455697,
+      "grad_norm": 0.40681353211402893,
+      "learning_rate": 0.00019392446391602695,
+      "loss": 0.7056689262390137,
+      "step": 2598
+    },
+    {
+      "epoch": 1.0970464135021096,
+      "grad_norm": 0.39278075098991394,
+      "learning_rate": 0.0001939079671201927,
+      "loss": 0.7557252645492554,
+      "step": 2600
+    },
+    {
+      "epoch": 1.0970464135021096,
+      "eval_loss": 0.8211485743522644,
+      "eval_runtime": 670.2276,
+      "eval_samples_per_second": 3.144,
+      "eval_steps_per_second": 3.144,
+      "step": 2600
+    },
+    {
+      "epoch": 1.0978902953586498,
+      "grad_norm": 0.4261230230331421,
+      "learning_rate": 0.00019389144866159398,
+      "loss": 0.7139819264411926,
+      "step": 2602
+    },
+    {
+      "epoch": 1.09873417721519,
+      "grad_norm": 0.4664439558982849,
+      "learning_rate": 0.0001938749085440412,
+      "loss": 0.7274705767631531,
+      "step": 2604
+    },
+    {
+      "epoch": 1.09957805907173,
+      "grad_norm": 0.46306154131889343,
+      "learning_rate": 0.00019385834677134998,
+      "loss": 0.7547051906585693,
+      "step": 2606
+    },
+    {
+      "epoch": 1.10042194092827,
+      "grad_norm": 0.39753687381744385,
+      "learning_rate": 0.00019384176334734073,
+      "loss": 0.6942694187164307,
+      "step": 2608
+    },
+    {
+      "epoch": 1.1012658227848102,
+      "grad_norm": 0.4556522071361542,
+      "learning_rate": 0.00019382515827583898,
+      "loss": 0.7161701321601868,
+      "step": 2610
+    },
+    {
+      "epoch": 1.1021097046413502,
+      "grad_norm": 0.3863290250301361,
+      "learning_rate": 0.0001938085315606751,
+      "loss": 0.7165710926055908,
+      "step": 2612
+    },
+    {
+      "epoch": 1.1029535864978903,
+      "grad_norm": 0.49601292610168457,
+      "learning_rate": 0.00019379188320568466,
+      "loss": 0.8080679774284363,
+      "step": 2614
+    },
+    {
+      "epoch": 1.1037974683544305,
+      "grad_norm": 0.4498465955257416,
+      "learning_rate": 0.00019377521321470805,
+      "loss": 0.7032502889633179,
+      "step": 2616
+    },
+    {
+      "epoch": 1.1046413502109704,
+      "grad_norm": 0.4537963569164276,
+      "learning_rate": 0.00019375852159159073,
+      "loss": 0.7076731324195862,
+      "step": 2618
+    },
+    {
+      "epoch": 1.1054852320675106,
+      "grad_norm": 0.4042968153953552,
+      "learning_rate": 0.0001937418083401832,
+      "loss": 0.7020825743675232,
+      "step": 2620
+    },
+    {
+      "epoch": 1.1063291139240505,
+      "grad_norm": 0.44306516647338867,
+      "learning_rate": 0.00019372507346434076,
+      "loss": 0.7222874760627747,
+      "step": 2622
+    },
+    {
+      "epoch": 1.1071729957805907,
+      "grad_norm": 0.3792516887187958,
+      "learning_rate": 0.00019370831696792392,
+      "loss": 0.6136211156845093,
+      "step": 2624
+    },
+    {
+      "epoch": 1.1080168776371309,
+      "grad_norm": 0.46655353903770447,
+      "learning_rate": 0.00019369153885479806,
+      "loss": 0.7358977198600769,
+      "step": 2626
+    },
+    {
+      "epoch": 1.1088607594936708,
+      "grad_norm": 0.4416999816894531,
+      "learning_rate": 0.0001936747391288336,
+      "loss": 0.7484551072120667,
+      "step": 2628
+    },
+    {
+      "epoch": 1.109704641350211,
+      "grad_norm": 0.4620150327682495,
+      "learning_rate": 0.00019365791779390584,
+      "loss": 0.6874802708625793,
+      "step": 2630
+    },
+    {
+      "epoch": 1.1105485232067511,
+      "grad_norm": 0.4357883334159851,
+      "learning_rate": 0.00019364107485389518,
+      "loss": 0.7168865203857422,
+      "step": 2632
+    },
+    {
+      "epoch": 1.111392405063291,
+      "grad_norm": 0.4581143856048584,
+      "learning_rate": 0.00019362421031268699,
+      "loss": 0.7533737421035767,
+      "step": 2634
+    },
+    {
+      "epoch": 1.1122362869198312,
+      "grad_norm": 0.46408653259277344,
+      "learning_rate": 0.00019360732417417152,
+      "loss": 0.8245398998260498,
+      "step": 2636
+    },
+    {
+      "epoch": 1.1130801687763714,
+      "grad_norm": 0.4639282524585724,
+      "learning_rate": 0.00019359041644224415,
+      "loss": 0.7095663547515869,
+      "step": 2638
+    },
+    {
+      "epoch": 1.1139240506329113,
+      "grad_norm": 0.4598273038864136,
+      "learning_rate": 0.00019357348712080517,
+      "loss": 0.7030285596847534,
+      "step": 2640
+    },
+    {
+      "epoch": 1.1147679324894515,
+      "grad_norm": 0.42993804812431335,
+      "learning_rate": 0.00019355653621375978,
+      "loss": 0.7158969640731812,
+      "step": 2642
+    },
+    {
+      "epoch": 1.1156118143459917,
+      "grad_norm": 0.36132168769836426,
+      "learning_rate": 0.0001935395637250183,
+      "loss": 0.6800820231437683,
+      "step": 2644
+    },
+    {
+      "epoch": 1.1164556962025316,
+      "grad_norm": 0.5004428029060364,
+      "learning_rate": 0.00019352256965849592,
+      "loss": 0.694972813129425,
+      "step": 2646
+    },
+    {
+      "epoch": 1.1172995780590718,
+      "grad_norm": 0.43715354800224304,
+      "learning_rate": 0.00019350555401811286,
+      "loss": 0.7164262533187866,
+      "step": 2648
+    },
+    {
+      "epoch": 1.1181434599156117,
+      "grad_norm": 0.42316755652427673,
+      "learning_rate": 0.00019348851680779433,
+      "loss": 0.7026737928390503,
+      "step": 2650
+    },
+    {
+      "epoch": 1.1189873417721519,
+      "grad_norm": 0.35848063230514526,
+      "learning_rate": 0.0001934714580314704,
+      "loss": 0.6774232387542725,
+      "step": 2652
+    },
+    {
+      "epoch": 1.119831223628692,
+      "grad_norm": 0.46613892912864685,
+      "learning_rate": 0.00019345437769307627,
+      "loss": 0.7789244055747986,
+      "step": 2654
+    },
+    {
+      "epoch": 1.120675105485232,
+      "grad_norm": 0.4298146367073059,
+      "learning_rate": 0.00019343727579655204,
+      "loss": 0.8647177815437317,
+      "step": 2656
+    },
+    {
+      "epoch": 1.1215189873417721,
+      "grad_norm": 0.4638161361217499,
+      "learning_rate": 0.00019342015234584276,
+      "loss": 0.6949124336242676,
+      "step": 2658
+    },
+    {
+      "epoch": 1.1223628691983123,
+      "grad_norm": 0.4079605042934418,
+      "learning_rate": 0.00019340300734489854,
+      "loss": 0.6677125096321106,
+      "step": 2660
+    },
+    {
+      "epoch": 1.1232067510548522,
+      "grad_norm": 0.46079063415527344,
+      "learning_rate": 0.00019338584079767434,
+      "loss": 0.7521067261695862,
+      "step": 2662
+    },
+    {
+      "epoch": 1.1240506329113924,
+      "grad_norm": 0.3935202956199646,
+      "learning_rate": 0.0001933686527081302,
+      "loss": 0.6829184889793396,
+      "step": 2664
+    },
+    {
+      "epoch": 1.1248945147679326,
+      "grad_norm": 0.45696479082107544,
+      "learning_rate": 0.000193351443080231,
+      "loss": 0.7122842669487,
+      "step": 2666
+    },
+    {
+      "epoch": 1.1257383966244725,
+      "grad_norm": 0.4078282415866852,
+      "learning_rate": 0.0001933342119179468,
+      "loss": 0.7316977977752686,
+      "step": 2668
+    },
+    {
+      "epoch": 1.1265822784810127,
+      "grad_norm": 0.4282447099685669,
+      "learning_rate": 0.0001933169592252524,
+      "loss": 0.7323395013809204,
+      "step": 2670
+    },
+    {
+      "epoch": 1.1274261603375528,
+      "grad_norm": 0.46398648619651794,
+      "learning_rate": 0.00019329968500612766,
+      "loss": 0.7177348732948303,
+      "step": 2672
+    },
+    {
+      "epoch": 1.1282700421940928,
+      "grad_norm": 0.4560202658176422,
+      "learning_rate": 0.0001932823892645575,
+      "loss": 0.7957183122634888,
+      "step": 2674
+    },
+    {
+      "epoch": 1.129113924050633,
+      "grad_norm": 0.4790743887424469,
+      "learning_rate": 0.00019326507200453162,
+      "loss": 0.762060284614563,
+      "step": 2676
+    },
+    {
+      "epoch": 1.129957805907173,
+      "grad_norm": 0.46524202823638916,
+      "learning_rate": 0.00019324773323004484,
+      "loss": 0.7227895259857178,
+      "step": 2678
+    },
+    {
+      "epoch": 1.130801687763713,
+      "grad_norm": 0.47912222146987915,
+      "learning_rate": 0.00019323037294509683,
+      "loss": 0.6968042254447937,
+      "step": 2680
+    },
+    {
+      "epoch": 1.1316455696202532,
+      "grad_norm": 0.4412849247455597,
+      "learning_rate": 0.00019321299115369233,
+      "loss": 0.6682016253471375,
+      "step": 2682
+    },
+    {
+      "epoch": 1.1324894514767934,
+      "grad_norm": 0.4138370752334595,
+      "learning_rate": 0.00019319558785984095,
+      "loss": 0.7287812829017639,
+      "step": 2684
+    },
+    {
+      "epoch": 1.1333333333333333,
+      "grad_norm": 0.391170859336853,
+      "learning_rate": 0.00019317816306755733,
+      "loss": 0.6547526121139526,
+      "step": 2686
+    },
+    {
+      "epoch": 1.1341772151898735,
+      "grad_norm": 0.45472806692123413,
+      "learning_rate": 0.00019316071678086098,
+      "loss": 0.6942042708396912,
+      "step": 2688
+    },
+    {
+      "epoch": 1.1350210970464134,
+      "grad_norm": 0.40321481227874756,
+      "learning_rate": 0.00019314324900377648,
+      "loss": 0.7417841553688049,
+      "step": 2690
+    },
+    {
+      "epoch": 1.1358649789029536,
+      "grad_norm": 0.3806784152984619,
+      "learning_rate": 0.00019312575974033327,
+      "loss": 0.7780525088310242,
+      "step": 2692
+    },
+    {
+      "epoch": 1.1367088607594937,
+      "grad_norm": 0.43763765692710876,
+      "learning_rate": 0.0001931082489945658,
+      "loss": 0.661472499370575,
+      "step": 2694
+    },
+    {
+      "epoch": 1.1375527426160337,
+      "grad_norm": 0.45410826802253723,
+      "learning_rate": 0.00019309071677051348,
+      "loss": 0.7443127632141113,
+      "step": 2696
+    },
+    {
+      "epoch": 1.1383966244725738,
+      "grad_norm": 0.47907447814941406,
+      "learning_rate": 0.00019307316307222062,
+      "loss": 0.744393527507782,
+      "step": 2698
+    },
+    {
+      "epoch": 1.139240506329114,
+      "grad_norm": 0.41723814606666565,
+      "learning_rate": 0.00019305558790373654,
+      "loss": 0.7042794227600098,
+      "step": 2700
+    },
+    {
+      "epoch": 1.139240506329114,
+      "eval_loss": 0.8155058026313782,
+      "eval_runtime": 678.284,
+      "eval_samples_per_second": 3.106,
+      "eval_steps_per_second": 3.106,
+      "step": 2700
+    },
+    {
+      "epoch": 1.140084388185654,
+      "grad_norm": 0.3863268196582794,
+      "learning_rate": 0.0001930379912691155,
+      "loss": 0.6715745329856873,
+      "step": 2702
+    },
+    {
+      "epoch": 1.140928270042194,
+      "grad_norm": 0.49994897842407227,
+      "learning_rate": 0.00019302037317241673,
+      "loss": 0.6959030628204346,
+      "step": 2704
+    },
+    {
+      "epoch": 1.1417721518987343,
+      "grad_norm": 0.46377572417259216,
+      "learning_rate": 0.00019300273361770432,
+      "loss": 0.6534222364425659,
+      "step": 2706
+    },
+    {
+      "epoch": 1.1426160337552742,
+      "grad_norm": 0.3841579556465149,
+      "learning_rate": 0.00019298507260904745,
+      "loss": 0.6750475764274597,
+      "step": 2708
+    },
+    {
+      "epoch": 1.1434599156118144,
+      "grad_norm": 0.4171544313430786,
+      "learning_rate": 0.00019296739015052015,
+      "loss": 0.7714900374412537,
+      "step": 2710
+    },
+    {
+      "epoch": 1.1443037974683543,
+      "grad_norm": 0.44384199380874634,
+      "learning_rate": 0.00019294968624620137,
+      "loss": 0.7281306982040405,
+      "step": 2712
+    },
+    {
+      "epoch": 1.1451476793248945,
+      "grad_norm": 0.43345192074775696,
+      "learning_rate": 0.00019293196090017513,
+      "loss": 0.7407262325286865,
+      "step": 2714
+    },
+    {
+      "epoch": 1.1459915611814346,
+      "grad_norm": 0.3897020220756531,
+      "learning_rate": 0.00019291421411653034,
+      "loss": 0.6794347167015076,
+      "step": 2716
+    },
+    {
+      "epoch": 1.1468354430379746,
+      "grad_norm": 0.4289143681526184,
+      "learning_rate": 0.00019289644589936075,
+      "loss": 0.7026289105415344,
+      "step": 2718
+    },
+    {
+      "epoch": 1.1476793248945147,
+      "grad_norm": 0.4783557057380676,
+      "learning_rate": 0.0001928786562527652,
+      "loss": 0.7613762617111206,
+      "step": 2720
+    },
+    {
+      "epoch": 1.148523206751055,
+      "grad_norm": 0.4152112901210785,
+      "learning_rate": 0.00019286084518084745,
+      "loss": 0.7123059630393982,
+      "step": 2722
+    },
+    {
+      "epoch": 1.1493670886075948,
+      "grad_norm": 0.4204908013343811,
+      "learning_rate": 0.0001928430126877161,
+      "loss": 0.7198041677474976,
+      "step": 2724
+    },
+    {
+      "epoch": 1.150210970464135,
+      "grad_norm": 0.41333863139152527,
+      "learning_rate": 0.00019282515877748485,
+      "loss": 0.6859839558601379,
+      "step": 2726
+    },
+    {
+      "epoch": 1.1510548523206752,
+      "grad_norm": 0.4041498005390167,
+      "learning_rate": 0.00019280728345427218,
+      "loss": 0.7185758948326111,
+      "step": 2728
+    },
+    {
+      "epoch": 1.1518987341772151,
+      "grad_norm": 0.4487358629703522,
+      "learning_rate": 0.00019278938672220167,
+      "loss": 0.7899971008300781,
+      "step": 2730
+    },
+    {
+      "epoch": 1.1527426160337553,
+      "grad_norm": 0.3510114252567291,
+      "learning_rate": 0.00019277146858540166,
+      "loss": 0.6266586780548096,
+      "step": 2732
+    },
+    {
+      "epoch": 1.1535864978902954,
+      "grad_norm": 0.4295555055141449,
+      "learning_rate": 0.00019275352904800553,
+      "loss": 0.7450110912322998,
+      "step": 2734
+    },
+    {
+      "epoch": 1.1544303797468354,
+      "grad_norm": 0.40629687905311584,
+      "learning_rate": 0.00019273556811415165,
+      "loss": 0.6875549554824829,
+      "step": 2736
+    },
+    {
+      "epoch": 1.1552742616033755,
+      "grad_norm": 0.4243362247943878,
+      "learning_rate": 0.00019271758578798322,
+      "loss": 0.8141522407531738,
+      "step": 2738
+    },
+    {
+      "epoch": 1.1561181434599157,
+      "grad_norm": 0.483306884765625,
+      "learning_rate": 0.00019269958207364843,
+      "loss": 0.6905186176300049,
+      "step": 2740
+    },
+    {
+      "epoch": 1.1569620253164556,
+      "grad_norm": 0.4303925335407257,
+      "learning_rate": 0.00019268155697530037,
+      "loss": 0.6709047555923462,
+      "step": 2742
+    },
+    {
+      "epoch": 1.1578059071729958,
+      "grad_norm": 0.4533381760120392,
+      "learning_rate": 0.0001926635104970971,
+      "loss": 0.7107465863227844,
+      "step": 2744
+    },
+    {
+      "epoch": 1.158649789029536,
+      "grad_norm": 0.4823286533355713,
+      "learning_rate": 0.0001926454426432016,
+      "loss": 0.8043986558914185,
+      "step": 2746
+    },
+    {
+      "epoch": 1.159493670886076,
+      "grad_norm": 0.3892867863178253,
+      "learning_rate": 0.00019262735341778179,
+      "loss": 0.7812165021896362,
+      "step": 2748
+    },
+    {
+      "epoch": 1.160337552742616,
+      "grad_norm": 0.46737003326416016,
+      "learning_rate": 0.00019260924282501045,
+      "loss": 0.7141886353492737,
+      "step": 2750
+    },
+    {
+      "epoch": 1.1611814345991562,
+      "grad_norm": 0.4335328936576843,
+      "learning_rate": 0.0001925911108690654,
+      "loss": 0.7528460621833801,
+      "step": 2752
+    },
+    {
+      "epoch": 1.1620253164556962,
+      "grad_norm": 0.3792484700679779,
+      "learning_rate": 0.00019257295755412932,
+      "loss": 0.7817611694335938,
+      "step": 2754
+    },
+    {
+      "epoch": 1.1628691983122363,
+      "grad_norm": 0.41246429085731506,
+      "learning_rate": 0.00019255478288438983,
+      "loss": 0.6904958486557007,
+      "step": 2756
+    },
+    {
+      "epoch": 1.1637130801687763,
+      "grad_norm": 0.44368481636047363,
+      "learning_rate": 0.00019253658686403948,
+      "loss": 0.7571630477905273,
+      "step": 2758
+    },
+    {
+      "epoch": 1.1645569620253164,
+      "grad_norm": 0.40121781826019287,
+      "learning_rate": 0.00019251836949727574,
+      "loss": 0.6747061014175415,
+      "step": 2760
+    },
+    {
+      "epoch": 1.1654008438818566,
+      "grad_norm": 0.4660188853740692,
+      "learning_rate": 0.000192500130788301,
+      "loss": 0.7722339034080505,
+      "step": 2762
+    },
+    {
+      "epoch": 1.1662447257383965,
+      "grad_norm": 0.41693782806396484,
+      "learning_rate": 0.0001924818707413226,
+      "loss": 0.8081457018852234,
+      "step": 2764
+    },
+    {
+      "epoch": 1.1670886075949367,
+      "grad_norm": 0.4274570345878601,
+      "learning_rate": 0.00019246358936055274,
+      "loss": 0.7745853066444397,
+      "step": 2766
+    },
+    {
+      "epoch": 1.1679324894514769,
+      "grad_norm": 0.4385969638824463,
+      "learning_rate": 0.00019244528665020863,
+      "loss": 0.7918146252632141,
+      "step": 2768
+    },
+    {
+      "epoch": 1.1687763713080168,
+      "grad_norm": 0.42832866311073303,
+      "learning_rate": 0.00019242696261451234,
+      "loss": 0.7382830381393433,
+      "step": 2770
+    },
+    {
+      "epoch": 1.169620253164557,
+      "grad_norm": 0.44177982211112976,
+      "learning_rate": 0.00019240861725769085,
+      "loss": 0.7501721978187561,
+      "step": 2772
+    },
+    {
+      "epoch": 1.1704641350210971,
+      "grad_norm": 0.4193457365036011,
+      "learning_rate": 0.00019239025058397611,
+      "loss": 0.6495388150215149,
+      "step": 2774
+    },
+    {
+      "epoch": 1.171308016877637,
+      "grad_norm": 0.5145925283432007,
+      "learning_rate": 0.000192371862597605,
+      "loss": 0.8291991353034973,
+      "step": 2776
+    },
+    {
+      "epoch": 1.1721518987341772,
+      "grad_norm": 0.40911251306533813,
+      "learning_rate": 0.00019235345330281914,
+      "loss": 0.7491340637207031,
+      "step": 2778
+    },
+    {
+      "epoch": 1.1729957805907172,
+      "grad_norm": 0.3920276165008545,
+      "learning_rate": 0.0001923350227038653,
+      "loss": 0.6315004825592041,
+      "step": 2780
+    },
+    {
+      "epoch": 1.1738396624472573,
+      "grad_norm": 0.442303329706192,
+      "learning_rate": 0.0001923165708049951,
+      "loss": 0.7538990378379822,
+      "step": 2782
+    },
+    {
+      "epoch": 1.1746835443037975,
+      "grad_norm": 0.4693523049354553,
+      "learning_rate": 0.00019229809761046497,
+      "loss": 0.7440794110298157,
+      "step": 2784
+    },
+    {
+      "epoch": 1.1755274261603375,
+      "grad_norm": 0.4948059618473053,
+      "learning_rate": 0.0001922796031245363,
+      "loss": 0.7342568635940552,
+      "step": 2786
+    },
+    {
+      "epoch": 1.1763713080168776,
+      "grad_norm": 0.3900465965270996,
+      "learning_rate": 0.00019226108735147547,
+      "loss": 0.6969661116600037,
+      "step": 2788
+    },
+    {
+      "epoch": 1.1772151898734178,
+      "grad_norm": 0.36153602600097656,
+      "learning_rate": 0.00019224255029555372,
+      "loss": 0.7097541689872742,
+      "step": 2790
+    },
+    {
+      "epoch": 1.1780590717299577,
+      "grad_norm": 0.41168832778930664,
+      "learning_rate": 0.00019222399196104715,
+      "loss": 0.8141831755638123,
+      "step": 2792
+    },
+    {
+      "epoch": 1.1789029535864979,
+      "grad_norm": 0.4542821943759918,
+      "learning_rate": 0.0001922054123522368,
+      "loss": 0.7458863258361816,
+      "step": 2794
+    },
+    {
+      "epoch": 1.179746835443038,
+      "grad_norm": 0.4026047885417938,
+      "learning_rate": 0.0001921868114734087,
+      "loss": 0.6535642147064209,
+      "step": 2796
+    },
+    {
+      "epoch": 1.180590717299578,
+      "grad_norm": 0.4417034387588501,
+      "learning_rate": 0.0001921681893288537,
+      "loss": 0.6248135566711426,
+      "step": 2798
+    },
+    {
+      "epoch": 1.1814345991561181,
+      "grad_norm": 0.45027074217796326,
+      "learning_rate": 0.00019214954592286748,
+      "loss": 0.6777785420417786,
+      "step": 2800
+    },
+    {
+      "epoch": 1.1814345991561181,
+      "eval_loss": 0.8124309182167053,
+      "eval_runtime": 688.4759,
+      "eval_samples_per_second": 3.06,
+      "eval_steps_per_second": 3.06,
+      "step": 2800
+    },
+    {
+      "epoch": 1.1822784810126583,
+      "grad_norm": 0.406865656375885,
+      "learning_rate": 0.00019213088125975082,
+      "loss": 0.6460301876068115,
+      "step": 2802
+    },
+    {
+      "epoch": 1.1831223628691983,
+      "grad_norm": 0.5052749514579773,
+      "learning_rate": 0.00019211219534380928,
+      "loss": 0.7292830944061279,
+      "step": 2804
+    },
+    {
+      "epoch": 1.1839662447257384,
+      "grad_norm": 0.4028380513191223,
+      "learning_rate": 0.00019209348817935328,
+      "loss": 0.7399409413337708,
+      "step": 2806
+    },
+    {
+      "epoch": 1.1848101265822786,
+      "grad_norm": 0.4522664248943329,
+      "learning_rate": 0.00019207475977069826,
+      "loss": 0.6885398030281067,
+      "step": 2808
+    },
+    {
+      "epoch": 1.1856540084388185,
+      "grad_norm": 0.419546514749527,
+      "learning_rate": 0.0001920560101221645,
+      "loss": 0.6962382197380066,
+      "step": 2810
+    },
+    {
+      "epoch": 1.1864978902953587,
+      "grad_norm": 0.48912352323532104,
+      "learning_rate": 0.00019203723923807715,
+      "loss": 0.7380965352058411,
+      "step": 2812
+    },
+    {
+      "epoch": 1.1873417721518988,
+      "grad_norm": 0.40905487537384033,
+      "learning_rate": 0.00019201844712276635,
+      "loss": 0.755382776260376,
+      "step": 2814
+    },
+    {
+      "epoch": 1.1881856540084388,
+      "grad_norm": 0.385623037815094,
+      "learning_rate": 0.000191999633780567,
+      "loss": 0.7066168189048767,
+      "step": 2816
+    },
+    {
+      "epoch": 1.189029535864979,
+      "grad_norm": 0.4488723576068878,
+      "learning_rate": 0.00019198079921581908,
+      "loss": 0.7172947525978088,
+      "step": 2818
+    },
+    {
+      "epoch": 1.189873417721519,
+      "grad_norm": 0.385277658700943,
+      "learning_rate": 0.00019196194343286727,
+      "loss": 0.6714195013046265,
+      "step": 2820
+    },
+    {
+      "epoch": 1.190717299578059,
+      "grad_norm": 0.4280855655670166,
+      "learning_rate": 0.00019194306643606127,
+      "loss": 0.7076966166496277,
+      "step": 2822
+    },
+    {
+      "epoch": 1.1915611814345992,
+      "grad_norm": 0.4626067578792572,
+      "learning_rate": 0.00019192416822975567,
+      "loss": 0.7926747798919678,
+      "step": 2824
+    },
+    {
+      "epoch": 1.1924050632911392,
+      "grad_norm": 0.3994208574295044,
+      "learning_rate": 0.00019190524881830987,
+      "loss": 0.7654290199279785,
+      "step": 2826
+    },
+    {
+      "epoch": 1.1932489451476793,
+      "grad_norm": 0.47346508502960205,
+      "learning_rate": 0.00019188630820608825,
+      "loss": 0.7857998609542847,
+      "step": 2828
+    },
+    {
+      "epoch": 1.1940928270042195,
+      "grad_norm": 0.446077823638916,
+      "learning_rate": 0.00019186734639746003,
+      "loss": 0.7603738903999329,
+      "step": 2830
+    },
+    {
+      "epoch": 1.1949367088607594,
+      "grad_norm": 0.41503727436065674,
+      "learning_rate": 0.00019184836339679936,
+      "loss": 0.790582537651062,
+      "step": 2832
+    },
+    {
+      "epoch": 1.1957805907172996,
+      "grad_norm": 0.43651700019836426,
+      "learning_rate": 0.00019182935920848522,
+      "loss": 0.7229864597320557,
+      "step": 2834
+    },
+    {
+      "epoch": 1.1966244725738397,
+      "grad_norm": 0.4409031867980957,
+      "learning_rate": 0.00019181033383690153,
+      "loss": 0.7817287445068359,
+      "step": 2836
+    },
+    {
+      "epoch": 1.1974683544303797,
+      "grad_norm": 0.3562015891075134,
+      "learning_rate": 0.0001917912872864371,
+      "loss": 0.6342297792434692,
+      "step": 2838
+    },
+    {
+      "epoch": 1.1983122362869199,
+      "grad_norm": 0.4767192602157593,
+      "learning_rate": 0.00019177221956148555,
+      "loss": 0.7065272927284241,
+      "step": 2840
+    },
+    {
+      "epoch": 1.1991561181434598,
+      "grad_norm": 0.45131945610046387,
+      "learning_rate": 0.00019175313066644546,
+      "loss": 0.777110755443573,
+      "step": 2842
+    },
+    {
+      "epoch": 1.2,
+      "grad_norm": 0.47566652297973633,
+      "learning_rate": 0.00019173402060572027,
+      "loss": 0.7456457018852234,
+      "step": 2844
+    },
+    {
+      "epoch": 1.2008438818565401,
+      "grad_norm": 0.48909857869148254,
+      "learning_rate": 0.00019171488938371834,
+      "loss": 0.8041547536849976,
+      "step": 2846
+    },
+    {
+      "epoch": 1.20168776371308,
+      "grad_norm": 0.436591774225235,
+      "learning_rate": 0.00019169573700485284,
+      "loss": 0.6786676049232483,
+      "step": 2848
+    },
+    {
+      "epoch": 1.2025316455696202,
+      "grad_norm": 0.44230160117149353,
+      "learning_rate": 0.00019167656347354183,
+      "loss": 0.7661880850791931,
+      "step": 2850
+    },
+    {
+      "epoch": 1.2033755274261604,
+      "grad_norm": 0.4844048321247101,
+      "learning_rate": 0.00019165736879420836,
+      "loss": 0.7369645833969116,
+      "step": 2852
+    },
+    {
+      "epoch": 1.2042194092827003,
+      "grad_norm": 0.4272955060005188,
+      "learning_rate": 0.0001916381529712802,
+      "loss": 0.7776681184768677,
+      "step": 2854
+    },
+    {
+      "epoch": 1.2050632911392405,
+      "grad_norm": 0.42392730712890625,
+      "learning_rate": 0.00019161891600919008,
+      "loss": 0.8019667267799377,
+      "step": 2856
+    },
+    {
+      "epoch": 1.2059071729957807,
+      "grad_norm": 0.4528326988220215,
+      "learning_rate": 0.00019159965791237565,
+      "loss": 0.7811952233314514,
+      "step": 2858
+    },
+    {
+      "epoch": 1.2067510548523206,
+      "grad_norm": 0.42840948700904846,
+      "learning_rate": 0.00019158037868527932,
+      "loss": 0.7244279980659485,
+      "step": 2860
+    },
+    {
+      "epoch": 1.2075949367088608,
+      "grad_norm": 0.4291226863861084,
+      "learning_rate": 0.00019156107833234845,
+      "loss": 0.7675760984420776,
+      "step": 2862
+    },
+    {
+      "epoch": 1.208438818565401,
+      "grad_norm": 0.4455093443393707,
+      "learning_rate": 0.00019154175685803527,
+      "loss": 0.7071483135223389,
+      "step": 2864
+    },
+    {
+      "epoch": 1.2092827004219409,
+      "grad_norm": 0.41356268525123596,
+      "learning_rate": 0.00019152241426679688,
+      "loss": 0.7554603815078735,
+      "step": 2866
+    },
+    {
+      "epoch": 1.210126582278481,
+      "grad_norm": 0.43875646591186523,
+      "learning_rate": 0.00019150305056309525,
+      "loss": 0.7198356986045837,
+      "step": 2868
+    },
+    {
+      "epoch": 1.2109704641350212,
+      "grad_norm": 0.454356849193573,
+      "learning_rate": 0.0001914836657513972,
+      "loss": 0.7393195629119873,
+      "step": 2870
+    },
+    {
+      "epoch": 1.2118143459915611,
+      "grad_norm": 0.37668535113334656,
+      "learning_rate": 0.00019146425983617444,
+      "loss": 0.7277671694755554,
+      "step": 2872
+    },
+    {
+      "epoch": 1.2126582278481013,
+      "grad_norm": 0.37869662046432495,
+      "learning_rate": 0.00019144483282190354,
+      "loss": 0.6783158183097839,
+      "step": 2874
+    },
+    {
+      "epoch": 1.2135021097046415,
+      "grad_norm": 0.4333525002002716,
+      "learning_rate": 0.00019142538471306596,
+      "loss": 0.7024772763252258,
+      "step": 2876
+    },
+    {
+      "epoch": 1.2143459915611814,
+      "grad_norm": 0.4693842828273773,
+      "learning_rate": 0.00019140591551414796,
+      "loss": 0.7387984395027161,
+      "step": 2878
+    },
+    {
+      "epoch": 1.2151898734177216,
+      "grad_norm": 0.432640939950943,
+      "learning_rate": 0.00019138642522964074,
+      "loss": 0.7597174644470215,
+      "step": 2880
+    },
+    {
+      "epoch": 1.2160337552742617,
+      "grad_norm": 0.432477205991745,
+      "learning_rate": 0.00019136691386404032,
+      "loss": 0.7884587049484253,
+      "step": 2882
+    },
+    {
+      "epoch": 1.2168776371308017,
+      "grad_norm": 0.39305904507637024,
+      "learning_rate": 0.00019134738142184763,
+      "loss": 0.6873269081115723,
+      "step": 2884
+    },
+    {
+      "epoch": 1.2177215189873418,
+      "grad_norm": 0.4485352635383606,
+      "learning_rate": 0.00019132782790756838,
+      "loss": 0.6523491144180298,
+      "step": 2886
+    },
+    {
+      "epoch": 1.2185654008438818,
+      "grad_norm": 0.4480868875980377,
+      "learning_rate": 0.00019130825332571322,
+      "loss": 0.7383941411972046,
+      "step": 2888
+    },
+    {
+      "epoch": 1.219409282700422,
+      "grad_norm": 0.44085049629211426,
+      "learning_rate": 0.00019128865768079764,
+      "loss": 0.7391335964202881,
+      "step": 2890
+    },
+    {
+      "epoch": 1.220253164556962,
+      "grad_norm": 0.4621105492115021,
+      "learning_rate": 0.00019126904097734198,
+      "loss": 0.7282017469406128,
+      "step": 2892
+    },
+    {
+      "epoch": 1.221097046413502,
+      "grad_norm": 0.41478198766708374,
+      "learning_rate": 0.00019124940321987136,
+      "loss": 0.7554038763046265,
+      "step": 2894
+    },
+    {
+      "epoch": 1.2219409282700422,
+      "grad_norm": 0.4143751263618469,
+      "learning_rate": 0.00019122974441291595,
+      "loss": 0.6826799511909485,
+      "step": 2896
+    },
+    {
+      "epoch": 1.2227848101265824,
+      "grad_norm": 0.45392894744873047,
+      "learning_rate": 0.00019121006456101057,
+      "loss": 0.751000702381134,
+      "step": 2898
+    },
+    {
+      "epoch": 1.2236286919831223,
+      "grad_norm": 0.46946296095848083,
+      "learning_rate": 0.00019119036366869506,
+      "loss": 0.7598516941070557,
+      "step": 2900
+    },
+    {
+      "epoch": 1.2236286919831223,
+      "eval_loss": 0.8077136278152466,
+      "eval_runtime": 685.6042,
+      "eval_samples_per_second": 3.073,
+      "eval_steps_per_second": 3.073,
+      "step": 2900
+    },
+    {
+      "epoch": 1.2244725738396625,
+      "grad_norm": 0.4316656291484833,
+      "learning_rate": 0.00019117064174051395,
+      "loss": 0.7518358826637268,
+      "step": 2902
+    },
+    {
+      "epoch": 1.2253164556962026,
+      "grad_norm": 0.47737377882003784,
+      "learning_rate": 0.0001911508987810168,
+      "loss": 0.7308215498924255,
+      "step": 2904
+    },
+    {
+      "epoch": 1.2261603375527426,
+      "grad_norm": 0.40906062722206116,
+      "learning_rate": 0.00019113113479475783,
+      "loss": 0.6996226906776428,
+      "step": 2906
+    },
+    {
+      "epoch": 1.2270042194092827,
+      "grad_norm": 0.517085075378418,
+      "learning_rate": 0.00019111134978629633,
+      "loss": 0.8364415168762207,
+      "step": 2908
+    },
+    {
+      "epoch": 1.2278481012658227,
+      "grad_norm": 0.425581693649292,
+      "learning_rate": 0.00019109154376019624,
+      "loss": 0.722011923789978,
+      "step": 2910
+    },
+    {
+      "epoch": 1.2286919831223628,
+      "grad_norm": 0.4651593565940857,
+      "learning_rate": 0.00019107171672102643,
+      "loss": 0.7586172223091125,
+      "step": 2912
+    },
+    {
+      "epoch": 1.229535864978903,
+      "grad_norm": 0.4578891396522522,
+      "learning_rate": 0.00019105186867336067,
+      "loss": 0.7387955188751221,
+      "step": 2914
+    },
+    {
+      "epoch": 1.230379746835443,
+      "grad_norm": 0.4312961995601654,
+      "learning_rate": 0.00019103199962177748,
+      "loss": 0.8036285638809204,
+      "step": 2916
+    },
+    {
+      "epoch": 1.231223628691983,
+      "grad_norm": 0.4355425536632538,
+      "learning_rate": 0.00019101210957086026,
+      "loss": 0.7356550097465515,
+      "step": 2918
+    },
+    {
+      "epoch": 1.2320675105485233,
+      "grad_norm": 0.424675852060318,
+      "learning_rate": 0.00019099219852519733,
+      "loss": 0.6730139255523682,
+      "step": 2920
+    },
+    {
+      "epoch": 1.2329113924050632,
+      "grad_norm": 0.41102367639541626,
+      "learning_rate": 0.0001909722664893817,
+      "loss": 0.6983892321586609,
+      "step": 2922
+    },
+    {
+      "epoch": 1.2337552742616034,
+      "grad_norm": 0.4088250398635864,
+      "learning_rate": 0.00019095231346801132,
+      "loss": 0.6297144889831543,
+      "step": 2924
+    },
+    {
+      "epoch": 1.2345991561181435,
+      "grad_norm": 0.3685890734195709,
+      "learning_rate": 0.00019093233946568906,
+      "loss": 0.7035690546035767,
+      "step": 2926
+    },
+    {
+      "epoch": 1.2354430379746835,
+      "grad_norm": 0.39438629150390625,
+      "learning_rate": 0.0001909123444870224,
+      "loss": 0.820853054523468,
+      "step": 2928
+    },
+    {
+      "epoch": 1.2362869198312236,
+      "grad_norm": 0.5041676163673401,
+      "learning_rate": 0.00019089232853662392,
+      "loss": 0.7884082794189453,
+      "step": 2930
+    },
+    {
+      "epoch": 1.2371308016877638,
+      "grad_norm": 0.40592801570892334,
+      "learning_rate": 0.00019087229161911086,
+      "loss": 0.773978590965271,
+      "step": 2932
+    },
+    {
+      "epoch": 1.2379746835443037,
+      "grad_norm": 0.41828030347824097,
+      "learning_rate": 0.00019085223373910536,
+      "loss": 0.6742141842842102,
+      "step": 2934
+    },
+    {
+      "epoch": 1.238818565400844,
+      "grad_norm": 0.39984026551246643,
+      "learning_rate": 0.0001908321549012344,
+      "loss": 0.7106125950813293,
+      "step": 2936
+    },
+    {
+      "epoch": 1.239662447257384,
+      "grad_norm": 0.4841296970844269,
+      "learning_rate": 0.00019081205511012973,
+      "loss": 0.7390915751457214,
+      "step": 2938
+    },
+    {
+      "epoch": 1.240506329113924,
+      "grad_norm": 0.39837637543678284,
+      "learning_rate": 0.00019079193437042806,
+      "loss": 0.6527198553085327,
+      "step": 2940
+    },
+    {
+      "epoch": 1.2413502109704642,
+      "grad_norm": 0.40916863083839417,
+      "learning_rate": 0.0001907717926867708,
+      "loss": 0.6939712762832642,
+      "step": 2942
+    },
+    {
+      "epoch": 1.2421940928270043,
+      "grad_norm": 0.4329306483268738,
+      "learning_rate": 0.0001907516300638043,
+      "loss": 0.7566131353378296,
+      "step": 2944
+    },
+    {
+      "epoch": 1.2430379746835443,
+      "grad_norm": 0.47983601689338684,
+      "learning_rate": 0.00019073144650617963,
+      "loss": 0.7893748879432678,
+      "step": 2946
+    },
+    {
+      "epoch": 1.2438818565400844,
+      "grad_norm": 0.4368246793746948,
+      "learning_rate": 0.0001907112420185528,
+      "loss": 0.8259525299072266,
+      "step": 2948
+    },
+    {
+      "epoch": 1.2447257383966246,
+      "grad_norm": 0.41272372007369995,
+      "learning_rate": 0.0001906910166055846,
+      "loss": 0.734692394733429,
+      "step": 2950
+    },
+    {
+      "epoch": 1.2455696202531645,
+      "grad_norm": 0.36929839849472046,
+      "learning_rate": 0.00019067077027194057,
+      "loss": 0.7091750502586365,
+      "step": 2952
+    },
+    {
+      "epoch": 1.2464135021097047,
+      "grad_norm": 0.4395252466201782,
+      "learning_rate": 0.00019065050302229125,
+      "loss": 0.6960210204124451,
+      "step": 2954
+    },
+    {
+      "epoch": 1.2472573839662446,
+      "grad_norm": 0.46784868836402893,
+      "learning_rate": 0.00019063021486131186,
+      "loss": 0.7817115187644958,
+      "step": 2956
+    },
+    {
+      "epoch": 1.2481012658227848,
+      "grad_norm": 0.4191588759422302,
+      "learning_rate": 0.00019060990579368244,
+      "loss": 0.7933676242828369,
+      "step": 2958
+    },
+    {
+      "epoch": 1.248945147679325,
+      "grad_norm": 0.3967524468898773,
+      "learning_rate": 0.000190589575824088,
+      "loss": 0.6856688857078552,
+      "step": 2960
+    },
+    {
+      "epoch": 1.249789029535865,
+      "grad_norm": 0.40346091985702515,
+      "learning_rate": 0.0001905692249572182,
+      "loss": 0.6783818602561951,
+      "step": 2962
+    },
+    {
+      "epoch": 1.250632911392405,
+      "grad_norm": 0.38349980115890503,
+      "learning_rate": 0.0001905488531977676,
+      "loss": 0.6997019052505493,
+      "step": 2964
+    },
+    {
+      "epoch": 1.2514767932489452,
+      "grad_norm": 0.44021397829055786,
+      "learning_rate": 0.0001905284605504356,
+      "loss": 0.6978800892829895,
+      "step": 2966
+    },
+    {
+      "epoch": 1.2523206751054852,
+      "grad_norm": 0.47491583228111267,
+      "learning_rate": 0.0001905080470199264,
+      "loss": 0.7007693648338318,
+      "step": 2968
+    },
+    {
+      "epoch": 1.2531645569620253,
+      "grad_norm": 0.4011451005935669,
+      "learning_rate": 0.00019048761261094898,
+      "loss": 0.7258896231651306,
+      "step": 2970
+    },
+    {
+      "epoch": 1.2540084388185653,
+      "grad_norm": 0.44206640124320984,
+      "learning_rate": 0.0001904671573282172,
+      "loss": 0.6746299266815186,
+      "step": 2972
+    },
+    {
+      "epoch": 1.2548523206751054,
+      "grad_norm": 0.4532456696033478,
+      "learning_rate": 0.00019044668117644965,
+      "loss": 0.7638527154922485,
+      "step": 2974
+    },
+    {
+      "epoch": 1.2556962025316456,
+      "grad_norm": 0.44301095604896545,
+      "learning_rate": 0.00019042618416036984,
+      "loss": 0.659598171710968,
+      "step": 2976
+    },
+    {
+      "epoch": 1.2565400843881855,
+      "grad_norm": 0.37427952885627747,
+      "learning_rate": 0.00019040566628470597,
+      "loss": 0.7132709622383118,
+      "step": 2978
+    },
+    {
+      "epoch": 1.2573839662447257,
+      "grad_norm": 0.4195408821105957,
+      "learning_rate": 0.0001903851275541912,
+      "loss": 0.8163022398948669,
+      "step": 2980
+    },
+    {
+      "epoch": 1.2582278481012659,
+      "grad_norm": 0.3951880633831024,
+      "learning_rate": 0.00019036456797356336,
+      "loss": 0.7481505870819092,
+      "step": 2982
+    },
+    {
+      "epoch": 1.2590717299578058,
+      "grad_norm": 0.4239100217819214,
+      "learning_rate": 0.00019034398754756522,
+      "loss": 0.750126838684082,
+      "step": 2984
+    },
+    {
+      "epoch": 1.259915611814346,
+      "grad_norm": 0.4209354817867279,
+      "learning_rate": 0.0001903233862809442,
+      "loss": 0.746433436870575,
+      "step": 2986
+    },
+    {
+      "epoch": 1.2607594936708861,
+      "grad_norm": 0.4188138544559479,
+      "learning_rate": 0.00019030276417845266,
+      "loss": 0.7104454040527344,
+      "step": 2988
+    },
+    {
+      "epoch": 1.261603375527426,
+      "grad_norm": 0.4531371593475342,
+      "learning_rate": 0.00019028212124484772,
+      "loss": 0.650738000869751,
+      "step": 2990
+    },
+    {
+      "epoch": 1.2624472573839662,
+      "grad_norm": 0.4672204256057739,
+      "learning_rate": 0.00019026145748489133,
+      "loss": 0.6870551109313965,
+      "step": 2992
+    },
+    {
+      "epoch": 1.2632911392405064,
+      "grad_norm": 0.40681278705596924,
+      "learning_rate": 0.00019024077290335018,
+      "loss": 0.6879403591156006,
+      "step": 2994
+    },
+    {
+      "epoch": 1.2641350210970463,
+      "grad_norm": 0.4491691291332245,
+      "learning_rate": 0.00019022006750499585,
+      "loss": 0.6886372566223145,
+      "step": 2996
+    },
+    {
+      "epoch": 1.2649789029535865,
+      "grad_norm": 0.4144807457923889,
+      "learning_rate": 0.0001901993412946046,
+      "loss": 0.7199324369430542,
+      "step": 2998
+    },
+    {
+      "epoch": 1.2658227848101267,
+      "grad_norm": 0.40370482206344604,
+      "learning_rate": 0.0001901785942769577,
+      "loss": 0.7105490565299988,
+      "step": 3000
+    },
+    {
+      "epoch": 1.2658227848101267,
+      "eval_loss": 0.8033165335655212,
+      "eval_runtime": 678.7554,
+      "eval_samples_per_second": 3.104,
+      "eval_steps_per_second": 3.104,
+      "step": 3000
+    },
+    {
+      "epoch": 1.2666666666666666,
+      "grad_norm": 0.42637890577316284,
+      "learning_rate": 0.00019015782645684097,
+      "loss": 0.7385403513908386,
+      "step": 3002
+    },
+    {
+      "epoch": 1.2675105485232068,
+      "grad_norm": 0.3709809482097626,
+      "learning_rate": 0.00019013703783904518,
+      "loss": 0.6395129561424255,
+      "step": 3004
+    },
+    {
+      "epoch": 1.268354430379747,
+      "grad_norm": 0.4642501175403595,
+      "learning_rate": 0.00019011622842836592,
+      "loss": 0.7123657464981079,
+      "step": 3006
+    },
+    {
+      "epoch": 1.2691983122362869,
+      "grad_norm": 0.4553911089897156,
+      "learning_rate": 0.00019009539822960342,
+      "loss": 0.744048535823822,
+      "step": 3008
+    },
+    {
+      "epoch": 1.270042194092827,
+      "grad_norm": 0.39334386587142944,
+      "learning_rate": 0.0001900745472475629,
+      "loss": 0.7181914448738098,
+      "step": 3010
+    },
+    {
+      "epoch": 1.2708860759493672,
+      "grad_norm": 0.40902766585350037,
+      "learning_rate": 0.00019005367548705426,
+      "loss": 0.7238107919692993,
+      "step": 3012
+    },
+    {
+      "epoch": 1.2717299578059071,
+      "grad_norm": 0.48296019434928894,
+      "learning_rate": 0.00019003278295289216,
+      "loss": 0.7479744553565979,
+      "step": 3014
+    },
+    {
+      "epoch": 1.2725738396624473,
+      "grad_norm": 0.4807029962539673,
+      "learning_rate": 0.00019001186964989617,
+      "loss": 0.829924464225769,
+      "step": 3016
+    },
+    {
+      "epoch": 1.2734177215189875,
+      "grad_norm": 0.4239083528518677,
+      "learning_rate": 0.00018999093558289055,
+      "loss": 0.7204707264900208,
+      "step": 3018
+    },
+    {
+      "epoch": 1.2742616033755274,
+      "grad_norm": 0.4687928259372711,
+      "learning_rate": 0.0001899699807567044,
+      "loss": 0.7668164968490601,
+      "step": 3020
+    },
+    {
+      "epoch": 1.2751054852320676,
+      "grad_norm": 0.41724732518196106,
+      "learning_rate": 0.0001899490051761716,
+      "loss": 0.766342043876648,
+      "step": 3022
+    },
+    {
+      "epoch": 1.2759493670886077,
+      "grad_norm": 0.4131613075733185,
+      "learning_rate": 0.00018992800884613082,
+      "loss": 0.6595507860183716,
+      "step": 3024
+    },
+    {
+      "epoch": 1.2767932489451477,
+      "grad_norm": 0.4156053364276886,
+      "learning_rate": 0.00018990699177142548,
+      "loss": 0.7419850826263428,
+      "step": 3026
+    },
+    {
+      "epoch": 1.2776371308016878,
+      "grad_norm": 0.41845977306365967,
+      "learning_rate": 0.00018988595395690388,
+      "loss": 0.67264324426651,
+      "step": 3028
+    },
+    {
+      "epoch": 1.2784810126582278,
+      "grad_norm": 0.3826757073402405,
+      "learning_rate": 0.00018986489540741893,
+      "loss": 0.6269281506538391,
+      "step": 3030
+    },
+    {
+      "epoch": 1.279324894514768,
+      "grad_norm": 0.4531428813934326,
+      "learning_rate": 0.00018984381612782854,
+      "loss": 0.71211177110672,
+      "step": 3032
+    },
+    {
+      "epoch": 1.2801687763713079,
+      "grad_norm": 0.3747255802154541,
+      "learning_rate": 0.00018982271612299524,
+      "loss": 0.635804295539856,
+      "step": 3034
+    },
+    {
+      "epoch": 1.281012658227848,
+      "grad_norm": 0.4079552888870239,
+      "learning_rate": 0.00018980159539778638,
+      "loss": 0.7269464731216431,
+      "step": 3036
+    },
+    {
+      "epoch": 1.2818565400843882,
+      "grad_norm": 0.4193854033946991,
+      "learning_rate": 0.00018978045395707418,
+      "loss": 0.7737077474594116,
+      "step": 3038
+    },
+    {
+      "epoch": 1.2827004219409281,
+      "grad_norm": 0.4182516038417816,
+      "learning_rate": 0.0001897592918057355,
+      "loss": 0.7458174824714661,
+      "step": 3040
+    },
+    {
+      "epoch": 1.2835443037974683,
+      "grad_norm": 0.43343567848205566,
+      "learning_rate": 0.00018973810894865205,
+      "loss": 0.6565808057785034,
+      "step": 3042
+    },
+    {
+      "epoch": 1.2843881856540085,
+      "grad_norm": 0.4294503331184387,
+      "learning_rate": 0.00018971690539071033,
+      "loss": 0.736944317817688,
+      "step": 3044
+    },
+    {
+      "epoch": 1.2852320675105484,
+      "grad_norm": 0.41330763697624207,
+      "learning_rate": 0.0001896956811368016,
+      "loss": 0.6921742558479309,
+      "step": 3046
+    },
+    {
+      "epoch": 1.2860759493670886,
+      "grad_norm": 0.44992703199386597,
+      "learning_rate": 0.0001896744361918218,
+      "loss": 0.7004357576370239,
+      "step": 3048
+    },
+    {
+      "epoch": 1.2869198312236287,
+      "grad_norm": 0.43748870491981506,
+      "learning_rate": 0.0001896531705606719,
+      "loss": 0.6767720580101013,
+      "step": 3050
+    },
+    {
+      "epoch": 1.2877637130801687,
+      "grad_norm": 0.47581973671913147,
+      "learning_rate": 0.0001896318842482573,
+      "loss": 0.7596892714500427,
+      "step": 3052
+    },
+    {
+      "epoch": 1.2886075949367088,
+      "grad_norm": 0.45806005597114563,
+      "learning_rate": 0.00018961057725948844,
+      "loss": 0.7650044560432434,
+      "step": 3054
+    },
+    {
+      "epoch": 1.289451476793249,
+      "grad_norm": 0.4291481375694275,
+      "learning_rate": 0.00018958924959928042,
+      "loss": 0.6654610633850098,
+      "step": 3056
+    },
+    {
+      "epoch": 1.290295358649789,
+      "grad_norm": 0.42454418540000916,
+      "learning_rate": 0.00018956790127255309,
+      "loss": 0.7079797387123108,
+      "step": 3058
+    },
+    {
+      "epoch": 1.2911392405063291,
+      "grad_norm": 0.46820637583732605,
+      "learning_rate": 0.00018954653228423114,
+      "loss": 0.7724313735961914,
+      "step": 3060
+    },
+    {
+      "epoch": 1.2919831223628693,
+      "grad_norm": 0.4440016746520996,
+      "learning_rate": 0.00018952514263924396,
+      "loss": 0.7746338248252869,
+      "step": 3062
+    },
+    {
+      "epoch": 1.2928270042194092,
+      "grad_norm": 0.4296419620513916,
+      "learning_rate": 0.00018950373234252572,
+      "loss": 0.7570929527282715,
+      "step": 3064
+    },
+    {
+      "epoch": 1.2936708860759494,
+      "grad_norm": 0.428780734539032,
+      "learning_rate": 0.0001894823013990154,
+      "loss": 0.7274802923202515,
+      "step": 3066
+    },
+    {
+      "epoch": 1.2945147679324895,
+      "grad_norm": 0.4111585021018982,
+      "learning_rate": 0.0001894608498136567,
+      "loss": 0.6595024466514587,
+      "step": 3068
+    },
+    {
+      "epoch": 1.2953586497890295,
+      "grad_norm": 0.4373013973236084,
+      "learning_rate": 0.00018943937759139802,
+      "loss": 0.7316440939903259,
+      "step": 3070
+    },
+    {
+      "epoch": 1.2962025316455696,
+      "grad_norm": 0.37705400586128235,
+      "learning_rate": 0.0001894178847371927,
+      "loss": 0.7072293162345886,
+      "step": 3072
+    },
+    {
+      "epoch": 1.2970464135021098,
+      "grad_norm": 0.4171125888824463,
+      "learning_rate": 0.00018939637125599864,
+      "loss": 0.7218101024627686,
+      "step": 3074
+    },
+    {
+      "epoch": 1.2978902953586497,
+      "grad_norm": 0.4264235198497772,
+      "learning_rate": 0.00018937483715277865,
+      "loss": 0.725140392780304,
+      "step": 3076
+    },
+    {
+      "epoch": 1.29873417721519,
+      "grad_norm": 0.4251643121242523,
+      "learning_rate": 0.0001893532824325002,
+      "loss": 0.7423256635665894,
+      "step": 3078
+    },
+    {
+      "epoch": 1.29957805907173,
+      "grad_norm": 0.43096452951431274,
+      "learning_rate": 0.00018933170710013554,
+      "loss": 0.7058811187744141,
+      "step": 3080
+    },
+    {
+      "epoch": 1.30042194092827,
+      "grad_norm": 0.466804176568985,
+      "learning_rate": 0.00018931011116066172,
+      "loss": 0.7334373593330383,
+      "step": 3082
+    },
+    {
+      "epoch": 1.3012658227848102,
+      "grad_norm": 0.36896654963493347,
+      "learning_rate": 0.00018928849461906047,
+      "loss": 0.6033462285995483,
+      "step": 3084
+    },
+    {
+      "epoch": 1.3021097046413503,
+      "grad_norm": 0.42875027656555176,
+      "learning_rate": 0.00018926685748031833,
+      "loss": 0.690375566482544,
+      "step": 3086
+    },
+    {
+      "epoch": 1.3029535864978903,
+      "grad_norm": 0.4041171967983246,
+      "learning_rate": 0.00018924519974942658,
+      "loss": 0.7468435168266296,
+      "step": 3088
+    },
+    {
+      "epoch": 1.3037974683544304,
+      "grad_norm": 0.4704097807407379,
+      "learning_rate": 0.00018922352143138126,
+      "loss": 0.7574344873428345,
+      "step": 3090
+    },
+    {
+      "epoch": 1.3046413502109704,
+      "grad_norm": 0.4391418397426605,
+      "learning_rate": 0.0001892018225311831,
+      "loss": 0.756452739238739,
+      "step": 3092
+    },
+    {
+      "epoch": 1.3054852320675105,
+      "grad_norm": 0.42684388160705566,
+      "learning_rate": 0.0001891801030538376,
+      "loss": 0.6620991230010986,
+      "step": 3094
+    },
+    {
+      "epoch": 1.3063291139240507,
+      "grad_norm": 0.4972277283668518,
+      "learning_rate": 0.00018915836300435507,
+      "loss": 0.7764125466346741,
+      "step": 3096
+    },
+    {
+      "epoch": 1.3071729957805907,
+      "grad_norm": 0.3905445337295532,
+      "learning_rate": 0.00018913660238775053,
+      "loss": 0.6557663083076477,
+      "step": 3098
+    },
+    {
+      "epoch": 1.3080168776371308,
+      "grad_norm": 0.4553062319755554,
+      "learning_rate": 0.0001891148212090437,
+      "loss": 0.6746718883514404,
+      "step": 3100
+    },
+    {
+      "epoch": 1.3080168776371308,
+      "eval_loss": 0.8010181784629822,
+      "eval_runtime": 668.1688,
+      "eval_samples_per_second": 3.153,
+      "eval_steps_per_second": 3.153,
+      "step": 3100
+    },
+    {
+      "epoch": 1.3088607594936708,
+      "grad_norm": 0.4254429340362549,
+      "learning_rate": 0.00018909301947325915,
+      "loss": 0.7916419506072998,
+      "step": 3102
+    },
+    {
+      "epoch": 1.309704641350211,
+      "grad_norm": 0.6118508577346802,
+      "learning_rate": 0.00018907119718542602,
+      "loss": 0.883936882019043,
+      "step": 3104
+    },
+    {
+      "epoch": 1.310548523206751,
+      "grad_norm": 0.43206971883773804,
+      "learning_rate": 0.00018904935435057837,
+      "loss": 0.7145928740501404,
+      "step": 3106
+    },
+    {
+      "epoch": 1.311392405063291,
+      "grad_norm": 0.39657703042030334,
+      "learning_rate": 0.0001890274909737549,
+      "loss": 0.7442842125892639,
+      "step": 3108
+    },
+    {
+      "epoch": 1.3122362869198312,
+      "grad_norm": 0.4087927043437958,
+      "learning_rate": 0.00018900560705999904,
+      "loss": 0.6815276145935059,
+      "step": 3110
+    },
+    {
+      "epoch": 1.3130801687763713,
+      "grad_norm": 0.430259108543396,
+      "learning_rate": 0.00018898370261435904,
+      "loss": 0.8153640627861023,
+      "step": 3112
+    },
+    {
+      "epoch": 1.3139240506329113,
+      "grad_norm": 0.47486868500709534,
+      "learning_rate": 0.0001889617776418878,
+      "loss": 0.7828266024589539,
+      "step": 3114
+    },
+    {
+      "epoch": 1.3147679324894515,
+      "grad_norm": 0.4929087162017822,
+      "learning_rate": 0.000188939832147643,
+      "loss": 0.7849065661430359,
+      "step": 3116
+    },
+    {
+      "epoch": 1.3156118143459916,
+      "grad_norm": 0.39038679003715515,
+      "learning_rate": 0.00018891786613668704,
+      "loss": 0.7262985706329346,
+      "step": 3118
+    },
+    {
+      "epoch": 1.3164556962025316,
+      "grad_norm": 0.4827214479446411,
+      "learning_rate": 0.0001888958796140871,
+      "loss": 0.8121856451034546,
+      "step": 3120
+    },
+    {
+      "epoch": 1.3172995780590717,
+      "grad_norm": 0.47739601135253906,
+      "learning_rate": 0.000188873872584915,
+      "loss": 0.7165566086769104,
+      "step": 3122
+    },
+    {
+      "epoch": 1.3181434599156119,
+      "grad_norm": 0.4274037778377533,
+      "learning_rate": 0.0001888518450542473,
+      "loss": 0.6698412299156189,
+      "step": 3124
+    },
+    {
+      "epoch": 1.3189873417721518,
+      "grad_norm": 0.44825276732444763,
+      "learning_rate": 0.00018882979702716544,
+      "loss": 0.7271128296852112,
+      "step": 3126
+    },
+    {
+      "epoch": 1.319831223628692,
+      "grad_norm": 0.42956918478012085,
+      "learning_rate": 0.00018880772850875537,
+      "loss": 0.7924301624298096,
+      "step": 3128
+    },
+    {
+      "epoch": 1.3206751054852321,
+      "grad_norm": 0.4848136305809021,
+      "learning_rate": 0.00018878563950410792,
+      "loss": 0.7846497893333435,
+      "step": 3130
+    },
+    {
+      "epoch": 1.321518987341772,
+      "grad_norm": 0.46651726961135864,
+      "learning_rate": 0.00018876353001831864,
+      "loss": 0.7262214422225952,
+      "step": 3132
+    },
+    {
+      "epoch": 1.3223628691983123,
+      "grad_norm": 0.44598284363746643,
+      "learning_rate": 0.0001887414000564877,
+      "loss": 0.7640091180801392,
+      "step": 3134
+    },
+    {
+      "epoch": 1.3232067510548524,
+      "grad_norm": 0.43744903802871704,
+      "learning_rate": 0.00018871924962372006,
+      "loss": 0.6960080862045288,
+      "step": 3136
+    },
+    {
+      "epoch": 1.3240506329113924,
+      "grad_norm": 0.4737822413444519,
+      "learning_rate": 0.00018869707872512544,
+      "loss": 0.7121244072914124,
+      "step": 3138
+    },
+    {
+      "epoch": 1.3248945147679325,
+      "grad_norm": 0.44452038407325745,
+      "learning_rate": 0.00018867488736581817,
+      "loss": 0.713148295879364,
+      "step": 3140
+    },
+    {
+      "epoch": 1.3257383966244727,
+      "grad_norm": 0.47142693400382996,
+      "learning_rate": 0.0001886526755509175,
+      "loss": 0.6904974579811096,
+      "step": 3142
+    },
+    {
+      "epoch": 1.3265822784810126,
+      "grad_norm": 0.40572887659072876,
+      "learning_rate": 0.00018863044328554712,
+      "loss": 0.7319386005401611,
+      "step": 3144
+    },
+    {
+      "epoch": 1.3274261603375528,
+      "grad_norm": 0.44173532724380493,
+      "learning_rate": 0.00018860819057483566,
+      "loss": 0.7268120646476746,
+      "step": 3146
+    },
+    {
+      "epoch": 1.328270042194093,
+      "grad_norm": 0.43025046586990356,
+      "learning_rate": 0.00018858591742391642,
+      "loss": 0.7129524946212769,
+      "step": 3148
+    },
+    {
+      "epoch": 1.3291139240506329,
+      "grad_norm": 0.41418036818504333,
+      "learning_rate": 0.00018856362383792732,
+      "loss": 0.6917402148246765,
+      "step": 3150
+    },
+    {
+      "epoch": 1.329957805907173,
+      "grad_norm": 0.46275854110717773,
+      "learning_rate": 0.00018854130982201111,
+      "loss": 0.7653918862342834,
+      "step": 3152
+    },
+    {
+      "epoch": 1.3308016877637132,
+      "grad_norm": 0.38927653431892395,
+      "learning_rate": 0.0001885189753813152,
+      "loss": 0.7311477065086365,
+      "step": 3154
+    },
+    {
+      "epoch": 1.3316455696202532,
+      "grad_norm": 0.44294974207878113,
+      "learning_rate": 0.0001884966205209917,
+      "loss": 0.7978816628456116,
+      "step": 3156
+    },
+    {
+      "epoch": 1.3324894514767933,
+      "grad_norm": 0.3772844970226288,
+      "learning_rate": 0.00018847424524619745,
+      "loss": 0.7497738599777222,
+      "step": 3158
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.4538787603378296,
+      "learning_rate": 0.000188451849562094,
+      "loss": 0.712806761264801,
+      "step": 3160
+    },
+    {
+      "epoch": 1.3341772151898734,
+      "grad_norm": 0.4194336533546448,
+      "learning_rate": 0.00018842943347384763,
+      "loss": 0.7282580137252808,
+      "step": 3162
+    },
+    {
+      "epoch": 1.3350210970464136,
+      "grad_norm": 0.4186438024044037,
+      "learning_rate": 0.00018840699698662925,
+      "loss": 0.797752320766449,
+      "step": 3164
+    },
+    {
+      "epoch": 1.3358649789029535,
+      "grad_norm": 0.47711506485939026,
+      "learning_rate": 0.00018838454010561457,
+      "loss": 0.8175379633903503,
+      "step": 3166
+    },
+    {
+      "epoch": 1.3367088607594937,
+      "grad_norm": 0.4380176067352295,
+      "learning_rate": 0.00018836206283598394,
+      "loss": 0.659176766872406,
+      "step": 3168
+    },
+    {
+      "epoch": 1.3375527426160336,
+      "grad_norm": 0.4505249559879303,
+      "learning_rate": 0.00018833956518292245,
+      "loss": 0.732314944267273,
+      "step": 3170
+    },
+    {
+      "epoch": 1.3383966244725738,
+      "grad_norm": 0.3832913935184479,
+      "learning_rate": 0.00018831704715161984,
+      "loss": 0.6791377067565918,
+      "step": 3172
+    },
+    {
+      "epoch": 1.339240506329114,
+      "grad_norm": 0.42996498942375183,
+      "learning_rate": 0.00018829450874727064,
+      "loss": 0.7395796775817871,
+      "step": 3174
+    },
+    {
+      "epoch": 1.340084388185654,
+      "grad_norm": 0.3926190435886383,
+      "learning_rate": 0.000188271949975074,
+      "loss": 0.6893631219863892,
+      "step": 3176
+    },
+    {
+      "epoch": 1.340928270042194,
+      "grad_norm": 0.4353938102722168,
+      "learning_rate": 0.0001882493708402338,
+      "loss": 0.6708382964134216,
+      "step": 3178
+    },
+    {
+      "epoch": 1.3417721518987342,
+      "grad_norm": 0.43043214082717896,
+      "learning_rate": 0.0001882267713479586,
+      "loss": 0.8447695374488831,
+      "step": 3180
+    },
+    {
+      "epoch": 1.3426160337552742,
+      "grad_norm": 0.40693536400794983,
+      "learning_rate": 0.00018820415150346163,
+      "loss": 0.7081091403961182,
+      "step": 3182
+    },
+    {
+      "epoch": 1.3434599156118143,
+      "grad_norm": 0.45037057995796204,
+      "learning_rate": 0.00018818151131196097,
+      "loss": 0.6822336316108704,
+      "step": 3184
+    },
+    {
+      "epoch": 1.3443037974683545,
+      "grad_norm": 0.3731948137283325,
+      "learning_rate": 0.00018815885077867915,
+      "loss": 0.6316909790039062,
+      "step": 3186
+    },
+    {
+      "epoch": 1.3451476793248944,
+      "grad_norm": 0.4248095750808716,
+      "learning_rate": 0.00018813616990884363,
+      "loss": 0.7452375292778015,
+      "step": 3188
+    },
+    {
+      "epoch": 1.3459915611814346,
+      "grad_norm": 0.39229142665863037,
+      "learning_rate": 0.00018811346870768638,
+      "loss": 0.6493782997131348,
+      "step": 3190
+    },
+    {
+      "epoch": 1.3468354430379748,
+      "grad_norm": 0.45015549659729004,
+      "learning_rate": 0.00018809074718044414,
+      "loss": 0.735163688659668,
+      "step": 3192
+    },
+    {
+      "epoch": 1.3476793248945147,
+      "grad_norm": 0.39756354689598083,
+      "learning_rate": 0.00018806800533235836,
+      "loss": 0.6656006574630737,
+      "step": 3194
+    },
+    {
+      "epoch": 1.3485232067510549,
+      "grad_norm": 0.4410010576248169,
+      "learning_rate": 0.0001880452431686751,
+      "loss": 0.7040987014770508,
+      "step": 3196
+    },
+    {
+      "epoch": 1.349367088607595,
+      "grad_norm": 0.48144352436065674,
+      "learning_rate": 0.0001880224606946452,
+      "loss": 0.7722331285476685,
+      "step": 3198
+    },
+    {
+      "epoch": 1.350210970464135,
+      "grad_norm": 0.38644957542419434,
+      "learning_rate": 0.00018799965791552414,
+      "loss": 0.6681689023971558,
+      "step": 3200
+    },
+    {
+      "epoch": 1.350210970464135,
+      "eval_loss": 0.797160804271698,
+      "eval_runtime": 680.976,
+      "eval_samples_per_second": 3.094,
+      "eval_steps_per_second": 3.094,
+      "step": 3200
+    },
+    {
+      "epoch": 1.3510548523206751,
+      "grad_norm": 0.4471014440059662,
+      "learning_rate": 0.00018797683483657201,
+      "loss": 0.6995527744293213,
+      "step": 3202
+    },
+    {
+      "epoch": 1.3518987341772153,
+      "grad_norm": 0.4065002202987671,
+      "learning_rate": 0.00018795399146305378,
+      "loss": 0.659622848033905,
+      "step": 3204
+    },
+    {
+      "epoch": 1.3527426160337552,
+      "grad_norm": 0.4787910580635071,
+      "learning_rate": 0.00018793112780023888,
+      "loss": 0.7927116751670837,
+      "step": 3206
+    },
+    {
+      "epoch": 1.3535864978902954,
+      "grad_norm": 0.43351876735687256,
+      "learning_rate": 0.0001879082438534015,
+      "loss": 0.6738835573196411,
+      "step": 3208
+    },
+    {
+      "epoch": 1.3544303797468356,
+      "grad_norm": 0.4275745749473572,
+      "learning_rate": 0.00018788533962782063,
+      "loss": 0.7785354852676392,
+      "step": 3210
+    },
+    {
+      "epoch": 1.3552742616033755,
+      "grad_norm": 0.4506145119667053,
+      "learning_rate": 0.00018786241512877975,
+      "loss": 0.7325424551963806,
+      "step": 3212
+    },
+    {
+      "epoch": 1.3561181434599157,
+      "grad_norm": 0.42040735483169556,
+      "learning_rate": 0.00018783947036156713,
+      "loss": 0.6898744702339172,
+      "step": 3214
+    },
+    {
+      "epoch": 1.3569620253164558,
+      "grad_norm": 0.4650438129901886,
+      "learning_rate": 0.0001878165053314757,
+      "loss": 0.7498815655708313,
+      "step": 3216
+    },
+    {
+      "epoch": 1.3578059071729958,
+      "grad_norm": 0.4194130599498749,
+      "learning_rate": 0.000187793520043803,
+      "loss": 0.6891883611679077,
+      "step": 3218
+    },
+    {
+      "epoch": 1.358649789029536,
+      "grad_norm": 0.4447539746761322,
+      "learning_rate": 0.0001877705145038514,
+      "loss": 0.753805935382843,
+      "step": 3220
+    },
+    {
+      "epoch": 1.3594936708860759,
+      "grad_norm": 0.4047110974788666,
+      "learning_rate": 0.00018774748871692773,
+      "loss": 0.6476293802261353,
+      "step": 3222
+    },
+    {
+      "epoch": 1.360337552742616,
+      "grad_norm": 0.39818957448005676,
+      "learning_rate": 0.00018772444268834364,
+      "loss": 0.6765931844711304,
+      "step": 3224
+    },
+    {
+      "epoch": 1.3611814345991562,
+      "grad_norm": 0.41843733191490173,
+      "learning_rate": 0.0001877013764234154,
+      "loss": 0.7360382676124573,
+      "step": 3226
+    },
+    {
+      "epoch": 1.3620253164556961,
+      "grad_norm": 0.41397789120674133,
+      "learning_rate": 0.00018767828992746394,
+      "loss": 0.7149524688720703,
+      "step": 3228
+    },
+    {
+      "epoch": 1.3628691983122363,
+      "grad_norm": 0.43585047125816345,
+      "learning_rate": 0.00018765518320581492,
+      "loss": 0.6932925581932068,
+      "step": 3230
+    },
+    {
+      "epoch": 1.3637130801687762,
+      "grad_norm": 0.3651057481765747,
+      "learning_rate": 0.00018763205626379857,
+      "loss": 0.6561182141304016,
+      "step": 3232
+    },
+    {
+      "epoch": 1.3645569620253164,
+      "grad_norm": 0.44937077164649963,
+      "learning_rate": 0.00018760890910674984,
+      "loss": 0.7436831593513489,
+      "step": 3234
+    },
+    {
+      "epoch": 1.3654008438818566,
+      "grad_norm": 0.5045920014381409,
+      "learning_rate": 0.00018758574174000832,
+      "loss": 0.7849522233009338,
+      "step": 3236
+    },
+    {
+      "epoch": 1.3662447257383965,
+      "grad_norm": 0.4625580906867981,
+      "learning_rate": 0.00018756255416891833,
+      "loss": 0.7501618266105652,
+      "step": 3238
+    },
+    {
+      "epoch": 1.3670886075949367,
+      "grad_norm": 0.3721269369125366,
+      "learning_rate": 0.00018753934639882875,
+      "loss": 0.7656083106994629,
+      "step": 3240
+    },
+    {
+      "epoch": 1.3679324894514768,
+      "grad_norm": 0.41721680760383606,
+      "learning_rate": 0.00018751611843509318,
+      "loss": 0.7816168069839478,
+      "step": 3242
+    },
+    {
+      "epoch": 1.3687763713080168,
+      "grad_norm": 0.8452199101448059,
+      "learning_rate": 0.00018749287028306988,
+      "loss": 0.7002755403518677,
+      "step": 3244
+    },
+    {
+      "epoch": 1.369620253164557,
+      "grad_norm": 0.4320507347583771,
+      "learning_rate": 0.00018746960194812172,
+      "loss": 0.7358270883560181,
+      "step": 3246
+    },
+    {
+      "epoch": 1.370464135021097,
+      "grad_norm": 0.43323323130607605,
+      "learning_rate": 0.00018744631343561627,
+      "loss": 0.6528644561767578,
+      "step": 3248
+    },
+    {
+      "epoch": 1.371308016877637,
+      "grad_norm": 0.41487041115760803,
+      "learning_rate": 0.0001874230047509258,
+      "loss": 0.7317517995834351,
+      "step": 3250
+    },
+    {
+      "epoch": 1.3721518987341772,
+      "grad_norm": 0.5085384249687195,
+      "learning_rate": 0.00018739967589942707,
+      "loss": 0.8236643671989441,
+      "step": 3252
+    },
+    {
+      "epoch": 1.3729957805907174,
+      "grad_norm": 0.4126952290534973,
+      "learning_rate": 0.00018737632688650168,
+      "loss": 0.730254054069519,
+      "step": 3254
+    },
+    {
+      "epoch": 1.3738396624472573,
+      "grad_norm": 0.5272142291069031,
+      "learning_rate": 0.00018735295771753573,
+      "loss": 0.7637890577316284,
+      "step": 3256
+    },
+    {
+      "epoch": 1.3746835443037975,
+      "grad_norm": 0.4319530725479126,
+      "learning_rate": 0.00018732956839792012,
+      "loss": 0.7906988263130188,
+      "step": 3258
+    },
+    {
+      "epoch": 1.3755274261603376,
+      "grad_norm": 0.4393143355846405,
+      "learning_rate": 0.00018730615893305025,
+      "loss": 0.7261547446250916,
+      "step": 3260
+    },
+    {
+      "epoch": 1.3763713080168776,
+      "grad_norm": 0.4161636233329773,
+      "learning_rate": 0.00018728272932832632,
+      "loss": 0.6857497692108154,
+      "step": 3262
+    },
+    {
+      "epoch": 1.3772151898734177,
+      "grad_norm": 0.48862120509147644,
+      "learning_rate": 0.00018725927958915297,
+      "loss": 0.7488857507705688,
+      "step": 3264
+    },
+    {
+      "epoch": 1.378059071729958,
+      "grad_norm": 0.4422304928302765,
+      "learning_rate": 0.0001872358097209397,
+      "loss": 0.7257721424102783,
+      "step": 3266
+    },
+    {
+      "epoch": 1.3789029535864978,
+      "grad_norm": 0.4577067494392395,
+      "learning_rate": 0.00018721231972910048,
+      "loss": 0.746697187423706,
+      "step": 3268
+    },
+    {
+      "epoch": 1.379746835443038,
+      "grad_norm": 0.42382001876831055,
+      "learning_rate": 0.00018718880961905406,
+      "loss": 0.768519401550293,
+      "step": 3270
+    },
+    {
+      "epoch": 1.3805907172995782,
+      "grad_norm": 0.43368014693260193,
+      "learning_rate": 0.00018716527939622378,
+      "loss": 0.7426102757453918,
+      "step": 3272
+    },
+    {
+      "epoch": 1.381434599156118,
+      "grad_norm": 0.3777220547199249,
+      "learning_rate": 0.00018714172906603757,
+      "loss": 0.7104197144508362,
+      "step": 3274
+    },
+    {
+      "epoch": 1.3822784810126583,
+      "grad_norm": 0.46071696281433105,
+      "learning_rate": 0.00018711815863392802,
+      "loss": 0.7502968907356262,
+      "step": 3276
+    },
+    {
+      "epoch": 1.3831223628691984,
+      "grad_norm": 0.42510855197906494,
+      "learning_rate": 0.00018709456810533245,
+      "loss": 0.6400540471076965,
+      "step": 3278
+    },
+    {
+      "epoch": 1.3839662447257384,
+      "grad_norm": 0.4217228293418884,
+      "learning_rate": 0.00018707095748569268,
+      "loss": 0.6648687720298767,
+      "step": 3280
+    },
+    {
+      "epoch": 1.3848101265822785,
+      "grad_norm": 0.48267799615859985,
+      "learning_rate": 0.00018704732678045526,
+      "loss": 0.7176171541213989,
+      "step": 3282
+    },
+    {
+      "epoch": 1.3856540084388187,
+      "grad_norm": 0.47114965319633484,
+      "learning_rate": 0.00018702367599507134,
+      "loss": 0.8402457237243652,
+      "step": 3284
+    },
+    {
+      "epoch": 1.3864978902953586,
+      "grad_norm": 0.5054492354393005,
+      "learning_rate": 0.00018700000513499667,
+      "loss": 0.8130836486816406,
+      "step": 3286
+    },
+    {
+      "epoch": 1.3873417721518988,
+      "grad_norm": 0.39435645937919617,
+      "learning_rate": 0.00018697631420569168,
+      "loss": 0.6432579755783081,
+      "step": 3288
+    },
+    {
+      "epoch": 1.3881856540084387,
+      "grad_norm": 0.3997228443622589,
+      "learning_rate": 0.00018695260321262142,
+      "loss": 0.6921643614768982,
+      "step": 3290
+    },
+    {
+      "epoch": 1.389029535864979,
+      "grad_norm": 0.37304237484931946,
+      "learning_rate": 0.00018692887216125557,
+      "loss": 0.7162153124809265,
+      "step": 3292
+    },
+    {
+      "epoch": 1.389873417721519,
+      "grad_norm": 0.7997981309890747,
+      "learning_rate": 0.00018690512105706838,
+      "loss": 0.7194803953170776,
+      "step": 3294
+    },
+    {
+      "epoch": 1.390717299578059,
+      "grad_norm": 0.41268959641456604,
+      "learning_rate": 0.00018688134990553883,
+      "loss": 0.6888322830200195,
+      "step": 3296
+    },
+    {
+      "epoch": 1.3915611814345992,
+      "grad_norm": 0.4485826790332794,
+      "learning_rate": 0.00018685755871215043,
+      "loss": 0.7387221455574036,
+      "step": 3298
+    },
+    {
+      "epoch": 1.3924050632911391,
+      "grad_norm": 0.47835296392440796,
+      "learning_rate": 0.0001868337474823914,
+      "loss": 0.8061952590942383,
+      "step": 3300
+    },
+    {
+      "epoch": 1.3924050632911391,
+      "eval_loss": 0.795619547367096,
+      "eval_runtime": 692.7157,
+      "eval_samples_per_second": 3.042,
+      "eval_steps_per_second": 3.042,
+      "step": 3300
+    },
+    {
+      "epoch": 1.3932489451476793,
+      "grad_norm": 0.4024396240711212,
+      "learning_rate": 0.00018680991622175446,
+      "loss": 0.7317537069320679,
+      "step": 3302
+    },
+    {
+      "epoch": 1.3940928270042194,
+      "grad_norm": 0.4526490271091461,
+      "learning_rate": 0.00018678606493573708,
+      "loss": 0.7470992207527161,
+      "step": 3304
+    },
+    {
+      "epoch": 1.3949367088607594,
+      "grad_norm": 0.5055129528045654,
+      "learning_rate": 0.00018676219362984126,
+      "loss": 0.7627943754196167,
+      "step": 3306
+    },
+    {
+      "epoch": 1.3957805907172995,
+      "grad_norm": 0.4030216634273529,
+      "learning_rate": 0.0001867383023095737,
+      "loss": 0.684033989906311,
+      "step": 3308
+    },
+    {
+      "epoch": 1.3966244725738397,
+      "grad_norm": 0.4353308379650116,
+      "learning_rate": 0.0001867143909804456,
+      "loss": 0.7366722822189331,
+      "step": 3310
+    },
+    {
+      "epoch": 1.3974683544303796,
+      "grad_norm": 0.5360553860664368,
+      "learning_rate": 0.0001866904596479729,
+      "loss": 0.8050951957702637,
+      "step": 3312
+    },
+    {
+      "epoch": 1.3983122362869198,
+      "grad_norm": 0.46081677079200745,
+      "learning_rate": 0.00018666650831767608,
+      "loss": 0.7227963805198669,
+      "step": 3314
+    },
+    {
+      "epoch": 1.39915611814346,
+      "grad_norm": 0.4325845539569855,
+      "learning_rate": 0.00018664253699508029,
+      "loss": 0.714216411113739,
+      "step": 3316
+    },
+    {
+      "epoch": 1.4,
+      "grad_norm": 0.4373100996017456,
+      "learning_rate": 0.0001866185456857152,
+      "loss": 0.7378593683242798,
+      "step": 3318
+    },
+    {
+      "epoch": 1.40084388185654,
+      "grad_norm": 0.43221279978752136,
+      "learning_rate": 0.00018659453439511512,
+      "loss": 0.7565981149673462,
+      "step": 3320
+    },
+    {
+      "epoch": 1.4016877637130802,
+      "grad_norm": 0.3938370943069458,
+      "learning_rate": 0.00018657050312881904,
+      "loss": 0.6409488916397095,
+      "step": 3322
+    },
+    {
+      "epoch": 1.4025316455696202,
+      "grad_norm": 0.43609440326690674,
+      "learning_rate": 0.00018654645189237055,
+      "loss": 0.6911652088165283,
+      "step": 3324
+    },
+    {
+      "epoch": 1.4033755274261603,
+      "grad_norm": 0.4526350200176239,
+      "learning_rate": 0.00018652238069131774,
+      "loss": 0.7275710701942444,
+      "step": 3326
+    },
+    {
+      "epoch": 1.4042194092827005,
+      "grad_norm": 0.3990516662597656,
+      "learning_rate": 0.0001864982895312134,
+      "loss": 0.6466895341873169,
+      "step": 3328
+    },
+    {
+      "epoch": 1.4050632911392404,
+      "grad_norm": 0.44321122765541077,
+      "learning_rate": 0.0001864741784176149,
+      "loss": 0.6520540714263916,
+      "step": 3330
+    },
+    {
+      "epoch": 1.4059071729957806,
+      "grad_norm": 0.44733941555023193,
+      "learning_rate": 0.0001864500473560842,
+      "loss": 0.7289671301841736,
+      "step": 3332
+    },
+    {
+      "epoch": 1.4067510548523208,
+      "grad_norm": 0.4569414556026459,
+      "learning_rate": 0.00018642589635218789,
+      "loss": 0.7664391398429871,
+      "step": 3334
+    },
+    {
+      "epoch": 1.4075949367088607,
+      "grad_norm": 0.4172186851501465,
+      "learning_rate": 0.00018640172541149709,
+      "loss": 0.7747462391853333,
+      "step": 3336
+    },
+    {
+      "epoch": 1.4084388185654009,
+      "grad_norm": 0.4540993869304657,
+      "learning_rate": 0.0001863775345395877,
+      "loss": 0.8337607383728027,
+      "step": 3338
+    },
+    {
+      "epoch": 1.409282700421941,
+      "grad_norm": 0.4306154251098633,
+      "learning_rate": 0.00018635332374203993,
+      "loss": 0.6851577758789062,
+      "step": 3340
+    },
+    {
+      "epoch": 1.410126582278481,
+      "grad_norm": 0.45615455508232117,
+      "learning_rate": 0.00018632909302443884,
+      "loss": 0.6841398477554321,
+      "step": 3342
+    },
+    {
+      "epoch": 1.4109704641350211,
+      "grad_norm": 0.35296666622161865,
+      "learning_rate": 0.00018630484239237397,
+      "loss": 0.6672312021255493,
+      "step": 3344
+    },
+    {
+      "epoch": 1.4118143459915613,
+      "grad_norm": 0.4424770176410675,
+      "learning_rate": 0.00018628057185143945,
+      "loss": 0.7027004361152649,
+      "step": 3346
+    },
+    {
+      "epoch": 1.4126582278481012,
+      "grad_norm": 0.429817259311676,
+      "learning_rate": 0.0001862562814072341,
+      "loss": 0.7397928237915039,
+      "step": 3348
+    },
+    {
+      "epoch": 1.4135021097046414,
+      "grad_norm": 0.4720051884651184,
+      "learning_rate": 0.00018623197106536118,
+      "loss": 0.8009215593338013,
+      "step": 3350
+    },
+    {
+      "epoch": 1.4143459915611816,
+      "grad_norm": 0.42197802662849426,
+      "learning_rate": 0.0001862076408314287,
+      "loss": 0.6875046491622925,
+      "step": 3352
+    },
+    {
+      "epoch": 1.4151898734177215,
+      "grad_norm": 0.3674047589302063,
+      "learning_rate": 0.00018618329071104907,
+      "loss": 0.7202261090278625,
+      "step": 3354
+    },
+    {
+      "epoch": 1.4160337552742617,
+      "grad_norm": 0.42327889800071716,
+      "learning_rate": 0.0001861589207098395,
+      "loss": 0.6981248259544373,
+      "step": 3356
+    },
+    {
+      "epoch": 1.4168776371308016,
+      "grad_norm": 0.40644150972366333,
+      "learning_rate": 0.00018613453083342162,
+      "loss": 0.7081356048583984,
+      "step": 3358
+    },
+    {
+      "epoch": 1.4177215189873418,
+      "grad_norm": 0.3948482275009155,
+      "learning_rate": 0.00018611012108742168,
+      "loss": 0.6420091986656189,
+      "step": 3360
+    },
+    {
+      "epoch": 1.4185654008438817,
+      "grad_norm": 0.44622674584388733,
+      "learning_rate": 0.00018608569147747064,
+      "loss": 0.7662023305892944,
+      "step": 3362
+    },
+    {
+      "epoch": 1.4194092827004219,
+      "grad_norm": 0.39772045612335205,
+      "learning_rate": 0.00018606124200920386,
+      "loss": 0.7243877649307251,
+      "step": 3364
+    },
+    {
+      "epoch": 1.420253164556962,
+      "grad_norm": 0.4284326136112213,
+      "learning_rate": 0.00018603677268826138,
+      "loss": 0.7399296164512634,
+      "step": 3366
+    },
+    {
+      "epoch": 1.421097046413502,
+      "grad_norm": 0.3749450445175171,
+      "learning_rate": 0.0001860122835202878,
+      "loss": 0.6978716850280762,
+      "step": 3368
+    },
+    {
+      "epoch": 1.4219409282700421,
+      "grad_norm": 0.46721795201301575,
+      "learning_rate": 0.00018598777451093234,
+      "loss": 0.7733059525489807,
+      "step": 3370
+    },
+    {
+      "epoch": 1.4227848101265823,
+      "grad_norm": 0.39606085419654846,
+      "learning_rate": 0.0001859632456658487,
+      "loss": 0.6721867918968201,
+      "step": 3372
+    },
+    {
+      "epoch": 1.4236286919831223,
+      "grad_norm": 0.3879879117012024,
+      "learning_rate": 0.00018593869699069527,
+      "loss": 0.715825080871582,
+      "step": 3374
+    },
+    {
+      "epoch": 1.4244725738396624,
+      "grad_norm": 0.3641676902770996,
+      "learning_rate": 0.0001859141284911349,
+      "loss": 0.6226941347122192,
+      "step": 3376
+    },
+    {
+      "epoch": 1.4253164556962026,
+      "grad_norm": 0.4612879753112793,
+      "learning_rate": 0.0001858895401728351,
+      "loss": 0.7972121238708496,
+      "step": 3378
+    },
+    {
+      "epoch": 1.4261603375527425,
+      "grad_norm": 0.4406103193759918,
+      "learning_rate": 0.00018586493204146795,
+      "loss": 0.6663658022880554,
+      "step": 3380
+    },
+    {
+      "epoch": 1.4270042194092827,
+      "grad_norm": 0.420205295085907,
+      "learning_rate": 0.00018584030410271,
+      "loss": 0.6867321133613586,
+      "step": 3382
+    },
+    {
+      "epoch": 1.4278481012658228,
+      "grad_norm": 0.4499560594558716,
+      "learning_rate": 0.00018581565636224253,
+      "loss": 0.66688072681427,
+      "step": 3384
+    },
+    {
+      "epoch": 1.4286919831223628,
+      "grad_norm": 0.42507919669151306,
+      "learning_rate": 0.00018579098882575123,
+      "loss": 0.6714702844619751,
+      "step": 3386
+    },
+    {
+      "epoch": 1.429535864978903,
+      "grad_norm": 0.43856513500213623,
+      "learning_rate": 0.0001857663014989265,
+      "loss": 0.7219351530075073,
+      "step": 3388
+    },
+    {
+      "epoch": 1.4303797468354431,
+      "grad_norm": 0.5167419910430908,
+      "learning_rate": 0.00018574159438746314,
+      "loss": 0.7157991528511047,
+      "step": 3390
+    },
+    {
+      "epoch": 1.431223628691983,
+      "grad_norm": 0.4443167746067047,
+      "learning_rate": 0.00018571686749706067,
+      "loss": 0.7509840726852417,
+      "step": 3392
+    },
+    {
+      "epoch": 1.4320675105485232,
+      "grad_norm": 0.41321757435798645,
+      "learning_rate": 0.0001856921208334231,
+      "loss": 0.756000816822052,
+      "step": 3394
+    },
+    {
+      "epoch": 1.4329113924050634,
+      "grad_norm": 0.4094153046607971,
+      "learning_rate": 0.000185667354402259,
+      "loss": 0.7047854065895081,
+      "step": 3396
+    },
+    {
+      "epoch": 1.4337552742616033,
+      "grad_norm": 0.4248482286930084,
+      "learning_rate": 0.0001856425682092815,
+      "loss": 0.7620865106582642,
+      "step": 3398
+    },
+    {
+      "epoch": 1.4345991561181435,
+      "grad_norm": 0.4190550148487091,
+      "learning_rate": 0.0001856177622602083,
+      "loss": 0.7441845536231995,
+      "step": 3400
+    },
+    {
+      "epoch": 1.4345991561181435,
+      "eval_loss": 0.7917885780334473,
+      "eval_runtime": 686.1689,
+      "eval_samples_per_second": 3.071,
+      "eval_steps_per_second": 3.071,
+      "step": 3400
+    },
+    {
+      "epoch": 1.4354430379746836,
+      "grad_norm": 0.45602327585220337,
+      "learning_rate": 0.00018559293656076166,
+      "loss": 0.7123498916625977,
+      "step": 3402
+    },
+    {
+      "epoch": 1.4362869198312236,
+      "grad_norm": 0.5209269523620605,
+      "learning_rate": 0.00018556809111666843,
+      "loss": 0.7096527814865112,
+      "step": 3404
+    },
+    {
+      "epoch": 1.4371308016877637,
+      "grad_norm": 0.44521182775497437,
+      "learning_rate": 0.00018554322593365995,
+      "loss": 0.7116838693618774,
+      "step": 3406
+    },
+    {
+      "epoch": 1.437974683544304,
+      "grad_norm": 0.44571003317832947,
+      "learning_rate": 0.00018551834101747212,
+      "loss": 0.7756352424621582,
+      "step": 3408
+    },
+    {
+      "epoch": 1.4388185654008439,
+      "grad_norm": 0.4030776023864746,
+      "learning_rate": 0.00018549343637384543,
+      "loss": 0.689633309841156,
+      "step": 3410
+    },
+    {
+      "epoch": 1.439662447257384,
+      "grad_norm": 0.4025530517101288,
+      "learning_rate": 0.0001854685120085249,
+      "loss": 0.7173827290534973,
+      "step": 3412
+    },
+    {
+      "epoch": 1.4405063291139242,
+      "grad_norm": 0.4376870095729828,
+      "learning_rate": 0.0001854435679272601,
+      "loss": 0.7212072610855103,
+      "step": 3414
+    },
+    {
+      "epoch": 1.4413502109704641,
+      "grad_norm": 0.4469084143638611,
+      "learning_rate": 0.00018541860413580515,
+      "loss": 0.8260660171508789,
+      "step": 3416
+    },
+    {
+      "epoch": 1.4421940928270043,
+      "grad_norm": 0.48033684492111206,
+      "learning_rate": 0.00018539362063991872,
+      "loss": 0.764930248260498,
+      "step": 3418
+    },
+    {
+      "epoch": 1.4430379746835442,
+      "grad_norm": 0.4428006708621979,
+      "learning_rate": 0.00018536861744536403,
+      "loss": 0.7436494827270508,
+      "step": 3420
+    },
+    {
+      "epoch": 1.4438818565400844,
+      "grad_norm": 0.41812989115715027,
+      "learning_rate": 0.00018534359455790885,
+      "loss": 0.7695552706718445,
+      "step": 3422
+    },
+    {
+      "epoch": 1.4447257383966245,
+      "grad_norm": 0.4243115186691284,
+      "learning_rate": 0.00018531855198332544,
+      "loss": 0.7430856823921204,
+      "step": 3424
+    },
+    {
+      "epoch": 1.4455696202531645,
+      "grad_norm": 0.44387027621269226,
+      "learning_rate": 0.00018529348972739067,
+      "loss": 0.7015851140022278,
+      "step": 3426
+    },
+    {
+      "epoch": 1.4464135021097047,
+      "grad_norm": 0.5105122923851013,
+      "learning_rate": 0.0001852684077958859,
+      "loss": 0.7073075771331787,
+      "step": 3428
+    },
+    {
+      "epoch": 1.4472573839662446,
+      "grad_norm": 0.40193185210227966,
+      "learning_rate": 0.00018524330619459708,
+      "loss": 0.7478930950164795,
+      "step": 3430
+    },
+    {
+      "epoch": 1.4481012658227848,
+      "grad_norm": 0.47935256361961365,
+      "learning_rate": 0.00018521818492931463,
+      "loss": 0.6987152099609375,
+      "step": 3432
+    },
+    {
+      "epoch": 1.448945147679325,
+      "grad_norm": 0.42298686504364014,
+      "learning_rate": 0.00018519304400583356,
+      "loss": 0.6677015423774719,
+      "step": 3434
+    },
+    {
+      "epoch": 1.4497890295358649,
+      "grad_norm": 0.37220221757888794,
+      "learning_rate": 0.0001851678834299534,
+      "loss": 0.5727065205574036,
+      "step": 3436
+    },
+    {
+      "epoch": 1.450632911392405,
+      "grad_norm": 0.4990854561328888,
+      "learning_rate": 0.00018514270320747823,
+      "loss": 0.7245276570320129,
+      "step": 3438
+    },
+    {
+      "epoch": 1.4514767932489452,
+      "grad_norm": 0.4366965889930725,
+      "learning_rate": 0.00018511750334421665,
+      "loss": 0.5907649993896484,
+      "step": 3440
+    },
+    {
+      "epoch": 1.4523206751054851,
+      "grad_norm": 0.4355267286300659,
+      "learning_rate": 0.00018509228384598172,
+      "loss": 0.6898093223571777,
+      "step": 3442
+    },
+    {
+      "epoch": 1.4531645569620253,
+      "grad_norm": 0.4055832028388977,
+      "learning_rate": 0.00018506704471859116,
+      "loss": 0.6242932081222534,
+      "step": 3444
+    },
+    {
+      "epoch": 1.4540084388185655,
+      "grad_norm": 0.42599600553512573,
+      "learning_rate": 0.00018504178596786712,
+      "loss": 0.7214970588684082,
+      "step": 3446
+    },
+    {
+      "epoch": 1.4548523206751054,
+      "grad_norm": 0.41228237748146057,
+      "learning_rate": 0.0001850165075996363,
+      "loss": 0.7120789885520935,
+      "step": 3448
+    },
+    {
+      "epoch": 1.4556962025316456,
+      "grad_norm": 0.4529595375061035,
+      "learning_rate": 0.00018499120961972996,
+      "loss": 0.7376183867454529,
+      "step": 3450
+    },
+    {
+      "epoch": 1.4565400843881857,
+      "grad_norm": 0.4260895550251007,
+      "learning_rate": 0.00018496589203398386,
+      "loss": 0.7321018576622009,
+      "step": 3452
+    },
+    {
+      "epoch": 1.4573839662447257,
+      "grad_norm": 0.46325981616973877,
+      "learning_rate": 0.00018494055484823824,
+      "loss": 0.7099342346191406,
+      "step": 3454
+    },
+    {
+      "epoch": 1.4582278481012658,
+      "grad_norm": 0.43322354555130005,
+      "learning_rate": 0.00018491519806833795,
+      "loss": 0.6663834452629089,
+      "step": 3456
+    },
+    {
+      "epoch": 1.459071729957806,
+      "grad_norm": 0.421683132648468,
+      "learning_rate": 0.0001848898217001323,
+      "loss": 0.6634764671325684,
+      "step": 3458
+    },
+    {
+      "epoch": 1.459915611814346,
+      "grad_norm": 0.44365957379341125,
+      "learning_rate": 0.00018486442574947511,
+      "loss": 0.697095513343811,
+      "step": 3460
+    },
+    {
+      "epoch": 1.460759493670886,
+      "grad_norm": 0.4892813265323639,
+      "learning_rate": 0.00018483901022222474,
+      "loss": 0.8009715676307678,
+      "step": 3462
+    },
+    {
+      "epoch": 1.4616033755274263,
+      "grad_norm": 0.4230288863182068,
+      "learning_rate": 0.00018481357512424408,
+      "loss": 0.631499707698822,
+      "step": 3464
+    },
+    {
+      "epoch": 1.4624472573839662,
+      "grad_norm": 0.49490758776664734,
+      "learning_rate": 0.00018478812046140056,
+      "loss": 0.8045525550842285,
+      "step": 3466
+    },
+    {
+      "epoch": 1.4632911392405064,
+      "grad_norm": 0.5035375952720642,
+      "learning_rate": 0.00018476264623956597,
+      "loss": 0.7299133539199829,
+      "step": 3468
+    },
+    {
+      "epoch": 1.4641350210970465,
+      "grad_norm": 0.45048409700393677,
+      "learning_rate": 0.00018473715246461684,
+      "loss": 0.7313228845596313,
+      "step": 3470
+    },
+    {
+      "epoch": 1.4649789029535865,
+      "grad_norm": 0.41414541006088257,
+      "learning_rate": 0.00018471163914243405,
+      "loss": 0.7515901923179626,
+      "step": 3472
+    },
+    {
+      "epoch": 1.4658227848101266,
+      "grad_norm": 0.4440646767616272,
+      "learning_rate": 0.00018468610627890298,
+      "loss": 0.7425532341003418,
+      "step": 3474
+    },
+    {
+      "epoch": 1.4666666666666668,
+      "grad_norm": 0.4749740660190582,
+      "learning_rate": 0.00018466055387991362,
+      "loss": 0.7509122490882874,
+      "step": 3476
+    },
+    {
+      "epoch": 1.4675105485232067,
+      "grad_norm": 0.498951256275177,
+      "learning_rate": 0.00018463498195136046,
+      "loss": 0.7395774126052856,
+      "step": 3478
+    },
+    {
+      "epoch": 1.4683544303797469,
+      "grad_norm": 0.45996859669685364,
+      "learning_rate": 0.00018460939049914237,
+      "loss": 0.7860153317451477,
+      "step": 3480
+    },
+    {
+      "epoch": 1.469198312236287,
+      "grad_norm": 0.5989604592323303,
+      "learning_rate": 0.0001845837795291629,
+      "loss": 0.8193565607070923,
+      "step": 3482
+    },
+    {
+      "epoch": 1.470042194092827,
+      "grad_norm": 0.3903721272945404,
+      "learning_rate": 0.0001845581490473299,
+      "loss": 0.6569210290908813,
+      "step": 3484
+    },
+    {
+      "epoch": 1.4708860759493672,
+      "grad_norm": 0.4316142201423645,
+      "learning_rate": 0.00018453249905955592,
+      "loss": 0.7226642370223999,
+      "step": 3486
+    },
+    {
+      "epoch": 1.471729957805907,
+      "grad_norm": 0.4447585344314575,
+      "learning_rate": 0.00018450682957175786,
+      "loss": 0.8134398460388184,
+      "step": 3488
+    },
+    {
+      "epoch": 1.4725738396624473,
+      "grad_norm": 0.42515450716018677,
+      "learning_rate": 0.0001844811405898572,
+      "loss": 0.7668985724449158,
+      "step": 3490
+    },
+    {
+      "epoch": 1.4734177215189874,
+      "grad_norm": 0.39883363246917725,
+      "learning_rate": 0.0001844554321197799,
+      "loss": 0.7289790511131287,
+      "step": 3492
+    },
+    {
+      "epoch": 1.4742616033755274,
+      "grad_norm": 0.4419080317020416,
+      "learning_rate": 0.00018442970416745645,
+      "loss": 0.7611824870109558,
+      "step": 3494
+    },
+    {
+      "epoch": 1.4751054852320675,
+      "grad_norm": 0.37460315227508545,
+      "learning_rate": 0.0001844039567388217,
+      "loss": 0.6926230788230896,
+      "step": 3496
+    },
+    {
+      "epoch": 1.4759493670886075,
+      "grad_norm": 0.47706854343414307,
+      "learning_rate": 0.00018437818983981513,
+      "loss": 0.7889325618743896,
+      "step": 3498
+    },
+    {
+      "epoch": 1.4767932489451476,
+      "grad_norm": 0.449028879404068,
+      "learning_rate": 0.00018435240347638074,
+      "loss": 0.7231656312942505,
+      "step": 3500
+    },
+    {
+      "epoch": 1.4767932489451476,
+      "eval_loss": 0.7902651429176331,
+      "eval_runtime": 672.312,
+      "eval_samples_per_second": 3.134,
+      "eval_steps_per_second": 3.134,
+      "step": 3500
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 14220,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 6,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 5,
+        "early_stopping_threshold": 0.001
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 0
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.0921947927763743e+18,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-3500/training_args.bin b/sft_qwen_14B_v2/checkpoints/checkpoint-3500/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..0bd5f45757bdc6c7fc2d67718b18675b51983e0e
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-3500/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:afae4a8ce1391f149eb00b14eed8f891c715e892ea546bf754d22db2c2bc5969
+size 4792
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-4000/README.md b/sft_qwen_14B_v2/checkpoints/checkpoint-4000/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..134a3a8531b68adda3fed2a98db824f7c1b7dcfd
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-4000/README.md
@@ -0,0 +1,207 @@
+---
+base_model: Models/Qwen2.5-Coder-14B-CPT
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Models/Qwen2.5-Coder-14B-CPT
+- lora
+- transformers
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.0
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-4000/adapter_config.json b/sft_qwen_14B_v2/checkpoints/checkpoint-4000/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..d6316742bc71d30704e2ccdb9080ed0b8062e454
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-4000/adapter_config.json
@@ -0,0 +1,43 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Models/Qwen2.5-Coder-14B-CPT",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.0",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "v_proj",
+    "k_proj",
+    "q_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-4000/adapter_model.safetensors b/sft_qwen_14B_v2/checkpoints/checkpoint-4000/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..4e60126d0e6c6f3073c02f93f4fadd608d862792
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-4000/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c0b4b4ab8835656ebacc3a666568fb7516cc0d8f445937c95d3c4bd92023329
+size 201378736
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-4000/optimizer.pt b/sft_qwen_14B_v2/checkpoints/checkpoint-4000/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..d01a9660b8ff0779ab72f976511bf543f7a0ebaf
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-4000/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c03540648772cff09199213d86878e9e3327878a3d8aaad80e2734fe7dc324f1
+size 402976786
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-4000/rng_state.pth b/sft_qwen_14B_v2/checkpoints/checkpoint-4000/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..5f80acae5861a2fda7b1bd4cad39d50a53ef3da2
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-4000/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:099dbb6b327a84989fe6c6e9aabafbab6293b89bc8cf6ae20258451890f15c96
+size 14244
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-4000/scheduler.pt b/sft_qwen_14B_v2/checkpoints/checkpoint-4000/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..7a98ef4622dc2c521d9c44fb0d80fec09a48e2cc
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-4000/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bccfebac5a89018bb22a2f7c343299ab5b1804b982de5ffaad552cc78488df8e
+size 1064
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-4000/trainer_state.json b/sft_qwen_14B_v2/checkpoints/checkpoint-4000/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..9368cc94c82f0e54b88b874cffc571fe489ccc9f
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-4000/trainer_state.json
@@ -0,0 +1,14363 @@
+{
+  "best_global_step": 4000,
+  "best_metric": 0.7721371650695801,
+  "best_model_checkpoint": "task2file/sft_qwen_14B_v2/checkpoints/checkpoint-4000",
+  "epoch": 1.6877637130801688,
+  "eval_steps": 100,
+  "global_step": 4000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0008438818565400844,
+      "grad_norm": 0.5386583805084229,
+      "learning_rate": 1.7574692442882248e-07,
+      "loss": 1.6941628456115723,
+      "step": 2
+    },
+    {
+      "epoch": 0.0016877637130801688,
+      "grad_norm": 0.5477277636528015,
+      "learning_rate": 5.272407732864675e-07,
+      "loss": 1.7132279872894287,
+      "step": 4
+    },
+    {
+      "epoch": 0.002531645569620253,
+      "grad_norm": 0.5390765070915222,
+      "learning_rate": 8.787346221441126e-07,
+      "loss": 1.641180396080017,
+      "step": 6
+    },
+    {
+      "epoch": 0.0033755274261603376,
+      "grad_norm": 0.5023683905601501,
+      "learning_rate": 1.2302284710017575e-06,
+      "loss": 1.5616240501403809,
+      "step": 8
+    },
+    {
+      "epoch": 0.004219409282700422,
+      "grad_norm": 0.4899154603481293,
+      "learning_rate": 1.5817223198594026e-06,
+      "loss": 1.572033405303955,
+      "step": 10
+    },
+    {
+      "epoch": 0.005063291139240506,
+      "grad_norm": 0.5239788293838501,
+      "learning_rate": 1.9332161687170474e-06,
+      "loss": 1.6242921352386475,
+      "step": 12
+    },
+    {
+      "epoch": 0.00590717299578059,
+      "grad_norm": 0.5172926783561707,
+      "learning_rate": 2.2847100175746925e-06,
+      "loss": 1.6800041198730469,
+      "step": 14
+    },
+    {
+      "epoch": 0.006751054852320675,
+      "grad_norm": 0.5539224743843079,
+      "learning_rate": 2.6362038664323376e-06,
+      "loss": 1.6450834274291992,
+      "step": 16
+    },
+    {
+      "epoch": 0.007594936708860759,
+      "grad_norm": 0.5255337953567505,
+      "learning_rate": 2.9876977152899827e-06,
+      "loss": 1.6673263311386108,
+      "step": 18
+    },
+    {
+      "epoch": 0.008438818565400843,
+      "grad_norm": 0.5074548721313477,
+      "learning_rate": 3.3391915641476277e-06,
+      "loss": 1.531802773475647,
+      "step": 20
+    },
+    {
+      "epoch": 0.009282700421940928,
+      "grad_norm": 0.4160279333591461,
+      "learning_rate": 3.6906854130052724e-06,
+      "loss": 1.599354863166809,
+      "step": 22
+    },
+    {
+      "epoch": 0.010126582278481013,
+      "grad_norm": 0.5716474652290344,
+      "learning_rate": 4.0421792618629174e-06,
+      "loss": 1.6700962781906128,
+      "step": 24
+    },
+    {
+      "epoch": 0.010970464135021098,
+      "grad_norm": 0.5148899555206299,
+      "learning_rate": 4.3936731107205625e-06,
+      "loss": 1.66217839717865,
+      "step": 26
+    },
+    {
+      "epoch": 0.01181434599156118,
+      "grad_norm": 0.575722336769104,
+      "learning_rate": 4.7451669595782076e-06,
+      "loss": 1.6692266464233398,
+      "step": 28
+    },
+    {
+      "epoch": 0.012658227848101266,
+      "grad_norm": 0.5345953106880188,
+      "learning_rate": 5.096660808435853e-06,
+      "loss": 1.5518689155578613,
+      "step": 30
+    },
+    {
+      "epoch": 0.01350210970464135,
+      "grad_norm": 0.4462043344974518,
+      "learning_rate": 5.448154657293498e-06,
+      "loss": 1.5930007696151733,
+      "step": 32
+    },
+    {
+      "epoch": 0.014345991561181435,
+      "grad_norm": 0.5119605660438538,
+      "learning_rate": 5.799648506151143e-06,
+      "loss": 1.6069684028625488,
+      "step": 34
+    },
+    {
+      "epoch": 0.015189873417721518,
+      "grad_norm": 0.5328608751296997,
+      "learning_rate": 6.151142355008788e-06,
+      "loss": 1.5838109254837036,
+      "step": 36
+    },
+    {
+      "epoch": 0.016033755274261603,
+      "grad_norm": 0.5065920352935791,
+      "learning_rate": 6.502636203866433e-06,
+      "loss": 1.608130931854248,
+      "step": 38
+    },
+    {
+      "epoch": 0.016877637130801686,
+      "grad_norm": 0.4479359984397888,
+      "learning_rate": 6.854130052724078e-06,
+      "loss": 1.5942182540893555,
+      "step": 40
+    },
+    {
+      "epoch": 0.017721518987341773,
+      "grad_norm": 0.42844903469085693,
+      "learning_rate": 7.205623901581722e-06,
+      "loss": 1.6441553831100464,
+      "step": 42
+    },
+    {
+      "epoch": 0.018565400843881856,
+      "grad_norm": 0.476630836725235,
+      "learning_rate": 7.557117750439367e-06,
+      "loss": 1.6068111658096313,
+      "step": 44
+    },
+    {
+      "epoch": 0.019409282700421943,
+      "grad_norm": 0.4532654881477356,
+      "learning_rate": 7.908611599297012e-06,
+      "loss": 1.6618021726608276,
+      "step": 46
+    },
+    {
+      "epoch": 0.020253164556962026,
+      "grad_norm": 0.3701118230819702,
+      "learning_rate": 8.260105448154657e-06,
+      "loss": 1.4730033874511719,
+      "step": 48
+    },
+    {
+      "epoch": 0.02109704641350211,
+      "grad_norm": 0.38471561670303345,
+      "learning_rate": 8.611599297012302e-06,
+      "loss": 1.4828267097473145,
+      "step": 50
+    },
+    {
+      "epoch": 0.021940928270042195,
+      "grad_norm": 0.3602336347103119,
+      "learning_rate": 8.963093145869948e-06,
+      "loss": 1.3877452611923218,
+      "step": 52
+    },
+    {
+      "epoch": 0.02278481012658228,
+      "grad_norm": 0.40318572521209717,
+      "learning_rate": 9.314586994727593e-06,
+      "loss": 1.49052894115448,
+      "step": 54
+    },
+    {
+      "epoch": 0.02362869198312236,
+      "grad_norm": 0.3223826587200165,
+      "learning_rate": 9.666080843585238e-06,
+      "loss": 1.4912524223327637,
+      "step": 56
+    },
+    {
+      "epoch": 0.024472573839662448,
+      "grad_norm": 0.3873065114021301,
+      "learning_rate": 1.0017574692442883e-05,
+      "loss": 1.526674509048462,
+      "step": 58
+    },
+    {
+      "epoch": 0.02531645569620253,
+      "grad_norm": 0.410159707069397,
+      "learning_rate": 1.0369068541300528e-05,
+      "loss": 1.4480271339416504,
+      "step": 60
+    },
+    {
+      "epoch": 0.026160337552742614,
+      "grad_norm": 0.3632003962993622,
+      "learning_rate": 1.0720562390158173e-05,
+      "loss": 1.4222990274429321,
+      "step": 62
+    },
+    {
+      "epoch": 0.0270042194092827,
+      "grad_norm": 0.33118435740470886,
+      "learning_rate": 1.1072056239015818e-05,
+      "loss": 1.387171745300293,
+      "step": 64
+    },
+    {
+      "epoch": 0.027848101265822784,
+      "grad_norm": 0.3301764726638794,
+      "learning_rate": 1.1423550087873463e-05,
+      "loss": 1.3523777723312378,
+      "step": 66
+    },
+    {
+      "epoch": 0.02869198312236287,
+      "grad_norm": 0.34342435002326965,
+      "learning_rate": 1.1775043936731108e-05,
+      "loss": 1.4515162706375122,
+      "step": 68
+    },
+    {
+      "epoch": 0.029535864978902954,
+      "grad_norm": 0.3243122100830078,
+      "learning_rate": 1.2126537785588753e-05,
+      "loss": 1.3509243726730347,
+      "step": 70
+    },
+    {
+      "epoch": 0.030379746835443037,
+      "grad_norm": 0.3450150787830353,
+      "learning_rate": 1.2478031634446398e-05,
+      "loss": 1.4936245679855347,
+      "step": 72
+    },
+    {
+      "epoch": 0.031223628691983123,
+      "grad_norm": 0.38912028074264526,
+      "learning_rate": 1.2829525483304042e-05,
+      "loss": 1.3419109582901,
+      "step": 74
+    },
+    {
+      "epoch": 0.032067510548523206,
+      "grad_norm": 0.3019310235977173,
+      "learning_rate": 1.3181019332161687e-05,
+      "loss": 1.4284154176712036,
+      "step": 76
+    },
+    {
+      "epoch": 0.03291139240506329,
+      "grad_norm": 0.37803682684898376,
+      "learning_rate": 1.3532513181019332e-05,
+      "loss": 1.4256561994552612,
+      "step": 78
+    },
+    {
+      "epoch": 0.03375527426160337,
+      "grad_norm": 0.34191736578941345,
+      "learning_rate": 1.3884007029876977e-05,
+      "loss": 1.3256909847259521,
+      "step": 80
+    },
+    {
+      "epoch": 0.03459915611814346,
+      "grad_norm": 0.35242700576782227,
+      "learning_rate": 1.4235500878734624e-05,
+      "loss": 1.2710685729980469,
+      "step": 82
+    },
+    {
+      "epoch": 0.035443037974683546,
+      "grad_norm": 0.38094228506088257,
+      "learning_rate": 1.4586994727592269e-05,
+      "loss": 1.253411889076233,
+      "step": 84
+    },
+    {
+      "epoch": 0.036286919831223625,
+      "grad_norm": 0.36837366223335266,
+      "learning_rate": 1.4938488576449914e-05,
+      "loss": 1.3064342737197876,
+      "step": 86
+    },
+    {
+      "epoch": 0.03713080168776371,
+      "grad_norm": 0.3443569242954254,
+      "learning_rate": 1.5289982425307557e-05,
+      "loss": 1.293562412261963,
+      "step": 88
+    },
+    {
+      "epoch": 0.0379746835443038,
+      "grad_norm": 0.3799338936805725,
+      "learning_rate": 1.5641476274165202e-05,
+      "loss": 1.3382648229599,
+      "step": 90
+    },
+    {
+      "epoch": 0.038818565400843885,
+      "grad_norm": 0.40501922369003296,
+      "learning_rate": 1.599297012302285e-05,
+      "loss": 1.3925724029541016,
+      "step": 92
+    },
+    {
+      "epoch": 0.039662447257383965,
+      "grad_norm": 0.4419630467891693,
+      "learning_rate": 1.6344463971880492e-05,
+      "loss": 1.357171893119812,
+      "step": 94
+    },
+    {
+      "epoch": 0.04050632911392405,
+      "grad_norm": 0.3619817793369293,
+      "learning_rate": 1.6695957820738137e-05,
+      "loss": 1.3029985427856445,
+      "step": 96
+    },
+    {
+      "epoch": 0.04135021097046414,
+      "grad_norm": 0.4851357340812683,
+      "learning_rate": 1.7047451669595782e-05,
+      "loss": 1.3498191833496094,
+      "step": 98
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "grad_norm": 0.418658584356308,
+      "learning_rate": 1.7398945518453427e-05,
+      "loss": 1.185287356376648,
+      "step": 100
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "eval_loss": 1.2979938983917236,
+      "eval_runtime": 682.1979,
+      "eval_samples_per_second": 3.089,
+      "eval_steps_per_second": 3.089,
+      "step": 100
+    },
+    {
+      "epoch": 0.043037974683544304,
+      "grad_norm": 0.4464418888092041,
+      "learning_rate": 1.7750439367311073e-05,
+      "loss": 1.2217272520065308,
+      "step": 102
+    },
+    {
+      "epoch": 0.04388185654008439,
+      "grad_norm": 0.4706237316131592,
+      "learning_rate": 1.8101933216168718e-05,
+      "loss": 1.2052050828933716,
+      "step": 104
+    },
+    {
+      "epoch": 0.04472573839662447,
+      "grad_norm": 0.46394404768943787,
+      "learning_rate": 1.8453427065026363e-05,
+      "loss": 1.221343994140625,
+      "step": 106
+    },
+    {
+      "epoch": 0.04556962025316456,
+      "grad_norm": 0.4726889431476593,
+      "learning_rate": 1.8804920913884008e-05,
+      "loss": 1.2387475967407227,
+      "step": 108
+    },
+    {
+      "epoch": 0.046413502109704644,
+      "grad_norm": 0.42130985856056213,
+      "learning_rate": 1.9156414762741653e-05,
+      "loss": 1.2851309776306152,
+      "step": 110
+    },
+    {
+      "epoch": 0.04725738396624472,
+      "grad_norm": 0.4504576623439789,
+      "learning_rate": 1.9507908611599298e-05,
+      "loss": 1.2753145694732666,
+      "step": 112
+    },
+    {
+      "epoch": 0.04810126582278481,
+      "grad_norm": 0.396085262298584,
+      "learning_rate": 1.9859402460456943e-05,
+      "loss": 1.2427717447280884,
+      "step": 114
+    },
+    {
+      "epoch": 0.048945147679324896,
+      "grad_norm": 0.5106491446495056,
+      "learning_rate": 2.0210896309314588e-05,
+      "loss": 1.2943825721740723,
+      "step": 116
+    },
+    {
+      "epoch": 0.049789029535864976,
+      "grad_norm": 0.42351317405700684,
+      "learning_rate": 2.0562390158172233e-05,
+      "loss": 1.263301134109497,
+      "step": 118
+    },
+    {
+      "epoch": 0.05063291139240506,
+      "grad_norm": 0.4403539299964905,
+      "learning_rate": 2.0913884007029878e-05,
+      "loss": 1.2647849321365356,
+      "step": 120
+    },
+    {
+      "epoch": 0.05147679324894515,
+      "grad_norm": 0.5260752439498901,
+      "learning_rate": 2.1265377855887523e-05,
+      "loss": 1.2351393699645996,
+      "step": 122
+    },
+    {
+      "epoch": 0.05232067510548523,
+      "grad_norm": 0.44978851079940796,
+      "learning_rate": 2.1616871704745168e-05,
+      "loss": 1.0384471416473389,
+      "step": 124
+    },
+    {
+      "epoch": 0.053164556962025315,
+      "grad_norm": 0.47732362151145935,
+      "learning_rate": 2.1968365553602813e-05,
+      "loss": 1.1518068313598633,
+      "step": 126
+    },
+    {
+      "epoch": 0.0540084388185654,
+      "grad_norm": 0.5473551750183105,
+      "learning_rate": 2.231985940246046e-05,
+      "loss": 1.2264912128448486,
+      "step": 128
+    },
+    {
+      "epoch": 0.05485232067510549,
+      "grad_norm": 0.4473855197429657,
+      "learning_rate": 2.2671353251318103e-05,
+      "loss": 1.1615246534347534,
+      "step": 130
+    },
+    {
+      "epoch": 0.05569620253164557,
+      "grad_norm": 0.5980377197265625,
+      "learning_rate": 2.302284710017575e-05,
+      "loss": 1.1334880590438843,
+      "step": 132
+    },
+    {
+      "epoch": 0.056540084388185655,
+      "grad_norm": 0.5987792015075684,
+      "learning_rate": 2.3374340949033394e-05,
+      "loss": 1.1546804904937744,
+      "step": 134
+    },
+    {
+      "epoch": 0.05738396624472574,
+      "grad_norm": 0.45355498790740967,
+      "learning_rate": 2.372583479789104e-05,
+      "loss": 1.194953441619873,
+      "step": 136
+    },
+    {
+      "epoch": 0.05822784810126582,
+      "grad_norm": 0.5373698472976685,
+      "learning_rate": 2.4077328646748684e-05,
+      "loss": 1.1067466735839844,
+      "step": 138
+    },
+    {
+      "epoch": 0.05907172995780591,
+      "grad_norm": 0.48734328150749207,
+      "learning_rate": 2.442882249560633e-05,
+      "loss": 1.188468098640442,
+      "step": 140
+    },
+    {
+      "epoch": 0.059915611814345994,
+      "grad_norm": 0.4692173898220062,
+      "learning_rate": 2.478031634446397e-05,
+      "loss": 1.1624362468719482,
+      "step": 142
+    },
+    {
+      "epoch": 0.060759493670886074,
+      "grad_norm": 0.532554030418396,
+      "learning_rate": 2.513181019332162e-05,
+      "loss": 1.0978907346725464,
+      "step": 144
+    },
+    {
+      "epoch": 0.06160337552742616,
+      "grad_norm": 0.5853802561759949,
+      "learning_rate": 2.5483304042179264e-05,
+      "loss": 1.2030781507492065,
+      "step": 146
+    },
+    {
+      "epoch": 0.06244725738396625,
+      "grad_norm": 0.5061611533164978,
+      "learning_rate": 2.583479789103691e-05,
+      "loss": 1.082366943359375,
+      "step": 148
+    },
+    {
+      "epoch": 0.06329113924050633,
+      "grad_norm": 0.49426141381263733,
+      "learning_rate": 2.6186291739894554e-05,
+      "loss": 1.10564386844635,
+      "step": 150
+    },
+    {
+      "epoch": 0.06413502109704641,
+      "grad_norm": 0.5846618413925171,
+      "learning_rate": 2.6537785588752196e-05,
+      "loss": 1.1992807388305664,
+      "step": 152
+    },
+    {
+      "epoch": 0.06497890295358649,
+      "grad_norm": 0.5517552495002747,
+      "learning_rate": 2.6889279437609844e-05,
+      "loss": 1.1757566928863525,
+      "step": 154
+    },
+    {
+      "epoch": 0.06582278481012659,
+      "grad_norm": 0.5667305588722229,
+      "learning_rate": 2.7240773286467486e-05,
+      "loss": 1.0548783540725708,
+      "step": 156
+    },
+    {
+      "epoch": 0.06666666666666667,
+      "grad_norm": 0.6760414242744446,
+      "learning_rate": 2.7592267135325134e-05,
+      "loss": 1.184364914894104,
+      "step": 158
+    },
+    {
+      "epoch": 0.06751054852320675,
+      "grad_norm": 0.5261430740356445,
+      "learning_rate": 2.7943760984182776e-05,
+      "loss": 1.1945042610168457,
+      "step": 160
+    },
+    {
+      "epoch": 0.06835443037974684,
+      "grad_norm": 0.6155015230178833,
+      "learning_rate": 2.8295254833040425e-05,
+      "loss": 1.2021973133087158,
+      "step": 162
+    },
+    {
+      "epoch": 0.06919831223628692,
+      "grad_norm": 0.6131619215011597,
+      "learning_rate": 2.8646748681898066e-05,
+      "loss": 1.144123911857605,
+      "step": 164
+    },
+    {
+      "epoch": 0.070042194092827,
+      "grad_norm": 0.5749185681343079,
+      "learning_rate": 2.8998242530755715e-05,
+      "loss": 1.1329256296157837,
+      "step": 166
+    },
+    {
+      "epoch": 0.07088607594936709,
+      "grad_norm": 0.5243118405342102,
+      "learning_rate": 2.9349736379613356e-05,
+      "loss": 1.0892387628555298,
+      "step": 168
+    },
+    {
+      "epoch": 0.07172995780590717,
+      "grad_norm": 0.7190104722976685,
+      "learning_rate": 2.9701230228471005e-05,
+      "loss": 1.163260817527771,
+      "step": 170
+    },
+    {
+      "epoch": 0.07257383966244725,
+      "grad_norm": 0.5486982464790344,
+      "learning_rate": 3.0052724077328647e-05,
+      "loss": 1.0880777835845947,
+      "step": 172
+    },
+    {
+      "epoch": 0.07341772151898734,
+      "grad_norm": 0.5020889043807983,
+      "learning_rate": 3.0404217926186295e-05,
+      "loss": 1.0433368682861328,
+      "step": 174
+    },
+    {
+      "epoch": 0.07426160337552742,
+      "grad_norm": 0.47329774498939514,
+      "learning_rate": 3.075571177504394e-05,
+      "loss": 1.0528991222381592,
+      "step": 176
+    },
+    {
+      "epoch": 0.0751054852320675,
+      "grad_norm": 0.6635547876358032,
+      "learning_rate": 3.110720562390158e-05,
+      "loss": 1.1627811193466187,
+      "step": 178
+    },
+    {
+      "epoch": 0.0759493670886076,
+      "grad_norm": 0.5624618530273438,
+      "learning_rate": 3.145869947275923e-05,
+      "loss": 1.084869384765625,
+      "step": 180
+    },
+    {
+      "epoch": 0.07679324894514768,
+      "grad_norm": 0.6029536724090576,
+      "learning_rate": 3.181019332161687e-05,
+      "loss": 1.2227671146392822,
+      "step": 182
+    },
+    {
+      "epoch": 0.07763713080168777,
+      "grad_norm": 0.930959939956665,
+      "learning_rate": 3.216168717047452e-05,
+      "loss": 1.0955452919006348,
+      "step": 184
+    },
+    {
+      "epoch": 0.07848101265822785,
+      "grad_norm": 0.5326952338218689,
+      "learning_rate": 3.251318101933216e-05,
+      "loss": 1.0640798807144165,
+      "step": 186
+    },
+    {
+      "epoch": 0.07932489451476793,
+      "grad_norm": 0.5484727621078491,
+      "learning_rate": 3.286467486818981e-05,
+      "loss": 1.0700589418411255,
+      "step": 188
+    },
+    {
+      "epoch": 0.08016877637130802,
+      "grad_norm": 0.605273425579071,
+      "learning_rate": 3.3216168717047456e-05,
+      "loss": 1.1593081951141357,
+      "step": 190
+    },
+    {
+      "epoch": 0.0810126582278481,
+      "grad_norm": 0.5704394578933716,
+      "learning_rate": 3.35676625659051e-05,
+      "loss": 1.1617076396942139,
+      "step": 192
+    },
+    {
+      "epoch": 0.08185654008438818,
+      "grad_norm": 0.5929452180862427,
+      "learning_rate": 3.3919156414762746e-05,
+      "loss": 1.1346839666366577,
+      "step": 194
+    },
+    {
+      "epoch": 0.08270042194092828,
+      "grad_norm": 0.5624077916145325,
+      "learning_rate": 3.427065026362039e-05,
+      "loss": 1.0934710502624512,
+      "step": 196
+    },
+    {
+      "epoch": 0.08354430379746836,
+      "grad_norm": 0.6717425584793091,
+      "learning_rate": 3.4622144112478036e-05,
+      "loss": 1.1810534000396729,
+      "step": 198
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "grad_norm": 0.5120199918746948,
+      "learning_rate": 3.4973637961335674e-05,
+      "loss": 1.1525514125823975,
+      "step": 200
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "eval_loss": 1.142486810684204,
+      "eval_runtime": 668.2356,
+      "eval_samples_per_second": 3.153,
+      "eval_steps_per_second": 3.153,
+      "step": 200
+    },
+    {
+      "epoch": 0.08523206751054853,
+      "grad_norm": 0.5144487023353577,
+      "learning_rate": 3.5325131810193326e-05,
+      "loss": 1.0243735313415527,
+      "step": 202
+    },
+    {
+      "epoch": 0.08607594936708861,
+      "grad_norm": 0.6325069069862366,
+      "learning_rate": 3.5676625659050964e-05,
+      "loss": 1.118743896484375,
+      "step": 204
+    },
+    {
+      "epoch": 0.08691983122362869,
+      "grad_norm": 0.5501633882522583,
+      "learning_rate": 3.6028119507908616e-05,
+      "loss": 1.0380504131317139,
+      "step": 206
+    },
+    {
+      "epoch": 0.08776371308016878,
+      "grad_norm": 0.6133899688720703,
+      "learning_rate": 3.6379613356766254e-05,
+      "loss": 0.9837555885314941,
+      "step": 208
+    },
+    {
+      "epoch": 0.08860759493670886,
+      "grad_norm": 0.5799810886383057,
+      "learning_rate": 3.6731107205623906e-05,
+      "loss": 1.090720295906067,
+      "step": 210
+    },
+    {
+      "epoch": 0.08945147679324894,
+      "grad_norm": 0.6039511561393738,
+      "learning_rate": 3.7082601054481544e-05,
+      "loss": 1.120232343673706,
+      "step": 212
+    },
+    {
+      "epoch": 0.09029535864978903,
+      "grad_norm": 0.5983024835586548,
+      "learning_rate": 3.7434094903339196e-05,
+      "loss": 1.096949815750122,
+      "step": 214
+    },
+    {
+      "epoch": 0.09113924050632911,
+      "grad_norm": 0.5641079545021057,
+      "learning_rate": 3.7785588752196835e-05,
+      "loss": 1.1226298809051514,
+      "step": 216
+    },
+    {
+      "epoch": 0.0919831223628692,
+      "grad_norm": 0.655717134475708,
+      "learning_rate": 3.8137082601054486e-05,
+      "loss": 1.1260643005371094,
+      "step": 218
+    },
+    {
+      "epoch": 0.09282700421940929,
+      "grad_norm": 0.6111898422241211,
+      "learning_rate": 3.848857644991213e-05,
+      "loss": 1.0777709484100342,
+      "step": 220
+    },
+    {
+      "epoch": 0.09367088607594937,
+      "grad_norm": 0.6821302771568298,
+      "learning_rate": 3.884007029876977e-05,
+      "loss": 1.10588800907135,
+      "step": 222
+    },
+    {
+      "epoch": 0.09451476793248945,
+      "grad_norm": 0.693175733089447,
+      "learning_rate": 3.919156414762742e-05,
+      "loss": 1.1498671770095825,
+      "step": 224
+    },
+    {
+      "epoch": 0.09535864978902954,
+      "grad_norm": 0.5288166403770447,
+      "learning_rate": 3.954305799648506e-05,
+      "loss": 1.0587562322616577,
+      "step": 226
+    },
+    {
+      "epoch": 0.09620253164556962,
+      "grad_norm": 0.6882867813110352,
+      "learning_rate": 3.989455184534271e-05,
+      "loss": 1.1107512712478638,
+      "step": 228
+    },
+    {
+      "epoch": 0.0970464135021097,
+      "grad_norm": 0.5834154486656189,
+      "learning_rate": 4.024604569420035e-05,
+      "loss": 1.020510196685791,
+      "step": 230
+    },
+    {
+      "epoch": 0.09789029535864979,
+      "grad_norm": 0.7157064080238342,
+      "learning_rate": 4.0597539543058e-05,
+      "loss": 1.0642449855804443,
+      "step": 232
+    },
+    {
+      "epoch": 0.09873417721518987,
+      "grad_norm": 0.6530708074569702,
+      "learning_rate": 4.094903339191564e-05,
+      "loss": 1.0359872579574585,
+      "step": 234
+    },
+    {
+      "epoch": 0.09957805907172995,
+      "grad_norm": 0.6329686045646667,
+      "learning_rate": 4.130052724077329e-05,
+      "loss": 1.050504446029663,
+      "step": 236
+    },
+    {
+      "epoch": 0.10042194092827005,
+      "grad_norm": 0.6597026586532593,
+      "learning_rate": 4.165202108963093e-05,
+      "loss": 1.2621175050735474,
+      "step": 238
+    },
+    {
+      "epoch": 0.10126582278481013,
+      "grad_norm": 0.6195225119590759,
+      "learning_rate": 4.200351493848858e-05,
+      "loss": 1.1218310594558716,
+      "step": 240
+    },
+    {
+      "epoch": 0.1021097046413502,
+      "grad_norm": 0.6764137744903564,
+      "learning_rate": 4.235500878734622e-05,
+      "loss": 1.1250728368759155,
+      "step": 242
+    },
+    {
+      "epoch": 0.1029535864978903,
+      "grad_norm": 0.552363395690918,
+      "learning_rate": 4.270650263620387e-05,
+      "loss": 1.028212308883667,
+      "step": 244
+    },
+    {
+      "epoch": 0.10379746835443038,
+      "grad_norm": 0.5620495676994324,
+      "learning_rate": 4.305799648506151e-05,
+      "loss": 1.0425450801849365,
+      "step": 246
+    },
+    {
+      "epoch": 0.10464135021097046,
+      "grad_norm": 0.6860032081604004,
+      "learning_rate": 4.3409490333919156e-05,
+      "loss": 1.144278883934021,
+      "step": 248
+    },
+    {
+      "epoch": 0.10548523206751055,
+      "grad_norm": 0.6033259034156799,
+      "learning_rate": 4.37609841827768e-05,
+      "loss": 1.1223982572555542,
+      "step": 250
+    },
+    {
+      "epoch": 0.10632911392405063,
+      "grad_norm": 0.6292146444320679,
+      "learning_rate": 4.4112478031634446e-05,
+      "loss": 1.1609960794448853,
+      "step": 252
+    },
+    {
+      "epoch": 0.10717299578059072,
+      "grad_norm": 0.7982883453369141,
+      "learning_rate": 4.44639718804921e-05,
+      "loss": 1.063547968864441,
+      "step": 254
+    },
+    {
+      "epoch": 0.1080168776371308,
+      "grad_norm": 0.7719110250473022,
+      "learning_rate": 4.4815465729349736e-05,
+      "loss": 1.0719804763793945,
+      "step": 256
+    },
+    {
+      "epoch": 0.10886075949367088,
+      "grad_norm": 0.6101011633872986,
+      "learning_rate": 4.516695957820739e-05,
+      "loss": 1.0778400897979736,
+      "step": 258
+    },
+    {
+      "epoch": 0.10970464135021098,
+      "grad_norm": 0.7300994396209717,
+      "learning_rate": 4.5518453427065026e-05,
+      "loss": 1.2129558324813843,
+      "step": 260
+    },
+    {
+      "epoch": 0.11054852320675106,
+      "grad_norm": 0.8348747491836548,
+      "learning_rate": 4.586994727592268e-05,
+      "loss": 1.221714735031128,
+      "step": 262
+    },
+    {
+      "epoch": 0.11139240506329114,
+      "grad_norm": 0.5445612072944641,
+      "learning_rate": 4.6221441124780316e-05,
+      "loss": 1.0187978744506836,
+      "step": 264
+    },
+    {
+      "epoch": 0.11223628691983123,
+      "grad_norm": 0.6230319738388062,
+      "learning_rate": 4.657293497363797e-05,
+      "loss": 1.096561312675476,
+      "step": 266
+    },
+    {
+      "epoch": 0.11308016877637131,
+      "grad_norm": 0.6231237649917603,
+      "learning_rate": 4.6924428822495606e-05,
+      "loss": 1.089842438697815,
+      "step": 268
+    },
+    {
+      "epoch": 0.11392405063291139,
+      "grad_norm": 0.7178627252578735,
+      "learning_rate": 4.727592267135325e-05,
+      "loss": 1.0696645975112915,
+      "step": 270
+    },
+    {
+      "epoch": 0.11476793248945148,
+      "grad_norm": 0.6895854473114014,
+      "learning_rate": 4.7627416520210896e-05,
+      "loss": 1.0511361360549927,
+      "step": 272
+    },
+    {
+      "epoch": 0.11561181434599156,
+      "grad_norm": 0.6046878695487976,
+      "learning_rate": 4.797891036906854e-05,
+      "loss": 1.1373958587646484,
+      "step": 274
+    },
+    {
+      "epoch": 0.11645569620253164,
+      "grad_norm": 0.6524552702903748,
+      "learning_rate": 4.833040421792619e-05,
+      "loss": 1.0734186172485352,
+      "step": 276
+    },
+    {
+      "epoch": 0.11729957805907174,
+      "grad_norm": 0.6331019997596741,
+      "learning_rate": 4.868189806678383e-05,
+      "loss": 1.123913049697876,
+      "step": 278
+    },
+    {
+      "epoch": 0.11814345991561181,
+      "grad_norm": 0.5919018983840942,
+      "learning_rate": 4.903339191564148e-05,
+      "loss": 1.0635710954666138,
+      "step": 280
+    },
+    {
+      "epoch": 0.1189873417721519,
+      "grad_norm": 0.6067633032798767,
+      "learning_rate": 4.938488576449912e-05,
+      "loss": 1.0429247617721558,
+      "step": 282
+    },
+    {
+      "epoch": 0.11983122362869199,
+      "grad_norm": 0.6583750247955322,
+      "learning_rate": 4.9736379613356774e-05,
+      "loss": 1.1397464275360107,
+      "step": 284
+    },
+    {
+      "epoch": 0.12067510548523207,
+      "grad_norm": 0.6200069785118103,
+      "learning_rate": 5.008787346221442e-05,
+      "loss": 1.0590803623199463,
+      "step": 286
+    },
+    {
+      "epoch": 0.12151898734177215,
+      "grad_norm": 0.6798665523529053,
+      "learning_rate": 5.0439367311072064e-05,
+      "loss": 1.1318789720535278,
+      "step": 288
+    },
+    {
+      "epoch": 0.12236286919831224,
+      "grad_norm": 0.7508794069290161,
+      "learning_rate": 5.07908611599297e-05,
+      "loss": 1.0934956073760986,
+      "step": 290
+    },
+    {
+      "epoch": 0.12320675105485232,
+      "grad_norm": 0.6901452541351318,
+      "learning_rate": 5.114235500878735e-05,
+      "loss": 1.163407802581787,
+      "step": 292
+    },
+    {
+      "epoch": 0.1240506329113924,
+      "grad_norm": 0.6423285603523254,
+      "learning_rate": 5.1493848857644985e-05,
+      "loss": 1.09059476852417,
+      "step": 294
+    },
+    {
+      "epoch": 0.1248945147679325,
+      "grad_norm": 0.6839275360107422,
+      "learning_rate": 5.1845342706502644e-05,
+      "loss": 1.0690211057662964,
+      "step": 296
+    },
+    {
+      "epoch": 0.1257383966244726,
+      "grad_norm": 0.6350128054618835,
+      "learning_rate": 5.219683655536028e-05,
+      "loss": 0.982322096824646,
+      "step": 298
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "grad_norm": 0.7136530876159668,
+      "learning_rate": 5.254833040421793e-05,
+      "loss": 1.1132930517196655,
+      "step": 300
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "eval_loss": 1.0952109098434448,
+      "eval_runtime": 677.0652,
+      "eval_samples_per_second": 3.112,
+      "eval_steps_per_second": 3.112,
+      "step": 300
+    },
+    {
+      "epoch": 0.12742616033755275,
+      "grad_norm": 0.7339721322059631,
+      "learning_rate": 5.289982425307557e-05,
+      "loss": 0.973595917224884,
+      "step": 302
+    },
+    {
+      "epoch": 0.12827004219409283,
+      "grad_norm": 0.5941481590270996,
+      "learning_rate": 5.3251318101933224e-05,
+      "loss": 0.9819849729537964,
+      "step": 304
+    },
+    {
+      "epoch": 0.1291139240506329,
+      "grad_norm": 0.7153938412666321,
+      "learning_rate": 5.360281195079086e-05,
+      "loss": 1.0315470695495605,
+      "step": 306
+    },
+    {
+      "epoch": 0.12995780590717299,
+      "grad_norm": 0.5167180299758911,
+      "learning_rate": 5.395430579964851e-05,
+      "loss": 0.9492001533508301,
+      "step": 308
+    },
+    {
+      "epoch": 0.1308016877637131,
+      "grad_norm": 0.6055944561958313,
+      "learning_rate": 5.430579964850615e-05,
+      "loss": 1.0156209468841553,
+      "step": 310
+    },
+    {
+      "epoch": 0.13164556962025317,
+      "grad_norm": 0.7662386298179626,
+      "learning_rate": 5.4657293497363805e-05,
+      "loss": 1.1791651248931885,
+      "step": 312
+    },
+    {
+      "epoch": 0.13248945147679325,
+      "grad_norm": 0.6065546274185181,
+      "learning_rate": 5.500878734622145e-05,
+      "loss": 1.0009297132492065,
+      "step": 314
+    },
+    {
+      "epoch": 0.13333333333333333,
+      "grad_norm": 0.604225754737854,
+      "learning_rate": 5.536028119507909e-05,
+      "loss": 1.0208244323730469,
+      "step": 316
+    },
+    {
+      "epoch": 0.1341772151898734,
+      "grad_norm": 0.6186763048171997,
+      "learning_rate": 5.571177504393673e-05,
+      "loss": 0.9968416690826416,
+      "step": 318
+    },
+    {
+      "epoch": 0.1350210970464135,
+      "grad_norm": 0.7100363969802856,
+      "learning_rate": 5.606326889279437e-05,
+      "loss": 0.9540256857872009,
+      "step": 320
+    },
+    {
+      "epoch": 0.1358649789029536,
+      "grad_norm": 0.6979711055755615,
+      "learning_rate": 5.641476274165203e-05,
+      "loss": 1.0631953477859497,
+      "step": 322
+    },
+    {
+      "epoch": 0.13670886075949368,
+      "grad_norm": 0.6237109303474426,
+      "learning_rate": 5.676625659050967e-05,
+      "loss": 1.0170501470565796,
+      "step": 324
+    },
+    {
+      "epoch": 0.13755274261603376,
+      "grad_norm": 0.6525548696517944,
+      "learning_rate": 5.711775043936731e-05,
+      "loss": 1.0715603828430176,
+      "step": 326
+    },
+    {
+      "epoch": 0.13839662447257384,
+      "grad_norm": 0.6869221329689026,
+      "learning_rate": 5.746924428822495e-05,
+      "loss": 1.0111541748046875,
+      "step": 328
+    },
+    {
+      "epoch": 0.13924050632911392,
+      "grad_norm": 0.553188145160675,
+      "learning_rate": 5.782073813708261e-05,
+      "loss": 1.0311682224273682,
+      "step": 330
+    },
+    {
+      "epoch": 0.140084388185654,
+      "grad_norm": 0.6760852932929993,
+      "learning_rate": 5.817223198594025e-05,
+      "loss": 1.0213634967803955,
+      "step": 332
+    },
+    {
+      "epoch": 0.1409282700421941,
+      "grad_norm": 0.5907419919967651,
+      "learning_rate": 5.8523725834797894e-05,
+      "loss": 0.9748594164848328,
+      "step": 334
+    },
+    {
+      "epoch": 0.14177215189873418,
+      "grad_norm": 0.7044920921325684,
+      "learning_rate": 5.887521968365554e-05,
+      "loss": 1.05863356590271,
+      "step": 336
+    },
+    {
+      "epoch": 0.14261603375527426,
+      "grad_norm": 0.679073691368103,
+      "learning_rate": 5.922671353251318e-05,
+      "loss": 1.1341127157211304,
+      "step": 338
+    },
+    {
+      "epoch": 0.14345991561181434,
+      "grad_norm": 0.7676237225532532,
+      "learning_rate": 5.957820738137083e-05,
+      "loss": 0.9540836215019226,
+      "step": 340
+    },
+    {
+      "epoch": 0.14430379746835442,
+      "grad_norm": 0.6313899755477905,
+      "learning_rate": 5.9929701230228474e-05,
+      "loss": 1.0585911273956299,
+      "step": 342
+    },
+    {
+      "epoch": 0.1451476793248945,
+      "grad_norm": 0.7123099565505981,
+      "learning_rate": 6.028119507908612e-05,
+      "loss": 1.0760118961334229,
+      "step": 344
+    },
+    {
+      "epoch": 0.1459915611814346,
+      "grad_norm": 0.585935652256012,
+      "learning_rate": 6.063268892794376e-05,
+      "loss": 1.036866307258606,
+      "step": 346
+    },
+    {
+      "epoch": 0.1468354430379747,
+      "grad_norm": 0.5643263459205627,
+      "learning_rate": 6.0984182776801416e-05,
+      "loss": 1.0242938995361328,
+      "step": 348
+    },
+    {
+      "epoch": 0.14767932489451477,
+      "grad_norm": 0.626761794090271,
+      "learning_rate": 6.133567662565906e-05,
+      "loss": 1.0497376918792725,
+      "step": 350
+    },
+    {
+      "epoch": 0.14852320675105485,
+      "grad_norm": 0.5106956958770752,
+      "learning_rate": 6.16871704745167e-05,
+      "loss": 0.9811885356903076,
+      "step": 352
+    },
+    {
+      "epoch": 0.14936708860759493,
+      "grad_norm": 0.6948089003562927,
+      "learning_rate": 6.203866432337434e-05,
+      "loss": 1.0715330839157104,
+      "step": 354
+    },
+    {
+      "epoch": 0.150210970464135,
+      "grad_norm": 0.699713945388794,
+      "learning_rate": 6.239015817223199e-05,
+      "loss": 1.0405226945877075,
+      "step": 356
+    },
+    {
+      "epoch": 0.15105485232067511,
+      "grad_norm": 0.6437667012214661,
+      "learning_rate": 6.274165202108964e-05,
+      "loss": 1.0490930080413818,
+      "step": 358
+    },
+    {
+      "epoch": 0.1518987341772152,
+      "grad_norm": 0.6952699422836304,
+      "learning_rate": 6.309314586994728e-05,
+      "loss": 0.9267548322677612,
+      "step": 360
+    },
+    {
+      "epoch": 0.15274261603375527,
+      "grad_norm": 0.6713186502456665,
+      "learning_rate": 6.344463971880492e-05,
+      "loss": 1.0427420139312744,
+      "step": 362
+    },
+    {
+      "epoch": 0.15358649789029535,
+      "grad_norm": 0.6750379800796509,
+      "learning_rate": 6.379613356766257e-05,
+      "loss": 1.048950433731079,
+      "step": 364
+    },
+    {
+      "epoch": 0.15443037974683543,
+      "grad_norm": 0.6053379774093628,
+      "learning_rate": 6.414762741652022e-05,
+      "loss": 1.0156004428863525,
+      "step": 366
+    },
+    {
+      "epoch": 0.15527426160337554,
+      "grad_norm": 0.8063633441925049,
+      "learning_rate": 6.449912126537786e-05,
+      "loss": 1.0020819902420044,
+      "step": 368
+    },
+    {
+      "epoch": 0.15611814345991562,
+      "grad_norm": 0.8027494549751282,
+      "learning_rate": 6.48506151142355e-05,
+      "loss": 1.055633783340454,
+      "step": 370
+    },
+    {
+      "epoch": 0.1569620253164557,
+      "grad_norm": 0.6580121517181396,
+      "learning_rate": 6.520210896309315e-05,
+      "loss": 1.0149940252304077,
+      "step": 372
+    },
+    {
+      "epoch": 0.15780590717299578,
+      "grad_norm": 0.6561233997344971,
+      "learning_rate": 6.55536028119508e-05,
+      "loss": 0.9769611954689026,
+      "step": 374
+    },
+    {
+      "epoch": 0.15864978902953586,
+      "grad_norm": 0.6444346308708191,
+      "learning_rate": 6.590509666080844e-05,
+      "loss": 0.9099349975585938,
+      "step": 376
+    },
+    {
+      "epoch": 0.15949367088607594,
+      "grad_norm": 0.5879359245300293,
+      "learning_rate": 6.625659050966608e-05,
+      "loss": 1.0797548294067383,
+      "step": 378
+    },
+    {
+      "epoch": 0.16033755274261605,
+      "grad_norm": 0.6994144916534424,
+      "learning_rate": 6.660808435852373e-05,
+      "loss": 1.0336791276931763,
+      "step": 380
+    },
+    {
+      "epoch": 0.16118143459915613,
+      "grad_norm": 0.6128669381141663,
+      "learning_rate": 6.695957820738138e-05,
+      "loss": 1.018118143081665,
+      "step": 382
+    },
+    {
+      "epoch": 0.1620253164556962,
+      "grad_norm": 1.0237540006637573,
+      "learning_rate": 6.731107205623902e-05,
+      "loss": 1.1405497789382935,
+      "step": 384
+    },
+    {
+      "epoch": 0.16286919831223629,
+      "grad_norm": 0.6091578006744385,
+      "learning_rate": 6.766256590509666e-05,
+      "loss": 1.0314189195632935,
+      "step": 386
+    },
+    {
+      "epoch": 0.16371308016877636,
+      "grad_norm": 0.5916037559509277,
+      "learning_rate": 6.801405975395431e-05,
+      "loss": 0.9564052820205688,
+      "step": 388
+    },
+    {
+      "epoch": 0.16455696202531644,
+      "grad_norm": 0.771653950214386,
+      "learning_rate": 6.836555360281195e-05,
+      "loss": 1.0023859739303589,
+      "step": 390
+    },
+    {
+      "epoch": 0.16540084388185655,
+      "grad_norm": 0.654658317565918,
+      "learning_rate": 6.87170474516696e-05,
+      "loss": 1.07024085521698,
+      "step": 392
+    },
+    {
+      "epoch": 0.16624472573839663,
+      "grad_norm": 0.6611968874931335,
+      "learning_rate": 6.906854130052724e-05,
+      "loss": 1.0552500486373901,
+      "step": 394
+    },
+    {
+      "epoch": 0.1670886075949367,
+      "grad_norm": 0.6955893039703369,
+      "learning_rate": 6.942003514938489e-05,
+      "loss": 1.0562875270843506,
+      "step": 396
+    },
+    {
+      "epoch": 0.1679324894514768,
+      "grad_norm": 0.6666058301925659,
+      "learning_rate": 6.977152899824253e-05,
+      "loss": 0.9850592017173767,
+      "step": 398
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "grad_norm": 0.6131711006164551,
+      "learning_rate": 7.012302284710018e-05,
+      "loss": 1.0077755451202393,
+      "step": 400
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "eval_loss": 1.0625108480453491,
+      "eval_runtime": 691.0068,
+      "eval_samples_per_second": 3.049,
+      "eval_steps_per_second": 3.049,
+      "step": 400
+    },
+    {
+      "epoch": 0.16962025316455695,
+      "grad_norm": 0.6286499500274658,
+      "learning_rate": 7.047451669595783e-05,
+      "loss": 1.1012427806854248,
+      "step": 402
+    },
+    {
+      "epoch": 0.17046413502109706,
+      "grad_norm": 0.6639351844787598,
+      "learning_rate": 7.082601054481547e-05,
+      "loss": 1.0379719734191895,
+      "step": 404
+    },
+    {
+      "epoch": 0.17130801687763714,
+      "grad_norm": 0.750401496887207,
+      "learning_rate": 7.117750439367311e-05,
+      "loss": 1.031856656074524,
+      "step": 406
+    },
+    {
+      "epoch": 0.17215189873417722,
+      "grad_norm": 0.8084847331047058,
+      "learning_rate": 7.152899824253075e-05,
+      "loss": 1.0493193864822388,
+      "step": 408
+    },
+    {
+      "epoch": 0.1729957805907173,
+      "grad_norm": 0.7448462247848511,
+      "learning_rate": 7.188049209138841e-05,
+      "loss": 1.1012418270111084,
+      "step": 410
+    },
+    {
+      "epoch": 0.17383966244725738,
+      "grad_norm": 0.5841867923736572,
+      "learning_rate": 7.223198594024605e-05,
+      "loss": 0.9926692247390747,
+      "step": 412
+    },
+    {
+      "epoch": 0.17468354430379746,
+      "grad_norm": 0.7125606536865234,
+      "learning_rate": 7.258347978910369e-05,
+      "loss": 1.0588877201080322,
+      "step": 414
+    },
+    {
+      "epoch": 0.17552742616033756,
+      "grad_norm": 0.5750942230224609,
+      "learning_rate": 7.293497363796134e-05,
+      "loss": 1.038270354270935,
+      "step": 416
+    },
+    {
+      "epoch": 0.17637130801687764,
+      "grad_norm": 0.565444827079773,
+      "learning_rate": 7.328646748681899e-05,
+      "loss": 0.9843021035194397,
+      "step": 418
+    },
+    {
+      "epoch": 0.17721518987341772,
+      "grad_norm": 0.5825693011283875,
+      "learning_rate": 7.363796133567663e-05,
+      "loss": 1.0731632709503174,
+      "step": 420
+    },
+    {
+      "epoch": 0.1780590717299578,
+      "grad_norm": 0.6267391443252563,
+      "learning_rate": 7.398945518453427e-05,
+      "loss": 1.0061273574829102,
+      "step": 422
+    },
+    {
+      "epoch": 0.17890295358649788,
+      "grad_norm": 0.6621372103691101,
+      "learning_rate": 7.434094903339192e-05,
+      "loss": 1.0461612939834595,
+      "step": 424
+    },
+    {
+      "epoch": 0.17974683544303796,
+      "grad_norm": 0.6635435223579407,
+      "learning_rate": 7.469244288224957e-05,
+      "loss": 0.9789207577705383,
+      "step": 426
+    },
+    {
+      "epoch": 0.18059071729957807,
+      "grad_norm": 0.6342346668243408,
+      "learning_rate": 7.504393673110721e-05,
+      "loss": 1.0527069568634033,
+      "step": 428
+    },
+    {
+      "epoch": 0.18143459915611815,
+      "grad_norm": 0.6762149930000305,
+      "learning_rate": 7.539543057996485e-05,
+      "loss": 0.9708702564239502,
+      "step": 430
+    },
+    {
+      "epoch": 0.18227848101265823,
+      "grad_norm": 0.7073282599449158,
+      "learning_rate": 7.57469244288225e-05,
+      "loss": 1.0509834289550781,
+      "step": 432
+    },
+    {
+      "epoch": 0.1831223628691983,
+      "grad_norm": 0.6917856931686401,
+      "learning_rate": 7.609841827768014e-05,
+      "loss": 1.0128819942474365,
+      "step": 434
+    },
+    {
+      "epoch": 0.1839662447257384,
+      "grad_norm": 0.5574942231178284,
+      "learning_rate": 7.644991212653779e-05,
+      "loss": 0.989395797252655,
+      "step": 436
+    },
+    {
+      "epoch": 0.1848101265822785,
+      "grad_norm": 0.640765905380249,
+      "learning_rate": 7.680140597539543e-05,
+      "loss": 0.9846042990684509,
+      "step": 438
+    },
+    {
+      "epoch": 0.18565400843881857,
+      "grad_norm": 0.6699127554893494,
+      "learning_rate": 7.715289982425308e-05,
+      "loss": 1.0344442129135132,
+      "step": 440
+    },
+    {
+      "epoch": 0.18649789029535865,
+      "grad_norm": 0.6164930462837219,
+      "learning_rate": 7.750439367311072e-05,
+      "loss": 1.0179373025894165,
+      "step": 442
+    },
+    {
+      "epoch": 0.18734177215189873,
+      "grad_norm": 0.6880720853805542,
+      "learning_rate": 7.785588752196837e-05,
+      "loss": 1.0518895387649536,
+      "step": 444
+    },
+    {
+      "epoch": 0.1881856540084388,
+      "grad_norm": 0.6501413583755493,
+      "learning_rate": 7.820738137082601e-05,
+      "loss": 1.0442606210708618,
+      "step": 446
+    },
+    {
+      "epoch": 0.1890295358649789,
+      "grad_norm": 0.6076085567474365,
+      "learning_rate": 7.855887521968366e-05,
+      "loss": 0.9828442335128784,
+      "step": 448
+    },
+    {
+      "epoch": 0.189873417721519,
+      "grad_norm": 0.6418202519416809,
+      "learning_rate": 7.89103690685413e-05,
+      "loss": 1.0573710203170776,
+      "step": 450
+    },
+    {
+      "epoch": 0.19071729957805908,
+      "grad_norm": 0.7055076360702515,
+      "learning_rate": 7.926186291739895e-05,
+      "loss": 1.0216103792190552,
+      "step": 452
+    },
+    {
+      "epoch": 0.19156118143459916,
+      "grad_norm": 0.5668330192565918,
+      "learning_rate": 7.961335676625659e-05,
+      "loss": 0.9837722778320312,
+      "step": 454
+    },
+    {
+      "epoch": 0.19240506329113924,
+      "grad_norm": 0.6419380307197571,
+      "learning_rate": 7.996485061511424e-05,
+      "loss": 1.0003894567489624,
+      "step": 456
+    },
+    {
+      "epoch": 0.19324894514767932,
+      "grad_norm": 0.5949198007583618,
+      "learning_rate": 8.031634446397188e-05,
+      "loss": 1.0609031915664673,
+      "step": 458
+    },
+    {
+      "epoch": 0.1940928270042194,
+      "grad_norm": 0.7032039761543274,
+      "learning_rate": 8.066783831282952e-05,
+      "loss": 1.0543403625488281,
+      "step": 460
+    },
+    {
+      "epoch": 0.1949367088607595,
+      "grad_norm": 0.5775868892669678,
+      "learning_rate": 8.101933216168718e-05,
+      "loss": 0.9819303154945374,
+      "step": 462
+    },
+    {
+      "epoch": 0.19578059071729959,
+      "grad_norm": 0.9301062226295471,
+      "learning_rate": 8.137082601054482e-05,
+      "loss": 1.0542067289352417,
+      "step": 464
+    },
+    {
+      "epoch": 0.19662447257383966,
+      "grad_norm": 0.6193217039108276,
+      "learning_rate": 8.172231985940246e-05,
+      "loss": 0.9966341257095337,
+      "step": 466
+    },
+    {
+      "epoch": 0.19746835443037974,
+      "grad_norm": 0.6286146640777588,
+      "learning_rate": 8.20738137082601e-05,
+      "loss": 1.0474121570587158,
+      "step": 468
+    },
+    {
+      "epoch": 0.19831223628691982,
+      "grad_norm": 0.7418972253799438,
+      "learning_rate": 8.242530755711776e-05,
+      "loss": 0.9549239277839661,
+      "step": 470
+    },
+    {
+      "epoch": 0.1991561181434599,
+      "grad_norm": 0.6122808456420898,
+      "learning_rate": 8.27768014059754e-05,
+      "loss": 1.0191338062286377,
+      "step": 472
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.6375362277030945,
+      "learning_rate": 8.312829525483304e-05,
+      "loss": 1.0987539291381836,
+      "step": 474
+    },
+    {
+      "epoch": 0.2008438818565401,
+      "grad_norm": 0.6459513306617737,
+      "learning_rate": 8.347978910369068e-05,
+      "loss": 1.0369136333465576,
+      "step": 476
+    },
+    {
+      "epoch": 0.20168776371308017,
+      "grad_norm": 0.7029640674591064,
+      "learning_rate": 8.383128295254833e-05,
+      "loss": 1.0582096576690674,
+      "step": 478
+    },
+    {
+      "epoch": 0.20253164556962025,
+      "grad_norm": 0.6345387697219849,
+      "learning_rate": 8.418277680140598e-05,
+      "loss": 1.022916316986084,
+      "step": 480
+    },
+    {
+      "epoch": 0.20337552742616033,
+      "grad_norm": 0.5764590501785278,
+      "learning_rate": 8.453427065026362e-05,
+      "loss": 0.973024308681488,
+      "step": 482
+    },
+    {
+      "epoch": 0.2042194092827004,
+      "grad_norm": 0.5884482860565186,
+      "learning_rate": 8.488576449912127e-05,
+      "loss": 1.0292812585830688,
+      "step": 484
+    },
+    {
+      "epoch": 0.20506329113924052,
+      "grad_norm": 0.616357147693634,
+      "learning_rate": 8.523725834797891e-05,
+      "loss": 1.0083447694778442,
+      "step": 486
+    },
+    {
+      "epoch": 0.2059071729957806,
+      "grad_norm": 0.7671196460723877,
+      "learning_rate": 8.558875219683656e-05,
+      "loss": 0.9936985373497009,
+      "step": 488
+    },
+    {
+      "epoch": 0.20675105485232068,
+      "grad_norm": 0.6197299957275391,
+      "learning_rate": 8.59402460456942e-05,
+      "loss": 1.051513910293579,
+      "step": 490
+    },
+    {
+      "epoch": 0.20759493670886076,
+      "grad_norm": 0.6912890672683716,
+      "learning_rate": 8.629173989455185e-05,
+      "loss": 0.9474978446960449,
+      "step": 492
+    },
+    {
+      "epoch": 0.20843881856540084,
+      "grad_norm": 0.6941592693328857,
+      "learning_rate": 8.664323374340949e-05,
+      "loss": 1.0671660900115967,
+      "step": 494
+    },
+    {
+      "epoch": 0.20928270042194091,
+      "grad_norm": 0.5889528393745422,
+      "learning_rate": 8.699472759226714e-05,
+      "loss": 1.0020159482955933,
+      "step": 496
+    },
+    {
+      "epoch": 0.21012658227848102,
+      "grad_norm": 0.6478549838066101,
+      "learning_rate": 8.734622144112478e-05,
+      "loss": 1.0165860652923584,
+      "step": 498
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "grad_norm": 0.6018255949020386,
+      "learning_rate": 8.769771528998243e-05,
+      "loss": 0.8798263072967529,
+      "step": 500
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "eval_loss": 1.042096495628357,
+      "eval_runtime": 692.4361,
+      "eval_samples_per_second": 3.043,
+      "eval_steps_per_second": 3.043,
+      "step": 500
+    },
+    {
+      "epoch": 0.21181434599156118,
+      "grad_norm": 0.578990638256073,
+      "learning_rate": 8.804920913884007e-05,
+      "loss": 1.092096209526062,
+      "step": 502
+    },
+    {
+      "epoch": 0.21265822784810126,
+      "grad_norm": 0.6597883701324463,
+      "learning_rate": 8.840070298769771e-05,
+      "loss": 1.0413451194763184,
+      "step": 504
+    },
+    {
+      "epoch": 0.21350210970464134,
+      "grad_norm": 0.6660305261611938,
+      "learning_rate": 8.875219683655536e-05,
+      "loss": 1.0073142051696777,
+      "step": 506
+    },
+    {
+      "epoch": 0.21434599156118145,
+      "grad_norm": 0.6283115148544312,
+      "learning_rate": 8.910369068541301e-05,
+      "loss": 1.0319768190383911,
+      "step": 508
+    },
+    {
+      "epoch": 0.21518987341772153,
+      "grad_norm": 0.6257343292236328,
+      "learning_rate": 8.945518453427065e-05,
+      "loss": 1.0046353340148926,
+      "step": 510
+    },
+    {
+      "epoch": 0.2160337552742616,
+      "grad_norm": 0.5530875325202942,
+      "learning_rate": 8.980667838312829e-05,
+      "loss": 0.9169099926948547,
+      "step": 512
+    },
+    {
+      "epoch": 0.2168776371308017,
+      "grad_norm": 0.5369633436203003,
+      "learning_rate": 9.015817223198594e-05,
+      "loss": 0.9081505537033081,
+      "step": 514
+    },
+    {
+      "epoch": 0.21772151898734177,
+      "grad_norm": 0.6618232131004333,
+      "learning_rate": 9.05096660808436e-05,
+      "loss": 0.9165045022964478,
+      "step": 516
+    },
+    {
+      "epoch": 0.21856540084388185,
+      "grad_norm": 0.600666344165802,
+      "learning_rate": 9.086115992970123e-05,
+      "loss": 0.91348797082901,
+      "step": 518
+    },
+    {
+      "epoch": 0.21940928270042195,
+      "grad_norm": 0.5919831991195679,
+      "learning_rate": 9.121265377855887e-05,
+      "loss": 1.006508469581604,
+      "step": 520
+    },
+    {
+      "epoch": 0.22025316455696203,
+      "grad_norm": 0.688058614730835,
+      "learning_rate": 9.156414762741654e-05,
+      "loss": 1.0013236999511719,
+      "step": 522
+    },
+    {
+      "epoch": 0.2210970464135021,
+      "grad_norm": 0.6721227765083313,
+      "learning_rate": 9.191564147627418e-05,
+      "loss": 1.0909923315048218,
+      "step": 524
+    },
+    {
+      "epoch": 0.2219409282700422,
+      "grad_norm": 0.5987313389778137,
+      "learning_rate": 9.226713532513181e-05,
+      "loss": 1.0117096900939941,
+      "step": 526
+    },
+    {
+      "epoch": 0.22278481012658227,
+      "grad_norm": 0.6191489696502686,
+      "learning_rate": 9.261862917398945e-05,
+      "loss": 1.0153647661209106,
+      "step": 528
+    },
+    {
+      "epoch": 0.22362869198312235,
+      "grad_norm": 0.6821563243865967,
+      "learning_rate": 9.29701230228471e-05,
+      "loss": 0.9649755954742432,
+      "step": 530
+    },
+    {
+      "epoch": 0.22447257383966246,
+      "grad_norm": 1.760398268699646,
+      "learning_rate": 9.332161687170476e-05,
+      "loss": 0.8673232197761536,
+      "step": 532
+    },
+    {
+      "epoch": 0.22531645569620254,
+      "grad_norm": 0.6670058369636536,
+      "learning_rate": 9.36731107205624e-05,
+      "loss": 0.9942440986633301,
+      "step": 534
+    },
+    {
+      "epoch": 0.22616033755274262,
+      "grad_norm": 0.7345916032791138,
+      "learning_rate": 9.402460456942003e-05,
+      "loss": 1.0364389419555664,
+      "step": 536
+    },
+    {
+      "epoch": 0.2270042194092827,
+      "grad_norm": 0.5946128964424133,
+      "learning_rate": 9.437609841827768e-05,
+      "loss": 0.9314924478530884,
+      "step": 538
+    },
+    {
+      "epoch": 0.22784810126582278,
+      "grad_norm": 0.5800848603248596,
+      "learning_rate": 9.472759226713534e-05,
+      "loss": 1.0694862604141235,
+      "step": 540
+    },
+    {
+      "epoch": 0.22869198312236286,
+      "grad_norm": 0.6712192893028259,
+      "learning_rate": 9.507908611599297e-05,
+      "loss": 1.03531014919281,
+      "step": 542
+    },
+    {
+      "epoch": 0.22953586497890296,
+      "grad_norm": 0.5641416311264038,
+      "learning_rate": 9.543057996485063e-05,
+      "loss": 0.9795235991477966,
+      "step": 544
+    },
+    {
+      "epoch": 0.23037974683544304,
+      "grad_norm": 0.50412517786026,
+      "learning_rate": 9.578207381370826e-05,
+      "loss": 0.9641494750976562,
+      "step": 546
+    },
+    {
+      "epoch": 0.23122362869198312,
+      "grad_norm": 0.579118549823761,
+      "learning_rate": 9.61335676625659e-05,
+      "loss": 0.9375281929969788,
+      "step": 548
+    },
+    {
+      "epoch": 0.2320675105485232,
+      "grad_norm": 0.5888341665267944,
+      "learning_rate": 9.648506151142355e-05,
+      "loss": 0.9414046406745911,
+      "step": 550
+    },
+    {
+      "epoch": 0.23291139240506328,
+      "grad_norm": 0.5595056414604187,
+      "learning_rate": 9.68365553602812e-05,
+      "loss": 0.9005617499351501,
+      "step": 552
+    },
+    {
+      "epoch": 0.23375527426160336,
+      "grad_norm": 0.6605326533317566,
+      "learning_rate": 9.718804920913884e-05,
+      "loss": 1.0283968448638916,
+      "step": 554
+    },
+    {
+      "epoch": 0.23459915611814347,
+      "grad_norm": 0.5657313466072083,
+      "learning_rate": 9.753954305799648e-05,
+      "loss": 1.0058249235153198,
+      "step": 556
+    },
+    {
+      "epoch": 0.23544303797468355,
+      "grad_norm": 0.5433364510536194,
+      "learning_rate": 9.789103690685413e-05,
+      "loss": 0.9835494756698608,
+      "step": 558
+    },
+    {
+      "epoch": 0.23628691983122363,
+      "grad_norm": 0.6129802465438843,
+      "learning_rate": 9.824253075571179e-05,
+      "loss": 1.054532527923584,
+      "step": 560
+    },
+    {
+      "epoch": 0.2371308016877637,
+      "grad_norm": 0.6496239304542542,
+      "learning_rate": 9.859402460456942e-05,
+      "loss": 1.0240973234176636,
+      "step": 562
+    },
+    {
+      "epoch": 0.2379746835443038,
+      "grad_norm": 0.6380873918533325,
+      "learning_rate": 9.894551845342706e-05,
+      "loss": 1.0229179859161377,
+      "step": 564
+    },
+    {
+      "epoch": 0.23881856540084387,
+      "grad_norm": 0.6151993870735168,
+      "learning_rate": 9.929701230228471e-05,
+      "loss": 1.0111570358276367,
+      "step": 566
+    },
+    {
+      "epoch": 0.23966244725738398,
+      "grad_norm": 0.5727584958076477,
+      "learning_rate": 9.964850615114237e-05,
+      "loss": 0.9450829029083252,
+      "step": 568
+    },
+    {
+      "epoch": 0.24050632911392406,
+      "grad_norm": 0.6620725989341736,
+      "learning_rate": 0.0001,
+      "loss": 0.9800319075584412,
+      "step": 570
+    },
+    {
+      "epoch": 0.24135021097046414,
+      "grad_norm": 0.6151163578033447,
+      "learning_rate": 0.00010035149384885764,
+      "loss": 0.9757438898086548,
+      "step": 572
+    },
+    {
+      "epoch": 0.24219409282700421,
+      "grad_norm": 0.5672140717506409,
+      "learning_rate": 0.0001007029876977153,
+      "loss": 0.9104921817779541,
+      "step": 574
+    },
+    {
+      "epoch": 0.2430379746835443,
+      "grad_norm": 0.5697256326675415,
+      "learning_rate": 0.00010105448154657293,
+      "loss": 1.1027376651763916,
+      "step": 576
+    },
+    {
+      "epoch": 0.2438818565400844,
+      "grad_norm": 0.5590381622314453,
+      "learning_rate": 0.00010140597539543057,
+      "loss": 1.0055404901504517,
+      "step": 578
+    },
+    {
+      "epoch": 0.24472573839662448,
+      "grad_norm": 0.5518567562103271,
+      "learning_rate": 0.00010175746924428824,
+      "loss": 1.020835518836975,
+      "step": 580
+    },
+    {
+      "epoch": 0.24556962025316456,
+      "grad_norm": 0.6338496208190918,
+      "learning_rate": 0.00010210896309314588,
+      "loss": 0.9528344869613647,
+      "step": 582
+    },
+    {
+      "epoch": 0.24641350210970464,
+      "grad_norm": 0.6497329473495483,
+      "learning_rate": 0.00010246045694200353,
+      "loss": 1.0088670253753662,
+      "step": 584
+    },
+    {
+      "epoch": 0.24725738396624472,
+      "grad_norm": 0.49888095259666443,
+      "learning_rate": 0.00010281195079086117,
+      "loss": 0.9961200952529907,
+      "step": 586
+    },
+    {
+      "epoch": 0.2481012658227848,
+      "grad_norm": 0.5680158734321594,
+      "learning_rate": 0.0001031634446397188,
+      "loss": 0.9635610580444336,
+      "step": 588
+    },
+    {
+      "epoch": 0.2489451476793249,
+      "grad_norm": 0.658168375492096,
+      "learning_rate": 0.00010351493848857646,
+      "loss": 0.9392287135124207,
+      "step": 590
+    },
+    {
+      "epoch": 0.249789029535865,
+      "grad_norm": 0.618262767791748,
+      "learning_rate": 0.0001038664323374341,
+      "loss": 0.9600516557693481,
+      "step": 592
+    },
+    {
+      "epoch": 0.25063291139240507,
+      "grad_norm": 0.6003909111022949,
+      "learning_rate": 0.00010421792618629173,
+      "loss": 1.005476713180542,
+      "step": 594
+    },
+    {
+      "epoch": 0.2514767932489452,
+      "grad_norm": 0.5437078475952148,
+      "learning_rate": 0.00010456942003514938,
+      "loss": 0.9523017406463623,
+      "step": 596
+    },
+    {
+      "epoch": 0.2523206751054852,
+      "grad_norm": 0.5524541735649109,
+      "learning_rate": 0.00010492091388400705,
+      "loss": 0.9526668787002563,
+      "step": 598
+    },
+    {
+      "epoch": 0.25316455696202533,
+      "grad_norm": 0.679504930973053,
+      "learning_rate": 0.00010527240773286469,
+      "loss": 1.019660472869873,
+      "step": 600
+    },
+    {
+      "epoch": 0.25316455696202533,
+      "eval_loss": 1.0193854570388794,
+      "eval_runtime": 677.9523,
+      "eval_samples_per_second": 3.108,
+      "eval_steps_per_second": 3.108,
+      "step": 600
+    },
+    {
+      "epoch": 0.2540084388185654,
+      "grad_norm": 0.5646136999130249,
+      "learning_rate": 0.00010562390158172233,
+      "loss": 0.9910882711410522,
+      "step": 602
+    },
+    {
+      "epoch": 0.2548523206751055,
+      "grad_norm": 0.5238093137741089,
+      "learning_rate": 0.00010597539543057998,
+      "loss": 0.9616432785987854,
+      "step": 604
+    },
+    {
+      "epoch": 0.25569620253164554,
+      "grad_norm": 0.7483857274055481,
+      "learning_rate": 0.00010632688927943762,
+      "loss": 1.0078275203704834,
+      "step": 606
+    },
+    {
+      "epoch": 0.25654008438818565,
+      "grad_norm": 0.578948974609375,
+      "learning_rate": 0.00010667838312829525,
+      "loss": 0.9827103018760681,
+      "step": 608
+    },
+    {
+      "epoch": 0.25738396624472576,
+      "grad_norm": 0.5525906085968018,
+      "learning_rate": 0.00010702987697715289,
+      "loss": 1.0423277616500854,
+      "step": 610
+    },
+    {
+      "epoch": 0.2582278481012658,
+      "grad_norm": 0.6721326112747192,
+      "learning_rate": 0.00010738137082601054,
+      "loss": 0.9561693072319031,
+      "step": 612
+    },
+    {
+      "epoch": 0.2590717299578059,
+      "grad_norm": 0.5701051354408264,
+      "learning_rate": 0.00010773286467486821,
+      "loss": 0.9602992534637451,
+      "step": 614
+    },
+    {
+      "epoch": 0.25991561181434597,
+      "grad_norm": 0.6349860429763794,
+      "learning_rate": 0.00010808435852372585,
+      "loss": 1.1422650814056396,
+      "step": 616
+    },
+    {
+      "epoch": 0.2607594936708861,
+      "grad_norm": 0.5496085286140442,
+      "learning_rate": 0.00010843585237258349,
+      "loss": 0.9762773513793945,
+      "step": 618
+    },
+    {
+      "epoch": 0.2616033755274262,
+      "grad_norm": 0.6080722808837891,
+      "learning_rate": 0.00010878734622144114,
+      "loss": 1.0133616924285889,
+      "step": 620
+    },
+    {
+      "epoch": 0.26244725738396624,
+      "grad_norm": 0.5450218915939331,
+      "learning_rate": 0.00010913884007029878,
+      "loss": 0.9385587573051453,
+      "step": 622
+    },
+    {
+      "epoch": 0.26329113924050634,
+      "grad_norm": 0.592106819152832,
+      "learning_rate": 0.00010949033391915641,
+      "loss": 0.9359989762306213,
+      "step": 624
+    },
+    {
+      "epoch": 0.2641350210970464,
+      "grad_norm": 0.6449427604675293,
+      "learning_rate": 0.00010984182776801407,
+      "loss": 1.0266027450561523,
+      "step": 626
+    },
+    {
+      "epoch": 0.2649789029535865,
+      "grad_norm": 0.538299560546875,
+      "learning_rate": 0.0001101933216168717,
+      "loss": 0.9303187131881714,
+      "step": 628
+    },
+    {
+      "epoch": 0.26582278481012656,
+      "grad_norm": 0.546316921710968,
+      "learning_rate": 0.00011054481546572934,
+      "loss": 0.9368857145309448,
+      "step": 630
+    },
+    {
+      "epoch": 0.26666666666666666,
+      "grad_norm": 0.5818730592727661,
+      "learning_rate": 0.00011089630931458701,
+      "loss": 0.9573145508766174,
+      "step": 632
+    },
+    {
+      "epoch": 0.26751054852320677,
+      "grad_norm": 0.5958262085914612,
+      "learning_rate": 0.00011124780316344465,
+      "loss": 0.9345449805259705,
+      "step": 634
+    },
+    {
+      "epoch": 0.2683544303797468,
+      "grad_norm": 0.6259077787399292,
+      "learning_rate": 0.0001115992970123023,
+      "loss": 1.0906590223312378,
+      "step": 636
+    },
+    {
+      "epoch": 0.26919831223628693,
+      "grad_norm": 0.589672863483429,
+      "learning_rate": 0.00011195079086115994,
+      "loss": 1.0757447481155396,
+      "step": 638
+    },
+    {
+      "epoch": 0.270042194092827,
+      "grad_norm": 0.5714080333709717,
+      "learning_rate": 0.00011230228471001758,
+      "loss": 0.9310855269432068,
+      "step": 640
+    },
+    {
+      "epoch": 0.2708860759493671,
+      "grad_norm": 0.45342639088630676,
+      "learning_rate": 0.00011265377855887523,
+      "loss": 0.9276360273361206,
+      "step": 642
+    },
+    {
+      "epoch": 0.2717299578059072,
+      "grad_norm": 0.6386750340461731,
+      "learning_rate": 0.00011300527240773287,
+      "loss": 1.084719181060791,
+      "step": 644
+    },
+    {
+      "epoch": 0.27257383966244725,
+      "grad_norm": 0.6446163654327393,
+      "learning_rate": 0.0001133567662565905,
+      "loss": 0.9763918519020081,
+      "step": 646
+    },
+    {
+      "epoch": 0.27341772151898736,
+      "grad_norm": 0.5925686359405518,
+      "learning_rate": 0.00011370826010544816,
+      "loss": 0.9517921805381775,
+      "step": 648
+    },
+    {
+      "epoch": 0.2742616033755274,
+      "grad_norm": 0.5399773716926575,
+      "learning_rate": 0.00011405975395430582,
+      "loss": 1.0587927103042603,
+      "step": 650
+    },
+    {
+      "epoch": 0.2751054852320675,
+      "grad_norm": 0.5872456431388855,
+      "learning_rate": 0.00011441124780316346,
+      "loss": 0.883341908454895,
+      "step": 652
+    },
+    {
+      "epoch": 0.2759493670886076,
+      "grad_norm": 0.5574564337730408,
+      "learning_rate": 0.0001147627416520211,
+      "loss": 1.0306891202926636,
+      "step": 654
+    },
+    {
+      "epoch": 0.2767932489451477,
+      "grad_norm": 0.47789013385772705,
+      "learning_rate": 0.00011511423550087874,
+      "loss": 0.8814032077789307,
+      "step": 656
+    },
+    {
+      "epoch": 0.2776371308016878,
+      "grad_norm": 0.5565530061721802,
+      "learning_rate": 0.00011546572934973639,
+      "loss": 0.9460552334785461,
+      "step": 658
+    },
+    {
+      "epoch": 0.27848101265822783,
+      "grad_norm": 0.5299761295318604,
+      "learning_rate": 0.00011581722319859403,
+      "loss": 0.9475110769271851,
+      "step": 660
+    },
+    {
+      "epoch": 0.27932489451476794,
+      "grad_norm": 0.6503344178199768,
+      "learning_rate": 0.00011616871704745166,
+      "loss": 1.0630913972854614,
+      "step": 662
+    },
+    {
+      "epoch": 0.280168776371308,
+      "grad_norm": 0.5794585943222046,
+      "learning_rate": 0.00011652021089630932,
+      "loss": 0.9389138221740723,
+      "step": 664
+    },
+    {
+      "epoch": 0.2810126582278481,
+      "grad_norm": 0.5762867331504822,
+      "learning_rate": 0.00011687170474516695,
+      "loss": 0.8934136033058167,
+      "step": 666
+    },
+    {
+      "epoch": 0.2818565400843882,
+      "grad_norm": 0.6565435528755188,
+      "learning_rate": 0.00011722319859402462,
+      "loss": 1.1072614192962646,
+      "step": 668
+    },
+    {
+      "epoch": 0.28270042194092826,
+      "grad_norm": 0.5819830298423767,
+      "learning_rate": 0.00011757469244288226,
+      "loss": 1.0501434803009033,
+      "step": 670
+    },
+    {
+      "epoch": 0.28354430379746837,
+      "grad_norm": 0.6071487069129944,
+      "learning_rate": 0.00011792618629173991,
+      "loss": 0.9880793690681458,
+      "step": 672
+    },
+    {
+      "epoch": 0.2843881856540084,
+      "grad_norm": 0.5765058398246765,
+      "learning_rate": 0.00011827768014059755,
+      "loss": 0.9670693874359131,
+      "step": 674
+    },
+    {
+      "epoch": 0.2852320675105485,
+      "grad_norm": 0.5245351791381836,
+      "learning_rate": 0.00011862917398945519,
+      "loss": 0.9602360725402832,
+      "step": 676
+    },
+    {
+      "epoch": 0.28607594936708863,
+      "grad_norm": 0.6189922094345093,
+      "learning_rate": 0.00011898066783831282,
+      "loss": 0.9684560894966125,
+      "step": 678
+    },
+    {
+      "epoch": 0.2869198312236287,
+      "grad_norm": 0.6138690710067749,
+      "learning_rate": 0.00011933216168717048,
+      "loss": 0.9465792775154114,
+      "step": 680
+    },
+    {
+      "epoch": 0.2877637130801688,
+      "grad_norm": 0.5371595621109009,
+      "learning_rate": 0.00011968365553602812,
+      "loss": 0.8495944738388062,
+      "step": 682
+    },
+    {
+      "epoch": 0.28860759493670884,
+      "grad_norm": 0.5549944639205933,
+      "learning_rate": 0.00012003514938488578,
+      "loss": 0.9663267135620117,
+      "step": 684
+    },
+    {
+      "epoch": 0.28945147679324895,
+      "grad_norm": 0.6484189033508301,
+      "learning_rate": 0.00012038664323374342,
+      "loss": 0.9736058712005615,
+      "step": 686
+    },
+    {
+      "epoch": 0.290295358649789,
+      "grad_norm": 0.540351390838623,
+      "learning_rate": 0.00012073813708260107,
+      "loss": 1.0591845512390137,
+      "step": 688
+    },
+    {
+      "epoch": 0.2911392405063291,
+      "grad_norm": 0.5657922029495239,
+      "learning_rate": 0.00012108963093145871,
+      "loss": 0.944908618927002,
+      "step": 690
+    },
+    {
+      "epoch": 0.2919831223628692,
+      "grad_norm": 0.6040505170822144,
+      "learning_rate": 0.00012144112478031635,
+      "loss": 1.0018219947814941,
+      "step": 692
+    },
+    {
+      "epoch": 0.29282700421940927,
+      "grad_norm": 0.5435477495193481,
+      "learning_rate": 0.000121792618629174,
+      "loss": 1.0351502895355225,
+      "step": 694
+    },
+    {
+      "epoch": 0.2936708860759494,
+      "grad_norm": 0.5712518692016602,
+      "learning_rate": 0.00012214411247803164,
+      "loss": 0.9935672283172607,
+      "step": 696
+    },
+    {
+      "epoch": 0.29451476793248943,
+      "grad_norm": 0.6138222813606262,
+      "learning_rate": 0.00012249560632688928,
+      "loss": 1.0165108442306519,
+      "step": 698
+    },
+    {
+      "epoch": 0.29535864978902954,
+      "grad_norm": 0.4495212435722351,
+      "learning_rate": 0.00012284710017574691,
+      "loss": 0.9334425926208496,
+      "step": 700
+    },
+    {
+      "epoch": 0.29535864978902954,
+      "eval_loss": 0.996929407119751,
+      "eval_runtime": 668.6398,
+      "eval_samples_per_second": 3.151,
+      "eval_steps_per_second": 3.151,
+      "step": 700
+    },
+    {
+      "epoch": 0.29620253164556964,
+      "grad_norm": 0.5321539044380188,
+      "learning_rate": 0.00012319859402460458,
+      "loss": 1.0516537427902222,
+      "step": 702
+    },
+    {
+      "epoch": 0.2970464135021097,
+      "grad_norm": 0.5716516971588135,
+      "learning_rate": 0.00012355008787346222,
+      "loss": 0.9387198686599731,
+      "step": 704
+    },
+    {
+      "epoch": 0.2978902953586498,
+      "grad_norm": 0.5617920160293579,
+      "learning_rate": 0.00012390158172231988,
+      "loss": 0.99737948179245,
+      "step": 706
+    },
+    {
+      "epoch": 0.29873417721518986,
+      "grad_norm": 0.4922899007797241,
+      "learning_rate": 0.00012425307557117752,
+      "loss": 0.9955025911331177,
+      "step": 708
+    },
+    {
+      "epoch": 0.29957805907172996,
+      "grad_norm": 0.543501615524292,
+      "learning_rate": 0.00012460456942003516,
+      "loss": 0.9124280214309692,
+      "step": 710
+    },
+    {
+      "epoch": 0.30042194092827,
+      "grad_norm": 0.49590054154396057,
+      "learning_rate": 0.0001249560632688928,
+      "loss": 0.9820216298103333,
+      "step": 712
+    },
+    {
+      "epoch": 0.3012658227848101,
+      "grad_norm": 0.5984305739402771,
+      "learning_rate": 0.00012530755711775044,
+      "loss": 1.0152074098587036,
+      "step": 714
+    },
+    {
+      "epoch": 0.30210970464135023,
+      "grad_norm": 0.9343504905700684,
+      "learning_rate": 0.00012565905096660807,
+      "loss": 1.0577725172042847,
+      "step": 716
+    },
+    {
+      "epoch": 0.3029535864978903,
+      "grad_norm": 0.5118702054023743,
+      "learning_rate": 0.0001260105448154657,
+      "loss": 0.9830358028411865,
+      "step": 718
+    },
+    {
+      "epoch": 0.3037974683544304,
+      "grad_norm": 0.4940392076969147,
+      "learning_rate": 0.00012636203866432338,
+      "loss": 0.9466043710708618,
+      "step": 720
+    },
+    {
+      "epoch": 0.30464135021097044,
+      "grad_norm": 0.5965693593025208,
+      "learning_rate": 0.00012671353251318104,
+      "loss": 1.015270709991455,
+      "step": 722
+    },
+    {
+      "epoch": 0.30548523206751055,
+      "grad_norm": 0.5020529627799988,
+      "learning_rate": 0.00012706502636203868,
+      "loss": 0.9703927636146545,
+      "step": 724
+    },
+    {
+      "epoch": 0.30632911392405066,
+      "grad_norm": 0.6067010164260864,
+      "learning_rate": 0.00012741652021089632,
+      "loss": 1.0255526304244995,
+      "step": 726
+    },
+    {
+      "epoch": 0.3071729957805907,
+      "grad_norm": 0.5931884050369263,
+      "learning_rate": 0.00012776801405975396,
+      "loss": 0.9335633516311646,
+      "step": 728
+    },
+    {
+      "epoch": 0.3080168776371308,
+      "grad_norm": 0.5938752293586731,
+      "learning_rate": 0.0001281195079086116,
+      "loss": 1.0921578407287598,
+      "step": 730
+    },
+    {
+      "epoch": 0.30886075949367087,
+      "grad_norm": 0.49728086590766907,
+      "learning_rate": 0.00012847100175746923,
+      "loss": 0.963066041469574,
+      "step": 732
+    },
+    {
+      "epoch": 0.309704641350211,
+      "grad_norm": 0.5452080965042114,
+      "learning_rate": 0.0001288224956063269,
+      "loss": 0.9513075351715088,
+      "step": 734
+    },
+    {
+      "epoch": 0.3105485232067511,
+      "grad_norm": 0.5497731566429138,
+      "learning_rate": 0.00012917398945518454,
+      "loss": 0.8576077222824097,
+      "step": 736
+    },
+    {
+      "epoch": 0.31139240506329113,
+      "grad_norm": 0.5580397248268127,
+      "learning_rate": 0.0001295254833040422,
+      "loss": 0.9542577862739563,
+      "step": 738
+    },
+    {
+      "epoch": 0.31223628691983124,
+      "grad_norm": 0.5890427827835083,
+      "learning_rate": 0.00012987697715289984,
+      "loss": 0.8992732167243958,
+      "step": 740
+    },
+    {
+      "epoch": 0.3130801687763713,
+      "grad_norm": 0.5942965745925903,
+      "learning_rate": 0.00013022847100175748,
+      "loss": 1.0322896242141724,
+      "step": 742
+    },
+    {
+      "epoch": 0.3139240506329114,
+      "grad_norm": 0.6341713070869446,
+      "learning_rate": 0.00013057996485061512,
+      "loss": 0.9217103719711304,
+      "step": 744
+    },
+    {
+      "epoch": 0.31476793248945145,
+      "grad_norm": 0.5294105410575867,
+      "learning_rate": 0.00013093145869947276,
+      "loss": 0.951789915561676,
+      "step": 746
+    },
+    {
+      "epoch": 0.31561181434599156,
+      "grad_norm": 0.6372058391571045,
+      "learning_rate": 0.0001312829525483304,
+      "loss": 0.9459875226020813,
+      "step": 748
+    },
+    {
+      "epoch": 0.31645569620253167,
+      "grad_norm": 0.5979796648025513,
+      "learning_rate": 0.00013163444639718806,
+      "loss": 0.9626097679138184,
+      "step": 750
+    },
+    {
+      "epoch": 0.3172995780590717,
+      "grad_norm": 0.5682399868965149,
+      "learning_rate": 0.0001319859402460457,
+      "loss": 1.0261781215667725,
+      "step": 752
+    },
+    {
+      "epoch": 0.3181434599156118,
+      "grad_norm": 0.5349125266075134,
+      "learning_rate": 0.00013233743409490336,
+      "loss": 0.9319828152656555,
+      "step": 754
+    },
+    {
+      "epoch": 0.3189873417721519,
+      "grad_norm": 0.6093934178352356,
+      "learning_rate": 0.000132688927943761,
+      "loss": 0.9216550588607788,
+      "step": 756
+    },
+    {
+      "epoch": 0.319831223628692,
+      "grad_norm": 0.5188612341880798,
+      "learning_rate": 0.00013304042179261864,
+      "loss": 0.901739776134491,
+      "step": 758
+    },
+    {
+      "epoch": 0.3206751054852321,
+      "grad_norm": 0.5877130627632141,
+      "learning_rate": 0.00013339191564147628,
+      "loss": 1.0362589359283447,
+      "step": 760
+    },
+    {
+      "epoch": 0.32151898734177214,
+      "grad_norm": 0.5542771816253662,
+      "learning_rate": 0.00013374340949033392,
+      "loss": 0.8787116408348083,
+      "step": 762
+    },
+    {
+      "epoch": 0.32236286919831225,
+      "grad_norm": 0.5084902048110962,
+      "learning_rate": 0.00013409490333919156,
+      "loss": 0.9237037301063538,
+      "step": 764
+    },
+    {
+      "epoch": 0.3232067510548523,
+      "grad_norm": 0.5461528301239014,
+      "learning_rate": 0.00013444639718804922,
+      "loss": 1.0150731801986694,
+      "step": 766
+    },
+    {
+      "epoch": 0.3240506329113924,
+      "grad_norm": 0.53483647108078,
+      "learning_rate": 0.00013479789103690686,
+      "loss": 0.8985214829444885,
+      "step": 768
+    },
+    {
+      "epoch": 0.32489451476793246,
+      "grad_norm": 0.5580531358718872,
+      "learning_rate": 0.0001351493848857645,
+      "loss": 1.0225775241851807,
+      "step": 770
+    },
+    {
+      "epoch": 0.32573839662447257,
+      "grad_norm": 0.5203377604484558,
+      "learning_rate": 0.00013550087873462216,
+      "loss": 0.9571293592453003,
+      "step": 772
+    },
+    {
+      "epoch": 0.3265822784810127,
+      "grad_norm": 0.5049671530723572,
+      "learning_rate": 0.0001358523725834798,
+      "loss": 1.0468909740447998,
+      "step": 774
+    },
+    {
+      "epoch": 0.32742616033755273,
+      "grad_norm": 0.4723063111305237,
+      "learning_rate": 0.00013620386643233744,
+      "loss": 0.7743215560913086,
+      "step": 776
+    },
+    {
+      "epoch": 0.32827004219409284,
+      "grad_norm": 0.6310980916023254,
+      "learning_rate": 0.00013655536028119508,
+      "loss": 1.021510362625122,
+      "step": 778
+    },
+    {
+      "epoch": 0.3291139240506329,
+      "grad_norm": 0.47066664695739746,
+      "learning_rate": 0.00013690685413005274,
+      "loss": 0.9134382605552673,
+      "step": 780
+    },
+    {
+      "epoch": 0.329957805907173,
+      "grad_norm": 0.5725092887878418,
+      "learning_rate": 0.00013725834797891038,
+      "loss": 0.9797834753990173,
+      "step": 782
+    },
+    {
+      "epoch": 0.3308016877637131,
+      "grad_norm": 0.5139563083648682,
+      "learning_rate": 0.00013760984182776802,
+      "loss": 0.9372621178627014,
+      "step": 784
+    },
+    {
+      "epoch": 0.33164556962025316,
+      "grad_norm": 0.5275821685791016,
+      "learning_rate": 0.00013796133567662566,
+      "loss": 0.9528245329856873,
+      "step": 786
+    },
+    {
+      "epoch": 0.33248945147679326,
+      "grad_norm": 0.5702582001686096,
+      "learning_rate": 0.0001383128295254833,
+      "loss": 0.9750176072120667,
+      "step": 788
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.5281293392181396,
+      "learning_rate": 0.00013866432337434096,
+      "loss": 0.9412306547164917,
+      "step": 790
+    },
+    {
+      "epoch": 0.3341772151898734,
+      "grad_norm": 0.5578986406326294,
+      "learning_rate": 0.0001390158172231986,
+      "loss": 0.997580885887146,
+      "step": 792
+    },
+    {
+      "epoch": 0.33502109704641353,
+      "grad_norm": 0.50461345911026,
+      "learning_rate": 0.00013936731107205624,
+      "loss": 0.9082320928573608,
+      "step": 794
+    },
+    {
+      "epoch": 0.3358649789029536,
+      "grad_norm": 0.5258530378341675,
+      "learning_rate": 0.0001397188049209139,
+      "loss": 1.0082844495773315,
+      "step": 796
+    },
+    {
+      "epoch": 0.3367088607594937,
+      "grad_norm": 0.5548169016838074,
+      "learning_rate": 0.00014007029876977154,
+      "loss": 0.9729003310203552,
+      "step": 798
+    },
+    {
+      "epoch": 0.33755274261603374,
+      "grad_norm": 0.48601076006889343,
+      "learning_rate": 0.00014042179261862918,
+      "loss": 0.9099526405334473,
+      "step": 800
+    },
+    {
+      "epoch": 0.33755274261603374,
+      "eval_loss": 0.9800403714179993,
+      "eval_runtime": 678.8306,
+      "eval_samples_per_second": 3.104,
+      "eval_steps_per_second": 3.104,
+      "step": 800
+    },
+    {
+      "epoch": 0.33839662447257385,
+      "grad_norm": 0.5413158535957336,
+      "learning_rate": 0.00014077328646748682,
+      "loss": 0.8610644936561584,
+      "step": 802
+    },
+    {
+      "epoch": 0.3392405063291139,
+      "grad_norm": 0.5147035717964172,
+      "learning_rate": 0.00014112478031634446,
+      "loss": 0.9584825038909912,
+      "step": 804
+    },
+    {
+      "epoch": 0.340084388185654,
+      "grad_norm": 0.5931771397590637,
+      "learning_rate": 0.0001414762741652021,
+      "loss": 1.0142558813095093,
+      "step": 806
+    },
+    {
+      "epoch": 0.3409282700421941,
+      "grad_norm": 0.5178377032279968,
+      "learning_rate": 0.00014182776801405976,
+      "loss": 1.0078763961791992,
+      "step": 808
+    },
+    {
+      "epoch": 0.34177215189873417,
+      "grad_norm": 0.5453237295150757,
+      "learning_rate": 0.0001421792618629174,
+      "loss": 0.9107215404510498,
+      "step": 810
+    },
+    {
+      "epoch": 0.3426160337552743,
+      "grad_norm": 0.5886152982711792,
+      "learning_rate": 0.00014253075571177506,
+      "loss": 0.9981362819671631,
+      "step": 812
+    },
+    {
+      "epoch": 0.3434599156118143,
+      "grad_norm": 0.48040178418159485,
+      "learning_rate": 0.0001428822495606327,
+      "loss": 0.9636131525039673,
+      "step": 814
+    },
+    {
+      "epoch": 0.34430379746835443,
+      "grad_norm": 0.5011753439903259,
+      "learning_rate": 0.00014323374340949034,
+      "loss": 0.9590586423873901,
+      "step": 816
+    },
+    {
+      "epoch": 0.34514767932489454,
+      "grad_norm": 0.57858806848526,
+      "learning_rate": 0.00014358523725834798,
+      "loss": 0.978246808052063,
+      "step": 818
+    },
+    {
+      "epoch": 0.3459915611814346,
+      "grad_norm": 0.46092939376831055,
+      "learning_rate": 0.00014393673110720562,
+      "loss": 0.8549934029579163,
+      "step": 820
+    },
+    {
+      "epoch": 0.3468354430379747,
+      "grad_norm": 0.5756489038467407,
+      "learning_rate": 0.00014428822495606326,
+      "loss": 0.9771265387535095,
+      "step": 822
+    },
+    {
+      "epoch": 0.34767932489451475,
+      "grad_norm": 0.5501731634140015,
+      "learning_rate": 0.00014463971880492092,
+      "loss": 0.8739748001098633,
+      "step": 824
+    },
+    {
+      "epoch": 0.34852320675105486,
+      "grad_norm": 0.5451868176460266,
+      "learning_rate": 0.0001449912126537786,
+      "loss": 0.9129468202590942,
+      "step": 826
+    },
+    {
+      "epoch": 0.3493670886075949,
+      "grad_norm": 0.4624619781970978,
+      "learning_rate": 0.00014534270650263623,
+      "loss": 0.9196704030036926,
+      "step": 828
+    },
+    {
+      "epoch": 0.350210970464135,
+      "grad_norm": 0.520878791809082,
+      "learning_rate": 0.00014569420035149386,
+      "loss": 0.9976527690887451,
+      "step": 830
+    },
+    {
+      "epoch": 0.3510548523206751,
+      "grad_norm": 0.4469171464443207,
+      "learning_rate": 0.0001460456942003515,
+      "loss": 0.7753443717956543,
+      "step": 832
+    },
+    {
+      "epoch": 0.3518987341772152,
+      "grad_norm": 0.5105249881744385,
+      "learning_rate": 0.00014639718804920914,
+      "loss": 0.9584846496582031,
+      "step": 834
+    },
+    {
+      "epoch": 0.3527426160337553,
+      "grad_norm": 0.5043913125991821,
+      "learning_rate": 0.00014674868189806678,
+      "loss": 0.9013500213623047,
+      "step": 836
+    },
+    {
+      "epoch": 0.35358649789029534,
+      "grad_norm": 0.575850784778595,
+      "learning_rate": 0.00014710017574692442,
+      "loss": 0.9775562286376953,
+      "step": 838
+    },
+    {
+      "epoch": 0.35443037974683544,
+      "grad_norm": 0.5128876566886902,
+      "learning_rate": 0.00014745166959578208,
+      "loss": 0.9278940558433533,
+      "step": 840
+    },
+    {
+      "epoch": 0.35527426160337555,
+      "grad_norm": 0.5757885575294495,
+      "learning_rate": 0.00014780316344463975,
+      "loss": 1.0091488361358643,
+      "step": 842
+    },
+    {
+      "epoch": 0.3561181434599156,
+      "grad_norm": 0.500934898853302,
+      "learning_rate": 0.00014815465729349739,
+      "loss": 0.9286836981773376,
+      "step": 844
+    },
+    {
+      "epoch": 0.3569620253164557,
+      "grad_norm": 0.5220686197280884,
+      "learning_rate": 0.00014850615114235502,
+      "loss": 0.9484171867370605,
+      "step": 846
+    },
+    {
+      "epoch": 0.35780590717299576,
+      "grad_norm": 0.5494697690010071,
+      "learning_rate": 0.00014885764499121266,
+      "loss": 1.0556397438049316,
+      "step": 848
+    },
+    {
+      "epoch": 0.35864978902953587,
+      "grad_norm": 0.46633943915367126,
+      "learning_rate": 0.0001492091388400703,
+      "loss": 0.933089017868042,
+      "step": 850
+    },
+    {
+      "epoch": 0.3594936708860759,
+      "grad_norm": 0.4864962697029114,
+      "learning_rate": 0.00014956063268892794,
+      "loss": 1.0016963481903076,
+      "step": 852
+    },
+    {
+      "epoch": 0.36033755274261603,
+      "grad_norm": 0.5032764673233032,
+      "learning_rate": 0.0001499121265377856,
+      "loss": 0.9085348844528198,
+      "step": 854
+    },
+    {
+      "epoch": 0.36118143459915614,
+      "grad_norm": 0.5424998998641968,
+      "learning_rate": 0.00015026362038664324,
+      "loss": 0.923502504825592,
+      "step": 856
+    },
+    {
+      "epoch": 0.3620253164556962,
+      "grad_norm": 0.5181655287742615,
+      "learning_rate": 0.00015061511423550088,
+      "loss": 0.919174313545227,
+      "step": 858
+    },
+    {
+      "epoch": 0.3628691983122363,
+      "grad_norm": 0.5129443407058716,
+      "learning_rate": 0.00015096660808435855,
+      "loss": 0.851981520652771,
+      "step": 860
+    },
+    {
+      "epoch": 0.36371308016877635,
+      "grad_norm": 0.49540698528289795,
+      "learning_rate": 0.00015131810193321618,
+      "loss": 0.8633858561515808,
+      "step": 862
+    },
+    {
+      "epoch": 0.36455696202531646,
+      "grad_norm": 0.4706701934337616,
+      "learning_rate": 0.00015166959578207382,
+      "loss": 0.9473677277565002,
+      "step": 864
+    },
+    {
+      "epoch": 0.36540084388185656,
+      "grad_norm": 0.587704598903656,
+      "learning_rate": 0.00015202108963093146,
+      "loss": 0.942383885383606,
+      "step": 866
+    },
+    {
+      "epoch": 0.3662447257383966,
+      "grad_norm": 0.5851273536682129,
+      "learning_rate": 0.0001523725834797891,
+      "loss": 0.9811251163482666,
+      "step": 868
+    },
+    {
+      "epoch": 0.3670886075949367,
+      "grad_norm": 0.46357613801956177,
+      "learning_rate": 0.00015272407732864676,
+      "loss": 0.8151084184646606,
+      "step": 870
+    },
+    {
+      "epoch": 0.3679324894514768,
+      "grad_norm": 0.47250670194625854,
+      "learning_rate": 0.0001530755711775044,
+      "loss": 0.9023333191871643,
+      "step": 872
+    },
+    {
+      "epoch": 0.3687763713080169,
+      "grad_norm": 0.47510042786598206,
+      "learning_rate": 0.00015342706502636204,
+      "loss": 0.8758499622344971,
+      "step": 874
+    },
+    {
+      "epoch": 0.369620253164557,
+      "grad_norm": 0.5687124729156494,
+      "learning_rate": 0.00015377855887521968,
+      "loss": 0.9837421774864197,
+      "step": 876
+    },
+    {
+      "epoch": 0.37046413502109704,
+      "grad_norm": 0.49064236879348755,
+      "learning_rate": 0.00015413005272407735,
+      "loss": 0.9084216356277466,
+      "step": 878
+    },
+    {
+      "epoch": 0.37130801687763715,
+      "grad_norm": 0.5288164615631104,
+      "learning_rate": 0.00015448154657293498,
+      "loss": 0.8425542712211609,
+      "step": 880
+    },
+    {
+      "epoch": 0.3721518987341772,
+      "grad_norm": 0.5446951985359192,
+      "learning_rate": 0.00015483304042179262,
+      "loss": 1.006197452545166,
+      "step": 882
+    },
+    {
+      "epoch": 0.3729957805907173,
+      "grad_norm": 0.46872642636299133,
+      "learning_rate": 0.00015518453427065026,
+      "loss": 0.9779444932937622,
+      "step": 884
+    },
+    {
+      "epoch": 0.37383966244725736,
+      "grad_norm": 0.4852714240550995,
+      "learning_rate": 0.00015553602811950793,
+      "loss": 0.9099963307380676,
+      "step": 886
+    },
+    {
+      "epoch": 0.37468354430379747,
+      "grad_norm": 0.5219841003417969,
+      "learning_rate": 0.00015588752196836556,
+      "loss": 0.9730570316314697,
+      "step": 888
+    },
+    {
+      "epoch": 0.3755274261603376,
+      "grad_norm": 0.5258626341819763,
+      "learning_rate": 0.0001562390158172232,
+      "loss": 0.9304586052894592,
+      "step": 890
+    },
+    {
+      "epoch": 0.3763713080168776,
+      "grad_norm": 0.4266716241836548,
+      "learning_rate": 0.00015659050966608084,
+      "loss": 0.8680716156959534,
+      "step": 892
+    },
+    {
+      "epoch": 0.37721518987341773,
+      "grad_norm": 0.46361327171325684,
+      "learning_rate": 0.0001569420035149385,
+      "loss": 0.8428018093109131,
+      "step": 894
+    },
+    {
+      "epoch": 0.3780590717299578,
+      "grad_norm": 0.5313687920570374,
+      "learning_rate": 0.00015729349736379614,
+      "loss": 0.8465750217437744,
+      "step": 896
+    },
+    {
+      "epoch": 0.3789029535864979,
+      "grad_norm": 0.47166210412979126,
+      "learning_rate": 0.00015764499121265378,
+      "loss": 0.9120327234268188,
+      "step": 898
+    },
+    {
+      "epoch": 0.379746835443038,
+      "grad_norm": 0.4647318720817566,
+      "learning_rate": 0.00015799648506151145,
+      "loss": 0.8950425982475281,
+      "step": 900
+    },
+    {
+      "epoch": 0.379746835443038,
+      "eval_loss": 0.9643027186393738,
+      "eval_runtime": 691.7929,
+      "eval_samples_per_second": 3.046,
+      "eval_steps_per_second": 3.046,
+      "step": 900
+    },
+    {
+      "epoch": 0.38059071729957805,
+      "grad_norm": 0.5445119738578796,
+      "learning_rate": 0.00015834797891036909,
+      "loss": 0.908163845539093,
+      "step": 902
+    },
+    {
+      "epoch": 0.38143459915611816,
+      "grad_norm": 0.4311858117580414,
+      "learning_rate": 0.00015869947275922672,
+      "loss": 0.8945821523666382,
+      "step": 904
+    },
+    {
+      "epoch": 0.3822784810126582,
+      "grad_norm": 0.5590984225273132,
+      "learning_rate": 0.00015905096660808436,
+      "loss": 0.9478458762168884,
+      "step": 906
+    },
+    {
+      "epoch": 0.3831223628691983,
+      "grad_norm": 0.5470241904258728,
+      "learning_rate": 0.000159402460456942,
+      "loss": 0.9259957671165466,
+      "step": 908
+    },
+    {
+      "epoch": 0.38396624472573837,
+      "grad_norm": 0.5498791337013245,
+      "learning_rate": 0.00015975395430579964,
+      "loss": 0.8824930787086487,
+      "step": 910
+    },
+    {
+      "epoch": 0.3848101265822785,
+      "grad_norm": 0.4779198467731476,
+      "learning_rate": 0.0001601054481546573,
+      "loss": 0.8842340707778931,
+      "step": 912
+    },
+    {
+      "epoch": 0.3856540084388186,
+      "grad_norm": 0.5390620827674866,
+      "learning_rate": 0.00016045694200351494,
+      "loss": 0.92950040102005,
+      "step": 914
+    },
+    {
+      "epoch": 0.38649789029535864,
+      "grad_norm": 0.505519688129425,
+      "learning_rate": 0.0001608084358523726,
+      "loss": 0.8420897126197815,
+      "step": 916
+    },
+    {
+      "epoch": 0.38734177215189874,
+      "grad_norm": 0.4463907778263092,
+      "learning_rate": 0.00016115992970123025,
+      "loss": 0.8256624341011047,
+      "step": 918
+    },
+    {
+      "epoch": 0.3881856540084388,
+      "grad_norm": 0.5321422219276428,
+      "learning_rate": 0.00016151142355008788,
+      "loss": 0.8701168298721313,
+      "step": 920
+    },
+    {
+      "epoch": 0.3890295358649789,
+      "grad_norm": 0.4343073070049286,
+      "learning_rate": 0.00016186291739894552,
+      "loss": 0.9019309878349304,
+      "step": 922
+    },
+    {
+      "epoch": 0.389873417721519,
+      "grad_norm": 0.5311984419822693,
+      "learning_rate": 0.00016221441124780316,
+      "loss": 0.8560551404953003,
+      "step": 924
+    },
+    {
+      "epoch": 0.39071729957805906,
+      "grad_norm": 0.588691771030426,
+      "learning_rate": 0.0001625659050966608,
+      "loss": 0.9700050354003906,
+      "step": 926
+    },
+    {
+      "epoch": 0.39156118143459917,
+      "grad_norm": 0.5425586104393005,
+      "learning_rate": 0.00016291739894551844,
+      "loss": 1.043768048286438,
+      "step": 928
+    },
+    {
+      "epoch": 0.3924050632911392,
+      "grad_norm": 0.5228736996650696,
+      "learning_rate": 0.0001632688927943761,
+      "loss": 0.9501712918281555,
+      "step": 930
+    },
+    {
+      "epoch": 0.39324894514767933,
+      "grad_norm": 0.48960360884666443,
+      "learning_rate": 0.00016362038664323377,
+      "loss": 0.9223058223724365,
+      "step": 932
+    },
+    {
+      "epoch": 0.39409282700421944,
+      "grad_norm": 0.45204755663871765,
+      "learning_rate": 0.0001639718804920914,
+      "loss": 0.9692960977554321,
+      "step": 934
+    },
+    {
+      "epoch": 0.3949367088607595,
+      "grad_norm": 0.5299274921417236,
+      "learning_rate": 0.00016432337434094905,
+      "loss": 0.9467466473579407,
+      "step": 936
+    },
+    {
+      "epoch": 0.3957805907172996,
+      "grad_norm": 0.5607715249061584,
+      "learning_rate": 0.00016467486818980668,
+      "loss": 0.9118053317070007,
+      "step": 938
+    },
+    {
+      "epoch": 0.39662447257383965,
+      "grad_norm": 0.5271831154823303,
+      "learning_rate": 0.00016502636203866432,
+      "loss": 0.9131460189819336,
+      "step": 940
+    },
+    {
+      "epoch": 0.39746835443037976,
+      "grad_norm": 0.5075286030769348,
+      "learning_rate": 0.00016537785588752196,
+      "loss": 0.9358300566673279,
+      "step": 942
+    },
+    {
+      "epoch": 0.3983122362869198,
+      "grad_norm": 0.515731155872345,
+      "learning_rate": 0.00016572934973637963,
+      "loss": 0.8908210396766663,
+      "step": 944
+    },
+    {
+      "epoch": 0.3991561181434599,
+      "grad_norm": 0.4856977164745331,
+      "learning_rate": 0.00016608084358523726,
+      "loss": 0.9775290489196777,
+      "step": 946
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.48846355080604553,
+      "learning_rate": 0.00016643233743409493,
+      "loss": 0.8957490921020508,
+      "step": 948
+    },
+    {
+      "epoch": 0.4008438818565401,
+      "grad_norm": 0.42990800738334656,
+      "learning_rate": 0.00016678383128295257,
+      "loss": 0.9036174416542053,
+      "step": 950
+    },
+    {
+      "epoch": 0.4016877637130802,
+      "grad_norm": 0.49552062153816223,
+      "learning_rate": 0.0001671353251318102,
+      "loss": 0.991032600402832,
+      "step": 952
+    },
+    {
+      "epoch": 0.40253164556962023,
+      "grad_norm": 0.4565040171146393,
+      "learning_rate": 0.00016748681898066784,
+      "loss": 0.823063313961029,
+      "step": 954
+    },
+    {
+      "epoch": 0.40337552742616034,
+      "grad_norm": 0.4290153682231903,
+      "learning_rate": 0.00016783831282952548,
+      "loss": 0.8785063624382019,
+      "step": 956
+    },
+    {
+      "epoch": 0.40421940928270045,
+      "grad_norm": 0.5419702529907227,
+      "learning_rate": 0.00016818980667838312,
+      "loss": 0.8763971924781799,
+      "step": 958
+    },
+    {
+      "epoch": 0.4050632911392405,
+      "grad_norm": 0.5177501440048218,
+      "learning_rate": 0.00016854130052724079,
+      "loss": 0.9470553398132324,
+      "step": 960
+    },
+    {
+      "epoch": 0.4059071729957806,
+      "grad_norm": 0.539725661277771,
+      "learning_rate": 0.00016889279437609842,
+      "loss": 0.9235025644302368,
+      "step": 962
+    },
+    {
+      "epoch": 0.40675105485232066,
+      "grad_norm": 0.5324983596801758,
+      "learning_rate": 0.0001692442882249561,
+      "loss": 1.0248996019363403,
+      "step": 964
+    },
+    {
+      "epoch": 0.40759493670886077,
+      "grad_norm": 0.4936407506465912,
+      "learning_rate": 0.00016959578207381373,
+      "loss": 0.9076873660087585,
+      "step": 966
+    },
+    {
+      "epoch": 0.4084388185654008,
+      "grad_norm": 0.4960501194000244,
+      "learning_rate": 0.00016994727592267137,
+      "loss": 0.9162673950195312,
+      "step": 968
+    },
+    {
+      "epoch": 0.4092827004219409,
+      "grad_norm": 0.45093682408332825,
+      "learning_rate": 0.000170298769771529,
+      "loss": 0.904100775718689,
+      "step": 970
+    },
+    {
+      "epoch": 0.41012658227848103,
+      "grad_norm": 0.4560275077819824,
+      "learning_rate": 0.00017065026362038664,
+      "loss": 0.8633337020874023,
+      "step": 972
+    },
+    {
+      "epoch": 0.4109704641350211,
+      "grad_norm": 0.44885000586509705,
+      "learning_rate": 0.00017100175746924428,
+      "loss": 0.8454209566116333,
+      "step": 974
+    },
+    {
+      "epoch": 0.4118143459915612,
+      "grad_norm": 0.47251659631729126,
+      "learning_rate": 0.00017135325131810195,
+      "loss": 0.824730396270752,
+      "step": 976
+    },
+    {
+      "epoch": 0.41265822784810124,
+      "grad_norm": 0.6597666144371033,
+      "learning_rate": 0.00017170474516695959,
+      "loss": 0.9496501684188843,
+      "step": 978
+    },
+    {
+      "epoch": 0.41350210970464135,
+      "grad_norm": 0.39806297421455383,
+      "learning_rate": 0.00017205623901581722,
+      "loss": 0.9419087171554565,
+      "step": 980
+    },
+    {
+      "epoch": 0.41434599156118146,
+      "grad_norm": 0.48231109976768494,
+      "learning_rate": 0.0001724077328646749,
+      "loss": 0.9182976484298706,
+      "step": 982
+    },
+    {
+      "epoch": 0.4151898734177215,
+      "grad_norm": 0.5438776612281799,
+      "learning_rate": 0.00017275922671353253,
+      "loss": 0.9386967420578003,
+      "step": 984
+    },
+    {
+      "epoch": 0.4160337552742616,
+      "grad_norm": 0.4959667921066284,
+      "learning_rate": 0.00017311072056239017,
+      "loss": 0.897849440574646,
+      "step": 986
+    },
+    {
+      "epoch": 0.41687763713080167,
+      "grad_norm": 0.43533357977867126,
+      "learning_rate": 0.0001734622144112478,
+      "loss": 0.8776953816413879,
+      "step": 988
+    },
+    {
+      "epoch": 0.4177215189873418,
+      "grad_norm": 0.47513946890830994,
+      "learning_rate": 0.00017381370826010547,
+      "loss": 0.9162989854812622,
+      "step": 990
+    },
+    {
+      "epoch": 0.41856540084388183,
+      "grad_norm": 0.4907188415527344,
+      "learning_rate": 0.0001741652021089631,
+      "loss": 0.9482660889625549,
+      "step": 992
+    },
+    {
+      "epoch": 0.41940928270042194,
+      "grad_norm": 0.44499966502189636,
+      "learning_rate": 0.00017451669595782075,
+      "loss": 0.8812930583953857,
+      "step": 994
+    },
+    {
+      "epoch": 0.42025316455696204,
+      "grad_norm": 0.4535730481147766,
+      "learning_rate": 0.00017486818980667838,
+      "loss": 0.9439874887466431,
+      "step": 996
+    },
+    {
+      "epoch": 0.4210970464135021,
+      "grad_norm": 0.5240745544433594,
+      "learning_rate": 0.00017521968365553602,
+      "loss": 0.8818395137786865,
+      "step": 998
+    },
+    {
+      "epoch": 0.4219409282700422,
+      "grad_norm": 0.5301211476325989,
+      "learning_rate": 0.0001755711775043937,
+      "loss": 0.886186957359314,
+      "step": 1000
+    },
+    {
+      "epoch": 0.4219409282700422,
+      "eval_loss": 0.9487298727035522,
+      "eval_runtime": 689.4288,
+      "eval_samples_per_second": 3.056,
+      "eval_steps_per_second": 3.056,
+      "step": 1000
+    },
+    {
+      "epoch": 0.42278481012658226,
+      "grad_norm": 0.47876957058906555,
+      "learning_rate": 0.00017592267135325133,
+      "loss": 0.8814021348953247,
+      "step": 1002
+    },
+    {
+      "epoch": 0.42362869198312236,
+      "grad_norm": 0.4929780960083008,
+      "learning_rate": 0.00017627416520210896,
+      "loss": 0.8295068740844727,
+      "step": 1004
+    },
+    {
+      "epoch": 0.42447257383966247,
+      "grad_norm": 0.41888436675071716,
+      "learning_rate": 0.00017662565905096663,
+      "loss": 0.8364827036857605,
+      "step": 1006
+    },
+    {
+      "epoch": 0.4253164556962025,
+      "grad_norm": 0.5175151824951172,
+      "learning_rate": 0.00017697715289982427,
+      "loss": 1.0167189836502075,
+      "step": 1008
+    },
+    {
+      "epoch": 0.42616033755274263,
+      "grad_norm": 0.4815356135368347,
+      "learning_rate": 0.0001773286467486819,
+      "loss": 0.8460752964019775,
+      "step": 1010
+    },
+    {
+      "epoch": 0.4270042194092827,
+      "grad_norm": 0.5210875272750854,
+      "learning_rate": 0.00017768014059753954,
+      "loss": 0.8595574498176575,
+      "step": 1012
+    },
+    {
+      "epoch": 0.4278481012658228,
+      "grad_norm": 0.4400486946105957,
+      "learning_rate": 0.00017803163444639718,
+      "loss": 0.8764723539352417,
+      "step": 1014
+    },
+    {
+      "epoch": 0.4286919831223629,
+      "grad_norm": 0.5282127857208252,
+      "learning_rate": 0.00017838312829525482,
+      "loss": 0.9706798791885376,
+      "step": 1016
+    },
+    {
+      "epoch": 0.42953586497890295,
+      "grad_norm": 0.47523441910743713,
+      "learning_rate": 0.00017873462214411249,
+      "loss": 0.8912002444267273,
+      "step": 1018
+    },
+    {
+      "epoch": 0.43037974683544306,
+      "grad_norm": 0.49640706181526184,
+      "learning_rate": 0.00017908611599297012,
+      "loss": 0.8835636973381042,
+      "step": 1020
+    },
+    {
+      "epoch": 0.4312236286919831,
+      "grad_norm": 0.5253039002418518,
+      "learning_rate": 0.0001794376098418278,
+      "loss": 0.8711735606193542,
+      "step": 1022
+    },
+    {
+      "epoch": 0.4320675105485232,
+      "grad_norm": 0.49285009503364563,
+      "learning_rate": 0.00017978910369068543,
+      "loss": 0.9064869284629822,
+      "step": 1024
+    },
+    {
+      "epoch": 0.43291139240506327,
+      "grad_norm": 0.48758041858673096,
+      "learning_rate": 0.00018014059753954307,
+      "loss": 0.8993359208106995,
+      "step": 1026
+    },
+    {
+      "epoch": 0.4337552742616034,
+      "grad_norm": 0.4815461039543152,
+      "learning_rate": 0.0001804920913884007,
+      "loss": 0.8747937679290771,
+      "step": 1028
+    },
+    {
+      "epoch": 0.4345991561181435,
+      "grad_norm": 0.4930349886417389,
+      "learning_rate": 0.00018084358523725834,
+      "loss": 0.8879084587097168,
+      "step": 1030
+    },
+    {
+      "epoch": 0.43544303797468353,
+      "grad_norm": 0.49303027987480164,
+      "learning_rate": 0.00018119507908611598,
+      "loss": 1.0023083686828613,
+      "step": 1032
+    },
+    {
+      "epoch": 0.43628691983122364,
+      "grad_norm": 0.5312249064445496,
+      "learning_rate": 0.00018154657293497365,
+      "loss": 0.938680112361908,
+      "step": 1034
+    },
+    {
+      "epoch": 0.4371308016877637,
+      "grad_norm": 0.5174582004547119,
+      "learning_rate": 0.0001818980667838313,
+      "loss": 0.854195773601532,
+      "step": 1036
+    },
+    {
+      "epoch": 0.4379746835443038,
+      "grad_norm": 0.5452545881271362,
+      "learning_rate": 0.00018224956063268895,
+      "loss": 0.9059375524520874,
+      "step": 1038
+    },
+    {
+      "epoch": 0.4388185654008439,
+      "grad_norm": 0.5480839014053345,
+      "learning_rate": 0.0001826010544815466,
+      "loss": 0.9708920121192932,
+      "step": 1040
+    },
+    {
+      "epoch": 0.43966244725738396,
+      "grad_norm": 0.44693151116371155,
+      "learning_rate": 0.00018295254833040423,
+      "loss": 0.8561046123504639,
+      "step": 1042
+    },
+    {
+      "epoch": 0.44050632911392407,
+      "grad_norm": 0.4024234116077423,
+      "learning_rate": 0.00018330404217926187,
+      "loss": 0.8811968564987183,
+      "step": 1044
+    },
+    {
+      "epoch": 0.4413502109704641,
+      "grad_norm": 0.5586408376693726,
+      "learning_rate": 0.0001836555360281195,
+      "loss": 0.978068470954895,
+      "step": 1046
+    },
+    {
+      "epoch": 0.4421940928270042,
+      "grad_norm": 0.4918624758720398,
+      "learning_rate": 0.00018400702987697714,
+      "loss": 0.8640981316566467,
+      "step": 1048
+    },
+    {
+      "epoch": 0.4430379746835443,
+      "grad_norm": 0.4335230588912964,
+      "learning_rate": 0.0001843585237258348,
+      "loss": 0.9077964425086975,
+      "step": 1050
+    },
+    {
+      "epoch": 0.4438818565400844,
+      "grad_norm": 0.5275123715400696,
+      "learning_rate": 0.00018471001757469247,
+      "loss": 0.960682213306427,
+      "step": 1052
+    },
+    {
+      "epoch": 0.4447257383966245,
+      "grad_norm": 0.5144415497779846,
+      "learning_rate": 0.0001850615114235501,
+      "loss": 0.8549577593803406,
+      "step": 1054
+    },
+    {
+      "epoch": 0.44556962025316454,
+      "grad_norm": 0.5166662931442261,
+      "learning_rate": 0.00018541300527240775,
+      "loss": 0.9320827126502991,
+      "step": 1056
+    },
+    {
+      "epoch": 0.44641350210970465,
+      "grad_norm": 0.5011980533599854,
+      "learning_rate": 0.0001857644991212654,
+      "loss": 0.9632431864738464,
+      "step": 1058
+    },
+    {
+      "epoch": 0.4472573839662447,
+      "grad_norm": 0.49312469363212585,
+      "learning_rate": 0.00018611599297012303,
+      "loss": 0.9442946910858154,
+      "step": 1060
+    },
+    {
+      "epoch": 0.4481012658227848,
+      "grad_norm": 0.49958568811416626,
+      "learning_rate": 0.00018646748681898066,
+      "loss": 0.952802300453186,
+      "step": 1062
+    },
+    {
+      "epoch": 0.4489451476793249,
+      "grad_norm": 0.49827462434768677,
+      "learning_rate": 0.00018681898066783833,
+      "loss": 0.9630650877952576,
+      "step": 1064
+    },
+    {
+      "epoch": 0.44978902953586497,
+      "grad_norm": 0.523980438709259,
+      "learning_rate": 0.00018717047451669597,
+      "loss": 0.9081395268440247,
+      "step": 1066
+    },
+    {
+      "epoch": 0.4506329113924051,
+      "grad_norm": 0.5108568668365479,
+      "learning_rate": 0.0001875219683655536,
+      "loss": 0.9398958683013916,
+      "step": 1068
+    },
+    {
+      "epoch": 0.45147679324894513,
+      "grad_norm": 0.4453965127468109,
+      "learning_rate": 0.00018787346221441127,
+      "loss": 0.93592768907547,
+      "step": 1070
+    },
+    {
+      "epoch": 0.45232067510548524,
+      "grad_norm": 0.4675683081150055,
+      "learning_rate": 0.0001882249560632689,
+      "loss": 0.8879633545875549,
+      "step": 1072
+    },
+    {
+      "epoch": 0.4531645569620253,
+      "grad_norm": 0.47398847341537476,
+      "learning_rate": 0.00018857644991212655,
+      "loss": 0.9966004490852356,
+      "step": 1074
+    },
+    {
+      "epoch": 0.4540084388185654,
+      "grad_norm": 0.46053192019462585,
+      "learning_rate": 0.0001889279437609842,
+      "loss": 0.9184179902076721,
+      "step": 1076
+    },
+    {
+      "epoch": 0.4548523206751055,
+      "grad_norm": 0.5601398348808289,
+      "learning_rate": 0.00018927943760984182,
+      "loss": 0.9434974193572998,
+      "step": 1078
+    },
+    {
+      "epoch": 0.45569620253164556,
+      "grad_norm": 0.48422637581825256,
+      "learning_rate": 0.0001896309314586995,
+      "loss": 0.9522465467453003,
+      "step": 1080
+    },
+    {
+      "epoch": 0.45654008438818566,
+      "grad_norm": 0.46280911564826965,
+      "learning_rate": 0.00018998242530755713,
+      "loss": 0.9315434694290161,
+      "step": 1082
+    },
+    {
+      "epoch": 0.4573839662447257,
+      "grad_norm": 0.4658683240413666,
+      "learning_rate": 0.00019033391915641477,
+      "loss": 0.8165783286094666,
+      "step": 1084
+    },
+    {
+      "epoch": 0.4582278481012658,
+      "grad_norm": 0.4192182123661041,
+      "learning_rate": 0.0001906854130052724,
+      "loss": 0.8024274110794067,
+      "step": 1086
+    },
+    {
+      "epoch": 0.45907172995780593,
+      "grad_norm": 0.4178735613822937,
+      "learning_rate": 0.00019103690685413007,
+      "loss": 0.8620653748512268,
+      "step": 1088
+    },
+    {
+      "epoch": 0.459915611814346,
+      "grad_norm": 0.4488574266433716,
+      "learning_rate": 0.0001913884007029877,
+      "loss": 0.9121530055999756,
+      "step": 1090
+    },
+    {
+      "epoch": 0.4607594936708861,
+      "grad_norm": 0.5164965987205505,
+      "learning_rate": 0.00019173989455184535,
+      "loss": 0.9496700763702393,
+      "step": 1092
+    },
+    {
+      "epoch": 0.46160337552742614,
+      "grad_norm": 0.41563132405281067,
+      "learning_rate": 0.00019209138840070299,
+      "loss": 0.8757708668708801,
+      "step": 1094
+    },
+    {
+      "epoch": 0.46244725738396625,
+      "grad_norm": 2.3222429752349854,
+      "learning_rate": 0.00019244288224956065,
+      "loss": 0.8406533002853394,
+      "step": 1096
+    },
+    {
+      "epoch": 0.46329113924050636,
+      "grad_norm": 0.4813845157623291,
+      "learning_rate": 0.0001927943760984183,
+      "loss": 0.9459465742111206,
+      "step": 1098
+    },
+    {
+      "epoch": 0.4641350210970464,
+      "grad_norm": 0.6233882308006287,
+      "learning_rate": 0.00019314586994727593,
+      "loss": 0.9376904368400574,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4641350210970464,
+      "eval_loss": 0.9357889294624329,
+      "eval_runtime": 676.9573,
+      "eval_samples_per_second": 3.112,
+      "eval_steps_per_second": 3.112,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4649789029535865,
+      "grad_norm": 0.5125579237937927,
+      "learning_rate": 0.00019349736379613357,
+      "loss": 0.8998825550079346,
+      "step": 1102
+    },
+    {
+      "epoch": 0.46582278481012657,
+      "grad_norm": 0.4534320831298828,
+      "learning_rate": 0.00019384885764499123,
+      "loss": 0.9213768839836121,
+      "step": 1104
+    },
+    {
+      "epoch": 0.4666666666666667,
+      "grad_norm": 0.4715143144130707,
+      "learning_rate": 0.00019420035149384887,
+      "loss": 0.9739661812782288,
+      "step": 1106
+    },
+    {
+      "epoch": 0.4675105485232067,
+      "grad_norm": 0.45529672503471375,
+      "learning_rate": 0.0001945518453427065,
+      "loss": 0.9113216400146484,
+      "step": 1108
+    },
+    {
+      "epoch": 0.46835443037974683,
+      "grad_norm": 0.48199015855789185,
+      "learning_rate": 0.00019490333919156417,
+      "loss": 0.92528235912323,
+      "step": 1110
+    },
+    {
+      "epoch": 0.46919831223628694,
+      "grad_norm": 0.4425188899040222,
+      "learning_rate": 0.0001952548330404218,
+      "loss": 0.8612716197967529,
+      "step": 1112
+    },
+    {
+      "epoch": 0.470042194092827,
+      "grad_norm": 0.4311593770980835,
+      "learning_rate": 0.00019560632688927945,
+      "loss": 0.8901699185371399,
+      "step": 1114
+    },
+    {
+      "epoch": 0.4708860759493671,
+      "grad_norm": 0.497806578874588,
+      "learning_rate": 0.0001959578207381371,
+      "loss": 0.9271994829177856,
+      "step": 1116
+    },
+    {
+      "epoch": 0.47172995780590715,
+      "grad_norm": 0.47149473428726196,
+      "learning_rate": 0.00019630931458699473,
+      "loss": 0.8740925788879395,
+      "step": 1118
+    },
+    {
+      "epoch": 0.47257383966244726,
+      "grad_norm": 0.5029966235160828,
+      "learning_rate": 0.00019666080843585236,
+      "loss": 0.9190115928649902,
+      "step": 1120
+    },
+    {
+      "epoch": 0.47341772151898737,
+      "grad_norm": 0.47972601652145386,
+      "learning_rate": 0.00019701230228471003,
+      "loss": 1.0043057203292847,
+      "step": 1122
+    },
+    {
+      "epoch": 0.4742616033755274,
+      "grad_norm": 0.45096471905708313,
+      "learning_rate": 0.00019736379613356767,
+      "loss": 0.9472925066947937,
+      "step": 1124
+    },
+    {
+      "epoch": 0.4751054852320675,
+      "grad_norm": 0.459852933883667,
+      "learning_rate": 0.00019771528998242533,
+      "loss": 0.8043124079704285,
+      "step": 1126
+    },
+    {
+      "epoch": 0.4759493670886076,
+      "grad_norm": 0.5313422679901123,
+      "learning_rate": 0.00019806678383128297,
+      "loss": 0.9662142395973206,
+      "step": 1128
+    },
+    {
+      "epoch": 0.4767932489451477,
+      "grad_norm": 0.43474531173706055,
+      "learning_rate": 0.0001984182776801406,
+      "loss": 0.8691151738166809,
+      "step": 1130
+    },
+    {
+      "epoch": 0.47763713080168774,
+      "grad_norm": 0.5232312083244324,
+      "learning_rate": 0.00019876977152899825,
+      "loss": 0.9024254083633423,
+      "step": 1132
+    },
+    {
+      "epoch": 0.47848101265822784,
+      "grad_norm": 0.47671905159950256,
+      "learning_rate": 0.0001991212653778559,
+      "loss": 0.8996873497962952,
+      "step": 1134
+    },
+    {
+      "epoch": 0.47932489451476795,
+      "grad_norm": 0.42289480566978455,
+      "learning_rate": 0.00019947275922671353,
+      "loss": 0.8336917757987976,
+      "step": 1136
+    },
+    {
+      "epoch": 0.480168776371308,
+      "grad_norm": 0.4700844883918762,
+      "learning_rate": 0.0001998242530755712,
+      "loss": 0.8943206071853638,
+      "step": 1138
+    },
+    {
+      "epoch": 0.4810126582278481,
+      "grad_norm": 0.5341399312019348,
+      "learning_rate": 0.00019999999711649004,
+      "loss": 0.9340365529060364,
+      "step": 1140
+    },
+    {
+      "epoch": 0.48185654008438816,
+      "grad_norm": 0.46169522404670715,
+      "learning_rate": 0.00019999997404841123,
+      "loss": 0.8241778016090393,
+      "step": 1142
+    },
+    {
+      "epoch": 0.48270042194092827,
+      "grad_norm": 0.5475223064422607,
+      "learning_rate": 0.00019999992791225896,
+      "loss": 1.0096158981323242,
+      "step": 1144
+    },
+    {
+      "epoch": 0.4835443037974684,
+      "grad_norm": 0.524641215801239,
+      "learning_rate": 0.00019999985870804385,
+      "loss": 0.9650378227233887,
+      "step": 1146
+    },
+    {
+      "epoch": 0.48438818565400843,
+      "grad_norm": 0.4326174855232239,
+      "learning_rate": 0.00019999976643578186,
+      "loss": 0.9003009796142578,
+      "step": 1148
+    },
+    {
+      "epoch": 0.48523206751054854,
+      "grad_norm": 0.49034059047698975,
+      "learning_rate": 0.0001999996510954943,
+      "loss": 0.9793432950973511,
+      "step": 1150
+    },
+    {
+      "epoch": 0.4860759493670886,
+      "grad_norm": 0.4760092496871948,
+      "learning_rate": 0.00019999951268720776,
+      "loss": 0.8793007731437683,
+      "step": 1152
+    },
+    {
+      "epoch": 0.4869198312236287,
+      "grad_norm": 0.44606879353523254,
+      "learning_rate": 0.00019999935121095417,
+      "loss": 0.888630747795105,
+      "step": 1154
+    },
+    {
+      "epoch": 0.4877637130801688,
+      "grad_norm": 0.5030332207679749,
+      "learning_rate": 0.0001999991666667708,
+      "loss": 0.9660000801086426,
+      "step": 1156
+    },
+    {
+      "epoch": 0.48860759493670886,
+      "grad_norm": 0.4295555055141449,
+      "learning_rate": 0.00019999895905470014,
+      "loss": 0.824654757976532,
+      "step": 1158
+    },
+    {
+      "epoch": 0.48945147679324896,
+      "grad_norm": 0.392167866230011,
+      "learning_rate": 0.0001999987283747902,
+      "loss": 0.8159562349319458,
+      "step": 1160
+    },
+    {
+      "epoch": 0.490295358649789,
+      "grad_norm": 0.484611839056015,
+      "learning_rate": 0.00019999847462709412,
+      "loss": 0.9630686044692993,
+      "step": 1162
+    },
+    {
+      "epoch": 0.4911392405063291,
+      "grad_norm": 0.45891445875167847,
+      "learning_rate": 0.00019999819781167042,
+      "loss": 0.8396129608154297,
+      "step": 1164
+    },
+    {
+      "epoch": 0.4919831223628692,
+      "grad_norm": 0.5553452968597412,
+      "learning_rate": 0.00019999789792858304,
+      "loss": 0.8194513916969299,
+      "step": 1166
+    },
+    {
+      "epoch": 0.4928270042194093,
+      "grad_norm": 0.38998672366142273,
+      "learning_rate": 0.00019999757497790106,
+      "loss": 0.9268721342086792,
+      "step": 1168
+    },
+    {
+      "epoch": 0.4936708860759494,
+      "grad_norm": 0.45445355772972107,
+      "learning_rate": 0.00019999722895969904,
+      "loss": 0.9204684495925903,
+      "step": 1170
+    },
+    {
+      "epoch": 0.49451476793248944,
+      "grad_norm": 0.484225332736969,
+      "learning_rate": 0.00019999685987405678,
+      "loss": 0.9270301461219788,
+      "step": 1172
+    },
+    {
+      "epoch": 0.49535864978902955,
+      "grad_norm": 0.45215415954589844,
+      "learning_rate": 0.00019999646772105942,
+      "loss": 0.8782645463943481,
+      "step": 1174
+    },
+    {
+      "epoch": 0.4962025316455696,
+      "grad_norm": 0.41578832268714905,
+      "learning_rate": 0.00019999605250079744,
+      "loss": 0.873112678527832,
+      "step": 1176
+    },
+    {
+      "epoch": 0.4970464135021097,
+      "grad_norm": 0.4632788300514221,
+      "learning_rate": 0.0001999956142133666,
+      "loss": 0.9593189358711243,
+      "step": 1178
+    },
+    {
+      "epoch": 0.4978902953586498,
+      "grad_norm": 0.5007622838020325,
+      "learning_rate": 0.000199995152858868,
+      "loss": 0.9215621948242188,
+      "step": 1180
+    },
+    {
+      "epoch": 0.49873417721518987,
+      "grad_norm": 0.5201935768127441,
+      "learning_rate": 0.0001999946684374081,
+      "loss": 0.8964219689369202,
+      "step": 1182
+    },
+    {
+      "epoch": 0.49957805907173,
+      "grad_norm": 0.5053967237472534,
+      "learning_rate": 0.0001999941609490986,
+      "loss": 0.8528663516044617,
+      "step": 1184
+    },
+    {
+      "epoch": 0.5004219409282701,
+      "grad_norm": 0.44397974014282227,
+      "learning_rate": 0.0001999936303940566,
+      "loss": 0.8214734792709351,
+      "step": 1186
+    },
+    {
+      "epoch": 0.5012658227848101,
+      "grad_norm": 0.4563101530075073,
+      "learning_rate": 0.0001999930767724045,
+      "loss": 0.8882166743278503,
+      "step": 1188
+    },
+    {
+      "epoch": 0.5021097046413502,
+      "grad_norm": 0.4666728675365448,
+      "learning_rate": 0.00019999250008426997,
+      "loss": 0.8882588148117065,
+      "step": 1190
+    },
+    {
+      "epoch": 0.5029535864978903,
+      "grad_norm": 0.4423629939556122,
+      "learning_rate": 0.00019999190032978607,
+      "loss": 0.8353691697120667,
+      "step": 1192
+    },
+    {
+      "epoch": 0.5037974683544304,
+      "grad_norm": 0.4514595866203308,
+      "learning_rate": 0.00019999127750909118,
+      "loss": 0.8721219301223755,
+      "step": 1194
+    },
+    {
+      "epoch": 0.5046413502109705,
+      "grad_norm": 0.48469996452331543,
+      "learning_rate": 0.00019999063162232886,
+      "loss": 0.902795135974884,
+      "step": 1196
+    },
+    {
+      "epoch": 0.5054852320675105,
+      "grad_norm": 0.4931983947753906,
+      "learning_rate": 0.00019998996266964823,
+      "loss": 0.8843175768852234,
+      "step": 1198
+    },
+    {
+      "epoch": 0.5063291139240507,
+      "grad_norm": 0.507408082485199,
+      "learning_rate": 0.00019998927065120357,
+      "loss": 0.9179208278656006,
+      "step": 1200
+    },
+    {
+      "epoch": 0.5063291139240507,
+      "eval_loss": 0.9224098324775696,
+      "eval_runtime": 669.7542,
+      "eval_samples_per_second": 3.146,
+      "eval_steps_per_second": 3.146,
+      "step": 1200
+    },
+    {
+      "epoch": 0.5071729957805907,
+      "grad_norm": 0.4024188220500946,
+      "learning_rate": 0.00019998855556715447,
+      "loss": 0.9078981876373291,
+      "step": 1202
+    },
+    {
+      "epoch": 0.5080168776371308,
+      "grad_norm": 0.6558667421340942,
+      "learning_rate": 0.00019998781741766594,
+      "loss": 0.8089252710342407,
+      "step": 1204
+    },
+    {
+      "epoch": 0.5088607594936709,
+      "grad_norm": 0.45128145813941956,
+      "learning_rate": 0.00019998705620290823,
+      "loss": 0.9316248297691345,
+      "step": 1206
+    },
+    {
+      "epoch": 0.509704641350211,
+      "grad_norm": 0.4733511507511139,
+      "learning_rate": 0.00019998627192305694,
+      "loss": 0.9047867655754089,
+      "step": 1208
+    },
+    {
+      "epoch": 0.510548523206751,
+      "grad_norm": 0.4026021659374237,
+      "learning_rate": 0.00019998546457829298,
+      "loss": 0.8507166504859924,
+      "step": 1210
+    },
+    {
+      "epoch": 0.5113924050632911,
+      "grad_norm": 0.43416184186935425,
+      "learning_rate": 0.00019998463416880262,
+      "loss": 0.8490090370178223,
+      "step": 1212
+    },
+    {
+      "epoch": 0.5122362869198313,
+      "grad_norm": 0.3678364157676697,
+      "learning_rate": 0.0001999837806947774,
+      "loss": 0.7730492949485779,
+      "step": 1214
+    },
+    {
+      "epoch": 0.5130801687763713,
+      "grad_norm": 0.49042677879333496,
+      "learning_rate": 0.00019998290415641415,
+      "loss": 0.9625269174575806,
+      "step": 1216
+    },
+    {
+      "epoch": 0.5139240506329114,
+      "grad_norm": 0.4576701819896698,
+      "learning_rate": 0.00019998200455391516,
+      "loss": 0.8961732983589172,
+      "step": 1218
+    },
+    {
+      "epoch": 0.5147679324894515,
+      "grad_norm": 0.4465518593788147,
+      "learning_rate": 0.00019998108188748793,
+      "loss": 0.8508996367454529,
+      "step": 1220
+    },
+    {
+      "epoch": 0.5156118143459916,
+      "grad_norm": 0.4267960488796234,
+      "learning_rate": 0.00019998013615734524,
+      "loss": 0.8664930462837219,
+      "step": 1222
+    },
+    {
+      "epoch": 0.5164556962025316,
+      "grad_norm": 0.39936602115631104,
+      "learning_rate": 0.0001999791673637053,
+      "loss": 0.8841317892074585,
+      "step": 1224
+    },
+    {
+      "epoch": 0.5172995780590718,
+      "grad_norm": 0.4442414343357086,
+      "learning_rate": 0.0001999781755067916,
+      "loss": 0.8601276874542236,
+      "step": 1226
+    },
+    {
+      "epoch": 0.5181434599156118,
+      "grad_norm": 0.4397028088569641,
+      "learning_rate": 0.00019997716058683292,
+      "loss": 0.8377046585083008,
+      "step": 1228
+    },
+    {
+      "epoch": 0.5189873417721519,
+      "grad_norm": 0.5327648520469666,
+      "learning_rate": 0.0001999761226040634,
+      "loss": 0.9853615164756775,
+      "step": 1230
+    },
+    {
+      "epoch": 0.5198312236286919,
+      "grad_norm": 0.40931421518325806,
+      "learning_rate": 0.00019997506155872244,
+      "loss": 0.9235715866088867,
+      "step": 1232
+    },
+    {
+      "epoch": 0.5206751054852321,
+      "grad_norm": 0.4599161148071289,
+      "learning_rate": 0.00019997397745105487,
+      "loss": 0.868396520614624,
+      "step": 1234
+    },
+    {
+      "epoch": 0.5215189873417722,
+      "grad_norm": 0.4813833236694336,
+      "learning_rate": 0.0001999728702813107,
+      "loss": 0.9946733117103577,
+      "step": 1236
+    },
+    {
+      "epoch": 0.5223628691983122,
+      "grad_norm": 0.43518269062042236,
+      "learning_rate": 0.00019997174004974543,
+      "loss": 0.8608635067939758,
+      "step": 1238
+    },
+    {
+      "epoch": 0.5232067510548524,
+      "grad_norm": 0.45646214485168457,
+      "learning_rate": 0.0001999705867566197,
+      "loss": 0.819354772567749,
+      "step": 1240
+    },
+    {
+      "epoch": 0.5240506329113924,
+      "grad_norm": 0.46834197640419006,
+      "learning_rate": 0.00019996941040219954,
+      "loss": 0.8934658765792847,
+      "step": 1242
+    },
+    {
+      "epoch": 0.5248945147679325,
+      "grad_norm": 0.46065032482147217,
+      "learning_rate": 0.0001999682109867564,
+      "loss": 0.8624778985977173,
+      "step": 1244
+    },
+    {
+      "epoch": 0.5257383966244725,
+      "grad_norm": 0.4506741166114807,
+      "learning_rate": 0.00019996698851056688,
+      "loss": 0.8540882468223572,
+      "step": 1246
+    },
+    {
+      "epoch": 0.5265822784810127,
+      "grad_norm": 0.4550519287586212,
+      "learning_rate": 0.00019996574297391302,
+      "loss": 0.9003006219863892,
+      "step": 1248
+    },
+    {
+      "epoch": 0.5274261603375527,
+      "grad_norm": 0.4097813367843628,
+      "learning_rate": 0.00019996447437708214,
+      "loss": 0.8490248918533325,
+      "step": 1250
+    },
+    {
+      "epoch": 0.5282700421940928,
+      "grad_norm": 0.4375711679458618,
+      "learning_rate": 0.00019996318272036688,
+      "loss": 0.9204569458961487,
+      "step": 1252
+    },
+    {
+      "epoch": 0.529113924050633,
+      "grad_norm": 0.4335198998451233,
+      "learning_rate": 0.00019996186800406518,
+      "loss": 0.8831789493560791,
+      "step": 1254
+    },
+    {
+      "epoch": 0.529957805907173,
+      "grad_norm": 0.4840933084487915,
+      "learning_rate": 0.00019996053022848035,
+      "loss": 0.8578064441680908,
+      "step": 1256
+    },
+    {
+      "epoch": 0.5308016877637131,
+      "grad_norm": 0.48911216855049133,
+      "learning_rate": 0.00019995916939392097,
+      "loss": 0.9235416650772095,
+      "step": 1258
+    },
+    {
+      "epoch": 0.5316455696202531,
+      "grad_norm": 0.41106730699539185,
+      "learning_rate": 0.00019995778550070098,
+      "loss": 0.7897764444351196,
+      "step": 1260
+    },
+    {
+      "epoch": 0.5324894514767933,
+      "grad_norm": 0.42233356833457947,
+      "learning_rate": 0.00019995637854913957,
+      "loss": 0.9392989873886108,
+      "step": 1262
+    },
+    {
+      "epoch": 0.5333333333333333,
+      "grad_norm": 0.39145180583000183,
+      "learning_rate": 0.0001999549485395613,
+      "loss": 0.9130091667175293,
+      "step": 1264
+    },
+    {
+      "epoch": 0.5341772151898734,
+      "grad_norm": 0.42780739068984985,
+      "learning_rate": 0.00019995349547229614,
+      "loss": 0.9690561890602112,
+      "step": 1266
+    },
+    {
+      "epoch": 0.5350210970464135,
+      "grad_norm": 0.4823327362537384,
+      "learning_rate": 0.0001999520193476792,
+      "loss": 0.9990131258964539,
+      "step": 1268
+    },
+    {
+      "epoch": 0.5358649789029536,
+      "grad_norm": 0.4669715166091919,
+      "learning_rate": 0.00019995052016605097,
+      "loss": 0.8343052864074707,
+      "step": 1270
+    },
+    {
+      "epoch": 0.5367088607594936,
+      "grad_norm": 0.4024743437767029,
+      "learning_rate": 0.0001999489979277574,
+      "loss": 0.9605970978736877,
+      "step": 1272
+    },
+    {
+      "epoch": 0.5375527426160338,
+      "grad_norm": 0.42063432931900024,
+      "learning_rate": 0.0001999474526331495,
+      "loss": 0.8164438009262085,
+      "step": 1274
+    },
+    {
+      "epoch": 0.5383966244725739,
+      "grad_norm": 0.47953495383262634,
+      "learning_rate": 0.00019994588428258383,
+      "loss": 0.879767656326294,
+      "step": 1276
+    },
+    {
+      "epoch": 0.5392405063291139,
+      "grad_norm": 0.39668557047843933,
+      "learning_rate": 0.00019994429287642214,
+      "loss": 0.7658726572990417,
+      "step": 1278
+    },
+    {
+      "epoch": 0.540084388185654,
+      "grad_norm": 0.5136121511459351,
+      "learning_rate": 0.00019994267841503155,
+      "loss": 0.9477384686470032,
+      "step": 1280
+    },
+    {
+      "epoch": 0.5409282700421941,
+      "grad_norm": 0.4039286673069,
+      "learning_rate": 0.00019994104089878452,
+      "loss": 0.7883896827697754,
+      "step": 1282
+    },
+    {
+      "epoch": 0.5417721518987342,
+      "grad_norm": 0.4425487816333771,
+      "learning_rate": 0.0001999393803280587,
+      "loss": 0.8526129126548767,
+      "step": 1284
+    },
+    {
+      "epoch": 0.5426160337552742,
+      "grad_norm": 0.4544559121131897,
+      "learning_rate": 0.00019993769670323725,
+      "loss": 0.8694683909416199,
+      "step": 1286
+    },
+    {
+      "epoch": 0.5434599156118144,
+      "grad_norm": 0.42872917652130127,
+      "learning_rate": 0.0001999359900247085,
+      "loss": 0.8211527466773987,
+      "step": 1288
+    },
+    {
+      "epoch": 0.5443037974683544,
+      "grad_norm": 0.41549673676490784,
+      "learning_rate": 0.00019993426029286617,
+      "loss": 0.7967932224273682,
+      "step": 1290
+    },
+    {
+      "epoch": 0.5451476793248945,
+      "grad_norm": 0.42888355255126953,
+      "learning_rate": 0.00019993250750810926,
+      "loss": 0.8291563987731934,
+      "step": 1292
+    },
+    {
+      "epoch": 0.5459915611814345,
+      "grad_norm": 0.44985485076904297,
+      "learning_rate": 0.0001999307316708421,
+      "loss": 0.8787803649902344,
+      "step": 1294
+    },
+    {
+      "epoch": 0.5468354430379747,
+      "grad_norm": 0.43326008319854736,
+      "learning_rate": 0.00019992893278147436,
+      "loss": 0.8375519514083862,
+      "step": 1296
+    },
+    {
+      "epoch": 0.5476793248945148,
+      "grad_norm": 0.406556099653244,
+      "learning_rate": 0.00019992711084042096,
+      "loss": 0.8538051247596741,
+      "step": 1298
+    },
+    {
+      "epoch": 0.5485232067510548,
+      "grad_norm": 0.46090877056121826,
+      "learning_rate": 0.00019992526584810226,
+      "loss": 0.9722132086753845,
+      "step": 1300
+    },
+    {
+      "epoch": 0.5485232067510548,
+      "eval_loss": 0.9068717360496521,
+      "eval_runtime": 680.7718,
+      "eval_samples_per_second": 3.095,
+      "eval_steps_per_second": 3.095,
+      "step": 1300
+    },
+    {
+      "epoch": 0.549367088607595,
+      "grad_norm": 0.44932207465171814,
+      "learning_rate": 0.00019992339780494382,
+      "loss": 0.8474152684211731,
+      "step": 1302
+    },
+    {
+      "epoch": 0.550210970464135,
+      "grad_norm": 0.5185276865959167,
+      "learning_rate": 0.0001999215067113766,
+      "loss": 0.8440775871276855,
+      "step": 1304
+    },
+    {
+      "epoch": 0.5510548523206751,
+      "grad_norm": 0.43811365962028503,
+      "learning_rate": 0.0001999195925678368,
+      "loss": 0.8848010897636414,
+      "step": 1306
+    },
+    {
+      "epoch": 0.5518987341772152,
+      "grad_norm": 0.4861660301685333,
+      "learning_rate": 0.00019991765537476596,
+      "loss": 0.8265100121498108,
+      "step": 1308
+    },
+    {
+      "epoch": 0.5527426160337553,
+      "grad_norm": 0.5068721771240234,
+      "learning_rate": 0.000199915695132611,
+      "loss": 0.9327963590621948,
+      "step": 1310
+    },
+    {
+      "epoch": 0.5535864978902953,
+      "grad_norm": 0.4902805685997009,
+      "learning_rate": 0.0001999137118418241,
+      "loss": 0.9315155744552612,
+      "step": 1312
+    },
+    {
+      "epoch": 0.5544303797468354,
+      "grad_norm": 0.4530964493751526,
+      "learning_rate": 0.00019991170550286272,
+      "loss": 0.9344852566719055,
+      "step": 1314
+    },
+    {
+      "epoch": 0.5552742616033756,
+      "grad_norm": 0.4106673300266266,
+      "learning_rate": 0.00019990967611618974,
+      "loss": 0.8326173424720764,
+      "step": 1316
+    },
+    {
+      "epoch": 0.5561181434599156,
+      "grad_norm": 0.4439711570739746,
+      "learning_rate": 0.0001999076236822733,
+      "loss": 0.8934909105300903,
+      "step": 1318
+    },
+    {
+      "epoch": 0.5569620253164557,
+      "grad_norm": 0.40245553851127625,
+      "learning_rate": 0.0001999055482015868,
+      "loss": 0.6691107153892517,
+      "step": 1320
+    },
+    {
+      "epoch": 0.5578059071729958,
+      "grad_norm": 0.43904435634613037,
+      "learning_rate": 0.0001999034496746091,
+      "loss": 0.8560307621955872,
+      "step": 1322
+    },
+    {
+      "epoch": 0.5586497890295359,
+      "grad_norm": 0.5131705403327942,
+      "learning_rate": 0.00019990132810182422,
+      "loss": 0.9024442434310913,
+      "step": 1324
+    },
+    {
+      "epoch": 0.5594936708860759,
+      "grad_norm": 0.4539531171321869,
+      "learning_rate": 0.00019989918348372156,
+      "loss": 0.8675428628921509,
+      "step": 1326
+    },
+    {
+      "epoch": 0.560337552742616,
+      "grad_norm": 0.44469985365867615,
+      "learning_rate": 0.00019989701582079591,
+      "loss": 0.8853429555892944,
+      "step": 1328
+    },
+    {
+      "epoch": 0.5611814345991561,
+      "grad_norm": 0.4485546946525574,
+      "learning_rate": 0.00019989482511354725,
+      "loss": 0.9675378799438477,
+      "step": 1330
+    },
+    {
+      "epoch": 0.5620253164556962,
+      "grad_norm": 0.43316468596458435,
+      "learning_rate": 0.00019989261136248097,
+      "loss": 0.8256655335426331,
+      "step": 1332
+    },
+    {
+      "epoch": 0.5628691983122363,
+      "grad_norm": 0.4315780699253082,
+      "learning_rate": 0.00019989037456810772,
+      "loss": 0.7727690935134888,
+      "step": 1334
+    },
+    {
+      "epoch": 0.5637130801687764,
+      "grad_norm": 0.45147502422332764,
+      "learning_rate": 0.00019988811473094348,
+      "loss": 0.9240618348121643,
+      "step": 1336
+    },
+    {
+      "epoch": 0.5645569620253165,
+      "grad_norm": 0.4606908857822418,
+      "learning_rate": 0.00019988583185150957,
+      "loss": 0.8005949258804321,
+      "step": 1338
+    },
+    {
+      "epoch": 0.5654008438818565,
+      "grad_norm": 0.48932451009750366,
+      "learning_rate": 0.00019988352593033255,
+      "loss": 0.9042718410491943,
+      "step": 1340
+    },
+    {
+      "epoch": 0.5662447257383966,
+      "grad_norm": 0.46022218465805054,
+      "learning_rate": 0.00019988119696794443,
+      "loss": 0.8724613785743713,
+      "step": 1342
+    },
+    {
+      "epoch": 0.5670886075949367,
+      "grad_norm": 0.4414621889591217,
+      "learning_rate": 0.00019987884496488242,
+      "loss": 0.8653066158294678,
+      "step": 1344
+    },
+    {
+      "epoch": 0.5679324894514768,
+      "grad_norm": 0.4490342140197754,
+      "learning_rate": 0.0001998764699216891,
+      "loss": 0.8843849301338196,
+      "step": 1346
+    },
+    {
+      "epoch": 0.5687763713080168,
+      "grad_norm": 0.47690069675445557,
+      "learning_rate": 0.00019987407183891232,
+      "loss": 0.8385393023490906,
+      "step": 1348
+    },
+    {
+      "epoch": 0.569620253164557,
+      "grad_norm": 0.41409361362457275,
+      "learning_rate": 0.00019987165071710527,
+      "loss": 0.77492755651474,
+      "step": 1350
+    },
+    {
+      "epoch": 0.570464135021097,
+      "grad_norm": 0.4151647984981537,
+      "learning_rate": 0.0001998692065568265,
+      "loss": 0.8368680477142334,
+      "step": 1352
+    },
+    {
+      "epoch": 0.5713080168776371,
+      "grad_norm": 0.4105067849159241,
+      "learning_rate": 0.0001998667393586398,
+      "loss": 0.833220362663269,
+      "step": 1354
+    },
+    {
+      "epoch": 0.5721518987341773,
+      "grad_norm": 0.44701820611953735,
+      "learning_rate": 0.0001998642491231143,
+      "loss": 0.8041552305221558,
+      "step": 1356
+    },
+    {
+      "epoch": 0.5729957805907173,
+      "grad_norm": 0.4621582329273224,
+      "learning_rate": 0.00019986173585082444,
+      "loss": 0.9052709937095642,
+      "step": 1358
+    },
+    {
+      "epoch": 0.5738396624472574,
+      "grad_norm": 0.4934713840484619,
+      "learning_rate": 0.00019985919954235005,
+      "loss": 0.8036054372787476,
+      "step": 1360
+    },
+    {
+      "epoch": 0.5746835443037974,
+      "grad_norm": 0.4557384252548218,
+      "learning_rate": 0.0001998566401982761,
+      "loss": 0.8340095281600952,
+      "step": 1362
+    },
+    {
+      "epoch": 0.5755274261603376,
+      "grad_norm": 0.5048403739929199,
+      "learning_rate": 0.00019985405781919305,
+      "loss": 0.8703633546829224,
+      "step": 1364
+    },
+    {
+      "epoch": 0.5763713080168776,
+      "grad_norm": 0.4707714319229126,
+      "learning_rate": 0.0001998514524056966,
+      "loss": 0.9115830659866333,
+      "step": 1366
+    },
+    {
+      "epoch": 0.5772151898734177,
+      "grad_norm": 0.4502186179161072,
+      "learning_rate": 0.00019984882395838777,
+      "loss": 0.8603149056434631,
+      "step": 1368
+    },
+    {
+      "epoch": 0.5780590717299579,
+      "grad_norm": 0.4229136109352112,
+      "learning_rate": 0.00019984617247787288,
+      "loss": 0.7482197284698486,
+      "step": 1370
+    },
+    {
+      "epoch": 0.5789029535864979,
+      "grad_norm": 0.49208030104637146,
+      "learning_rate": 0.0001998434979647636,
+      "loss": 0.9071239233016968,
+      "step": 1372
+    },
+    {
+      "epoch": 0.579746835443038,
+      "grad_norm": 0.47393161058425903,
+      "learning_rate": 0.00019984080041967687,
+      "loss": 0.9370644092559814,
+      "step": 1374
+    },
+    {
+      "epoch": 0.580590717299578,
+      "grad_norm": 0.4483324885368347,
+      "learning_rate": 0.00019983807984323492,
+      "loss": 0.8536180257797241,
+      "step": 1376
+    },
+    {
+      "epoch": 0.5814345991561182,
+      "grad_norm": 0.4163796007633209,
+      "learning_rate": 0.00019983533623606543,
+      "loss": 0.7360405325889587,
+      "step": 1378
+    },
+    {
+      "epoch": 0.5822784810126582,
+      "grad_norm": 0.44147446751594543,
+      "learning_rate": 0.00019983256959880122,
+      "loss": 0.8162824511528015,
+      "step": 1380
+    },
+    {
+      "epoch": 0.5831223628691983,
+      "grad_norm": 0.44443491101264954,
+      "learning_rate": 0.0001998297799320805,
+      "loss": 0.9169327616691589,
+      "step": 1382
+    },
+    {
+      "epoch": 0.5839662447257384,
+      "grad_norm": 0.4641949236392975,
+      "learning_rate": 0.00019982696723654686,
+      "loss": 0.8875693678855896,
+      "step": 1384
+    },
+    {
+      "epoch": 0.5848101265822785,
+      "grad_norm": 0.4327974319458008,
+      "learning_rate": 0.00019982413151284906,
+      "loss": 0.7747344970703125,
+      "step": 1386
+    },
+    {
+      "epoch": 0.5856540084388185,
+      "grad_norm": 0.44200772047042847,
+      "learning_rate": 0.0001998212727616413,
+      "loss": 0.8490481972694397,
+      "step": 1388
+    },
+    {
+      "epoch": 0.5864978902953587,
+      "grad_norm": 0.39084959030151367,
+      "learning_rate": 0.000199818390983583,
+      "loss": 0.8014808297157288,
+      "step": 1390
+    },
+    {
+      "epoch": 0.5873417721518988,
+      "grad_norm": 0.3691277801990509,
+      "learning_rate": 0.00019981548617933897,
+      "loss": 0.8650928139686584,
+      "step": 1392
+    },
+    {
+      "epoch": 0.5881856540084388,
+      "grad_norm": 0.480002760887146,
+      "learning_rate": 0.00019981255834957925,
+      "loss": 0.7713267803192139,
+      "step": 1394
+    },
+    {
+      "epoch": 0.5890295358649789,
+      "grad_norm": 0.4397226572036743,
+      "learning_rate": 0.00019980960749497927,
+      "loss": 0.9053656458854675,
+      "step": 1396
+    },
+    {
+      "epoch": 0.589873417721519,
+      "grad_norm": 0.4263727366924286,
+      "learning_rate": 0.00019980663361621973,
+      "loss": 0.8737669587135315,
+      "step": 1398
+    },
+    {
+      "epoch": 0.5907172995780591,
+      "grad_norm": 0.4468817710876465,
+      "learning_rate": 0.0001998036367139866,
+      "loss": 0.9396650791168213,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5907172995780591,
+      "eval_loss": 0.8971880674362183,
+      "eval_runtime": 692.8046,
+      "eval_samples_per_second": 3.041,
+      "eval_steps_per_second": 3.041,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5915611814345991,
+      "grad_norm": 0.4128144383430481,
+      "learning_rate": 0.00019980061678897127,
+      "loss": 0.926134467124939,
+      "step": 1402
+    },
+    {
+      "epoch": 0.5924050632911393,
+      "grad_norm": 0.45524629950523376,
+      "learning_rate": 0.00019979757384187035,
+      "loss": 0.826360285282135,
+      "step": 1404
+    },
+    {
+      "epoch": 0.5932489451476793,
+      "grad_norm": 0.42521336674690247,
+      "learning_rate": 0.0001997945078733858,
+      "loss": 0.8147702813148499,
+      "step": 1406
+    },
+    {
+      "epoch": 0.5940928270042194,
+      "grad_norm": 0.4172525107860565,
+      "learning_rate": 0.00019979141888422489,
+      "loss": 0.7711596488952637,
+      "step": 1408
+    },
+    {
+      "epoch": 0.5949367088607594,
+      "grad_norm": 0.46193623542785645,
+      "learning_rate": 0.00019978830687510013,
+      "loss": 0.7875980138778687,
+      "step": 1410
+    },
+    {
+      "epoch": 0.5957805907172996,
+      "grad_norm": 0.4455825686454773,
+      "learning_rate": 0.00019978517184672946,
+      "loss": 0.8590195178985596,
+      "step": 1412
+    },
+    {
+      "epoch": 0.5966244725738397,
+      "grad_norm": 0.4266505539417267,
+      "learning_rate": 0.00019978201379983608,
+      "loss": 0.9318227767944336,
+      "step": 1414
+    },
+    {
+      "epoch": 0.5974683544303797,
+      "grad_norm": 0.42627251148223877,
+      "learning_rate": 0.00019977883273514843,
+      "loss": 0.8644474744796753,
+      "step": 1416
+    },
+    {
+      "epoch": 0.5983122362869199,
+      "grad_norm": 0.393873393535614,
+      "learning_rate": 0.00019977562865340038,
+      "loss": 0.7760446071624756,
+      "step": 1418
+    },
+    {
+      "epoch": 0.5991561181434599,
+      "grad_norm": 0.44334208965301514,
+      "learning_rate": 0.000199772401555331,
+      "loss": 0.8071584701538086,
+      "step": 1420
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.39390063285827637,
+      "learning_rate": 0.0001997691514416848,
+      "loss": 0.9076889753341675,
+      "step": 1422
+    },
+    {
+      "epoch": 0.60084388185654,
+      "grad_norm": 0.46753421425819397,
+      "learning_rate": 0.00019976587831321144,
+      "loss": 0.8534318804740906,
+      "step": 1424
+    },
+    {
+      "epoch": 0.6016877637130802,
+      "grad_norm": 0.4840681254863739,
+      "learning_rate": 0.00019976258217066598,
+      "loss": 0.9049596190452576,
+      "step": 1426
+    },
+    {
+      "epoch": 0.6025316455696202,
+      "grad_norm": 0.4740557372570038,
+      "learning_rate": 0.00019975926301480878,
+      "loss": 0.8520918488502502,
+      "step": 1428
+    },
+    {
+      "epoch": 0.6033755274261603,
+      "grad_norm": 0.46498584747314453,
+      "learning_rate": 0.00019975592084640553,
+      "loss": 0.8221305012702942,
+      "step": 1430
+    },
+    {
+      "epoch": 0.6042194092827005,
+      "grad_norm": 0.40885692834854126,
+      "learning_rate": 0.0001997525556662272,
+      "loss": 0.8132198452949524,
+      "step": 1432
+    },
+    {
+      "epoch": 0.6050632911392405,
+      "grad_norm": 0.4375821053981781,
+      "learning_rate": 0.00019974916747505006,
+      "loss": 0.870534360408783,
+      "step": 1434
+    },
+    {
+      "epoch": 0.6059071729957806,
+      "grad_norm": 0.4780935049057007,
+      "learning_rate": 0.00019974575627365569,
+      "loss": 0.8825759291648865,
+      "step": 1436
+    },
+    {
+      "epoch": 0.6067510548523207,
+      "grad_norm": 0.4623970687389374,
+      "learning_rate": 0.00019974232206283098,
+      "loss": 0.8187641501426697,
+      "step": 1438
+    },
+    {
+      "epoch": 0.6075949367088608,
+      "grad_norm": 0.42172595858573914,
+      "learning_rate": 0.0001997388648433682,
+      "loss": 0.8569821119308472,
+      "step": 1440
+    },
+    {
+      "epoch": 0.6084388185654008,
+      "grad_norm": 0.49386659264564514,
+      "learning_rate": 0.0001997353846160648,
+      "loss": 0.9173614382743835,
+      "step": 1442
+    },
+    {
+      "epoch": 0.6092827004219409,
+      "grad_norm": 0.47122615575790405,
+      "learning_rate": 0.00019973188138172363,
+      "loss": 0.9444975256919861,
+      "step": 1444
+    },
+    {
+      "epoch": 0.610126582278481,
+      "grad_norm": 0.4831530451774597,
+      "learning_rate": 0.0001997283551411528,
+      "loss": 0.9537245631217957,
+      "step": 1446
+    },
+    {
+      "epoch": 0.6109704641350211,
+      "grad_norm": 0.44460317492485046,
+      "learning_rate": 0.0001997248058951658,
+      "loss": 0.8481367230415344,
+      "step": 1448
+    },
+    {
+      "epoch": 0.6118143459915611,
+      "grad_norm": 0.4315263628959656,
+      "learning_rate": 0.00019972123364458128,
+      "loss": 0.8572371006011963,
+      "step": 1450
+    },
+    {
+      "epoch": 0.6126582278481013,
+      "grad_norm": 0.38559582829475403,
+      "learning_rate": 0.00019971763839022336,
+      "loss": 0.762590765953064,
+      "step": 1452
+    },
+    {
+      "epoch": 0.6135021097046414,
+      "grad_norm": 0.3820290267467499,
+      "learning_rate": 0.00019971402013292138,
+      "loss": 0.7879500389099121,
+      "step": 1454
+    },
+    {
+      "epoch": 0.6143459915611814,
+      "grad_norm": 0.42509570717811584,
+      "learning_rate": 0.00019971037887351,
+      "loss": 0.8696863055229187,
+      "step": 1456
+    },
+    {
+      "epoch": 0.6151898734177215,
+      "grad_norm": 0.46750548481941223,
+      "learning_rate": 0.00019970671461282916,
+      "loss": 0.8425862789154053,
+      "step": 1458
+    },
+    {
+      "epoch": 0.6160337552742616,
+      "grad_norm": 0.3937121033668518,
+      "learning_rate": 0.0001997030273517242,
+      "loss": 0.8028931021690369,
+      "step": 1460
+    },
+    {
+      "epoch": 0.6168776371308017,
+      "grad_norm": 0.4361220896244049,
+      "learning_rate": 0.00019969931709104565,
+      "loss": 0.8662509918212891,
+      "step": 1462
+    },
+    {
+      "epoch": 0.6177215189873417,
+      "grad_norm": 0.4737963378429413,
+      "learning_rate": 0.00019969558383164943,
+      "loss": 0.8962596654891968,
+      "step": 1464
+    },
+    {
+      "epoch": 0.6185654008438819,
+      "grad_norm": 0.3952867388725281,
+      "learning_rate": 0.00019969182757439668,
+      "loss": 0.8630867600440979,
+      "step": 1466
+    },
+    {
+      "epoch": 0.619409282700422,
+      "grad_norm": 0.41688743233680725,
+      "learning_rate": 0.00019968804832015393,
+      "loss": 0.8759240508079529,
+      "step": 1468
+    },
+    {
+      "epoch": 0.620253164556962,
+      "grad_norm": 0.4599224627017975,
+      "learning_rate": 0.00019968424606979298,
+      "loss": 0.8459385633468628,
+      "step": 1470
+    },
+    {
+      "epoch": 0.6210970464135022,
+      "grad_norm": 0.5476765632629395,
+      "learning_rate": 0.00019968042082419094,
+      "loss": 0.8844659328460693,
+      "step": 1472
+    },
+    {
+      "epoch": 0.6219409282700422,
+      "grad_norm": 0.46202269196510315,
+      "learning_rate": 0.00019967657258423022,
+      "loss": 0.860946536064148,
+      "step": 1474
+    },
+    {
+      "epoch": 0.6227848101265823,
+      "grad_norm": 0.39446666836738586,
+      "learning_rate": 0.00019967270135079853,
+      "loss": 0.7587860822677612,
+      "step": 1476
+    },
+    {
+      "epoch": 0.6236286919831223,
+      "grad_norm": 0.4500375986099243,
+      "learning_rate": 0.0001996688071247889,
+      "loss": 0.9143000245094299,
+      "step": 1478
+    },
+    {
+      "epoch": 0.6244725738396625,
+      "grad_norm": 0.4203545153141022,
+      "learning_rate": 0.0001996648899070996,
+      "loss": 0.7865519523620605,
+      "step": 1480
+    },
+    {
+      "epoch": 0.6253164556962025,
+      "grad_norm": 0.45623889565467834,
+      "learning_rate": 0.00019966094969863432,
+      "loss": 0.9192912578582764,
+      "step": 1482
+    },
+    {
+      "epoch": 0.6261603375527426,
+      "grad_norm": 0.4139637053012848,
+      "learning_rate": 0.00019965698650030195,
+      "loss": 0.8459161520004272,
+      "step": 1484
+    },
+    {
+      "epoch": 0.6270042194092827,
+      "grad_norm": 0.4410824179649353,
+      "learning_rate": 0.00019965300031301678,
+      "loss": 0.8159077763557434,
+      "step": 1486
+    },
+    {
+      "epoch": 0.6278481012658228,
+      "grad_norm": 0.43821197748184204,
+      "learning_rate": 0.00019964899113769828,
+      "loss": 0.884467363357544,
+      "step": 1488
+    },
+    {
+      "epoch": 0.6286919831223629,
+      "grad_norm": 0.4692091643810272,
+      "learning_rate": 0.00019964495897527133,
+      "loss": 0.8693601489067078,
+      "step": 1490
+    },
+    {
+      "epoch": 0.6295358649789029,
+      "grad_norm": 0.40715447068214417,
+      "learning_rate": 0.00019964090382666608,
+      "loss": 0.8897743225097656,
+      "step": 1492
+    },
+    {
+      "epoch": 0.6303797468354431,
+      "grad_norm": 0.4123334586620331,
+      "learning_rate": 0.0001996368256928179,
+      "loss": 0.8274733424186707,
+      "step": 1494
+    },
+    {
+      "epoch": 0.6312236286919831,
+      "grad_norm": 0.44347211718559265,
+      "learning_rate": 0.00019963272457466767,
+      "loss": 0.8158749938011169,
+      "step": 1496
+    },
+    {
+      "epoch": 0.6320675105485232,
+      "grad_norm": 0.4700329601764679,
+      "learning_rate": 0.00019962860047316135,
+      "loss": 0.8439569473266602,
+      "step": 1498
+    },
+    {
+      "epoch": 0.6329113924050633,
+      "grad_norm": 0.46415844559669495,
+      "learning_rate": 0.00019962445338925027,
+      "loss": 0.8316822648048401,
+      "step": 1500
+    },
+    {
+      "epoch": 0.6329113924050633,
+      "eval_loss": 0.887488842010498,
+      "eval_runtime": 686.2804,
+      "eval_samples_per_second": 3.07,
+      "eval_steps_per_second": 3.07,
+      "step": 1500
+    },
+    {
+      "epoch": 0.6337552742616034,
+      "grad_norm": 0.44183167815208435,
+      "learning_rate": 0.0001996202833238911,
+      "loss": 0.9121994972229004,
+      "step": 1502
+    },
+    {
+      "epoch": 0.6345991561181434,
+      "grad_norm": 0.4279208779335022,
+      "learning_rate": 0.00019961609027804587,
+      "loss": 0.9160211086273193,
+      "step": 1504
+    },
+    {
+      "epoch": 0.6354430379746835,
+      "grad_norm": 0.3967169523239136,
+      "learning_rate": 0.00019961187425268176,
+      "loss": 0.861677348613739,
+      "step": 1506
+    },
+    {
+      "epoch": 0.6362869198312237,
+      "grad_norm": 0.4796451926231384,
+      "learning_rate": 0.0001996076352487713,
+      "loss": 0.8478423357009888,
+      "step": 1508
+    },
+    {
+      "epoch": 0.6371308016877637,
+      "grad_norm": 0.3861426115036011,
+      "learning_rate": 0.00019960337326729245,
+      "loss": 0.8739159107208252,
+      "step": 1510
+    },
+    {
+      "epoch": 0.6379746835443038,
+      "grad_norm": 0.4567820429801941,
+      "learning_rate": 0.00019959908830922824,
+      "loss": 0.7693920731544495,
+      "step": 1512
+    },
+    {
+      "epoch": 0.6388185654008439,
+      "grad_norm": 0.4457108676433563,
+      "learning_rate": 0.00019959478037556724,
+      "loss": 0.8482301831245422,
+      "step": 1514
+    },
+    {
+      "epoch": 0.639662447257384,
+      "grad_norm": 0.43918928503990173,
+      "learning_rate": 0.00019959044946730314,
+      "loss": 0.8626812696456909,
+      "step": 1516
+    },
+    {
+      "epoch": 0.640506329113924,
+      "grad_norm": 0.4095900058746338,
+      "learning_rate": 0.00019958609558543504,
+      "loss": 0.8342230916023254,
+      "step": 1518
+    },
+    {
+      "epoch": 0.6413502109704642,
+      "grad_norm": 0.4181270897388458,
+      "learning_rate": 0.00019958171873096724,
+      "loss": 0.8794118165969849,
+      "step": 1520
+    },
+    {
+      "epoch": 0.6421940928270042,
+      "grad_norm": 0.4750959277153015,
+      "learning_rate": 0.00019957731890490947,
+      "loss": 0.7787677049636841,
+      "step": 1522
+    },
+    {
+      "epoch": 0.6430379746835443,
+      "grad_norm": 0.5129296779632568,
+      "learning_rate": 0.00019957289610827663,
+      "loss": 0.9084368348121643,
+      "step": 1524
+    },
+    {
+      "epoch": 0.6438818565400843,
+      "grad_norm": 0.4829029142856598,
+      "learning_rate": 0.000199568450342089,
+      "loss": 0.9066952466964722,
+      "step": 1526
+    },
+    {
+      "epoch": 0.6447257383966245,
+      "grad_norm": 0.4225057363510132,
+      "learning_rate": 0.00019956398160737214,
+      "loss": 0.8127874732017517,
+      "step": 1528
+    },
+    {
+      "epoch": 0.6455696202531646,
+      "grad_norm": 0.4231826364994049,
+      "learning_rate": 0.00019955948990515684,
+      "loss": 0.8781921863555908,
+      "step": 1530
+    },
+    {
+      "epoch": 0.6464135021097046,
+      "grad_norm": 0.4436216950416565,
+      "learning_rate": 0.00019955497523647933,
+      "loss": 0.8136996626853943,
+      "step": 1532
+    },
+    {
+      "epoch": 0.6472573839662448,
+      "grad_norm": 0.4018244743347168,
+      "learning_rate": 0.00019955043760238098,
+      "loss": 0.8003877401351929,
+      "step": 1534
+    },
+    {
+      "epoch": 0.6481012658227848,
+      "grad_norm": 0.40024611353874207,
+      "learning_rate": 0.0001995458770039086,
+      "loss": 0.8996267914772034,
+      "step": 1536
+    },
+    {
+      "epoch": 0.6489451476793249,
+      "grad_norm": 0.42845702171325684,
+      "learning_rate": 0.00019954129344211424,
+      "loss": 0.8610522150993347,
+      "step": 1538
+    },
+    {
+      "epoch": 0.6497890295358649,
+      "grad_norm": 0.4258182644844055,
+      "learning_rate": 0.00019953668691805517,
+      "loss": 0.8991837501525879,
+      "step": 1540
+    },
+    {
+      "epoch": 0.6506329113924051,
+      "grad_norm": 0.4483257830142975,
+      "learning_rate": 0.0001995320574327941,
+      "loss": 0.8150189518928528,
+      "step": 1542
+    },
+    {
+      "epoch": 0.6514767932489451,
+      "grad_norm": 0.4320627748966217,
+      "learning_rate": 0.0001995274049873989,
+      "loss": 0.8572840690612793,
+      "step": 1544
+    },
+    {
+      "epoch": 0.6523206751054852,
+      "grad_norm": 0.39950594305992126,
+      "learning_rate": 0.00019952272958294288,
+      "loss": 0.7766129374504089,
+      "step": 1546
+    },
+    {
+      "epoch": 0.6531645569620254,
+      "grad_norm": 0.4130117893218994,
+      "learning_rate": 0.00019951803122050446,
+      "loss": 0.9032199382781982,
+      "step": 1548
+    },
+    {
+      "epoch": 0.6540084388185654,
+      "grad_norm": 0.4436919391155243,
+      "learning_rate": 0.00019951330990116754,
+      "loss": 0.8361048102378845,
+      "step": 1550
+    },
+    {
+      "epoch": 0.6548523206751055,
+      "grad_norm": 0.4229227900505066,
+      "learning_rate": 0.00019950856562602121,
+      "loss": 0.8661768436431885,
+      "step": 1552
+    },
+    {
+      "epoch": 0.6556962025316456,
+      "grad_norm": 0.4839727580547333,
+      "learning_rate": 0.00019950379839615988,
+      "loss": 0.9737826585769653,
+      "step": 1554
+    },
+    {
+      "epoch": 0.6565400843881857,
+      "grad_norm": 0.3859386742115021,
+      "learning_rate": 0.00019949900821268328,
+      "loss": 0.7542453408241272,
+      "step": 1556
+    },
+    {
+      "epoch": 0.6573839662447257,
+      "grad_norm": 0.39643988013267517,
+      "learning_rate": 0.0001994941950766964,
+      "loss": 0.9226290583610535,
+      "step": 1558
+    },
+    {
+      "epoch": 0.6582278481012658,
+      "grad_norm": 0.4261355400085449,
+      "learning_rate": 0.00019948935898930954,
+      "loss": 0.8375602960586548,
+      "step": 1560
+    },
+    {
+      "epoch": 0.6590717299578059,
+      "grad_norm": 0.445931077003479,
+      "learning_rate": 0.00019948449995163826,
+      "loss": 0.7847310304641724,
+      "step": 1562
+    },
+    {
+      "epoch": 0.659915611814346,
+      "grad_norm": 0.4680459797382355,
+      "learning_rate": 0.00019947961796480353,
+      "loss": 0.851150631904602,
+      "step": 1564
+    },
+    {
+      "epoch": 0.660759493670886,
+      "grad_norm": 0.6826074719429016,
+      "learning_rate": 0.00019947471302993143,
+      "loss": 0.8259562253952026,
+      "step": 1566
+    },
+    {
+      "epoch": 0.6616033755274262,
+      "grad_norm": 0.47369205951690674,
+      "learning_rate": 0.00019946978514815352,
+      "loss": 0.879021406173706,
+      "step": 1568
+    },
+    {
+      "epoch": 0.6624472573839663,
+      "grad_norm": 0.41111937165260315,
+      "learning_rate": 0.0001994648343206065,
+      "loss": 0.9296056032180786,
+      "step": 1570
+    },
+    {
+      "epoch": 0.6632911392405063,
+      "grad_norm": 0.4297783076763153,
+      "learning_rate": 0.00019945986054843248,
+      "loss": 0.873822808265686,
+      "step": 1572
+    },
+    {
+      "epoch": 0.6641350210970464,
+      "grad_norm": 0.48124316334724426,
+      "learning_rate": 0.00019945486383277878,
+      "loss": 0.937151312828064,
+      "step": 1574
+    },
+    {
+      "epoch": 0.6649789029535865,
+      "grad_norm": 0.4048190116882324,
+      "learning_rate": 0.00019944984417479805,
+      "loss": 0.8398929834365845,
+      "step": 1576
+    },
+    {
+      "epoch": 0.6658227848101266,
+      "grad_norm": 0.4023774266242981,
+      "learning_rate": 0.00019944480157564826,
+      "loss": 0.8269520998001099,
+      "step": 1578
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.3784184157848358,
+      "learning_rate": 0.00019943973603649262,
+      "loss": 0.8237608671188354,
+      "step": 1580
+    },
+    {
+      "epoch": 0.6675105485232068,
+      "grad_norm": 0.4448573887348175,
+      "learning_rate": 0.00019943464755849965,
+      "loss": 0.8469099998474121,
+      "step": 1582
+    },
+    {
+      "epoch": 0.6683544303797468,
+      "grad_norm": 0.4690774083137512,
+      "learning_rate": 0.00019942953614284317,
+      "loss": 0.8630822896957397,
+      "step": 1584
+    },
+    {
+      "epoch": 0.6691983122362869,
+      "grad_norm": 0.43670853972435,
+      "learning_rate": 0.00019942440179070227,
+      "loss": 0.8904082179069519,
+      "step": 1586
+    },
+    {
+      "epoch": 0.6700421940928271,
+      "grad_norm": 0.4236849844455719,
+      "learning_rate": 0.00019941924450326138,
+      "loss": 0.8355640172958374,
+      "step": 1588
+    },
+    {
+      "epoch": 0.6708860759493671,
+      "grad_norm": 0.48624539375305176,
+      "learning_rate": 0.00019941406428171013,
+      "loss": 0.8898435831069946,
+      "step": 1590
+    },
+    {
+      "epoch": 0.6717299578059072,
+      "grad_norm": 0.4453780949115753,
+      "learning_rate": 0.00019940886112724355,
+      "loss": 0.9000012278556824,
+      "step": 1592
+    },
+    {
+      "epoch": 0.6725738396624472,
+      "grad_norm": 0.4417199194431305,
+      "learning_rate": 0.00019940363504106193,
+      "loss": 0.8171058893203735,
+      "step": 1594
+    },
+    {
+      "epoch": 0.6734177215189874,
+      "grad_norm": 0.45055362582206726,
+      "learning_rate": 0.00019939838602437074,
+      "loss": 0.8703798055648804,
+      "step": 1596
+    },
+    {
+      "epoch": 0.6742616033755274,
+      "grad_norm": 0.42316532135009766,
+      "learning_rate": 0.00019939311407838088,
+      "loss": 0.8677281737327576,
+      "step": 1598
+    },
+    {
+      "epoch": 0.6751054852320675,
+      "grad_norm": 0.3935781717300415,
+      "learning_rate": 0.0001993878192043085,
+      "loss": 0.7314013242721558,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6751054852320675,
+      "eval_loss": 0.8769772052764893,
+      "eval_runtime": 677.9338,
+      "eval_samples_per_second": 3.108,
+      "eval_steps_per_second": 3.108,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6759493670886076,
+      "grad_norm": 0.4197012186050415,
+      "learning_rate": 0.000199382501403375,
+      "loss": 0.8075380325317383,
+      "step": 1602
+    },
+    {
+      "epoch": 0.6767932489451477,
+      "grad_norm": 0.45227327942848206,
+      "learning_rate": 0.00019937716067680713,
+      "loss": 0.7885794639587402,
+      "step": 1604
+    },
+    {
+      "epoch": 0.6776371308016877,
+      "grad_norm": 0.45660385489463806,
+      "learning_rate": 0.00019937179702583682,
+      "loss": 0.868261992931366,
+      "step": 1606
+    },
+    {
+      "epoch": 0.6784810126582278,
+      "grad_norm": 0.3955288231372833,
+      "learning_rate": 0.00019936641045170144,
+      "loss": 0.8306655287742615,
+      "step": 1608
+    },
+    {
+      "epoch": 0.679324894514768,
+      "grad_norm": 0.3651324510574341,
+      "learning_rate": 0.00019936100095564353,
+      "loss": 0.7505315542221069,
+      "step": 1610
+    },
+    {
+      "epoch": 0.680168776371308,
+      "grad_norm": 0.4240955412387848,
+      "learning_rate": 0.00019935556853891096,
+      "loss": 0.8792756795883179,
+      "step": 1612
+    },
+    {
+      "epoch": 0.6810126582278481,
+      "grad_norm": 0.4142507314682007,
+      "learning_rate": 0.00019935011320275687,
+      "loss": 0.7328272461891174,
+      "step": 1614
+    },
+    {
+      "epoch": 0.6818565400843882,
+      "grad_norm": 0.48339322209358215,
+      "learning_rate": 0.00019934463494843975,
+      "loss": 0.8414849042892456,
+      "step": 1616
+    },
+    {
+      "epoch": 0.6827004219409283,
+      "grad_norm": 0.4312443435192108,
+      "learning_rate": 0.00019933913377722328,
+      "loss": 0.8588716983795166,
+      "step": 1618
+    },
+    {
+      "epoch": 0.6835443037974683,
+      "grad_norm": 0.43849512934684753,
+      "learning_rate": 0.0001993336096903765,
+      "loss": 0.846511960029602,
+      "step": 1620
+    },
+    {
+      "epoch": 0.6843881856540084,
+      "grad_norm": 0.45767566561698914,
+      "learning_rate": 0.0001993280626891737,
+      "loss": 0.9352323412895203,
+      "step": 1622
+    },
+    {
+      "epoch": 0.6852320675105485,
+      "grad_norm": 0.438412606716156,
+      "learning_rate": 0.00019932249277489444,
+      "loss": 0.8884757161140442,
+      "step": 1624
+    },
+    {
+      "epoch": 0.6860759493670886,
+      "grad_norm": 0.45893922448158264,
+      "learning_rate": 0.0001993168999488237,
+      "loss": 0.7739649415016174,
+      "step": 1626
+    },
+    {
+      "epoch": 0.6869198312236287,
+      "grad_norm": 0.38631772994995117,
+      "learning_rate": 0.0001993112842122515,
+      "loss": 0.7948258519172668,
+      "step": 1628
+    },
+    {
+      "epoch": 0.6877637130801688,
+      "grad_norm": 0.3824027180671692,
+      "learning_rate": 0.00019930564556647334,
+      "loss": 0.750511884689331,
+      "step": 1630
+    },
+    {
+      "epoch": 0.6886075949367089,
+      "grad_norm": 0.3980776071548462,
+      "learning_rate": 0.00019929998401278996,
+      "loss": 0.9006738662719727,
+      "step": 1632
+    },
+    {
+      "epoch": 0.6894514767932489,
+      "grad_norm": 0.41419896483421326,
+      "learning_rate": 0.00019929429955250734,
+      "loss": 0.6996869444847107,
+      "step": 1634
+    },
+    {
+      "epoch": 0.6902953586497891,
+      "grad_norm": 0.43783676624298096,
+      "learning_rate": 0.00019928859218693682,
+      "loss": 0.8523393273353577,
+      "step": 1636
+    },
+    {
+      "epoch": 0.6911392405063291,
+      "grad_norm": 0.4378126561641693,
+      "learning_rate": 0.00019928286191739497,
+      "loss": 0.8885987401008606,
+      "step": 1638
+    },
+    {
+      "epoch": 0.6919831223628692,
+      "grad_norm": 0.4276133179664612,
+      "learning_rate": 0.00019927710874520363,
+      "loss": 0.8563809990882874,
+      "step": 1640
+    },
+    {
+      "epoch": 0.6928270042194092,
+      "grad_norm": 0.4600282311439514,
+      "learning_rate": 0.00019927133267168992,
+      "loss": 0.854114294052124,
+      "step": 1642
+    },
+    {
+      "epoch": 0.6936708860759494,
+      "grad_norm": 0.5058461427688599,
+      "learning_rate": 0.00019926553369818635,
+      "loss": 0.8557889461517334,
+      "step": 1644
+    },
+    {
+      "epoch": 0.6945147679324895,
+      "grad_norm": 0.4947412312030792,
+      "learning_rate": 0.00019925971182603057,
+      "loss": 0.8622767925262451,
+      "step": 1646
+    },
+    {
+      "epoch": 0.6953586497890295,
+      "grad_norm": 0.42555496096611023,
+      "learning_rate": 0.00019925386705656558,
+      "loss": 0.7836558818817139,
+      "step": 1648
+    },
+    {
+      "epoch": 0.6962025316455697,
+      "grad_norm": 0.47560831904411316,
+      "learning_rate": 0.00019924799939113967,
+      "loss": 0.7877846956253052,
+      "step": 1650
+    },
+    {
+      "epoch": 0.6970464135021097,
+      "grad_norm": 0.391702264547348,
+      "learning_rate": 0.0001992421088311064,
+      "loss": 0.775681734085083,
+      "step": 1652
+    },
+    {
+      "epoch": 0.6978902953586498,
+      "grad_norm": 0.463679701089859,
+      "learning_rate": 0.0001992361953778246,
+      "loss": 0.8809882402420044,
+      "step": 1654
+    },
+    {
+      "epoch": 0.6987341772151898,
+      "grad_norm": 0.44096484780311584,
+      "learning_rate": 0.00019923025903265836,
+      "loss": 0.8513299226760864,
+      "step": 1656
+    },
+    {
+      "epoch": 0.69957805907173,
+      "grad_norm": 0.47818854451179504,
+      "learning_rate": 0.0001992242997969772,
+      "loss": 0.8309667706489563,
+      "step": 1658
+    },
+    {
+      "epoch": 0.70042194092827,
+      "grad_norm": 0.4192790985107422,
+      "learning_rate": 0.00019921831767215565,
+      "loss": 0.8817558884620667,
+      "step": 1660
+    },
+    {
+      "epoch": 0.7012658227848101,
+      "grad_norm": 0.45285239815711975,
+      "learning_rate": 0.00019921231265957373,
+      "loss": 0.8315755724906921,
+      "step": 1662
+    },
+    {
+      "epoch": 0.7021097046413503,
+      "grad_norm": 0.47118625044822693,
+      "learning_rate": 0.00019920628476061673,
+      "loss": 0.8698058724403381,
+      "step": 1664
+    },
+    {
+      "epoch": 0.7029535864978903,
+      "grad_norm": 0.4769238829612732,
+      "learning_rate": 0.0001992002339766751,
+      "loss": 0.953620970249176,
+      "step": 1666
+    },
+    {
+      "epoch": 0.7037974683544304,
+      "grad_norm": 0.39397355914115906,
+      "learning_rate": 0.00019919416030914468,
+      "loss": 0.8536615371704102,
+      "step": 1668
+    },
+    {
+      "epoch": 0.7046413502109705,
+      "grad_norm": 0.3659648299217224,
+      "learning_rate": 0.00019918806375942655,
+      "loss": 0.8159828782081604,
+      "step": 1670
+    },
+    {
+      "epoch": 0.7054852320675106,
+      "grad_norm": 0.4381789565086365,
+      "learning_rate": 0.00019918194432892703,
+      "loss": 0.782463550567627,
+      "step": 1672
+    },
+    {
+      "epoch": 0.7063291139240506,
+      "grad_norm": 0.44745752215385437,
+      "learning_rate": 0.00019917580201905778,
+      "loss": 0.9314272999763489,
+      "step": 1674
+    },
+    {
+      "epoch": 0.7071729957805907,
+      "grad_norm": 0.4515164792537689,
+      "learning_rate": 0.00019916963683123574,
+      "loss": 0.860816240310669,
+      "step": 1676
+    },
+    {
+      "epoch": 0.7080168776371308,
+      "grad_norm": 0.42735886573791504,
+      "learning_rate": 0.00019916344876688303,
+      "loss": 0.8008456826210022,
+      "step": 1678
+    },
+    {
+      "epoch": 0.7088607594936709,
+      "grad_norm": 0.418275386095047,
+      "learning_rate": 0.00019915723782742718,
+      "loss": 0.7796595096588135,
+      "step": 1680
+    },
+    {
+      "epoch": 0.7097046413502109,
+      "grad_norm": 0.44347989559173584,
+      "learning_rate": 0.00019915100401430089,
+      "loss": 0.9006506204605103,
+      "step": 1682
+    },
+    {
+      "epoch": 0.7105485232067511,
+      "grad_norm": 0.3932702839374542,
+      "learning_rate": 0.00019914474732894225,
+      "loss": 0.7750146985054016,
+      "step": 1684
+    },
+    {
+      "epoch": 0.7113924050632912,
+      "grad_norm": 0.4810502231121063,
+      "learning_rate": 0.00019913846777279447,
+      "loss": 0.7508097290992737,
+      "step": 1686
+    },
+    {
+      "epoch": 0.7122362869198312,
+      "grad_norm": 0.4166778028011322,
+      "learning_rate": 0.00019913216534730622,
+      "loss": 0.7926796078681946,
+      "step": 1688
+    },
+    {
+      "epoch": 0.7130801687763713,
+      "grad_norm": 0.47900474071502686,
+      "learning_rate": 0.00019912584005393123,
+      "loss": 0.8363928198814392,
+      "step": 1690
+    },
+    {
+      "epoch": 0.7139240506329114,
+      "grad_norm": 0.435143381357193,
+      "learning_rate": 0.00019911949189412872,
+      "loss": 0.8649623990058899,
+      "step": 1692
+    },
+    {
+      "epoch": 0.7147679324894515,
+      "grad_norm": 0.4070208966732025,
+      "learning_rate": 0.00019911312086936305,
+      "loss": 0.8196499943733215,
+      "step": 1694
+    },
+    {
+      "epoch": 0.7156118143459915,
+      "grad_norm": 0.4717724621295929,
+      "learning_rate": 0.0001991067269811039,
+      "loss": 0.9157413840293884,
+      "step": 1696
+    },
+    {
+      "epoch": 0.7164556962025317,
+      "grad_norm": 0.3942880928516388,
+      "learning_rate": 0.0001991003102308262,
+      "loss": 0.7880456447601318,
+      "step": 1698
+    },
+    {
+      "epoch": 0.7172995780590717,
+      "grad_norm": 0.4037776589393616,
+      "learning_rate": 0.0001990938706200102,
+      "loss": 0.8088646531105042,
+      "step": 1700
+    },
+    {
+      "epoch": 0.7172995780590717,
+      "eval_loss": 0.8708170056343079,
+      "eval_runtime": 670.3019,
+      "eval_samples_per_second": 3.143,
+      "eval_steps_per_second": 3.143,
+      "step": 1700
+    },
+    {
+      "epoch": 0.7181434599156118,
+      "grad_norm": 0.4443536400794983,
+      "learning_rate": 0.0001990874081501414,
+      "loss": 0.8740324378013611,
+      "step": 1702
+    },
+    {
+      "epoch": 0.7189873417721518,
+      "grad_norm": 0.45089733600616455,
+      "learning_rate": 0.00019908092282271053,
+      "loss": 0.8305023908615112,
+      "step": 1704
+    },
+    {
+      "epoch": 0.719831223628692,
+      "grad_norm": 0.3903036415576935,
+      "learning_rate": 0.00019907441463921366,
+      "loss": 0.7718394994735718,
+      "step": 1706
+    },
+    {
+      "epoch": 0.7206751054852321,
+      "grad_norm": 0.42990729212760925,
+      "learning_rate": 0.00019906788360115208,
+      "loss": 0.843289315700531,
+      "step": 1708
+    },
+    {
+      "epoch": 0.7215189873417721,
+      "grad_norm": 0.4287833571434021,
+      "learning_rate": 0.0001990613297100324,
+      "loss": 0.8332282900810242,
+      "step": 1710
+    },
+    {
+      "epoch": 0.7223628691983123,
+      "grad_norm": 0.4557269513607025,
+      "learning_rate": 0.00019905475296736646,
+      "loss": 0.784180223941803,
+      "step": 1712
+    },
+    {
+      "epoch": 0.7232067510548523,
+      "grad_norm": 0.4001525640487671,
+      "learning_rate": 0.0001990481533746714,
+      "loss": 0.7807716131210327,
+      "step": 1714
+    },
+    {
+      "epoch": 0.7240506329113924,
+      "grad_norm": 0.5208709836006165,
+      "learning_rate": 0.0001990415309334696,
+      "loss": 0.7936414480209351,
+      "step": 1716
+    },
+    {
+      "epoch": 0.7248945147679325,
+      "grad_norm": 0.46469953656196594,
+      "learning_rate": 0.00019903488564528875,
+      "loss": 0.7964845895767212,
+      "step": 1718
+    },
+    {
+      "epoch": 0.7257383966244726,
+      "grad_norm": 0.405773401260376,
+      "learning_rate": 0.00019902821751166178,
+      "loss": 0.8302215337753296,
+      "step": 1720
+    },
+    {
+      "epoch": 0.7265822784810126,
+      "grad_norm": 0.3874973654747009,
+      "learning_rate": 0.00019902152653412692,
+      "loss": 0.8486512899398804,
+      "step": 1722
+    },
+    {
+      "epoch": 0.7274261603375527,
+      "grad_norm": 0.46865755319595337,
+      "learning_rate": 0.00019901481271422758,
+      "loss": 0.837841272354126,
+      "step": 1724
+    },
+    {
+      "epoch": 0.7282700421940929,
+      "grad_norm": 0.4139123558998108,
+      "learning_rate": 0.0001990080760535126,
+      "loss": 0.7640942335128784,
+      "step": 1726
+    },
+    {
+      "epoch": 0.7291139240506329,
+      "grad_norm": 0.5003024935722351,
+      "learning_rate": 0.000199001316553536,
+      "loss": 0.8236244320869446,
+      "step": 1728
+    },
+    {
+      "epoch": 0.729957805907173,
+      "grad_norm": 0.38495776057243347,
+      "learning_rate": 0.00019899453421585696,
+      "loss": 0.7792612314224243,
+      "step": 1730
+    },
+    {
+      "epoch": 0.7308016877637131,
+      "grad_norm": 0.4462733268737793,
+      "learning_rate": 0.00019898772904204014,
+      "loss": 0.7733872532844543,
+      "step": 1732
+    },
+    {
+      "epoch": 0.7316455696202532,
+      "grad_norm": 0.398248553276062,
+      "learning_rate": 0.00019898090103365532,
+      "loss": 0.8933147192001343,
+      "step": 1734
+    },
+    {
+      "epoch": 0.7324894514767932,
+      "grad_norm": 0.4231175482273102,
+      "learning_rate": 0.00019897405019227756,
+      "loss": 0.8198356628417969,
+      "step": 1736
+    },
+    {
+      "epoch": 0.7333333333333333,
+      "grad_norm": 0.43639084696769714,
+      "learning_rate": 0.0001989671765194873,
+      "loss": 0.8266391158103943,
+      "step": 1738
+    },
+    {
+      "epoch": 0.7341772151898734,
+      "grad_norm": 0.4235835671424866,
+      "learning_rate": 0.00019896028001687008,
+      "loss": 0.7839997410774231,
+      "step": 1740
+    },
+    {
+      "epoch": 0.7350210970464135,
+      "grad_norm": 0.4033111035823822,
+      "learning_rate": 0.00019895336068601687,
+      "loss": 0.7153518199920654,
+      "step": 1742
+    },
+    {
+      "epoch": 0.7358649789029535,
+      "grad_norm": 0.44281265139579773,
+      "learning_rate": 0.00019894641852852377,
+      "loss": 0.9079289436340332,
+      "step": 1744
+    },
+    {
+      "epoch": 0.7367088607594937,
+      "grad_norm": 0.4882747232913971,
+      "learning_rate": 0.0001989394535459922,
+      "loss": 0.8645254373550415,
+      "step": 1746
+    },
+    {
+      "epoch": 0.7375527426160338,
+      "grad_norm": 0.34858641028404236,
+      "learning_rate": 0.00019893246574002887,
+      "loss": 0.8134214878082275,
+      "step": 1748
+    },
+    {
+      "epoch": 0.7383966244725738,
+      "grad_norm": 0.4273303747177124,
+      "learning_rate": 0.00019892545511224574,
+      "loss": 0.8142663240432739,
+      "step": 1750
+    },
+    {
+      "epoch": 0.739240506329114,
+      "grad_norm": 0.4289957284927368,
+      "learning_rate": 0.00019891842166426004,
+      "loss": 0.7957643866539001,
+      "step": 1752
+    },
+    {
+      "epoch": 0.740084388185654,
+      "grad_norm": 0.4812224805355072,
+      "learning_rate": 0.00019891136539769416,
+      "loss": 0.9000363945960999,
+      "step": 1754
+    },
+    {
+      "epoch": 0.7409282700421941,
+      "grad_norm": 0.4078022837638855,
+      "learning_rate": 0.00019890428631417595,
+      "loss": 0.7719792723655701,
+      "step": 1756
+    },
+    {
+      "epoch": 0.7417721518987341,
+      "grad_norm": 0.40169858932495117,
+      "learning_rate": 0.00019889718441533838,
+      "loss": 0.7681847214698792,
+      "step": 1758
+    },
+    {
+      "epoch": 0.7426160337552743,
+      "grad_norm": 0.3713594377040863,
+      "learning_rate": 0.00019889005970281972,
+      "loss": 0.7536827325820923,
+      "step": 1760
+    },
+    {
+      "epoch": 0.7434599156118143,
+      "grad_norm": 0.4183627665042877,
+      "learning_rate": 0.0001988829121782635,
+      "loss": 0.9000160694122314,
+      "step": 1762
+    },
+    {
+      "epoch": 0.7443037974683544,
+      "grad_norm": 0.4241594672203064,
+      "learning_rate": 0.00019887574184331851,
+      "loss": 0.8047307133674622,
+      "step": 1764
+    },
+    {
+      "epoch": 0.7451476793248946,
+      "grad_norm": 0.4249233603477478,
+      "learning_rate": 0.00019886854869963883,
+      "loss": 0.8932583928108215,
+      "step": 1766
+    },
+    {
+      "epoch": 0.7459915611814346,
+      "grad_norm": 0.35815340280532837,
+      "learning_rate": 0.00019886133274888378,
+      "loss": 0.7684977054595947,
+      "step": 1768
+    },
+    {
+      "epoch": 0.7468354430379747,
+      "grad_norm": 0.41035401821136475,
+      "learning_rate": 0.00019885409399271795,
+      "loss": 0.8644338250160217,
+      "step": 1770
+    },
+    {
+      "epoch": 0.7476793248945147,
+      "grad_norm": 0.5450286269187927,
+      "learning_rate": 0.00019884683243281116,
+      "loss": 0.8035860657691956,
+      "step": 1772
+    },
+    {
+      "epoch": 0.7485232067510549,
+      "grad_norm": 0.5280999541282654,
+      "learning_rate": 0.0001988395480708385,
+      "loss": 0.8861207962036133,
+      "step": 1774
+    },
+    {
+      "epoch": 0.7493670886075949,
+      "grad_norm": 0.42665231227874756,
+      "learning_rate": 0.00019883224090848036,
+      "loss": 0.7506847977638245,
+      "step": 1776
+    },
+    {
+      "epoch": 0.750210970464135,
+      "grad_norm": 0.39029282331466675,
+      "learning_rate": 0.00019882491094742232,
+      "loss": 0.7662046551704407,
+      "step": 1778
+    },
+    {
+      "epoch": 0.7510548523206751,
+      "grad_norm": 0.43870019912719727,
+      "learning_rate": 0.00019881755818935534,
+      "loss": 0.7590143084526062,
+      "step": 1780
+    },
+    {
+      "epoch": 0.7518987341772152,
+      "grad_norm": 0.4195050001144409,
+      "learning_rate": 0.0001988101826359755,
+      "loss": 0.9340365529060364,
+      "step": 1782
+    },
+    {
+      "epoch": 0.7527426160337553,
+      "grad_norm": 0.4236123263835907,
+      "learning_rate": 0.0001988027842889842,
+      "loss": 0.7969209551811218,
+      "step": 1784
+    },
+    {
+      "epoch": 0.7535864978902953,
+      "grad_norm": 0.4274357259273529,
+      "learning_rate": 0.00019879536315008808,
+      "loss": 0.7892382144927979,
+      "step": 1786
+    },
+    {
+      "epoch": 0.7544303797468355,
+      "grad_norm": 0.4622916579246521,
+      "learning_rate": 0.00019878791922099912,
+      "loss": 0.8752562403678894,
+      "step": 1788
+    },
+    {
+      "epoch": 0.7552742616033755,
+      "grad_norm": 0.3648734986782074,
+      "learning_rate": 0.00019878045250343445,
+      "loss": 0.813011884689331,
+      "step": 1790
+    },
+    {
+      "epoch": 0.7561181434599156,
+      "grad_norm": 0.41361239552497864,
+      "learning_rate": 0.00019877296299911648,
+      "loss": 0.7310198545455933,
+      "step": 1792
+    },
+    {
+      "epoch": 0.7569620253164557,
+      "grad_norm": 0.4113222062587738,
+      "learning_rate": 0.00019876545070977294,
+      "loss": 0.858386754989624,
+      "step": 1794
+    },
+    {
+      "epoch": 0.7578059071729958,
+      "grad_norm": 0.4295370876789093,
+      "learning_rate": 0.00019875791563713676,
+      "loss": 0.8377325534820557,
+      "step": 1796
+    },
+    {
+      "epoch": 0.7586497890295358,
+      "grad_norm": 0.4227522015571594,
+      "learning_rate": 0.0001987503577829461,
+      "loss": 0.9139418005943298,
+      "step": 1798
+    },
+    {
+      "epoch": 0.759493670886076,
+      "grad_norm": 0.3638151288032532,
+      "learning_rate": 0.00019874277714894442,
+      "loss": 0.6994872689247131,
+      "step": 1800
+    },
+    {
+      "epoch": 0.759493670886076,
+      "eval_loss": 0.8625519275665283,
+      "eval_runtime": 686.4271,
+      "eval_samples_per_second": 3.07,
+      "eval_steps_per_second": 3.07,
+      "step": 1800
+    },
+    {
+      "epoch": 0.760337552742616,
+      "grad_norm": 0.46954411268234253,
+      "learning_rate": 0.00019873517373688047,
+      "loss": 0.8681167960166931,
+      "step": 1802
+    },
+    {
+      "epoch": 0.7611814345991561,
+      "grad_norm": 0.41958674788475037,
+      "learning_rate": 0.00019872754754850819,
+      "loss": 0.7420852184295654,
+      "step": 1804
+    },
+    {
+      "epoch": 0.7620253164556962,
+      "grad_norm": 0.431226521730423,
+      "learning_rate": 0.00019871989858558678,
+      "loss": 0.8018608093261719,
+      "step": 1806
+    },
+    {
+      "epoch": 0.7628691983122363,
+      "grad_norm": 0.45129454135894775,
+      "learning_rate": 0.0001987122268498807,
+      "loss": 0.8793904781341553,
+      "step": 1808
+    },
+    {
+      "epoch": 0.7637130801687764,
+      "grad_norm": 0.44278961420059204,
+      "learning_rate": 0.00019870453234315972,
+      "loss": 0.8416730761528015,
+      "step": 1810
+    },
+    {
+      "epoch": 0.7645569620253164,
+      "grad_norm": 0.4689098000526428,
+      "learning_rate": 0.00019869681506719876,
+      "loss": 0.8225743174552917,
+      "step": 1812
+    },
+    {
+      "epoch": 0.7654008438818566,
+      "grad_norm": 0.4413386881351471,
+      "learning_rate": 0.00019868907502377806,
+      "loss": 0.7871913313865662,
+      "step": 1814
+    },
+    {
+      "epoch": 0.7662447257383966,
+      "grad_norm": 0.48395273089408875,
+      "learning_rate": 0.00019868131221468315,
+      "loss": 0.8616237044334412,
+      "step": 1816
+    },
+    {
+      "epoch": 0.7670886075949367,
+      "grad_norm": 0.370263934135437,
+      "learning_rate": 0.0001986735266417047,
+      "loss": 0.7243452668190002,
+      "step": 1818
+    },
+    {
+      "epoch": 0.7679324894514767,
+      "grad_norm": 0.4522445797920227,
+      "learning_rate": 0.00019866571830663875,
+      "loss": 0.8623812198638916,
+      "step": 1820
+    },
+    {
+      "epoch": 0.7687763713080169,
+      "grad_norm": 0.43821004033088684,
+      "learning_rate": 0.00019865788721128643,
+      "loss": 0.776618480682373,
+      "step": 1822
+    },
+    {
+      "epoch": 0.769620253164557,
+      "grad_norm": 0.3825650215148926,
+      "learning_rate": 0.00019865003335745436,
+      "loss": 0.8722774982452393,
+      "step": 1824
+    },
+    {
+      "epoch": 0.770464135021097,
+      "grad_norm": 0.43684205412864685,
+      "learning_rate": 0.00019864215674695416,
+      "loss": 0.907802939414978,
+      "step": 1826
+    },
+    {
+      "epoch": 0.7713080168776372,
+      "grad_norm": 0.4128975570201874,
+      "learning_rate": 0.00019863425738160287,
+      "loss": 0.8505539298057556,
+      "step": 1828
+    },
+    {
+      "epoch": 0.7721518987341772,
+      "grad_norm": 0.4254065454006195,
+      "learning_rate": 0.0001986263352632227,
+      "loss": 0.83274906873703,
+      "step": 1830
+    },
+    {
+      "epoch": 0.7729957805907173,
+      "grad_norm": 0.4137375056743622,
+      "learning_rate": 0.00019861839039364118,
+      "loss": 0.8172078728675842,
+      "step": 1832
+    },
+    {
+      "epoch": 0.7738396624472574,
+      "grad_norm": 0.4387703239917755,
+      "learning_rate": 0.00019861042277469096,
+      "loss": 0.8610842227935791,
+      "step": 1834
+    },
+    {
+      "epoch": 0.7746835443037975,
+      "grad_norm": 0.398952454328537,
+      "learning_rate": 0.00019860243240821005,
+      "loss": 0.7961188554763794,
+      "step": 1836
+    },
+    {
+      "epoch": 0.7755274261603375,
+      "grad_norm": 0.4079921543598175,
+      "learning_rate": 0.0001985944192960417,
+      "loss": 0.8078321218490601,
+      "step": 1838
+    },
+    {
+      "epoch": 0.7763713080168776,
+      "grad_norm": 0.4701489508152008,
+      "learning_rate": 0.00019858638344003433,
+      "loss": 0.8743909001350403,
+      "step": 1840
+    },
+    {
+      "epoch": 0.7772151898734178,
+      "grad_norm": 0.5391269326210022,
+      "learning_rate": 0.00019857832484204173,
+      "loss": 0.7681707143783569,
+      "step": 1842
+    },
+    {
+      "epoch": 0.7780590717299578,
+      "grad_norm": 0.36250734329223633,
+      "learning_rate": 0.00019857024350392277,
+      "loss": 0.7817525863647461,
+      "step": 1844
+    },
+    {
+      "epoch": 0.7789029535864979,
+      "grad_norm": 0.41237136721611023,
+      "learning_rate": 0.00019856213942754175,
+      "loss": 0.7614551782608032,
+      "step": 1846
+    },
+    {
+      "epoch": 0.779746835443038,
+      "grad_norm": 0.3968869745731354,
+      "learning_rate": 0.00019855401261476807,
+      "loss": 0.7492313981056213,
+      "step": 1848
+    },
+    {
+      "epoch": 0.7805907172995781,
+      "grad_norm": 0.441476434469223,
+      "learning_rate": 0.00019854586306747646,
+      "loss": 0.8615695238113403,
+      "step": 1850
+    },
+    {
+      "epoch": 0.7814345991561181,
+      "grad_norm": 0.4486154317855835,
+      "learning_rate": 0.00019853769078754686,
+      "loss": 0.8416724801063538,
+      "step": 1852
+    },
+    {
+      "epoch": 0.7822784810126582,
+      "grad_norm": 0.45148879289627075,
+      "learning_rate": 0.0001985294957768644,
+      "loss": 0.8987806439399719,
+      "step": 1854
+    },
+    {
+      "epoch": 0.7831223628691983,
+      "grad_norm": 0.39330095052719116,
+      "learning_rate": 0.0001985212780373196,
+      "loss": 0.7597590088844299,
+      "step": 1856
+    },
+    {
+      "epoch": 0.7839662447257384,
+      "grad_norm": 0.4260408580303192,
+      "learning_rate": 0.00019851303757080808,
+      "loss": 0.8429927229881287,
+      "step": 1858
+    },
+    {
+      "epoch": 0.7848101265822784,
+      "grad_norm": 0.37519171833992004,
+      "learning_rate": 0.00019850477437923075,
+      "loss": 0.7443564534187317,
+      "step": 1860
+    },
+    {
+      "epoch": 0.7856540084388186,
+      "grad_norm": 0.4991084337234497,
+      "learning_rate": 0.00019849648846449383,
+      "loss": 0.934238612651825,
+      "step": 1862
+    },
+    {
+      "epoch": 0.7864978902953587,
+      "grad_norm": 0.4711335003376007,
+      "learning_rate": 0.00019848817982850867,
+      "loss": 0.8938905000686646,
+      "step": 1864
+    },
+    {
+      "epoch": 0.7873417721518987,
+      "grad_norm": 0.4145370423793793,
+      "learning_rate": 0.00019847984847319193,
+      "loss": 0.8745643496513367,
+      "step": 1866
+    },
+    {
+      "epoch": 0.7881856540084389,
+      "grad_norm": 0.3444044888019562,
+      "learning_rate": 0.00019847149440046548,
+      "loss": 0.7178786396980286,
+      "step": 1868
+    },
+    {
+      "epoch": 0.7890295358649789,
+      "grad_norm": 0.34632962942123413,
+      "learning_rate": 0.00019846311761225646,
+      "loss": 0.9114303588867188,
+      "step": 1870
+    },
+    {
+      "epoch": 0.789873417721519,
+      "grad_norm": 0.38630741834640503,
+      "learning_rate": 0.0001984547181104972,
+      "loss": 0.836281955242157,
+      "step": 1872
+    },
+    {
+      "epoch": 0.790717299578059,
+      "grad_norm": 0.40947434306144714,
+      "learning_rate": 0.00019844629589712534,
+      "loss": 0.8247858881950378,
+      "step": 1874
+    },
+    {
+      "epoch": 0.7915611814345992,
+      "grad_norm": 0.45878684520721436,
+      "learning_rate": 0.00019843785097408372,
+      "loss": 0.8446351289749146,
+      "step": 1876
+    },
+    {
+      "epoch": 0.7924050632911392,
+      "grad_norm": 0.37202852964401245,
+      "learning_rate": 0.0001984293833433204,
+      "loss": 0.8567686676979065,
+      "step": 1878
+    },
+    {
+      "epoch": 0.7932489451476793,
+      "grad_norm": 0.3839069902896881,
+      "learning_rate": 0.00019842089300678876,
+      "loss": 0.8134047985076904,
+      "step": 1880
+    },
+    {
+      "epoch": 0.7940928270042195,
+      "grad_norm": 0.4018687605857849,
+      "learning_rate": 0.00019841237996644727,
+      "loss": 0.8670110106468201,
+      "step": 1882
+    },
+    {
+      "epoch": 0.7949367088607595,
+      "grad_norm": 0.372089684009552,
+      "learning_rate": 0.0001984038442242598,
+      "loss": 0.7975355386734009,
+      "step": 1884
+    },
+    {
+      "epoch": 0.7957805907172996,
+      "grad_norm": 0.4492705166339874,
+      "learning_rate": 0.0001983952857821953,
+      "loss": 0.8755611777305603,
+      "step": 1886
+    },
+    {
+      "epoch": 0.7966244725738396,
+      "grad_norm": 0.451054185628891,
+      "learning_rate": 0.00019838670464222813,
+      "loss": 0.8680241107940674,
+      "step": 1888
+    },
+    {
+      "epoch": 0.7974683544303798,
+      "grad_norm": 0.4036900997161865,
+      "learning_rate": 0.00019837810080633773,
+      "loss": 0.7990990877151489,
+      "step": 1890
+    },
+    {
+      "epoch": 0.7983122362869198,
+      "grad_norm": 0.46125802397727966,
+      "learning_rate": 0.0001983694742765089,
+      "loss": 0.8598953485488892,
+      "step": 1892
+    },
+    {
+      "epoch": 0.7991561181434599,
+      "grad_norm": 0.4458785951137543,
+      "learning_rate": 0.00019836082505473153,
+      "loss": 0.7707474827766418,
+      "step": 1894
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.44501590728759766,
+      "learning_rate": 0.0001983521531430009,
+      "loss": 0.8069534301757812,
+      "step": 1896
+    },
+    {
+      "epoch": 0.8008438818565401,
+      "grad_norm": 0.400771826505661,
+      "learning_rate": 0.0001983434585433174,
+      "loss": 0.862334668636322,
+      "step": 1898
+    },
+    {
+      "epoch": 0.8016877637130801,
+      "grad_norm": 0.39193347096443176,
+      "learning_rate": 0.00019833474125768676,
+      "loss": 0.8125433325767517,
+      "step": 1900
+    },
+    {
+      "epoch": 0.8016877637130801,
+      "eval_loss": 0.8546335697174072,
+      "eval_runtime": 688.5301,
+      "eval_samples_per_second": 3.06,
+      "eval_steps_per_second": 3.06,
+      "step": 1900
+    },
+    {
+      "epoch": 0.8025316455696202,
+      "grad_norm": 0.441129207611084,
+      "learning_rate": 0.00019832600128811986,
+      "loss": 0.8246012330055237,
+      "step": 1902
+    },
+    {
+      "epoch": 0.8033755274261604,
+      "grad_norm": 0.4089467227458954,
+      "learning_rate": 0.00019831723863663285,
+      "loss": 0.8387641310691833,
+      "step": 1904
+    },
+    {
+      "epoch": 0.8042194092827004,
+      "grad_norm": 0.4871654510498047,
+      "learning_rate": 0.0001983084533052471,
+      "loss": 0.7891429662704468,
+      "step": 1906
+    },
+    {
+      "epoch": 0.8050632911392405,
+      "grad_norm": 0.40085604786872864,
+      "learning_rate": 0.00019829964529598923,
+      "loss": 0.8008774518966675,
+      "step": 1908
+    },
+    {
+      "epoch": 0.8059071729957806,
+      "grad_norm": 0.38775014877319336,
+      "learning_rate": 0.00019829081461089109,
+      "loss": 0.7921834588050842,
+      "step": 1910
+    },
+    {
+      "epoch": 0.8067510548523207,
+      "grad_norm": 0.37830594182014465,
+      "learning_rate": 0.0001982819612519897,
+      "loss": 0.7752519845962524,
+      "step": 1912
+    },
+    {
+      "epoch": 0.8075949367088607,
+      "grad_norm": 0.41137194633483887,
+      "learning_rate": 0.0001982730852213274,
+      "loss": 0.7958255410194397,
+      "step": 1914
+    },
+    {
+      "epoch": 0.8084388185654009,
+      "grad_norm": 0.4871830940246582,
+      "learning_rate": 0.0001982641865209517,
+      "loss": 0.8177281618118286,
+      "step": 1916
+    },
+    {
+      "epoch": 0.809282700421941,
+      "grad_norm": 0.4157388210296631,
+      "learning_rate": 0.0001982552651529154,
+      "loss": 0.7726616859436035,
+      "step": 1918
+    },
+    {
+      "epoch": 0.810126582278481,
+      "grad_norm": 0.32338014245033264,
+      "learning_rate": 0.00019824632111927645,
+      "loss": 0.6601400375366211,
+      "step": 1920
+    },
+    {
+      "epoch": 0.810970464135021,
+      "grad_norm": 0.4642949104309082,
+      "learning_rate": 0.00019823735442209804,
+      "loss": 0.7866622805595398,
+      "step": 1922
+    },
+    {
+      "epoch": 0.8118143459915612,
+      "grad_norm": 0.43558797240257263,
+      "learning_rate": 0.00019822836506344865,
+      "loss": 0.7857986092567444,
+      "step": 1924
+    },
+    {
+      "epoch": 0.8126582278481013,
+      "grad_norm": 0.3996953070163727,
+      "learning_rate": 0.000198219353045402,
+      "loss": 0.8407763838768005,
+      "step": 1926
+    },
+    {
+      "epoch": 0.8135021097046413,
+      "grad_norm": 0.4415414333343506,
+      "learning_rate": 0.00019821031837003686,
+      "loss": 0.8433752655982971,
+      "step": 1928
+    },
+    {
+      "epoch": 0.8143459915611815,
+      "grad_norm": 0.37329310178756714,
+      "learning_rate": 0.00019820126103943752,
+      "loss": 0.6941158175468445,
+      "step": 1930
+    },
+    {
+      "epoch": 0.8151898734177215,
+      "grad_norm": 0.4482601284980774,
+      "learning_rate": 0.00019819218105569323,
+      "loss": 0.775604248046875,
+      "step": 1932
+    },
+    {
+      "epoch": 0.8160337552742616,
+      "grad_norm": 0.426194429397583,
+      "learning_rate": 0.00019818307842089854,
+      "loss": 0.8496418595314026,
+      "step": 1934
+    },
+    {
+      "epoch": 0.8168776371308016,
+      "grad_norm": 0.42379093170166016,
+      "learning_rate": 0.00019817395313715335,
+      "loss": 0.7767958045005798,
+      "step": 1936
+    },
+    {
+      "epoch": 0.8177215189873418,
+      "grad_norm": 0.4149782061576843,
+      "learning_rate": 0.0001981648052065626,
+      "loss": 0.7347666025161743,
+      "step": 1938
+    },
+    {
+      "epoch": 0.8185654008438819,
+      "grad_norm": 0.3927431106567383,
+      "learning_rate": 0.0001981556346312366,
+      "loss": 0.7839647531509399,
+      "step": 1940
+    },
+    {
+      "epoch": 0.8194092827004219,
+      "grad_norm": 0.4483136236667633,
+      "learning_rate": 0.00019814644141329083,
+      "loss": 0.8870531916618347,
+      "step": 1942
+    },
+    {
+      "epoch": 0.8202531645569621,
+      "grad_norm": 0.4400147497653961,
+      "learning_rate": 0.00019813722555484594,
+      "loss": 0.8646620512008667,
+      "step": 1944
+    },
+    {
+      "epoch": 0.8210970464135021,
+      "grad_norm": 0.469971626996994,
+      "learning_rate": 0.00019812798705802785,
+      "loss": 0.9576541781425476,
+      "step": 1946
+    },
+    {
+      "epoch": 0.8219409282700422,
+      "grad_norm": 0.388113409280777,
+      "learning_rate": 0.00019811872592496776,
+      "loss": 0.7956477403640747,
+      "step": 1948
+    },
+    {
+      "epoch": 0.8227848101265823,
+      "grad_norm": 0.4271424114704132,
+      "learning_rate": 0.00019810944215780201,
+      "loss": 0.8440352082252502,
+      "step": 1950
+    },
+    {
+      "epoch": 0.8236286919831224,
+      "grad_norm": 0.4097966253757477,
+      "learning_rate": 0.00019810013575867217,
+      "loss": 0.8589065074920654,
+      "step": 1952
+    },
+    {
+      "epoch": 0.8244725738396624,
+      "grad_norm": 0.41550710797309875,
+      "learning_rate": 0.00019809080672972504,
+      "loss": 0.8606626987457275,
+      "step": 1954
+    },
+    {
+      "epoch": 0.8253164556962025,
+      "grad_norm": 0.3680916130542755,
+      "learning_rate": 0.0001980814550731127,
+      "loss": 0.7582436800003052,
+      "step": 1956
+    },
+    {
+      "epoch": 0.8261603375527427,
+      "grad_norm": 0.573297917842865,
+      "learning_rate": 0.00019807208079099234,
+      "loss": 0.8345186710357666,
+      "step": 1958
+    },
+    {
+      "epoch": 0.8270042194092827,
+      "grad_norm": 0.43631303310394287,
+      "learning_rate": 0.00019806268388552646,
+      "loss": 0.889683723449707,
+      "step": 1960
+    },
+    {
+      "epoch": 0.8278481012658228,
+      "grad_norm": 0.442531019449234,
+      "learning_rate": 0.00019805326435888275,
+      "loss": 0.8092973232269287,
+      "step": 1962
+    },
+    {
+      "epoch": 0.8286919831223629,
+      "grad_norm": 0.45020824670791626,
+      "learning_rate": 0.0001980438222132341,
+      "loss": 0.8919535279273987,
+      "step": 1964
+    },
+    {
+      "epoch": 0.829535864978903,
+      "grad_norm": 0.43131786584854126,
+      "learning_rate": 0.0001980343574507586,
+      "loss": 0.7652381658554077,
+      "step": 1966
+    },
+    {
+      "epoch": 0.830379746835443,
+      "grad_norm": 0.4461326599121094,
+      "learning_rate": 0.00019802487007363967,
+      "loss": 0.8512389659881592,
+      "step": 1968
+    },
+    {
+      "epoch": 0.8312236286919831,
+      "grad_norm": 0.4676894247531891,
+      "learning_rate": 0.0001980153600840658,
+      "loss": 0.8681327700614929,
+      "step": 1970
+    },
+    {
+      "epoch": 0.8320675105485232,
+      "grad_norm": 0.38376888632774353,
+      "learning_rate": 0.00019800582748423075,
+      "loss": 0.7476955652236938,
+      "step": 1972
+    },
+    {
+      "epoch": 0.8329113924050633,
+      "grad_norm": 0.37173229455947876,
+      "learning_rate": 0.00019799627227633354,
+      "loss": 0.814192533493042,
+      "step": 1974
+    },
+    {
+      "epoch": 0.8337552742616033,
+      "grad_norm": 0.4345237910747528,
+      "learning_rate": 0.00019798669446257844,
+      "loss": 0.7668994069099426,
+      "step": 1976
+    },
+    {
+      "epoch": 0.8345991561181435,
+      "grad_norm": 0.45872583985328674,
+      "learning_rate": 0.0001979770940451747,
+      "loss": 0.8321775197982788,
+      "step": 1978
+    },
+    {
+      "epoch": 0.8354430379746836,
+      "grad_norm": 0.40618205070495605,
+      "learning_rate": 0.0001979674710263371,
+      "loss": 0.7824342250823975,
+      "step": 1980
+    },
+    {
+      "epoch": 0.8362869198312236,
+      "grad_norm": 0.38377392292022705,
+      "learning_rate": 0.00019795782540828544,
+      "loss": 0.8752480745315552,
+      "step": 1982
+    },
+    {
+      "epoch": 0.8371308016877637,
+      "grad_norm": 0.36340072751045227,
+      "learning_rate": 0.00019794815719324476,
+      "loss": 0.66700679063797,
+      "step": 1984
+    },
+    {
+      "epoch": 0.8379746835443038,
+      "grad_norm": 0.4638384282588959,
+      "learning_rate": 0.00019793846638344536,
+      "loss": 0.8582209944725037,
+      "step": 1986
+    },
+    {
+      "epoch": 0.8388185654008439,
+      "grad_norm": 0.38364699482917786,
+      "learning_rate": 0.00019792875298112268,
+      "loss": 0.7724968791007996,
+      "step": 1988
+    },
+    {
+      "epoch": 0.8396624472573839,
+      "grad_norm": 0.439730703830719,
+      "learning_rate": 0.00019791901698851749,
+      "loss": 0.8542404174804688,
+      "step": 1990
+    },
+    {
+      "epoch": 0.8405063291139241,
+      "grad_norm": 0.4622722566127777,
+      "learning_rate": 0.00019790925840787563,
+      "loss": 0.8424296975135803,
+      "step": 1992
+    },
+    {
+      "epoch": 0.8413502109704641,
+      "grad_norm": 0.4491989314556122,
+      "learning_rate": 0.0001978994772414482,
+      "loss": 0.8400710225105286,
+      "step": 1994
+    },
+    {
+      "epoch": 0.8421940928270042,
+      "grad_norm": 0.372295081615448,
+      "learning_rate": 0.0001978896734914916,
+      "loss": 0.7265452146530151,
+      "step": 1996
+    },
+    {
+      "epoch": 0.8430379746835444,
+      "grad_norm": 0.4328666925430298,
+      "learning_rate": 0.00019787984716026732,
+      "loss": 0.8458228707313538,
+      "step": 1998
+    },
+    {
+      "epoch": 0.8438818565400844,
+      "grad_norm": 0.42246317863464355,
+      "learning_rate": 0.00019786999825004216,
+      "loss": 0.8517491817474365,
+      "step": 2000
+    },
+    {
+      "epoch": 0.8438818565400844,
+      "eval_loss": 0.8460908532142639,
+      "eval_runtime": 685.2518,
+      "eval_samples_per_second": 3.075,
+      "eval_steps_per_second": 3.075,
+      "step": 2000
+    },
+    {
+      "epoch": 0.8447257383966245,
+      "grad_norm": 0.3927323818206787,
+      "learning_rate": 0.00019786012676308802,
+      "loss": 0.8045481443405151,
+      "step": 2002
+    },
+    {
+      "epoch": 0.8455696202531645,
+      "grad_norm": 0.38189369440078735,
+      "learning_rate": 0.00019785023270168202,
+      "loss": 0.7509838938713074,
+      "step": 2004
+    },
+    {
+      "epoch": 0.8464135021097047,
+      "grad_norm": 0.40383535623550415,
+      "learning_rate": 0.00019784031606810663,
+      "loss": 0.7730163335800171,
+      "step": 2006
+    },
+    {
+      "epoch": 0.8472573839662447,
+      "grad_norm": 0.3665419816970825,
+      "learning_rate": 0.00019783037686464935,
+      "loss": 0.7096828818321228,
+      "step": 2008
+    },
+    {
+      "epoch": 0.8481012658227848,
+      "grad_norm": 0.36623379588127136,
+      "learning_rate": 0.00019782041509360302,
+      "loss": 0.8186401128768921,
+      "step": 2010
+    },
+    {
+      "epoch": 0.8489451476793249,
+      "grad_norm": 0.477255642414093,
+      "learning_rate": 0.0001978104307572656,
+      "loss": 0.8502879738807678,
+      "step": 2012
+    },
+    {
+      "epoch": 0.849789029535865,
+      "grad_norm": 0.44456636905670166,
+      "learning_rate": 0.00019780042385794032,
+      "loss": 0.9010884761810303,
+      "step": 2014
+    },
+    {
+      "epoch": 0.850632911392405,
+      "grad_norm": 0.42378175258636475,
+      "learning_rate": 0.00019779039439793552,
+      "loss": 0.774183452129364,
+      "step": 2016
+    },
+    {
+      "epoch": 0.8514767932489451,
+      "grad_norm": 0.43021997809410095,
+      "learning_rate": 0.00019778034237956482,
+      "loss": 0.8487694263458252,
+      "step": 2018
+    },
+    {
+      "epoch": 0.8523206751054853,
+      "grad_norm": 0.38188812136650085,
+      "learning_rate": 0.00019777026780514704,
+      "loss": 0.8004868030548096,
+      "step": 2020
+    },
+    {
+      "epoch": 0.8531645569620253,
+      "grad_norm": 0.3684130609035492,
+      "learning_rate": 0.0001977601706770062,
+      "loss": 0.7680953145027161,
+      "step": 2022
+    },
+    {
+      "epoch": 0.8540084388185654,
+      "grad_norm": 0.444844514131546,
+      "learning_rate": 0.00019775005099747153,
+      "loss": 0.8990264534950256,
+      "step": 2024
+    },
+    {
+      "epoch": 0.8548523206751055,
+      "grad_norm": 0.3432365655899048,
+      "learning_rate": 0.00019773990876887737,
+      "loss": 0.7676048874855042,
+      "step": 2026
+    },
+    {
+      "epoch": 0.8556962025316456,
+      "grad_norm": 0.3728618025779724,
+      "learning_rate": 0.00019772974399356343,
+      "loss": 0.8546769618988037,
+      "step": 2028
+    },
+    {
+      "epoch": 0.8565400843881856,
+      "grad_norm": 0.42767980694770813,
+      "learning_rate": 0.00019771955667387449,
+      "loss": 0.8009977340698242,
+      "step": 2030
+    },
+    {
+      "epoch": 0.8573839662447258,
+      "grad_norm": 0.4125710129737854,
+      "learning_rate": 0.00019770934681216055,
+      "loss": 0.8424299955368042,
+      "step": 2032
+    },
+    {
+      "epoch": 0.8582278481012658,
+      "grad_norm": 0.4399266839027405,
+      "learning_rate": 0.00019769911441077685,
+      "loss": 0.7940502762794495,
+      "step": 2034
+    },
+    {
+      "epoch": 0.8590717299578059,
+      "grad_norm": 0.41797709465026855,
+      "learning_rate": 0.0001976888594720838,
+      "loss": 0.9424314498901367,
+      "step": 2036
+    },
+    {
+      "epoch": 0.859915611814346,
+      "grad_norm": 0.5245009660720825,
+      "learning_rate": 0.00019767858199844698,
+      "loss": 0.9102896451950073,
+      "step": 2038
+    },
+    {
+      "epoch": 0.8607594936708861,
+      "grad_norm": 0.4092593193054199,
+      "learning_rate": 0.00019766828199223728,
+      "loss": 0.7434720396995544,
+      "step": 2040
+    },
+    {
+      "epoch": 0.8616033755274262,
+      "grad_norm": 0.40787473320961,
+      "learning_rate": 0.00019765795945583068,
+      "loss": 0.7441987991333008,
+      "step": 2042
+    },
+    {
+      "epoch": 0.8624472573839662,
+      "grad_norm": 0.42438021302223206,
+      "learning_rate": 0.0001976476143916084,
+      "loss": 0.8317872881889343,
+      "step": 2044
+    },
+    {
+      "epoch": 0.8632911392405064,
+      "grad_norm": 0.38261914253234863,
+      "learning_rate": 0.00019763724680195682,
+      "loss": 0.764509379863739,
+      "step": 2046
+    },
+    {
+      "epoch": 0.8641350210970464,
+      "grad_norm": 0.43295741081237793,
+      "learning_rate": 0.00019762685668926758,
+      "loss": 0.8666167855262756,
+      "step": 2048
+    },
+    {
+      "epoch": 0.8649789029535865,
+      "grad_norm": 0.36384403705596924,
+      "learning_rate": 0.00019761644405593743,
+      "loss": 0.7606754302978516,
+      "step": 2050
+    },
+    {
+      "epoch": 0.8658227848101265,
+      "grad_norm": 0.4296814501285553,
+      "learning_rate": 0.0001976060089043684,
+      "loss": 0.827495813369751,
+      "step": 2052
+    },
+    {
+      "epoch": 0.8666666666666667,
+      "grad_norm": 0.3831455111503601,
+      "learning_rate": 0.0001975955512369677,
+      "loss": 0.7418577075004578,
+      "step": 2054
+    },
+    {
+      "epoch": 0.8675105485232067,
+      "grad_norm": 0.44710344076156616,
+      "learning_rate": 0.0001975850710561477,
+      "loss": 0.8603045344352722,
+      "step": 2056
+    },
+    {
+      "epoch": 0.8683544303797468,
+      "grad_norm": 0.44012558460235596,
+      "learning_rate": 0.00019757456836432593,
+      "loss": 0.900658905506134,
+      "step": 2058
+    },
+    {
+      "epoch": 0.869198312236287,
+      "grad_norm": 0.42166298627853394,
+      "learning_rate": 0.0001975640431639252,
+      "loss": 0.8416497707366943,
+      "step": 2060
+    },
+    {
+      "epoch": 0.870042194092827,
+      "grad_norm": 0.40824490785598755,
+      "learning_rate": 0.00019755349545737345,
+      "loss": 0.8281066417694092,
+      "step": 2062
+    },
+    {
+      "epoch": 0.8708860759493671,
+      "grad_norm": 0.4272223711013794,
+      "learning_rate": 0.00019754292524710388,
+      "loss": 0.83869469165802,
+      "step": 2064
+    },
+    {
+      "epoch": 0.8717299578059071,
+      "grad_norm": 0.3732798397541046,
+      "learning_rate": 0.00019753233253555478,
+      "loss": 0.8337644934654236,
+      "step": 2066
+    },
+    {
+      "epoch": 0.8725738396624473,
+      "grad_norm": 0.4048069417476654,
+      "learning_rate": 0.00019752171732516973,
+      "loss": 0.7677452564239502,
+      "step": 2068
+    },
+    {
+      "epoch": 0.8734177215189873,
+      "grad_norm": 0.43628430366516113,
+      "learning_rate": 0.00019751107961839742,
+      "loss": 0.8320774435997009,
+      "step": 2070
+    },
+    {
+      "epoch": 0.8742616033755274,
+      "grad_norm": 0.43674105405807495,
+      "learning_rate": 0.00019750041941769174,
+      "loss": 0.8480262160301208,
+      "step": 2072
+    },
+    {
+      "epoch": 0.8751054852320675,
+      "grad_norm": 0.3678584098815918,
+      "learning_rate": 0.0001974897367255119,
+      "loss": 0.7598503828048706,
+      "step": 2074
+    },
+    {
+      "epoch": 0.8759493670886076,
+      "grad_norm": 0.4316411316394806,
+      "learning_rate": 0.0001974790315443221,
+      "loss": 0.7907084822654724,
+      "step": 2076
+    },
+    {
+      "epoch": 0.8767932489451477,
+      "grad_norm": 0.426465779542923,
+      "learning_rate": 0.00019746830387659182,
+      "loss": 0.8287727236747742,
+      "step": 2078
+    },
+    {
+      "epoch": 0.8776371308016878,
+      "grad_norm": 0.447337806224823,
+      "learning_rate": 0.00019745755372479578,
+      "loss": 0.8464577794075012,
+      "step": 2080
+    },
+    {
+      "epoch": 0.8784810126582279,
+      "grad_norm": 0.37877580523490906,
+      "learning_rate": 0.0001974467810914138,
+      "loss": 0.7383086681365967,
+      "step": 2082
+    },
+    {
+      "epoch": 0.8793248945147679,
+      "grad_norm": 0.3914664089679718,
+      "learning_rate": 0.0001974359859789309,
+      "loss": 0.729221522808075,
+      "step": 2084
+    },
+    {
+      "epoch": 0.880168776371308,
+      "grad_norm": 0.39767366647720337,
+      "learning_rate": 0.00019742516838983733,
+      "loss": 0.692332923412323,
+      "step": 2086
+    },
+    {
+      "epoch": 0.8810126582278481,
+      "grad_norm": 0.4396754205226898,
+      "learning_rate": 0.0001974143283266285,
+      "loss": 0.8540068864822388,
+      "step": 2088
+    },
+    {
+      "epoch": 0.8818565400843882,
+      "grad_norm": 0.4968172609806061,
+      "learning_rate": 0.00019740346579180502,
+      "loss": 0.8570998907089233,
+      "step": 2090
+    },
+    {
+      "epoch": 0.8827004219409282,
+      "grad_norm": 0.41686809062957764,
+      "learning_rate": 0.00019739258078787264,
+      "loss": 0.784139096736908,
+      "step": 2092
+    },
+    {
+      "epoch": 0.8835443037974684,
+      "grad_norm": 0.38788047432899475,
+      "learning_rate": 0.0001973816733173423,
+      "loss": 0.8521397113800049,
+      "step": 2094
+    },
+    {
+      "epoch": 0.8843881856540085,
+      "grad_norm": 0.38369178771972656,
+      "learning_rate": 0.00019737074338273023,
+      "loss": 0.8457935452461243,
+      "step": 2096
+    },
+    {
+      "epoch": 0.8852320675105485,
+      "grad_norm": 0.38869398832321167,
+      "learning_rate": 0.0001973597909865577,
+      "loss": 0.7425959706306458,
+      "step": 2098
+    },
+    {
+      "epoch": 0.8860759493670886,
+      "grad_norm": 0.4067917466163635,
+      "learning_rate": 0.00019734881613135122,
+      "loss": 0.7742393016815186,
+      "step": 2100
+    },
+    {
+      "epoch": 0.8860759493670886,
+      "eval_loss": 0.8401098847389221,
+      "eval_runtime": 669.1149,
+      "eval_samples_per_second": 3.149,
+      "eval_steps_per_second": 3.149,
+      "step": 2100
+    },
+    {
+      "epoch": 0.8869198312236287,
+      "grad_norm": 0.4282098412513733,
+      "learning_rate": 0.00019733781881964242,
+      "loss": 0.9099814295768738,
+      "step": 2102
+    },
+    {
+      "epoch": 0.8877637130801688,
+      "grad_norm": 0.33882173895835876,
+      "learning_rate": 0.00019732679905396826,
+      "loss": 0.7458856105804443,
+      "step": 2104
+    },
+    {
+      "epoch": 0.8886075949367088,
+      "grad_norm": 0.43959516286849976,
+      "learning_rate": 0.00019731575683687078,
+      "loss": 0.7829693555831909,
+      "step": 2106
+    },
+    {
+      "epoch": 0.889451476793249,
+      "grad_norm": 0.40046483278274536,
+      "learning_rate": 0.00019730469217089716,
+      "loss": 0.7898997068405151,
+      "step": 2108
+    },
+    {
+      "epoch": 0.890295358649789,
+      "grad_norm": 0.40572553873062134,
+      "learning_rate": 0.00019729360505859984,
+      "loss": 0.7874186635017395,
+      "step": 2110
+    },
+    {
+      "epoch": 0.8911392405063291,
+      "grad_norm": 0.42086347937583923,
+      "learning_rate": 0.0001972824955025364,
+      "loss": 0.8499898910522461,
+      "step": 2112
+    },
+    {
+      "epoch": 0.8919831223628693,
+      "grad_norm": 0.38845229148864746,
+      "learning_rate": 0.00019727136350526956,
+      "loss": 0.7940409779548645,
+      "step": 2114
+    },
+    {
+      "epoch": 0.8928270042194093,
+      "grad_norm": 0.41575267910957336,
+      "learning_rate": 0.00019726020906936728,
+      "loss": 0.8279082775115967,
+      "step": 2116
+    },
+    {
+      "epoch": 0.8936708860759494,
+      "grad_norm": 0.37615057826042175,
+      "learning_rate": 0.0001972490321974027,
+      "loss": 0.8217757940292358,
+      "step": 2118
+    },
+    {
+      "epoch": 0.8945147679324894,
+      "grad_norm": 0.9318385124206543,
+      "learning_rate": 0.00019723783289195415,
+      "loss": 0.7637750506401062,
+      "step": 2120
+    },
+    {
+      "epoch": 0.8953586497890296,
+      "grad_norm": 0.41948962211608887,
+      "learning_rate": 0.000197226611155605,
+      "loss": 0.8702366948127747,
+      "step": 2122
+    },
+    {
+      "epoch": 0.8962025316455696,
+      "grad_norm": 0.4718417525291443,
+      "learning_rate": 0.00019721536699094392,
+      "loss": 0.8146984577178955,
+      "step": 2124
+    },
+    {
+      "epoch": 0.8970464135021097,
+      "grad_norm": 0.4470657408237457,
+      "learning_rate": 0.00019720410040056474,
+      "loss": 0.8646075129508972,
+      "step": 2126
+    },
+    {
+      "epoch": 0.8978902953586498,
+      "grad_norm": 0.4176540970802307,
+      "learning_rate": 0.00019719281138706643,
+      "loss": 0.7925992012023926,
+      "step": 2128
+    },
+    {
+      "epoch": 0.8987341772151899,
+      "grad_norm": 0.4854719042778015,
+      "learning_rate": 0.00019718149995305315,
+      "loss": 0.8142143487930298,
+      "step": 2130
+    },
+    {
+      "epoch": 0.8995780590717299,
+      "grad_norm": 0.4270440340042114,
+      "learning_rate": 0.00019717016610113427,
+      "loss": 0.8050890564918518,
+      "step": 2132
+    },
+    {
+      "epoch": 0.90042194092827,
+      "grad_norm": 0.37710970640182495,
+      "learning_rate": 0.00019715880983392422,
+      "loss": 0.7888354659080505,
+      "step": 2134
+    },
+    {
+      "epoch": 0.9012658227848102,
+      "grad_norm": 0.4181252717971802,
+      "learning_rate": 0.00019714743115404273,
+      "loss": 0.7348225712776184,
+      "step": 2136
+    },
+    {
+      "epoch": 0.9021097046413502,
+      "grad_norm": 0.4475359618663788,
+      "learning_rate": 0.0001971360300641146,
+      "loss": 0.7729623913764954,
+      "step": 2138
+    },
+    {
+      "epoch": 0.9029535864978903,
+      "grad_norm": 0.4052332937717438,
+      "learning_rate": 0.00019712460656676992,
+      "loss": 0.8066384792327881,
+      "step": 2140
+    },
+    {
+      "epoch": 0.9037974683544304,
+      "grad_norm": 0.4567393362522125,
+      "learning_rate": 0.0001971131606646438,
+      "loss": 0.8053154349327087,
+      "step": 2142
+    },
+    {
+      "epoch": 0.9046413502109705,
+      "grad_norm": 0.4304409325122833,
+      "learning_rate": 0.00019710169236037656,
+      "loss": 0.7242560386657715,
+      "step": 2144
+    },
+    {
+      "epoch": 0.9054852320675105,
+      "grad_norm": 0.4785008132457733,
+      "learning_rate": 0.0001970902016566138,
+      "loss": 0.8475301265716553,
+      "step": 2146
+    },
+    {
+      "epoch": 0.9063291139240506,
+      "grad_norm": 0.45226508378982544,
+      "learning_rate": 0.00019707868855600618,
+      "loss": 0.928544282913208,
+      "step": 2148
+    },
+    {
+      "epoch": 0.9071729957805907,
+      "grad_norm": 0.3794468641281128,
+      "learning_rate": 0.00019706715306120956,
+      "loss": 0.8135722875595093,
+      "step": 2150
+    },
+    {
+      "epoch": 0.9080168776371308,
+      "grad_norm": 0.4005141258239746,
+      "learning_rate": 0.0001970555951748849,
+      "loss": 0.7808290719985962,
+      "step": 2152
+    },
+    {
+      "epoch": 0.9088607594936708,
+      "grad_norm": 0.37929919362068176,
+      "learning_rate": 0.00019704401489969842,
+      "loss": 0.8292917013168335,
+      "step": 2154
+    },
+    {
+      "epoch": 0.909704641350211,
+      "grad_norm": 0.4458678066730499,
+      "learning_rate": 0.0001970324122383215,
+      "loss": 0.772884726524353,
+      "step": 2156
+    },
+    {
+      "epoch": 0.9105485232067511,
+      "grad_norm": 0.41907885670661926,
+      "learning_rate": 0.0001970207871934306,
+      "loss": 0.8204470872879028,
+      "step": 2158
+    },
+    {
+      "epoch": 0.9113924050632911,
+      "grad_norm": 0.40610212087631226,
+      "learning_rate": 0.00019700913976770741,
+      "loss": 0.8361451625823975,
+      "step": 2160
+    },
+    {
+      "epoch": 0.9122362869198313,
+      "grad_norm": 0.4213273227214813,
+      "learning_rate": 0.00019699746996383878,
+      "loss": 0.8456075191497803,
+      "step": 2162
+    },
+    {
+      "epoch": 0.9130801687763713,
+      "grad_norm": 0.49264612793922424,
+      "learning_rate": 0.0001969857777845167,
+      "loss": 0.8731603622436523,
+      "step": 2164
+    },
+    {
+      "epoch": 0.9139240506329114,
+      "grad_norm": 0.42002132534980774,
+      "learning_rate": 0.00019697406323243834,
+      "loss": 0.8149666786193848,
+      "step": 2166
+    },
+    {
+      "epoch": 0.9147679324894514,
+      "grad_norm": 0.37501421570777893,
+      "learning_rate": 0.000196962326310306,
+      "loss": 0.7508546710014343,
+      "step": 2168
+    },
+    {
+      "epoch": 0.9156118143459916,
+      "grad_norm": 0.3706159293651581,
+      "learning_rate": 0.00019695056702082717,
+      "loss": 0.830280601978302,
+      "step": 2170
+    },
+    {
+      "epoch": 0.9164556962025316,
+      "grad_norm": 0.3873467445373535,
+      "learning_rate": 0.00019693878536671452,
+      "loss": 0.7297167778015137,
+      "step": 2172
+    },
+    {
+      "epoch": 0.9172995780590717,
+      "grad_norm": 0.42876651883125305,
+      "learning_rate": 0.00019692698135068582,
+      "loss": 0.7924292087554932,
+      "step": 2174
+    },
+    {
+      "epoch": 0.9181434599156119,
+      "grad_norm": 0.4259159564971924,
+      "learning_rate": 0.00019691515497546406,
+      "loss": 0.8098247051239014,
+      "step": 2176
+    },
+    {
+      "epoch": 0.9189873417721519,
+      "grad_norm": 0.38152721524238586,
+      "learning_rate": 0.00019690330624377728,
+      "loss": 0.6789063215255737,
+      "step": 2178
+    },
+    {
+      "epoch": 0.919831223628692,
+      "grad_norm": 0.4135437607765198,
+      "learning_rate": 0.00019689143515835887,
+      "loss": 0.8582620620727539,
+      "step": 2180
+    },
+    {
+      "epoch": 0.920675105485232,
+      "grad_norm": 0.3818798065185547,
+      "learning_rate": 0.0001968795417219472,
+      "loss": 0.8059743642807007,
+      "step": 2182
+    },
+    {
+      "epoch": 0.9215189873417722,
+      "grad_norm": 0.39021286368370056,
+      "learning_rate": 0.00019686762593728582,
+      "loss": 0.7432259917259216,
+      "step": 2184
+    },
+    {
+      "epoch": 0.9223628691983122,
+      "grad_norm": 0.37864068150520325,
+      "learning_rate": 0.00019685568780712355,
+      "loss": 0.7561086416244507,
+      "step": 2186
+    },
+    {
+      "epoch": 0.9232067510548523,
+      "grad_norm": 0.450847864151001,
+      "learning_rate": 0.00019684372733421425,
+      "loss": 0.7552722692489624,
+      "step": 2188
+    },
+    {
+      "epoch": 0.9240506329113924,
+      "grad_norm": 0.37909507751464844,
+      "learning_rate": 0.00019683174452131695,
+      "loss": 0.6905336380004883,
+      "step": 2190
+    },
+    {
+      "epoch": 0.9248945147679325,
+      "grad_norm": 0.4439173638820648,
+      "learning_rate": 0.0001968197393711959,
+      "loss": 0.7710593938827515,
+      "step": 2192
+    },
+    {
+      "epoch": 0.9257383966244725,
+      "grad_norm": 0.3908376395702362,
+      "learning_rate": 0.00019680771188662044,
+      "loss": 0.8212814331054688,
+      "step": 2194
+    },
+    {
+      "epoch": 0.9265822784810127,
+      "grad_norm": 0.3528299033641815,
+      "learning_rate": 0.00019679566207036507,
+      "loss": 0.751355767250061,
+      "step": 2196
+    },
+    {
+      "epoch": 0.9274261603375528,
+      "grad_norm": 0.4351397156715393,
+      "learning_rate": 0.00019678358992520945,
+      "loss": 0.8204057812690735,
+      "step": 2198
+    },
+    {
+      "epoch": 0.9282700421940928,
+      "grad_norm": 0.45566943287849426,
+      "learning_rate": 0.0001967714954539384,
+      "loss": 0.763841450214386,
+      "step": 2200
+    },
+    {
+      "epoch": 0.9282700421940928,
+      "eval_loss": 0.8336610198020935,
+      "eval_runtime": 674.5134,
+      "eval_samples_per_second": 3.124,
+      "eval_steps_per_second": 3.124,
+      "step": 2200
+    },
+    {
+      "epoch": 0.9291139240506329,
+      "grad_norm": 0.4516221284866333,
+      "learning_rate": 0.00019675937865934188,
+      "loss": 0.861711323261261,
+      "step": 2202
+    },
+    {
+      "epoch": 0.929957805907173,
+      "grad_norm": 0.43421855568885803,
+      "learning_rate": 0.00019674723954421502,
+      "loss": 0.8062718510627747,
+      "step": 2204
+    },
+    {
+      "epoch": 0.9308016877637131,
+      "grad_norm": 0.42442476749420166,
+      "learning_rate": 0.00019673507811135803,
+      "loss": 0.8069695234298706,
+      "step": 2206
+    },
+    {
+      "epoch": 0.9316455696202531,
+      "grad_norm": 0.3976045250892639,
+      "learning_rate": 0.00019672289436357637,
+      "loss": 0.7336108684539795,
+      "step": 2208
+    },
+    {
+      "epoch": 0.9324894514767933,
+      "grad_norm": 0.4131270945072174,
+      "learning_rate": 0.00019671068830368058,
+      "loss": 0.7559890151023865,
+      "step": 2210
+    },
+    {
+      "epoch": 0.9333333333333333,
+      "grad_norm": 0.4599294364452362,
+      "learning_rate": 0.00019669845993448634,
+      "loss": 0.8447075486183167,
+      "step": 2212
+    },
+    {
+      "epoch": 0.9341772151898734,
+      "grad_norm": 0.4291204810142517,
+      "learning_rate": 0.00019668620925881455,
+      "loss": 0.8329233527183533,
+      "step": 2214
+    },
+    {
+      "epoch": 0.9350210970464135,
+      "grad_norm": 0.4226539731025696,
+      "learning_rate": 0.00019667393627949116,
+      "loss": 0.749232292175293,
+      "step": 2216
+    },
+    {
+      "epoch": 0.9358649789029536,
+      "grad_norm": 0.4159580171108246,
+      "learning_rate": 0.00019666164099934732,
+      "loss": 0.7470011115074158,
+      "step": 2218
+    },
+    {
+      "epoch": 0.9367088607594937,
+      "grad_norm": 0.4160766303539276,
+      "learning_rate": 0.00019664932342121935,
+      "loss": 0.8094202280044556,
+      "step": 2220
+    },
+    {
+      "epoch": 0.9375527426160337,
+      "grad_norm": 0.45284298062324524,
+      "learning_rate": 0.00019663698354794863,
+      "loss": 0.8398169875144958,
+      "step": 2222
+    },
+    {
+      "epoch": 0.9383966244725739,
+      "grad_norm": 0.3960540294647217,
+      "learning_rate": 0.00019662462138238178,
+      "loss": 0.7652443051338196,
+      "step": 2224
+    },
+    {
+      "epoch": 0.9392405063291139,
+      "grad_norm": 0.4269275367259979,
+      "learning_rate": 0.00019661223692737047,
+      "loss": 0.7909489870071411,
+      "step": 2226
+    },
+    {
+      "epoch": 0.940084388185654,
+      "grad_norm": 0.38187873363494873,
+      "learning_rate": 0.00019659983018577157,
+      "loss": 0.7821935415267944,
+      "step": 2228
+    },
+    {
+      "epoch": 0.9409282700421941,
+      "grad_norm": 0.45898866653442383,
+      "learning_rate": 0.0001965874011604471,
+      "loss": 0.7792721390724182,
+      "step": 2230
+    },
+    {
+      "epoch": 0.9417721518987342,
+      "grad_norm": 0.4740699827671051,
+      "learning_rate": 0.00019657494985426418,
+      "loss": 0.885725200176239,
+      "step": 2232
+    },
+    {
+      "epoch": 0.9426160337552743,
+      "grad_norm": 0.3912813365459442,
+      "learning_rate": 0.00019656247627009506,
+      "loss": 0.7898101806640625,
+      "step": 2234
+    },
+    {
+      "epoch": 0.9434599156118143,
+      "grad_norm": 0.4552418291568756,
+      "learning_rate": 0.0001965499804108172,
+      "loss": 0.8646330833435059,
+      "step": 2236
+    },
+    {
+      "epoch": 0.9443037974683545,
+      "grad_norm": 0.4178548753261566,
+      "learning_rate": 0.0001965374622793131,
+      "loss": 0.8068296313285828,
+      "step": 2238
+    },
+    {
+      "epoch": 0.9451476793248945,
+      "grad_norm": 0.38649076223373413,
+      "learning_rate": 0.00019652492187847057,
+      "loss": 0.7953271865844727,
+      "step": 2240
+    },
+    {
+      "epoch": 0.9459915611814346,
+      "grad_norm": 0.39785075187683105,
+      "learning_rate": 0.0001965123592111823,
+      "loss": 0.8423402309417725,
+      "step": 2242
+    },
+    {
+      "epoch": 0.9468354430379747,
+      "grad_norm": 0.47455263137817383,
+      "learning_rate": 0.00019649977428034632,
+      "loss": 0.8883589506149292,
+      "step": 2244
+    },
+    {
+      "epoch": 0.9476793248945148,
+      "grad_norm": 0.3866841197013855,
+      "learning_rate": 0.0001964871670888657,
+      "loss": 0.7232863903045654,
+      "step": 2246
+    },
+    {
+      "epoch": 0.9485232067510548,
+      "grad_norm": 0.3950609564781189,
+      "learning_rate": 0.00019647453763964876,
+      "loss": 0.7265347838401794,
+      "step": 2248
+    },
+    {
+      "epoch": 0.9493670886075949,
+      "grad_norm": 0.4315829575061798,
+      "learning_rate": 0.0001964618859356088,
+      "loss": 0.8000131249427795,
+      "step": 2250
+    },
+    {
+      "epoch": 0.950210970464135,
+      "grad_norm": 0.403984934091568,
+      "learning_rate": 0.00019644921197966433,
+      "loss": 0.823086678981781,
+      "step": 2252
+    },
+    {
+      "epoch": 0.9510548523206751,
+      "grad_norm": 0.4311080574989319,
+      "learning_rate": 0.00019643651577473897,
+      "loss": 0.8457553386688232,
+      "step": 2254
+    },
+    {
+      "epoch": 0.9518987341772152,
+      "grad_norm": 0.3884856104850769,
+      "learning_rate": 0.00019642379732376158,
+      "loss": 0.7866189479827881,
+      "step": 2256
+    },
+    {
+      "epoch": 0.9527426160337553,
+      "grad_norm": 0.3384067714214325,
+      "learning_rate": 0.00019641105662966595,
+      "loss": 0.7095813751220703,
+      "step": 2258
+    },
+    {
+      "epoch": 0.9535864978902954,
+      "grad_norm": 0.40426746010780334,
+      "learning_rate": 0.00019639829369539117,
+      "loss": 0.8271632194519043,
+      "step": 2260
+    },
+    {
+      "epoch": 0.9544303797468354,
+      "grad_norm": 0.45495131611824036,
+      "learning_rate": 0.00019638550852388144,
+      "loss": 0.8204861283302307,
+      "step": 2262
+    },
+    {
+      "epoch": 0.9552742616033755,
+      "grad_norm": 0.47201067209243774,
+      "learning_rate": 0.00019637270111808598,
+      "loss": 0.884763777256012,
+      "step": 2264
+    },
+    {
+      "epoch": 0.9561181434599156,
+      "grad_norm": 0.3959396779537201,
+      "learning_rate": 0.0001963598714809592,
+      "loss": 0.7630926370620728,
+      "step": 2266
+    },
+    {
+      "epoch": 0.9569620253164557,
+      "grad_norm": 0.4197136461734772,
+      "learning_rate": 0.00019634701961546077,
+      "loss": 0.797870397567749,
+      "step": 2268
+    },
+    {
+      "epoch": 0.9578059071729957,
+      "grad_norm": 0.381290078163147,
+      "learning_rate": 0.00019633414552455526,
+      "loss": 0.7755846977233887,
+      "step": 2270
+    },
+    {
+      "epoch": 0.9586497890295359,
+      "grad_norm": 0.3868124783039093,
+      "learning_rate": 0.00019632124921121255,
+      "loss": 0.7553264498710632,
+      "step": 2272
+    },
+    {
+      "epoch": 0.959493670886076,
+      "grad_norm": 0.4350631833076477,
+      "learning_rate": 0.00019630833067840747,
+      "loss": 0.7792160511016846,
+      "step": 2274
+    },
+    {
+      "epoch": 0.960337552742616,
+      "grad_norm": 0.4717157781124115,
+      "learning_rate": 0.00019629538992912016,
+      "loss": 0.878220796585083,
+      "step": 2276
+    },
+    {
+      "epoch": 0.9611814345991562,
+      "grad_norm": 0.45545122027397156,
+      "learning_rate": 0.00019628242696633584,
+      "loss": 0.8891014456748962,
+      "step": 2278
+    },
+    {
+      "epoch": 0.9620253164556962,
+      "grad_norm": 0.4421025514602661,
+      "learning_rate": 0.00019626944179304467,
+      "loss": 0.8024723529815674,
+      "step": 2280
+    },
+    {
+      "epoch": 0.9628691983122363,
+      "grad_norm": 0.40385812520980835,
+      "learning_rate": 0.00019625643441224222,
+      "loss": 0.8036280274391174,
+      "step": 2282
+    },
+    {
+      "epoch": 0.9637130801687763,
+      "grad_norm": 0.3866552412509918,
+      "learning_rate": 0.00019624340482692898,
+      "loss": 0.8072395324707031,
+      "step": 2284
+    },
+    {
+      "epoch": 0.9645569620253165,
+      "grad_norm": 0.4113233685493469,
+      "learning_rate": 0.0001962303530401106,
+      "loss": 0.8046114444732666,
+      "step": 2286
+    },
+    {
+      "epoch": 0.9654008438818565,
+      "grad_norm": 0.40720227360725403,
+      "learning_rate": 0.00019621727905479798,
+      "loss": 0.8434037566184998,
+      "step": 2288
+    },
+    {
+      "epoch": 0.9662447257383966,
+      "grad_norm": 0.35759273171424866,
+      "learning_rate": 0.00019620418287400693,
+      "loss": 0.7832419276237488,
+      "step": 2290
+    },
+    {
+      "epoch": 0.9670886075949368,
+      "grad_norm": 0.4170442223548889,
+      "learning_rate": 0.0001961910645007585,
+      "loss": 0.7975693345069885,
+      "step": 2292
+    },
+    {
+      "epoch": 0.9679324894514768,
+      "grad_norm": 0.4126700162887573,
+      "learning_rate": 0.00019617792393807895,
+      "loss": 0.8554494976997375,
+      "step": 2294
+    },
+    {
+      "epoch": 0.9687763713080169,
+      "grad_norm": 0.4231695830821991,
+      "learning_rate": 0.00019616476118899942,
+      "loss": 0.824539065361023,
+      "step": 2296
+    },
+    {
+      "epoch": 0.9696202531645569,
+      "grad_norm": 0.3556209206581116,
+      "learning_rate": 0.0001961515762565564,
+      "loss": 0.7008040547370911,
+      "step": 2298
+    },
+    {
+      "epoch": 0.9704641350210971,
+      "grad_norm": 0.4370400309562683,
+      "learning_rate": 0.00019613836914379132,
+      "loss": 0.8192558288574219,
+      "step": 2300
+    },
+    {
+      "epoch": 0.9704641350210971,
+      "eval_loss": 0.8281980156898499,
+      "eval_runtime": 688.6136,
+      "eval_samples_per_second": 3.06,
+      "eval_steps_per_second": 3.06,
+      "step": 2300
+    },
+    {
+      "epoch": 0.9713080168776371,
+      "grad_norm": 0.4207614064216614,
+      "learning_rate": 0.00019612513985375088,
+      "loss": 0.878928005695343,
+      "step": 2302
+    },
+    {
+      "epoch": 0.9721518987341772,
+      "grad_norm": 0.37607815861701965,
+      "learning_rate": 0.0001961118883894868,
+      "loss": 0.736182689666748,
+      "step": 2304
+    },
+    {
+      "epoch": 0.9729957805907173,
+      "grad_norm": 0.43311476707458496,
+      "learning_rate": 0.00019609861475405592,
+      "loss": 0.8403618335723877,
+      "step": 2306
+    },
+    {
+      "epoch": 0.9738396624472574,
+      "grad_norm": 0.4533176124095917,
+      "learning_rate": 0.00019608531895052021,
+      "loss": 0.8414564728736877,
+      "step": 2308
+    },
+    {
+      "epoch": 0.9746835443037974,
+      "grad_norm": 0.4025697410106659,
+      "learning_rate": 0.00019607200098194677,
+      "loss": 0.8105775117874146,
+      "step": 2310
+    },
+    {
+      "epoch": 0.9755274261603376,
+      "grad_norm": 0.37894198298454285,
+      "learning_rate": 0.00019605866085140784,
+      "loss": 0.8053821325302124,
+      "step": 2312
+    },
+    {
+      "epoch": 0.9763713080168777,
+      "grad_norm": 0.46265891194343567,
+      "learning_rate": 0.0001960452985619807,
+      "loss": 0.8092884421348572,
+      "step": 2314
+    },
+    {
+      "epoch": 0.9772151898734177,
+      "grad_norm": 0.3770054280757904,
+      "learning_rate": 0.0001960319141167477,
+      "loss": 0.7456095218658447,
+      "step": 2316
+    },
+    {
+      "epoch": 0.9780590717299578,
+      "grad_norm": 0.44084450602531433,
+      "learning_rate": 0.0001960185075187965,
+      "loss": 0.7739883661270142,
+      "step": 2318
+    },
+    {
+      "epoch": 0.9789029535864979,
+      "grad_norm": 0.3880341351032257,
+      "learning_rate": 0.00019600507877121965,
+      "loss": 0.6961302757263184,
+      "step": 2320
+    },
+    {
+      "epoch": 0.979746835443038,
+      "grad_norm": 0.468037486076355,
+      "learning_rate": 0.00019599162787711495,
+      "loss": 0.8447984457015991,
+      "step": 2322
+    },
+    {
+      "epoch": 0.980590717299578,
+      "grad_norm": 0.4123631417751312,
+      "learning_rate": 0.00019597815483958527,
+      "loss": 0.8155615329742432,
+      "step": 2324
+    },
+    {
+      "epoch": 0.9814345991561182,
+      "grad_norm": 0.3827875554561615,
+      "learning_rate": 0.00019596465966173853,
+      "loss": 0.7990952730178833,
+      "step": 2326
+    },
+    {
+      "epoch": 0.9822784810126582,
+      "grad_norm": 0.4644222557544708,
+      "learning_rate": 0.00019595114234668788,
+      "loss": 0.8899750709533691,
+      "step": 2328
+    },
+    {
+      "epoch": 0.9831223628691983,
+      "grad_norm": 0.43699249625205994,
+      "learning_rate": 0.00019593760289755144,
+      "loss": 0.7863160371780396,
+      "step": 2330
+    },
+    {
+      "epoch": 0.9839662447257383,
+      "grad_norm": 0.40713244676589966,
+      "learning_rate": 0.00019592404131745254,
+      "loss": 0.7960637807846069,
+      "step": 2332
+    },
+    {
+      "epoch": 0.9848101265822785,
+      "grad_norm": 0.37364068627357483,
+      "learning_rate": 0.00019591045760951957,
+      "loss": 0.714144229888916,
+      "step": 2334
+    },
+    {
+      "epoch": 0.9856540084388186,
+      "grad_norm": 0.4094736576080322,
+      "learning_rate": 0.00019589685177688599,
+      "loss": 0.7311791777610779,
+      "step": 2336
+    },
+    {
+      "epoch": 0.9864978902953586,
+      "grad_norm": 0.4275180697441101,
+      "learning_rate": 0.0001958832238226905,
+      "loss": 0.8191342353820801,
+      "step": 2338
+    },
+    {
+      "epoch": 0.9873417721518988,
+      "grad_norm": 0.4627651274204254,
+      "learning_rate": 0.00019586957375007668,
+      "loss": 0.7476106286048889,
+      "step": 2340
+    },
+    {
+      "epoch": 0.9881856540084388,
+      "grad_norm": 0.44676414132118225,
+      "learning_rate": 0.00019585590156219346,
+      "loss": 0.8257967233657837,
+      "step": 2342
+    },
+    {
+      "epoch": 0.9890295358649789,
+      "grad_norm": 0.40667134523391724,
+      "learning_rate": 0.00019584220726219466,
+      "loss": 0.7922079563140869,
+      "step": 2344
+    },
+    {
+      "epoch": 0.9898734177215189,
+      "grad_norm": 0.40865346789360046,
+      "learning_rate": 0.00019582849085323934,
+      "loss": 0.7892434597015381,
+      "step": 2346
+    },
+    {
+      "epoch": 0.9907172995780591,
+      "grad_norm": 0.43309709429740906,
+      "learning_rate": 0.00019581475233849163,
+      "loss": 0.8766644597053528,
+      "step": 2348
+    },
+    {
+      "epoch": 0.9915611814345991,
+      "grad_norm": 0.41558781266212463,
+      "learning_rate": 0.00019580099172112069,
+      "loss": 0.9183312058448792,
+      "step": 2350
+    },
+    {
+      "epoch": 0.9924050632911392,
+      "grad_norm": 0.432748019695282,
+      "learning_rate": 0.00019578720900430082,
+      "loss": 0.7905668020248413,
+      "step": 2352
+    },
+    {
+      "epoch": 0.9932489451476794,
+      "grad_norm": 0.4116985499858856,
+      "learning_rate": 0.0001957734041912115,
+      "loss": 0.8279712200164795,
+      "step": 2354
+    },
+    {
+      "epoch": 0.9940928270042194,
+      "grad_norm": 0.3927275240421295,
+      "learning_rate": 0.00019575957728503718,
+      "loss": 0.7451362609863281,
+      "step": 2356
+    },
+    {
+      "epoch": 0.9949367088607595,
+      "grad_norm": 0.3389509320259094,
+      "learning_rate": 0.0001957457282889675,
+      "loss": 0.6769390106201172,
+      "step": 2358
+    },
+    {
+      "epoch": 0.9957805907172996,
+      "grad_norm": 0.4220774471759796,
+      "learning_rate": 0.00019573185720619712,
+      "loss": 0.7590897083282471,
+      "step": 2360
+    },
+    {
+      "epoch": 0.9966244725738397,
+      "grad_norm": 0.4299897849559784,
+      "learning_rate": 0.00019571796403992583,
+      "loss": 0.8301284313201904,
+      "step": 2362
+    },
+    {
+      "epoch": 0.9974683544303797,
+      "grad_norm": 0.4031789302825928,
+      "learning_rate": 0.00019570404879335856,
+      "loss": 0.8019471764564514,
+      "step": 2364
+    },
+    {
+      "epoch": 0.9983122362869198,
+      "grad_norm": 0.3957567811012268,
+      "learning_rate": 0.00019569011146970526,
+      "loss": 0.7427027821540833,
+      "step": 2366
+    },
+    {
+      "epoch": 0.99915611814346,
+      "grad_norm": 0.41741347312927246,
+      "learning_rate": 0.000195676152072181,
+      "loss": 0.8221617341041565,
+      "step": 2368
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.47121551632881165,
+      "learning_rate": 0.00019566217060400595,
+      "loss": 0.8143455982208252,
+      "step": 2370
+    },
+    {
+      "epoch": 1.0008438818565402,
+      "grad_norm": 0.3854716122150421,
+      "learning_rate": 0.0001956481670684054,
+      "loss": 0.6971694231033325,
+      "step": 2372
+    },
+    {
+      "epoch": 1.00168776371308,
+      "grad_norm": 0.4029138386249542,
+      "learning_rate": 0.00019563414146860964,
+      "loss": 0.6719155311584473,
+      "step": 2374
+    },
+    {
+      "epoch": 1.0025316455696203,
+      "grad_norm": 0.40136656165122986,
+      "learning_rate": 0.00019562009380785412,
+      "loss": 0.7161872386932373,
+      "step": 2376
+    },
+    {
+      "epoch": 1.0033755274261604,
+      "grad_norm": 0.40685850381851196,
+      "learning_rate": 0.0001956060240893794,
+      "loss": 0.71699059009552,
+      "step": 2378
+    },
+    {
+      "epoch": 1.0042194092827004,
+      "grad_norm": 0.46153366565704346,
+      "learning_rate": 0.00019559193231643104,
+      "loss": 0.8030059337615967,
+      "step": 2380
+    },
+    {
+      "epoch": 1.0050632911392405,
+      "grad_norm": 0.45048823952674866,
+      "learning_rate": 0.0001955778184922598,
+      "loss": 0.7666479349136353,
+      "step": 2382
+    },
+    {
+      "epoch": 1.0059071729957807,
+      "grad_norm": 0.4260413348674774,
+      "learning_rate": 0.00019556368262012142,
+      "loss": 0.7481234073638916,
+      "step": 2384
+    },
+    {
+      "epoch": 1.0067510548523206,
+      "grad_norm": 0.4819575548171997,
+      "learning_rate": 0.0001955495247032768,
+      "loss": 0.7943820953369141,
+      "step": 2386
+    },
+    {
+      "epoch": 1.0075949367088608,
+      "grad_norm": 0.45789605379104614,
+      "learning_rate": 0.0001955353447449919,
+      "loss": 0.6401634812355042,
+      "step": 2388
+    },
+    {
+      "epoch": 1.0084388185654007,
+      "grad_norm": 0.37019097805023193,
+      "learning_rate": 0.00019552114274853778,
+      "loss": 0.7058886885643005,
+      "step": 2390
+    },
+    {
+      "epoch": 1.009282700421941,
+      "grad_norm": 0.44416263699531555,
+      "learning_rate": 0.0001955069187171905,
+      "loss": 0.7631242871284485,
+      "step": 2392
+    },
+    {
+      "epoch": 1.010126582278481,
+      "grad_norm": 0.4478379786014557,
+      "learning_rate": 0.00019549267265423138,
+      "loss": 0.7873892784118652,
+      "step": 2394
+    },
+    {
+      "epoch": 1.010970464135021,
+      "grad_norm": 0.4596130847930908,
+      "learning_rate": 0.00019547840456294658,
+      "loss": 0.7882059216499329,
+      "step": 2396
+    },
+    {
+      "epoch": 1.0118143459915612,
+      "grad_norm": 0.4062999188899994,
+      "learning_rate": 0.0001954641144466276,
+      "loss": 0.6358574628829956,
+      "step": 2398
+    },
+    {
+      "epoch": 1.0126582278481013,
+      "grad_norm": 0.47627049684524536,
+      "learning_rate": 0.0001954498023085708,
+      "loss": 0.7803886532783508,
+      "step": 2400
+    },
+    {
+      "epoch": 1.0126582278481013,
+      "eval_loss": 0.8250564932823181,
+      "eval_runtime": 691.5833,
+      "eval_samples_per_second": 3.047,
+      "eval_steps_per_second": 3.047,
+      "step": 2400
+    },
+    {
+      "epoch": 1.0135021097046413,
+      "grad_norm": 0.40690696239471436,
+      "learning_rate": 0.00019543546815207777,
+      "loss": 0.7394700050354004,
+      "step": 2402
+    },
+    {
+      "epoch": 1.0143459915611814,
+      "grad_norm": 0.45543980598449707,
+      "learning_rate": 0.0001954211119804551,
+      "loss": 0.7096555829048157,
+      "step": 2404
+    },
+    {
+      "epoch": 1.0151898734177216,
+      "grad_norm": 0.4367314279079437,
+      "learning_rate": 0.00019540673379701454,
+      "loss": 0.7961843013763428,
+      "step": 2406
+    },
+    {
+      "epoch": 1.0160337552742615,
+      "grad_norm": 0.4458340108394623,
+      "learning_rate": 0.00019539233360507278,
+      "loss": 0.743843138217926,
+      "step": 2408
+    },
+    {
+      "epoch": 1.0168776371308017,
+      "grad_norm": 0.3741893470287323,
+      "learning_rate": 0.0001953779114079517,
+      "loss": 0.7273756265640259,
+      "step": 2410
+    },
+    {
+      "epoch": 1.0177215189873419,
+      "grad_norm": 0.4340079426765442,
+      "learning_rate": 0.00019536346720897823,
+      "loss": 0.7922506928443909,
+      "step": 2412
+    },
+    {
+      "epoch": 1.0185654008438818,
+      "grad_norm": 0.4518018662929535,
+      "learning_rate": 0.00019534900101148437,
+      "loss": 0.6492037773132324,
+      "step": 2414
+    },
+    {
+      "epoch": 1.019409282700422,
+      "grad_norm": 0.4160727560520172,
+      "learning_rate": 0.00019533451281880716,
+      "loss": 0.7422950863838196,
+      "step": 2416
+    },
+    {
+      "epoch": 1.0202531645569621,
+      "grad_norm": 0.35953211784362793,
+      "learning_rate": 0.00019532000263428883,
+      "loss": 0.6489207148551941,
+      "step": 2418
+    },
+    {
+      "epoch": 1.021097046413502,
+      "grad_norm": 0.38774433732032776,
+      "learning_rate": 0.0001953054704612765,
+      "loss": 0.7449417114257812,
+      "step": 2420
+    },
+    {
+      "epoch": 1.0219409282700422,
+      "grad_norm": 0.48185858130455017,
+      "learning_rate": 0.0001952909163031225,
+      "loss": 0.8481494188308716,
+      "step": 2422
+    },
+    {
+      "epoch": 1.0227848101265822,
+      "grad_norm": 0.4098491072654724,
+      "learning_rate": 0.00019527634016318425,
+      "loss": 0.7442291378974915,
+      "step": 2424
+    },
+    {
+      "epoch": 1.0236286919831223,
+      "grad_norm": 0.41284480690956116,
+      "learning_rate": 0.00019526174204482412,
+      "loss": 0.7564480900764465,
+      "step": 2426
+    },
+    {
+      "epoch": 1.0244725738396625,
+      "grad_norm": 0.44948479533195496,
+      "learning_rate": 0.00019524712195140965,
+      "loss": 0.7234854102134705,
+      "step": 2428
+    },
+    {
+      "epoch": 1.0253164556962024,
+      "grad_norm": 0.39939233660697937,
+      "learning_rate": 0.0001952324798863134,
+      "loss": 0.7638729810714722,
+      "step": 2430
+    },
+    {
+      "epoch": 1.0261603375527426,
+      "grad_norm": 0.40807482600212097,
+      "learning_rate": 0.00019521781585291298,
+      "loss": 0.6922630071640015,
+      "step": 2432
+    },
+    {
+      "epoch": 1.0270042194092828,
+      "grad_norm": 0.45735791325569153,
+      "learning_rate": 0.00019520312985459116,
+      "loss": 0.762996256351471,
+      "step": 2434
+    },
+    {
+      "epoch": 1.0278481012658227,
+      "grad_norm": 0.4280409812927246,
+      "learning_rate": 0.0001951884218947357,
+      "loss": 0.7811596989631653,
+      "step": 2436
+    },
+    {
+      "epoch": 1.0286919831223629,
+      "grad_norm": 0.41957560181617737,
+      "learning_rate": 0.00019517369197673943,
+      "loss": 0.7278305292129517,
+      "step": 2438
+    },
+    {
+      "epoch": 1.029535864978903,
+      "grad_norm": 0.4832586646080017,
+      "learning_rate": 0.00019515894010400028,
+      "loss": 0.807456910610199,
+      "step": 2440
+    },
+    {
+      "epoch": 1.030379746835443,
+      "grad_norm": 0.46448788046836853,
+      "learning_rate": 0.00019514416627992122,
+      "loss": 0.7421302795410156,
+      "step": 2442
+    },
+    {
+      "epoch": 1.0312236286919831,
+      "grad_norm": 0.41362619400024414,
+      "learning_rate": 0.00019512937050791024,
+      "loss": 0.6892213225364685,
+      "step": 2444
+    },
+    {
+      "epoch": 1.0320675105485233,
+      "grad_norm": 0.4158109128475189,
+      "learning_rate": 0.0001951145527913805,
+      "loss": 0.7305331826210022,
+      "step": 2446
+    },
+    {
+      "epoch": 1.0329113924050632,
+      "grad_norm": 0.4725087881088257,
+      "learning_rate": 0.00019509971313375011,
+      "loss": 0.7682753205299377,
+      "step": 2448
+    },
+    {
+      "epoch": 1.0337552742616034,
+      "grad_norm": 0.4341438114643097,
+      "learning_rate": 0.00019508485153844237,
+      "loss": 0.747041642665863,
+      "step": 2450
+    },
+    {
+      "epoch": 1.0345991561181433,
+      "grad_norm": 0.44515377283096313,
+      "learning_rate": 0.00019506996800888548,
+      "loss": 0.7312062382698059,
+      "step": 2452
+    },
+    {
+      "epoch": 1.0354430379746835,
+      "grad_norm": 0.5313915014266968,
+      "learning_rate": 0.00019505506254851283,
+      "loss": 0.7861107587814331,
+      "step": 2454
+    },
+    {
+      "epoch": 1.0362869198312237,
+      "grad_norm": 0.6094246506690979,
+      "learning_rate": 0.00019504013516076285,
+      "loss": 0.8458115458488464,
+      "step": 2456
+    },
+    {
+      "epoch": 1.0371308016877636,
+      "grad_norm": 0.36419153213500977,
+      "learning_rate": 0.00019502518584907895,
+      "loss": 0.6091256141662598,
+      "step": 2458
+    },
+    {
+      "epoch": 1.0379746835443038,
+      "grad_norm": 0.399271696805954,
+      "learning_rate": 0.00019501021461690966,
+      "loss": 0.6703370809555054,
+      "step": 2460
+    },
+    {
+      "epoch": 1.038818565400844,
+      "grad_norm": 0.38560932874679565,
+      "learning_rate": 0.00019499522146770856,
+      "loss": 0.7220943570137024,
+      "step": 2462
+    },
+    {
+      "epoch": 1.0396624472573839,
+      "grad_norm": 0.44199803471565247,
+      "learning_rate": 0.00019498020640493428,
+      "loss": 0.7501406073570251,
+      "step": 2464
+    },
+    {
+      "epoch": 1.040506329113924,
+      "grad_norm": 0.47394421696662903,
+      "learning_rate": 0.00019496516943205054,
+      "loss": 0.7957925796508789,
+      "step": 2466
+    },
+    {
+      "epoch": 1.0413502109704642,
+      "grad_norm": 0.5274425745010376,
+      "learning_rate": 0.00019495011055252603,
+      "loss": 0.7964824438095093,
+      "step": 2468
+    },
+    {
+      "epoch": 1.0421940928270041,
+      "grad_norm": 0.44004881381988525,
+      "learning_rate": 0.0001949350297698346,
+      "loss": 0.7478582262992859,
+      "step": 2470
+    },
+    {
+      "epoch": 1.0430379746835443,
+      "grad_norm": 0.40662670135498047,
+      "learning_rate": 0.000194919927087455,
+      "loss": 0.738339900970459,
+      "step": 2472
+    },
+    {
+      "epoch": 1.0438818565400845,
+      "grad_norm": 0.43934181332588196,
+      "learning_rate": 0.00019490480250887123,
+      "loss": 0.817171037197113,
+      "step": 2474
+    },
+    {
+      "epoch": 1.0447257383966244,
+      "grad_norm": 0.4111221432685852,
+      "learning_rate": 0.00019488965603757223,
+      "loss": 0.7413169145584106,
+      "step": 2476
+    },
+    {
+      "epoch": 1.0455696202531646,
+      "grad_norm": 0.3770996630191803,
+      "learning_rate": 0.00019487448767705194,
+      "loss": 0.7893102169036865,
+      "step": 2478
+    },
+    {
+      "epoch": 1.0464135021097047,
+      "grad_norm": 0.4524461627006531,
+      "learning_rate": 0.00019485929743080945,
+      "loss": 0.7037841081619263,
+      "step": 2480
+    },
+    {
+      "epoch": 1.0472573839662447,
+      "grad_norm": 0.4232378900051117,
+      "learning_rate": 0.00019484408530234885,
+      "loss": 0.7456525564193726,
+      "step": 2482
+    },
+    {
+      "epoch": 1.0481012658227848,
+      "grad_norm": 0.3842810392379761,
+      "learning_rate": 0.00019482885129517929,
+      "loss": 0.7263808250427246,
+      "step": 2484
+    },
+    {
+      "epoch": 1.048945147679325,
+      "grad_norm": 0.43203726410865784,
+      "learning_rate": 0.00019481359541281495,
+      "loss": 0.7101474404335022,
+      "step": 2486
+    },
+    {
+      "epoch": 1.049789029535865,
+      "grad_norm": 0.46944454312324524,
+      "learning_rate": 0.00019479831765877509,
+      "loss": 0.8561921715736389,
+      "step": 2488
+    },
+    {
+      "epoch": 1.0506329113924051,
+      "grad_norm": 0.4455019533634186,
+      "learning_rate": 0.00019478301803658397,
+      "loss": 0.7206679582595825,
+      "step": 2490
+    },
+    {
+      "epoch": 1.051476793248945,
+      "grad_norm": 0.4253813326358795,
+      "learning_rate": 0.00019476769654977095,
+      "loss": 0.7271745800971985,
+      "step": 2492
+    },
+    {
+      "epoch": 1.0523206751054852,
+      "grad_norm": 0.42758429050445557,
+      "learning_rate": 0.00019475235320187033,
+      "loss": 0.7358576059341431,
+      "step": 2494
+    },
+    {
+      "epoch": 1.0531645569620254,
+      "grad_norm": 0.400634765625,
+      "learning_rate": 0.00019473698799642164,
+      "loss": 0.6617647409439087,
+      "step": 2496
+    },
+    {
+      "epoch": 1.0540084388185653,
+      "grad_norm": 0.4132104814052582,
+      "learning_rate": 0.00019472160093696926,
+      "loss": 0.6900970935821533,
+      "step": 2498
+    },
+    {
+      "epoch": 1.0548523206751055,
+      "grad_norm": 0.5116495490074158,
+      "learning_rate": 0.00019470619202706268,
+      "loss": 0.7352861762046814,
+      "step": 2500
+    },
+    {
+      "epoch": 1.0548523206751055,
+      "eval_loss": 0.8249453902244568,
+      "eval_runtime": 679.4446,
+      "eval_samples_per_second": 3.101,
+      "eval_steps_per_second": 3.101,
+      "step": 2500
+    },
+    {
+      "epoch": 1.0556962025316456,
+      "grad_norm": 0.42822444438934326,
+      "learning_rate": 0.0001946907612702565,
+      "loss": 0.7583479881286621,
+      "step": 2502
+    },
+    {
+      "epoch": 1.0565400843881856,
+      "grad_norm": 0.4788980782032013,
+      "learning_rate": 0.00019467530867011025,
+      "loss": 0.777511477470398,
+      "step": 2504
+    },
+    {
+      "epoch": 1.0573839662447257,
+      "grad_norm": 0.4210498034954071,
+      "learning_rate": 0.00019465983423018856,
+      "loss": 0.7452852129936218,
+      "step": 2506
+    },
+    {
+      "epoch": 1.058227848101266,
+      "grad_norm": 0.43219906091690063,
+      "learning_rate": 0.0001946443379540611,
+      "loss": 0.7540958523750305,
+      "step": 2508
+    },
+    {
+      "epoch": 1.0590717299578059,
+      "grad_norm": 0.4361002445220947,
+      "learning_rate": 0.00019462881984530253,
+      "loss": 0.7175624966621399,
+      "step": 2510
+    },
+    {
+      "epoch": 1.059915611814346,
+      "grad_norm": 0.4575091600418091,
+      "learning_rate": 0.00019461327990749264,
+      "loss": 0.7342076301574707,
+      "step": 2512
+    },
+    {
+      "epoch": 1.0607594936708862,
+      "grad_norm": 0.4823324978351593,
+      "learning_rate": 0.00019459771814421615,
+      "loss": 0.8255488872528076,
+      "step": 2514
+    },
+    {
+      "epoch": 1.0616033755274261,
+      "grad_norm": 0.41693785786628723,
+      "learning_rate": 0.00019458213455906284,
+      "loss": 0.6624729633331299,
+      "step": 2516
+    },
+    {
+      "epoch": 1.0624472573839663,
+      "grad_norm": 0.41319167613983154,
+      "learning_rate": 0.00019456652915562762,
+      "loss": 0.8224266767501831,
+      "step": 2518
+    },
+    {
+      "epoch": 1.0632911392405062,
+      "grad_norm": 0.4405190050601959,
+      "learning_rate": 0.00019455090193751023,
+      "loss": 0.7582900524139404,
+      "step": 2520
+    },
+    {
+      "epoch": 1.0641350210970464,
+      "grad_norm": 0.4564387798309326,
+      "learning_rate": 0.0001945352529083157,
+      "loss": 0.7948101162910461,
+      "step": 2522
+    },
+    {
+      "epoch": 1.0649789029535865,
+      "grad_norm": 0.4247225224971771,
+      "learning_rate": 0.00019451958207165394,
+      "loss": 0.7670508027076721,
+      "step": 2524
+    },
+    {
+      "epoch": 1.0658227848101265,
+      "grad_norm": 0.42070862650871277,
+      "learning_rate": 0.00019450388943113983,
+      "loss": 0.7753698229789734,
+      "step": 2526
+    },
+    {
+      "epoch": 1.0666666666666667,
+      "grad_norm": 0.3729179799556732,
+      "learning_rate": 0.00019448817499039342,
+      "loss": 0.710904598236084,
+      "step": 2528
+    },
+    {
+      "epoch": 1.0675105485232068,
+      "grad_norm": 0.40901753306388855,
+      "learning_rate": 0.0001944724387530397,
+      "loss": 0.7000213265419006,
+      "step": 2530
+    },
+    {
+      "epoch": 1.0683544303797468,
+      "grad_norm": 0.4481421113014221,
+      "learning_rate": 0.00019445668072270877,
+      "loss": 0.7374410629272461,
+      "step": 2532
+    },
+    {
+      "epoch": 1.069198312236287,
+      "grad_norm": 0.4839915335178375,
+      "learning_rate": 0.00019444090090303567,
+      "loss": 0.746309220790863,
+      "step": 2534
+    },
+    {
+      "epoch": 1.070042194092827,
+      "grad_norm": 0.4982132613658905,
+      "learning_rate": 0.00019442509929766048,
+      "loss": 0.7448083162307739,
+      "step": 2536
+    },
+    {
+      "epoch": 1.070886075949367,
+      "grad_norm": 0.3987475037574768,
+      "learning_rate": 0.00019440927591022838,
+      "loss": 0.7052555680274963,
+      "step": 2538
+    },
+    {
+      "epoch": 1.0717299578059072,
+      "grad_norm": 0.4089896082878113,
+      "learning_rate": 0.00019439343074438944,
+      "loss": 0.7090281248092651,
+      "step": 2540
+    },
+    {
+      "epoch": 1.0725738396624473,
+      "grad_norm": 0.3703136146068573,
+      "learning_rate": 0.00019437756380379895,
+      "loss": 0.7530601024627686,
+      "step": 2542
+    },
+    {
+      "epoch": 1.0734177215189873,
+      "grad_norm": 0.4015701413154602,
+      "learning_rate": 0.00019436167509211702,
+      "loss": 0.7597941160202026,
+      "step": 2544
+    },
+    {
+      "epoch": 1.0742616033755275,
+      "grad_norm": 0.3915148675441742,
+      "learning_rate": 0.00019434576461300888,
+      "loss": 0.7168124318122864,
+      "step": 2546
+    },
+    {
+      "epoch": 1.0751054852320676,
+      "grad_norm": 0.44643113017082214,
+      "learning_rate": 0.00019432983237014482,
+      "loss": 0.7762725949287415,
+      "step": 2548
+    },
+    {
+      "epoch": 1.0759493670886076,
+      "grad_norm": 0.46874719858169556,
+      "learning_rate": 0.00019431387836720004,
+      "loss": 0.7879725098609924,
+      "step": 2550
+    },
+    {
+      "epoch": 1.0767932489451477,
+      "grad_norm": 0.4143001437187195,
+      "learning_rate": 0.00019429790260785487,
+      "loss": 0.7351391315460205,
+      "step": 2552
+    },
+    {
+      "epoch": 1.0776371308016879,
+      "grad_norm": 0.4569888412952423,
+      "learning_rate": 0.0001942819050957946,
+      "loss": 0.86613929271698,
+      "step": 2554
+    },
+    {
+      "epoch": 1.0784810126582278,
+      "grad_norm": 0.41729897260665894,
+      "learning_rate": 0.00019426588583470954,
+      "loss": 0.7122719287872314,
+      "step": 2556
+    },
+    {
+      "epoch": 1.079324894514768,
+      "grad_norm": 0.42295604944229126,
+      "learning_rate": 0.000194249844828295,
+      "loss": 0.6605332493782043,
+      "step": 2558
+    },
+    {
+      "epoch": 1.080168776371308,
+      "grad_norm": 0.45787641406059265,
+      "learning_rate": 0.00019423378208025137,
+      "loss": 0.7839207649230957,
+      "step": 2560
+    },
+    {
+      "epoch": 1.081012658227848,
+      "grad_norm": 0.43370622396469116,
+      "learning_rate": 0.00019421769759428404,
+      "loss": 0.7557987570762634,
+      "step": 2562
+    },
+    {
+      "epoch": 1.0818565400843883,
+      "grad_norm": 0.4371975362300873,
+      "learning_rate": 0.00019420159137410335,
+      "loss": 0.7349345088005066,
+      "step": 2564
+    },
+    {
+      "epoch": 1.0827004219409282,
+      "grad_norm": 0.468730628490448,
+      "learning_rate": 0.0001941854634234247,
+      "loss": 0.7024962902069092,
+      "step": 2566
+    },
+    {
+      "epoch": 1.0835443037974684,
+      "grad_norm": 0.4257284998893738,
+      "learning_rate": 0.0001941693137459685,
+      "loss": 0.6981064081192017,
+      "step": 2568
+    },
+    {
+      "epoch": 1.0843881856540085,
+      "grad_norm": 0.4322902262210846,
+      "learning_rate": 0.00019415314234546015,
+      "loss": 0.7899637818336487,
+      "step": 2570
+    },
+    {
+      "epoch": 1.0852320675105485,
+      "grad_norm": 0.44384217262268066,
+      "learning_rate": 0.00019413694922563014,
+      "loss": 0.7964152693748474,
+      "step": 2572
+    },
+    {
+      "epoch": 1.0860759493670886,
+      "grad_norm": 0.47391360998153687,
+      "learning_rate": 0.0001941207343902139,
+      "loss": 0.7902941107749939,
+      "step": 2574
+    },
+    {
+      "epoch": 1.0869198312236288,
+      "grad_norm": 0.4575883746147156,
+      "learning_rate": 0.00019410449784295182,
+      "loss": 0.7818595170974731,
+      "step": 2576
+    },
+    {
+      "epoch": 1.0877637130801687,
+      "grad_norm": 0.41236087679862976,
+      "learning_rate": 0.0001940882395875894,
+      "loss": 0.7516021728515625,
+      "step": 2578
+    },
+    {
+      "epoch": 1.0886075949367089,
+      "grad_norm": 0.3817494213581085,
+      "learning_rate": 0.0001940719596278771,
+      "loss": 0.6870401501655579,
+      "step": 2580
+    },
+    {
+      "epoch": 1.0894514767932488,
+      "grad_norm": 0.4630168378353119,
+      "learning_rate": 0.00019405565796757043,
+      "loss": 0.7437341809272766,
+      "step": 2582
+    },
+    {
+      "epoch": 1.090295358649789,
+      "grad_norm": 0.4063820540904999,
+      "learning_rate": 0.00019403933461042982,
+      "loss": 0.7101680040359497,
+      "step": 2584
+    },
+    {
+      "epoch": 1.0911392405063292,
+      "grad_norm": 0.44423359632492065,
+      "learning_rate": 0.0001940229895602208,
+      "loss": 0.6894780397415161,
+      "step": 2586
+    },
+    {
+      "epoch": 1.091983122362869,
+      "grad_norm": 0.4267353117465973,
+      "learning_rate": 0.0001940066228207138,
+      "loss": 0.7277120351791382,
+      "step": 2588
+    },
+    {
+      "epoch": 1.0928270042194093,
+      "grad_norm": 0.44747647643089294,
+      "learning_rate": 0.00019399023439568434,
+      "loss": 0.6512353420257568,
+      "step": 2590
+    },
+    {
+      "epoch": 1.0936708860759494,
+      "grad_norm": 0.4557100832462311,
+      "learning_rate": 0.00019397382428891296,
+      "loss": 0.7850068807601929,
+      "step": 2592
+    },
+    {
+      "epoch": 1.0945147679324894,
+      "grad_norm": 0.44980114698410034,
+      "learning_rate": 0.0001939573925041851,
+      "loss": 0.786868691444397,
+      "step": 2594
+    },
+    {
+      "epoch": 1.0953586497890295,
+      "grad_norm": 0.45374399423599243,
+      "learning_rate": 0.00019394093904529126,
+      "loss": 0.8260899186134338,
+      "step": 2596
+    },
+    {
+      "epoch": 1.0962025316455697,
+      "grad_norm": 0.40681353211402893,
+      "learning_rate": 0.00019392446391602695,
+      "loss": 0.7056689262390137,
+      "step": 2598
+    },
+    {
+      "epoch": 1.0970464135021096,
+      "grad_norm": 0.39278075098991394,
+      "learning_rate": 0.0001939079671201927,
+      "loss": 0.7557252645492554,
+      "step": 2600
+    },
+    {
+      "epoch": 1.0970464135021096,
+      "eval_loss": 0.8211485743522644,
+      "eval_runtime": 670.2276,
+      "eval_samples_per_second": 3.144,
+      "eval_steps_per_second": 3.144,
+      "step": 2600
+    },
+    {
+      "epoch": 1.0978902953586498,
+      "grad_norm": 0.4261230230331421,
+      "learning_rate": 0.00019389144866159398,
+      "loss": 0.7139819264411926,
+      "step": 2602
+    },
+    {
+      "epoch": 1.09873417721519,
+      "grad_norm": 0.4664439558982849,
+      "learning_rate": 0.0001938749085440412,
+      "loss": 0.7274705767631531,
+      "step": 2604
+    },
+    {
+      "epoch": 1.09957805907173,
+      "grad_norm": 0.46306154131889343,
+      "learning_rate": 0.00019385834677134998,
+      "loss": 0.7547051906585693,
+      "step": 2606
+    },
+    {
+      "epoch": 1.10042194092827,
+      "grad_norm": 0.39753687381744385,
+      "learning_rate": 0.00019384176334734073,
+      "loss": 0.6942694187164307,
+      "step": 2608
+    },
+    {
+      "epoch": 1.1012658227848102,
+      "grad_norm": 0.4556522071361542,
+      "learning_rate": 0.00019382515827583898,
+      "loss": 0.7161701321601868,
+      "step": 2610
+    },
+    {
+      "epoch": 1.1021097046413502,
+      "grad_norm": 0.3863290250301361,
+      "learning_rate": 0.0001938085315606751,
+      "loss": 0.7165710926055908,
+      "step": 2612
+    },
+    {
+      "epoch": 1.1029535864978903,
+      "grad_norm": 0.49601292610168457,
+      "learning_rate": 0.00019379188320568466,
+      "loss": 0.8080679774284363,
+      "step": 2614
+    },
+    {
+      "epoch": 1.1037974683544305,
+      "grad_norm": 0.4498465955257416,
+      "learning_rate": 0.00019377521321470805,
+      "loss": 0.7032502889633179,
+      "step": 2616
+    },
+    {
+      "epoch": 1.1046413502109704,
+      "grad_norm": 0.4537963569164276,
+      "learning_rate": 0.00019375852159159073,
+      "loss": 0.7076731324195862,
+      "step": 2618
+    },
+    {
+      "epoch": 1.1054852320675106,
+      "grad_norm": 0.4042968153953552,
+      "learning_rate": 0.0001937418083401832,
+      "loss": 0.7020825743675232,
+      "step": 2620
+    },
+    {
+      "epoch": 1.1063291139240505,
+      "grad_norm": 0.44306516647338867,
+      "learning_rate": 0.00019372507346434076,
+      "loss": 0.7222874760627747,
+      "step": 2622
+    },
+    {
+      "epoch": 1.1071729957805907,
+      "grad_norm": 0.3792516887187958,
+      "learning_rate": 0.00019370831696792392,
+      "loss": 0.6136211156845093,
+      "step": 2624
+    },
+    {
+      "epoch": 1.1080168776371309,
+      "grad_norm": 0.46655353903770447,
+      "learning_rate": 0.00019369153885479806,
+      "loss": 0.7358977198600769,
+      "step": 2626
+    },
+    {
+      "epoch": 1.1088607594936708,
+      "grad_norm": 0.4416999816894531,
+      "learning_rate": 0.0001936747391288336,
+      "loss": 0.7484551072120667,
+      "step": 2628
+    },
+    {
+      "epoch": 1.109704641350211,
+      "grad_norm": 0.4620150327682495,
+      "learning_rate": 0.00019365791779390584,
+      "loss": 0.6874802708625793,
+      "step": 2630
+    },
+    {
+      "epoch": 1.1105485232067511,
+      "grad_norm": 0.4357883334159851,
+      "learning_rate": 0.00019364107485389518,
+      "loss": 0.7168865203857422,
+      "step": 2632
+    },
+    {
+      "epoch": 1.111392405063291,
+      "grad_norm": 0.4581143856048584,
+      "learning_rate": 0.00019362421031268699,
+      "loss": 0.7533737421035767,
+      "step": 2634
+    },
+    {
+      "epoch": 1.1122362869198312,
+      "grad_norm": 0.46408653259277344,
+      "learning_rate": 0.00019360732417417152,
+      "loss": 0.8245398998260498,
+      "step": 2636
+    },
+    {
+      "epoch": 1.1130801687763714,
+      "grad_norm": 0.4639282524585724,
+      "learning_rate": 0.00019359041644224415,
+      "loss": 0.7095663547515869,
+      "step": 2638
+    },
+    {
+      "epoch": 1.1139240506329113,
+      "grad_norm": 0.4598273038864136,
+      "learning_rate": 0.00019357348712080517,
+      "loss": 0.7030285596847534,
+      "step": 2640
+    },
+    {
+      "epoch": 1.1147679324894515,
+      "grad_norm": 0.42993804812431335,
+      "learning_rate": 0.00019355653621375978,
+      "loss": 0.7158969640731812,
+      "step": 2642
+    },
+    {
+      "epoch": 1.1156118143459917,
+      "grad_norm": 0.36132168769836426,
+      "learning_rate": 0.0001935395637250183,
+      "loss": 0.6800820231437683,
+      "step": 2644
+    },
+    {
+      "epoch": 1.1164556962025316,
+      "grad_norm": 0.5004428029060364,
+      "learning_rate": 0.00019352256965849592,
+      "loss": 0.694972813129425,
+      "step": 2646
+    },
+    {
+      "epoch": 1.1172995780590718,
+      "grad_norm": 0.43715354800224304,
+      "learning_rate": 0.00019350555401811286,
+      "loss": 0.7164262533187866,
+      "step": 2648
+    },
+    {
+      "epoch": 1.1181434599156117,
+      "grad_norm": 0.42316755652427673,
+      "learning_rate": 0.00019348851680779433,
+      "loss": 0.7026737928390503,
+      "step": 2650
+    },
+    {
+      "epoch": 1.1189873417721519,
+      "grad_norm": 0.35848063230514526,
+      "learning_rate": 0.0001934714580314704,
+      "loss": 0.6774232387542725,
+      "step": 2652
+    },
+    {
+      "epoch": 1.119831223628692,
+      "grad_norm": 0.46613892912864685,
+      "learning_rate": 0.00019345437769307627,
+      "loss": 0.7789244055747986,
+      "step": 2654
+    },
+    {
+      "epoch": 1.120675105485232,
+      "grad_norm": 0.4298146367073059,
+      "learning_rate": 0.00019343727579655204,
+      "loss": 0.8647177815437317,
+      "step": 2656
+    },
+    {
+      "epoch": 1.1215189873417721,
+      "grad_norm": 0.4638161361217499,
+      "learning_rate": 0.00019342015234584276,
+      "loss": 0.6949124336242676,
+      "step": 2658
+    },
+    {
+      "epoch": 1.1223628691983123,
+      "grad_norm": 0.4079605042934418,
+      "learning_rate": 0.00019340300734489854,
+      "loss": 0.6677125096321106,
+      "step": 2660
+    },
+    {
+      "epoch": 1.1232067510548522,
+      "grad_norm": 0.46079063415527344,
+      "learning_rate": 0.00019338584079767434,
+      "loss": 0.7521067261695862,
+      "step": 2662
+    },
+    {
+      "epoch": 1.1240506329113924,
+      "grad_norm": 0.3935202956199646,
+      "learning_rate": 0.0001933686527081302,
+      "loss": 0.6829184889793396,
+      "step": 2664
+    },
+    {
+      "epoch": 1.1248945147679326,
+      "grad_norm": 0.45696479082107544,
+      "learning_rate": 0.000193351443080231,
+      "loss": 0.7122842669487,
+      "step": 2666
+    },
+    {
+      "epoch": 1.1257383966244725,
+      "grad_norm": 0.4078282415866852,
+      "learning_rate": 0.0001933342119179468,
+      "loss": 0.7316977977752686,
+      "step": 2668
+    },
+    {
+      "epoch": 1.1265822784810127,
+      "grad_norm": 0.4282447099685669,
+      "learning_rate": 0.0001933169592252524,
+      "loss": 0.7323395013809204,
+      "step": 2670
+    },
+    {
+      "epoch": 1.1274261603375528,
+      "grad_norm": 0.46398648619651794,
+      "learning_rate": 0.00019329968500612766,
+      "loss": 0.7177348732948303,
+      "step": 2672
+    },
+    {
+      "epoch": 1.1282700421940928,
+      "grad_norm": 0.4560202658176422,
+      "learning_rate": 0.0001932823892645575,
+      "loss": 0.7957183122634888,
+      "step": 2674
+    },
+    {
+      "epoch": 1.129113924050633,
+      "grad_norm": 0.4790743887424469,
+      "learning_rate": 0.00019326507200453162,
+      "loss": 0.762060284614563,
+      "step": 2676
+    },
+    {
+      "epoch": 1.129957805907173,
+      "grad_norm": 0.46524202823638916,
+      "learning_rate": 0.00019324773323004484,
+      "loss": 0.7227895259857178,
+      "step": 2678
+    },
+    {
+      "epoch": 1.130801687763713,
+      "grad_norm": 0.47912222146987915,
+      "learning_rate": 0.00019323037294509683,
+      "loss": 0.6968042254447937,
+      "step": 2680
+    },
+    {
+      "epoch": 1.1316455696202532,
+      "grad_norm": 0.4412849247455597,
+      "learning_rate": 0.00019321299115369233,
+      "loss": 0.6682016253471375,
+      "step": 2682
+    },
+    {
+      "epoch": 1.1324894514767934,
+      "grad_norm": 0.4138370752334595,
+      "learning_rate": 0.00019319558785984095,
+      "loss": 0.7287812829017639,
+      "step": 2684
+    },
+    {
+      "epoch": 1.1333333333333333,
+      "grad_norm": 0.391170859336853,
+      "learning_rate": 0.00019317816306755733,
+      "loss": 0.6547526121139526,
+      "step": 2686
+    },
+    {
+      "epoch": 1.1341772151898735,
+      "grad_norm": 0.45472806692123413,
+      "learning_rate": 0.00019316071678086098,
+      "loss": 0.6942042708396912,
+      "step": 2688
+    },
+    {
+      "epoch": 1.1350210970464134,
+      "grad_norm": 0.40321481227874756,
+      "learning_rate": 0.00019314324900377648,
+      "loss": 0.7417841553688049,
+      "step": 2690
+    },
+    {
+      "epoch": 1.1358649789029536,
+      "grad_norm": 0.3806784152984619,
+      "learning_rate": 0.00019312575974033327,
+      "loss": 0.7780525088310242,
+      "step": 2692
+    },
+    {
+      "epoch": 1.1367088607594937,
+      "grad_norm": 0.43763765692710876,
+      "learning_rate": 0.0001931082489945658,
+      "loss": 0.661472499370575,
+      "step": 2694
+    },
+    {
+      "epoch": 1.1375527426160337,
+      "grad_norm": 0.45410826802253723,
+      "learning_rate": 0.00019309071677051348,
+      "loss": 0.7443127632141113,
+      "step": 2696
+    },
+    {
+      "epoch": 1.1383966244725738,
+      "grad_norm": 0.47907447814941406,
+      "learning_rate": 0.00019307316307222062,
+      "loss": 0.744393527507782,
+      "step": 2698
+    },
+    {
+      "epoch": 1.139240506329114,
+      "grad_norm": 0.41723814606666565,
+      "learning_rate": 0.00019305558790373654,
+      "loss": 0.7042794227600098,
+      "step": 2700
+    },
+    {
+      "epoch": 1.139240506329114,
+      "eval_loss": 0.8155058026313782,
+      "eval_runtime": 678.284,
+      "eval_samples_per_second": 3.106,
+      "eval_steps_per_second": 3.106,
+      "step": 2700
+    },
+    {
+      "epoch": 1.140084388185654,
+      "grad_norm": 0.3863268196582794,
+      "learning_rate": 0.0001930379912691155,
+      "loss": 0.6715745329856873,
+      "step": 2702
+    },
+    {
+      "epoch": 1.140928270042194,
+      "grad_norm": 0.49994897842407227,
+      "learning_rate": 0.00019302037317241673,
+      "loss": 0.6959030628204346,
+      "step": 2704
+    },
+    {
+      "epoch": 1.1417721518987343,
+      "grad_norm": 0.46377572417259216,
+      "learning_rate": 0.00019300273361770432,
+      "loss": 0.6534222364425659,
+      "step": 2706
+    },
+    {
+      "epoch": 1.1426160337552742,
+      "grad_norm": 0.3841579556465149,
+      "learning_rate": 0.00019298507260904745,
+      "loss": 0.6750475764274597,
+      "step": 2708
+    },
+    {
+      "epoch": 1.1434599156118144,
+      "grad_norm": 0.4171544313430786,
+      "learning_rate": 0.00019296739015052015,
+      "loss": 0.7714900374412537,
+      "step": 2710
+    },
+    {
+      "epoch": 1.1443037974683543,
+      "grad_norm": 0.44384199380874634,
+      "learning_rate": 0.00019294968624620137,
+      "loss": 0.7281306982040405,
+      "step": 2712
+    },
+    {
+      "epoch": 1.1451476793248945,
+      "grad_norm": 0.43345192074775696,
+      "learning_rate": 0.00019293196090017513,
+      "loss": 0.7407262325286865,
+      "step": 2714
+    },
+    {
+      "epoch": 1.1459915611814346,
+      "grad_norm": 0.3897020220756531,
+      "learning_rate": 0.00019291421411653034,
+      "loss": 0.6794347167015076,
+      "step": 2716
+    },
+    {
+      "epoch": 1.1468354430379746,
+      "grad_norm": 0.4289143681526184,
+      "learning_rate": 0.00019289644589936075,
+      "loss": 0.7026289105415344,
+      "step": 2718
+    },
+    {
+      "epoch": 1.1476793248945147,
+      "grad_norm": 0.4783557057380676,
+      "learning_rate": 0.0001928786562527652,
+      "loss": 0.7613762617111206,
+      "step": 2720
+    },
+    {
+      "epoch": 1.148523206751055,
+      "grad_norm": 0.4152112901210785,
+      "learning_rate": 0.00019286084518084745,
+      "loss": 0.7123059630393982,
+      "step": 2722
+    },
+    {
+      "epoch": 1.1493670886075948,
+      "grad_norm": 0.4204908013343811,
+      "learning_rate": 0.0001928430126877161,
+      "loss": 0.7198041677474976,
+      "step": 2724
+    },
+    {
+      "epoch": 1.150210970464135,
+      "grad_norm": 0.41333863139152527,
+      "learning_rate": 0.00019282515877748485,
+      "loss": 0.6859839558601379,
+      "step": 2726
+    },
+    {
+      "epoch": 1.1510548523206752,
+      "grad_norm": 0.4041498005390167,
+      "learning_rate": 0.00019280728345427218,
+      "loss": 0.7185758948326111,
+      "step": 2728
+    },
+    {
+      "epoch": 1.1518987341772151,
+      "grad_norm": 0.4487358629703522,
+      "learning_rate": 0.00019278938672220167,
+      "loss": 0.7899971008300781,
+      "step": 2730
+    },
+    {
+      "epoch": 1.1527426160337553,
+      "grad_norm": 0.3510114252567291,
+      "learning_rate": 0.00019277146858540166,
+      "loss": 0.6266586780548096,
+      "step": 2732
+    },
+    {
+      "epoch": 1.1535864978902954,
+      "grad_norm": 0.4295555055141449,
+      "learning_rate": 0.00019275352904800553,
+      "loss": 0.7450110912322998,
+      "step": 2734
+    },
+    {
+      "epoch": 1.1544303797468354,
+      "grad_norm": 0.40629687905311584,
+      "learning_rate": 0.00019273556811415165,
+      "loss": 0.6875549554824829,
+      "step": 2736
+    },
+    {
+      "epoch": 1.1552742616033755,
+      "grad_norm": 0.4243362247943878,
+      "learning_rate": 0.00019271758578798322,
+      "loss": 0.8141522407531738,
+      "step": 2738
+    },
+    {
+      "epoch": 1.1561181434599157,
+      "grad_norm": 0.483306884765625,
+      "learning_rate": 0.00019269958207364843,
+      "loss": 0.6905186176300049,
+      "step": 2740
+    },
+    {
+      "epoch": 1.1569620253164556,
+      "grad_norm": 0.4303925335407257,
+      "learning_rate": 0.00019268155697530037,
+      "loss": 0.6709047555923462,
+      "step": 2742
+    },
+    {
+      "epoch": 1.1578059071729958,
+      "grad_norm": 0.4533381760120392,
+      "learning_rate": 0.0001926635104970971,
+      "loss": 0.7107465863227844,
+      "step": 2744
+    },
+    {
+      "epoch": 1.158649789029536,
+      "grad_norm": 0.4823286533355713,
+      "learning_rate": 0.0001926454426432016,
+      "loss": 0.8043986558914185,
+      "step": 2746
+    },
+    {
+      "epoch": 1.159493670886076,
+      "grad_norm": 0.3892867863178253,
+      "learning_rate": 0.00019262735341778179,
+      "loss": 0.7812165021896362,
+      "step": 2748
+    },
+    {
+      "epoch": 1.160337552742616,
+      "grad_norm": 0.46737003326416016,
+      "learning_rate": 0.00019260924282501045,
+      "loss": 0.7141886353492737,
+      "step": 2750
+    },
+    {
+      "epoch": 1.1611814345991562,
+      "grad_norm": 0.4335328936576843,
+      "learning_rate": 0.0001925911108690654,
+      "loss": 0.7528460621833801,
+      "step": 2752
+    },
+    {
+      "epoch": 1.1620253164556962,
+      "grad_norm": 0.3792484700679779,
+      "learning_rate": 0.00019257295755412932,
+      "loss": 0.7817611694335938,
+      "step": 2754
+    },
+    {
+      "epoch": 1.1628691983122363,
+      "grad_norm": 0.41246429085731506,
+      "learning_rate": 0.00019255478288438983,
+      "loss": 0.6904958486557007,
+      "step": 2756
+    },
+    {
+      "epoch": 1.1637130801687763,
+      "grad_norm": 0.44368481636047363,
+      "learning_rate": 0.00019253658686403948,
+      "loss": 0.7571630477905273,
+      "step": 2758
+    },
+    {
+      "epoch": 1.1645569620253164,
+      "grad_norm": 0.40121781826019287,
+      "learning_rate": 0.00019251836949727574,
+      "loss": 0.6747061014175415,
+      "step": 2760
+    },
+    {
+      "epoch": 1.1654008438818566,
+      "grad_norm": 0.4660188853740692,
+      "learning_rate": 0.000192500130788301,
+      "loss": 0.7722339034080505,
+      "step": 2762
+    },
+    {
+      "epoch": 1.1662447257383965,
+      "grad_norm": 0.41693782806396484,
+      "learning_rate": 0.0001924818707413226,
+      "loss": 0.8081457018852234,
+      "step": 2764
+    },
+    {
+      "epoch": 1.1670886075949367,
+      "grad_norm": 0.4274570345878601,
+      "learning_rate": 0.00019246358936055274,
+      "loss": 0.7745853066444397,
+      "step": 2766
+    },
+    {
+      "epoch": 1.1679324894514769,
+      "grad_norm": 0.4385969638824463,
+      "learning_rate": 0.00019244528665020863,
+      "loss": 0.7918146252632141,
+      "step": 2768
+    },
+    {
+      "epoch": 1.1687763713080168,
+      "grad_norm": 0.42832866311073303,
+      "learning_rate": 0.00019242696261451234,
+      "loss": 0.7382830381393433,
+      "step": 2770
+    },
+    {
+      "epoch": 1.169620253164557,
+      "grad_norm": 0.44177982211112976,
+      "learning_rate": 0.00019240861725769085,
+      "loss": 0.7501721978187561,
+      "step": 2772
+    },
+    {
+      "epoch": 1.1704641350210971,
+      "grad_norm": 0.4193457365036011,
+      "learning_rate": 0.00019239025058397611,
+      "loss": 0.6495388150215149,
+      "step": 2774
+    },
+    {
+      "epoch": 1.171308016877637,
+      "grad_norm": 0.5145925283432007,
+      "learning_rate": 0.000192371862597605,
+      "loss": 0.8291991353034973,
+      "step": 2776
+    },
+    {
+      "epoch": 1.1721518987341772,
+      "grad_norm": 0.40911251306533813,
+      "learning_rate": 0.00019235345330281914,
+      "loss": 0.7491340637207031,
+      "step": 2778
+    },
+    {
+      "epoch": 1.1729957805907172,
+      "grad_norm": 0.3920276165008545,
+      "learning_rate": 0.0001923350227038653,
+      "loss": 0.6315004825592041,
+      "step": 2780
+    },
+    {
+      "epoch": 1.1738396624472573,
+      "grad_norm": 0.442303329706192,
+      "learning_rate": 0.0001923165708049951,
+      "loss": 0.7538990378379822,
+      "step": 2782
+    },
+    {
+      "epoch": 1.1746835443037975,
+      "grad_norm": 0.4693523049354553,
+      "learning_rate": 0.00019229809761046497,
+      "loss": 0.7440794110298157,
+      "step": 2784
+    },
+    {
+      "epoch": 1.1755274261603375,
+      "grad_norm": 0.4948059618473053,
+      "learning_rate": 0.0001922796031245363,
+      "loss": 0.7342568635940552,
+      "step": 2786
+    },
+    {
+      "epoch": 1.1763713080168776,
+      "grad_norm": 0.3900465965270996,
+      "learning_rate": 0.00019226108735147547,
+      "loss": 0.6969661116600037,
+      "step": 2788
+    },
+    {
+      "epoch": 1.1772151898734178,
+      "grad_norm": 0.36153602600097656,
+      "learning_rate": 0.00019224255029555372,
+      "loss": 0.7097541689872742,
+      "step": 2790
+    },
+    {
+      "epoch": 1.1780590717299577,
+      "grad_norm": 0.41168832778930664,
+      "learning_rate": 0.00019222399196104715,
+      "loss": 0.8141831755638123,
+      "step": 2792
+    },
+    {
+      "epoch": 1.1789029535864979,
+      "grad_norm": 0.4542821943759918,
+      "learning_rate": 0.0001922054123522368,
+      "loss": 0.7458863258361816,
+      "step": 2794
+    },
+    {
+      "epoch": 1.179746835443038,
+      "grad_norm": 0.4026047885417938,
+      "learning_rate": 0.0001921868114734087,
+      "loss": 0.6535642147064209,
+      "step": 2796
+    },
+    {
+      "epoch": 1.180590717299578,
+      "grad_norm": 0.4417034387588501,
+      "learning_rate": 0.0001921681893288537,
+      "loss": 0.6248135566711426,
+      "step": 2798
+    },
+    {
+      "epoch": 1.1814345991561181,
+      "grad_norm": 0.45027074217796326,
+      "learning_rate": 0.00019214954592286748,
+      "loss": 0.6777785420417786,
+      "step": 2800
+    },
+    {
+      "epoch": 1.1814345991561181,
+      "eval_loss": 0.8124309182167053,
+      "eval_runtime": 688.4759,
+      "eval_samples_per_second": 3.06,
+      "eval_steps_per_second": 3.06,
+      "step": 2800
+    },
+    {
+      "epoch": 1.1822784810126583,
+      "grad_norm": 0.406865656375885,
+      "learning_rate": 0.00019213088125975082,
+      "loss": 0.6460301876068115,
+      "step": 2802
+    },
+    {
+      "epoch": 1.1831223628691983,
+      "grad_norm": 0.5052749514579773,
+      "learning_rate": 0.00019211219534380928,
+      "loss": 0.7292830944061279,
+      "step": 2804
+    },
+    {
+      "epoch": 1.1839662447257384,
+      "grad_norm": 0.4028380513191223,
+      "learning_rate": 0.00019209348817935328,
+      "loss": 0.7399409413337708,
+      "step": 2806
+    },
+    {
+      "epoch": 1.1848101265822786,
+      "grad_norm": 0.4522664248943329,
+      "learning_rate": 0.00019207475977069826,
+      "loss": 0.6885398030281067,
+      "step": 2808
+    },
+    {
+      "epoch": 1.1856540084388185,
+      "grad_norm": 0.419546514749527,
+      "learning_rate": 0.0001920560101221645,
+      "loss": 0.6962382197380066,
+      "step": 2810
+    },
+    {
+      "epoch": 1.1864978902953587,
+      "grad_norm": 0.48912352323532104,
+      "learning_rate": 0.00019203723923807715,
+      "loss": 0.7380965352058411,
+      "step": 2812
+    },
+    {
+      "epoch": 1.1873417721518988,
+      "grad_norm": 0.40905487537384033,
+      "learning_rate": 0.00019201844712276635,
+      "loss": 0.755382776260376,
+      "step": 2814
+    },
+    {
+      "epoch": 1.1881856540084388,
+      "grad_norm": 0.385623037815094,
+      "learning_rate": 0.000191999633780567,
+      "loss": 0.7066168189048767,
+      "step": 2816
+    },
+    {
+      "epoch": 1.189029535864979,
+      "grad_norm": 0.4488723576068878,
+      "learning_rate": 0.00019198079921581908,
+      "loss": 0.7172947525978088,
+      "step": 2818
+    },
+    {
+      "epoch": 1.189873417721519,
+      "grad_norm": 0.385277658700943,
+      "learning_rate": 0.00019196194343286727,
+      "loss": 0.6714195013046265,
+      "step": 2820
+    },
+    {
+      "epoch": 1.190717299578059,
+      "grad_norm": 0.4280855655670166,
+      "learning_rate": 0.00019194306643606127,
+      "loss": 0.7076966166496277,
+      "step": 2822
+    },
+    {
+      "epoch": 1.1915611814345992,
+      "grad_norm": 0.4626067578792572,
+      "learning_rate": 0.00019192416822975567,
+      "loss": 0.7926747798919678,
+      "step": 2824
+    },
+    {
+      "epoch": 1.1924050632911392,
+      "grad_norm": 0.3994208574295044,
+      "learning_rate": 0.00019190524881830987,
+      "loss": 0.7654290199279785,
+      "step": 2826
+    },
+    {
+      "epoch": 1.1932489451476793,
+      "grad_norm": 0.47346508502960205,
+      "learning_rate": 0.00019188630820608825,
+      "loss": 0.7857998609542847,
+      "step": 2828
+    },
+    {
+      "epoch": 1.1940928270042195,
+      "grad_norm": 0.446077823638916,
+      "learning_rate": 0.00019186734639746003,
+      "loss": 0.7603738903999329,
+      "step": 2830
+    },
+    {
+      "epoch": 1.1949367088607594,
+      "grad_norm": 0.41503727436065674,
+      "learning_rate": 0.00019184836339679936,
+      "loss": 0.790582537651062,
+      "step": 2832
+    },
+    {
+      "epoch": 1.1957805907172996,
+      "grad_norm": 0.43651700019836426,
+      "learning_rate": 0.00019182935920848522,
+      "loss": 0.7229864597320557,
+      "step": 2834
+    },
+    {
+      "epoch": 1.1966244725738397,
+      "grad_norm": 0.4409031867980957,
+      "learning_rate": 0.00019181033383690153,
+      "loss": 0.7817287445068359,
+      "step": 2836
+    },
+    {
+      "epoch": 1.1974683544303797,
+      "grad_norm": 0.3562015891075134,
+      "learning_rate": 0.0001917912872864371,
+      "loss": 0.6342297792434692,
+      "step": 2838
+    },
+    {
+      "epoch": 1.1983122362869199,
+      "grad_norm": 0.4767192602157593,
+      "learning_rate": 0.00019177221956148555,
+      "loss": 0.7065272927284241,
+      "step": 2840
+    },
+    {
+      "epoch": 1.1991561181434598,
+      "grad_norm": 0.45131945610046387,
+      "learning_rate": 0.00019175313066644546,
+      "loss": 0.777110755443573,
+      "step": 2842
+    },
+    {
+      "epoch": 1.2,
+      "grad_norm": 0.47566652297973633,
+      "learning_rate": 0.00019173402060572027,
+      "loss": 0.7456457018852234,
+      "step": 2844
+    },
+    {
+      "epoch": 1.2008438818565401,
+      "grad_norm": 0.48909857869148254,
+      "learning_rate": 0.00019171488938371834,
+      "loss": 0.8041547536849976,
+      "step": 2846
+    },
+    {
+      "epoch": 1.20168776371308,
+      "grad_norm": 0.436591774225235,
+      "learning_rate": 0.00019169573700485284,
+      "loss": 0.6786676049232483,
+      "step": 2848
+    },
+    {
+      "epoch": 1.2025316455696202,
+      "grad_norm": 0.44230160117149353,
+      "learning_rate": 0.00019167656347354183,
+      "loss": 0.7661880850791931,
+      "step": 2850
+    },
+    {
+      "epoch": 1.2033755274261604,
+      "grad_norm": 0.4844048321247101,
+      "learning_rate": 0.00019165736879420836,
+      "loss": 0.7369645833969116,
+      "step": 2852
+    },
+    {
+      "epoch": 1.2042194092827003,
+      "grad_norm": 0.4272955060005188,
+      "learning_rate": 0.0001916381529712802,
+      "loss": 0.7776681184768677,
+      "step": 2854
+    },
+    {
+      "epoch": 1.2050632911392405,
+      "grad_norm": 0.42392730712890625,
+      "learning_rate": 0.00019161891600919008,
+      "loss": 0.8019667267799377,
+      "step": 2856
+    },
+    {
+      "epoch": 1.2059071729957807,
+      "grad_norm": 0.4528326988220215,
+      "learning_rate": 0.00019159965791237565,
+      "loss": 0.7811952233314514,
+      "step": 2858
+    },
+    {
+      "epoch": 1.2067510548523206,
+      "grad_norm": 0.42840948700904846,
+      "learning_rate": 0.00019158037868527932,
+      "loss": 0.7244279980659485,
+      "step": 2860
+    },
+    {
+      "epoch": 1.2075949367088608,
+      "grad_norm": 0.4291226863861084,
+      "learning_rate": 0.00019156107833234845,
+      "loss": 0.7675760984420776,
+      "step": 2862
+    },
+    {
+      "epoch": 1.208438818565401,
+      "grad_norm": 0.4455093443393707,
+      "learning_rate": 0.00019154175685803527,
+      "loss": 0.7071483135223389,
+      "step": 2864
+    },
+    {
+      "epoch": 1.2092827004219409,
+      "grad_norm": 0.41356268525123596,
+      "learning_rate": 0.00019152241426679688,
+      "loss": 0.7554603815078735,
+      "step": 2866
+    },
+    {
+      "epoch": 1.210126582278481,
+      "grad_norm": 0.43875646591186523,
+      "learning_rate": 0.00019150305056309525,
+      "loss": 0.7198356986045837,
+      "step": 2868
+    },
+    {
+      "epoch": 1.2109704641350212,
+      "grad_norm": 0.454356849193573,
+      "learning_rate": 0.0001914836657513972,
+      "loss": 0.7393195629119873,
+      "step": 2870
+    },
+    {
+      "epoch": 1.2118143459915611,
+      "grad_norm": 0.37668535113334656,
+      "learning_rate": 0.00019146425983617444,
+      "loss": 0.7277671694755554,
+      "step": 2872
+    },
+    {
+      "epoch": 1.2126582278481013,
+      "grad_norm": 0.37869662046432495,
+      "learning_rate": 0.00019144483282190354,
+      "loss": 0.6783158183097839,
+      "step": 2874
+    },
+    {
+      "epoch": 1.2135021097046415,
+      "grad_norm": 0.4333525002002716,
+      "learning_rate": 0.00019142538471306596,
+      "loss": 0.7024772763252258,
+      "step": 2876
+    },
+    {
+      "epoch": 1.2143459915611814,
+      "grad_norm": 0.4693842828273773,
+      "learning_rate": 0.00019140591551414796,
+      "loss": 0.7387984395027161,
+      "step": 2878
+    },
+    {
+      "epoch": 1.2151898734177216,
+      "grad_norm": 0.432640939950943,
+      "learning_rate": 0.00019138642522964074,
+      "loss": 0.7597174644470215,
+      "step": 2880
+    },
+    {
+      "epoch": 1.2160337552742617,
+      "grad_norm": 0.432477205991745,
+      "learning_rate": 0.00019136691386404032,
+      "loss": 0.7884587049484253,
+      "step": 2882
+    },
+    {
+      "epoch": 1.2168776371308017,
+      "grad_norm": 0.39305904507637024,
+      "learning_rate": 0.00019134738142184763,
+      "loss": 0.6873269081115723,
+      "step": 2884
+    },
+    {
+      "epoch": 1.2177215189873418,
+      "grad_norm": 0.4485352635383606,
+      "learning_rate": 0.00019132782790756838,
+      "loss": 0.6523491144180298,
+      "step": 2886
+    },
+    {
+      "epoch": 1.2185654008438818,
+      "grad_norm": 0.4480868875980377,
+      "learning_rate": 0.00019130825332571322,
+      "loss": 0.7383941411972046,
+      "step": 2888
+    },
+    {
+      "epoch": 1.219409282700422,
+      "grad_norm": 0.44085049629211426,
+      "learning_rate": 0.00019128865768079764,
+      "loss": 0.7391335964202881,
+      "step": 2890
+    },
+    {
+      "epoch": 1.220253164556962,
+      "grad_norm": 0.4621105492115021,
+      "learning_rate": 0.00019126904097734198,
+      "loss": 0.7282017469406128,
+      "step": 2892
+    },
+    {
+      "epoch": 1.221097046413502,
+      "grad_norm": 0.41478198766708374,
+      "learning_rate": 0.00019124940321987136,
+      "loss": 0.7554038763046265,
+      "step": 2894
+    },
+    {
+      "epoch": 1.2219409282700422,
+      "grad_norm": 0.4143751263618469,
+      "learning_rate": 0.00019122974441291595,
+      "loss": 0.6826799511909485,
+      "step": 2896
+    },
+    {
+      "epoch": 1.2227848101265824,
+      "grad_norm": 0.45392894744873047,
+      "learning_rate": 0.00019121006456101057,
+      "loss": 0.751000702381134,
+      "step": 2898
+    },
+    {
+      "epoch": 1.2236286919831223,
+      "grad_norm": 0.46946296095848083,
+      "learning_rate": 0.00019119036366869506,
+      "loss": 0.7598516941070557,
+      "step": 2900
+    },
+    {
+      "epoch": 1.2236286919831223,
+      "eval_loss": 0.8077136278152466,
+      "eval_runtime": 685.6042,
+      "eval_samples_per_second": 3.073,
+      "eval_steps_per_second": 3.073,
+      "step": 2900
+    },
+    {
+      "epoch": 1.2244725738396625,
+      "grad_norm": 0.4316656291484833,
+      "learning_rate": 0.00019117064174051395,
+      "loss": 0.7518358826637268,
+      "step": 2902
+    },
+    {
+      "epoch": 1.2253164556962026,
+      "grad_norm": 0.47737377882003784,
+      "learning_rate": 0.0001911508987810168,
+      "loss": 0.7308215498924255,
+      "step": 2904
+    },
+    {
+      "epoch": 1.2261603375527426,
+      "grad_norm": 0.40906062722206116,
+      "learning_rate": 0.00019113113479475783,
+      "loss": 0.6996226906776428,
+      "step": 2906
+    },
+    {
+      "epoch": 1.2270042194092827,
+      "grad_norm": 0.517085075378418,
+      "learning_rate": 0.00019111134978629633,
+      "loss": 0.8364415168762207,
+      "step": 2908
+    },
+    {
+      "epoch": 1.2278481012658227,
+      "grad_norm": 0.425581693649292,
+      "learning_rate": 0.00019109154376019624,
+      "loss": 0.722011923789978,
+      "step": 2910
+    },
+    {
+      "epoch": 1.2286919831223628,
+      "grad_norm": 0.4651593565940857,
+      "learning_rate": 0.00019107171672102643,
+      "loss": 0.7586172223091125,
+      "step": 2912
+    },
+    {
+      "epoch": 1.229535864978903,
+      "grad_norm": 0.4578891396522522,
+      "learning_rate": 0.00019105186867336067,
+      "loss": 0.7387955188751221,
+      "step": 2914
+    },
+    {
+      "epoch": 1.230379746835443,
+      "grad_norm": 0.4312961995601654,
+      "learning_rate": 0.00019103199962177748,
+      "loss": 0.8036285638809204,
+      "step": 2916
+    },
+    {
+      "epoch": 1.231223628691983,
+      "grad_norm": 0.4355425536632538,
+      "learning_rate": 0.00019101210957086026,
+      "loss": 0.7356550097465515,
+      "step": 2918
+    },
+    {
+      "epoch": 1.2320675105485233,
+      "grad_norm": 0.424675852060318,
+      "learning_rate": 0.00019099219852519733,
+      "loss": 0.6730139255523682,
+      "step": 2920
+    },
+    {
+      "epoch": 1.2329113924050632,
+      "grad_norm": 0.41102367639541626,
+      "learning_rate": 0.0001909722664893817,
+      "loss": 0.6983892321586609,
+      "step": 2922
+    },
+    {
+      "epoch": 1.2337552742616034,
+      "grad_norm": 0.4088250398635864,
+      "learning_rate": 0.00019095231346801132,
+      "loss": 0.6297144889831543,
+      "step": 2924
+    },
+    {
+      "epoch": 1.2345991561181435,
+      "grad_norm": 0.3685890734195709,
+      "learning_rate": 0.00019093233946568906,
+      "loss": 0.7035690546035767,
+      "step": 2926
+    },
+    {
+      "epoch": 1.2354430379746835,
+      "grad_norm": 0.39438629150390625,
+      "learning_rate": 0.0001909123444870224,
+      "loss": 0.820853054523468,
+      "step": 2928
+    },
+    {
+      "epoch": 1.2362869198312236,
+      "grad_norm": 0.5041676163673401,
+      "learning_rate": 0.00019089232853662392,
+      "loss": 0.7884082794189453,
+      "step": 2930
+    },
+    {
+      "epoch": 1.2371308016877638,
+      "grad_norm": 0.40592801570892334,
+      "learning_rate": 0.00019087229161911086,
+      "loss": 0.773978590965271,
+      "step": 2932
+    },
+    {
+      "epoch": 1.2379746835443037,
+      "grad_norm": 0.41828030347824097,
+      "learning_rate": 0.00019085223373910536,
+      "loss": 0.6742141842842102,
+      "step": 2934
+    },
+    {
+      "epoch": 1.238818565400844,
+      "grad_norm": 0.39984026551246643,
+      "learning_rate": 0.0001908321549012344,
+      "loss": 0.7106125950813293,
+      "step": 2936
+    },
+    {
+      "epoch": 1.239662447257384,
+      "grad_norm": 0.4841296970844269,
+      "learning_rate": 0.00019081205511012973,
+      "loss": 0.7390915751457214,
+      "step": 2938
+    },
+    {
+      "epoch": 1.240506329113924,
+      "grad_norm": 0.39837637543678284,
+      "learning_rate": 0.00019079193437042806,
+      "loss": 0.6527198553085327,
+      "step": 2940
+    },
+    {
+      "epoch": 1.2413502109704642,
+      "grad_norm": 0.40916863083839417,
+      "learning_rate": 0.0001907717926867708,
+      "loss": 0.6939712762832642,
+      "step": 2942
+    },
+    {
+      "epoch": 1.2421940928270043,
+      "grad_norm": 0.4329306483268738,
+      "learning_rate": 0.0001907516300638043,
+      "loss": 0.7566131353378296,
+      "step": 2944
+    },
+    {
+      "epoch": 1.2430379746835443,
+      "grad_norm": 0.47983601689338684,
+      "learning_rate": 0.00019073144650617963,
+      "loss": 0.7893748879432678,
+      "step": 2946
+    },
+    {
+      "epoch": 1.2438818565400844,
+      "grad_norm": 0.4368246793746948,
+      "learning_rate": 0.0001907112420185528,
+      "loss": 0.8259525299072266,
+      "step": 2948
+    },
+    {
+      "epoch": 1.2447257383966246,
+      "grad_norm": 0.41272372007369995,
+      "learning_rate": 0.0001906910166055846,
+      "loss": 0.734692394733429,
+      "step": 2950
+    },
+    {
+      "epoch": 1.2455696202531645,
+      "grad_norm": 0.36929839849472046,
+      "learning_rate": 0.00019067077027194057,
+      "loss": 0.7091750502586365,
+      "step": 2952
+    },
+    {
+      "epoch": 1.2464135021097047,
+      "grad_norm": 0.4395252466201782,
+      "learning_rate": 0.00019065050302229125,
+      "loss": 0.6960210204124451,
+      "step": 2954
+    },
+    {
+      "epoch": 1.2472573839662446,
+      "grad_norm": 0.46784868836402893,
+      "learning_rate": 0.00019063021486131186,
+      "loss": 0.7817115187644958,
+      "step": 2956
+    },
+    {
+      "epoch": 1.2481012658227848,
+      "grad_norm": 0.4191588759422302,
+      "learning_rate": 0.00019060990579368244,
+      "loss": 0.7933676242828369,
+      "step": 2958
+    },
+    {
+      "epoch": 1.248945147679325,
+      "grad_norm": 0.3967524468898773,
+      "learning_rate": 0.000190589575824088,
+      "loss": 0.6856688857078552,
+      "step": 2960
+    },
+    {
+      "epoch": 1.249789029535865,
+      "grad_norm": 0.40346091985702515,
+      "learning_rate": 0.0001905692249572182,
+      "loss": 0.6783818602561951,
+      "step": 2962
+    },
+    {
+      "epoch": 1.250632911392405,
+      "grad_norm": 0.38349980115890503,
+      "learning_rate": 0.0001905488531977676,
+      "loss": 0.6997019052505493,
+      "step": 2964
+    },
+    {
+      "epoch": 1.2514767932489452,
+      "grad_norm": 0.44021397829055786,
+      "learning_rate": 0.0001905284605504356,
+      "loss": 0.6978800892829895,
+      "step": 2966
+    },
+    {
+      "epoch": 1.2523206751054852,
+      "grad_norm": 0.47491583228111267,
+      "learning_rate": 0.0001905080470199264,
+      "loss": 0.7007693648338318,
+      "step": 2968
+    },
+    {
+      "epoch": 1.2531645569620253,
+      "grad_norm": 0.4011451005935669,
+      "learning_rate": 0.00019048761261094898,
+      "loss": 0.7258896231651306,
+      "step": 2970
+    },
+    {
+      "epoch": 1.2540084388185653,
+      "grad_norm": 0.44206640124320984,
+      "learning_rate": 0.0001904671573282172,
+      "loss": 0.6746299266815186,
+      "step": 2972
+    },
+    {
+      "epoch": 1.2548523206751054,
+      "grad_norm": 0.4532456696033478,
+      "learning_rate": 0.00019044668117644965,
+      "loss": 0.7638527154922485,
+      "step": 2974
+    },
+    {
+      "epoch": 1.2556962025316456,
+      "grad_norm": 0.44301095604896545,
+      "learning_rate": 0.00019042618416036984,
+      "loss": 0.659598171710968,
+      "step": 2976
+    },
+    {
+      "epoch": 1.2565400843881855,
+      "grad_norm": 0.37427952885627747,
+      "learning_rate": 0.00019040566628470597,
+      "loss": 0.7132709622383118,
+      "step": 2978
+    },
+    {
+      "epoch": 1.2573839662447257,
+      "grad_norm": 0.4195408821105957,
+      "learning_rate": 0.0001903851275541912,
+      "loss": 0.8163022398948669,
+      "step": 2980
+    },
+    {
+      "epoch": 1.2582278481012659,
+      "grad_norm": 0.3951880633831024,
+      "learning_rate": 0.00019036456797356336,
+      "loss": 0.7481505870819092,
+      "step": 2982
+    },
+    {
+      "epoch": 1.2590717299578058,
+      "grad_norm": 0.4239100217819214,
+      "learning_rate": 0.00019034398754756522,
+      "loss": 0.750126838684082,
+      "step": 2984
+    },
+    {
+      "epoch": 1.259915611814346,
+      "grad_norm": 0.4209354817867279,
+      "learning_rate": 0.0001903233862809442,
+      "loss": 0.746433436870575,
+      "step": 2986
+    },
+    {
+      "epoch": 1.2607594936708861,
+      "grad_norm": 0.4188138544559479,
+      "learning_rate": 0.00019030276417845266,
+      "loss": 0.7104454040527344,
+      "step": 2988
+    },
+    {
+      "epoch": 1.261603375527426,
+      "grad_norm": 0.4531371593475342,
+      "learning_rate": 0.00019028212124484772,
+      "loss": 0.650738000869751,
+      "step": 2990
+    },
+    {
+      "epoch": 1.2624472573839662,
+      "grad_norm": 0.4672204256057739,
+      "learning_rate": 0.00019026145748489133,
+      "loss": 0.6870551109313965,
+      "step": 2992
+    },
+    {
+      "epoch": 1.2632911392405064,
+      "grad_norm": 0.40681278705596924,
+      "learning_rate": 0.00019024077290335018,
+      "loss": 0.6879403591156006,
+      "step": 2994
+    },
+    {
+      "epoch": 1.2641350210970463,
+      "grad_norm": 0.4491691291332245,
+      "learning_rate": 0.00019022006750499585,
+      "loss": 0.6886372566223145,
+      "step": 2996
+    },
+    {
+      "epoch": 1.2649789029535865,
+      "grad_norm": 0.4144807457923889,
+      "learning_rate": 0.0001901993412946046,
+      "loss": 0.7199324369430542,
+      "step": 2998
+    },
+    {
+      "epoch": 1.2658227848101267,
+      "grad_norm": 0.40370482206344604,
+      "learning_rate": 0.0001901785942769577,
+      "loss": 0.7105490565299988,
+      "step": 3000
+    },
+    {
+      "epoch": 1.2658227848101267,
+      "eval_loss": 0.8033165335655212,
+      "eval_runtime": 678.7554,
+      "eval_samples_per_second": 3.104,
+      "eval_steps_per_second": 3.104,
+      "step": 3000
+    },
+    {
+      "epoch": 1.2666666666666666,
+      "grad_norm": 0.42637890577316284,
+      "learning_rate": 0.00019015782645684097,
+      "loss": 0.7385403513908386,
+      "step": 3002
+    },
+    {
+      "epoch": 1.2675105485232068,
+      "grad_norm": 0.3709809482097626,
+      "learning_rate": 0.00019013703783904518,
+      "loss": 0.6395129561424255,
+      "step": 3004
+    },
+    {
+      "epoch": 1.268354430379747,
+      "grad_norm": 0.4642501175403595,
+      "learning_rate": 0.00019011622842836592,
+      "loss": 0.7123657464981079,
+      "step": 3006
+    },
+    {
+      "epoch": 1.2691983122362869,
+      "grad_norm": 0.4553911089897156,
+      "learning_rate": 0.00019009539822960342,
+      "loss": 0.744048535823822,
+      "step": 3008
+    },
+    {
+      "epoch": 1.270042194092827,
+      "grad_norm": 0.39334386587142944,
+      "learning_rate": 0.0001900745472475629,
+      "loss": 0.7181914448738098,
+      "step": 3010
+    },
+    {
+      "epoch": 1.2708860759493672,
+      "grad_norm": 0.40902766585350037,
+      "learning_rate": 0.00019005367548705426,
+      "loss": 0.7238107919692993,
+      "step": 3012
+    },
+    {
+      "epoch": 1.2717299578059071,
+      "grad_norm": 0.48296019434928894,
+      "learning_rate": 0.00019003278295289216,
+      "loss": 0.7479744553565979,
+      "step": 3014
+    },
+    {
+      "epoch": 1.2725738396624473,
+      "grad_norm": 0.4807029962539673,
+      "learning_rate": 0.00019001186964989617,
+      "loss": 0.829924464225769,
+      "step": 3016
+    },
+    {
+      "epoch": 1.2734177215189875,
+      "grad_norm": 0.4239083528518677,
+      "learning_rate": 0.00018999093558289055,
+      "loss": 0.7204707264900208,
+      "step": 3018
+    },
+    {
+      "epoch": 1.2742616033755274,
+      "grad_norm": 0.4687928259372711,
+      "learning_rate": 0.0001899699807567044,
+      "loss": 0.7668164968490601,
+      "step": 3020
+    },
+    {
+      "epoch": 1.2751054852320676,
+      "grad_norm": 0.41724732518196106,
+      "learning_rate": 0.0001899490051761716,
+      "loss": 0.766342043876648,
+      "step": 3022
+    },
+    {
+      "epoch": 1.2759493670886077,
+      "grad_norm": 0.4131613075733185,
+      "learning_rate": 0.00018992800884613082,
+      "loss": 0.6595507860183716,
+      "step": 3024
+    },
+    {
+      "epoch": 1.2767932489451477,
+      "grad_norm": 0.4156053364276886,
+      "learning_rate": 0.00018990699177142548,
+      "loss": 0.7419850826263428,
+      "step": 3026
+    },
+    {
+      "epoch": 1.2776371308016878,
+      "grad_norm": 0.41845977306365967,
+      "learning_rate": 0.00018988595395690388,
+      "loss": 0.67264324426651,
+      "step": 3028
+    },
+    {
+      "epoch": 1.2784810126582278,
+      "grad_norm": 0.3826757073402405,
+      "learning_rate": 0.00018986489540741893,
+      "loss": 0.6269281506538391,
+      "step": 3030
+    },
+    {
+      "epoch": 1.279324894514768,
+      "grad_norm": 0.4531428813934326,
+      "learning_rate": 0.00018984381612782854,
+      "loss": 0.71211177110672,
+      "step": 3032
+    },
+    {
+      "epoch": 1.2801687763713079,
+      "grad_norm": 0.3747255802154541,
+      "learning_rate": 0.00018982271612299524,
+      "loss": 0.635804295539856,
+      "step": 3034
+    },
+    {
+      "epoch": 1.281012658227848,
+      "grad_norm": 0.4079552888870239,
+      "learning_rate": 0.00018980159539778638,
+      "loss": 0.7269464731216431,
+      "step": 3036
+    },
+    {
+      "epoch": 1.2818565400843882,
+      "grad_norm": 0.4193854033946991,
+      "learning_rate": 0.00018978045395707418,
+      "loss": 0.7737077474594116,
+      "step": 3038
+    },
+    {
+      "epoch": 1.2827004219409281,
+      "grad_norm": 0.4182516038417816,
+      "learning_rate": 0.0001897592918057355,
+      "loss": 0.7458174824714661,
+      "step": 3040
+    },
+    {
+      "epoch": 1.2835443037974683,
+      "grad_norm": 0.43343567848205566,
+      "learning_rate": 0.00018973810894865205,
+      "loss": 0.6565808057785034,
+      "step": 3042
+    },
+    {
+      "epoch": 1.2843881856540085,
+      "grad_norm": 0.4294503331184387,
+      "learning_rate": 0.00018971690539071033,
+      "loss": 0.736944317817688,
+      "step": 3044
+    },
+    {
+      "epoch": 1.2852320675105484,
+      "grad_norm": 0.41330763697624207,
+      "learning_rate": 0.0001896956811368016,
+      "loss": 0.6921742558479309,
+      "step": 3046
+    },
+    {
+      "epoch": 1.2860759493670886,
+      "grad_norm": 0.44992703199386597,
+      "learning_rate": 0.0001896744361918218,
+      "loss": 0.7004357576370239,
+      "step": 3048
+    },
+    {
+      "epoch": 1.2869198312236287,
+      "grad_norm": 0.43748870491981506,
+      "learning_rate": 0.0001896531705606719,
+      "loss": 0.6767720580101013,
+      "step": 3050
+    },
+    {
+      "epoch": 1.2877637130801687,
+      "grad_norm": 0.47581973671913147,
+      "learning_rate": 0.0001896318842482573,
+      "loss": 0.7596892714500427,
+      "step": 3052
+    },
+    {
+      "epoch": 1.2886075949367088,
+      "grad_norm": 0.45806005597114563,
+      "learning_rate": 0.00018961057725948844,
+      "loss": 0.7650044560432434,
+      "step": 3054
+    },
+    {
+      "epoch": 1.289451476793249,
+      "grad_norm": 0.4291481375694275,
+      "learning_rate": 0.00018958924959928042,
+      "loss": 0.6654610633850098,
+      "step": 3056
+    },
+    {
+      "epoch": 1.290295358649789,
+      "grad_norm": 0.42454418540000916,
+      "learning_rate": 0.00018956790127255309,
+      "loss": 0.7079797387123108,
+      "step": 3058
+    },
+    {
+      "epoch": 1.2911392405063291,
+      "grad_norm": 0.46820637583732605,
+      "learning_rate": 0.00018954653228423114,
+      "loss": 0.7724313735961914,
+      "step": 3060
+    },
+    {
+      "epoch": 1.2919831223628693,
+      "grad_norm": 0.4440016746520996,
+      "learning_rate": 0.00018952514263924396,
+      "loss": 0.7746338248252869,
+      "step": 3062
+    },
+    {
+      "epoch": 1.2928270042194092,
+      "grad_norm": 0.4296419620513916,
+      "learning_rate": 0.00018950373234252572,
+      "loss": 0.7570929527282715,
+      "step": 3064
+    },
+    {
+      "epoch": 1.2936708860759494,
+      "grad_norm": 0.428780734539032,
+      "learning_rate": 0.0001894823013990154,
+      "loss": 0.7274802923202515,
+      "step": 3066
+    },
+    {
+      "epoch": 1.2945147679324895,
+      "grad_norm": 0.4111585021018982,
+      "learning_rate": 0.0001894608498136567,
+      "loss": 0.6595024466514587,
+      "step": 3068
+    },
+    {
+      "epoch": 1.2953586497890295,
+      "grad_norm": 0.4373013973236084,
+      "learning_rate": 0.00018943937759139802,
+      "loss": 0.7316440939903259,
+      "step": 3070
+    },
+    {
+      "epoch": 1.2962025316455696,
+      "grad_norm": 0.37705400586128235,
+      "learning_rate": 0.0001894178847371927,
+      "loss": 0.7072293162345886,
+      "step": 3072
+    },
+    {
+      "epoch": 1.2970464135021098,
+      "grad_norm": 0.4171125888824463,
+      "learning_rate": 0.00018939637125599864,
+      "loss": 0.7218101024627686,
+      "step": 3074
+    },
+    {
+      "epoch": 1.2978902953586497,
+      "grad_norm": 0.4264235198497772,
+      "learning_rate": 0.00018937483715277865,
+      "loss": 0.725140392780304,
+      "step": 3076
+    },
+    {
+      "epoch": 1.29873417721519,
+      "grad_norm": 0.4251643121242523,
+      "learning_rate": 0.0001893532824325002,
+      "loss": 0.7423256635665894,
+      "step": 3078
+    },
+    {
+      "epoch": 1.29957805907173,
+      "grad_norm": 0.43096452951431274,
+      "learning_rate": 0.00018933170710013554,
+      "loss": 0.7058811187744141,
+      "step": 3080
+    },
+    {
+      "epoch": 1.30042194092827,
+      "grad_norm": 0.466804176568985,
+      "learning_rate": 0.00018931011116066172,
+      "loss": 0.7334373593330383,
+      "step": 3082
+    },
+    {
+      "epoch": 1.3012658227848102,
+      "grad_norm": 0.36896654963493347,
+      "learning_rate": 0.00018928849461906047,
+      "loss": 0.6033462285995483,
+      "step": 3084
+    },
+    {
+      "epoch": 1.3021097046413503,
+      "grad_norm": 0.42875027656555176,
+      "learning_rate": 0.00018926685748031833,
+      "loss": 0.690375566482544,
+      "step": 3086
+    },
+    {
+      "epoch": 1.3029535864978903,
+      "grad_norm": 0.4041171967983246,
+      "learning_rate": 0.00018924519974942658,
+      "loss": 0.7468435168266296,
+      "step": 3088
+    },
+    {
+      "epoch": 1.3037974683544304,
+      "grad_norm": 0.4704097807407379,
+      "learning_rate": 0.00018922352143138126,
+      "loss": 0.7574344873428345,
+      "step": 3090
+    },
+    {
+      "epoch": 1.3046413502109704,
+      "grad_norm": 0.4391418397426605,
+      "learning_rate": 0.0001892018225311831,
+      "loss": 0.756452739238739,
+      "step": 3092
+    },
+    {
+      "epoch": 1.3054852320675105,
+      "grad_norm": 0.42684388160705566,
+      "learning_rate": 0.0001891801030538376,
+      "loss": 0.6620991230010986,
+      "step": 3094
+    },
+    {
+      "epoch": 1.3063291139240507,
+      "grad_norm": 0.4972277283668518,
+      "learning_rate": 0.00018915836300435507,
+      "loss": 0.7764125466346741,
+      "step": 3096
+    },
+    {
+      "epoch": 1.3071729957805907,
+      "grad_norm": 0.3905445337295532,
+      "learning_rate": 0.00018913660238775053,
+      "loss": 0.6557663083076477,
+      "step": 3098
+    },
+    {
+      "epoch": 1.3080168776371308,
+      "grad_norm": 0.4553062319755554,
+      "learning_rate": 0.0001891148212090437,
+      "loss": 0.6746718883514404,
+      "step": 3100
+    },
+    {
+      "epoch": 1.3080168776371308,
+      "eval_loss": 0.8010181784629822,
+      "eval_runtime": 668.1688,
+      "eval_samples_per_second": 3.153,
+      "eval_steps_per_second": 3.153,
+      "step": 3100
+    },
+    {
+      "epoch": 1.3088607594936708,
+      "grad_norm": 0.4254429340362549,
+      "learning_rate": 0.00018909301947325915,
+      "loss": 0.7916419506072998,
+      "step": 3102
+    },
+    {
+      "epoch": 1.309704641350211,
+      "grad_norm": 0.6118508577346802,
+      "learning_rate": 0.00018907119718542602,
+      "loss": 0.883936882019043,
+      "step": 3104
+    },
+    {
+      "epoch": 1.310548523206751,
+      "grad_norm": 0.43206971883773804,
+      "learning_rate": 0.00018904935435057837,
+      "loss": 0.7145928740501404,
+      "step": 3106
+    },
+    {
+      "epoch": 1.311392405063291,
+      "grad_norm": 0.39657703042030334,
+      "learning_rate": 0.0001890274909737549,
+      "loss": 0.7442842125892639,
+      "step": 3108
+    },
+    {
+      "epoch": 1.3122362869198312,
+      "grad_norm": 0.4087927043437958,
+      "learning_rate": 0.00018900560705999904,
+      "loss": 0.6815276145935059,
+      "step": 3110
+    },
+    {
+      "epoch": 1.3130801687763713,
+      "grad_norm": 0.430259108543396,
+      "learning_rate": 0.00018898370261435904,
+      "loss": 0.8153640627861023,
+      "step": 3112
+    },
+    {
+      "epoch": 1.3139240506329113,
+      "grad_norm": 0.47486868500709534,
+      "learning_rate": 0.0001889617776418878,
+      "loss": 0.7828266024589539,
+      "step": 3114
+    },
+    {
+      "epoch": 1.3147679324894515,
+      "grad_norm": 0.4929087162017822,
+      "learning_rate": 0.000188939832147643,
+      "loss": 0.7849065661430359,
+      "step": 3116
+    },
+    {
+      "epoch": 1.3156118143459916,
+      "grad_norm": 0.39038679003715515,
+      "learning_rate": 0.00018891786613668704,
+      "loss": 0.7262985706329346,
+      "step": 3118
+    },
+    {
+      "epoch": 1.3164556962025316,
+      "grad_norm": 0.4827214479446411,
+      "learning_rate": 0.0001888958796140871,
+      "loss": 0.8121856451034546,
+      "step": 3120
+    },
+    {
+      "epoch": 1.3172995780590717,
+      "grad_norm": 0.47739601135253906,
+      "learning_rate": 0.000188873872584915,
+      "loss": 0.7165566086769104,
+      "step": 3122
+    },
+    {
+      "epoch": 1.3181434599156119,
+      "grad_norm": 0.4274037778377533,
+      "learning_rate": 0.0001888518450542473,
+      "loss": 0.6698412299156189,
+      "step": 3124
+    },
+    {
+      "epoch": 1.3189873417721518,
+      "grad_norm": 0.44825276732444763,
+      "learning_rate": 0.00018882979702716544,
+      "loss": 0.7271128296852112,
+      "step": 3126
+    },
+    {
+      "epoch": 1.319831223628692,
+      "grad_norm": 0.42956918478012085,
+      "learning_rate": 0.00018880772850875537,
+      "loss": 0.7924301624298096,
+      "step": 3128
+    },
+    {
+      "epoch": 1.3206751054852321,
+      "grad_norm": 0.4848136305809021,
+      "learning_rate": 0.00018878563950410792,
+      "loss": 0.7846497893333435,
+      "step": 3130
+    },
+    {
+      "epoch": 1.321518987341772,
+      "grad_norm": 0.46651726961135864,
+      "learning_rate": 0.00018876353001831864,
+      "loss": 0.7262214422225952,
+      "step": 3132
+    },
+    {
+      "epoch": 1.3223628691983123,
+      "grad_norm": 0.44598284363746643,
+      "learning_rate": 0.0001887414000564877,
+      "loss": 0.7640091180801392,
+      "step": 3134
+    },
+    {
+      "epoch": 1.3232067510548524,
+      "grad_norm": 0.43744903802871704,
+      "learning_rate": 0.00018871924962372006,
+      "loss": 0.6960080862045288,
+      "step": 3136
+    },
+    {
+      "epoch": 1.3240506329113924,
+      "grad_norm": 0.4737822413444519,
+      "learning_rate": 0.00018869707872512544,
+      "loss": 0.7121244072914124,
+      "step": 3138
+    },
+    {
+      "epoch": 1.3248945147679325,
+      "grad_norm": 0.44452038407325745,
+      "learning_rate": 0.00018867488736581817,
+      "loss": 0.713148295879364,
+      "step": 3140
+    },
+    {
+      "epoch": 1.3257383966244727,
+      "grad_norm": 0.47142693400382996,
+      "learning_rate": 0.0001886526755509175,
+      "loss": 0.6904974579811096,
+      "step": 3142
+    },
+    {
+      "epoch": 1.3265822784810126,
+      "grad_norm": 0.40572887659072876,
+      "learning_rate": 0.00018863044328554712,
+      "loss": 0.7319386005401611,
+      "step": 3144
+    },
+    {
+      "epoch": 1.3274261603375528,
+      "grad_norm": 0.44173532724380493,
+      "learning_rate": 0.00018860819057483566,
+      "loss": 0.7268120646476746,
+      "step": 3146
+    },
+    {
+      "epoch": 1.328270042194093,
+      "grad_norm": 0.43025046586990356,
+      "learning_rate": 0.00018858591742391642,
+      "loss": 0.7129524946212769,
+      "step": 3148
+    },
+    {
+      "epoch": 1.3291139240506329,
+      "grad_norm": 0.41418036818504333,
+      "learning_rate": 0.00018856362383792732,
+      "loss": 0.6917402148246765,
+      "step": 3150
+    },
+    {
+      "epoch": 1.329957805907173,
+      "grad_norm": 0.46275854110717773,
+      "learning_rate": 0.00018854130982201111,
+      "loss": 0.7653918862342834,
+      "step": 3152
+    },
+    {
+      "epoch": 1.3308016877637132,
+      "grad_norm": 0.38927653431892395,
+      "learning_rate": 0.0001885189753813152,
+      "loss": 0.7311477065086365,
+      "step": 3154
+    },
+    {
+      "epoch": 1.3316455696202532,
+      "grad_norm": 0.44294974207878113,
+      "learning_rate": 0.0001884966205209917,
+      "loss": 0.7978816628456116,
+      "step": 3156
+    },
+    {
+      "epoch": 1.3324894514767933,
+      "grad_norm": 0.3772844970226288,
+      "learning_rate": 0.00018847424524619745,
+      "loss": 0.7497738599777222,
+      "step": 3158
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.4538787603378296,
+      "learning_rate": 0.000188451849562094,
+      "loss": 0.712806761264801,
+      "step": 3160
+    },
+    {
+      "epoch": 1.3341772151898734,
+      "grad_norm": 0.4194336533546448,
+      "learning_rate": 0.00018842943347384763,
+      "loss": 0.7282580137252808,
+      "step": 3162
+    },
+    {
+      "epoch": 1.3350210970464136,
+      "grad_norm": 0.4186438024044037,
+      "learning_rate": 0.00018840699698662925,
+      "loss": 0.797752320766449,
+      "step": 3164
+    },
+    {
+      "epoch": 1.3358649789029535,
+      "grad_norm": 0.47711506485939026,
+      "learning_rate": 0.00018838454010561457,
+      "loss": 0.8175379633903503,
+      "step": 3166
+    },
+    {
+      "epoch": 1.3367088607594937,
+      "grad_norm": 0.4380176067352295,
+      "learning_rate": 0.00018836206283598394,
+      "loss": 0.659176766872406,
+      "step": 3168
+    },
+    {
+      "epoch": 1.3375527426160336,
+      "grad_norm": 0.4505249559879303,
+      "learning_rate": 0.00018833956518292245,
+      "loss": 0.732314944267273,
+      "step": 3170
+    },
+    {
+      "epoch": 1.3383966244725738,
+      "grad_norm": 0.3832913935184479,
+      "learning_rate": 0.00018831704715161984,
+      "loss": 0.6791377067565918,
+      "step": 3172
+    },
+    {
+      "epoch": 1.339240506329114,
+      "grad_norm": 0.42996498942375183,
+      "learning_rate": 0.00018829450874727064,
+      "loss": 0.7395796775817871,
+      "step": 3174
+    },
+    {
+      "epoch": 1.340084388185654,
+      "grad_norm": 0.3926190435886383,
+      "learning_rate": 0.000188271949975074,
+      "loss": 0.6893631219863892,
+      "step": 3176
+    },
+    {
+      "epoch": 1.340928270042194,
+      "grad_norm": 0.4353938102722168,
+      "learning_rate": 0.0001882493708402338,
+      "loss": 0.6708382964134216,
+      "step": 3178
+    },
+    {
+      "epoch": 1.3417721518987342,
+      "grad_norm": 0.43043214082717896,
+      "learning_rate": 0.0001882267713479586,
+      "loss": 0.8447695374488831,
+      "step": 3180
+    },
+    {
+      "epoch": 1.3426160337552742,
+      "grad_norm": 0.40693536400794983,
+      "learning_rate": 0.00018820415150346163,
+      "loss": 0.7081091403961182,
+      "step": 3182
+    },
+    {
+      "epoch": 1.3434599156118143,
+      "grad_norm": 0.45037057995796204,
+      "learning_rate": 0.00018818151131196097,
+      "loss": 0.6822336316108704,
+      "step": 3184
+    },
+    {
+      "epoch": 1.3443037974683545,
+      "grad_norm": 0.3731948137283325,
+      "learning_rate": 0.00018815885077867915,
+      "loss": 0.6316909790039062,
+      "step": 3186
+    },
+    {
+      "epoch": 1.3451476793248944,
+      "grad_norm": 0.4248095750808716,
+      "learning_rate": 0.00018813616990884363,
+      "loss": 0.7452375292778015,
+      "step": 3188
+    },
+    {
+      "epoch": 1.3459915611814346,
+      "grad_norm": 0.39229142665863037,
+      "learning_rate": 0.00018811346870768638,
+      "loss": 0.6493782997131348,
+      "step": 3190
+    },
+    {
+      "epoch": 1.3468354430379748,
+      "grad_norm": 0.45015549659729004,
+      "learning_rate": 0.00018809074718044414,
+      "loss": 0.735163688659668,
+      "step": 3192
+    },
+    {
+      "epoch": 1.3476793248945147,
+      "grad_norm": 0.39756354689598083,
+      "learning_rate": 0.00018806800533235836,
+      "loss": 0.6656006574630737,
+      "step": 3194
+    },
+    {
+      "epoch": 1.3485232067510549,
+      "grad_norm": 0.4410010576248169,
+      "learning_rate": 0.0001880452431686751,
+      "loss": 0.7040987014770508,
+      "step": 3196
+    },
+    {
+      "epoch": 1.349367088607595,
+      "grad_norm": 0.48144352436065674,
+      "learning_rate": 0.0001880224606946452,
+      "loss": 0.7722331285476685,
+      "step": 3198
+    },
+    {
+      "epoch": 1.350210970464135,
+      "grad_norm": 0.38644957542419434,
+      "learning_rate": 0.00018799965791552414,
+      "loss": 0.6681689023971558,
+      "step": 3200
+    },
+    {
+      "epoch": 1.350210970464135,
+      "eval_loss": 0.797160804271698,
+      "eval_runtime": 680.976,
+      "eval_samples_per_second": 3.094,
+      "eval_steps_per_second": 3.094,
+      "step": 3200
+    },
+    {
+      "epoch": 1.3510548523206751,
+      "grad_norm": 0.4471014440059662,
+      "learning_rate": 0.00018797683483657201,
+      "loss": 0.6995527744293213,
+      "step": 3202
+    },
+    {
+      "epoch": 1.3518987341772153,
+      "grad_norm": 0.4065002202987671,
+      "learning_rate": 0.00018795399146305378,
+      "loss": 0.659622848033905,
+      "step": 3204
+    },
+    {
+      "epoch": 1.3527426160337552,
+      "grad_norm": 0.4787910580635071,
+      "learning_rate": 0.00018793112780023888,
+      "loss": 0.7927116751670837,
+      "step": 3206
+    },
+    {
+      "epoch": 1.3535864978902954,
+      "grad_norm": 0.43351876735687256,
+      "learning_rate": 0.0001879082438534015,
+      "loss": 0.6738835573196411,
+      "step": 3208
+    },
+    {
+      "epoch": 1.3544303797468356,
+      "grad_norm": 0.4275745749473572,
+      "learning_rate": 0.00018788533962782063,
+      "loss": 0.7785354852676392,
+      "step": 3210
+    },
+    {
+      "epoch": 1.3552742616033755,
+      "grad_norm": 0.4506145119667053,
+      "learning_rate": 0.00018786241512877975,
+      "loss": 0.7325424551963806,
+      "step": 3212
+    },
+    {
+      "epoch": 1.3561181434599157,
+      "grad_norm": 0.42040735483169556,
+      "learning_rate": 0.00018783947036156713,
+      "loss": 0.6898744702339172,
+      "step": 3214
+    },
+    {
+      "epoch": 1.3569620253164558,
+      "grad_norm": 0.4650438129901886,
+      "learning_rate": 0.0001878165053314757,
+      "loss": 0.7498815655708313,
+      "step": 3216
+    },
+    {
+      "epoch": 1.3578059071729958,
+      "grad_norm": 0.4194130599498749,
+      "learning_rate": 0.000187793520043803,
+      "loss": 0.6891883611679077,
+      "step": 3218
+    },
+    {
+      "epoch": 1.358649789029536,
+      "grad_norm": 0.4447539746761322,
+      "learning_rate": 0.0001877705145038514,
+      "loss": 0.753805935382843,
+      "step": 3220
+    },
+    {
+      "epoch": 1.3594936708860759,
+      "grad_norm": 0.4047110974788666,
+      "learning_rate": 0.00018774748871692773,
+      "loss": 0.6476293802261353,
+      "step": 3222
+    },
+    {
+      "epoch": 1.360337552742616,
+      "grad_norm": 0.39818957448005676,
+      "learning_rate": 0.00018772444268834364,
+      "loss": 0.6765931844711304,
+      "step": 3224
+    },
+    {
+      "epoch": 1.3611814345991562,
+      "grad_norm": 0.41843733191490173,
+      "learning_rate": 0.0001877013764234154,
+      "loss": 0.7360382676124573,
+      "step": 3226
+    },
+    {
+      "epoch": 1.3620253164556961,
+      "grad_norm": 0.41397789120674133,
+      "learning_rate": 0.00018767828992746394,
+      "loss": 0.7149524688720703,
+      "step": 3228
+    },
+    {
+      "epoch": 1.3628691983122363,
+      "grad_norm": 0.43585047125816345,
+      "learning_rate": 0.00018765518320581492,
+      "loss": 0.6932925581932068,
+      "step": 3230
+    },
+    {
+      "epoch": 1.3637130801687762,
+      "grad_norm": 0.3651057481765747,
+      "learning_rate": 0.00018763205626379857,
+      "loss": 0.6561182141304016,
+      "step": 3232
+    },
+    {
+      "epoch": 1.3645569620253164,
+      "grad_norm": 0.44937077164649963,
+      "learning_rate": 0.00018760890910674984,
+      "loss": 0.7436831593513489,
+      "step": 3234
+    },
+    {
+      "epoch": 1.3654008438818566,
+      "grad_norm": 0.5045920014381409,
+      "learning_rate": 0.00018758574174000832,
+      "loss": 0.7849522233009338,
+      "step": 3236
+    },
+    {
+      "epoch": 1.3662447257383965,
+      "grad_norm": 0.4625580906867981,
+      "learning_rate": 0.00018756255416891833,
+      "loss": 0.7501618266105652,
+      "step": 3238
+    },
+    {
+      "epoch": 1.3670886075949367,
+      "grad_norm": 0.3721269369125366,
+      "learning_rate": 0.00018753934639882875,
+      "loss": 0.7656083106994629,
+      "step": 3240
+    },
+    {
+      "epoch": 1.3679324894514768,
+      "grad_norm": 0.41721680760383606,
+      "learning_rate": 0.00018751611843509318,
+      "loss": 0.7816168069839478,
+      "step": 3242
+    },
+    {
+      "epoch": 1.3687763713080168,
+      "grad_norm": 0.8452199101448059,
+      "learning_rate": 0.00018749287028306988,
+      "loss": 0.7002755403518677,
+      "step": 3244
+    },
+    {
+      "epoch": 1.369620253164557,
+      "grad_norm": 0.4320507347583771,
+      "learning_rate": 0.00018746960194812172,
+      "loss": 0.7358270883560181,
+      "step": 3246
+    },
+    {
+      "epoch": 1.370464135021097,
+      "grad_norm": 0.43323323130607605,
+      "learning_rate": 0.00018744631343561627,
+      "loss": 0.6528644561767578,
+      "step": 3248
+    },
+    {
+      "epoch": 1.371308016877637,
+      "grad_norm": 0.41487041115760803,
+      "learning_rate": 0.0001874230047509258,
+      "loss": 0.7317517995834351,
+      "step": 3250
+    },
+    {
+      "epoch": 1.3721518987341772,
+      "grad_norm": 0.5085384249687195,
+      "learning_rate": 0.00018739967589942707,
+      "loss": 0.8236643671989441,
+      "step": 3252
+    },
+    {
+      "epoch": 1.3729957805907174,
+      "grad_norm": 0.4126952290534973,
+      "learning_rate": 0.00018737632688650168,
+      "loss": 0.730254054069519,
+      "step": 3254
+    },
+    {
+      "epoch": 1.3738396624472573,
+      "grad_norm": 0.5272142291069031,
+      "learning_rate": 0.00018735295771753573,
+      "loss": 0.7637890577316284,
+      "step": 3256
+    },
+    {
+      "epoch": 1.3746835443037975,
+      "grad_norm": 0.4319530725479126,
+      "learning_rate": 0.00018732956839792012,
+      "loss": 0.7906988263130188,
+      "step": 3258
+    },
+    {
+      "epoch": 1.3755274261603376,
+      "grad_norm": 0.4393143355846405,
+      "learning_rate": 0.00018730615893305025,
+      "loss": 0.7261547446250916,
+      "step": 3260
+    },
+    {
+      "epoch": 1.3763713080168776,
+      "grad_norm": 0.4161636233329773,
+      "learning_rate": 0.00018728272932832632,
+      "loss": 0.6857497692108154,
+      "step": 3262
+    },
+    {
+      "epoch": 1.3772151898734177,
+      "grad_norm": 0.48862120509147644,
+      "learning_rate": 0.00018725927958915297,
+      "loss": 0.7488857507705688,
+      "step": 3264
+    },
+    {
+      "epoch": 1.378059071729958,
+      "grad_norm": 0.4422304928302765,
+      "learning_rate": 0.0001872358097209397,
+      "loss": 0.7257721424102783,
+      "step": 3266
+    },
+    {
+      "epoch": 1.3789029535864978,
+      "grad_norm": 0.4577067494392395,
+      "learning_rate": 0.00018721231972910048,
+      "loss": 0.746697187423706,
+      "step": 3268
+    },
+    {
+      "epoch": 1.379746835443038,
+      "grad_norm": 0.42382001876831055,
+      "learning_rate": 0.00018718880961905406,
+      "loss": 0.768519401550293,
+      "step": 3270
+    },
+    {
+      "epoch": 1.3805907172995782,
+      "grad_norm": 0.43368014693260193,
+      "learning_rate": 0.00018716527939622378,
+      "loss": 0.7426102757453918,
+      "step": 3272
+    },
+    {
+      "epoch": 1.381434599156118,
+      "grad_norm": 0.3777220547199249,
+      "learning_rate": 0.00018714172906603757,
+      "loss": 0.7104197144508362,
+      "step": 3274
+    },
+    {
+      "epoch": 1.3822784810126583,
+      "grad_norm": 0.46071696281433105,
+      "learning_rate": 0.00018711815863392802,
+      "loss": 0.7502968907356262,
+      "step": 3276
+    },
+    {
+      "epoch": 1.3831223628691984,
+      "grad_norm": 0.42510855197906494,
+      "learning_rate": 0.00018709456810533245,
+      "loss": 0.6400540471076965,
+      "step": 3278
+    },
+    {
+      "epoch": 1.3839662447257384,
+      "grad_norm": 0.4217228293418884,
+      "learning_rate": 0.00018707095748569268,
+      "loss": 0.6648687720298767,
+      "step": 3280
+    },
+    {
+      "epoch": 1.3848101265822785,
+      "grad_norm": 0.48267799615859985,
+      "learning_rate": 0.00018704732678045526,
+      "loss": 0.7176171541213989,
+      "step": 3282
+    },
+    {
+      "epoch": 1.3856540084388187,
+      "grad_norm": 0.47114965319633484,
+      "learning_rate": 0.00018702367599507134,
+      "loss": 0.8402457237243652,
+      "step": 3284
+    },
+    {
+      "epoch": 1.3864978902953586,
+      "grad_norm": 0.5054492354393005,
+      "learning_rate": 0.00018700000513499667,
+      "loss": 0.8130836486816406,
+      "step": 3286
+    },
+    {
+      "epoch": 1.3873417721518988,
+      "grad_norm": 0.39435645937919617,
+      "learning_rate": 0.00018697631420569168,
+      "loss": 0.6432579755783081,
+      "step": 3288
+    },
+    {
+      "epoch": 1.3881856540084387,
+      "grad_norm": 0.3997228443622589,
+      "learning_rate": 0.00018695260321262142,
+      "loss": 0.6921643614768982,
+      "step": 3290
+    },
+    {
+      "epoch": 1.389029535864979,
+      "grad_norm": 0.37304237484931946,
+      "learning_rate": 0.00018692887216125557,
+      "loss": 0.7162153124809265,
+      "step": 3292
+    },
+    {
+      "epoch": 1.389873417721519,
+      "grad_norm": 0.7997981309890747,
+      "learning_rate": 0.00018690512105706838,
+      "loss": 0.7194803953170776,
+      "step": 3294
+    },
+    {
+      "epoch": 1.390717299578059,
+      "grad_norm": 0.41268959641456604,
+      "learning_rate": 0.00018688134990553883,
+      "loss": 0.6888322830200195,
+      "step": 3296
+    },
+    {
+      "epoch": 1.3915611814345992,
+      "grad_norm": 0.4485826790332794,
+      "learning_rate": 0.00018685755871215043,
+      "loss": 0.7387221455574036,
+      "step": 3298
+    },
+    {
+      "epoch": 1.3924050632911391,
+      "grad_norm": 0.47835296392440796,
+      "learning_rate": 0.0001868337474823914,
+      "loss": 0.8061952590942383,
+      "step": 3300
+    },
+    {
+      "epoch": 1.3924050632911391,
+      "eval_loss": 0.795619547367096,
+      "eval_runtime": 692.7157,
+      "eval_samples_per_second": 3.042,
+      "eval_steps_per_second": 3.042,
+      "step": 3300
+    },
+    {
+      "epoch": 1.3932489451476793,
+      "grad_norm": 0.4024396240711212,
+      "learning_rate": 0.00018680991622175446,
+      "loss": 0.7317537069320679,
+      "step": 3302
+    },
+    {
+      "epoch": 1.3940928270042194,
+      "grad_norm": 0.4526490271091461,
+      "learning_rate": 0.00018678606493573708,
+      "loss": 0.7470992207527161,
+      "step": 3304
+    },
+    {
+      "epoch": 1.3949367088607594,
+      "grad_norm": 0.5055129528045654,
+      "learning_rate": 0.00018676219362984126,
+      "loss": 0.7627943754196167,
+      "step": 3306
+    },
+    {
+      "epoch": 1.3957805907172995,
+      "grad_norm": 0.4030216634273529,
+      "learning_rate": 0.0001867383023095737,
+      "loss": 0.684033989906311,
+      "step": 3308
+    },
+    {
+      "epoch": 1.3966244725738397,
+      "grad_norm": 0.4353308379650116,
+      "learning_rate": 0.0001867143909804456,
+      "loss": 0.7366722822189331,
+      "step": 3310
+    },
+    {
+      "epoch": 1.3974683544303796,
+      "grad_norm": 0.5360553860664368,
+      "learning_rate": 0.0001866904596479729,
+      "loss": 0.8050951957702637,
+      "step": 3312
+    },
+    {
+      "epoch": 1.3983122362869198,
+      "grad_norm": 0.46081677079200745,
+      "learning_rate": 0.00018666650831767608,
+      "loss": 0.7227963805198669,
+      "step": 3314
+    },
+    {
+      "epoch": 1.39915611814346,
+      "grad_norm": 0.4325845539569855,
+      "learning_rate": 0.00018664253699508029,
+      "loss": 0.714216411113739,
+      "step": 3316
+    },
+    {
+      "epoch": 1.4,
+      "grad_norm": 0.4373100996017456,
+      "learning_rate": 0.0001866185456857152,
+      "loss": 0.7378593683242798,
+      "step": 3318
+    },
+    {
+      "epoch": 1.40084388185654,
+      "grad_norm": 0.43221279978752136,
+      "learning_rate": 0.00018659453439511512,
+      "loss": 0.7565981149673462,
+      "step": 3320
+    },
+    {
+      "epoch": 1.4016877637130802,
+      "grad_norm": 0.3938370943069458,
+      "learning_rate": 0.00018657050312881904,
+      "loss": 0.6409488916397095,
+      "step": 3322
+    },
+    {
+      "epoch": 1.4025316455696202,
+      "grad_norm": 0.43609440326690674,
+      "learning_rate": 0.00018654645189237055,
+      "loss": 0.6911652088165283,
+      "step": 3324
+    },
+    {
+      "epoch": 1.4033755274261603,
+      "grad_norm": 0.4526350200176239,
+      "learning_rate": 0.00018652238069131774,
+      "loss": 0.7275710701942444,
+      "step": 3326
+    },
+    {
+      "epoch": 1.4042194092827005,
+      "grad_norm": 0.3990516662597656,
+      "learning_rate": 0.0001864982895312134,
+      "loss": 0.6466895341873169,
+      "step": 3328
+    },
+    {
+      "epoch": 1.4050632911392404,
+      "grad_norm": 0.44321122765541077,
+      "learning_rate": 0.0001864741784176149,
+      "loss": 0.6520540714263916,
+      "step": 3330
+    },
+    {
+      "epoch": 1.4059071729957806,
+      "grad_norm": 0.44733941555023193,
+      "learning_rate": 0.0001864500473560842,
+      "loss": 0.7289671301841736,
+      "step": 3332
+    },
+    {
+      "epoch": 1.4067510548523208,
+      "grad_norm": 0.4569414556026459,
+      "learning_rate": 0.00018642589635218789,
+      "loss": 0.7664391398429871,
+      "step": 3334
+    },
+    {
+      "epoch": 1.4075949367088607,
+      "grad_norm": 0.4172186851501465,
+      "learning_rate": 0.00018640172541149709,
+      "loss": 0.7747462391853333,
+      "step": 3336
+    },
+    {
+      "epoch": 1.4084388185654009,
+      "grad_norm": 0.4540993869304657,
+      "learning_rate": 0.0001863775345395877,
+      "loss": 0.8337607383728027,
+      "step": 3338
+    },
+    {
+      "epoch": 1.409282700421941,
+      "grad_norm": 0.4306154251098633,
+      "learning_rate": 0.00018635332374203993,
+      "loss": 0.6851577758789062,
+      "step": 3340
+    },
+    {
+      "epoch": 1.410126582278481,
+      "grad_norm": 0.45615455508232117,
+      "learning_rate": 0.00018632909302443884,
+      "loss": 0.6841398477554321,
+      "step": 3342
+    },
+    {
+      "epoch": 1.4109704641350211,
+      "grad_norm": 0.35296666622161865,
+      "learning_rate": 0.00018630484239237397,
+      "loss": 0.6672312021255493,
+      "step": 3344
+    },
+    {
+      "epoch": 1.4118143459915613,
+      "grad_norm": 0.4424770176410675,
+      "learning_rate": 0.00018628057185143945,
+      "loss": 0.7027004361152649,
+      "step": 3346
+    },
+    {
+      "epoch": 1.4126582278481012,
+      "grad_norm": 0.429817259311676,
+      "learning_rate": 0.0001862562814072341,
+      "loss": 0.7397928237915039,
+      "step": 3348
+    },
+    {
+      "epoch": 1.4135021097046414,
+      "grad_norm": 0.4720051884651184,
+      "learning_rate": 0.00018623197106536118,
+      "loss": 0.8009215593338013,
+      "step": 3350
+    },
+    {
+      "epoch": 1.4143459915611816,
+      "grad_norm": 0.42197802662849426,
+      "learning_rate": 0.0001862076408314287,
+      "loss": 0.6875046491622925,
+      "step": 3352
+    },
+    {
+      "epoch": 1.4151898734177215,
+      "grad_norm": 0.3674047589302063,
+      "learning_rate": 0.00018618329071104907,
+      "loss": 0.7202261090278625,
+      "step": 3354
+    },
+    {
+      "epoch": 1.4160337552742617,
+      "grad_norm": 0.42327889800071716,
+      "learning_rate": 0.0001861589207098395,
+      "loss": 0.6981248259544373,
+      "step": 3356
+    },
+    {
+      "epoch": 1.4168776371308016,
+      "grad_norm": 0.40644150972366333,
+      "learning_rate": 0.00018613453083342162,
+      "loss": 0.7081356048583984,
+      "step": 3358
+    },
+    {
+      "epoch": 1.4177215189873418,
+      "grad_norm": 0.3948482275009155,
+      "learning_rate": 0.00018611012108742168,
+      "loss": 0.6420091986656189,
+      "step": 3360
+    },
+    {
+      "epoch": 1.4185654008438817,
+      "grad_norm": 0.44622674584388733,
+      "learning_rate": 0.00018608569147747064,
+      "loss": 0.7662023305892944,
+      "step": 3362
+    },
+    {
+      "epoch": 1.4194092827004219,
+      "grad_norm": 0.39772045612335205,
+      "learning_rate": 0.00018606124200920386,
+      "loss": 0.7243877649307251,
+      "step": 3364
+    },
+    {
+      "epoch": 1.420253164556962,
+      "grad_norm": 0.4284326136112213,
+      "learning_rate": 0.00018603677268826138,
+      "loss": 0.7399296164512634,
+      "step": 3366
+    },
+    {
+      "epoch": 1.421097046413502,
+      "grad_norm": 0.3749450445175171,
+      "learning_rate": 0.0001860122835202878,
+      "loss": 0.6978716850280762,
+      "step": 3368
+    },
+    {
+      "epoch": 1.4219409282700421,
+      "grad_norm": 0.46721795201301575,
+      "learning_rate": 0.00018598777451093234,
+      "loss": 0.7733059525489807,
+      "step": 3370
+    },
+    {
+      "epoch": 1.4227848101265823,
+      "grad_norm": 0.39606085419654846,
+      "learning_rate": 0.0001859632456658487,
+      "loss": 0.6721867918968201,
+      "step": 3372
+    },
+    {
+      "epoch": 1.4236286919831223,
+      "grad_norm": 0.3879879117012024,
+      "learning_rate": 0.00018593869699069527,
+      "loss": 0.715825080871582,
+      "step": 3374
+    },
+    {
+      "epoch": 1.4244725738396624,
+      "grad_norm": 0.3641676902770996,
+      "learning_rate": 0.0001859141284911349,
+      "loss": 0.6226941347122192,
+      "step": 3376
+    },
+    {
+      "epoch": 1.4253164556962026,
+      "grad_norm": 0.4612879753112793,
+      "learning_rate": 0.0001858895401728351,
+      "loss": 0.7972121238708496,
+      "step": 3378
+    },
+    {
+      "epoch": 1.4261603375527425,
+      "grad_norm": 0.4406103193759918,
+      "learning_rate": 0.00018586493204146795,
+      "loss": 0.6663658022880554,
+      "step": 3380
+    },
+    {
+      "epoch": 1.4270042194092827,
+      "grad_norm": 0.420205295085907,
+      "learning_rate": 0.00018584030410271,
+      "loss": 0.6867321133613586,
+      "step": 3382
+    },
+    {
+      "epoch": 1.4278481012658228,
+      "grad_norm": 0.4499560594558716,
+      "learning_rate": 0.00018581565636224253,
+      "loss": 0.66688072681427,
+      "step": 3384
+    },
+    {
+      "epoch": 1.4286919831223628,
+      "grad_norm": 0.42507919669151306,
+      "learning_rate": 0.00018579098882575123,
+      "loss": 0.6714702844619751,
+      "step": 3386
+    },
+    {
+      "epoch": 1.429535864978903,
+      "grad_norm": 0.43856513500213623,
+      "learning_rate": 0.0001857663014989265,
+      "loss": 0.7219351530075073,
+      "step": 3388
+    },
+    {
+      "epoch": 1.4303797468354431,
+      "grad_norm": 0.5167419910430908,
+      "learning_rate": 0.00018574159438746314,
+      "loss": 0.7157991528511047,
+      "step": 3390
+    },
+    {
+      "epoch": 1.431223628691983,
+      "grad_norm": 0.4443167746067047,
+      "learning_rate": 0.00018571686749706067,
+      "loss": 0.7509840726852417,
+      "step": 3392
+    },
+    {
+      "epoch": 1.4320675105485232,
+      "grad_norm": 0.41321757435798645,
+      "learning_rate": 0.0001856921208334231,
+      "loss": 0.756000816822052,
+      "step": 3394
+    },
+    {
+      "epoch": 1.4329113924050634,
+      "grad_norm": 0.4094153046607971,
+      "learning_rate": 0.000185667354402259,
+      "loss": 0.7047854065895081,
+      "step": 3396
+    },
+    {
+      "epoch": 1.4337552742616033,
+      "grad_norm": 0.4248482286930084,
+      "learning_rate": 0.0001856425682092815,
+      "loss": 0.7620865106582642,
+      "step": 3398
+    },
+    {
+      "epoch": 1.4345991561181435,
+      "grad_norm": 0.4190550148487091,
+      "learning_rate": 0.0001856177622602083,
+      "loss": 0.7441845536231995,
+      "step": 3400
+    },
+    {
+      "epoch": 1.4345991561181435,
+      "eval_loss": 0.7917885780334473,
+      "eval_runtime": 686.1689,
+      "eval_samples_per_second": 3.071,
+      "eval_steps_per_second": 3.071,
+      "step": 3400
+    },
+    {
+      "epoch": 1.4354430379746836,
+      "grad_norm": 0.45602327585220337,
+      "learning_rate": 0.00018559293656076166,
+      "loss": 0.7123498916625977,
+      "step": 3402
+    },
+    {
+      "epoch": 1.4362869198312236,
+      "grad_norm": 0.5209269523620605,
+      "learning_rate": 0.00018556809111666843,
+      "loss": 0.7096527814865112,
+      "step": 3404
+    },
+    {
+      "epoch": 1.4371308016877637,
+      "grad_norm": 0.44521182775497437,
+      "learning_rate": 0.00018554322593365995,
+      "loss": 0.7116838693618774,
+      "step": 3406
+    },
+    {
+      "epoch": 1.437974683544304,
+      "grad_norm": 0.44571003317832947,
+      "learning_rate": 0.00018551834101747212,
+      "loss": 0.7756352424621582,
+      "step": 3408
+    },
+    {
+      "epoch": 1.4388185654008439,
+      "grad_norm": 0.4030776023864746,
+      "learning_rate": 0.00018549343637384543,
+      "loss": 0.689633309841156,
+      "step": 3410
+    },
+    {
+      "epoch": 1.439662447257384,
+      "grad_norm": 0.4025530517101288,
+      "learning_rate": 0.0001854685120085249,
+      "loss": 0.7173827290534973,
+      "step": 3412
+    },
+    {
+      "epoch": 1.4405063291139242,
+      "grad_norm": 0.4376870095729828,
+      "learning_rate": 0.0001854435679272601,
+      "loss": 0.7212072610855103,
+      "step": 3414
+    },
+    {
+      "epoch": 1.4413502109704641,
+      "grad_norm": 0.4469084143638611,
+      "learning_rate": 0.00018541860413580515,
+      "loss": 0.8260660171508789,
+      "step": 3416
+    },
+    {
+      "epoch": 1.4421940928270043,
+      "grad_norm": 0.48033684492111206,
+      "learning_rate": 0.00018539362063991872,
+      "loss": 0.764930248260498,
+      "step": 3418
+    },
+    {
+      "epoch": 1.4430379746835442,
+      "grad_norm": 0.4428006708621979,
+      "learning_rate": 0.00018536861744536403,
+      "loss": 0.7436494827270508,
+      "step": 3420
+    },
+    {
+      "epoch": 1.4438818565400844,
+      "grad_norm": 0.41812989115715027,
+      "learning_rate": 0.00018534359455790885,
+      "loss": 0.7695552706718445,
+      "step": 3422
+    },
+    {
+      "epoch": 1.4447257383966245,
+      "grad_norm": 0.4243115186691284,
+      "learning_rate": 0.00018531855198332544,
+      "loss": 0.7430856823921204,
+      "step": 3424
+    },
+    {
+      "epoch": 1.4455696202531645,
+      "grad_norm": 0.44387027621269226,
+      "learning_rate": 0.00018529348972739067,
+      "loss": 0.7015851140022278,
+      "step": 3426
+    },
+    {
+      "epoch": 1.4464135021097047,
+      "grad_norm": 0.5105122923851013,
+      "learning_rate": 0.0001852684077958859,
+      "loss": 0.7073075771331787,
+      "step": 3428
+    },
+    {
+      "epoch": 1.4472573839662446,
+      "grad_norm": 0.40193185210227966,
+      "learning_rate": 0.00018524330619459708,
+      "loss": 0.7478930950164795,
+      "step": 3430
+    },
+    {
+      "epoch": 1.4481012658227848,
+      "grad_norm": 0.47935256361961365,
+      "learning_rate": 0.00018521818492931463,
+      "loss": 0.6987152099609375,
+      "step": 3432
+    },
+    {
+      "epoch": 1.448945147679325,
+      "grad_norm": 0.42298686504364014,
+      "learning_rate": 0.00018519304400583356,
+      "loss": 0.6677015423774719,
+      "step": 3434
+    },
+    {
+      "epoch": 1.4497890295358649,
+      "grad_norm": 0.37220221757888794,
+      "learning_rate": 0.0001851678834299534,
+      "loss": 0.5727065205574036,
+      "step": 3436
+    },
+    {
+      "epoch": 1.450632911392405,
+      "grad_norm": 0.4990854561328888,
+      "learning_rate": 0.00018514270320747823,
+      "loss": 0.7245276570320129,
+      "step": 3438
+    },
+    {
+      "epoch": 1.4514767932489452,
+      "grad_norm": 0.4366965889930725,
+      "learning_rate": 0.00018511750334421665,
+      "loss": 0.5907649993896484,
+      "step": 3440
+    },
+    {
+      "epoch": 1.4523206751054851,
+      "grad_norm": 0.4355267286300659,
+      "learning_rate": 0.00018509228384598172,
+      "loss": 0.6898093223571777,
+      "step": 3442
+    },
+    {
+      "epoch": 1.4531645569620253,
+      "grad_norm": 0.4055832028388977,
+      "learning_rate": 0.00018506704471859116,
+      "loss": 0.6242932081222534,
+      "step": 3444
+    },
+    {
+      "epoch": 1.4540084388185655,
+      "grad_norm": 0.42599600553512573,
+      "learning_rate": 0.00018504178596786712,
+      "loss": 0.7214970588684082,
+      "step": 3446
+    },
+    {
+      "epoch": 1.4548523206751054,
+      "grad_norm": 0.41228237748146057,
+      "learning_rate": 0.0001850165075996363,
+      "loss": 0.7120789885520935,
+      "step": 3448
+    },
+    {
+      "epoch": 1.4556962025316456,
+      "grad_norm": 0.4529595375061035,
+      "learning_rate": 0.00018499120961972996,
+      "loss": 0.7376183867454529,
+      "step": 3450
+    },
+    {
+      "epoch": 1.4565400843881857,
+      "grad_norm": 0.4260895550251007,
+      "learning_rate": 0.00018496589203398386,
+      "loss": 0.7321018576622009,
+      "step": 3452
+    },
+    {
+      "epoch": 1.4573839662447257,
+      "grad_norm": 0.46325981616973877,
+      "learning_rate": 0.00018494055484823824,
+      "loss": 0.7099342346191406,
+      "step": 3454
+    },
+    {
+      "epoch": 1.4582278481012658,
+      "grad_norm": 0.43322354555130005,
+      "learning_rate": 0.00018491519806833795,
+      "loss": 0.6663834452629089,
+      "step": 3456
+    },
+    {
+      "epoch": 1.459071729957806,
+      "grad_norm": 0.421683132648468,
+      "learning_rate": 0.0001848898217001323,
+      "loss": 0.6634764671325684,
+      "step": 3458
+    },
+    {
+      "epoch": 1.459915611814346,
+      "grad_norm": 0.44365957379341125,
+      "learning_rate": 0.00018486442574947511,
+      "loss": 0.697095513343811,
+      "step": 3460
+    },
+    {
+      "epoch": 1.460759493670886,
+      "grad_norm": 0.4892813265323639,
+      "learning_rate": 0.00018483901022222474,
+      "loss": 0.8009715676307678,
+      "step": 3462
+    },
+    {
+      "epoch": 1.4616033755274263,
+      "grad_norm": 0.4230288863182068,
+      "learning_rate": 0.00018481357512424408,
+      "loss": 0.631499707698822,
+      "step": 3464
+    },
+    {
+      "epoch": 1.4624472573839662,
+      "grad_norm": 0.49490758776664734,
+      "learning_rate": 0.00018478812046140056,
+      "loss": 0.8045525550842285,
+      "step": 3466
+    },
+    {
+      "epoch": 1.4632911392405064,
+      "grad_norm": 0.5035375952720642,
+      "learning_rate": 0.00018476264623956597,
+      "loss": 0.7299133539199829,
+      "step": 3468
+    },
+    {
+      "epoch": 1.4641350210970465,
+      "grad_norm": 0.45048409700393677,
+      "learning_rate": 0.00018473715246461684,
+      "loss": 0.7313228845596313,
+      "step": 3470
+    },
+    {
+      "epoch": 1.4649789029535865,
+      "grad_norm": 0.41414541006088257,
+      "learning_rate": 0.00018471163914243405,
+      "loss": 0.7515901923179626,
+      "step": 3472
+    },
+    {
+      "epoch": 1.4658227848101266,
+      "grad_norm": 0.4440646767616272,
+      "learning_rate": 0.00018468610627890298,
+      "loss": 0.7425532341003418,
+      "step": 3474
+    },
+    {
+      "epoch": 1.4666666666666668,
+      "grad_norm": 0.4749740660190582,
+      "learning_rate": 0.00018466055387991362,
+      "loss": 0.7509122490882874,
+      "step": 3476
+    },
+    {
+      "epoch": 1.4675105485232067,
+      "grad_norm": 0.498951256275177,
+      "learning_rate": 0.00018463498195136046,
+      "loss": 0.7395774126052856,
+      "step": 3478
+    },
+    {
+      "epoch": 1.4683544303797469,
+      "grad_norm": 0.45996859669685364,
+      "learning_rate": 0.00018460939049914237,
+      "loss": 0.7860153317451477,
+      "step": 3480
+    },
+    {
+      "epoch": 1.469198312236287,
+      "grad_norm": 0.5989604592323303,
+      "learning_rate": 0.0001845837795291629,
+      "loss": 0.8193565607070923,
+      "step": 3482
+    },
+    {
+      "epoch": 1.470042194092827,
+      "grad_norm": 0.3903721272945404,
+      "learning_rate": 0.0001845581490473299,
+      "loss": 0.6569210290908813,
+      "step": 3484
+    },
+    {
+      "epoch": 1.4708860759493672,
+      "grad_norm": 0.4316142201423645,
+      "learning_rate": 0.00018453249905955592,
+      "loss": 0.7226642370223999,
+      "step": 3486
+    },
+    {
+      "epoch": 1.471729957805907,
+      "grad_norm": 0.4447585344314575,
+      "learning_rate": 0.00018450682957175786,
+      "loss": 0.8134398460388184,
+      "step": 3488
+    },
+    {
+      "epoch": 1.4725738396624473,
+      "grad_norm": 0.42515450716018677,
+      "learning_rate": 0.0001844811405898572,
+      "loss": 0.7668985724449158,
+      "step": 3490
+    },
+    {
+      "epoch": 1.4734177215189874,
+      "grad_norm": 0.39883363246917725,
+      "learning_rate": 0.0001844554321197799,
+      "loss": 0.7289790511131287,
+      "step": 3492
+    },
+    {
+      "epoch": 1.4742616033755274,
+      "grad_norm": 0.4419080317020416,
+      "learning_rate": 0.00018442970416745645,
+      "loss": 0.7611824870109558,
+      "step": 3494
+    },
+    {
+      "epoch": 1.4751054852320675,
+      "grad_norm": 0.37460315227508545,
+      "learning_rate": 0.0001844039567388217,
+      "loss": 0.6926230788230896,
+      "step": 3496
+    },
+    {
+      "epoch": 1.4759493670886075,
+      "grad_norm": 0.47706854343414307,
+      "learning_rate": 0.00018437818983981513,
+      "loss": 0.7889325618743896,
+      "step": 3498
+    },
+    {
+      "epoch": 1.4767932489451476,
+      "grad_norm": 0.449028879404068,
+      "learning_rate": 0.00018435240347638074,
+      "loss": 0.7231656312942505,
+      "step": 3500
+    },
+    {
+      "epoch": 1.4767932489451476,
+      "eval_loss": 0.7902651429176331,
+      "eval_runtime": 672.312,
+      "eval_samples_per_second": 3.134,
+      "eval_steps_per_second": 3.134,
+      "step": 3500
+    },
+    {
+      "epoch": 1.4776371308016878,
+      "grad_norm": 0.44085755944252014,
+      "learning_rate": 0.00018432659765446686,
+      "loss": 0.7464927434921265,
+      "step": 3502
+    },
+    {
+      "epoch": 1.4784810126582277,
+      "grad_norm": 0.4243546724319458,
+      "learning_rate": 0.0001843007723800264,
+      "loss": 0.7312393188476562,
+      "step": 3504
+    },
+    {
+      "epoch": 1.479324894514768,
+      "grad_norm": 0.4437529444694519,
+      "learning_rate": 0.00018427492765901677,
+      "loss": 0.7169944643974304,
+      "step": 3506
+    },
+    {
+      "epoch": 1.480168776371308,
+      "grad_norm": 0.3639313280582428,
+      "learning_rate": 0.0001842490634973999,
+      "loss": 0.6169725060462952,
+      "step": 3508
+    },
+    {
+      "epoch": 1.481012658227848,
+      "grad_norm": 0.4452592730522156,
+      "learning_rate": 0.0001842231799011421,
+      "loss": 0.7034649848937988,
+      "step": 3510
+    },
+    {
+      "epoch": 1.4818565400843882,
+      "grad_norm": 0.5229591131210327,
+      "learning_rate": 0.00018419727687621421,
+      "loss": 0.7535615563392639,
+      "step": 3512
+    },
+    {
+      "epoch": 1.4827004219409283,
+      "grad_norm": 0.40748512744903564,
+      "learning_rate": 0.00018417135442859163,
+      "loss": 0.6858918070793152,
+      "step": 3514
+    },
+    {
+      "epoch": 1.4835443037974683,
+      "grad_norm": 0.4532773792743683,
+      "learning_rate": 0.0001841454125642541,
+      "loss": 0.7120109796524048,
+      "step": 3516
+    },
+    {
+      "epoch": 1.4843881856540084,
+      "grad_norm": 0.4281020760536194,
+      "learning_rate": 0.00018411945128918593,
+      "loss": 0.6837015748023987,
+      "step": 3518
+    },
+    {
+      "epoch": 1.4852320675105486,
+      "grad_norm": 0.4891088306903839,
+      "learning_rate": 0.0001840934706093759,
+      "loss": 0.6639618873596191,
+      "step": 3520
+    },
+    {
+      "epoch": 1.4860759493670885,
+      "grad_norm": 0.44358906149864197,
+      "learning_rate": 0.00018406747053081724,
+      "loss": 0.693483829498291,
+      "step": 3522
+    },
+    {
+      "epoch": 1.4869198312236287,
+      "grad_norm": 0.40134626626968384,
+      "learning_rate": 0.00018404145105950767,
+      "loss": 0.7046290636062622,
+      "step": 3524
+    },
+    {
+      "epoch": 1.4877637130801689,
+      "grad_norm": 0.4601934552192688,
+      "learning_rate": 0.0001840154122014494,
+      "loss": 0.6991957426071167,
+      "step": 3526
+    },
+    {
+      "epoch": 1.4886075949367088,
+      "grad_norm": 0.39107242226600647,
+      "learning_rate": 0.00018398935396264906,
+      "loss": 0.667023241519928,
+      "step": 3528
+    },
+    {
+      "epoch": 1.489451476793249,
+      "grad_norm": 0.4498196542263031,
+      "learning_rate": 0.00018396327634911784,
+      "loss": 0.7187332510948181,
+      "step": 3530
+    },
+    {
+      "epoch": 1.4902953586497891,
+      "grad_norm": 0.41769179701805115,
+      "learning_rate": 0.0001839371793668713,
+      "loss": 0.6854014992713928,
+      "step": 3532
+    },
+    {
+      "epoch": 1.491139240506329,
+      "grad_norm": 0.40170612931251526,
+      "learning_rate": 0.0001839110630219295,
+      "loss": 0.6643120050430298,
+      "step": 3534
+    },
+    {
+      "epoch": 1.4919831223628692,
+      "grad_norm": 0.43697425723075867,
+      "learning_rate": 0.00018388492732031701,
+      "loss": 0.6822295188903809,
+      "step": 3536
+    },
+    {
+      "epoch": 1.4928270042194094,
+      "grad_norm": 0.4400080442428589,
+      "learning_rate": 0.00018385877226806288,
+      "loss": 0.7500558495521545,
+      "step": 3538
+    },
+    {
+      "epoch": 1.4936708860759493,
+      "grad_norm": 0.5162575244903564,
+      "learning_rate": 0.00018383259787120048,
+      "loss": 0.7742161750793457,
+      "step": 3540
+    },
+    {
+      "epoch": 1.4945147679324895,
+      "grad_norm": 0.4234789311885834,
+      "learning_rate": 0.0001838064041357678,
+      "loss": 0.7334147095680237,
+      "step": 3542
+    },
+    {
+      "epoch": 1.4953586497890297,
+      "grad_norm": 0.40562525391578674,
+      "learning_rate": 0.0001837801910678072,
+      "loss": 0.691303014755249,
+      "step": 3544
+    },
+    {
+      "epoch": 1.4962025316455696,
+      "grad_norm": 0.3651523292064667,
+      "learning_rate": 0.00018375395867336556,
+      "loss": 0.6478174328804016,
+      "step": 3546
+    },
+    {
+      "epoch": 1.4970464135021098,
+      "grad_norm": 0.45428016781806946,
+      "learning_rate": 0.00018372770695849417,
+      "loss": 0.6800187826156616,
+      "step": 3548
+    },
+    {
+      "epoch": 1.49789029535865,
+      "grad_norm": 0.4258574843406677,
+      "learning_rate": 0.00018370143592924882,
+      "loss": 0.7386951446533203,
+      "step": 3550
+    },
+    {
+      "epoch": 1.4987341772151899,
+      "grad_norm": 0.4260908365249634,
+      "learning_rate": 0.00018367514559168972,
+      "loss": 0.757290780544281,
+      "step": 3552
+    },
+    {
+      "epoch": 1.49957805907173,
+      "grad_norm": 0.4037017226219177,
+      "learning_rate": 0.00018364883595188152,
+      "loss": 0.7254101037979126,
+      "step": 3554
+    },
+    {
+      "epoch": 1.5004219409282702,
+      "grad_norm": 0.4212391674518585,
+      "learning_rate": 0.0001836225070158934,
+      "loss": 0.7342239618301392,
+      "step": 3556
+    },
+    {
+      "epoch": 1.5012658227848101,
+      "grad_norm": 0.4902292490005493,
+      "learning_rate": 0.0001835961587897989,
+      "loss": 0.7147436738014221,
+      "step": 3558
+    },
+    {
+      "epoch": 1.50210970464135,
+      "grad_norm": 0.39230459928512573,
+      "learning_rate": 0.00018356979127967605,
+      "loss": 0.7473883628845215,
+      "step": 3560
+    },
+    {
+      "epoch": 1.5029535864978905,
+      "grad_norm": 0.4203200340270996,
+      "learning_rate": 0.00018354340449160735,
+      "loss": 0.7026845812797546,
+      "step": 3562
+    },
+    {
+      "epoch": 1.5037974683544304,
+      "grad_norm": 0.44363775849342346,
+      "learning_rate": 0.00018351699843167972,
+      "loss": 0.7426526546478271,
+      "step": 3564
+    },
+    {
+      "epoch": 1.5046413502109703,
+      "grad_norm": 0.4479754865169525,
+      "learning_rate": 0.0001834905731059845,
+      "loss": 0.6800932884216309,
+      "step": 3566
+    },
+    {
+      "epoch": 1.5054852320675105,
+      "grad_norm": 0.43898341059684753,
+      "learning_rate": 0.0001834641285206176,
+      "loss": 0.6681315898895264,
+      "step": 3568
+    },
+    {
+      "epoch": 1.5063291139240507,
+      "grad_norm": 0.44789746403694153,
+      "learning_rate": 0.00018343766468167916,
+      "loss": 0.7319603562355042,
+      "step": 3570
+    },
+    {
+      "epoch": 1.5071729957805906,
+      "grad_norm": 0.44450023770332336,
+      "learning_rate": 0.00018341118159527392,
+      "loss": 0.7205765843391418,
+      "step": 3572
+    },
+    {
+      "epoch": 1.5080168776371308,
+      "grad_norm": 0.4461601674556732,
+      "learning_rate": 0.00018338467926751103,
+      "loss": 0.8053180575370789,
+      "step": 3574
+    },
+    {
+      "epoch": 1.508860759493671,
+      "grad_norm": 0.415216863155365,
+      "learning_rate": 0.00018335815770450408,
+      "loss": 0.6944835186004639,
+      "step": 3576
+    },
+    {
+      "epoch": 1.5097046413502109,
+      "grad_norm": 0.48354068398475647,
+      "learning_rate": 0.00018333161691237105,
+      "loss": 0.7333522439002991,
+      "step": 3578
+    },
+    {
+      "epoch": 1.510548523206751,
+      "grad_norm": 0.44450995326042175,
+      "learning_rate": 0.00018330505689723446,
+      "loss": 0.7036639451980591,
+      "step": 3580
+    },
+    {
+      "epoch": 1.5113924050632912,
+      "grad_norm": 0.355945348739624,
+      "learning_rate": 0.00018327847766522111,
+      "loss": 0.6589286923408508,
+      "step": 3582
+    },
+    {
+      "epoch": 1.5122362869198311,
+      "grad_norm": 0.4532116651535034,
+      "learning_rate": 0.00018325187922246236,
+      "loss": 0.7284913063049316,
+      "step": 3584
+    },
+    {
+      "epoch": 1.5130801687763713,
+      "grad_norm": 0.45910802483558655,
+      "learning_rate": 0.00018322526157509396,
+      "loss": 0.7253853678703308,
+      "step": 3586
+    },
+    {
+      "epoch": 1.5139240506329115,
+      "grad_norm": 0.4729263186454773,
+      "learning_rate": 0.0001831986247292561,
+      "loss": 0.7626010179519653,
+      "step": 3588
+    },
+    {
+      "epoch": 1.5147679324894514,
+      "grad_norm": 0.42071840167045593,
+      "learning_rate": 0.00018317196869109336,
+      "loss": 0.705466091632843,
+      "step": 3590
+    },
+    {
+      "epoch": 1.5156118143459916,
+      "grad_norm": 0.40275856852531433,
+      "learning_rate": 0.00018314529346675479,
+      "loss": 0.6900550723075867,
+      "step": 3592
+    },
+    {
+      "epoch": 1.5164556962025317,
+      "grad_norm": 0.43557414412498474,
+      "learning_rate": 0.00018311859906239387,
+      "loss": 0.6951512694358826,
+      "step": 3594
+    },
+    {
+      "epoch": 1.5172995780590717,
+      "grad_norm": 0.4007355570793152,
+      "learning_rate": 0.00018309188548416844,
+      "loss": 0.7453750967979431,
+      "step": 3596
+    },
+    {
+      "epoch": 1.5181434599156118,
+      "grad_norm": 0.4686451256275177,
+      "learning_rate": 0.00018306515273824086,
+      "loss": 0.7549434900283813,
+      "step": 3598
+    },
+    {
+      "epoch": 1.518987341772152,
+      "grad_norm": 0.47308704257011414,
+      "learning_rate": 0.0001830384008307778,
+      "loss": 0.7708864212036133,
+      "step": 3600
+    },
+    {
+      "epoch": 1.518987341772152,
+      "eval_loss": 0.785450279712677,
+      "eval_runtime": 675.2312,
+      "eval_samples_per_second": 3.12,
+      "eval_steps_per_second": 3.12,
+      "step": 3600
+    },
+    {
+      "epoch": 1.519831223628692,
+      "grad_norm": 0.4343389570713043,
+      "learning_rate": 0.0001830116297679505,
+      "loss": 0.7501338124275208,
+      "step": 3602
+    },
+    {
+      "epoch": 1.520675105485232,
+      "grad_norm": 0.4971345067024231,
+      "learning_rate": 0.00018298483955593445,
+      "loss": 0.7358015179634094,
+      "step": 3604
+    },
+    {
+      "epoch": 1.5215189873417723,
+      "grad_norm": 0.42013493180274963,
+      "learning_rate": 0.0001829580302009097,
+      "loss": 0.7972728610038757,
+      "step": 3606
+    },
+    {
+      "epoch": 1.5223628691983122,
+      "grad_norm": 0.44829806685447693,
+      "learning_rate": 0.00018293120170906062,
+      "loss": 0.5908644199371338,
+      "step": 3608
+    },
+    {
+      "epoch": 1.5232067510548524,
+      "grad_norm": 0.47412917017936707,
+      "learning_rate": 0.000182904354086576,
+      "loss": 0.7872797846794128,
+      "step": 3610
+    },
+    {
+      "epoch": 1.5240506329113925,
+      "grad_norm": 0.4224211275577545,
+      "learning_rate": 0.0001828774873396491,
+      "loss": 0.7327616214752197,
+      "step": 3612
+    },
+    {
+      "epoch": 1.5248945147679325,
+      "grad_norm": 0.4055147171020508,
+      "learning_rate": 0.00018285060147447756,
+      "loss": 0.6649892330169678,
+      "step": 3614
+    },
+    {
+      "epoch": 1.5257383966244724,
+      "grad_norm": 0.4058035910129547,
+      "learning_rate": 0.00018282369649726346,
+      "loss": 0.6939427852630615,
+      "step": 3616
+    },
+    {
+      "epoch": 1.5265822784810128,
+      "grad_norm": 0.4192068874835968,
+      "learning_rate": 0.00018279677241421325,
+      "loss": 0.6351069211959839,
+      "step": 3618
+    },
+    {
+      "epoch": 1.5274261603375527,
+      "grad_norm": 0.4659916162490845,
+      "learning_rate": 0.00018276982923153776,
+      "loss": 0.6963213682174683,
+      "step": 3620
+    },
+    {
+      "epoch": 1.5282700421940927,
+      "grad_norm": 0.4993051290512085,
+      "learning_rate": 0.00018274286695545229,
+      "loss": 0.7463255524635315,
+      "step": 3622
+    },
+    {
+      "epoch": 1.529113924050633,
+      "grad_norm": 0.39232251048088074,
+      "learning_rate": 0.00018271588559217654,
+      "loss": 0.6566499471664429,
+      "step": 3624
+    },
+    {
+      "epoch": 1.529957805907173,
+      "grad_norm": 0.4482063353061676,
+      "learning_rate": 0.00018268888514793457,
+      "loss": 0.7695494890213013,
+      "step": 3626
+    },
+    {
+      "epoch": 1.530801687763713,
+      "grad_norm": 0.4000336825847626,
+      "learning_rate": 0.00018266186562895483,
+      "loss": 0.7357239723205566,
+      "step": 3628
+    },
+    {
+      "epoch": 1.5316455696202531,
+      "grad_norm": 0.4220462441444397,
+      "learning_rate": 0.0001826348270414703,
+      "loss": 0.7234170436859131,
+      "step": 3630
+    },
+    {
+      "epoch": 1.5324894514767933,
+      "grad_norm": 0.4787490665912628,
+      "learning_rate": 0.00018260776939171814,
+      "loss": 0.7864061594009399,
+      "step": 3632
+    },
+    {
+      "epoch": 1.5333333333333332,
+      "grad_norm": 0.3670683205127716,
+      "learning_rate": 0.00018258069268594013,
+      "loss": 0.670537531375885,
+      "step": 3634
+    },
+    {
+      "epoch": 1.5341772151898734,
+      "grad_norm": 0.41599538922309875,
+      "learning_rate": 0.0001825535969303823,
+      "loss": 0.7535691261291504,
+      "step": 3636
+    },
+    {
+      "epoch": 1.5350210970464135,
+      "grad_norm": 0.5033245086669922,
+      "learning_rate": 0.00018252648213129514,
+      "loss": 0.7553712725639343,
+      "step": 3638
+    },
+    {
+      "epoch": 1.5358649789029535,
+      "grad_norm": 0.4235823452472687,
+      "learning_rate": 0.0001824993482949335,
+      "loss": 0.7149441242218018,
+      "step": 3640
+    },
+    {
+      "epoch": 1.5367088607594936,
+      "grad_norm": 0.37993329763412476,
+      "learning_rate": 0.00018247219542755664,
+      "loss": 0.735568642616272,
+      "step": 3642
+    },
+    {
+      "epoch": 1.5375527426160338,
+      "grad_norm": 0.39534425735473633,
+      "learning_rate": 0.00018244502353542818,
+      "loss": 0.693510115146637,
+      "step": 3644
+    },
+    {
+      "epoch": 1.5383966244725737,
+      "grad_norm": 0.45447441935539246,
+      "learning_rate": 0.00018241783262481622,
+      "loss": 0.7261984348297119,
+      "step": 3646
+    },
+    {
+      "epoch": 1.539240506329114,
+      "grad_norm": 0.4131409823894501,
+      "learning_rate": 0.0001823906227019931,
+      "loss": 0.7518824338912964,
+      "step": 3648
+    },
+    {
+      "epoch": 1.540084388185654,
+      "grad_norm": 0.46603885293006897,
+      "learning_rate": 0.00018236339377323567,
+      "loss": 0.7650659680366516,
+      "step": 3650
+    },
+    {
+      "epoch": 1.540928270042194,
+      "grad_norm": 0.3998747766017914,
+      "learning_rate": 0.00018233614584482513,
+      "loss": 0.6436064839363098,
+      "step": 3652
+    },
+    {
+      "epoch": 1.5417721518987342,
+      "grad_norm": 0.4314839839935303,
+      "learning_rate": 0.00018230887892304703,
+      "loss": 0.7068451642990112,
+      "step": 3654
+    },
+    {
+      "epoch": 1.5426160337552743,
+      "grad_norm": 0.4324610233306885,
+      "learning_rate": 0.0001822815930141913,
+      "loss": 0.7342977523803711,
+      "step": 3656
+    },
+    {
+      "epoch": 1.5434599156118143,
+      "grad_norm": 0.4385458827018738,
+      "learning_rate": 0.0001822542881245524,
+      "loss": 0.6482785940170288,
+      "step": 3658
+    },
+    {
+      "epoch": 1.5443037974683544,
+      "grad_norm": 0.3736587166786194,
+      "learning_rate": 0.0001822269642604289,
+      "loss": 0.6319621205329895,
+      "step": 3660
+    },
+    {
+      "epoch": 1.5451476793248946,
+      "grad_norm": 0.42113035917282104,
+      "learning_rate": 0.00018219962142812393,
+      "loss": 0.6603307723999023,
+      "step": 3662
+    },
+    {
+      "epoch": 1.5459915611814345,
+      "grad_norm": 0.422158420085907,
+      "learning_rate": 0.00018217225963394496,
+      "loss": 0.6874772906303406,
+      "step": 3664
+    },
+    {
+      "epoch": 1.5468354430379747,
+      "grad_norm": 0.42724940180778503,
+      "learning_rate": 0.00018214487888420388,
+      "loss": 0.7231854796409607,
+      "step": 3666
+    },
+    {
+      "epoch": 1.5476793248945149,
+      "grad_norm": 0.3747844398021698,
+      "learning_rate": 0.00018211747918521685,
+      "loss": 0.6418231129646301,
+      "step": 3668
+    },
+    {
+      "epoch": 1.5485232067510548,
+      "grad_norm": 0.37007269263267517,
+      "learning_rate": 0.00018209006054330446,
+      "loss": 0.7019823789596558,
+      "step": 3670
+    },
+    {
+      "epoch": 1.549367088607595,
+      "grad_norm": 0.409701943397522,
+      "learning_rate": 0.00018206262296479168,
+      "loss": 0.6632764339447021,
+      "step": 3672
+    },
+    {
+      "epoch": 1.5502109704641351,
+      "grad_norm": 0.49566221237182617,
+      "learning_rate": 0.00018203516645600782,
+      "loss": 0.7327075600624084,
+      "step": 3674
+    },
+    {
+      "epoch": 1.551054852320675,
+      "grad_norm": 0.4549940526485443,
+      "learning_rate": 0.0001820076910232866,
+      "loss": 0.6535149812698364,
+      "step": 3676
+    },
+    {
+      "epoch": 1.5518987341772152,
+      "grad_norm": 0.42619478702545166,
+      "learning_rate": 0.000181980196672966,
+      "loss": 0.7761320471763611,
+      "step": 3678
+    },
+    {
+      "epoch": 1.5527426160337554,
+      "grad_norm": 0.4332887828350067,
+      "learning_rate": 0.00018195268341138852,
+      "loss": 0.6728626489639282,
+      "step": 3680
+    },
+    {
+      "epoch": 1.5535864978902953,
+      "grad_norm": 0.4972975552082062,
+      "learning_rate": 0.0001819251512449009,
+      "loss": 0.7369232177734375,
+      "step": 3682
+    },
+    {
+      "epoch": 1.5544303797468353,
+      "grad_norm": 0.3619556725025177,
+      "learning_rate": 0.00018189760017985428,
+      "loss": 0.6703663468360901,
+      "step": 3684
+    },
+    {
+      "epoch": 1.5552742616033757,
+      "grad_norm": 0.46435829997062683,
+      "learning_rate": 0.00018187003022260417,
+      "loss": 0.7183756828308105,
+      "step": 3686
+    },
+    {
+      "epoch": 1.5561181434599156,
+      "grad_norm": 0.4605063498020172,
+      "learning_rate": 0.00018184244137951045,
+      "loss": 0.680042564868927,
+      "step": 3688
+    },
+    {
+      "epoch": 1.5569620253164556,
+      "grad_norm": 0.4023124873638153,
+      "learning_rate": 0.00018181483365693732,
+      "loss": 0.679821789264679,
+      "step": 3690
+    },
+    {
+      "epoch": 1.557805907172996,
+      "grad_norm": 0.459499716758728,
+      "learning_rate": 0.00018178720706125333,
+      "loss": 0.7408921718597412,
+      "step": 3692
+    },
+    {
+      "epoch": 1.5586497890295359,
+      "grad_norm": 0.4397135078907013,
+      "learning_rate": 0.00018175956159883146,
+      "loss": 0.6926741003990173,
+      "step": 3694
+    },
+    {
+      "epoch": 1.5594936708860758,
+      "grad_norm": 0.4476170837879181,
+      "learning_rate": 0.0001817318972760489,
+      "loss": 0.752724826335907,
+      "step": 3696
+    },
+    {
+      "epoch": 1.560337552742616,
+      "grad_norm": 0.43684154748916626,
+      "learning_rate": 0.00018170421409928735,
+      "loss": 0.7217722535133362,
+      "step": 3698
+    },
+    {
+      "epoch": 1.5611814345991561,
+      "grad_norm": 0.4118131995201111,
+      "learning_rate": 0.00018167651207493278,
+      "loss": 0.7264936566352844,
+      "step": 3700
+    },
+    {
+      "epoch": 1.5611814345991561,
+      "eval_loss": 0.7854447960853577,
+      "eval_runtime": 687.7907,
+      "eval_samples_per_second": 3.063,
+      "eval_steps_per_second": 3.063,
+      "step": 3700
+    },
+    {
+      "epoch": 1.562025316455696,
+      "grad_norm": 0.40639030933380127,
+      "learning_rate": 0.0001816487912093755,
+      "loss": 0.701478123664856,
+      "step": 3702
+    },
+    {
+      "epoch": 1.5628691983122363,
+      "grad_norm": 0.4205699563026428,
+      "learning_rate": 0.00018162105150901017,
+      "loss": 0.696711003780365,
+      "step": 3704
+    },
+    {
+      "epoch": 1.5637130801687764,
+      "grad_norm": 0.46437183022499084,
+      "learning_rate": 0.0001815932929802358,
+      "loss": 0.7130293250083923,
+      "step": 3706
+    },
+    {
+      "epoch": 1.5645569620253164,
+      "grad_norm": 0.4413023591041565,
+      "learning_rate": 0.0001815655156294558,
+      "loss": 0.7180310487747192,
+      "step": 3708
+    },
+    {
+      "epoch": 1.5654008438818565,
+      "grad_norm": 0.3964478075504303,
+      "learning_rate": 0.00018153771946307783,
+      "loss": 0.6883325576782227,
+      "step": 3710
+    },
+    {
+      "epoch": 1.5662447257383967,
+      "grad_norm": 0.34486109018325806,
+      "learning_rate": 0.00018150990448751394,
+      "loss": 0.6683162450790405,
+      "step": 3712
+    },
+    {
+      "epoch": 1.5670886075949366,
+      "grad_norm": 0.4209035336971283,
+      "learning_rate": 0.00018148207070918052,
+      "loss": 0.6969789862632751,
+      "step": 3714
+    },
+    {
+      "epoch": 1.5679324894514768,
+      "grad_norm": 0.41674816608428955,
+      "learning_rate": 0.00018145421813449829,
+      "loss": 0.714320182800293,
+      "step": 3716
+    },
+    {
+      "epoch": 1.568776371308017,
+      "grad_norm": 0.4365934729576111,
+      "learning_rate": 0.00018142634676989228,
+      "loss": 0.7125279307365417,
+      "step": 3718
+    },
+    {
+      "epoch": 1.5696202531645569,
+      "grad_norm": 0.43599721789360046,
+      "learning_rate": 0.00018139845662179188,
+      "loss": 0.6787398457527161,
+      "step": 3720
+    },
+    {
+      "epoch": 1.570464135021097,
+      "grad_norm": 0.36320003867149353,
+      "learning_rate": 0.0001813705476966308,
+      "loss": 0.6401328444480896,
+      "step": 3722
+    },
+    {
+      "epoch": 1.5713080168776372,
+      "grad_norm": 0.43504419922828674,
+      "learning_rate": 0.00018134262000084718,
+      "loss": 0.6875129342079163,
+      "step": 3724
+    },
+    {
+      "epoch": 1.5721518987341772,
+      "grad_norm": 0.476723313331604,
+      "learning_rate": 0.00018131467354088332,
+      "loss": 0.6903918385505676,
+      "step": 3726
+    },
+    {
+      "epoch": 1.5729957805907173,
+      "grad_norm": 0.4465550184249878,
+      "learning_rate": 0.00018128670832318592,
+      "loss": 0.690018892288208,
+      "step": 3728
+    },
+    {
+      "epoch": 1.5738396624472575,
+      "grad_norm": 0.43816477060317993,
+      "learning_rate": 0.0001812587243542061,
+      "loss": 0.6912158727645874,
+      "step": 3730
+    },
+    {
+      "epoch": 1.5746835443037974,
+      "grad_norm": 0.40546250343322754,
+      "learning_rate": 0.00018123072164039913,
+      "loss": 0.7415492534637451,
+      "step": 3732
+    },
+    {
+      "epoch": 1.5755274261603376,
+      "grad_norm": 0.4253178536891937,
+      "learning_rate": 0.00018120270018822477,
+      "loss": 0.7664727568626404,
+      "step": 3734
+    },
+    {
+      "epoch": 1.5763713080168777,
+      "grad_norm": 0.503524661064148,
+      "learning_rate": 0.000181174660004147,
+      "loss": 0.782779335975647,
+      "step": 3736
+    },
+    {
+      "epoch": 1.5772151898734177,
+      "grad_norm": 0.5260395407676697,
+      "learning_rate": 0.00018114660109463414,
+      "loss": 0.806460976600647,
+      "step": 3738
+    },
+    {
+      "epoch": 1.5780590717299579,
+      "grad_norm": 0.3641359806060791,
+      "learning_rate": 0.0001811185234661589,
+      "loss": 0.6859704852104187,
+      "step": 3740
+    },
+    {
+      "epoch": 1.578902953586498,
+      "grad_norm": 0.4468916058540344,
+      "learning_rate": 0.00018109042712519818,
+      "loss": 0.7798553109169006,
+      "step": 3742
+    },
+    {
+      "epoch": 1.579746835443038,
+      "grad_norm": 0.4078207015991211,
+      "learning_rate": 0.00018106231207823327,
+      "loss": 0.7086471915245056,
+      "step": 3744
+    },
+    {
+      "epoch": 1.580590717299578,
+      "grad_norm": 0.39144742488861084,
+      "learning_rate": 0.00018103417833174981,
+      "loss": 0.6669760346412659,
+      "step": 3746
+    },
+    {
+      "epoch": 1.5814345991561183,
+      "grad_norm": 0.4366845190525055,
+      "learning_rate": 0.0001810060258922377,
+      "loss": 0.7296954989433289,
+      "step": 3748
+    },
+    {
+      "epoch": 1.5822784810126582,
+      "grad_norm": 0.46795082092285156,
+      "learning_rate": 0.00018097785476619118,
+      "loss": 0.7663856148719788,
+      "step": 3750
+    },
+    {
+      "epoch": 1.5831223628691982,
+      "grad_norm": 0.40331873297691345,
+      "learning_rate": 0.00018094966496010878,
+      "loss": 0.7105234861373901,
+      "step": 3752
+    },
+    {
+      "epoch": 1.5839662447257385,
+      "grad_norm": 0.38210469484329224,
+      "learning_rate": 0.00018092145648049334,
+      "loss": 0.666635274887085,
+      "step": 3754
+    },
+    {
+      "epoch": 1.5848101265822785,
+      "grad_norm": 0.45946040749549866,
+      "learning_rate": 0.000180893229333852,
+      "loss": 0.7287175059318542,
+      "step": 3756
+    },
+    {
+      "epoch": 1.5856540084388184,
+      "grad_norm": 0.4035777747631073,
+      "learning_rate": 0.00018086498352669624,
+      "loss": 0.7145618796348572,
+      "step": 3758
+    },
+    {
+      "epoch": 1.5864978902953588,
+      "grad_norm": 0.42843320965766907,
+      "learning_rate": 0.0001808367190655418,
+      "loss": 0.6550979018211365,
+      "step": 3760
+    },
+    {
+      "epoch": 1.5873417721518988,
+      "grad_norm": 0.4016468822956085,
+      "learning_rate": 0.0001808084359569088,
+      "loss": 0.7040430307388306,
+      "step": 3762
+    },
+    {
+      "epoch": 1.5881856540084387,
+      "grad_norm": 0.4126848578453064,
+      "learning_rate": 0.00018078013420732157,
+      "loss": 0.7523179054260254,
+      "step": 3764
+    },
+    {
+      "epoch": 1.5890295358649789,
+      "grad_norm": 0.45878592133522034,
+      "learning_rate": 0.00018075181382330878,
+      "loss": 0.7497967481613159,
+      "step": 3766
+    },
+    {
+      "epoch": 1.589873417721519,
+      "grad_norm": 0.44179272651672363,
+      "learning_rate": 0.00018072347481140345,
+      "loss": 0.7694459557533264,
+      "step": 3768
+    },
+    {
+      "epoch": 1.590717299578059,
+      "grad_norm": 0.3706412613391876,
+      "learning_rate": 0.00018069511717814276,
+      "loss": 0.6766634583473206,
+      "step": 3770
+    },
+    {
+      "epoch": 1.5915611814345991,
+      "grad_norm": 0.40555453300476074,
+      "learning_rate": 0.0001806667409300683,
+      "loss": 0.6394028067588806,
+      "step": 3772
+    },
+    {
+      "epoch": 1.5924050632911393,
+      "grad_norm": 0.3752219080924988,
+      "learning_rate": 0.000180638346073726,
+      "loss": 0.6775637865066528,
+      "step": 3774
+    },
+    {
+      "epoch": 1.5932489451476792,
+      "grad_norm": 0.40960127115249634,
+      "learning_rate": 0.00018060993261566593,
+      "loss": 0.7200632095336914,
+      "step": 3776
+    },
+    {
+      "epoch": 1.5940928270042194,
+      "grad_norm": 0.4215535521507263,
+      "learning_rate": 0.00018058150056244254,
+      "loss": 0.7314276695251465,
+      "step": 3778
+    },
+    {
+      "epoch": 1.5949367088607596,
+      "grad_norm": 0.42734605073928833,
+      "learning_rate": 0.00018055304992061456,
+      "loss": 0.7954897284507751,
+      "step": 3780
+    },
+    {
+      "epoch": 1.5957805907172995,
+      "grad_norm": 0.3980397582054138,
+      "learning_rate": 0.00018052458069674505,
+      "loss": 0.6552245020866394,
+      "step": 3782
+    },
+    {
+      "epoch": 1.5966244725738397,
+      "grad_norm": 0.44602933526039124,
+      "learning_rate": 0.00018049609289740124,
+      "loss": 0.7638092041015625,
+      "step": 3784
+    },
+    {
+      "epoch": 1.5974683544303798,
+      "grad_norm": 0.43835216760635376,
+      "learning_rate": 0.00018046758652915474,
+      "loss": 0.6886481046676636,
+      "step": 3786
+    },
+    {
+      "epoch": 1.5983122362869198,
+      "grad_norm": 0.38229402899742126,
+      "learning_rate": 0.00018043906159858147,
+      "loss": 0.6726123690605164,
+      "step": 3788
+    },
+    {
+      "epoch": 1.59915611814346,
+      "grad_norm": 0.43049392104148865,
+      "learning_rate": 0.00018041051811226155,
+      "loss": 0.7416896224021912,
+      "step": 3790
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 0.4027816951274872,
+      "learning_rate": 0.00018038195607677943,
+      "loss": 0.682691752910614,
+      "step": 3792
+    },
+    {
+      "epoch": 1.60084388185654,
+      "grad_norm": 0.4143599569797516,
+      "learning_rate": 0.00018035337549872376,
+      "loss": 0.7199558615684509,
+      "step": 3794
+    },
+    {
+      "epoch": 1.6016877637130802,
+      "grad_norm": 0.41221415996551514,
+      "learning_rate": 0.00018032477638468762,
+      "loss": 0.6586267948150635,
+      "step": 3796
+    },
+    {
+      "epoch": 1.6025316455696204,
+      "grad_norm": 0.40655994415283203,
+      "learning_rate": 0.0001802961587412682,
+      "loss": 0.7579189538955688,
+      "step": 3798
+    },
+    {
+      "epoch": 1.6033755274261603,
+      "grad_norm": 0.4163990914821625,
+      "learning_rate": 0.00018026752257506708,
+      "loss": 0.7246071100234985,
+      "step": 3800
+    },
+    {
+      "epoch": 1.6033755274261603,
+      "eval_loss": 0.778353214263916,
+      "eval_runtime": 692.5522,
+      "eval_samples_per_second": 3.042,
+      "eval_steps_per_second": 3.042,
+      "step": 3800
+    },
+    {
+      "epoch": 1.6042194092827005,
+      "grad_norm": 0.378628671169281,
+      "learning_rate": 0.0001802388678926901,
+      "loss": 0.7114223837852478,
+      "step": 3802
+    },
+    {
+      "epoch": 1.6050632911392406,
+      "grad_norm": 0.5050750374794006,
+      "learning_rate": 0.0001802101947007473,
+      "loss": 0.7519527077674866,
+      "step": 3804
+    },
+    {
+      "epoch": 1.6059071729957806,
+      "grad_norm": 0.5003950595855713,
+      "learning_rate": 0.00018018150300585305,
+      "loss": 0.7553771138191223,
+      "step": 3806
+    },
+    {
+      "epoch": 1.6067510548523207,
+      "grad_norm": 0.3851728141307831,
+      "learning_rate": 0.00018015279281462595,
+      "loss": 0.6584326028823853,
+      "step": 3808
+    },
+    {
+      "epoch": 1.6075949367088609,
+      "grad_norm": 0.41256240010261536,
+      "learning_rate": 0.00018012406413368893,
+      "loss": 0.7215287089347839,
+      "step": 3810
+    },
+    {
+      "epoch": 1.6084388185654008,
+      "grad_norm": 0.41479718685150146,
+      "learning_rate": 0.00018009531696966912,
+      "loss": 0.7753872275352478,
+      "step": 3812
+    },
+    {
+      "epoch": 1.6092827004219408,
+      "grad_norm": 0.398509681224823,
+      "learning_rate": 0.00018006655132919797,
+      "loss": 0.6984832882881165,
+      "step": 3814
+    },
+    {
+      "epoch": 1.6101265822784812,
+      "grad_norm": 0.42210307717323303,
+      "learning_rate": 0.0001800377672189111,
+      "loss": 0.8117644786834717,
+      "step": 3816
+    },
+    {
+      "epoch": 1.610970464135021,
+      "grad_norm": 0.3714556396007538,
+      "learning_rate": 0.0001800089646454485,
+      "loss": 0.6302977204322815,
+      "step": 3818
+    },
+    {
+      "epoch": 1.611814345991561,
+      "grad_norm": 0.4478890299797058,
+      "learning_rate": 0.00017998014361545438,
+      "loss": 0.7327325940132141,
+      "step": 3820
+    },
+    {
+      "epoch": 1.6126582278481014,
+      "grad_norm": 0.43516477942466736,
+      "learning_rate": 0.00017995130413557713,
+      "loss": 0.6877713203430176,
+      "step": 3822
+    },
+    {
+      "epoch": 1.6135021097046414,
+      "grad_norm": 0.42912542819976807,
+      "learning_rate": 0.0001799224462124695,
+      "loss": 0.6797093152999878,
+      "step": 3824
+    },
+    {
+      "epoch": 1.6143459915611813,
+      "grad_norm": 0.40646228194236755,
+      "learning_rate": 0.00017989356985278853,
+      "loss": 0.7402470111846924,
+      "step": 3826
+    },
+    {
+      "epoch": 1.6151898734177215,
+      "grad_norm": 0.42576655745506287,
+      "learning_rate": 0.00017986467506319535,
+      "loss": 0.6698583364486694,
+      "step": 3828
+    },
+    {
+      "epoch": 1.6160337552742616,
+      "grad_norm": 0.4298654794692993,
+      "learning_rate": 0.0001798357618503555,
+      "loss": 0.6736086010932922,
+      "step": 3830
+    },
+    {
+      "epoch": 1.6168776371308016,
+      "grad_norm": 0.4205976128578186,
+      "learning_rate": 0.0001798068302209386,
+      "loss": 0.7724249958992004,
+      "step": 3832
+    },
+    {
+      "epoch": 1.6177215189873417,
+      "grad_norm": 0.39669108390808105,
+      "learning_rate": 0.00017977788018161872,
+      "loss": 0.6468521952629089,
+      "step": 3834
+    },
+    {
+      "epoch": 1.618565400843882,
+      "grad_norm": 0.428551584482193,
+      "learning_rate": 0.00017974891173907405,
+      "loss": 0.7086918950080872,
+      "step": 3836
+    },
+    {
+      "epoch": 1.6194092827004218,
+      "grad_norm": 0.43773409724235535,
+      "learning_rate": 0.00017971992489998703,
+      "loss": 0.6483711004257202,
+      "step": 3838
+    },
+    {
+      "epoch": 1.620253164556962,
+      "grad_norm": 0.40540891885757446,
+      "learning_rate": 0.00017969091967104438,
+      "loss": 0.6947301030158997,
+      "step": 3840
+    },
+    {
+      "epoch": 1.6210970464135022,
+      "grad_norm": 0.3685019016265869,
+      "learning_rate": 0.00017966189605893708,
+      "loss": 0.667517900466919,
+      "step": 3842
+    },
+    {
+      "epoch": 1.621940928270042,
+      "grad_norm": 0.44813284277915955,
+      "learning_rate": 0.0001796328540703603,
+      "loss": 0.6957448124885559,
+      "step": 3844
+    },
+    {
+      "epoch": 1.6227848101265823,
+      "grad_norm": 0.4524349868297577,
+      "learning_rate": 0.00017960379371201345,
+      "loss": 0.7735851407051086,
+      "step": 3846
+    },
+    {
+      "epoch": 1.6236286919831224,
+      "grad_norm": 0.42349138855934143,
+      "learning_rate": 0.0001795747149906002,
+      "loss": 0.7052056193351746,
+      "step": 3848
+    },
+    {
+      "epoch": 1.6244725738396624,
+      "grad_norm": 0.47317537665367126,
+      "learning_rate": 0.0001795456179128285,
+      "loss": 0.7402311563491821,
+      "step": 3850
+    },
+    {
+      "epoch": 1.6253164556962025,
+      "grad_norm": 0.38915711641311646,
+      "learning_rate": 0.0001795165024854104,
+      "loss": 0.6743243932723999,
+      "step": 3852
+    },
+    {
+      "epoch": 1.6261603375527427,
+      "grad_norm": 0.4430672526359558,
+      "learning_rate": 0.00017948736871506235,
+      "loss": 0.6983860731124878,
+      "step": 3854
+    },
+    {
+      "epoch": 1.6270042194092826,
+      "grad_norm": 0.44286689162254333,
+      "learning_rate": 0.00017945821660850494,
+      "loss": 0.8012514710426331,
+      "step": 3856
+    },
+    {
+      "epoch": 1.6278481012658228,
+      "grad_norm": 0.43364086747169495,
+      "learning_rate": 0.00017942904617246295,
+      "loss": 0.8064966201782227,
+      "step": 3858
+    },
+    {
+      "epoch": 1.628691983122363,
+      "grad_norm": 0.45848122239112854,
+      "learning_rate": 0.00017939985741366547,
+      "loss": 0.7069969177246094,
+      "step": 3860
+    },
+    {
+      "epoch": 1.629535864978903,
+      "grad_norm": 0.4199717044830322,
+      "learning_rate": 0.0001793706503388458,
+      "loss": 0.6722052097320557,
+      "step": 3862
+    },
+    {
+      "epoch": 1.630379746835443,
+      "grad_norm": 0.3820365071296692,
+      "learning_rate": 0.00017934142495474142,
+      "loss": 0.663442075252533,
+      "step": 3864
+    },
+    {
+      "epoch": 1.6312236286919832,
+      "grad_norm": 0.4470944106578827,
+      "learning_rate": 0.00017931218126809412,
+      "loss": 0.70677250623703,
+      "step": 3866
+    },
+    {
+      "epoch": 1.6320675105485232,
+      "grad_norm": 0.3941600024700165,
+      "learning_rate": 0.0001792829192856498,
+      "loss": 0.7125341892242432,
+      "step": 3868
+    },
+    {
+      "epoch": 1.6329113924050633,
+      "grad_norm": 0.44924396276474,
+      "learning_rate": 0.00017925363901415863,
+      "loss": 0.7245699763298035,
+      "step": 3870
+    },
+    {
+      "epoch": 1.6337552742616035,
+      "grad_norm": 0.4070027470588684,
+      "learning_rate": 0.00017922434046037508,
+      "loss": 0.6977120637893677,
+      "step": 3872
+    },
+    {
+      "epoch": 1.6345991561181434,
+      "grad_norm": 0.45040908455848694,
+      "learning_rate": 0.00017919502363105772,
+      "loss": 0.7631482481956482,
+      "step": 3874
+    },
+    {
+      "epoch": 1.6354430379746834,
+      "grad_norm": 0.4083724617958069,
+      "learning_rate": 0.00017916568853296934,
+      "loss": 0.6656452417373657,
+      "step": 3876
+    },
+    {
+      "epoch": 1.6362869198312238,
+      "grad_norm": 0.3762308359146118,
+      "learning_rate": 0.00017913633517287704,
+      "loss": 0.6302213072776794,
+      "step": 3878
+    },
+    {
+      "epoch": 1.6371308016877637,
+      "grad_norm": 0.36366912722587585,
+      "learning_rate": 0.00017910696355755205,
+      "loss": 0.7119566202163696,
+      "step": 3880
+    },
+    {
+      "epoch": 1.6379746835443036,
+      "grad_norm": 0.43168532848358154,
+      "learning_rate": 0.00017907757369376985,
+      "loss": 0.7271678447723389,
+      "step": 3882
+    },
+    {
+      "epoch": 1.638818565400844,
+      "grad_norm": 0.37720009684562683,
+      "learning_rate": 0.00017904816558831014,
+      "loss": 0.6340095400810242,
+      "step": 3884
+    },
+    {
+      "epoch": 1.639662447257384,
+      "grad_norm": 0.36539995670318604,
+      "learning_rate": 0.00017901873924795677,
+      "loss": 0.6910572052001953,
+      "step": 3886
+    },
+    {
+      "epoch": 1.640506329113924,
+      "grad_norm": 0.42437905073165894,
+      "learning_rate": 0.00017898929467949782,
+      "loss": 0.7570282220840454,
+      "step": 3888
+    },
+    {
+      "epoch": 1.6413502109704643,
+      "grad_norm": 0.43482181429862976,
+      "learning_rate": 0.00017895983188972558,
+      "loss": 0.7835264205932617,
+      "step": 3890
+    },
+    {
+      "epoch": 1.6421940928270042,
+      "grad_norm": 0.40460237860679626,
+      "learning_rate": 0.00017893035088543662,
+      "loss": 0.7058968544006348,
+      "step": 3892
+    },
+    {
+      "epoch": 1.6430379746835442,
+      "grad_norm": 0.39779701828956604,
+      "learning_rate": 0.00017890085167343158,
+      "loss": 0.723969042301178,
+      "step": 3894
+    },
+    {
+      "epoch": 1.6438818565400843,
+      "grad_norm": 0.46131494641304016,
+      "learning_rate": 0.00017887133426051536,
+      "loss": 0.7773922681808472,
+      "step": 3896
+    },
+    {
+      "epoch": 1.6447257383966245,
+      "grad_norm": 0.48412877321243286,
+      "learning_rate": 0.0001788417986534971,
+      "loss": 0.7742162942886353,
+      "step": 3898
+    },
+    {
+      "epoch": 1.6455696202531644,
+      "grad_norm": 0.40083935856819153,
+      "learning_rate": 0.00017881224485919006,
+      "loss": 0.6674869656562805,
+      "step": 3900
+    },
+    {
+      "epoch": 1.6455696202531644,
+      "eval_loss": 0.7763351202011108,
+      "eval_runtime": 682.0824,
+      "eval_samples_per_second": 3.089,
+      "eval_steps_per_second": 3.089,
+      "step": 3900
+    },
+    {
+      "epoch": 1.6464135021097046,
+      "grad_norm": 0.3944476842880249,
+      "learning_rate": 0.00017878267288441176,
+      "loss": 0.6777821183204651,
+      "step": 3902
+    },
+    {
+      "epoch": 1.6472573839662448,
+      "grad_norm": 0.44863224029541016,
+      "learning_rate": 0.0001787530827359839,
+      "loss": 0.7398322224617004,
+      "step": 3904
+    },
+    {
+      "epoch": 1.6481012658227847,
+      "grad_norm": 0.4665529131889343,
+      "learning_rate": 0.00017872347442073232,
+      "loss": 0.7203068733215332,
+      "step": 3906
+    },
+    {
+      "epoch": 1.6489451476793249,
+      "grad_norm": 0.3603050410747528,
+      "learning_rate": 0.00017869384794548708,
+      "loss": 0.6232115030288696,
+      "step": 3908
+    },
+    {
+      "epoch": 1.649789029535865,
+      "grad_norm": 0.3545325696468353,
+      "learning_rate": 0.0001786642033170825,
+      "loss": 0.6120040416717529,
+      "step": 3910
+    },
+    {
+      "epoch": 1.650632911392405,
+      "grad_norm": 0.3947487473487854,
+      "learning_rate": 0.00017863454054235697,
+      "loss": 0.7285997867584229,
+      "step": 3912
+    },
+    {
+      "epoch": 1.6514767932489451,
+      "grad_norm": 0.33153054118156433,
+      "learning_rate": 0.00017860485962815312,
+      "loss": 0.6486109495162964,
+      "step": 3914
+    },
+    {
+      "epoch": 1.6523206751054853,
+      "grad_norm": 0.3886185884475708,
+      "learning_rate": 0.00017857516058131782,
+      "loss": 0.6870376467704773,
+      "step": 3916
+    },
+    {
+      "epoch": 1.6531645569620252,
+      "grad_norm": 0.46399059891700745,
+      "learning_rate": 0.00017854544340870201,
+      "loss": 0.7351298928260803,
+      "step": 3918
+    },
+    {
+      "epoch": 1.6540084388185654,
+      "grad_norm": 0.49280229210853577,
+      "learning_rate": 0.00017851570811716093,
+      "loss": 0.7283915281295776,
+      "step": 3920
+    },
+    {
+      "epoch": 1.6548523206751056,
+      "grad_norm": 0.41570642590522766,
+      "learning_rate": 0.00017848595471355388,
+      "loss": 0.7436906099319458,
+      "step": 3922
+    },
+    {
+      "epoch": 1.6556962025316455,
+      "grad_norm": 0.44591766595840454,
+      "learning_rate": 0.00017845618320474444,
+      "loss": 0.7427505254745483,
+      "step": 3924
+    },
+    {
+      "epoch": 1.6565400843881857,
+      "grad_norm": 0.4804318845272064,
+      "learning_rate": 0.00017842639359760032,
+      "loss": 0.7642725706100464,
+      "step": 3926
+    },
+    {
+      "epoch": 1.6573839662447258,
+      "grad_norm": 0.4164349138736725,
+      "learning_rate": 0.00017839658589899343,
+      "loss": 0.7097790837287903,
+      "step": 3928
+    },
+    {
+      "epoch": 1.6582278481012658,
+      "grad_norm": 0.5307883620262146,
+      "learning_rate": 0.00017836676011579976,
+      "loss": 0.8346033096313477,
+      "step": 3930
+    },
+    {
+      "epoch": 1.659071729957806,
+      "grad_norm": 0.37231504917144775,
+      "learning_rate": 0.0001783369162548996,
+      "loss": 0.6327160596847534,
+      "step": 3932
+    },
+    {
+      "epoch": 1.659915611814346,
+      "grad_norm": 0.4178757965564728,
+      "learning_rate": 0.00017830705432317738,
+      "loss": 0.6441237330436707,
+      "step": 3934
+    },
+    {
+      "epoch": 1.660759493670886,
+      "grad_norm": 0.42698073387145996,
+      "learning_rate": 0.00017827717432752162,
+      "loss": 0.7267951369285583,
+      "step": 3936
+    },
+    {
+      "epoch": 1.6616033755274262,
+      "grad_norm": 0.48763108253479004,
+      "learning_rate": 0.0001782472762748251,
+      "loss": 0.7248380780220032,
+      "step": 3938
+    },
+    {
+      "epoch": 1.6624472573839664,
+      "grad_norm": 0.38882505893707275,
+      "learning_rate": 0.0001782173601719847,
+      "loss": 0.6665275692939758,
+      "step": 3940
+    },
+    {
+      "epoch": 1.6632911392405063,
+      "grad_norm": 0.4684160649776459,
+      "learning_rate": 0.0001781874260259015,
+      "loss": 0.8017846941947937,
+      "step": 3942
+    },
+    {
+      "epoch": 1.6641350210970463,
+      "grad_norm": 0.4200204908847809,
+      "learning_rate": 0.00017815747384348075,
+      "loss": 0.6810902953147888,
+      "step": 3944
+    },
+    {
+      "epoch": 1.6649789029535866,
+      "grad_norm": 0.45843279361724854,
+      "learning_rate": 0.0001781275036316318,
+      "loss": 0.714082658290863,
+      "step": 3946
+    },
+    {
+      "epoch": 1.6658227848101266,
+      "grad_norm": 0.522654116153717,
+      "learning_rate": 0.00017809751539726824,
+      "loss": 0.842085599899292,
+      "step": 3948
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 0.43468809127807617,
+      "learning_rate": 0.00017806750914730777,
+      "loss": 0.6917972564697266,
+      "step": 3950
+    },
+    {
+      "epoch": 1.667510548523207,
+      "grad_norm": 0.4071558117866516,
+      "learning_rate": 0.00017803748488867224,
+      "loss": 0.6676424741744995,
+      "step": 3952
+    },
+    {
+      "epoch": 1.6683544303797468,
+      "grad_norm": 0.4478702247142792,
+      "learning_rate": 0.00017800744262828772,
+      "loss": 0.7563748359680176,
+      "step": 3954
+    },
+    {
+      "epoch": 1.6691983122362868,
+      "grad_norm": 0.42986053228378296,
+      "learning_rate": 0.00017797738237308432,
+      "loss": 0.685714066028595,
+      "step": 3956
+    },
+    {
+      "epoch": 1.6700421940928272,
+      "grad_norm": 0.4410569667816162,
+      "learning_rate": 0.0001779473041299964,
+      "loss": 0.7753605842590332,
+      "step": 3958
+    },
+    {
+      "epoch": 1.6708860759493671,
+      "grad_norm": 0.44055965542793274,
+      "learning_rate": 0.00017791720790596242,
+      "loss": 0.7826018333435059,
+      "step": 3960
+    },
+    {
+      "epoch": 1.671729957805907,
+      "grad_norm": 0.4660222828388214,
+      "learning_rate": 0.00017788709370792502,
+      "loss": 0.7012629508972168,
+      "step": 3962
+    },
+    {
+      "epoch": 1.6725738396624472,
+      "grad_norm": 0.3655732274055481,
+      "learning_rate": 0.00017785696154283093,
+      "loss": 0.6647007465362549,
+      "step": 3964
+    },
+    {
+      "epoch": 1.6734177215189874,
+      "grad_norm": 0.45735979080200195,
+      "learning_rate": 0.0001778268114176311,
+      "loss": 0.7238346338272095,
+      "step": 3966
+    },
+    {
+      "epoch": 1.6742616033755273,
+      "grad_norm": 0.4894043207168579,
+      "learning_rate": 0.00017779664333928053,
+      "loss": 0.7595300078392029,
+      "step": 3968
+    },
+    {
+      "epoch": 1.6751054852320675,
+      "grad_norm": 0.3613647222518921,
+      "learning_rate": 0.00017776645731473848,
+      "loss": 0.6534504294395447,
+      "step": 3970
+    },
+    {
+      "epoch": 1.6759493670886076,
+      "grad_norm": 0.4402320683002472,
+      "learning_rate": 0.00017773625335096828,
+      "loss": 0.7699148058891296,
+      "step": 3972
+    },
+    {
+      "epoch": 1.6767932489451476,
+      "grad_norm": 0.4391469955444336,
+      "learning_rate": 0.0001777060314549374,
+      "loss": 0.771800696849823,
+      "step": 3974
+    },
+    {
+      "epoch": 1.6776371308016877,
+      "grad_norm": 0.41348180174827576,
+      "learning_rate": 0.00017767579163361735,
+      "loss": 0.6957393884658813,
+      "step": 3976
+    },
+    {
+      "epoch": 1.678481012658228,
+      "grad_norm": 0.4115046560764313,
+      "learning_rate": 0.00017764553389398407,
+      "loss": 0.6666664481163025,
+      "step": 3978
+    },
+    {
+      "epoch": 1.6793248945147679,
+      "grad_norm": 0.44647306203842163,
+      "learning_rate": 0.00017761525824301725,
+      "loss": 0.7364352345466614,
+      "step": 3980
+    },
+    {
+      "epoch": 1.680168776371308,
+      "grad_norm": 0.424234002828598,
+      "learning_rate": 0.00017758496468770103,
+      "loss": 0.7229774594306946,
+      "step": 3982
+    },
+    {
+      "epoch": 1.6810126582278482,
+      "grad_norm": 0.4850127696990967,
+      "learning_rate": 0.00017755465323502346,
+      "loss": 0.6992678642272949,
+      "step": 3984
+    },
+    {
+      "epoch": 1.6818565400843881,
+      "grad_norm": 0.44637173414230347,
+      "learning_rate": 0.00017752432389197688,
+      "loss": 0.6957634091377258,
+      "step": 3986
+    },
+    {
+      "epoch": 1.6827004219409283,
+      "grad_norm": 0.477393239736557,
+      "learning_rate": 0.00017749397666555768,
+      "loss": 0.6842859983444214,
+      "step": 3988
+    },
+    {
+      "epoch": 1.6835443037974684,
+      "grad_norm": 0.3634146451950073,
+      "learning_rate": 0.00017746361156276632,
+      "loss": 0.6478199362754822,
+      "step": 3990
+    },
+    {
+      "epoch": 1.6843881856540084,
+      "grad_norm": 0.4806389808654785,
+      "learning_rate": 0.00017743322859060752,
+      "loss": 0.7166973352432251,
+      "step": 3992
+    },
+    {
+      "epoch": 1.6852320675105485,
+      "grad_norm": 0.4273380935192108,
+      "learning_rate": 0.00017740282775609,
+      "loss": 0.6624643206596375,
+      "step": 3994
+    },
+    {
+      "epoch": 1.6860759493670887,
+      "grad_norm": 0.4340716600418091,
+      "learning_rate": 0.0001773724090662267,
+      "loss": 0.6408489346504211,
+      "step": 3996
+    },
+    {
+      "epoch": 1.6869198312236287,
+      "grad_norm": 0.50480717420578,
+      "learning_rate": 0.00017734197252803458,
+      "loss": 0.8096797466278076,
+      "step": 3998
+    },
+    {
+      "epoch": 1.6877637130801688,
+      "grad_norm": 0.4549327790737152,
+      "learning_rate": 0.00017731151814853475,
+      "loss": 0.6902434825897217,
+      "step": 4000
+    },
+    {
+      "epoch": 1.6877637130801688,
+      "eval_loss": 0.7721371650695801,
+      "eval_runtime": 668.395,
+      "eval_samples_per_second": 3.152,
+      "eval_steps_per_second": 3.152,
+      "step": 4000
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 14220,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 6,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 5,
+        "early_stopping_threshold": 0.001
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 0
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.3909119356077015e+18,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-4000/training_args.bin b/sft_qwen_14B_v2/checkpoints/checkpoint-4000/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..0bd5f45757bdc6c7fc2d67718b18675b51983e0e
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-4000/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:afae4a8ce1391f149eb00b14eed8f891c715e892ea546bf754d22db2c2bc5969
+size 4792
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-4500/README.md b/sft_qwen_14B_v2/checkpoints/checkpoint-4500/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..134a3a8531b68adda3fed2a98db824f7c1b7dcfd
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-4500/README.md
@@ -0,0 +1,207 @@
+---
+base_model: Models/Qwen2.5-Coder-14B-CPT
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Models/Qwen2.5-Coder-14B-CPT
+- lora
+- transformers
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.0
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-4500/adapter_config.json b/sft_qwen_14B_v2/checkpoints/checkpoint-4500/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..d6316742bc71d30704e2ccdb9080ed0b8062e454
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-4500/adapter_config.json
@@ -0,0 +1,43 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Models/Qwen2.5-Coder-14B-CPT",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.0",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "v_proj",
+    "k_proj",
+    "q_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-4500/adapter_model.safetensors b/sft_qwen_14B_v2/checkpoints/checkpoint-4500/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..fd323e50fdfb9ae59802526dced950bad901fdf6
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-4500/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c19a29b40a83cc48759a1d5af0d41a375dcc4f15306fcdc65ea10ce6044a6f47
+size 201378736
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-4500/optimizer.pt b/sft_qwen_14B_v2/checkpoints/checkpoint-4500/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..c86e415ae09c37077060d5e0ff8c9e409ae6699a
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-4500/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f775e0e1732e49a2d23041c0f639fa3b90e82287d672f13a34f05f0d39347d1
+size 402976786
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-4500/rng_state.pth b/sft_qwen_14B_v2/checkpoints/checkpoint-4500/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..a963d2d404489f327cf3d47030d84d6198819dce
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-4500/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cadd1d4964d81ffd554b895540fe42c724ce67c8ce385b329c23f9ba4322912d
+size 14244
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-4500/scheduler.pt b/sft_qwen_14B_v2/checkpoints/checkpoint-4500/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..92b48b5d43a9bf67f6d7cc05b7d1d1e29b5181aa
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-4500/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:315c9d74357f9543b8b6474791ceab2fe082042e289fc2e3547f3a4b6c9b01b2
+size 1064
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-4500/trainer_state.json b/sft_qwen_14B_v2/checkpoints/checkpoint-4500/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..7f96ea7ab5456683a9d04285c0eb68c419dc188e
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-4500/trainer_state.json
@@ -0,0 +1,16153 @@
+{
+  "best_global_step": 4500,
+  "best_metric": 0.7600579857826233,
+  "best_model_checkpoint": "task2file/sft_qwen_14B_v2/checkpoints/checkpoint-4500",
+  "epoch": 1.8987341772151898,
+  "eval_steps": 100,
+  "global_step": 4500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0008438818565400844,
+      "grad_norm": 0.5386583805084229,
+      "learning_rate": 1.7574692442882248e-07,
+      "loss": 1.6941628456115723,
+      "step": 2
+    },
+    {
+      "epoch": 0.0016877637130801688,
+      "grad_norm": 0.5477277636528015,
+      "learning_rate": 5.272407732864675e-07,
+      "loss": 1.7132279872894287,
+      "step": 4
+    },
+    {
+      "epoch": 0.002531645569620253,
+      "grad_norm": 0.5390765070915222,
+      "learning_rate": 8.787346221441126e-07,
+      "loss": 1.641180396080017,
+      "step": 6
+    },
+    {
+      "epoch": 0.0033755274261603376,
+      "grad_norm": 0.5023683905601501,
+      "learning_rate": 1.2302284710017575e-06,
+      "loss": 1.5616240501403809,
+      "step": 8
+    },
+    {
+      "epoch": 0.004219409282700422,
+      "grad_norm": 0.4899154603481293,
+      "learning_rate": 1.5817223198594026e-06,
+      "loss": 1.572033405303955,
+      "step": 10
+    },
+    {
+      "epoch": 0.005063291139240506,
+      "grad_norm": 0.5239788293838501,
+      "learning_rate": 1.9332161687170474e-06,
+      "loss": 1.6242921352386475,
+      "step": 12
+    },
+    {
+      "epoch": 0.00590717299578059,
+      "grad_norm": 0.5172926783561707,
+      "learning_rate": 2.2847100175746925e-06,
+      "loss": 1.6800041198730469,
+      "step": 14
+    },
+    {
+      "epoch": 0.006751054852320675,
+      "grad_norm": 0.5539224743843079,
+      "learning_rate": 2.6362038664323376e-06,
+      "loss": 1.6450834274291992,
+      "step": 16
+    },
+    {
+      "epoch": 0.007594936708860759,
+      "grad_norm": 0.5255337953567505,
+      "learning_rate": 2.9876977152899827e-06,
+      "loss": 1.6673263311386108,
+      "step": 18
+    },
+    {
+      "epoch": 0.008438818565400843,
+      "grad_norm": 0.5074548721313477,
+      "learning_rate": 3.3391915641476277e-06,
+      "loss": 1.531802773475647,
+      "step": 20
+    },
+    {
+      "epoch": 0.009282700421940928,
+      "grad_norm": 0.4160279333591461,
+      "learning_rate": 3.6906854130052724e-06,
+      "loss": 1.599354863166809,
+      "step": 22
+    },
+    {
+      "epoch": 0.010126582278481013,
+      "grad_norm": 0.5716474652290344,
+      "learning_rate": 4.0421792618629174e-06,
+      "loss": 1.6700962781906128,
+      "step": 24
+    },
+    {
+      "epoch": 0.010970464135021098,
+      "grad_norm": 0.5148899555206299,
+      "learning_rate": 4.3936731107205625e-06,
+      "loss": 1.66217839717865,
+      "step": 26
+    },
+    {
+      "epoch": 0.01181434599156118,
+      "grad_norm": 0.575722336769104,
+      "learning_rate": 4.7451669595782076e-06,
+      "loss": 1.6692266464233398,
+      "step": 28
+    },
+    {
+      "epoch": 0.012658227848101266,
+      "grad_norm": 0.5345953106880188,
+      "learning_rate": 5.096660808435853e-06,
+      "loss": 1.5518689155578613,
+      "step": 30
+    },
+    {
+      "epoch": 0.01350210970464135,
+      "grad_norm": 0.4462043344974518,
+      "learning_rate": 5.448154657293498e-06,
+      "loss": 1.5930007696151733,
+      "step": 32
+    },
+    {
+      "epoch": 0.014345991561181435,
+      "grad_norm": 0.5119605660438538,
+      "learning_rate": 5.799648506151143e-06,
+      "loss": 1.6069684028625488,
+      "step": 34
+    },
+    {
+      "epoch": 0.015189873417721518,
+      "grad_norm": 0.5328608751296997,
+      "learning_rate": 6.151142355008788e-06,
+      "loss": 1.5838109254837036,
+      "step": 36
+    },
+    {
+      "epoch": 0.016033755274261603,
+      "grad_norm": 0.5065920352935791,
+      "learning_rate": 6.502636203866433e-06,
+      "loss": 1.608130931854248,
+      "step": 38
+    },
+    {
+      "epoch": 0.016877637130801686,
+      "grad_norm": 0.4479359984397888,
+      "learning_rate": 6.854130052724078e-06,
+      "loss": 1.5942182540893555,
+      "step": 40
+    },
+    {
+      "epoch": 0.017721518987341773,
+      "grad_norm": 0.42844903469085693,
+      "learning_rate": 7.205623901581722e-06,
+      "loss": 1.6441553831100464,
+      "step": 42
+    },
+    {
+      "epoch": 0.018565400843881856,
+      "grad_norm": 0.476630836725235,
+      "learning_rate": 7.557117750439367e-06,
+      "loss": 1.6068111658096313,
+      "step": 44
+    },
+    {
+      "epoch": 0.019409282700421943,
+      "grad_norm": 0.4532654881477356,
+      "learning_rate": 7.908611599297012e-06,
+      "loss": 1.6618021726608276,
+      "step": 46
+    },
+    {
+      "epoch": 0.020253164556962026,
+      "grad_norm": 0.3701118230819702,
+      "learning_rate": 8.260105448154657e-06,
+      "loss": 1.4730033874511719,
+      "step": 48
+    },
+    {
+      "epoch": 0.02109704641350211,
+      "grad_norm": 0.38471561670303345,
+      "learning_rate": 8.611599297012302e-06,
+      "loss": 1.4828267097473145,
+      "step": 50
+    },
+    {
+      "epoch": 0.021940928270042195,
+      "grad_norm": 0.3602336347103119,
+      "learning_rate": 8.963093145869948e-06,
+      "loss": 1.3877452611923218,
+      "step": 52
+    },
+    {
+      "epoch": 0.02278481012658228,
+      "grad_norm": 0.40318572521209717,
+      "learning_rate": 9.314586994727593e-06,
+      "loss": 1.49052894115448,
+      "step": 54
+    },
+    {
+      "epoch": 0.02362869198312236,
+      "grad_norm": 0.3223826587200165,
+      "learning_rate": 9.666080843585238e-06,
+      "loss": 1.4912524223327637,
+      "step": 56
+    },
+    {
+      "epoch": 0.024472573839662448,
+      "grad_norm": 0.3873065114021301,
+      "learning_rate": 1.0017574692442883e-05,
+      "loss": 1.526674509048462,
+      "step": 58
+    },
+    {
+      "epoch": 0.02531645569620253,
+      "grad_norm": 0.410159707069397,
+      "learning_rate": 1.0369068541300528e-05,
+      "loss": 1.4480271339416504,
+      "step": 60
+    },
+    {
+      "epoch": 0.026160337552742614,
+      "grad_norm": 0.3632003962993622,
+      "learning_rate": 1.0720562390158173e-05,
+      "loss": 1.4222990274429321,
+      "step": 62
+    },
+    {
+      "epoch": 0.0270042194092827,
+      "grad_norm": 0.33118435740470886,
+      "learning_rate": 1.1072056239015818e-05,
+      "loss": 1.387171745300293,
+      "step": 64
+    },
+    {
+      "epoch": 0.027848101265822784,
+      "grad_norm": 0.3301764726638794,
+      "learning_rate": 1.1423550087873463e-05,
+      "loss": 1.3523777723312378,
+      "step": 66
+    },
+    {
+      "epoch": 0.02869198312236287,
+      "grad_norm": 0.34342435002326965,
+      "learning_rate": 1.1775043936731108e-05,
+      "loss": 1.4515162706375122,
+      "step": 68
+    },
+    {
+      "epoch": 0.029535864978902954,
+      "grad_norm": 0.3243122100830078,
+      "learning_rate": 1.2126537785588753e-05,
+      "loss": 1.3509243726730347,
+      "step": 70
+    },
+    {
+      "epoch": 0.030379746835443037,
+      "grad_norm": 0.3450150787830353,
+      "learning_rate": 1.2478031634446398e-05,
+      "loss": 1.4936245679855347,
+      "step": 72
+    },
+    {
+      "epoch": 0.031223628691983123,
+      "grad_norm": 0.38912028074264526,
+      "learning_rate": 1.2829525483304042e-05,
+      "loss": 1.3419109582901,
+      "step": 74
+    },
+    {
+      "epoch": 0.032067510548523206,
+      "grad_norm": 0.3019310235977173,
+      "learning_rate": 1.3181019332161687e-05,
+      "loss": 1.4284154176712036,
+      "step": 76
+    },
+    {
+      "epoch": 0.03291139240506329,
+      "grad_norm": 0.37803682684898376,
+      "learning_rate": 1.3532513181019332e-05,
+      "loss": 1.4256561994552612,
+      "step": 78
+    },
+    {
+      "epoch": 0.03375527426160337,
+      "grad_norm": 0.34191736578941345,
+      "learning_rate": 1.3884007029876977e-05,
+      "loss": 1.3256909847259521,
+      "step": 80
+    },
+    {
+      "epoch": 0.03459915611814346,
+      "grad_norm": 0.35242700576782227,
+      "learning_rate": 1.4235500878734624e-05,
+      "loss": 1.2710685729980469,
+      "step": 82
+    },
+    {
+      "epoch": 0.035443037974683546,
+      "grad_norm": 0.38094228506088257,
+      "learning_rate": 1.4586994727592269e-05,
+      "loss": 1.253411889076233,
+      "step": 84
+    },
+    {
+      "epoch": 0.036286919831223625,
+      "grad_norm": 0.36837366223335266,
+      "learning_rate": 1.4938488576449914e-05,
+      "loss": 1.3064342737197876,
+      "step": 86
+    },
+    {
+      "epoch": 0.03713080168776371,
+      "grad_norm": 0.3443569242954254,
+      "learning_rate": 1.5289982425307557e-05,
+      "loss": 1.293562412261963,
+      "step": 88
+    },
+    {
+      "epoch": 0.0379746835443038,
+      "grad_norm": 0.3799338936805725,
+      "learning_rate": 1.5641476274165202e-05,
+      "loss": 1.3382648229599,
+      "step": 90
+    },
+    {
+      "epoch": 0.038818565400843885,
+      "grad_norm": 0.40501922369003296,
+      "learning_rate": 1.599297012302285e-05,
+      "loss": 1.3925724029541016,
+      "step": 92
+    },
+    {
+      "epoch": 0.039662447257383965,
+      "grad_norm": 0.4419630467891693,
+      "learning_rate": 1.6344463971880492e-05,
+      "loss": 1.357171893119812,
+      "step": 94
+    },
+    {
+      "epoch": 0.04050632911392405,
+      "grad_norm": 0.3619817793369293,
+      "learning_rate": 1.6695957820738137e-05,
+      "loss": 1.3029985427856445,
+      "step": 96
+    },
+    {
+      "epoch": 0.04135021097046414,
+      "grad_norm": 0.4851357340812683,
+      "learning_rate": 1.7047451669595782e-05,
+      "loss": 1.3498191833496094,
+      "step": 98
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "grad_norm": 0.418658584356308,
+      "learning_rate": 1.7398945518453427e-05,
+      "loss": 1.185287356376648,
+      "step": 100
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "eval_loss": 1.2979938983917236,
+      "eval_runtime": 682.1979,
+      "eval_samples_per_second": 3.089,
+      "eval_steps_per_second": 3.089,
+      "step": 100
+    },
+    {
+      "epoch": 0.043037974683544304,
+      "grad_norm": 0.4464418888092041,
+      "learning_rate": 1.7750439367311073e-05,
+      "loss": 1.2217272520065308,
+      "step": 102
+    },
+    {
+      "epoch": 0.04388185654008439,
+      "grad_norm": 0.4706237316131592,
+      "learning_rate": 1.8101933216168718e-05,
+      "loss": 1.2052050828933716,
+      "step": 104
+    },
+    {
+      "epoch": 0.04472573839662447,
+      "grad_norm": 0.46394404768943787,
+      "learning_rate": 1.8453427065026363e-05,
+      "loss": 1.221343994140625,
+      "step": 106
+    },
+    {
+      "epoch": 0.04556962025316456,
+      "grad_norm": 0.4726889431476593,
+      "learning_rate": 1.8804920913884008e-05,
+      "loss": 1.2387475967407227,
+      "step": 108
+    },
+    {
+      "epoch": 0.046413502109704644,
+      "grad_norm": 0.42130985856056213,
+      "learning_rate": 1.9156414762741653e-05,
+      "loss": 1.2851309776306152,
+      "step": 110
+    },
+    {
+      "epoch": 0.04725738396624472,
+      "grad_norm": 0.4504576623439789,
+      "learning_rate": 1.9507908611599298e-05,
+      "loss": 1.2753145694732666,
+      "step": 112
+    },
+    {
+      "epoch": 0.04810126582278481,
+      "grad_norm": 0.396085262298584,
+      "learning_rate": 1.9859402460456943e-05,
+      "loss": 1.2427717447280884,
+      "step": 114
+    },
+    {
+      "epoch": 0.048945147679324896,
+      "grad_norm": 0.5106491446495056,
+      "learning_rate": 2.0210896309314588e-05,
+      "loss": 1.2943825721740723,
+      "step": 116
+    },
+    {
+      "epoch": 0.049789029535864976,
+      "grad_norm": 0.42351317405700684,
+      "learning_rate": 2.0562390158172233e-05,
+      "loss": 1.263301134109497,
+      "step": 118
+    },
+    {
+      "epoch": 0.05063291139240506,
+      "grad_norm": 0.4403539299964905,
+      "learning_rate": 2.0913884007029878e-05,
+      "loss": 1.2647849321365356,
+      "step": 120
+    },
+    {
+      "epoch": 0.05147679324894515,
+      "grad_norm": 0.5260752439498901,
+      "learning_rate": 2.1265377855887523e-05,
+      "loss": 1.2351393699645996,
+      "step": 122
+    },
+    {
+      "epoch": 0.05232067510548523,
+      "grad_norm": 0.44978851079940796,
+      "learning_rate": 2.1616871704745168e-05,
+      "loss": 1.0384471416473389,
+      "step": 124
+    },
+    {
+      "epoch": 0.053164556962025315,
+      "grad_norm": 0.47732362151145935,
+      "learning_rate": 2.1968365553602813e-05,
+      "loss": 1.1518068313598633,
+      "step": 126
+    },
+    {
+      "epoch": 0.0540084388185654,
+      "grad_norm": 0.5473551750183105,
+      "learning_rate": 2.231985940246046e-05,
+      "loss": 1.2264912128448486,
+      "step": 128
+    },
+    {
+      "epoch": 0.05485232067510549,
+      "grad_norm": 0.4473855197429657,
+      "learning_rate": 2.2671353251318103e-05,
+      "loss": 1.1615246534347534,
+      "step": 130
+    },
+    {
+      "epoch": 0.05569620253164557,
+      "grad_norm": 0.5980377197265625,
+      "learning_rate": 2.302284710017575e-05,
+      "loss": 1.1334880590438843,
+      "step": 132
+    },
+    {
+      "epoch": 0.056540084388185655,
+      "grad_norm": 0.5987792015075684,
+      "learning_rate": 2.3374340949033394e-05,
+      "loss": 1.1546804904937744,
+      "step": 134
+    },
+    {
+      "epoch": 0.05738396624472574,
+      "grad_norm": 0.45355498790740967,
+      "learning_rate": 2.372583479789104e-05,
+      "loss": 1.194953441619873,
+      "step": 136
+    },
+    {
+      "epoch": 0.05822784810126582,
+      "grad_norm": 0.5373698472976685,
+      "learning_rate": 2.4077328646748684e-05,
+      "loss": 1.1067466735839844,
+      "step": 138
+    },
+    {
+      "epoch": 0.05907172995780591,
+      "grad_norm": 0.48734328150749207,
+      "learning_rate": 2.442882249560633e-05,
+      "loss": 1.188468098640442,
+      "step": 140
+    },
+    {
+      "epoch": 0.059915611814345994,
+      "grad_norm": 0.4692173898220062,
+      "learning_rate": 2.478031634446397e-05,
+      "loss": 1.1624362468719482,
+      "step": 142
+    },
+    {
+      "epoch": 0.060759493670886074,
+      "grad_norm": 0.532554030418396,
+      "learning_rate": 2.513181019332162e-05,
+      "loss": 1.0978907346725464,
+      "step": 144
+    },
+    {
+      "epoch": 0.06160337552742616,
+      "grad_norm": 0.5853802561759949,
+      "learning_rate": 2.5483304042179264e-05,
+      "loss": 1.2030781507492065,
+      "step": 146
+    },
+    {
+      "epoch": 0.06244725738396625,
+      "grad_norm": 0.5061611533164978,
+      "learning_rate": 2.583479789103691e-05,
+      "loss": 1.082366943359375,
+      "step": 148
+    },
+    {
+      "epoch": 0.06329113924050633,
+      "grad_norm": 0.49426141381263733,
+      "learning_rate": 2.6186291739894554e-05,
+      "loss": 1.10564386844635,
+      "step": 150
+    },
+    {
+      "epoch": 0.06413502109704641,
+      "grad_norm": 0.5846618413925171,
+      "learning_rate": 2.6537785588752196e-05,
+      "loss": 1.1992807388305664,
+      "step": 152
+    },
+    {
+      "epoch": 0.06497890295358649,
+      "grad_norm": 0.5517552495002747,
+      "learning_rate": 2.6889279437609844e-05,
+      "loss": 1.1757566928863525,
+      "step": 154
+    },
+    {
+      "epoch": 0.06582278481012659,
+      "grad_norm": 0.5667305588722229,
+      "learning_rate": 2.7240773286467486e-05,
+      "loss": 1.0548783540725708,
+      "step": 156
+    },
+    {
+      "epoch": 0.06666666666666667,
+      "grad_norm": 0.6760414242744446,
+      "learning_rate": 2.7592267135325134e-05,
+      "loss": 1.184364914894104,
+      "step": 158
+    },
+    {
+      "epoch": 0.06751054852320675,
+      "grad_norm": 0.5261430740356445,
+      "learning_rate": 2.7943760984182776e-05,
+      "loss": 1.1945042610168457,
+      "step": 160
+    },
+    {
+      "epoch": 0.06835443037974684,
+      "grad_norm": 0.6155015230178833,
+      "learning_rate": 2.8295254833040425e-05,
+      "loss": 1.2021973133087158,
+      "step": 162
+    },
+    {
+      "epoch": 0.06919831223628692,
+      "grad_norm": 0.6131619215011597,
+      "learning_rate": 2.8646748681898066e-05,
+      "loss": 1.144123911857605,
+      "step": 164
+    },
+    {
+      "epoch": 0.070042194092827,
+      "grad_norm": 0.5749185681343079,
+      "learning_rate": 2.8998242530755715e-05,
+      "loss": 1.1329256296157837,
+      "step": 166
+    },
+    {
+      "epoch": 0.07088607594936709,
+      "grad_norm": 0.5243118405342102,
+      "learning_rate": 2.9349736379613356e-05,
+      "loss": 1.0892387628555298,
+      "step": 168
+    },
+    {
+      "epoch": 0.07172995780590717,
+      "grad_norm": 0.7190104722976685,
+      "learning_rate": 2.9701230228471005e-05,
+      "loss": 1.163260817527771,
+      "step": 170
+    },
+    {
+      "epoch": 0.07257383966244725,
+      "grad_norm": 0.5486982464790344,
+      "learning_rate": 3.0052724077328647e-05,
+      "loss": 1.0880777835845947,
+      "step": 172
+    },
+    {
+      "epoch": 0.07341772151898734,
+      "grad_norm": 0.5020889043807983,
+      "learning_rate": 3.0404217926186295e-05,
+      "loss": 1.0433368682861328,
+      "step": 174
+    },
+    {
+      "epoch": 0.07426160337552742,
+      "grad_norm": 0.47329774498939514,
+      "learning_rate": 3.075571177504394e-05,
+      "loss": 1.0528991222381592,
+      "step": 176
+    },
+    {
+      "epoch": 0.0751054852320675,
+      "grad_norm": 0.6635547876358032,
+      "learning_rate": 3.110720562390158e-05,
+      "loss": 1.1627811193466187,
+      "step": 178
+    },
+    {
+      "epoch": 0.0759493670886076,
+      "grad_norm": 0.5624618530273438,
+      "learning_rate": 3.145869947275923e-05,
+      "loss": 1.084869384765625,
+      "step": 180
+    },
+    {
+      "epoch": 0.07679324894514768,
+      "grad_norm": 0.6029536724090576,
+      "learning_rate": 3.181019332161687e-05,
+      "loss": 1.2227671146392822,
+      "step": 182
+    },
+    {
+      "epoch": 0.07763713080168777,
+      "grad_norm": 0.930959939956665,
+      "learning_rate": 3.216168717047452e-05,
+      "loss": 1.0955452919006348,
+      "step": 184
+    },
+    {
+      "epoch": 0.07848101265822785,
+      "grad_norm": 0.5326952338218689,
+      "learning_rate": 3.251318101933216e-05,
+      "loss": 1.0640798807144165,
+      "step": 186
+    },
+    {
+      "epoch": 0.07932489451476793,
+      "grad_norm": 0.5484727621078491,
+      "learning_rate": 3.286467486818981e-05,
+      "loss": 1.0700589418411255,
+      "step": 188
+    },
+    {
+      "epoch": 0.08016877637130802,
+      "grad_norm": 0.605273425579071,
+      "learning_rate": 3.3216168717047456e-05,
+      "loss": 1.1593081951141357,
+      "step": 190
+    },
+    {
+      "epoch": 0.0810126582278481,
+      "grad_norm": 0.5704394578933716,
+      "learning_rate": 3.35676625659051e-05,
+      "loss": 1.1617076396942139,
+      "step": 192
+    },
+    {
+      "epoch": 0.08185654008438818,
+      "grad_norm": 0.5929452180862427,
+      "learning_rate": 3.3919156414762746e-05,
+      "loss": 1.1346839666366577,
+      "step": 194
+    },
+    {
+      "epoch": 0.08270042194092828,
+      "grad_norm": 0.5624077916145325,
+      "learning_rate": 3.427065026362039e-05,
+      "loss": 1.0934710502624512,
+      "step": 196
+    },
+    {
+      "epoch": 0.08354430379746836,
+      "grad_norm": 0.6717425584793091,
+      "learning_rate": 3.4622144112478036e-05,
+      "loss": 1.1810534000396729,
+      "step": 198
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "grad_norm": 0.5120199918746948,
+      "learning_rate": 3.4973637961335674e-05,
+      "loss": 1.1525514125823975,
+      "step": 200
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "eval_loss": 1.142486810684204,
+      "eval_runtime": 668.2356,
+      "eval_samples_per_second": 3.153,
+      "eval_steps_per_second": 3.153,
+      "step": 200
+    },
+    {
+      "epoch": 0.08523206751054853,
+      "grad_norm": 0.5144487023353577,
+      "learning_rate": 3.5325131810193326e-05,
+      "loss": 1.0243735313415527,
+      "step": 202
+    },
+    {
+      "epoch": 0.08607594936708861,
+      "grad_norm": 0.6325069069862366,
+      "learning_rate": 3.5676625659050964e-05,
+      "loss": 1.118743896484375,
+      "step": 204
+    },
+    {
+      "epoch": 0.08691983122362869,
+      "grad_norm": 0.5501633882522583,
+      "learning_rate": 3.6028119507908616e-05,
+      "loss": 1.0380504131317139,
+      "step": 206
+    },
+    {
+      "epoch": 0.08776371308016878,
+      "grad_norm": 0.6133899688720703,
+      "learning_rate": 3.6379613356766254e-05,
+      "loss": 0.9837555885314941,
+      "step": 208
+    },
+    {
+      "epoch": 0.08860759493670886,
+      "grad_norm": 0.5799810886383057,
+      "learning_rate": 3.6731107205623906e-05,
+      "loss": 1.090720295906067,
+      "step": 210
+    },
+    {
+      "epoch": 0.08945147679324894,
+      "grad_norm": 0.6039511561393738,
+      "learning_rate": 3.7082601054481544e-05,
+      "loss": 1.120232343673706,
+      "step": 212
+    },
+    {
+      "epoch": 0.09029535864978903,
+      "grad_norm": 0.5983024835586548,
+      "learning_rate": 3.7434094903339196e-05,
+      "loss": 1.096949815750122,
+      "step": 214
+    },
+    {
+      "epoch": 0.09113924050632911,
+      "grad_norm": 0.5641079545021057,
+      "learning_rate": 3.7785588752196835e-05,
+      "loss": 1.1226298809051514,
+      "step": 216
+    },
+    {
+      "epoch": 0.0919831223628692,
+      "grad_norm": 0.655717134475708,
+      "learning_rate": 3.8137082601054486e-05,
+      "loss": 1.1260643005371094,
+      "step": 218
+    },
+    {
+      "epoch": 0.09282700421940929,
+      "grad_norm": 0.6111898422241211,
+      "learning_rate": 3.848857644991213e-05,
+      "loss": 1.0777709484100342,
+      "step": 220
+    },
+    {
+      "epoch": 0.09367088607594937,
+      "grad_norm": 0.6821302771568298,
+      "learning_rate": 3.884007029876977e-05,
+      "loss": 1.10588800907135,
+      "step": 222
+    },
+    {
+      "epoch": 0.09451476793248945,
+      "grad_norm": 0.693175733089447,
+      "learning_rate": 3.919156414762742e-05,
+      "loss": 1.1498671770095825,
+      "step": 224
+    },
+    {
+      "epoch": 0.09535864978902954,
+      "grad_norm": 0.5288166403770447,
+      "learning_rate": 3.954305799648506e-05,
+      "loss": 1.0587562322616577,
+      "step": 226
+    },
+    {
+      "epoch": 0.09620253164556962,
+      "grad_norm": 0.6882867813110352,
+      "learning_rate": 3.989455184534271e-05,
+      "loss": 1.1107512712478638,
+      "step": 228
+    },
+    {
+      "epoch": 0.0970464135021097,
+      "grad_norm": 0.5834154486656189,
+      "learning_rate": 4.024604569420035e-05,
+      "loss": 1.020510196685791,
+      "step": 230
+    },
+    {
+      "epoch": 0.09789029535864979,
+      "grad_norm": 0.7157064080238342,
+      "learning_rate": 4.0597539543058e-05,
+      "loss": 1.0642449855804443,
+      "step": 232
+    },
+    {
+      "epoch": 0.09873417721518987,
+      "grad_norm": 0.6530708074569702,
+      "learning_rate": 4.094903339191564e-05,
+      "loss": 1.0359872579574585,
+      "step": 234
+    },
+    {
+      "epoch": 0.09957805907172995,
+      "grad_norm": 0.6329686045646667,
+      "learning_rate": 4.130052724077329e-05,
+      "loss": 1.050504446029663,
+      "step": 236
+    },
+    {
+      "epoch": 0.10042194092827005,
+      "grad_norm": 0.6597026586532593,
+      "learning_rate": 4.165202108963093e-05,
+      "loss": 1.2621175050735474,
+      "step": 238
+    },
+    {
+      "epoch": 0.10126582278481013,
+      "grad_norm": 0.6195225119590759,
+      "learning_rate": 4.200351493848858e-05,
+      "loss": 1.1218310594558716,
+      "step": 240
+    },
+    {
+      "epoch": 0.1021097046413502,
+      "grad_norm": 0.6764137744903564,
+      "learning_rate": 4.235500878734622e-05,
+      "loss": 1.1250728368759155,
+      "step": 242
+    },
+    {
+      "epoch": 0.1029535864978903,
+      "grad_norm": 0.552363395690918,
+      "learning_rate": 4.270650263620387e-05,
+      "loss": 1.028212308883667,
+      "step": 244
+    },
+    {
+      "epoch": 0.10379746835443038,
+      "grad_norm": 0.5620495676994324,
+      "learning_rate": 4.305799648506151e-05,
+      "loss": 1.0425450801849365,
+      "step": 246
+    },
+    {
+      "epoch": 0.10464135021097046,
+      "grad_norm": 0.6860032081604004,
+      "learning_rate": 4.3409490333919156e-05,
+      "loss": 1.144278883934021,
+      "step": 248
+    },
+    {
+      "epoch": 0.10548523206751055,
+      "grad_norm": 0.6033259034156799,
+      "learning_rate": 4.37609841827768e-05,
+      "loss": 1.1223982572555542,
+      "step": 250
+    },
+    {
+      "epoch": 0.10632911392405063,
+      "grad_norm": 0.6292146444320679,
+      "learning_rate": 4.4112478031634446e-05,
+      "loss": 1.1609960794448853,
+      "step": 252
+    },
+    {
+      "epoch": 0.10717299578059072,
+      "grad_norm": 0.7982883453369141,
+      "learning_rate": 4.44639718804921e-05,
+      "loss": 1.063547968864441,
+      "step": 254
+    },
+    {
+      "epoch": 0.1080168776371308,
+      "grad_norm": 0.7719110250473022,
+      "learning_rate": 4.4815465729349736e-05,
+      "loss": 1.0719804763793945,
+      "step": 256
+    },
+    {
+      "epoch": 0.10886075949367088,
+      "grad_norm": 0.6101011633872986,
+      "learning_rate": 4.516695957820739e-05,
+      "loss": 1.0778400897979736,
+      "step": 258
+    },
+    {
+      "epoch": 0.10970464135021098,
+      "grad_norm": 0.7300994396209717,
+      "learning_rate": 4.5518453427065026e-05,
+      "loss": 1.2129558324813843,
+      "step": 260
+    },
+    {
+      "epoch": 0.11054852320675106,
+      "grad_norm": 0.8348747491836548,
+      "learning_rate": 4.586994727592268e-05,
+      "loss": 1.221714735031128,
+      "step": 262
+    },
+    {
+      "epoch": 0.11139240506329114,
+      "grad_norm": 0.5445612072944641,
+      "learning_rate": 4.6221441124780316e-05,
+      "loss": 1.0187978744506836,
+      "step": 264
+    },
+    {
+      "epoch": 0.11223628691983123,
+      "grad_norm": 0.6230319738388062,
+      "learning_rate": 4.657293497363797e-05,
+      "loss": 1.096561312675476,
+      "step": 266
+    },
+    {
+      "epoch": 0.11308016877637131,
+      "grad_norm": 0.6231237649917603,
+      "learning_rate": 4.6924428822495606e-05,
+      "loss": 1.089842438697815,
+      "step": 268
+    },
+    {
+      "epoch": 0.11392405063291139,
+      "grad_norm": 0.7178627252578735,
+      "learning_rate": 4.727592267135325e-05,
+      "loss": 1.0696645975112915,
+      "step": 270
+    },
+    {
+      "epoch": 0.11476793248945148,
+      "grad_norm": 0.6895854473114014,
+      "learning_rate": 4.7627416520210896e-05,
+      "loss": 1.0511361360549927,
+      "step": 272
+    },
+    {
+      "epoch": 0.11561181434599156,
+      "grad_norm": 0.6046878695487976,
+      "learning_rate": 4.797891036906854e-05,
+      "loss": 1.1373958587646484,
+      "step": 274
+    },
+    {
+      "epoch": 0.11645569620253164,
+      "grad_norm": 0.6524552702903748,
+      "learning_rate": 4.833040421792619e-05,
+      "loss": 1.0734186172485352,
+      "step": 276
+    },
+    {
+      "epoch": 0.11729957805907174,
+      "grad_norm": 0.6331019997596741,
+      "learning_rate": 4.868189806678383e-05,
+      "loss": 1.123913049697876,
+      "step": 278
+    },
+    {
+      "epoch": 0.11814345991561181,
+      "grad_norm": 0.5919018983840942,
+      "learning_rate": 4.903339191564148e-05,
+      "loss": 1.0635710954666138,
+      "step": 280
+    },
+    {
+      "epoch": 0.1189873417721519,
+      "grad_norm": 0.6067633032798767,
+      "learning_rate": 4.938488576449912e-05,
+      "loss": 1.0429247617721558,
+      "step": 282
+    },
+    {
+      "epoch": 0.11983122362869199,
+      "grad_norm": 0.6583750247955322,
+      "learning_rate": 4.9736379613356774e-05,
+      "loss": 1.1397464275360107,
+      "step": 284
+    },
+    {
+      "epoch": 0.12067510548523207,
+      "grad_norm": 0.6200069785118103,
+      "learning_rate": 5.008787346221442e-05,
+      "loss": 1.0590803623199463,
+      "step": 286
+    },
+    {
+      "epoch": 0.12151898734177215,
+      "grad_norm": 0.6798665523529053,
+      "learning_rate": 5.0439367311072064e-05,
+      "loss": 1.1318789720535278,
+      "step": 288
+    },
+    {
+      "epoch": 0.12236286919831224,
+      "grad_norm": 0.7508794069290161,
+      "learning_rate": 5.07908611599297e-05,
+      "loss": 1.0934956073760986,
+      "step": 290
+    },
+    {
+      "epoch": 0.12320675105485232,
+      "grad_norm": 0.6901452541351318,
+      "learning_rate": 5.114235500878735e-05,
+      "loss": 1.163407802581787,
+      "step": 292
+    },
+    {
+      "epoch": 0.1240506329113924,
+      "grad_norm": 0.6423285603523254,
+      "learning_rate": 5.1493848857644985e-05,
+      "loss": 1.09059476852417,
+      "step": 294
+    },
+    {
+      "epoch": 0.1248945147679325,
+      "grad_norm": 0.6839275360107422,
+      "learning_rate": 5.1845342706502644e-05,
+      "loss": 1.0690211057662964,
+      "step": 296
+    },
+    {
+      "epoch": 0.1257383966244726,
+      "grad_norm": 0.6350128054618835,
+      "learning_rate": 5.219683655536028e-05,
+      "loss": 0.982322096824646,
+      "step": 298
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "grad_norm": 0.7136530876159668,
+      "learning_rate": 5.254833040421793e-05,
+      "loss": 1.1132930517196655,
+      "step": 300
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "eval_loss": 1.0952109098434448,
+      "eval_runtime": 677.0652,
+      "eval_samples_per_second": 3.112,
+      "eval_steps_per_second": 3.112,
+      "step": 300
+    },
+    {
+      "epoch": 0.12742616033755275,
+      "grad_norm": 0.7339721322059631,
+      "learning_rate": 5.289982425307557e-05,
+      "loss": 0.973595917224884,
+      "step": 302
+    },
+    {
+      "epoch": 0.12827004219409283,
+      "grad_norm": 0.5941481590270996,
+      "learning_rate": 5.3251318101933224e-05,
+      "loss": 0.9819849729537964,
+      "step": 304
+    },
+    {
+      "epoch": 0.1291139240506329,
+      "grad_norm": 0.7153938412666321,
+      "learning_rate": 5.360281195079086e-05,
+      "loss": 1.0315470695495605,
+      "step": 306
+    },
+    {
+      "epoch": 0.12995780590717299,
+      "grad_norm": 0.5167180299758911,
+      "learning_rate": 5.395430579964851e-05,
+      "loss": 0.9492001533508301,
+      "step": 308
+    },
+    {
+      "epoch": 0.1308016877637131,
+      "grad_norm": 0.6055944561958313,
+      "learning_rate": 5.430579964850615e-05,
+      "loss": 1.0156209468841553,
+      "step": 310
+    },
+    {
+      "epoch": 0.13164556962025317,
+      "grad_norm": 0.7662386298179626,
+      "learning_rate": 5.4657293497363805e-05,
+      "loss": 1.1791651248931885,
+      "step": 312
+    },
+    {
+      "epoch": 0.13248945147679325,
+      "grad_norm": 0.6065546274185181,
+      "learning_rate": 5.500878734622145e-05,
+      "loss": 1.0009297132492065,
+      "step": 314
+    },
+    {
+      "epoch": 0.13333333333333333,
+      "grad_norm": 0.604225754737854,
+      "learning_rate": 5.536028119507909e-05,
+      "loss": 1.0208244323730469,
+      "step": 316
+    },
+    {
+      "epoch": 0.1341772151898734,
+      "grad_norm": 0.6186763048171997,
+      "learning_rate": 5.571177504393673e-05,
+      "loss": 0.9968416690826416,
+      "step": 318
+    },
+    {
+      "epoch": 0.1350210970464135,
+      "grad_norm": 0.7100363969802856,
+      "learning_rate": 5.606326889279437e-05,
+      "loss": 0.9540256857872009,
+      "step": 320
+    },
+    {
+      "epoch": 0.1358649789029536,
+      "grad_norm": 0.6979711055755615,
+      "learning_rate": 5.641476274165203e-05,
+      "loss": 1.0631953477859497,
+      "step": 322
+    },
+    {
+      "epoch": 0.13670886075949368,
+      "grad_norm": 0.6237109303474426,
+      "learning_rate": 5.676625659050967e-05,
+      "loss": 1.0170501470565796,
+      "step": 324
+    },
+    {
+      "epoch": 0.13755274261603376,
+      "grad_norm": 0.6525548696517944,
+      "learning_rate": 5.711775043936731e-05,
+      "loss": 1.0715603828430176,
+      "step": 326
+    },
+    {
+      "epoch": 0.13839662447257384,
+      "grad_norm": 0.6869221329689026,
+      "learning_rate": 5.746924428822495e-05,
+      "loss": 1.0111541748046875,
+      "step": 328
+    },
+    {
+      "epoch": 0.13924050632911392,
+      "grad_norm": 0.553188145160675,
+      "learning_rate": 5.782073813708261e-05,
+      "loss": 1.0311682224273682,
+      "step": 330
+    },
+    {
+      "epoch": 0.140084388185654,
+      "grad_norm": 0.6760852932929993,
+      "learning_rate": 5.817223198594025e-05,
+      "loss": 1.0213634967803955,
+      "step": 332
+    },
+    {
+      "epoch": 0.1409282700421941,
+      "grad_norm": 0.5907419919967651,
+      "learning_rate": 5.8523725834797894e-05,
+      "loss": 0.9748594164848328,
+      "step": 334
+    },
+    {
+      "epoch": 0.14177215189873418,
+      "grad_norm": 0.7044920921325684,
+      "learning_rate": 5.887521968365554e-05,
+      "loss": 1.05863356590271,
+      "step": 336
+    },
+    {
+      "epoch": 0.14261603375527426,
+      "grad_norm": 0.679073691368103,
+      "learning_rate": 5.922671353251318e-05,
+      "loss": 1.1341127157211304,
+      "step": 338
+    },
+    {
+      "epoch": 0.14345991561181434,
+      "grad_norm": 0.7676237225532532,
+      "learning_rate": 5.957820738137083e-05,
+      "loss": 0.9540836215019226,
+      "step": 340
+    },
+    {
+      "epoch": 0.14430379746835442,
+      "grad_norm": 0.6313899755477905,
+      "learning_rate": 5.9929701230228474e-05,
+      "loss": 1.0585911273956299,
+      "step": 342
+    },
+    {
+      "epoch": 0.1451476793248945,
+      "grad_norm": 0.7123099565505981,
+      "learning_rate": 6.028119507908612e-05,
+      "loss": 1.0760118961334229,
+      "step": 344
+    },
+    {
+      "epoch": 0.1459915611814346,
+      "grad_norm": 0.585935652256012,
+      "learning_rate": 6.063268892794376e-05,
+      "loss": 1.036866307258606,
+      "step": 346
+    },
+    {
+      "epoch": 0.1468354430379747,
+      "grad_norm": 0.5643263459205627,
+      "learning_rate": 6.0984182776801416e-05,
+      "loss": 1.0242938995361328,
+      "step": 348
+    },
+    {
+      "epoch": 0.14767932489451477,
+      "grad_norm": 0.626761794090271,
+      "learning_rate": 6.133567662565906e-05,
+      "loss": 1.0497376918792725,
+      "step": 350
+    },
+    {
+      "epoch": 0.14852320675105485,
+      "grad_norm": 0.5106956958770752,
+      "learning_rate": 6.16871704745167e-05,
+      "loss": 0.9811885356903076,
+      "step": 352
+    },
+    {
+      "epoch": 0.14936708860759493,
+      "grad_norm": 0.6948089003562927,
+      "learning_rate": 6.203866432337434e-05,
+      "loss": 1.0715330839157104,
+      "step": 354
+    },
+    {
+      "epoch": 0.150210970464135,
+      "grad_norm": 0.699713945388794,
+      "learning_rate": 6.239015817223199e-05,
+      "loss": 1.0405226945877075,
+      "step": 356
+    },
+    {
+      "epoch": 0.15105485232067511,
+      "grad_norm": 0.6437667012214661,
+      "learning_rate": 6.274165202108964e-05,
+      "loss": 1.0490930080413818,
+      "step": 358
+    },
+    {
+      "epoch": 0.1518987341772152,
+      "grad_norm": 0.6952699422836304,
+      "learning_rate": 6.309314586994728e-05,
+      "loss": 0.9267548322677612,
+      "step": 360
+    },
+    {
+      "epoch": 0.15274261603375527,
+      "grad_norm": 0.6713186502456665,
+      "learning_rate": 6.344463971880492e-05,
+      "loss": 1.0427420139312744,
+      "step": 362
+    },
+    {
+      "epoch": 0.15358649789029535,
+      "grad_norm": 0.6750379800796509,
+      "learning_rate": 6.379613356766257e-05,
+      "loss": 1.048950433731079,
+      "step": 364
+    },
+    {
+      "epoch": 0.15443037974683543,
+      "grad_norm": 0.6053379774093628,
+      "learning_rate": 6.414762741652022e-05,
+      "loss": 1.0156004428863525,
+      "step": 366
+    },
+    {
+      "epoch": 0.15527426160337554,
+      "grad_norm": 0.8063633441925049,
+      "learning_rate": 6.449912126537786e-05,
+      "loss": 1.0020819902420044,
+      "step": 368
+    },
+    {
+      "epoch": 0.15611814345991562,
+      "grad_norm": 0.8027494549751282,
+      "learning_rate": 6.48506151142355e-05,
+      "loss": 1.055633783340454,
+      "step": 370
+    },
+    {
+      "epoch": 0.1569620253164557,
+      "grad_norm": 0.6580121517181396,
+      "learning_rate": 6.520210896309315e-05,
+      "loss": 1.0149940252304077,
+      "step": 372
+    },
+    {
+      "epoch": 0.15780590717299578,
+      "grad_norm": 0.6561233997344971,
+      "learning_rate": 6.55536028119508e-05,
+      "loss": 0.9769611954689026,
+      "step": 374
+    },
+    {
+      "epoch": 0.15864978902953586,
+      "grad_norm": 0.6444346308708191,
+      "learning_rate": 6.590509666080844e-05,
+      "loss": 0.9099349975585938,
+      "step": 376
+    },
+    {
+      "epoch": 0.15949367088607594,
+      "grad_norm": 0.5879359245300293,
+      "learning_rate": 6.625659050966608e-05,
+      "loss": 1.0797548294067383,
+      "step": 378
+    },
+    {
+      "epoch": 0.16033755274261605,
+      "grad_norm": 0.6994144916534424,
+      "learning_rate": 6.660808435852373e-05,
+      "loss": 1.0336791276931763,
+      "step": 380
+    },
+    {
+      "epoch": 0.16118143459915613,
+      "grad_norm": 0.6128669381141663,
+      "learning_rate": 6.695957820738138e-05,
+      "loss": 1.018118143081665,
+      "step": 382
+    },
+    {
+      "epoch": 0.1620253164556962,
+      "grad_norm": 1.0237540006637573,
+      "learning_rate": 6.731107205623902e-05,
+      "loss": 1.1405497789382935,
+      "step": 384
+    },
+    {
+      "epoch": 0.16286919831223629,
+      "grad_norm": 0.6091578006744385,
+      "learning_rate": 6.766256590509666e-05,
+      "loss": 1.0314189195632935,
+      "step": 386
+    },
+    {
+      "epoch": 0.16371308016877636,
+      "grad_norm": 0.5916037559509277,
+      "learning_rate": 6.801405975395431e-05,
+      "loss": 0.9564052820205688,
+      "step": 388
+    },
+    {
+      "epoch": 0.16455696202531644,
+      "grad_norm": 0.771653950214386,
+      "learning_rate": 6.836555360281195e-05,
+      "loss": 1.0023859739303589,
+      "step": 390
+    },
+    {
+      "epoch": 0.16540084388185655,
+      "grad_norm": 0.654658317565918,
+      "learning_rate": 6.87170474516696e-05,
+      "loss": 1.07024085521698,
+      "step": 392
+    },
+    {
+      "epoch": 0.16624472573839663,
+      "grad_norm": 0.6611968874931335,
+      "learning_rate": 6.906854130052724e-05,
+      "loss": 1.0552500486373901,
+      "step": 394
+    },
+    {
+      "epoch": 0.1670886075949367,
+      "grad_norm": 0.6955893039703369,
+      "learning_rate": 6.942003514938489e-05,
+      "loss": 1.0562875270843506,
+      "step": 396
+    },
+    {
+      "epoch": 0.1679324894514768,
+      "grad_norm": 0.6666058301925659,
+      "learning_rate": 6.977152899824253e-05,
+      "loss": 0.9850592017173767,
+      "step": 398
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "grad_norm": 0.6131711006164551,
+      "learning_rate": 7.012302284710018e-05,
+      "loss": 1.0077755451202393,
+      "step": 400
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "eval_loss": 1.0625108480453491,
+      "eval_runtime": 691.0068,
+      "eval_samples_per_second": 3.049,
+      "eval_steps_per_second": 3.049,
+      "step": 400
+    },
+    {
+      "epoch": 0.16962025316455695,
+      "grad_norm": 0.6286499500274658,
+      "learning_rate": 7.047451669595783e-05,
+      "loss": 1.1012427806854248,
+      "step": 402
+    },
+    {
+      "epoch": 0.17046413502109706,
+      "grad_norm": 0.6639351844787598,
+      "learning_rate": 7.082601054481547e-05,
+      "loss": 1.0379719734191895,
+      "step": 404
+    },
+    {
+      "epoch": 0.17130801687763714,
+      "grad_norm": 0.750401496887207,
+      "learning_rate": 7.117750439367311e-05,
+      "loss": 1.031856656074524,
+      "step": 406
+    },
+    {
+      "epoch": 0.17215189873417722,
+      "grad_norm": 0.8084847331047058,
+      "learning_rate": 7.152899824253075e-05,
+      "loss": 1.0493193864822388,
+      "step": 408
+    },
+    {
+      "epoch": 0.1729957805907173,
+      "grad_norm": 0.7448462247848511,
+      "learning_rate": 7.188049209138841e-05,
+      "loss": 1.1012418270111084,
+      "step": 410
+    },
+    {
+      "epoch": 0.17383966244725738,
+      "grad_norm": 0.5841867923736572,
+      "learning_rate": 7.223198594024605e-05,
+      "loss": 0.9926692247390747,
+      "step": 412
+    },
+    {
+      "epoch": 0.17468354430379746,
+      "grad_norm": 0.7125606536865234,
+      "learning_rate": 7.258347978910369e-05,
+      "loss": 1.0588877201080322,
+      "step": 414
+    },
+    {
+      "epoch": 0.17552742616033756,
+      "grad_norm": 0.5750942230224609,
+      "learning_rate": 7.293497363796134e-05,
+      "loss": 1.038270354270935,
+      "step": 416
+    },
+    {
+      "epoch": 0.17637130801687764,
+      "grad_norm": 0.565444827079773,
+      "learning_rate": 7.328646748681899e-05,
+      "loss": 0.9843021035194397,
+      "step": 418
+    },
+    {
+      "epoch": 0.17721518987341772,
+      "grad_norm": 0.5825693011283875,
+      "learning_rate": 7.363796133567663e-05,
+      "loss": 1.0731632709503174,
+      "step": 420
+    },
+    {
+      "epoch": 0.1780590717299578,
+      "grad_norm": 0.6267391443252563,
+      "learning_rate": 7.398945518453427e-05,
+      "loss": 1.0061273574829102,
+      "step": 422
+    },
+    {
+      "epoch": 0.17890295358649788,
+      "grad_norm": 0.6621372103691101,
+      "learning_rate": 7.434094903339192e-05,
+      "loss": 1.0461612939834595,
+      "step": 424
+    },
+    {
+      "epoch": 0.17974683544303796,
+      "grad_norm": 0.6635435223579407,
+      "learning_rate": 7.469244288224957e-05,
+      "loss": 0.9789207577705383,
+      "step": 426
+    },
+    {
+      "epoch": 0.18059071729957807,
+      "grad_norm": 0.6342346668243408,
+      "learning_rate": 7.504393673110721e-05,
+      "loss": 1.0527069568634033,
+      "step": 428
+    },
+    {
+      "epoch": 0.18143459915611815,
+      "grad_norm": 0.6762149930000305,
+      "learning_rate": 7.539543057996485e-05,
+      "loss": 0.9708702564239502,
+      "step": 430
+    },
+    {
+      "epoch": 0.18227848101265823,
+      "grad_norm": 0.7073282599449158,
+      "learning_rate": 7.57469244288225e-05,
+      "loss": 1.0509834289550781,
+      "step": 432
+    },
+    {
+      "epoch": 0.1831223628691983,
+      "grad_norm": 0.6917856931686401,
+      "learning_rate": 7.609841827768014e-05,
+      "loss": 1.0128819942474365,
+      "step": 434
+    },
+    {
+      "epoch": 0.1839662447257384,
+      "grad_norm": 0.5574942231178284,
+      "learning_rate": 7.644991212653779e-05,
+      "loss": 0.989395797252655,
+      "step": 436
+    },
+    {
+      "epoch": 0.1848101265822785,
+      "grad_norm": 0.640765905380249,
+      "learning_rate": 7.680140597539543e-05,
+      "loss": 0.9846042990684509,
+      "step": 438
+    },
+    {
+      "epoch": 0.18565400843881857,
+      "grad_norm": 0.6699127554893494,
+      "learning_rate": 7.715289982425308e-05,
+      "loss": 1.0344442129135132,
+      "step": 440
+    },
+    {
+      "epoch": 0.18649789029535865,
+      "grad_norm": 0.6164930462837219,
+      "learning_rate": 7.750439367311072e-05,
+      "loss": 1.0179373025894165,
+      "step": 442
+    },
+    {
+      "epoch": 0.18734177215189873,
+      "grad_norm": 0.6880720853805542,
+      "learning_rate": 7.785588752196837e-05,
+      "loss": 1.0518895387649536,
+      "step": 444
+    },
+    {
+      "epoch": 0.1881856540084388,
+      "grad_norm": 0.6501413583755493,
+      "learning_rate": 7.820738137082601e-05,
+      "loss": 1.0442606210708618,
+      "step": 446
+    },
+    {
+      "epoch": 0.1890295358649789,
+      "grad_norm": 0.6076085567474365,
+      "learning_rate": 7.855887521968366e-05,
+      "loss": 0.9828442335128784,
+      "step": 448
+    },
+    {
+      "epoch": 0.189873417721519,
+      "grad_norm": 0.6418202519416809,
+      "learning_rate": 7.89103690685413e-05,
+      "loss": 1.0573710203170776,
+      "step": 450
+    },
+    {
+      "epoch": 0.19071729957805908,
+      "grad_norm": 0.7055076360702515,
+      "learning_rate": 7.926186291739895e-05,
+      "loss": 1.0216103792190552,
+      "step": 452
+    },
+    {
+      "epoch": 0.19156118143459916,
+      "grad_norm": 0.5668330192565918,
+      "learning_rate": 7.961335676625659e-05,
+      "loss": 0.9837722778320312,
+      "step": 454
+    },
+    {
+      "epoch": 0.19240506329113924,
+      "grad_norm": 0.6419380307197571,
+      "learning_rate": 7.996485061511424e-05,
+      "loss": 1.0003894567489624,
+      "step": 456
+    },
+    {
+      "epoch": 0.19324894514767932,
+      "grad_norm": 0.5949198007583618,
+      "learning_rate": 8.031634446397188e-05,
+      "loss": 1.0609031915664673,
+      "step": 458
+    },
+    {
+      "epoch": 0.1940928270042194,
+      "grad_norm": 0.7032039761543274,
+      "learning_rate": 8.066783831282952e-05,
+      "loss": 1.0543403625488281,
+      "step": 460
+    },
+    {
+      "epoch": 0.1949367088607595,
+      "grad_norm": 0.5775868892669678,
+      "learning_rate": 8.101933216168718e-05,
+      "loss": 0.9819303154945374,
+      "step": 462
+    },
+    {
+      "epoch": 0.19578059071729959,
+      "grad_norm": 0.9301062226295471,
+      "learning_rate": 8.137082601054482e-05,
+      "loss": 1.0542067289352417,
+      "step": 464
+    },
+    {
+      "epoch": 0.19662447257383966,
+      "grad_norm": 0.6193217039108276,
+      "learning_rate": 8.172231985940246e-05,
+      "loss": 0.9966341257095337,
+      "step": 466
+    },
+    {
+      "epoch": 0.19746835443037974,
+      "grad_norm": 0.6286146640777588,
+      "learning_rate": 8.20738137082601e-05,
+      "loss": 1.0474121570587158,
+      "step": 468
+    },
+    {
+      "epoch": 0.19831223628691982,
+      "grad_norm": 0.7418972253799438,
+      "learning_rate": 8.242530755711776e-05,
+      "loss": 0.9549239277839661,
+      "step": 470
+    },
+    {
+      "epoch": 0.1991561181434599,
+      "grad_norm": 0.6122808456420898,
+      "learning_rate": 8.27768014059754e-05,
+      "loss": 1.0191338062286377,
+      "step": 472
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.6375362277030945,
+      "learning_rate": 8.312829525483304e-05,
+      "loss": 1.0987539291381836,
+      "step": 474
+    },
+    {
+      "epoch": 0.2008438818565401,
+      "grad_norm": 0.6459513306617737,
+      "learning_rate": 8.347978910369068e-05,
+      "loss": 1.0369136333465576,
+      "step": 476
+    },
+    {
+      "epoch": 0.20168776371308017,
+      "grad_norm": 0.7029640674591064,
+      "learning_rate": 8.383128295254833e-05,
+      "loss": 1.0582096576690674,
+      "step": 478
+    },
+    {
+      "epoch": 0.20253164556962025,
+      "grad_norm": 0.6345387697219849,
+      "learning_rate": 8.418277680140598e-05,
+      "loss": 1.022916316986084,
+      "step": 480
+    },
+    {
+      "epoch": 0.20337552742616033,
+      "grad_norm": 0.5764590501785278,
+      "learning_rate": 8.453427065026362e-05,
+      "loss": 0.973024308681488,
+      "step": 482
+    },
+    {
+      "epoch": 0.2042194092827004,
+      "grad_norm": 0.5884482860565186,
+      "learning_rate": 8.488576449912127e-05,
+      "loss": 1.0292812585830688,
+      "step": 484
+    },
+    {
+      "epoch": 0.20506329113924052,
+      "grad_norm": 0.616357147693634,
+      "learning_rate": 8.523725834797891e-05,
+      "loss": 1.0083447694778442,
+      "step": 486
+    },
+    {
+      "epoch": 0.2059071729957806,
+      "grad_norm": 0.7671196460723877,
+      "learning_rate": 8.558875219683656e-05,
+      "loss": 0.9936985373497009,
+      "step": 488
+    },
+    {
+      "epoch": 0.20675105485232068,
+      "grad_norm": 0.6197299957275391,
+      "learning_rate": 8.59402460456942e-05,
+      "loss": 1.051513910293579,
+      "step": 490
+    },
+    {
+      "epoch": 0.20759493670886076,
+      "grad_norm": 0.6912890672683716,
+      "learning_rate": 8.629173989455185e-05,
+      "loss": 0.9474978446960449,
+      "step": 492
+    },
+    {
+      "epoch": 0.20843881856540084,
+      "grad_norm": 0.6941592693328857,
+      "learning_rate": 8.664323374340949e-05,
+      "loss": 1.0671660900115967,
+      "step": 494
+    },
+    {
+      "epoch": 0.20928270042194091,
+      "grad_norm": 0.5889528393745422,
+      "learning_rate": 8.699472759226714e-05,
+      "loss": 1.0020159482955933,
+      "step": 496
+    },
+    {
+      "epoch": 0.21012658227848102,
+      "grad_norm": 0.6478549838066101,
+      "learning_rate": 8.734622144112478e-05,
+      "loss": 1.0165860652923584,
+      "step": 498
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "grad_norm": 0.6018255949020386,
+      "learning_rate": 8.769771528998243e-05,
+      "loss": 0.8798263072967529,
+      "step": 500
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "eval_loss": 1.042096495628357,
+      "eval_runtime": 692.4361,
+      "eval_samples_per_second": 3.043,
+      "eval_steps_per_second": 3.043,
+      "step": 500
+    },
+    {
+      "epoch": 0.21181434599156118,
+      "grad_norm": 0.578990638256073,
+      "learning_rate": 8.804920913884007e-05,
+      "loss": 1.092096209526062,
+      "step": 502
+    },
+    {
+      "epoch": 0.21265822784810126,
+      "grad_norm": 0.6597883701324463,
+      "learning_rate": 8.840070298769771e-05,
+      "loss": 1.0413451194763184,
+      "step": 504
+    },
+    {
+      "epoch": 0.21350210970464134,
+      "grad_norm": 0.6660305261611938,
+      "learning_rate": 8.875219683655536e-05,
+      "loss": 1.0073142051696777,
+      "step": 506
+    },
+    {
+      "epoch": 0.21434599156118145,
+      "grad_norm": 0.6283115148544312,
+      "learning_rate": 8.910369068541301e-05,
+      "loss": 1.0319768190383911,
+      "step": 508
+    },
+    {
+      "epoch": 0.21518987341772153,
+      "grad_norm": 0.6257343292236328,
+      "learning_rate": 8.945518453427065e-05,
+      "loss": 1.0046353340148926,
+      "step": 510
+    },
+    {
+      "epoch": 0.2160337552742616,
+      "grad_norm": 0.5530875325202942,
+      "learning_rate": 8.980667838312829e-05,
+      "loss": 0.9169099926948547,
+      "step": 512
+    },
+    {
+      "epoch": 0.2168776371308017,
+      "grad_norm": 0.5369633436203003,
+      "learning_rate": 9.015817223198594e-05,
+      "loss": 0.9081505537033081,
+      "step": 514
+    },
+    {
+      "epoch": 0.21772151898734177,
+      "grad_norm": 0.6618232131004333,
+      "learning_rate": 9.05096660808436e-05,
+      "loss": 0.9165045022964478,
+      "step": 516
+    },
+    {
+      "epoch": 0.21856540084388185,
+      "grad_norm": 0.600666344165802,
+      "learning_rate": 9.086115992970123e-05,
+      "loss": 0.91348797082901,
+      "step": 518
+    },
+    {
+      "epoch": 0.21940928270042195,
+      "grad_norm": 0.5919831991195679,
+      "learning_rate": 9.121265377855887e-05,
+      "loss": 1.006508469581604,
+      "step": 520
+    },
+    {
+      "epoch": 0.22025316455696203,
+      "grad_norm": 0.688058614730835,
+      "learning_rate": 9.156414762741654e-05,
+      "loss": 1.0013236999511719,
+      "step": 522
+    },
+    {
+      "epoch": 0.2210970464135021,
+      "grad_norm": 0.6721227765083313,
+      "learning_rate": 9.191564147627418e-05,
+      "loss": 1.0909923315048218,
+      "step": 524
+    },
+    {
+      "epoch": 0.2219409282700422,
+      "grad_norm": 0.5987313389778137,
+      "learning_rate": 9.226713532513181e-05,
+      "loss": 1.0117096900939941,
+      "step": 526
+    },
+    {
+      "epoch": 0.22278481012658227,
+      "grad_norm": 0.6191489696502686,
+      "learning_rate": 9.261862917398945e-05,
+      "loss": 1.0153647661209106,
+      "step": 528
+    },
+    {
+      "epoch": 0.22362869198312235,
+      "grad_norm": 0.6821563243865967,
+      "learning_rate": 9.29701230228471e-05,
+      "loss": 0.9649755954742432,
+      "step": 530
+    },
+    {
+      "epoch": 0.22447257383966246,
+      "grad_norm": 1.760398268699646,
+      "learning_rate": 9.332161687170476e-05,
+      "loss": 0.8673232197761536,
+      "step": 532
+    },
+    {
+      "epoch": 0.22531645569620254,
+      "grad_norm": 0.6670058369636536,
+      "learning_rate": 9.36731107205624e-05,
+      "loss": 0.9942440986633301,
+      "step": 534
+    },
+    {
+      "epoch": 0.22616033755274262,
+      "grad_norm": 0.7345916032791138,
+      "learning_rate": 9.402460456942003e-05,
+      "loss": 1.0364389419555664,
+      "step": 536
+    },
+    {
+      "epoch": 0.2270042194092827,
+      "grad_norm": 0.5946128964424133,
+      "learning_rate": 9.437609841827768e-05,
+      "loss": 0.9314924478530884,
+      "step": 538
+    },
+    {
+      "epoch": 0.22784810126582278,
+      "grad_norm": 0.5800848603248596,
+      "learning_rate": 9.472759226713534e-05,
+      "loss": 1.0694862604141235,
+      "step": 540
+    },
+    {
+      "epoch": 0.22869198312236286,
+      "grad_norm": 0.6712192893028259,
+      "learning_rate": 9.507908611599297e-05,
+      "loss": 1.03531014919281,
+      "step": 542
+    },
+    {
+      "epoch": 0.22953586497890296,
+      "grad_norm": 0.5641416311264038,
+      "learning_rate": 9.543057996485063e-05,
+      "loss": 0.9795235991477966,
+      "step": 544
+    },
+    {
+      "epoch": 0.23037974683544304,
+      "grad_norm": 0.50412517786026,
+      "learning_rate": 9.578207381370826e-05,
+      "loss": 0.9641494750976562,
+      "step": 546
+    },
+    {
+      "epoch": 0.23122362869198312,
+      "grad_norm": 0.579118549823761,
+      "learning_rate": 9.61335676625659e-05,
+      "loss": 0.9375281929969788,
+      "step": 548
+    },
+    {
+      "epoch": 0.2320675105485232,
+      "grad_norm": 0.5888341665267944,
+      "learning_rate": 9.648506151142355e-05,
+      "loss": 0.9414046406745911,
+      "step": 550
+    },
+    {
+      "epoch": 0.23291139240506328,
+      "grad_norm": 0.5595056414604187,
+      "learning_rate": 9.68365553602812e-05,
+      "loss": 0.9005617499351501,
+      "step": 552
+    },
+    {
+      "epoch": 0.23375527426160336,
+      "grad_norm": 0.6605326533317566,
+      "learning_rate": 9.718804920913884e-05,
+      "loss": 1.0283968448638916,
+      "step": 554
+    },
+    {
+      "epoch": 0.23459915611814347,
+      "grad_norm": 0.5657313466072083,
+      "learning_rate": 9.753954305799648e-05,
+      "loss": 1.0058249235153198,
+      "step": 556
+    },
+    {
+      "epoch": 0.23544303797468355,
+      "grad_norm": 0.5433364510536194,
+      "learning_rate": 9.789103690685413e-05,
+      "loss": 0.9835494756698608,
+      "step": 558
+    },
+    {
+      "epoch": 0.23628691983122363,
+      "grad_norm": 0.6129802465438843,
+      "learning_rate": 9.824253075571179e-05,
+      "loss": 1.054532527923584,
+      "step": 560
+    },
+    {
+      "epoch": 0.2371308016877637,
+      "grad_norm": 0.6496239304542542,
+      "learning_rate": 9.859402460456942e-05,
+      "loss": 1.0240973234176636,
+      "step": 562
+    },
+    {
+      "epoch": 0.2379746835443038,
+      "grad_norm": 0.6380873918533325,
+      "learning_rate": 9.894551845342706e-05,
+      "loss": 1.0229179859161377,
+      "step": 564
+    },
+    {
+      "epoch": 0.23881856540084387,
+      "grad_norm": 0.6151993870735168,
+      "learning_rate": 9.929701230228471e-05,
+      "loss": 1.0111570358276367,
+      "step": 566
+    },
+    {
+      "epoch": 0.23966244725738398,
+      "grad_norm": 0.5727584958076477,
+      "learning_rate": 9.964850615114237e-05,
+      "loss": 0.9450829029083252,
+      "step": 568
+    },
+    {
+      "epoch": 0.24050632911392406,
+      "grad_norm": 0.6620725989341736,
+      "learning_rate": 0.0001,
+      "loss": 0.9800319075584412,
+      "step": 570
+    },
+    {
+      "epoch": 0.24135021097046414,
+      "grad_norm": 0.6151163578033447,
+      "learning_rate": 0.00010035149384885764,
+      "loss": 0.9757438898086548,
+      "step": 572
+    },
+    {
+      "epoch": 0.24219409282700421,
+      "grad_norm": 0.5672140717506409,
+      "learning_rate": 0.0001007029876977153,
+      "loss": 0.9104921817779541,
+      "step": 574
+    },
+    {
+      "epoch": 0.2430379746835443,
+      "grad_norm": 0.5697256326675415,
+      "learning_rate": 0.00010105448154657293,
+      "loss": 1.1027376651763916,
+      "step": 576
+    },
+    {
+      "epoch": 0.2438818565400844,
+      "grad_norm": 0.5590381622314453,
+      "learning_rate": 0.00010140597539543057,
+      "loss": 1.0055404901504517,
+      "step": 578
+    },
+    {
+      "epoch": 0.24472573839662448,
+      "grad_norm": 0.5518567562103271,
+      "learning_rate": 0.00010175746924428824,
+      "loss": 1.020835518836975,
+      "step": 580
+    },
+    {
+      "epoch": 0.24556962025316456,
+      "grad_norm": 0.6338496208190918,
+      "learning_rate": 0.00010210896309314588,
+      "loss": 0.9528344869613647,
+      "step": 582
+    },
+    {
+      "epoch": 0.24641350210970464,
+      "grad_norm": 0.6497329473495483,
+      "learning_rate": 0.00010246045694200353,
+      "loss": 1.0088670253753662,
+      "step": 584
+    },
+    {
+      "epoch": 0.24725738396624472,
+      "grad_norm": 0.49888095259666443,
+      "learning_rate": 0.00010281195079086117,
+      "loss": 0.9961200952529907,
+      "step": 586
+    },
+    {
+      "epoch": 0.2481012658227848,
+      "grad_norm": 0.5680158734321594,
+      "learning_rate": 0.0001031634446397188,
+      "loss": 0.9635610580444336,
+      "step": 588
+    },
+    {
+      "epoch": 0.2489451476793249,
+      "grad_norm": 0.658168375492096,
+      "learning_rate": 0.00010351493848857646,
+      "loss": 0.9392287135124207,
+      "step": 590
+    },
+    {
+      "epoch": 0.249789029535865,
+      "grad_norm": 0.618262767791748,
+      "learning_rate": 0.0001038664323374341,
+      "loss": 0.9600516557693481,
+      "step": 592
+    },
+    {
+      "epoch": 0.25063291139240507,
+      "grad_norm": 0.6003909111022949,
+      "learning_rate": 0.00010421792618629173,
+      "loss": 1.005476713180542,
+      "step": 594
+    },
+    {
+      "epoch": 0.2514767932489452,
+      "grad_norm": 0.5437078475952148,
+      "learning_rate": 0.00010456942003514938,
+      "loss": 0.9523017406463623,
+      "step": 596
+    },
+    {
+      "epoch": 0.2523206751054852,
+      "grad_norm": 0.5524541735649109,
+      "learning_rate": 0.00010492091388400705,
+      "loss": 0.9526668787002563,
+      "step": 598
+    },
+    {
+      "epoch": 0.25316455696202533,
+      "grad_norm": 0.679504930973053,
+      "learning_rate": 0.00010527240773286469,
+      "loss": 1.019660472869873,
+      "step": 600
+    },
+    {
+      "epoch": 0.25316455696202533,
+      "eval_loss": 1.0193854570388794,
+      "eval_runtime": 677.9523,
+      "eval_samples_per_second": 3.108,
+      "eval_steps_per_second": 3.108,
+      "step": 600
+    },
+    {
+      "epoch": 0.2540084388185654,
+      "grad_norm": 0.5646136999130249,
+      "learning_rate": 0.00010562390158172233,
+      "loss": 0.9910882711410522,
+      "step": 602
+    },
+    {
+      "epoch": 0.2548523206751055,
+      "grad_norm": 0.5238093137741089,
+      "learning_rate": 0.00010597539543057998,
+      "loss": 0.9616432785987854,
+      "step": 604
+    },
+    {
+      "epoch": 0.25569620253164554,
+      "grad_norm": 0.7483857274055481,
+      "learning_rate": 0.00010632688927943762,
+      "loss": 1.0078275203704834,
+      "step": 606
+    },
+    {
+      "epoch": 0.25654008438818565,
+      "grad_norm": 0.578948974609375,
+      "learning_rate": 0.00010667838312829525,
+      "loss": 0.9827103018760681,
+      "step": 608
+    },
+    {
+      "epoch": 0.25738396624472576,
+      "grad_norm": 0.5525906085968018,
+      "learning_rate": 0.00010702987697715289,
+      "loss": 1.0423277616500854,
+      "step": 610
+    },
+    {
+      "epoch": 0.2582278481012658,
+      "grad_norm": 0.6721326112747192,
+      "learning_rate": 0.00010738137082601054,
+      "loss": 0.9561693072319031,
+      "step": 612
+    },
+    {
+      "epoch": 0.2590717299578059,
+      "grad_norm": 0.5701051354408264,
+      "learning_rate": 0.00010773286467486821,
+      "loss": 0.9602992534637451,
+      "step": 614
+    },
+    {
+      "epoch": 0.25991561181434597,
+      "grad_norm": 0.6349860429763794,
+      "learning_rate": 0.00010808435852372585,
+      "loss": 1.1422650814056396,
+      "step": 616
+    },
+    {
+      "epoch": 0.2607594936708861,
+      "grad_norm": 0.5496085286140442,
+      "learning_rate": 0.00010843585237258349,
+      "loss": 0.9762773513793945,
+      "step": 618
+    },
+    {
+      "epoch": 0.2616033755274262,
+      "grad_norm": 0.6080722808837891,
+      "learning_rate": 0.00010878734622144114,
+      "loss": 1.0133616924285889,
+      "step": 620
+    },
+    {
+      "epoch": 0.26244725738396624,
+      "grad_norm": 0.5450218915939331,
+      "learning_rate": 0.00010913884007029878,
+      "loss": 0.9385587573051453,
+      "step": 622
+    },
+    {
+      "epoch": 0.26329113924050634,
+      "grad_norm": 0.592106819152832,
+      "learning_rate": 0.00010949033391915641,
+      "loss": 0.9359989762306213,
+      "step": 624
+    },
+    {
+      "epoch": 0.2641350210970464,
+      "grad_norm": 0.6449427604675293,
+      "learning_rate": 0.00010984182776801407,
+      "loss": 1.0266027450561523,
+      "step": 626
+    },
+    {
+      "epoch": 0.2649789029535865,
+      "grad_norm": 0.538299560546875,
+      "learning_rate": 0.0001101933216168717,
+      "loss": 0.9303187131881714,
+      "step": 628
+    },
+    {
+      "epoch": 0.26582278481012656,
+      "grad_norm": 0.546316921710968,
+      "learning_rate": 0.00011054481546572934,
+      "loss": 0.9368857145309448,
+      "step": 630
+    },
+    {
+      "epoch": 0.26666666666666666,
+      "grad_norm": 0.5818730592727661,
+      "learning_rate": 0.00011089630931458701,
+      "loss": 0.9573145508766174,
+      "step": 632
+    },
+    {
+      "epoch": 0.26751054852320677,
+      "grad_norm": 0.5958262085914612,
+      "learning_rate": 0.00011124780316344465,
+      "loss": 0.9345449805259705,
+      "step": 634
+    },
+    {
+      "epoch": 0.2683544303797468,
+      "grad_norm": 0.6259077787399292,
+      "learning_rate": 0.0001115992970123023,
+      "loss": 1.0906590223312378,
+      "step": 636
+    },
+    {
+      "epoch": 0.26919831223628693,
+      "grad_norm": 0.589672863483429,
+      "learning_rate": 0.00011195079086115994,
+      "loss": 1.0757447481155396,
+      "step": 638
+    },
+    {
+      "epoch": 0.270042194092827,
+      "grad_norm": 0.5714080333709717,
+      "learning_rate": 0.00011230228471001758,
+      "loss": 0.9310855269432068,
+      "step": 640
+    },
+    {
+      "epoch": 0.2708860759493671,
+      "grad_norm": 0.45342639088630676,
+      "learning_rate": 0.00011265377855887523,
+      "loss": 0.9276360273361206,
+      "step": 642
+    },
+    {
+      "epoch": 0.2717299578059072,
+      "grad_norm": 0.6386750340461731,
+      "learning_rate": 0.00011300527240773287,
+      "loss": 1.084719181060791,
+      "step": 644
+    },
+    {
+      "epoch": 0.27257383966244725,
+      "grad_norm": 0.6446163654327393,
+      "learning_rate": 0.0001133567662565905,
+      "loss": 0.9763918519020081,
+      "step": 646
+    },
+    {
+      "epoch": 0.27341772151898736,
+      "grad_norm": 0.5925686359405518,
+      "learning_rate": 0.00011370826010544816,
+      "loss": 0.9517921805381775,
+      "step": 648
+    },
+    {
+      "epoch": 0.2742616033755274,
+      "grad_norm": 0.5399773716926575,
+      "learning_rate": 0.00011405975395430582,
+      "loss": 1.0587927103042603,
+      "step": 650
+    },
+    {
+      "epoch": 0.2751054852320675,
+      "grad_norm": 0.5872456431388855,
+      "learning_rate": 0.00011441124780316346,
+      "loss": 0.883341908454895,
+      "step": 652
+    },
+    {
+      "epoch": 0.2759493670886076,
+      "grad_norm": 0.5574564337730408,
+      "learning_rate": 0.0001147627416520211,
+      "loss": 1.0306891202926636,
+      "step": 654
+    },
+    {
+      "epoch": 0.2767932489451477,
+      "grad_norm": 0.47789013385772705,
+      "learning_rate": 0.00011511423550087874,
+      "loss": 0.8814032077789307,
+      "step": 656
+    },
+    {
+      "epoch": 0.2776371308016878,
+      "grad_norm": 0.5565530061721802,
+      "learning_rate": 0.00011546572934973639,
+      "loss": 0.9460552334785461,
+      "step": 658
+    },
+    {
+      "epoch": 0.27848101265822783,
+      "grad_norm": 0.5299761295318604,
+      "learning_rate": 0.00011581722319859403,
+      "loss": 0.9475110769271851,
+      "step": 660
+    },
+    {
+      "epoch": 0.27932489451476794,
+      "grad_norm": 0.6503344178199768,
+      "learning_rate": 0.00011616871704745166,
+      "loss": 1.0630913972854614,
+      "step": 662
+    },
+    {
+      "epoch": 0.280168776371308,
+      "grad_norm": 0.5794585943222046,
+      "learning_rate": 0.00011652021089630932,
+      "loss": 0.9389138221740723,
+      "step": 664
+    },
+    {
+      "epoch": 0.2810126582278481,
+      "grad_norm": 0.5762867331504822,
+      "learning_rate": 0.00011687170474516695,
+      "loss": 0.8934136033058167,
+      "step": 666
+    },
+    {
+      "epoch": 0.2818565400843882,
+      "grad_norm": 0.6565435528755188,
+      "learning_rate": 0.00011722319859402462,
+      "loss": 1.1072614192962646,
+      "step": 668
+    },
+    {
+      "epoch": 0.28270042194092826,
+      "grad_norm": 0.5819830298423767,
+      "learning_rate": 0.00011757469244288226,
+      "loss": 1.0501434803009033,
+      "step": 670
+    },
+    {
+      "epoch": 0.28354430379746837,
+      "grad_norm": 0.6071487069129944,
+      "learning_rate": 0.00011792618629173991,
+      "loss": 0.9880793690681458,
+      "step": 672
+    },
+    {
+      "epoch": 0.2843881856540084,
+      "grad_norm": 0.5765058398246765,
+      "learning_rate": 0.00011827768014059755,
+      "loss": 0.9670693874359131,
+      "step": 674
+    },
+    {
+      "epoch": 0.2852320675105485,
+      "grad_norm": 0.5245351791381836,
+      "learning_rate": 0.00011862917398945519,
+      "loss": 0.9602360725402832,
+      "step": 676
+    },
+    {
+      "epoch": 0.28607594936708863,
+      "grad_norm": 0.6189922094345093,
+      "learning_rate": 0.00011898066783831282,
+      "loss": 0.9684560894966125,
+      "step": 678
+    },
+    {
+      "epoch": 0.2869198312236287,
+      "grad_norm": 0.6138690710067749,
+      "learning_rate": 0.00011933216168717048,
+      "loss": 0.9465792775154114,
+      "step": 680
+    },
+    {
+      "epoch": 0.2877637130801688,
+      "grad_norm": 0.5371595621109009,
+      "learning_rate": 0.00011968365553602812,
+      "loss": 0.8495944738388062,
+      "step": 682
+    },
+    {
+      "epoch": 0.28860759493670884,
+      "grad_norm": 0.5549944639205933,
+      "learning_rate": 0.00012003514938488578,
+      "loss": 0.9663267135620117,
+      "step": 684
+    },
+    {
+      "epoch": 0.28945147679324895,
+      "grad_norm": 0.6484189033508301,
+      "learning_rate": 0.00012038664323374342,
+      "loss": 0.9736058712005615,
+      "step": 686
+    },
+    {
+      "epoch": 0.290295358649789,
+      "grad_norm": 0.540351390838623,
+      "learning_rate": 0.00012073813708260107,
+      "loss": 1.0591845512390137,
+      "step": 688
+    },
+    {
+      "epoch": 0.2911392405063291,
+      "grad_norm": 0.5657922029495239,
+      "learning_rate": 0.00012108963093145871,
+      "loss": 0.944908618927002,
+      "step": 690
+    },
+    {
+      "epoch": 0.2919831223628692,
+      "grad_norm": 0.6040505170822144,
+      "learning_rate": 0.00012144112478031635,
+      "loss": 1.0018219947814941,
+      "step": 692
+    },
+    {
+      "epoch": 0.29282700421940927,
+      "grad_norm": 0.5435477495193481,
+      "learning_rate": 0.000121792618629174,
+      "loss": 1.0351502895355225,
+      "step": 694
+    },
+    {
+      "epoch": 0.2936708860759494,
+      "grad_norm": 0.5712518692016602,
+      "learning_rate": 0.00012214411247803164,
+      "loss": 0.9935672283172607,
+      "step": 696
+    },
+    {
+      "epoch": 0.29451476793248943,
+      "grad_norm": 0.6138222813606262,
+      "learning_rate": 0.00012249560632688928,
+      "loss": 1.0165108442306519,
+      "step": 698
+    },
+    {
+      "epoch": 0.29535864978902954,
+      "grad_norm": 0.4495212435722351,
+      "learning_rate": 0.00012284710017574691,
+      "loss": 0.9334425926208496,
+      "step": 700
+    },
+    {
+      "epoch": 0.29535864978902954,
+      "eval_loss": 0.996929407119751,
+      "eval_runtime": 668.6398,
+      "eval_samples_per_second": 3.151,
+      "eval_steps_per_second": 3.151,
+      "step": 700
+    },
+    {
+      "epoch": 0.29620253164556964,
+      "grad_norm": 0.5321539044380188,
+      "learning_rate": 0.00012319859402460458,
+      "loss": 1.0516537427902222,
+      "step": 702
+    },
+    {
+      "epoch": 0.2970464135021097,
+      "grad_norm": 0.5716516971588135,
+      "learning_rate": 0.00012355008787346222,
+      "loss": 0.9387198686599731,
+      "step": 704
+    },
+    {
+      "epoch": 0.2978902953586498,
+      "grad_norm": 0.5617920160293579,
+      "learning_rate": 0.00012390158172231988,
+      "loss": 0.99737948179245,
+      "step": 706
+    },
+    {
+      "epoch": 0.29873417721518986,
+      "grad_norm": 0.4922899007797241,
+      "learning_rate": 0.00012425307557117752,
+      "loss": 0.9955025911331177,
+      "step": 708
+    },
+    {
+      "epoch": 0.29957805907172996,
+      "grad_norm": 0.543501615524292,
+      "learning_rate": 0.00012460456942003516,
+      "loss": 0.9124280214309692,
+      "step": 710
+    },
+    {
+      "epoch": 0.30042194092827,
+      "grad_norm": 0.49590054154396057,
+      "learning_rate": 0.0001249560632688928,
+      "loss": 0.9820216298103333,
+      "step": 712
+    },
+    {
+      "epoch": 0.3012658227848101,
+      "grad_norm": 0.5984305739402771,
+      "learning_rate": 0.00012530755711775044,
+      "loss": 1.0152074098587036,
+      "step": 714
+    },
+    {
+      "epoch": 0.30210970464135023,
+      "grad_norm": 0.9343504905700684,
+      "learning_rate": 0.00012565905096660807,
+      "loss": 1.0577725172042847,
+      "step": 716
+    },
+    {
+      "epoch": 0.3029535864978903,
+      "grad_norm": 0.5118702054023743,
+      "learning_rate": 0.0001260105448154657,
+      "loss": 0.9830358028411865,
+      "step": 718
+    },
+    {
+      "epoch": 0.3037974683544304,
+      "grad_norm": 0.4940392076969147,
+      "learning_rate": 0.00012636203866432338,
+      "loss": 0.9466043710708618,
+      "step": 720
+    },
+    {
+      "epoch": 0.30464135021097044,
+      "grad_norm": 0.5965693593025208,
+      "learning_rate": 0.00012671353251318104,
+      "loss": 1.015270709991455,
+      "step": 722
+    },
+    {
+      "epoch": 0.30548523206751055,
+      "grad_norm": 0.5020529627799988,
+      "learning_rate": 0.00012706502636203868,
+      "loss": 0.9703927636146545,
+      "step": 724
+    },
+    {
+      "epoch": 0.30632911392405066,
+      "grad_norm": 0.6067010164260864,
+      "learning_rate": 0.00012741652021089632,
+      "loss": 1.0255526304244995,
+      "step": 726
+    },
+    {
+      "epoch": 0.3071729957805907,
+      "grad_norm": 0.5931884050369263,
+      "learning_rate": 0.00012776801405975396,
+      "loss": 0.9335633516311646,
+      "step": 728
+    },
+    {
+      "epoch": 0.3080168776371308,
+      "grad_norm": 0.5938752293586731,
+      "learning_rate": 0.0001281195079086116,
+      "loss": 1.0921578407287598,
+      "step": 730
+    },
+    {
+      "epoch": 0.30886075949367087,
+      "grad_norm": 0.49728086590766907,
+      "learning_rate": 0.00012847100175746923,
+      "loss": 0.963066041469574,
+      "step": 732
+    },
+    {
+      "epoch": 0.309704641350211,
+      "grad_norm": 0.5452080965042114,
+      "learning_rate": 0.0001288224956063269,
+      "loss": 0.9513075351715088,
+      "step": 734
+    },
+    {
+      "epoch": 0.3105485232067511,
+      "grad_norm": 0.5497731566429138,
+      "learning_rate": 0.00012917398945518454,
+      "loss": 0.8576077222824097,
+      "step": 736
+    },
+    {
+      "epoch": 0.31139240506329113,
+      "grad_norm": 0.5580397248268127,
+      "learning_rate": 0.0001295254833040422,
+      "loss": 0.9542577862739563,
+      "step": 738
+    },
+    {
+      "epoch": 0.31223628691983124,
+      "grad_norm": 0.5890427827835083,
+      "learning_rate": 0.00012987697715289984,
+      "loss": 0.8992732167243958,
+      "step": 740
+    },
+    {
+      "epoch": 0.3130801687763713,
+      "grad_norm": 0.5942965745925903,
+      "learning_rate": 0.00013022847100175748,
+      "loss": 1.0322896242141724,
+      "step": 742
+    },
+    {
+      "epoch": 0.3139240506329114,
+      "grad_norm": 0.6341713070869446,
+      "learning_rate": 0.00013057996485061512,
+      "loss": 0.9217103719711304,
+      "step": 744
+    },
+    {
+      "epoch": 0.31476793248945145,
+      "grad_norm": 0.5294105410575867,
+      "learning_rate": 0.00013093145869947276,
+      "loss": 0.951789915561676,
+      "step": 746
+    },
+    {
+      "epoch": 0.31561181434599156,
+      "grad_norm": 0.6372058391571045,
+      "learning_rate": 0.0001312829525483304,
+      "loss": 0.9459875226020813,
+      "step": 748
+    },
+    {
+      "epoch": 0.31645569620253167,
+      "grad_norm": 0.5979796648025513,
+      "learning_rate": 0.00013163444639718806,
+      "loss": 0.9626097679138184,
+      "step": 750
+    },
+    {
+      "epoch": 0.3172995780590717,
+      "grad_norm": 0.5682399868965149,
+      "learning_rate": 0.0001319859402460457,
+      "loss": 1.0261781215667725,
+      "step": 752
+    },
+    {
+      "epoch": 0.3181434599156118,
+      "grad_norm": 0.5349125266075134,
+      "learning_rate": 0.00013233743409490336,
+      "loss": 0.9319828152656555,
+      "step": 754
+    },
+    {
+      "epoch": 0.3189873417721519,
+      "grad_norm": 0.6093934178352356,
+      "learning_rate": 0.000132688927943761,
+      "loss": 0.9216550588607788,
+      "step": 756
+    },
+    {
+      "epoch": 0.319831223628692,
+      "grad_norm": 0.5188612341880798,
+      "learning_rate": 0.00013304042179261864,
+      "loss": 0.901739776134491,
+      "step": 758
+    },
+    {
+      "epoch": 0.3206751054852321,
+      "grad_norm": 0.5877130627632141,
+      "learning_rate": 0.00013339191564147628,
+      "loss": 1.0362589359283447,
+      "step": 760
+    },
+    {
+      "epoch": 0.32151898734177214,
+      "grad_norm": 0.5542771816253662,
+      "learning_rate": 0.00013374340949033392,
+      "loss": 0.8787116408348083,
+      "step": 762
+    },
+    {
+      "epoch": 0.32236286919831225,
+      "grad_norm": 0.5084902048110962,
+      "learning_rate": 0.00013409490333919156,
+      "loss": 0.9237037301063538,
+      "step": 764
+    },
+    {
+      "epoch": 0.3232067510548523,
+      "grad_norm": 0.5461528301239014,
+      "learning_rate": 0.00013444639718804922,
+      "loss": 1.0150731801986694,
+      "step": 766
+    },
+    {
+      "epoch": 0.3240506329113924,
+      "grad_norm": 0.53483647108078,
+      "learning_rate": 0.00013479789103690686,
+      "loss": 0.8985214829444885,
+      "step": 768
+    },
+    {
+      "epoch": 0.32489451476793246,
+      "grad_norm": 0.5580531358718872,
+      "learning_rate": 0.0001351493848857645,
+      "loss": 1.0225775241851807,
+      "step": 770
+    },
+    {
+      "epoch": 0.32573839662447257,
+      "grad_norm": 0.5203377604484558,
+      "learning_rate": 0.00013550087873462216,
+      "loss": 0.9571293592453003,
+      "step": 772
+    },
+    {
+      "epoch": 0.3265822784810127,
+      "grad_norm": 0.5049671530723572,
+      "learning_rate": 0.0001358523725834798,
+      "loss": 1.0468909740447998,
+      "step": 774
+    },
+    {
+      "epoch": 0.32742616033755273,
+      "grad_norm": 0.4723063111305237,
+      "learning_rate": 0.00013620386643233744,
+      "loss": 0.7743215560913086,
+      "step": 776
+    },
+    {
+      "epoch": 0.32827004219409284,
+      "grad_norm": 0.6310980916023254,
+      "learning_rate": 0.00013655536028119508,
+      "loss": 1.021510362625122,
+      "step": 778
+    },
+    {
+      "epoch": 0.3291139240506329,
+      "grad_norm": 0.47066664695739746,
+      "learning_rate": 0.00013690685413005274,
+      "loss": 0.9134382605552673,
+      "step": 780
+    },
+    {
+      "epoch": 0.329957805907173,
+      "grad_norm": 0.5725092887878418,
+      "learning_rate": 0.00013725834797891038,
+      "loss": 0.9797834753990173,
+      "step": 782
+    },
+    {
+      "epoch": 0.3308016877637131,
+      "grad_norm": 0.5139563083648682,
+      "learning_rate": 0.00013760984182776802,
+      "loss": 0.9372621178627014,
+      "step": 784
+    },
+    {
+      "epoch": 0.33164556962025316,
+      "grad_norm": 0.5275821685791016,
+      "learning_rate": 0.00013796133567662566,
+      "loss": 0.9528245329856873,
+      "step": 786
+    },
+    {
+      "epoch": 0.33248945147679326,
+      "grad_norm": 0.5702582001686096,
+      "learning_rate": 0.0001383128295254833,
+      "loss": 0.9750176072120667,
+      "step": 788
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.5281293392181396,
+      "learning_rate": 0.00013866432337434096,
+      "loss": 0.9412306547164917,
+      "step": 790
+    },
+    {
+      "epoch": 0.3341772151898734,
+      "grad_norm": 0.5578986406326294,
+      "learning_rate": 0.0001390158172231986,
+      "loss": 0.997580885887146,
+      "step": 792
+    },
+    {
+      "epoch": 0.33502109704641353,
+      "grad_norm": 0.50461345911026,
+      "learning_rate": 0.00013936731107205624,
+      "loss": 0.9082320928573608,
+      "step": 794
+    },
+    {
+      "epoch": 0.3358649789029536,
+      "grad_norm": 0.5258530378341675,
+      "learning_rate": 0.0001397188049209139,
+      "loss": 1.0082844495773315,
+      "step": 796
+    },
+    {
+      "epoch": 0.3367088607594937,
+      "grad_norm": 0.5548169016838074,
+      "learning_rate": 0.00014007029876977154,
+      "loss": 0.9729003310203552,
+      "step": 798
+    },
+    {
+      "epoch": 0.33755274261603374,
+      "grad_norm": 0.48601076006889343,
+      "learning_rate": 0.00014042179261862918,
+      "loss": 0.9099526405334473,
+      "step": 800
+    },
+    {
+      "epoch": 0.33755274261603374,
+      "eval_loss": 0.9800403714179993,
+      "eval_runtime": 678.8306,
+      "eval_samples_per_second": 3.104,
+      "eval_steps_per_second": 3.104,
+      "step": 800
+    },
+    {
+      "epoch": 0.33839662447257385,
+      "grad_norm": 0.5413158535957336,
+      "learning_rate": 0.00014077328646748682,
+      "loss": 0.8610644936561584,
+      "step": 802
+    },
+    {
+      "epoch": 0.3392405063291139,
+      "grad_norm": 0.5147035717964172,
+      "learning_rate": 0.00014112478031634446,
+      "loss": 0.9584825038909912,
+      "step": 804
+    },
+    {
+      "epoch": 0.340084388185654,
+      "grad_norm": 0.5931771397590637,
+      "learning_rate": 0.0001414762741652021,
+      "loss": 1.0142558813095093,
+      "step": 806
+    },
+    {
+      "epoch": 0.3409282700421941,
+      "grad_norm": 0.5178377032279968,
+      "learning_rate": 0.00014182776801405976,
+      "loss": 1.0078763961791992,
+      "step": 808
+    },
+    {
+      "epoch": 0.34177215189873417,
+      "grad_norm": 0.5453237295150757,
+      "learning_rate": 0.0001421792618629174,
+      "loss": 0.9107215404510498,
+      "step": 810
+    },
+    {
+      "epoch": 0.3426160337552743,
+      "grad_norm": 0.5886152982711792,
+      "learning_rate": 0.00014253075571177506,
+      "loss": 0.9981362819671631,
+      "step": 812
+    },
+    {
+      "epoch": 0.3434599156118143,
+      "grad_norm": 0.48040178418159485,
+      "learning_rate": 0.0001428822495606327,
+      "loss": 0.9636131525039673,
+      "step": 814
+    },
+    {
+      "epoch": 0.34430379746835443,
+      "grad_norm": 0.5011753439903259,
+      "learning_rate": 0.00014323374340949034,
+      "loss": 0.9590586423873901,
+      "step": 816
+    },
+    {
+      "epoch": 0.34514767932489454,
+      "grad_norm": 0.57858806848526,
+      "learning_rate": 0.00014358523725834798,
+      "loss": 0.978246808052063,
+      "step": 818
+    },
+    {
+      "epoch": 0.3459915611814346,
+      "grad_norm": 0.46092939376831055,
+      "learning_rate": 0.00014393673110720562,
+      "loss": 0.8549934029579163,
+      "step": 820
+    },
+    {
+      "epoch": 0.3468354430379747,
+      "grad_norm": 0.5756489038467407,
+      "learning_rate": 0.00014428822495606326,
+      "loss": 0.9771265387535095,
+      "step": 822
+    },
+    {
+      "epoch": 0.34767932489451475,
+      "grad_norm": 0.5501731634140015,
+      "learning_rate": 0.00014463971880492092,
+      "loss": 0.8739748001098633,
+      "step": 824
+    },
+    {
+      "epoch": 0.34852320675105486,
+      "grad_norm": 0.5451868176460266,
+      "learning_rate": 0.0001449912126537786,
+      "loss": 0.9129468202590942,
+      "step": 826
+    },
+    {
+      "epoch": 0.3493670886075949,
+      "grad_norm": 0.4624619781970978,
+      "learning_rate": 0.00014534270650263623,
+      "loss": 0.9196704030036926,
+      "step": 828
+    },
+    {
+      "epoch": 0.350210970464135,
+      "grad_norm": 0.520878791809082,
+      "learning_rate": 0.00014569420035149386,
+      "loss": 0.9976527690887451,
+      "step": 830
+    },
+    {
+      "epoch": 0.3510548523206751,
+      "grad_norm": 0.4469171464443207,
+      "learning_rate": 0.0001460456942003515,
+      "loss": 0.7753443717956543,
+      "step": 832
+    },
+    {
+      "epoch": 0.3518987341772152,
+      "grad_norm": 0.5105249881744385,
+      "learning_rate": 0.00014639718804920914,
+      "loss": 0.9584846496582031,
+      "step": 834
+    },
+    {
+      "epoch": 0.3527426160337553,
+      "grad_norm": 0.5043913125991821,
+      "learning_rate": 0.00014674868189806678,
+      "loss": 0.9013500213623047,
+      "step": 836
+    },
+    {
+      "epoch": 0.35358649789029534,
+      "grad_norm": 0.575850784778595,
+      "learning_rate": 0.00014710017574692442,
+      "loss": 0.9775562286376953,
+      "step": 838
+    },
+    {
+      "epoch": 0.35443037974683544,
+      "grad_norm": 0.5128876566886902,
+      "learning_rate": 0.00014745166959578208,
+      "loss": 0.9278940558433533,
+      "step": 840
+    },
+    {
+      "epoch": 0.35527426160337555,
+      "grad_norm": 0.5757885575294495,
+      "learning_rate": 0.00014780316344463975,
+      "loss": 1.0091488361358643,
+      "step": 842
+    },
+    {
+      "epoch": 0.3561181434599156,
+      "grad_norm": 0.500934898853302,
+      "learning_rate": 0.00014815465729349739,
+      "loss": 0.9286836981773376,
+      "step": 844
+    },
+    {
+      "epoch": 0.3569620253164557,
+      "grad_norm": 0.5220686197280884,
+      "learning_rate": 0.00014850615114235502,
+      "loss": 0.9484171867370605,
+      "step": 846
+    },
+    {
+      "epoch": 0.35780590717299576,
+      "grad_norm": 0.5494697690010071,
+      "learning_rate": 0.00014885764499121266,
+      "loss": 1.0556397438049316,
+      "step": 848
+    },
+    {
+      "epoch": 0.35864978902953587,
+      "grad_norm": 0.46633943915367126,
+      "learning_rate": 0.0001492091388400703,
+      "loss": 0.933089017868042,
+      "step": 850
+    },
+    {
+      "epoch": 0.3594936708860759,
+      "grad_norm": 0.4864962697029114,
+      "learning_rate": 0.00014956063268892794,
+      "loss": 1.0016963481903076,
+      "step": 852
+    },
+    {
+      "epoch": 0.36033755274261603,
+      "grad_norm": 0.5032764673233032,
+      "learning_rate": 0.0001499121265377856,
+      "loss": 0.9085348844528198,
+      "step": 854
+    },
+    {
+      "epoch": 0.36118143459915614,
+      "grad_norm": 0.5424998998641968,
+      "learning_rate": 0.00015026362038664324,
+      "loss": 0.923502504825592,
+      "step": 856
+    },
+    {
+      "epoch": 0.3620253164556962,
+      "grad_norm": 0.5181655287742615,
+      "learning_rate": 0.00015061511423550088,
+      "loss": 0.919174313545227,
+      "step": 858
+    },
+    {
+      "epoch": 0.3628691983122363,
+      "grad_norm": 0.5129443407058716,
+      "learning_rate": 0.00015096660808435855,
+      "loss": 0.851981520652771,
+      "step": 860
+    },
+    {
+      "epoch": 0.36371308016877635,
+      "grad_norm": 0.49540698528289795,
+      "learning_rate": 0.00015131810193321618,
+      "loss": 0.8633858561515808,
+      "step": 862
+    },
+    {
+      "epoch": 0.36455696202531646,
+      "grad_norm": 0.4706701934337616,
+      "learning_rate": 0.00015166959578207382,
+      "loss": 0.9473677277565002,
+      "step": 864
+    },
+    {
+      "epoch": 0.36540084388185656,
+      "grad_norm": 0.587704598903656,
+      "learning_rate": 0.00015202108963093146,
+      "loss": 0.942383885383606,
+      "step": 866
+    },
+    {
+      "epoch": 0.3662447257383966,
+      "grad_norm": 0.5851273536682129,
+      "learning_rate": 0.0001523725834797891,
+      "loss": 0.9811251163482666,
+      "step": 868
+    },
+    {
+      "epoch": 0.3670886075949367,
+      "grad_norm": 0.46357613801956177,
+      "learning_rate": 0.00015272407732864676,
+      "loss": 0.8151084184646606,
+      "step": 870
+    },
+    {
+      "epoch": 0.3679324894514768,
+      "grad_norm": 0.47250670194625854,
+      "learning_rate": 0.0001530755711775044,
+      "loss": 0.9023333191871643,
+      "step": 872
+    },
+    {
+      "epoch": 0.3687763713080169,
+      "grad_norm": 0.47510042786598206,
+      "learning_rate": 0.00015342706502636204,
+      "loss": 0.8758499622344971,
+      "step": 874
+    },
+    {
+      "epoch": 0.369620253164557,
+      "grad_norm": 0.5687124729156494,
+      "learning_rate": 0.00015377855887521968,
+      "loss": 0.9837421774864197,
+      "step": 876
+    },
+    {
+      "epoch": 0.37046413502109704,
+      "grad_norm": 0.49064236879348755,
+      "learning_rate": 0.00015413005272407735,
+      "loss": 0.9084216356277466,
+      "step": 878
+    },
+    {
+      "epoch": 0.37130801687763715,
+      "grad_norm": 0.5288164615631104,
+      "learning_rate": 0.00015448154657293498,
+      "loss": 0.8425542712211609,
+      "step": 880
+    },
+    {
+      "epoch": 0.3721518987341772,
+      "grad_norm": 0.5446951985359192,
+      "learning_rate": 0.00015483304042179262,
+      "loss": 1.006197452545166,
+      "step": 882
+    },
+    {
+      "epoch": 0.3729957805907173,
+      "grad_norm": 0.46872642636299133,
+      "learning_rate": 0.00015518453427065026,
+      "loss": 0.9779444932937622,
+      "step": 884
+    },
+    {
+      "epoch": 0.37383966244725736,
+      "grad_norm": 0.4852714240550995,
+      "learning_rate": 0.00015553602811950793,
+      "loss": 0.9099963307380676,
+      "step": 886
+    },
+    {
+      "epoch": 0.37468354430379747,
+      "grad_norm": 0.5219841003417969,
+      "learning_rate": 0.00015588752196836556,
+      "loss": 0.9730570316314697,
+      "step": 888
+    },
+    {
+      "epoch": 0.3755274261603376,
+      "grad_norm": 0.5258626341819763,
+      "learning_rate": 0.0001562390158172232,
+      "loss": 0.9304586052894592,
+      "step": 890
+    },
+    {
+      "epoch": 0.3763713080168776,
+      "grad_norm": 0.4266716241836548,
+      "learning_rate": 0.00015659050966608084,
+      "loss": 0.8680716156959534,
+      "step": 892
+    },
+    {
+      "epoch": 0.37721518987341773,
+      "grad_norm": 0.46361327171325684,
+      "learning_rate": 0.0001569420035149385,
+      "loss": 0.8428018093109131,
+      "step": 894
+    },
+    {
+      "epoch": 0.3780590717299578,
+      "grad_norm": 0.5313687920570374,
+      "learning_rate": 0.00015729349736379614,
+      "loss": 0.8465750217437744,
+      "step": 896
+    },
+    {
+      "epoch": 0.3789029535864979,
+      "grad_norm": 0.47166210412979126,
+      "learning_rate": 0.00015764499121265378,
+      "loss": 0.9120327234268188,
+      "step": 898
+    },
+    {
+      "epoch": 0.379746835443038,
+      "grad_norm": 0.4647318720817566,
+      "learning_rate": 0.00015799648506151145,
+      "loss": 0.8950425982475281,
+      "step": 900
+    },
+    {
+      "epoch": 0.379746835443038,
+      "eval_loss": 0.9643027186393738,
+      "eval_runtime": 691.7929,
+      "eval_samples_per_second": 3.046,
+      "eval_steps_per_second": 3.046,
+      "step": 900
+    },
+    {
+      "epoch": 0.38059071729957805,
+      "grad_norm": 0.5445119738578796,
+      "learning_rate": 0.00015834797891036909,
+      "loss": 0.908163845539093,
+      "step": 902
+    },
+    {
+      "epoch": 0.38143459915611816,
+      "grad_norm": 0.4311858117580414,
+      "learning_rate": 0.00015869947275922672,
+      "loss": 0.8945821523666382,
+      "step": 904
+    },
+    {
+      "epoch": 0.3822784810126582,
+      "grad_norm": 0.5590984225273132,
+      "learning_rate": 0.00015905096660808436,
+      "loss": 0.9478458762168884,
+      "step": 906
+    },
+    {
+      "epoch": 0.3831223628691983,
+      "grad_norm": 0.5470241904258728,
+      "learning_rate": 0.000159402460456942,
+      "loss": 0.9259957671165466,
+      "step": 908
+    },
+    {
+      "epoch": 0.38396624472573837,
+      "grad_norm": 0.5498791337013245,
+      "learning_rate": 0.00015975395430579964,
+      "loss": 0.8824930787086487,
+      "step": 910
+    },
+    {
+      "epoch": 0.3848101265822785,
+      "grad_norm": 0.4779198467731476,
+      "learning_rate": 0.0001601054481546573,
+      "loss": 0.8842340707778931,
+      "step": 912
+    },
+    {
+      "epoch": 0.3856540084388186,
+      "grad_norm": 0.5390620827674866,
+      "learning_rate": 0.00016045694200351494,
+      "loss": 0.92950040102005,
+      "step": 914
+    },
+    {
+      "epoch": 0.38649789029535864,
+      "grad_norm": 0.505519688129425,
+      "learning_rate": 0.0001608084358523726,
+      "loss": 0.8420897126197815,
+      "step": 916
+    },
+    {
+      "epoch": 0.38734177215189874,
+      "grad_norm": 0.4463907778263092,
+      "learning_rate": 0.00016115992970123025,
+      "loss": 0.8256624341011047,
+      "step": 918
+    },
+    {
+      "epoch": 0.3881856540084388,
+      "grad_norm": 0.5321422219276428,
+      "learning_rate": 0.00016151142355008788,
+      "loss": 0.8701168298721313,
+      "step": 920
+    },
+    {
+      "epoch": 0.3890295358649789,
+      "grad_norm": 0.4343073070049286,
+      "learning_rate": 0.00016186291739894552,
+      "loss": 0.9019309878349304,
+      "step": 922
+    },
+    {
+      "epoch": 0.389873417721519,
+      "grad_norm": 0.5311984419822693,
+      "learning_rate": 0.00016221441124780316,
+      "loss": 0.8560551404953003,
+      "step": 924
+    },
+    {
+      "epoch": 0.39071729957805906,
+      "grad_norm": 0.588691771030426,
+      "learning_rate": 0.0001625659050966608,
+      "loss": 0.9700050354003906,
+      "step": 926
+    },
+    {
+      "epoch": 0.39156118143459917,
+      "grad_norm": 0.5425586104393005,
+      "learning_rate": 0.00016291739894551844,
+      "loss": 1.043768048286438,
+      "step": 928
+    },
+    {
+      "epoch": 0.3924050632911392,
+      "grad_norm": 0.5228736996650696,
+      "learning_rate": 0.0001632688927943761,
+      "loss": 0.9501712918281555,
+      "step": 930
+    },
+    {
+      "epoch": 0.39324894514767933,
+      "grad_norm": 0.48960360884666443,
+      "learning_rate": 0.00016362038664323377,
+      "loss": 0.9223058223724365,
+      "step": 932
+    },
+    {
+      "epoch": 0.39409282700421944,
+      "grad_norm": 0.45204755663871765,
+      "learning_rate": 0.0001639718804920914,
+      "loss": 0.9692960977554321,
+      "step": 934
+    },
+    {
+      "epoch": 0.3949367088607595,
+      "grad_norm": 0.5299274921417236,
+      "learning_rate": 0.00016432337434094905,
+      "loss": 0.9467466473579407,
+      "step": 936
+    },
+    {
+      "epoch": 0.3957805907172996,
+      "grad_norm": 0.5607715249061584,
+      "learning_rate": 0.00016467486818980668,
+      "loss": 0.9118053317070007,
+      "step": 938
+    },
+    {
+      "epoch": 0.39662447257383965,
+      "grad_norm": 0.5271831154823303,
+      "learning_rate": 0.00016502636203866432,
+      "loss": 0.9131460189819336,
+      "step": 940
+    },
+    {
+      "epoch": 0.39746835443037976,
+      "grad_norm": 0.5075286030769348,
+      "learning_rate": 0.00016537785588752196,
+      "loss": 0.9358300566673279,
+      "step": 942
+    },
+    {
+      "epoch": 0.3983122362869198,
+      "grad_norm": 0.515731155872345,
+      "learning_rate": 0.00016572934973637963,
+      "loss": 0.8908210396766663,
+      "step": 944
+    },
+    {
+      "epoch": 0.3991561181434599,
+      "grad_norm": 0.4856977164745331,
+      "learning_rate": 0.00016608084358523726,
+      "loss": 0.9775290489196777,
+      "step": 946
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.48846355080604553,
+      "learning_rate": 0.00016643233743409493,
+      "loss": 0.8957490921020508,
+      "step": 948
+    },
+    {
+      "epoch": 0.4008438818565401,
+      "grad_norm": 0.42990800738334656,
+      "learning_rate": 0.00016678383128295257,
+      "loss": 0.9036174416542053,
+      "step": 950
+    },
+    {
+      "epoch": 0.4016877637130802,
+      "grad_norm": 0.49552062153816223,
+      "learning_rate": 0.0001671353251318102,
+      "loss": 0.991032600402832,
+      "step": 952
+    },
+    {
+      "epoch": 0.40253164556962023,
+      "grad_norm": 0.4565040171146393,
+      "learning_rate": 0.00016748681898066784,
+      "loss": 0.823063313961029,
+      "step": 954
+    },
+    {
+      "epoch": 0.40337552742616034,
+      "grad_norm": 0.4290153682231903,
+      "learning_rate": 0.00016783831282952548,
+      "loss": 0.8785063624382019,
+      "step": 956
+    },
+    {
+      "epoch": 0.40421940928270045,
+      "grad_norm": 0.5419702529907227,
+      "learning_rate": 0.00016818980667838312,
+      "loss": 0.8763971924781799,
+      "step": 958
+    },
+    {
+      "epoch": 0.4050632911392405,
+      "grad_norm": 0.5177501440048218,
+      "learning_rate": 0.00016854130052724079,
+      "loss": 0.9470553398132324,
+      "step": 960
+    },
+    {
+      "epoch": 0.4059071729957806,
+      "grad_norm": 0.539725661277771,
+      "learning_rate": 0.00016889279437609842,
+      "loss": 0.9235025644302368,
+      "step": 962
+    },
+    {
+      "epoch": 0.40675105485232066,
+      "grad_norm": 0.5324983596801758,
+      "learning_rate": 0.0001692442882249561,
+      "loss": 1.0248996019363403,
+      "step": 964
+    },
+    {
+      "epoch": 0.40759493670886077,
+      "grad_norm": 0.4936407506465912,
+      "learning_rate": 0.00016959578207381373,
+      "loss": 0.9076873660087585,
+      "step": 966
+    },
+    {
+      "epoch": 0.4084388185654008,
+      "grad_norm": 0.4960501194000244,
+      "learning_rate": 0.00016994727592267137,
+      "loss": 0.9162673950195312,
+      "step": 968
+    },
+    {
+      "epoch": 0.4092827004219409,
+      "grad_norm": 0.45093682408332825,
+      "learning_rate": 0.000170298769771529,
+      "loss": 0.904100775718689,
+      "step": 970
+    },
+    {
+      "epoch": 0.41012658227848103,
+      "grad_norm": 0.4560275077819824,
+      "learning_rate": 0.00017065026362038664,
+      "loss": 0.8633337020874023,
+      "step": 972
+    },
+    {
+      "epoch": 0.4109704641350211,
+      "grad_norm": 0.44885000586509705,
+      "learning_rate": 0.00017100175746924428,
+      "loss": 0.8454209566116333,
+      "step": 974
+    },
+    {
+      "epoch": 0.4118143459915612,
+      "grad_norm": 0.47251659631729126,
+      "learning_rate": 0.00017135325131810195,
+      "loss": 0.824730396270752,
+      "step": 976
+    },
+    {
+      "epoch": 0.41265822784810124,
+      "grad_norm": 0.6597666144371033,
+      "learning_rate": 0.00017170474516695959,
+      "loss": 0.9496501684188843,
+      "step": 978
+    },
+    {
+      "epoch": 0.41350210970464135,
+      "grad_norm": 0.39806297421455383,
+      "learning_rate": 0.00017205623901581722,
+      "loss": 0.9419087171554565,
+      "step": 980
+    },
+    {
+      "epoch": 0.41434599156118146,
+      "grad_norm": 0.48231109976768494,
+      "learning_rate": 0.0001724077328646749,
+      "loss": 0.9182976484298706,
+      "step": 982
+    },
+    {
+      "epoch": 0.4151898734177215,
+      "grad_norm": 0.5438776612281799,
+      "learning_rate": 0.00017275922671353253,
+      "loss": 0.9386967420578003,
+      "step": 984
+    },
+    {
+      "epoch": 0.4160337552742616,
+      "grad_norm": 0.4959667921066284,
+      "learning_rate": 0.00017311072056239017,
+      "loss": 0.897849440574646,
+      "step": 986
+    },
+    {
+      "epoch": 0.41687763713080167,
+      "grad_norm": 0.43533357977867126,
+      "learning_rate": 0.0001734622144112478,
+      "loss": 0.8776953816413879,
+      "step": 988
+    },
+    {
+      "epoch": 0.4177215189873418,
+      "grad_norm": 0.47513946890830994,
+      "learning_rate": 0.00017381370826010547,
+      "loss": 0.9162989854812622,
+      "step": 990
+    },
+    {
+      "epoch": 0.41856540084388183,
+      "grad_norm": 0.4907188415527344,
+      "learning_rate": 0.0001741652021089631,
+      "loss": 0.9482660889625549,
+      "step": 992
+    },
+    {
+      "epoch": 0.41940928270042194,
+      "grad_norm": 0.44499966502189636,
+      "learning_rate": 0.00017451669595782075,
+      "loss": 0.8812930583953857,
+      "step": 994
+    },
+    {
+      "epoch": 0.42025316455696204,
+      "grad_norm": 0.4535730481147766,
+      "learning_rate": 0.00017486818980667838,
+      "loss": 0.9439874887466431,
+      "step": 996
+    },
+    {
+      "epoch": 0.4210970464135021,
+      "grad_norm": 0.5240745544433594,
+      "learning_rate": 0.00017521968365553602,
+      "loss": 0.8818395137786865,
+      "step": 998
+    },
+    {
+      "epoch": 0.4219409282700422,
+      "grad_norm": 0.5301211476325989,
+      "learning_rate": 0.0001755711775043937,
+      "loss": 0.886186957359314,
+      "step": 1000
+    },
+    {
+      "epoch": 0.4219409282700422,
+      "eval_loss": 0.9487298727035522,
+      "eval_runtime": 689.4288,
+      "eval_samples_per_second": 3.056,
+      "eval_steps_per_second": 3.056,
+      "step": 1000
+    },
+    {
+      "epoch": 0.42278481012658226,
+      "grad_norm": 0.47876957058906555,
+      "learning_rate": 0.00017592267135325133,
+      "loss": 0.8814021348953247,
+      "step": 1002
+    },
+    {
+      "epoch": 0.42362869198312236,
+      "grad_norm": 0.4929780960083008,
+      "learning_rate": 0.00017627416520210896,
+      "loss": 0.8295068740844727,
+      "step": 1004
+    },
+    {
+      "epoch": 0.42447257383966247,
+      "grad_norm": 0.41888436675071716,
+      "learning_rate": 0.00017662565905096663,
+      "loss": 0.8364827036857605,
+      "step": 1006
+    },
+    {
+      "epoch": 0.4253164556962025,
+      "grad_norm": 0.5175151824951172,
+      "learning_rate": 0.00017697715289982427,
+      "loss": 1.0167189836502075,
+      "step": 1008
+    },
+    {
+      "epoch": 0.42616033755274263,
+      "grad_norm": 0.4815356135368347,
+      "learning_rate": 0.0001773286467486819,
+      "loss": 0.8460752964019775,
+      "step": 1010
+    },
+    {
+      "epoch": 0.4270042194092827,
+      "grad_norm": 0.5210875272750854,
+      "learning_rate": 0.00017768014059753954,
+      "loss": 0.8595574498176575,
+      "step": 1012
+    },
+    {
+      "epoch": 0.4278481012658228,
+      "grad_norm": 0.4400486946105957,
+      "learning_rate": 0.00017803163444639718,
+      "loss": 0.8764723539352417,
+      "step": 1014
+    },
+    {
+      "epoch": 0.4286919831223629,
+      "grad_norm": 0.5282127857208252,
+      "learning_rate": 0.00017838312829525482,
+      "loss": 0.9706798791885376,
+      "step": 1016
+    },
+    {
+      "epoch": 0.42953586497890295,
+      "grad_norm": 0.47523441910743713,
+      "learning_rate": 0.00017873462214411249,
+      "loss": 0.8912002444267273,
+      "step": 1018
+    },
+    {
+      "epoch": 0.43037974683544306,
+      "grad_norm": 0.49640706181526184,
+      "learning_rate": 0.00017908611599297012,
+      "loss": 0.8835636973381042,
+      "step": 1020
+    },
+    {
+      "epoch": 0.4312236286919831,
+      "grad_norm": 0.5253039002418518,
+      "learning_rate": 0.0001794376098418278,
+      "loss": 0.8711735606193542,
+      "step": 1022
+    },
+    {
+      "epoch": 0.4320675105485232,
+      "grad_norm": 0.49285009503364563,
+      "learning_rate": 0.00017978910369068543,
+      "loss": 0.9064869284629822,
+      "step": 1024
+    },
+    {
+      "epoch": 0.43291139240506327,
+      "grad_norm": 0.48758041858673096,
+      "learning_rate": 0.00018014059753954307,
+      "loss": 0.8993359208106995,
+      "step": 1026
+    },
+    {
+      "epoch": 0.4337552742616034,
+      "grad_norm": 0.4815461039543152,
+      "learning_rate": 0.0001804920913884007,
+      "loss": 0.8747937679290771,
+      "step": 1028
+    },
+    {
+      "epoch": 0.4345991561181435,
+      "grad_norm": 0.4930349886417389,
+      "learning_rate": 0.00018084358523725834,
+      "loss": 0.8879084587097168,
+      "step": 1030
+    },
+    {
+      "epoch": 0.43544303797468353,
+      "grad_norm": 0.49303027987480164,
+      "learning_rate": 0.00018119507908611598,
+      "loss": 1.0023083686828613,
+      "step": 1032
+    },
+    {
+      "epoch": 0.43628691983122364,
+      "grad_norm": 0.5312249064445496,
+      "learning_rate": 0.00018154657293497365,
+      "loss": 0.938680112361908,
+      "step": 1034
+    },
+    {
+      "epoch": 0.4371308016877637,
+      "grad_norm": 0.5174582004547119,
+      "learning_rate": 0.0001818980667838313,
+      "loss": 0.854195773601532,
+      "step": 1036
+    },
+    {
+      "epoch": 0.4379746835443038,
+      "grad_norm": 0.5452545881271362,
+      "learning_rate": 0.00018224956063268895,
+      "loss": 0.9059375524520874,
+      "step": 1038
+    },
+    {
+      "epoch": 0.4388185654008439,
+      "grad_norm": 0.5480839014053345,
+      "learning_rate": 0.0001826010544815466,
+      "loss": 0.9708920121192932,
+      "step": 1040
+    },
+    {
+      "epoch": 0.43966244725738396,
+      "grad_norm": 0.44693151116371155,
+      "learning_rate": 0.00018295254833040423,
+      "loss": 0.8561046123504639,
+      "step": 1042
+    },
+    {
+      "epoch": 0.44050632911392407,
+      "grad_norm": 0.4024234116077423,
+      "learning_rate": 0.00018330404217926187,
+      "loss": 0.8811968564987183,
+      "step": 1044
+    },
+    {
+      "epoch": 0.4413502109704641,
+      "grad_norm": 0.5586408376693726,
+      "learning_rate": 0.0001836555360281195,
+      "loss": 0.978068470954895,
+      "step": 1046
+    },
+    {
+      "epoch": 0.4421940928270042,
+      "grad_norm": 0.4918624758720398,
+      "learning_rate": 0.00018400702987697714,
+      "loss": 0.8640981316566467,
+      "step": 1048
+    },
+    {
+      "epoch": 0.4430379746835443,
+      "grad_norm": 0.4335230588912964,
+      "learning_rate": 0.0001843585237258348,
+      "loss": 0.9077964425086975,
+      "step": 1050
+    },
+    {
+      "epoch": 0.4438818565400844,
+      "grad_norm": 0.5275123715400696,
+      "learning_rate": 0.00018471001757469247,
+      "loss": 0.960682213306427,
+      "step": 1052
+    },
+    {
+      "epoch": 0.4447257383966245,
+      "grad_norm": 0.5144415497779846,
+      "learning_rate": 0.0001850615114235501,
+      "loss": 0.8549577593803406,
+      "step": 1054
+    },
+    {
+      "epoch": 0.44556962025316454,
+      "grad_norm": 0.5166662931442261,
+      "learning_rate": 0.00018541300527240775,
+      "loss": 0.9320827126502991,
+      "step": 1056
+    },
+    {
+      "epoch": 0.44641350210970465,
+      "grad_norm": 0.5011980533599854,
+      "learning_rate": 0.0001857644991212654,
+      "loss": 0.9632431864738464,
+      "step": 1058
+    },
+    {
+      "epoch": 0.4472573839662447,
+      "grad_norm": 0.49312469363212585,
+      "learning_rate": 0.00018611599297012303,
+      "loss": 0.9442946910858154,
+      "step": 1060
+    },
+    {
+      "epoch": 0.4481012658227848,
+      "grad_norm": 0.49958568811416626,
+      "learning_rate": 0.00018646748681898066,
+      "loss": 0.952802300453186,
+      "step": 1062
+    },
+    {
+      "epoch": 0.4489451476793249,
+      "grad_norm": 0.49827462434768677,
+      "learning_rate": 0.00018681898066783833,
+      "loss": 0.9630650877952576,
+      "step": 1064
+    },
+    {
+      "epoch": 0.44978902953586497,
+      "grad_norm": 0.523980438709259,
+      "learning_rate": 0.00018717047451669597,
+      "loss": 0.9081395268440247,
+      "step": 1066
+    },
+    {
+      "epoch": 0.4506329113924051,
+      "grad_norm": 0.5108568668365479,
+      "learning_rate": 0.0001875219683655536,
+      "loss": 0.9398958683013916,
+      "step": 1068
+    },
+    {
+      "epoch": 0.45147679324894513,
+      "grad_norm": 0.4453965127468109,
+      "learning_rate": 0.00018787346221441127,
+      "loss": 0.93592768907547,
+      "step": 1070
+    },
+    {
+      "epoch": 0.45232067510548524,
+      "grad_norm": 0.4675683081150055,
+      "learning_rate": 0.0001882249560632689,
+      "loss": 0.8879633545875549,
+      "step": 1072
+    },
+    {
+      "epoch": 0.4531645569620253,
+      "grad_norm": 0.47398847341537476,
+      "learning_rate": 0.00018857644991212655,
+      "loss": 0.9966004490852356,
+      "step": 1074
+    },
+    {
+      "epoch": 0.4540084388185654,
+      "grad_norm": 0.46053192019462585,
+      "learning_rate": 0.0001889279437609842,
+      "loss": 0.9184179902076721,
+      "step": 1076
+    },
+    {
+      "epoch": 0.4548523206751055,
+      "grad_norm": 0.5601398348808289,
+      "learning_rate": 0.00018927943760984182,
+      "loss": 0.9434974193572998,
+      "step": 1078
+    },
+    {
+      "epoch": 0.45569620253164556,
+      "grad_norm": 0.48422637581825256,
+      "learning_rate": 0.0001896309314586995,
+      "loss": 0.9522465467453003,
+      "step": 1080
+    },
+    {
+      "epoch": 0.45654008438818566,
+      "grad_norm": 0.46280911564826965,
+      "learning_rate": 0.00018998242530755713,
+      "loss": 0.9315434694290161,
+      "step": 1082
+    },
+    {
+      "epoch": 0.4573839662447257,
+      "grad_norm": 0.4658683240413666,
+      "learning_rate": 0.00019033391915641477,
+      "loss": 0.8165783286094666,
+      "step": 1084
+    },
+    {
+      "epoch": 0.4582278481012658,
+      "grad_norm": 0.4192182123661041,
+      "learning_rate": 0.0001906854130052724,
+      "loss": 0.8024274110794067,
+      "step": 1086
+    },
+    {
+      "epoch": 0.45907172995780593,
+      "grad_norm": 0.4178735613822937,
+      "learning_rate": 0.00019103690685413007,
+      "loss": 0.8620653748512268,
+      "step": 1088
+    },
+    {
+      "epoch": 0.459915611814346,
+      "grad_norm": 0.4488574266433716,
+      "learning_rate": 0.0001913884007029877,
+      "loss": 0.9121530055999756,
+      "step": 1090
+    },
+    {
+      "epoch": 0.4607594936708861,
+      "grad_norm": 0.5164965987205505,
+      "learning_rate": 0.00019173989455184535,
+      "loss": 0.9496700763702393,
+      "step": 1092
+    },
+    {
+      "epoch": 0.46160337552742614,
+      "grad_norm": 0.41563132405281067,
+      "learning_rate": 0.00019209138840070299,
+      "loss": 0.8757708668708801,
+      "step": 1094
+    },
+    {
+      "epoch": 0.46244725738396625,
+      "grad_norm": 2.3222429752349854,
+      "learning_rate": 0.00019244288224956065,
+      "loss": 0.8406533002853394,
+      "step": 1096
+    },
+    {
+      "epoch": 0.46329113924050636,
+      "grad_norm": 0.4813845157623291,
+      "learning_rate": 0.0001927943760984183,
+      "loss": 0.9459465742111206,
+      "step": 1098
+    },
+    {
+      "epoch": 0.4641350210970464,
+      "grad_norm": 0.6233882308006287,
+      "learning_rate": 0.00019314586994727593,
+      "loss": 0.9376904368400574,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4641350210970464,
+      "eval_loss": 0.9357889294624329,
+      "eval_runtime": 676.9573,
+      "eval_samples_per_second": 3.112,
+      "eval_steps_per_second": 3.112,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4649789029535865,
+      "grad_norm": 0.5125579237937927,
+      "learning_rate": 0.00019349736379613357,
+      "loss": 0.8998825550079346,
+      "step": 1102
+    },
+    {
+      "epoch": 0.46582278481012657,
+      "grad_norm": 0.4534320831298828,
+      "learning_rate": 0.00019384885764499123,
+      "loss": 0.9213768839836121,
+      "step": 1104
+    },
+    {
+      "epoch": 0.4666666666666667,
+      "grad_norm": 0.4715143144130707,
+      "learning_rate": 0.00019420035149384887,
+      "loss": 0.9739661812782288,
+      "step": 1106
+    },
+    {
+      "epoch": 0.4675105485232067,
+      "grad_norm": 0.45529672503471375,
+      "learning_rate": 0.0001945518453427065,
+      "loss": 0.9113216400146484,
+      "step": 1108
+    },
+    {
+      "epoch": 0.46835443037974683,
+      "grad_norm": 0.48199015855789185,
+      "learning_rate": 0.00019490333919156417,
+      "loss": 0.92528235912323,
+      "step": 1110
+    },
+    {
+      "epoch": 0.46919831223628694,
+      "grad_norm": 0.4425188899040222,
+      "learning_rate": 0.0001952548330404218,
+      "loss": 0.8612716197967529,
+      "step": 1112
+    },
+    {
+      "epoch": 0.470042194092827,
+      "grad_norm": 0.4311593770980835,
+      "learning_rate": 0.00019560632688927945,
+      "loss": 0.8901699185371399,
+      "step": 1114
+    },
+    {
+      "epoch": 0.4708860759493671,
+      "grad_norm": 0.497806578874588,
+      "learning_rate": 0.0001959578207381371,
+      "loss": 0.9271994829177856,
+      "step": 1116
+    },
+    {
+      "epoch": 0.47172995780590715,
+      "grad_norm": 0.47149473428726196,
+      "learning_rate": 0.00019630931458699473,
+      "loss": 0.8740925788879395,
+      "step": 1118
+    },
+    {
+      "epoch": 0.47257383966244726,
+      "grad_norm": 0.5029966235160828,
+      "learning_rate": 0.00019666080843585236,
+      "loss": 0.9190115928649902,
+      "step": 1120
+    },
+    {
+      "epoch": 0.47341772151898737,
+      "grad_norm": 0.47972601652145386,
+      "learning_rate": 0.00019701230228471003,
+      "loss": 1.0043057203292847,
+      "step": 1122
+    },
+    {
+      "epoch": 0.4742616033755274,
+      "grad_norm": 0.45096471905708313,
+      "learning_rate": 0.00019736379613356767,
+      "loss": 0.9472925066947937,
+      "step": 1124
+    },
+    {
+      "epoch": 0.4751054852320675,
+      "grad_norm": 0.459852933883667,
+      "learning_rate": 0.00019771528998242533,
+      "loss": 0.8043124079704285,
+      "step": 1126
+    },
+    {
+      "epoch": 0.4759493670886076,
+      "grad_norm": 0.5313422679901123,
+      "learning_rate": 0.00019806678383128297,
+      "loss": 0.9662142395973206,
+      "step": 1128
+    },
+    {
+      "epoch": 0.4767932489451477,
+      "grad_norm": 0.43474531173706055,
+      "learning_rate": 0.0001984182776801406,
+      "loss": 0.8691151738166809,
+      "step": 1130
+    },
+    {
+      "epoch": 0.47763713080168774,
+      "grad_norm": 0.5232312083244324,
+      "learning_rate": 0.00019876977152899825,
+      "loss": 0.9024254083633423,
+      "step": 1132
+    },
+    {
+      "epoch": 0.47848101265822784,
+      "grad_norm": 0.47671905159950256,
+      "learning_rate": 0.0001991212653778559,
+      "loss": 0.8996873497962952,
+      "step": 1134
+    },
+    {
+      "epoch": 0.47932489451476795,
+      "grad_norm": 0.42289480566978455,
+      "learning_rate": 0.00019947275922671353,
+      "loss": 0.8336917757987976,
+      "step": 1136
+    },
+    {
+      "epoch": 0.480168776371308,
+      "grad_norm": 0.4700844883918762,
+      "learning_rate": 0.0001998242530755712,
+      "loss": 0.8943206071853638,
+      "step": 1138
+    },
+    {
+      "epoch": 0.4810126582278481,
+      "grad_norm": 0.5341399312019348,
+      "learning_rate": 0.00019999999711649004,
+      "loss": 0.9340365529060364,
+      "step": 1140
+    },
+    {
+      "epoch": 0.48185654008438816,
+      "grad_norm": 0.46169522404670715,
+      "learning_rate": 0.00019999997404841123,
+      "loss": 0.8241778016090393,
+      "step": 1142
+    },
+    {
+      "epoch": 0.48270042194092827,
+      "grad_norm": 0.5475223064422607,
+      "learning_rate": 0.00019999992791225896,
+      "loss": 1.0096158981323242,
+      "step": 1144
+    },
+    {
+      "epoch": 0.4835443037974684,
+      "grad_norm": 0.524641215801239,
+      "learning_rate": 0.00019999985870804385,
+      "loss": 0.9650378227233887,
+      "step": 1146
+    },
+    {
+      "epoch": 0.48438818565400843,
+      "grad_norm": 0.4326174855232239,
+      "learning_rate": 0.00019999976643578186,
+      "loss": 0.9003009796142578,
+      "step": 1148
+    },
+    {
+      "epoch": 0.48523206751054854,
+      "grad_norm": 0.49034059047698975,
+      "learning_rate": 0.0001999996510954943,
+      "loss": 0.9793432950973511,
+      "step": 1150
+    },
+    {
+      "epoch": 0.4860759493670886,
+      "grad_norm": 0.4760092496871948,
+      "learning_rate": 0.00019999951268720776,
+      "loss": 0.8793007731437683,
+      "step": 1152
+    },
+    {
+      "epoch": 0.4869198312236287,
+      "grad_norm": 0.44606879353523254,
+      "learning_rate": 0.00019999935121095417,
+      "loss": 0.888630747795105,
+      "step": 1154
+    },
+    {
+      "epoch": 0.4877637130801688,
+      "grad_norm": 0.5030332207679749,
+      "learning_rate": 0.0001999991666667708,
+      "loss": 0.9660000801086426,
+      "step": 1156
+    },
+    {
+      "epoch": 0.48860759493670886,
+      "grad_norm": 0.4295555055141449,
+      "learning_rate": 0.00019999895905470014,
+      "loss": 0.824654757976532,
+      "step": 1158
+    },
+    {
+      "epoch": 0.48945147679324896,
+      "grad_norm": 0.392167866230011,
+      "learning_rate": 0.0001999987283747902,
+      "loss": 0.8159562349319458,
+      "step": 1160
+    },
+    {
+      "epoch": 0.490295358649789,
+      "grad_norm": 0.484611839056015,
+      "learning_rate": 0.00019999847462709412,
+      "loss": 0.9630686044692993,
+      "step": 1162
+    },
+    {
+      "epoch": 0.4911392405063291,
+      "grad_norm": 0.45891445875167847,
+      "learning_rate": 0.00019999819781167042,
+      "loss": 0.8396129608154297,
+      "step": 1164
+    },
+    {
+      "epoch": 0.4919831223628692,
+      "grad_norm": 0.5553452968597412,
+      "learning_rate": 0.00019999789792858304,
+      "loss": 0.8194513916969299,
+      "step": 1166
+    },
+    {
+      "epoch": 0.4928270042194093,
+      "grad_norm": 0.38998672366142273,
+      "learning_rate": 0.00019999757497790106,
+      "loss": 0.9268721342086792,
+      "step": 1168
+    },
+    {
+      "epoch": 0.4936708860759494,
+      "grad_norm": 0.45445355772972107,
+      "learning_rate": 0.00019999722895969904,
+      "loss": 0.9204684495925903,
+      "step": 1170
+    },
+    {
+      "epoch": 0.49451476793248944,
+      "grad_norm": 0.484225332736969,
+      "learning_rate": 0.00019999685987405678,
+      "loss": 0.9270301461219788,
+      "step": 1172
+    },
+    {
+      "epoch": 0.49535864978902955,
+      "grad_norm": 0.45215415954589844,
+      "learning_rate": 0.00019999646772105942,
+      "loss": 0.8782645463943481,
+      "step": 1174
+    },
+    {
+      "epoch": 0.4962025316455696,
+      "grad_norm": 0.41578832268714905,
+      "learning_rate": 0.00019999605250079744,
+      "loss": 0.873112678527832,
+      "step": 1176
+    },
+    {
+      "epoch": 0.4970464135021097,
+      "grad_norm": 0.4632788300514221,
+      "learning_rate": 0.0001999956142133666,
+      "loss": 0.9593189358711243,
+      "step": 1178
+    },
+    {
+      "epoch": 0.4978902953586498,
+      "grad_norm": 0.5007622838020325,
+      "learning_rate": 0.000199995152858868,
+      "loss": 0.9215621948242188,
+      "step": 1180
+    },
+    {
+      "epoch": 0.49873417721518987,
+      "grad_norm": 0.5201935768127441,
+      "learning_rate": 0.0001999946684374081,
+      "loss": 0.8964219689369202,
+      "step": 1182
+    },
+    {
+      "epoch": 0.49957805907173,
+      "grad_norm": 0.5053967237472534,
+      "learning_rate": 0.0001999941609490986,
+      "loss": 0.8528663516044617,
+      "step": 1184
+    },
+    {
+      "epoch": 0.5004219409282701,
+      "grad_norm": 0.44397974014282227,
+      "learning_rate": 0.0001999936303940566,
+      "loss": 0.8214734792709351,
+      "step": 1186
+    },
+    {
+      "epoch": 0.5012658227848101,
+      "grad_norm": 0.4563101530075073,
+      "learning_rate": 0.0001999930767724045,
+      "loss": 0.8882166743278503,
+      "step": 1188
+    },
+    {
+      "epoch": 0.5021097046413502,
+      "grad_norm": 0.4666728675365448,
+      "learning_rate": 0.00019999250008426997,
+      "loss": 0.8882588148117065,
+      "step": 1190
+    },
+    {
+      "epoch": 0.5029535864978903,
+      "grad_norm": 0.4423629939556122,
+      "learning_rate": 0.00019999190032978607,
+      "loss": 0.8353691697120667,
+      "step": 1192
+    },
+    {
+      "epoch": 0.5037974683544304,
+      "grad_norm": 0.4514595866203308,
+      "learning_rate": 0.00019999127750909118,
+      "loss": 0.8721219301223755,
+      "step": 1194
+    },
+    {
+      "epoch": 0.5046413502109705,
+      "grad_norm": 0.48469996452331543,
+      "learning_rate": 0.00019999063162232886,
+      "loss": 0.902795135974884,
+      "step": 1196
+    },
+    {
+      "epoch": 0.5054852320675105,
+      "grad_norm": 0.4931983947753906,
+      "learning_rate": 0.00019998996266964823,
+      "loss": 0.8843175768852234,
+      "step": 1198
+    },
+    {
+      "epoch": 0.5063291139240507,
+      "grad_norm": 0.507408082485199,
+      "learning_rate": 0.00019998927065120357,
+      "loss": 0.9179208278656006,
+      "step": 1200
+    },
+    {
+      "epoch": 0.5063291139240507,
+      "eval_loss": 0.9224098324775696,
+      "eval_runtime": 669.7542,
+      "eval_samples_per_second": 3.146,
+      "eval_steps_per_second": 3.146,
+      "step": 1200
+    },
+    {
+      "epoch": 0.5071729957805907,
+      "grad_norm": 0.4024188220500946,
+      "learning_rate": 0.00019998855556715447,
+      "loss": 0.9078981876373291,
+      "step": 1202
+    },
+    {
+      "epoch": 0.5080168776371308,
+      "grad_norm": 0.6558667421340942,
+      "learning_rate": 0.00019998781741766594,
+      "loss": 0.8089252710342407,
+      "step": 1204
+    },
+    {
+      "epoch": 0.5088607594936709,
+      "grad_norm": 0.45128145813941956,
+      "learning_rate": 0.00019998705620290823,
+      "loss": 0.9316248297691345,
+      "step": 1206
+    },
+    {
+      "epoch": 0.509704641350211,
+      "grad_norm": 0.4733511507511139,
+      "learning_rate": 0.00019998627192305694,
+      "loss": 0.9047867655754089,
+      "step": 1208
+    },
+    {
+      "epoch": 0.510548523206751,
+      "grad_norm": 0.4026021659374237,
+      "learning_rate": 0.00019998546457829298,
+      "loss": 0.8507166504859924,
+      "step": 1210
+    },
+    {
+      "epoch": 0.5113924050632911,
+      "grad_norm": 0.43416184186935425,
+      "learning_rate": 0.00019998463416880262,
+      "loss": 0.8490090370178223,
+      "step": 1212
+    },
+    {
+      "epoch": 0.5122362869198313,
+      "grad_norm": 0.3678364157676697,
+      "learning_rate": 0.0001999837806947774,
+      "loss": 0.7730492949485779,
+      "step": 1214
+    },
+    {
+      "epoch": 0.5130801687763713,
+      "grad_norm": 0.49042677879333496,
+      "learning_rate": 0.00019998290415641415,
+      "loss": 0.9625269174575806,
+      "step": 1216
+    },
+    {
+      "epoch": 0.5139240506329114,
+      "grad_norm": 0.4576701819896698,
+      "learning_rate": 0.00019998200455391516,
+      "loss": 0.8961732983589172,
+      "step": 1218
+    },
+    {
+      "epoch": 0.5147679324894515,
+      "grad_norm": 0.4465518593788147,
+      "learning_rate": 0.00019998108188748793,
+      "loss": 0.8508996367454529,
+      "step": 1220
+    },
+    {
+      "epoch": 0.5156118143459916,
+      "grad_norm": 0.4267960488796234,
+      "learning_rate": 0.00019998013615734524,
+      "loss": 0.8664930462837219,
+      "step": 1222
+    },
+    {
+      "epoch": 0.5164556962025316,
+      "grad_norm": 0.39936602115631104,
+      "learning_rate": 0.0001999791673637053,
+      "loss": 0.8841317892074585,
+      "step": 1224
+    },
+    {
+      "epoch": 0.5172995780590718,
+      "grad_norm": 0.4442414343357086,
+      "learning_rate": 0.0001999781755067916,
+      "loss": 0.8601276874542236,
+      "step": 1226
+    },
+    {
+      "epoch": 0.5181434599156118,
+      "grad_norm": 0.4397028088569641,
+      "learning_rate": 0.00019997716058683292,
+      "loss": 0.8377046585083008,
+      "step": 1228
+    },
+    {
+      "epoch": 0.5189873417721519,
+      "grad_norm": 0.5327648520469666,
+      "learning_rate": 0.0001999761226040634,
+      "loss": 0.9853615164756775,
+      "step": 1230
+    },
+    {
+      "epoch": 0.5198312236286919,
+      "grad_norm": 0.40931421518325806,
+      "learning_rate": 0.00019997506155872244,
+      "loss": 0.9235715866088867,
+      "step": 1232
+    },
+    {
+      "epoch": 0.5206751054852321,
+      "grad_norm": 0.4599161148071289,
+      "learning_rate": 0.00019997397745105487,
+      "loss": 0.868396520614624,
+      "step": 1234
+    },
+    {
+      "epoch": 0.5215189873417722,
+      "grad_norm": 0.4813833236694336,
+      "learning_rate": 0.0001999728702813107,
+      "loss": 0.9946733117103577,
+      "step": 1236
+    },
+    {
+      "epoch": 0.5223628691983122,
+      "grad_norm": 0.43518269062042236,
+      "learning_rate": 0.00019997174004974543,
+      "loss": 0.8608635067939758,
+      "step": 1238
+    },
+    {
+      "epoch": 0.5232067510548524,
+      "grad_norm": 0.45646214485168457,
+      "learning_rate": 0.0001999705867566197,
+      "loss": 0.819354772567749,
+      "step": 1240
+    },
+    {
+      "epoch": 0.5240506329113924,
+      "grad_norm": 0.46834197640419006,
+      "learning_rate": 0.00019996941040219954,
+      "loss": 0.8934658765792847,
+      "step": 1242
+    },
+    {
+      "epoch": 0.5248945147679325,
+      "grad_norm": 0.46065032482147217,
+      "learning_rate": 0.0001999682109867564,
+      "loss": 0.8624778985977173,
+      "step": 1244
+    },
+    {
+      "epoch": 0.5257383966244725,
+      "grad_norm": 0.4506741166114807,
+      "learning_rate": 0.00019996698851056688,
+      "loss": 0.8540882468223572,
+      "step": 1246
+    },
+    {
+      "epoch": 0.5265822784810127,
+      "grad_norm": 0.4550519287586212,
+      "learning_rate": 0.00019996574297391302,
+      "loss": 0.9003006219863892,
+      "step": 1248
+    },
+    {
+      "epoch": 0.5274261603375527,
+      "grad_norm": 0.4097813367843628,
+      "learning_rate": 0.00019996447437708214,
+      "loss": 0.8490248918533325,
+      "step": 1250
+    },
+    {
+      "epoch": 0.5282700421940928,
+      "grad_norm": 0.4375711679458618,
+      "learning_rate": 0.00019996318272036688,
+      "loss": 0.9204569458961487,
+      "step": 1252
+    },
+    {
+      "epoch": 0.529113924050633,
+      "grad_norm": 0.4335198998451233,
+      "learning_rate": 0.00019996186800406518,
+      "loss": 0.8831789493560791,
+      "step": 1254
+    },
+    {
+      "epoch": 0.529957805907173,
+      "grad_norm": 0.4840933084487915,
+      "learning_rate": 0.00019996053022848035,
+      "loss": 0.8578064441680908,
+      "step": 1256
+    },
+    {
+      "epoch": 0.5308016877637131,
+      "grad_norm": 0.48911216855049133,
+      "learning_rate": 0.00019995916939392097,
+      "loss": 0.9235416650772095,
+      "step": 1258
+    },
+    {
+      "epoch": 0.5316455696202531,
+      "grad_norm": 0.41106730699539185,
+      "learning_rate": 0.00019995778550070098,
+      "loss": 0.7897764444351196,
+      "step": 1260
+    },
+    {
+      "epoch": 0.5324894514767933,
+      "grad_norm": 0.42233356833457947,
+      "learning_rate": 0.00019995637854913957,
+      "loss": 0.9392989873886108,
+      "step": 1262
+    },
+    {
+      "epoch": 0.5333333333333333,
+      "grad_norm": 0.39145180583000183,
+      "learning_rate": 0.0001999549485395613,
+      "loss": 0.9130091667175293,
+      "step": 1264
+    },
+    {
+      "epoch": 0.5341772151898734,
+      "grad_norm": 0.42780739068984985,
+      "learning_rate": 0.00019995349547229614,
+      "loss": 0.9690561890602112,
+      "step": 1266
+    },
+    {
+      "epoch": 0.5350210970464135,
+      "grad_norm": 0.4823327362537384,
+      "learning_rate": 0.0001999520193476792,
+      "loss": 0.9990131258964539,
+      "step": 1268
+    },
+    {
+      "epoch": 0.5358649789029536,
+      "grad_norm": 0.4669715166091919,
+      "learning_rate": 0.00019995052016605097,
+      "loss": 0.8343052864074707,
+      "step": 1270
+    },
+    {
+      "epoch": 0.5367088607594936,
+      "grad_norm": 0.4024743437767029,
+      "learning_rate": 0.0001999489979277574,
+      "loss": 0.9605970978736877,
+      "step": 1272
+    },
+    {
+      "epoch": 0.5375527426160338,
+      "grad_norm": 0.42063432931900024,
+      "learning_rate": 0.0001999474526331495,
+      "loss": 0.8164438009262085,
+      "step": 1274
+    },
+    {
+      "epoch": 0.5383966244725739,
+      "grad_norm": 0.47953495383262634,
+      "learning_rate": 0.00019994588428258383,
+      "loss": 0.879767656326294,
+      "step": 1276
+    },
+    {
+      "epoch": 0.5392405063291139,
+      "grad_norm": 0.39668557047843933,
+      "learning_rate": 0.00019994429287642214,
+      "loss": 0.7658726572990417,
+      "step": 1278
+    },
+    {
+      "epoch": 0.540084388185654,
+      "grad_norm": 0.5136121511459351,
+      "learning_rate": 0.00019994267841503155,
+      "loss": 0.9477384686470032,
+      "step": 1280
+    },
+    {
+      "epoch": 0.5409282700421941,
+      "grad_norm": 0.4039286673069,
+      "learning_rate": 0.00019994104089878452,
+      "loss": 0.7883896827697754,
+      "step": 1282
+    },
+    {
+      "epoch": 0.5417721518987342,
+      "grad_norm": 0.4425487816333771,
+      "learning_rate": 0.0001999393803280587,
+      "loss": 0.8526129126548767,
+      "step": 1284
+    },
+    {
+      "epoch": 0.5426160337552742,
+      "grad_norm": 0.4544559121131897,
+      "learning_rate": 0.00019993769670323725,
+      "loss": 0.8694683909416199,
+      "step": 1286
+    },
+    {
+      "epoch": 0.5434599156118144,
+      "grad_norm": 0.42872917652130127,
+      "learning_rate": 0.0001999359900247085,
+      "loss": 0.8211527466773987,
+      "step": 1288
+    },
+    {
+      "epoch": 0.5443037974683544,
+      "grad_norm": 0.41549673676490784,
+      "learning_rate": 0.00019993426029286617,
+      "loss": 0.7967932224273682,
+      "step": 1290
+    },
+    {
+      "epoch": 0.5451476793248945,
+      "grad_norm": 0.42888355255126953,
+      "learning_rate": 0.00019993250750810926,
+      "loss": 0.8291563987731934,
+      "step": 1292
+    },
+    {
+      "epoch": 0.5459915611814345,
+      "grad_norm": 0.44985485076904297,
+      "learning_rate": 0.0001999307316708421,
+      "loss": 0.8787803649902344,
+      "step": 1294
+    },
+    {
+      "epoch": 0.5468354430379747,
+      "grad_norm": 0.43326008319854736,
+      "learning_rate": 0.00019992893278147436,
+      "loss": 0.8375519514083862,
+      "step": 1296
+    },
+    {
+      "epoch": 0.5476793248945148,
+      "grad_norm": 0.406556099653244,
+      "learning_rate": 0.00019992711084042096,
+      "loss": 0.8538051247596741,
+      "step": 1298
+    },
+    {
+      "epoch": 0.5485232067510548,
+      "grad_norm": 0.46090877056121826,
+      "learning_rate": 0.00019992526584810226,
+      "loss": 0.9722132086753845,
+      "step": 1300
+    },
+    {
+      "epoch": 0.5485232067510548,
+      "eval_loss": 0.9068717360496521,
+      "eval_runtime": 680.7718,
+      "eval_samples_per_second": 3.095,
+      "eval_steps_per_second": 3.095,
+      "step": 1300
+    },
+    {
+      "epoch": 0.549367088607595,
+      "grad_norm": 0.44932207465171814,
+      "learning_rate": 0.00019992339780494382,
+      "loss": 0.8474152684211731,
+      "step": 1302
+    },
+    {
+      "epoch": 0.550210970464135,
+      "grad_norm": 0.5185276865959167,
+      "learning_rate": 0.0001999215067113766,
+      "loss": 0.8440775871276855,
+      "step": 1304
+    },
+    {
+      "epoch": 0.5510548523206751,
+      "grad_norm": 0.43811365962028503,
+      "learning_rate": 0.0001999195925678368,
+      "loss": 0.8848010897636414,
+      "step": 1306
+    },
+    {
+      "epoch": 0.5518987341772152,
+      "grad_norm": 0.4861660301685333,
+      "learning_rate": 0.00019991765537476596,
+      "loss": 0.8265100121498108,
+      "step": 1308
+    },
+    {
+      "epoch": 0.5527426160337553,
+      "grad_norm": 0.5068721771240234,
+      "learning_rate": 0.000199915695132611,
+      "loss": 0.9327963590621948,
+      "step": 1310
+    },
+    {
+      "epoch": 0.5535864978902953,
+      "grad_norm": 0.4902805685997009,
+      "learning_rate": 0.0001999137118418241,
+      "loss": 0.9315155744552612,
+      "step": 1312
+    },
+    {
+      "epoch": 0.5544303797468354,
+      "grad_norm": 0.4530964493751526,
+      "learning_rate": 0.00019991170550286272,
+      "loss": 0.9344852566719055,
+      "step": 1314
+    },
+    {
+      "epoch": 0.5552742616033756,
+      "grad_norm": 0.4106673300266266,
+      "learning_rate": 0.00019990967611618974,
+      "loss": 0.8326173424720764,
+      "step": 1316
+    },
+    {
+      "epoch": 0.5561181434599156,
+      "grad_norm": 0.4439711570739746,
+      "learning_rate": 0.0001999076236822733,
+      "loss": 0.8934909105300903,
+      "step": 1318
+    },
+    {
+      "epoch": 0.5569620253164557,
+      "grad_norm": 0.40245553851127625,
+      "learning_rate": 0.0001999055482015868,
+      "loss": 0.6691107153892517,
+      "step": 1320
+    },
+    {
+      "epoch": 0.5578059071729958,
+      "grad_norm": 0.43904435634613037,
+      "learning_rate": 0.0001999034496746091,
+      "loss": 0.8560307621955872,
+      "step": 1322
+    },
+    {
+      "epoch": 0.5586497890295359,
+      "grad_norm": 0.5131705403327942,
+      "learning_rate": 0.00019990132810182422,
+      "loss": 0.9024442434310913,
+      "step": 1324
+    },
+    {
+      "epoch": 0.5594936708860759,
+      "grad_norm": 0.4539531171321869,
+      "learning_rate": 0.00019989918348372156,
+      "loss": 0.8675428628921509,
+      "step": 1326
+    },
+    {
+      "epoch": 0.560337552742616,
+      "grad_norm": 0.44469985365867615,
+      "learning_rate": 0.00019989701582079591,
+      "loss": 0.8853429555892944,
+      "step": 1328
+    },
+    {
+      "epoch": 0.5611814345991561,
+      "grad_norm": 0.4485546946525574,
+      "learning_rate": 0.00019989482511354725,
+      "loss": 0.9675378799438477,
+      "step": 1330
+    },
+    {
+      "epoch": 0.5620253164556962,
+      "grad_norm": 0.43316468596458435,
+      "learning_rate": 0.00019989261136248097,
+      "loss": 0.8256655335426331,
+      "step": 1332
+    },
+    {
+      "epoch": 0.5628691983122363,
+      "grad_norm": 0.4315780699253082,
+      "learning_rate": 0.00019989037456810772,
+      "loss": 0.7727690935134888,
+      "step": 1334
+    },
+    {
+      "epoch": 0.5637130801687764,
+      "grad_norm": 0.45147502422332764,
+      "learning_rate": 0.00019988811473094348,
+      "loss": 0.9240618348121643,
+      "step": 1336
+    },
+    {
+      "epoch": 0.5645569620253165,
+      "grad_norm": 0.4606908857822418,
+      "learning_rate": 0.00019988583185150957,
+      "loss": 0.8005949258804321,
+      "step": 1338
+    },
+    {
+      "epoch": 0.5654008438818565,
+      "grad_norm": 0.48932451009750366,
+      "learning_rate": 0.00019988352593033255,
+      "loss": 0.9042718410491943,
+      "step": 1340
+    },
+    {
+      "epoch": 0.5662447257383966,
+      "grad_norm": 0.46022218465805054,
+      "learning_rate": 0.00019988119696794443,
+      "loss": 0.8724613785743713,
+      "step": 1342
+    },
+    {
+      "epoch": 0.5670886075949367,
+      "grad_norm": 0.4414621889591217,
+      "learning_rate": 0.00019987884496488242,
+      "loss": 0.8653066158294678,
+      "step": 1344
+    },
+    {
+      "epoch": 0.5679324894514768,
+      "grad_norm": 0.4490342140197754,
+      "learning_rate": 0.0001998764699216891,
+      "loss": 0.8843849301338196,
+      "step": 1346
+    },
+    {
+      "epoch": 0.5687763713080168,
+      "grad_norm": 0.47690069675445557,
+      "learning_rate": 0.00019987407183891232,
+      "loss": 0.8385393023490906,
+      "step": 1348
+    },
+    {
+      "epoch": 0.569620253164557,
+      "grad_norm": 0.41409361362457275,
+      "learning_rate": 0.00019987165071710527,
+      "loss": 0.77492755651474,
+      "step": 1350
+    },
+    {
+      "epoch": 0.570464135021097,
+      "grad_norm": 0.4151647984981537,
+      "learning_rate": 0.0001998692065568265,
+      "loss": 0.8368680477142334,
+      "step": 1352
+    },
+    {
+      "epoch": 0.5713080168776371,
+      "grad_norm": 0.4105067849159241,
+      "learning_rate": 0.0001998667393586398,
+      "loss": 0.833220362663269,
+      "step": 1354
+    },
+    {
+      "epoch": 0.5721518987341773,
+      "grad_norm": 0.44701820611953735,
+      "learning_rate": 0.0001998642491231143,
+      "loss": 0.8041552305221558,
+      "step": 1356
+    },
+    {
+      "epoch": 0.5729957805907173,
+      "grad_norm": 0.4621582329273224,
+      "learning_rate": 0.00019986173585082444,
+      "loss": 0.9052709937095642,
+      "step": 1358
+    },
+    {
+      "epoch": 0.5738396624472574,
+      "grad_norm": 0.4934713840484619,
+      "learning_rate": 0.00019985919954235005,
+      "loss": 0.8036054372787476,
+      "step": 1360
+    },
+    {
+      "epoch": 0.5746835443037974,
+      "grad_norm": 0.4557384252548218,
+      "learning_rate": 0.0001998566401982761,
+      "loss": 0.8340095281600952,
+      "step": 1362
+    },
+    {
+      "epoch": 0.5755274261603376,
+      "grad_norm": 0.5048403739929199,
+      "learning_rate": 0.00019985405781919305,
+      "loss": 0.8703633546829224,
+      "step": 1364
+    },
+    {
+      "epoch": 0.5763713080168776,
+      "grad_norm": 0.4707714319229126,
+      "learning_rate": 0.0001998514524056966,
+      "loss": 0.9115830659866333,
+      "step": 1366
+    },
+    {
+      "epoch": 0.5772151898734177,
+      "grad_norm": 0.4502186179161072,
+      "learning_rate": 0.00019984882395838777,
+      "loss": 0.8603149056434631,
+      "step": 1368
+    },
+    {
+      "epoch": 0.5780590717299579,
+      "grad_norm": 0.4229136109352112,
+      "learning_rate": 0.00019984617247787288,
+      "loss": 0.7482197284698486,
+      "step": 1370
+    },
+    {
+      "epoch": 0.5789029535864979,
+      "grad_norm": 0.49208030104637146,
+      "learning_rate": 0.0001998434979647636,
+      "loss": 0.9071239233016968,
+      "step": 1372
+    },
+    {
+      "epoch": 0.579746835443038,
+      "grad_norm": 0.47393161058425903,
+      "learning_rate": 0.00019984080041967687,
+      "loss": 0.9370644092559814,
+      "step": 1374
+    },
+    {
+      "epoch": 0.580590717299578,
+      "grad_norm": 0.4483324885368347,
+      "learning_rate": 0.00019983807984323492,
+      "loss": 0.8536180257797241,
+      "step": 1376
+    },
+    {
+      "epoch": 0.5814345991561182,
+      "grad_norm": 0.4163796007633209,
+      "learning_rate": 0.00019983533623606543,
+      "loss": 0.7360405325889587,
+      "step": 1378
+    },
+    {
+      "epoch": 0.5822784810126582,
+      "grad_norm": 0.44147446751594543,
+      "learning_rate": 0.00019983256959880122,
+      "loss": 0.8162824511528015,
+      "step": 1380
+    },
+    {
+      "epoch": 0.5831223628691983,
+      "grad_norm": 0.44443491101264954,
+      "learning_rate": 0.0001998297799320805,
+      "loss": 0.9169327616691589,
+      "step": 1382
+    },
+    {
+      "epoch": 0.5839662447257384,
+      "grad_norm": 0.4641949236392975,
+      "learning_rate": 0.00019982696723654686,
+      "loss": 0.8875693678855896,
+      "step": 1384
+    },
+    {
+      "epoch": 0.5848101265822785,
+      "grad_norm": 0.4327974319458008,
+      "learning_rate": 0.00019982413151284906,
+      "loss": 0.7747344970703125,
+      "step": 1386
+    },
+    {
+      "epoch": 0.5856540084388185,
+      "grad_norm": 0.44200772047042847,
+      "learning_rate": 0.0001998212727616413,
+      "loss": 0.8490481972694397,
+      "step": 1388
+    },
+    {
+      "epoch": 0.5864978902953587,
+      "grad_norm": 0.39084959030151367,
+      "learning_rate": 0.000199818390983583,
+      "loss": 0.8014808297157288,
+      "step": 1390
+    },
+    {
+      "epoch": 0.5873417721518988,
+      "grad_norm": 0.3691277801990509,
+      "learning_rate": 0.00019981548617933897,
+      "loss": 0.8650928139686584,
+      "step": 1392
+    },
+    {
+      "epoch": 0.5881856540084388,
+      "grad_norm": 0.480002760887146,
+      "learning_rate": 0.00019981255834957925,
+      "loss": 0.7713267803192139,
+      "step": 1394
+    },
+    {
+      "epoch": 0.5890295358649789,
+      "grad_norm": 0.4397226572036743,
+      "learning_rate": 0.00019980960749497927,
+      "loss": 0.9053656458854675,
+      "step": 1396
+    },
+    {
+      "epoch": 0.589873417721519,
+      "grad_norm": 0.4263727366924286,
+      "learning_rate": 0.00019980663361621973,
+      "loss": 0.8737669587135315,
+      "step": 1398
+    },
+    {
+      "epoch": 0.5907172995780591,
+      "grad_norm": 0.4468817710876465,
+      "learning_rate": 0.0001998036367139866,
+      "loss": 0.9396650791168213,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5907172995780591,
+      "eval_loss": 0.8971880674362183,
+      "eval_runtime": 692.8046,
+      "eval_samples_per_second": 3.041,
+      "eval_steps_per_second": 3.041,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5915611814345991,
+      "grad_norm": 0.4128144383430481,
+      "learning_rate": 0.00019980061678897127,
+      "loss": 0.926134467124939,
+      "step": 1402
+    },
+    {
+      "epoch": 0.5924050632911393,
+      "grad_norm": 0.45524629950523376,
+      "learning_rate": 0.00019979757384187035,
+      "loss": 0.826360285282135,
+      "step": 1404
+    },
+    {
+      "epoch": 0.5932489451476793,
+      "grad_norm": 0.42521336674690247,
+      "learning_rate": 0.0001997945078733858,
+      "loss": 0.8147702813148499,
+      "step": 1406
+    },
+    {
+      "epoch": 0.5940928270042194,
+      "grad_norm": 0.4172525107860565,
+      "learning_rate": 0.00019979141888422489,
+      "loss": 0.7711596488952637,
+      "step": 1408
+    },
+    {
+      "epoch": 0.5949367088607594,
+      "grad_norm": 0.46193623542785645,
+      "learning_rate": 0.00019978830687510013,
+      "loss": 0.7875980138778687,
+      "step": 1410
+    },
+    {
+      "epoch": 0.5957805907172996,
+      "grad_norm": 0.4455825686454773,
+      "learning_rate": 0.00019978517184672946,
+      "loss": 0.8590195178985596,
+      "step": 1412
+    },
+    {
+      "epoch": 0.5966244725738397,
+      "grad_norm": 0.4266505539417267,
+      "learning_rate": 0.00019978201379983608,
+      "loss": 0.9318227767944336,
+      "step": 1414
+    },
+    {
+      "epoch": 0.5974683544303797,
+      "grad_norm": 0.42627251148223877,
+      "learning_rate": 0.00019977883273514843,
+      "loss": 0.8644474744796753,
+      "step": 1416
+    },
+    {
+      "epoch": 0.5983122362869199,
+      "grad_norm": 0.393873393535614,
+      "learning_rate": 0.00019977562865340038,
+      "loss": 0.7760446071624756,
+      "step": 1418
+    },
+    {
+      "epoch": 0.5991561181434599,
+      "grad_norm": 0.44334208965301514,
+      "learning_rate": 0.000199772401555331,
+      "loss": 0.8071584701538086,
+      "step": 1420
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.39390063285827637,
+      "learning_rate": 0.0001997691514416848,
+      "loss": 0.9076889753341675,
+      "step": 1422
+    },
+    {
+      "epoch": 0.60084388185654,
+      "grad_norm": 0.46753421425819397,
+      "learning_rate": 0.00019976587831321144,
+      "loss": 0.8534318804740906,
+      "step": 1424
+    },
+    {
+      "epoch": 0.6016877637130802,
+      "grad_norm": 0.4840681254863739,
+      "learning_rate": 0.00019976258217066598,
+      "loss": 0.9049596190452576,
+      "step": 1426
+    },
+    {
+      "epoch": 0.6025316455696202,
+      "grad_norm": 0.4740557372570038,
+      "learning_rate": 0.00019975926301480878,
+      "loss": 0.8520918488502502,
+      "step": 1428
+    },
+    {
+      "epoch": 0.6033755274261603,
+      "grad_norm": 0.46498584747314453,
+      "learning_rate": 0.00019975592084640553,
+      "loss": 0.8221305012702942,
+      "step": 1430
+    },
+    {
+      "epoch": 0.6042194092827005,
+      "grad_norm": 0.40885692834854126,
+      "learning_rate": 0.0001997525556662272,
+      "loss": 0.8132198452949524,
+      "step": 1432
+    },
+    {
+      "epoch": 0.6050632911392405,
+      "grad_norm": 0.4375821053981781,
+      "learning_rate": 0.00019974916747505006,
+      "loss": 0.870534360408783,
+      "step": 1434
+    },
+    {
+      "epoch": 0.6059071729957806,
+      "grad_norm": 0.4780935049057007,
+      "learning_rate": 0.00019974575627365569,
+      "loss": 0.8825759291648865,
+      "step": 1436
+    },
+    {
+      "epoch": 0.6067510548523207,
+      "grad_norm": 0.4623970687389374,
+      "learning_rate": 0.00019974232206283098,
+      "loss": 0.8187641501426697,
+      "step": 1438
+    },
+    {
+      "epoch": 0.6075949367088608,
+      "grad_norm": 0.42172595858573914,
+      "learning_rate": 0.0001997388648433682,
+      "loss": 0.8569821119308472,
+      "step": 1440
+    },
+    {
+      "epoch": 0.6084388185654008,
+      "grad_norm": 0.49386659264564514,
+      "learning_rate": 0.0001997353846160648,
+      "loss": 0.9173614382743835,
+      "step": 1442
+    },
+    {
+      "epoch": 0.6092827004219409,
+      "grad_norm": 0.47122615575790405,
+      "learning_rate": 0.00019973188138172363,
+      "loss": 0.9444975256919861,
+      "step": 1444
+    },
+    {
+      "epoch": 0.610126582278481,
+      "grad_norm": 0.4831530451774597,
+      "learning_rate": 0.0001997283551411528,
+      "loss": 0.9537245631217957,
+      "step": 1446
+    },
+    {
+      "epoch": 0.6109704641350211,
+      "grad_norm": 0.44460317492485046,
+      "learning_rate": 0.0001997248058951658,
+      "loss": 0.8481367230415344,
+      "step": 1448
+    },
+    {
+      "epoch": 0.6118143459915611,
+      "grad_norm": 0.4315263628959656,
+      "learning_rate": 0.00019972123364458128,
+      "loss": 0.8572371006011963,
+      "step": 1450
+    },
+    {
+      "epoch": 0.6126582278481013,
+      "grad_norm": 0.38559582829475403,
+      "learning_rate": 0.00019971763839022336,
+      "loss": 0.762590765953064,
+      "step": 1452
+    },
+    {
+      "epoch": 0.6135021097046414,
+      "grad_norm": 0.3820290267467499,
+      "learning_rate": 0.00019971402013292138,
+      "loss": 0.7879500389099121,
+      "step": 1454
+    },
+    {
+      "epoch": 0.6143459915611814,
+      "grad_norm": 0.42509570717811584,
+      "learning_rate": 0.00019971037887351,
+      "loss": 0.8696863055229187,
+      "step": 1456
+    },
+    {
+      "epoch": 0.6151898734177215,
+      "grad_norm": 0.46750548481941223,
+      "learning_rate": 0.00019970671461282916,
+      "loss": 0.8425862789154053,
+      "step": 1458
+    },
+    {
+      "epoch": 0.6160337552742616,
+      "grad_norm": 0.3937121033668518,
+      "learning_rate": 0.0001997030273517242,
+      "loss": 0.8028931021690369,
+      "step": 1460
+    },
+    {
+      "epoch": 0.6168776371308017,
+      "grad_norm": 0.4361220896244049,
+      "learning_rate": 0.00019969931709104565,
+      "loss": 0.8662509918212891,
+      "step": 1462
+    },
+    {
+      "epoch": 0.6177215189873417,
+      "grad_norm": 0.4737963378429413,
+      "learning_rate": 0.00019969558383164943,
+      "loss": 0.8962596654891968,
+      "step": 1464
+    },
+    {
+      "epoch": 0.6185654008438819,
+      "grad_norm": 0.3952867388725281,
+      "learning_rate": 0.00019969182757439668,
+      "loss": 0.8630867600440979,
+      "step": 1466
+    },
+    {
+      "epoch": 0.619409282700422,
+      "grad_norm": 0.41688743233680725,
+      "learning_rate": 0.00019968804832015393,
+      "loss": 0.8759240508079529,
+      "step": 1468
+    },
+    {
+      "epoch": 0.620253164556962,
+      "grad_norm": 0.4599224627017975,
+      "learning_rate": 0.00019968424606979298,
+      "loss": 0.8459385633468628,
+      "step": 1470
+    },
+    {
+      "epoch": 0.6210970464135022,
+      "grad_norm": 0.5476765632629395,
+      "learning_rate": 0.00019968042082419094,
+      "loss": 0.8844659328460693,
+      "step": 1472
+    },
+    {
+      "epoch": 0.6219409282700422,
+      "grad_norm": 0.46202269196510315,
+      "learning_rate": 0.00019967657258423022,
+      "loss": 0.860946536064148,
+      "step": 1474
+    },
+    {
+      "epoch": 0.6227848101265823,
+      "grad_norm": 0.39446666836738586,
+      "learning_rate": 0.00019967270135079853,
+      "loss": 0.7587860822677612,
+      "step": 1476
+    },
+    {
+      "epoch": 0.6236286919831223,
+      "grad_norm": 0.4500375986099243,
+      "learning_rate": 0.0001996688071247889,
+      "loss": 0.9143000245094299,
+      "step": 1478
+    },
+    {
+      "epoch": 0.6244725738396625,
+      "grad_norm": 0.4203545153141022,
+      "learning_rate": 0.0001996648899070996,
+      "loss": 0.7865519523620605,
+      "step": 1480
+    },
+    {
+      "epoch": 0.6253164556962025,
+      "grad_norm": 0.45623889565467834,
+      "learning_rate": 0.00019966094969863432,
+      "loss": 0.9192912578582764,
+      "step": 1482
+    },
+    {
+      "epoch": 0.6261603375527426,
+      "grad_norm": 0.4139637053012848,
+      "learning_rate": 0.00019965698650030195,
+      "loss": 0.8459161520004272,
+      "step": 1484
+    },
+    {
+      "epoch": 0.6270042194092827,
+      "grad_norm": 0.4410824179649353,
+      "learning_rate": 0.00019965300031301678,
+      "loss": 0.8159077763557434,
+      "step": 1486
+    },
+    {
+      "epoch": 0.6278481012658228,
+      "grad_norm": 0.43821197748184204,
+      "learning_rate": 0.00019964899113769828,
+      "loss": 0.884467363357544,
+      "step": 1488
+    },
+    {
+      "epoch": 0.6286919831223629,
+      "grad_norm": 0.4692091643810272,
+      "learning_rate": 0.00019964495897527133,
+      "loss": 0.8693601489067078,
+      "step": 1490
+    },
+    {
+      "epoch": 0.6295358649789029,
+      "grad_norm": 0.40715447068214417,
+      "learning_rate": 0.00019964090382666608,
+      "loss": 0.8897743225097656,
+      "step": 1492
+    },
+    {
+      "epoch": 0.6303797468354431,
+      "grad_norm": 0.4123334586620331,
+      "learning_rate": 0.0001996368256928179,
+      "loss": 0.8274733424186707,
+      "step": 1494
+    },
+    {
+      "epoch": 0.6312236286919831,
+      "grad_norm": 0.44347211718559265,
+      "learning_rate": 0.00019963272457466767,
+      "loss": 0.8158749938011169,
+      "step": 1496
+    },
+    {
+      "epoch": 0.6320675105485232,
+      "grad_norm": 0.4700329601764679,
+      "learning_rate": 0.00019962860047316135,
+      "loss": 0.8439569473266602,
+      "step": 1498
+    },
+    {
+      "epoch": 0.6329113924050633,
+      "grad_norm": 0.46415844559669495,
+      "learning_rate": 0.00019962445338925027,
+      "loss": 0.8316822648048401,
+      "step": 1500
+    },
+    {
+      "epoch": 0.6329113924050633,
+      "eval_loss": 0.887488842010498,
+      "eval_runtime": 686.2804,
+      "eval_samples_per_second": 3.07,
+      "eval_steps_per_second": 3.07,
+      "step": 1500
+    },
+    {
+      "epoch": 0.6337552742616034,
+      "grad_norm": 0.44183167815208435,
+      "learning_rate": 0.0001996202833238911,
+      "loss": 0.9121994972229004,
+      "step": 1502
+    },
+    {
+      "epoch": 0.6345991561181434,
+      "grad_norm": 0.4279208779335022,
+      "learning_rate": 0.00019961609027804587,
+      "loss": 0.9160211086273193,
+      "step": 1504
+    },
+    {
+      "epoch": 0.6354430379746835,
+      "grad_norm": 0.3967169523239136,
+      "learning_rate": 0.00019961187425268176,
+      "loss": 0.861677348613739,
+      "step": 1506
+    },
+    {
+      "epoch": 0.6362869198312237,
+      "grad_norm": 0.4796451926231384,
+      "learning_rate": 0.0001996076352487713,
+      "loss": 0.8478423357009888,
+      "step": 1508
+    },
+    {
+      "epoch": 0.6371308016877637,
+      "grad_norm": 0.3861426115036011,
+      "learning_rate": 0.00019960337326729245,
+      "loss": 0.8739159107208252,
+      "step": 1510
+    },
+    {
+      "epoch": 0.6379746835443038,
+      "grad_norm": 0.4567820429801941,
+      "learning_rate": 0.00019959908830922824,
+      "loss": 0.7693920731544495,
+      "step": 1512
+    },
+    {
+      "epoch": 0.6388185654008439,
+      "grad_norm": 0.4457108676433563,
+      "learning_rate": 0.00019959478037556724,
+      "loss": 0.8482301831245422,
+      "step": 1514
+    },
+    {
+      "epoch": 0.639662447257384,
+      "grad_norm": 0.43918928503990173,
+      "learning_rate": 0.00019959044946730314,
+      "loss": 0.8626812696456909,
+      "step": 1516
+    },
+    {
+      "epoch": 0.640506329113924,
+      "grad_norm": 0.4095900058746338,
+      "learning_rate": 0.00019958609558543504,
+      "loss": 0.8342230916023254,
+      "step": 1518
+    },
+    {
+      "epoch": 0.6413502109704642,
+      "grad_norm": 0.4181270897388458,
+      "learning_rate": 0.00019958171873096724,
+      "loss": 0.8794118165969849,
+      "step": 1520
+    },
+    {
+      "epoch": 0.6421940928270042,
+      "grad_norm": 0.4750959277153015,
+      "learning_rate": 0.00019957731890490947,
+      "loss": 0.7787677049636841,
+      "step": 1522
+    },
+    {
+      "epoch": 0.6430379746835443,
+      "grad_norm": 0.5129296779632568,
+      "learning_rate": 0.00019957289610827663,
+      "loss": 0.9084368348121643,
+      "step": 1524
+    },
+    {
+      "epoch": 0.6438818565400843,
+      "grad_norm": 0.4829029142856598,
+      "learning_rate": 0.000199568450342089,
+      "loss": 0.9066952466964722,
+      "step": 1526
+    },
+    {
+      "epoch": 0.6447257383966245,
+      "grad_norm": 0.4225057363510132,
+      "learning_rate": 0.00019956398160737214,
+      "loss": 0.8127874732017517,
+      "step": 1528
+    },
+    {
+      "epoch": 0.6455696202531646,
+      "grad_norm": 0.4231826364994049,
+      "learning_rate": 0.00019955948990515684,
+      "loss": 0.8781921863555908,
+      "step": 1530
+    },
+    {
+      "epoch": 0.6464135021097046,
+      "grad_norm": 0.4436216950416565,
+      "learning_rate": 0.00019955497523647933,
+      "loss": 0.8136996626853943,
+      "step": 1532
+    },
+    {
+      "epoch": 0.6472573839662448,
+      "grad_norm": 0.4018244743347168,
+      "learning_rate": 0.00019955043760238098,
+      "loss": 0.8003877401351929,
+      "step": 1534
+    },
+    {
+      "epoch": 0.6481012658227848,
+      "grad_norm": 0.40024611353874207,
+      "learning_rate": 0.0001995458770039086,
+      "loss": 0.8996267914772034,
+      "step": 1536
+    },
+    {
+      "epoch": 0.6489451476793249,
+      "grad_norm": 0.42845702171325684,
+      "learning_rate": 0.00019954129344211424,
+      "loss": 0.8610522150993347,
+      "step": 1538
+    },
+    {
+      "epoch": 0.6497890295358649,
+      "grad_norm": 0.4258182644844055,
+      "learning_rate": 0.00019953668691805517,
+      "loss": 0.8991837501525879,
+      "step": 1540
+    },
+    {
+      "epoch": 0.6506329113924051,
+      "grad_norm": 0.4483257830142975,
+      "learning_rate": 0.0001995320574327941,
+      "loss": 0.8150189518928528,
+      "step": 1542
+    },
+    {
+      "epoch": 0.6514767932489451,
+      "grad_norm": 0.4320627748966217,
+      "learning_rate": 0.0001995274049873989,
+      "loss": 0.8572840690612793,
+      "step": 1544
+    },
+    {
+      "epoch": 0.6523206751054852,
+      "grad_norm": 0.39950594305992126,
+      "learning_rate": 0.00019952272958294288,
+      "loss": 0.7766129374504089,
+      "step": 1546
+    },
+    {
+      "epoch": 0.6531645569620254,
+      "grad_norm": 0.4130117893218994,
+      "learning_rate": 0.00019951803122050446,
+      "loss": 0.9032199382781982,
+      "step": 1548
+    },
+    {
+      "epoch": 0.6540084388185654,
+      "grad_norm": 0.4436919391155243,
+      "learning_rate": 0.00019951330990116754,
+      "loss": 0.8361048102378845,
+      "step": 1550
+    },
+    {
+      "epoch": 0.6548523206751055,
+      "grad_norm": 0.4229227900505066,
+      "learning_rate": 0.00019950856562602121,
+      "loss": 0.8661768436431885,
+      "step": 1552
+    },
+    {
+      "epoch": 0.6556962025316456,
+      "grad_norm": 0.4839727580547333,
+      "learning_rate": 0.00019950379839615988,
+      "loss": 0.9737826585769653,
+      "step": 1554
+    },
+    {
+      "epoch": 0.6565400843881857,
+      "grad_norm": 0.3859386742115021,
+      "learning_rate": 0.00019949900821268328,
+      "loss": 0.7542453408241272,
+      "step": 1556
+    },
+    {
+      "epoch": 0.6573839662447257,
+      "grad_norm": 0.39643988013267517,
+      "learning_rate": 0.0001994941950766964,
+      "loss": 0.9226290583610535,
+      "step": 1558
+    },
+    {
+      "epoch": 0.6582278481012658,
+      "grad_norm": 0.4261355400085449,
+      "learning_rate": 0.00019948935898930954,
+      "loss": 0.8375602960586548,
+      "step": 1560
+    },
+    {
+      "epoch": 0.6590717299578059,
+      "grad_norm": 0.445931077003479,
+      "learning_rate": 0.00019948449995163826,
+      "loss": 0.7847310304641724,
+      "step": 1562
+    },
+    {
+      "epoch": 0.659915611814346,
+      "grad_norm": 0.4680459797382355,
+      "learning_rate": 0.00019947961796480353,
+      "loss": 0.851150631904602,
+      "step": 1564
+    },
+    {
+      "epoch": 0.660759493670886,
+      "grad_norm": 0.6826074719429016,
+      "learning_rate": 0.00019947471302993143,
+      "loss": 0.8259562253952026,
+      "step": 1566
+    },
+    {
+      "epoch": 0.6616033755274262,
+      "grad_norm": 0.47369205951690674,
+      "learning_rate": 0.00019946978514815352,
+      "loss": 0.879021406173706,
+      "step": 1568
+    },
+    {
+      "epoch": 0.6624472573839663,
+      "grad_norm": 0.41111937165260315,
+      "learning_rate": 0.0001994648343206065,
+      "loss": 0.9296056032180786,
+      "step": 1570
+    },
+    {
+      "epoch": 0.6632911392405063,
+      "grad_norm": 0.4297783076763153,
+      "learning_rate": 0.00019945986054843248,
+      "loss": 0.873822808265686,
+      "step": 1572
+    },
+    {
+      "epoch": 0.6641350210970464,
+      "grad_norm": 0.48124316334724426,
+      "learning_rate": 0.00019945486383277878,
+      "loss": 0.937151312828064,
+      "step": 1574
+    },
+    {
+      "epoch": 0.6649789029535865,
+      "grad_norm": 0.4048190116882324,
+      "learning_rate": 0.00019944984417479805,
+      "loss": 0.8398929834365845,
+      "step": 1576
+    },
+    {
+      "epoch": 0.6658227848101266,
+      "grad_norm": 0.4023774266242981,
+      "learning_rate": 0.00019944480157564826,
+      "loss": 0.8269520998001099,
+      "step": 1578
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.3784184157848358,
+      "learning_rate": 0.00019943973603649262,
+      "loss": 0.8237608671188354,
+      "step": 1580
+    },
+    {
+      "epoch": 0.6675105485232068,
+      "grad_norm": 0.4448573887348175,
+      "learning_rate": 0.00019943464755849965,
+      "loss": 0.8469099998474121,
+      "step": 1582
+    },
+    {
+      "epoch": 0.6683544303797468,
+      "grad_norm": 0.4690774083137512,
+      "learning_rate": 0.00019942953614284317,
+      "loss": 0.8630822896957397,
+      "step": 1584
+    },
+    {
+      "epoch": 0.6691983122362869,
+      "grad_norm": 0.43670853972435,
+      "learning_rate": 0.00019942440179070227,
+      "loss": 0.8904082179069519,
+      "step": 1586
+    },
+    {
+      "epoch": 0.6700421940928271,
+      "grad_norm": 0.4236849844455719,
+      "learning_rate": 0.00019941924450326138,
+      "loss": 0.8355640172958374,
+      "step": 1588
+    },
+    {
+      "epoch": 0.6708860759493671,
+      "grad_norm": 0.48624539375305176,
+      "learning_rate": 0.00019941406428171013,
+      "loss": 0.8898435831069946,
+      "step": 1590
+    },
+    {
+      "epoch": 0.6717299578059072,
+      "grad_norm": 0.4453780949115753,
+      "learning_rate": 0.00019940886112724355,
+      "loss": 0.9000012278556824,
+      "step": 1592
+    },
+    {
+      "epoch": 0.6725738396624472,
+      "grad_norm": 0.4417199194431305,
+      "learning_rate": 0.00019940363504106193,
+      "loss": 0.8171058893203735,
+      "step": 1594
+    },
+    {
+      "epoch": 0.6734177215189874,
+      "grad_norm": 0.45055362582206726,
+      "learning_rate": 0.00019939838602437074,
+      "loss": 0.8703798055648804,
+      "step": 1596
+    },
+    {
+      "epoch": 0.6742616033755274,
+      "grad_norm": 0.42316532135009766,
+      "learning_rate": 0.00019939311407838088,
+      "loss": 0.8677281737327576,
+      "step": 1598
+    },
+    {
+      "epoch": 0.6751054852320675,
+      "grad_norm": 0.3935781717300415,
+      "learning_rate": 0.0001993878192043085,
+      "loss": 0.7314013242721558,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6751054852320675,
+      "eval_loss": 0.8769772052764893,
+      "eval_runtime": 677.9338,
+      "eval_samples_per_second": 3.108,
+      "eval_steps_per_second": 3.108,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6759493670886076,
+      "grad_norm": 0.4197012186050415,
+      "learning_rate": 0.000199382501403375,
+      "loss": 0.8075380325317383,
+      "step": 1602
+    },
+    {
+      "epoch": 0.6767932489451477,
+      "grad_norm": 0.45227327942848206,
+      "learning_rate": 0.00019937716067680713,
+      "loss": 0.7885794639587402,
+      "step": 1604
+    },
+    {
+      "epoch": 0.6776371308016877,
+      "grad_norm": 0.45660385489463806,
+      "learning_rate": 0.00019937179702583682,
+      "loss": 0.868261992931366,
+      "step": 1606
+    },
+    {
+      "epoch": 0.6784810126582278,
+      "grad_norm": 0.3955288231372833,
+      "learning_rate": 0.00019936641045170144,
+      "loss": 0.8306655287742615,
+      "step": 1608
+    },
+    {
+      "epoch": 0.679324894514768,
+      "grad_norm": 0.3651324510574341,
+      "learning_rate": 0.00019936100095564353,
+      "loss": 0.7505315542221069,
+      "step": 1610
+    },
+    {
+      "epoch": 0.680168776371308,
+      "grad_norm": 0.4240955412387848,
+      "learning_rate": 0.00019935556853891096,
+      "loss": 0.8792756795883179,
+      "step": 1612
+    },
+    {
+      "epoch": 0.6810126582278481,
+      "grad_norm": 0.4142507314682007,
+      "learning_rate": 0.00019935011320275687,
+      "loss": 0.7328272461891174,
+      "step": 1614
+    },
+    {
+      "epoch": 0.6818565400843882,
+      "grad_norm": 0.48339322209358215,
+      "learning_rate": 0.00019934463494843975,
+      "loss": 0.8414849042892456,
+      "step": 1616
+    },
+    {
+      "epoch": 0.6827004219409283,
+      "grad_norm": 0.4312443435192108,
+      "learning_rate": 0.00019933913377722328,
+      "loss": 0.8588716983795166,
+      "step": 1618
+    },
+    {
+      "epoch": 0.6835443037974683,
+      "grad_norm": 0.43849512934684753,
+      "learning_rate": 0.0001993336096903765,
+      "loss": 0.846511960029602,
+      "step": 1620
+    },
+    {
+      "epoch": 0.6843881856540084,
+      "grad_norm": 0.45767566561698914,
+      "learning_rate": 0.0001993280626891737,
+      "loss": 0.9352323412895203,
+      "step": 1622
+    },
+    {
+      "epoch": 0.6852320675105485,
+      "grad_norm": 0.438412606716156,
+      "learning_rate": 0.00019932249277489444,
+      "loss": 0.8884757161140442,
+      "step": 1624
+    },
+    {
+      "epoch": 0.6860759493670886,
+      "grad_norm": 0.45893922448158264,
+      "learning_rate": 0.0001993168999488237,
+      "loss": 0.7739649415016174,
+      "step": 1626
+    },
+    {
+      "epoch": 0.6869198312236287,
+      "grad_norm": 0.38631772994995117,
+      "learning_rate": 0.0001993112842122515,
+      "loss": 0.7948258519172668,
+      "step": 1628
+    },
+    {
+      "epoch": 0.6877637130801688,
+      "grad_norm": 0.3824027180671692,
+      "learning_rate": 0.00019930564556647334,
+      "loss": 0.750511884689331,
+      "step": 1630
+    },
+    {
+      "epoch": 0.6886075949367089,
+      "grad_norm": 0.3980776071548462,
+      "learning_rate": 0.00019929998401278996,
+      "loss": 0.9006738662719727,
+      "step": 1632
+    },
+    {
+      "epoch": 0.6894514767932489,
+      "grad_norm": 0.41419896483421326,
+      "learning_rate": 0.00019929429955250734,
+      "loss": 0.6996869444847107,
+      "step": 1634
+    },
+    {
+      "epoch": 0.6902953586497891,
+      "grad_norm": 0.43783676624298096,
+      "learning_rate": 0.00019928859218693682,
+      "loss": 0.8523393273353577,
+      "step": 1636
+    },
+    {
+      "epoch": 0.6911392405063291,
+      "grad_norm": 0.4378126561641693,
+      "learning_rate": 0.00019928286191739497,
+      "loss": 0.8885987401008606,
+      "step": 1638
+    },
+    {
+      "epoch": 0.6919831223628692,
+      "grad_norm": 0.4276133179664612,
+      "learning_rate": 0.00019927710874520363,
+      "loss": 0.8563809990882874,
+      "step": 1640
+    },
+    {
+      "epoch": 0.6928270042194092,
+      "grad_norm": 0.4600282311439514,
+      "learning_rate": 0.00019927133267168992,
+      "loss": 0.854114294052124,
+      "step": 1642
+    },
+    {
+      "epoch": 0.6936708860759494,
+      "grad_norm": 0.5058461427688599,
+      "learning_rate": 0.00019926553369818635,
+      "loss": 0.8557889461517334,
+      "step": 1644
+    },
+    {
+      "epoch": 0.6945147679324895,
+      "grad_norm": 0.4947412312030792,
+      "learning_rate": 0.00019925971182603057,
+      "loss": 0.8622767925262451,
+      "step": 1646
+    },
+    {
+      "epoch": 0.6953586497890295,
+      "grad_norm": 0.42555496096611023,
+      "learning_rate": 0.00019925386705656558,
+      "loss": 0.7836558818817139,
+      "step": 1648
+    },
+    {
+      "epoch": 0.6962025316455697,
+      "grad_norm": 0.47560831904411316,
+      "learning_rate": 0.00019924799939113967,
+      "loss": 0.7877846956253052,
+      "step": 1650
+    },
+    {
+      "epoch": 0.6970464135021097,
+      "grad_norm": 0.391702264547348,
+      "learning_rate": 0.0001992421088311064,
+      "loss": 0.775681734085083,
+      "step": 1652
+    },
+    {
+      "epoch": 0.6978902953586498,
+      "grad_norm": 0.463679701089859,
+      "learning_rate": 0.0001992361953778246,
+      "loss": 0.8809882402420044,
+      "step": 1654
+    },
+    {
+      "epoch": 0.6987341772151898,
+      "grad_norm": 0.44096484780311584,
+      "learning_rate": 0.00019923025903265836,
+      "loss": 0.8513299226760864,
+      "step": 1656
+    },
+    {
+      "epoch": 0.69957805907173,
+      "grad_norm": 0.47818854451179504,
+      "learning_rate": 0.0001992242997969772,
+      "loss": 0.8309667706489563,
+      "step": 1658
+    },
+    {
+      "epoch": 0.70042194092827,
+      "grad_norm": 0.4192790985107422,
+      "learning_rate": 0.00019921831767215565,
+      "loss": 0.8817558884620667,
+      "step": 1660
+    },
+    {
+      "epoch": 0.7012658227848101,
+      "grad_norm": 0.45285239815711975,
+      "learning_rate": 0.00019921231265957373,
+      "loss": 0.8315755724906921,
+      "step": 1662
+    },
+    {
+      "epoch": 0.7021097046413503,
+      "grad_norm": 0.47118625044822693,
+      "learning_rate": 0.00019920628476061673,
+      "loss": 0.8698058724403381,
+      "step": 1664
+    },
+    {
+      "epoch": 0.7029535864978903,
+      "grad_norm": 0.4769238829612732,
+      "learning_rate": 0.0001992002339766751,
+      "loss": 0.953620970249176,
+      "step": 1666
+    },
+    {
+      "epoch": 0.7037974683544304,
+      "grad_norm": 0.39397355914115906,
+      "learning_rate": 0.00019919416030914468,
+      "loss": 0.8536615371704102,
+      "step": 1668
+    },
+    {
+      "epoch": 0.7046413502109705,
+      "grad_norm": 0.3659648299217224,
+      "learning_rate": 0.00019918806375942655,
+      "loss": 0.8159828782081604,
+      "step": 1670
+    },
+    {
+      "epoch": 0.7054852320675106,
+      "grad_norm": 0.4381789565086365,
+      "learning_rate": 0.00019918194432892703,
+      "loss": 0.782463550567627,
+      "step": 1672
+    },
+    {
+      "epoch": 0.7063291139240506,
+      "grad_norm": 0.44745752215385437,
+      "learning_rate": 0.00019917580201905778,
+      "loss": 0.9314272999763489,
+      "step": 1674
+    },
+    {
+      "epoch": 0.7071729957805907,
+      "grad_norm": 0.4515164792537689,
+      "learning_rate": 0.00019916963683123574,
+      "loss": 0.860816240310669,
+      "step": 1676
+    },
+    {
+      "epoch": 0.7080168776371308,
+      "grad_norm": 0.42735886573791504,
+      "learning_rate": 0.00019916344876688303,
+      "loss": 0.8008456826210022,
+      "step": 1678
+    },
+    {
+      "epoch": 0.7088607594936709,
+      "grad_norm": 0.418275386095047,
+      "learning_rate": 0.00019915723782742718,
+      "loss": 0.7796595096588135,
+      "step": 1680
+    },
+    {
+      "epoch": 0.7097046413502109,
+      "grad_norm": 0.44347989559173584,
+      "learning_rate": 0.00019915100401430089,
+      "loss": 0.9006506204605103,
+      "step": 1682
+    },
+    {
+      "epoch": 0.7105485232067511,
+      "grad_norm": 0.3932702839374542,
+      "learning_rate": 0.00019914474732894225,
+      "loss": 0.7750146985054016,
+      "step": 1684
+    },
+    {
+      "epoch": 0.7113924050632912,
+      "grad_norm": 0.4810502231121063,
+      "learning_rate": 0.00019913846777279447,
+      "loss": 0.7508097290992737,
+      "step": 1686
+    },
+    {
+      "epoch": 0.7122362869198312,
+      "grad_norm": 0.4166778028011322,
+      "learning_rate": 0.00019913216534730622,
+      "loss": 0.7926796078681946,
+      "step": 1688
+    },
+    {
+      "epoch": 0.7130801687763713,
+      "grad_norm": 0.47900474071502686,
+      "learning_rate": 0.00019912584005393123,
+      "loss": 0.8363928198814392,
+      "step": 1690
+    },
+    {
+      "epoch": 0.7139240506329114,
+      "grad_norm": 0.435143381357193,
+      "learning_rate": 0.00019911949189412872,
+      "loss": 0.8649623990058899,
+      "step": 1692
+    },
+    {
+      "epoch": 0.7147679324894515,
+      "grad_norm": 0.4070208966732025,
+      "learning_rate": 0.00019911312086936305,
+      "loss": 0.8196499943733215,
+      "step": 1694
+    },
+    {
+      "epoch": 0.7156118143459915,
+      "grad_norm": 0.4717724621295929,
+      "learning_rate": 0.0001991067269811039,
+      "loss": 0.9157413840293884,
+      "step": 1696
+    },
+    {
+      "epoch": 0.7164556962025317,
+      "grad_norm": 0.3942880928516388,
+      "learning_rate": 0.0001991003102308262,
+      "loss": 0.7880456447601318,
+      "step": 1698
+    },
+    {
+      "epoch": 0.7172995780590717,
+      "grad_norm": 0.4037776589393616,
+      "learning_rate": 0.0001990938706200102,
+      "loss": 0.8088646531105042,
+      "step": 1700
+    },
+    {
+      "epoch": 0.7172995780590717,
+      "eval_loss": 0.8708170056343079,
+      "eval_runtime": 670.3019,
+      "eval_samples_per_second": 3.143,
+      "eval_steps_per_second": 3.143,
+      "step": 1700
+    },
+    {
+      "epoch": 0.7181434599156118,
+      "grad_norm": 0.4443536400794983,
+      "learning_rate": 0.0001990874081501414,
+      "loss": 0.8740324378013611,
+      "step": 1702
+    },
+    {
+      "epoch": 0.7189873417721518,
+      "grad_norm": 0.45089733600616455,
+      "learning_rate": 0.00019908092282271053,
+      "loss": 0.8305023908615112,
+      "step": 1704
+    },
+    {
+      "epoch": 0.719831223628692,
+      "grad_norm": 0.3903036415576935,
+      "learning_rate": 0.00019907441463921366,
+      "loss": 0.7718394994735718,
+      "step": 1706
+    },
+    {
+      "epoch": 0.7206751054852321,
+      "grad_norm": 0.42990729212760925,
+      "learning_rate": 0.00019906788360115208,
+      "loss": 0.843289315700531,
+      "step": 1708
+    },
+    {
+      "epoch": 0.7215189873417721,
+      "grad_norm": 0.4287833571434021,
+      "learning_rate": 0.0001990613297100324,
+      "loss": 0.8332282900810242,
+      "step": 1710
+    },
+    {
+      "epoch": 0.7223628691983123,
+      "grad_norm": 0.4557269513607025,
+      "learning_rate": 0.00019905475296736646,
+      "loss": 0.784180223941803,
+      "step": 1712
+    },
+    {
+      "epoch": 0.7232067510548523,
+      "grad_norm": 0.4001525640487671,
+      "learning_rate": 0.0001990481533746714,
+      "loss": 0.7807716131210327,
+      "step": 1714
+    },
+    {
+      "epoch": 0.7240506329113924,
+      "grad_norm": 0.5208709836006165,
+      "learning_rate": 0.0001990415309334696,
+      "loss": 0.7936414480209351,
+      "step": 1716
+    },
+    {
+      "epoch": 0.7248945147679325,
+      "grad_norm": 0.46469953656196594,
+      "learning_rate": 0.00019903488564528875,
+      "loss": 0.7964845895767212,
+      "step": 1718
+    },
+    {
+      "epoch": 0.7257383966244726,
+      "grad_norm": 0.405773401260376,
+      "learning_rate": 0.00019902821751166178,
+      "loss": 0.8302215337753296,
+      "step": 1720
+    },
+    {
+      "epoch": 0.7265822784810126,
+      "grad_norm": 0.3874973654747009,
+      "learning_rate": 0.00019902152653412692,
+      "loss": 0.8486512899398804,
+      "step": 1722
+    },
+    {
+      "epoch": 0.7274261603375527,
+      "grad_norm": 0.46865755319595337,
+      "learning_rate": 0.00019901481271422758,
+      "loss": 0.837841272354126,
+      "step": 1724
+    },
+    {
+      "epoch": 0.7282700421940929,
+      "grad_norm": 0.4139123558998108,
+      "learning_rate": 0.0001990080760535126,
+      "loss": 0.7640942335128784,
+      "step": 1726
+    },
+    {
+      "epoch": 0.7291139240506329,
+      "grad_norm": 0.5003024935722351,
+      "learning_rate": 0.000199001316553536,
+      "loss": 0.8236244320869446,
+      "step": 1728
+    },
+    {
+      "epoch": 0.729957805907173,
+      "grad_norm": 0.38495776057243347,
+      "learning_rate": 0.00019899453421585696,
+      "loss": 0.7792612314224243,
+      "step": 1730
+    },
+    {
+      "epoch": 0.7308016877637131,
+      "grad_norm": 0.4462733268737793,
+      "learning_rate": 0.00019898772904204014,
+      "loss": 0.7733872532844543,
+      "step": 1732
+    },
+    {
+      "epoch": 0.7316455696202532,
+      "grad_norm": 0.398248553276062,
+      "learning_rate": 0.00019898090103365532,
+      "loss": 0.8933147192001343,
+      "step": 1734
+    },
+    {
+      "epoch": 0.7324894514767932,
+      "grad_norm": 0.4231175482273102,
+      "learning_rate": 0.00019897405019227756,
+      "loss": 0.8198356628417969,
+      "step": 1736
+    },
+    {
+      "epoch": 0.7333333333333333,
+      "grad_norm": 0.43639084696769714,
+      "learning_rate": 0.0001989671765194873,
+      "loss": 0.8266391158103943,
+      "step": 1738
+    },
+    {
+      "epoch": 0.7341772151898734,
+      "grad_norm": 0.4235835671424866,
+      "learning_rate": 0.00019896028001687008,
+      "loss": 0.7839997410774231,
+      "step": 1740
+    },
+    {
+      "epoch": 0.7350210970464135,
+      "grad_norm": 0.4033111035823822,
+      "learning_rate": 0.00019895336068601687,
+      "loss": 0.7153518199920654,
+      "step": 1742
+    },
+    {
+      "epoch": 0.7358649789029535,
+      "grad_norm": 0.44281265139579773,
+      "learning_rate": 0.00019894641852852377,
+      "loss": 0.9079289436340332,
+      "step": 1744
+    },
+    {
+      "epoch": 0.7367088607594937,
+      "grad_norm": 0.4882747232913971,
+      "learning_rate": 0.0001989394535459922,
+      "loss": 0.8645254373550415,
+      "step": 1746
+    },
+    {
+      "epoch": 0.7375527426160338,
+      "grad_norm": 0.34858641028404236,
+      "learning_rate": 0.00019893246574002887,
+      "loss": 0.8134214878082275,
+      "step": 1748
+    },
+    {
+      "epoch": 0.7383966244725738,
+      "grad_norm": 0.4273303747177124,
+      "learning_rate": 0.00019892545511224574,
+      "loss": 0.8142663240432739,
+      "step": 1750
+    },
+    {
+      "epoch": 0.739240506329114,
+      "grad_norm": 0.4289957284927368,
+      "learning_rate": 0.00019891842166426004,
+      "loss": 0.7957643866539001,
+      "step": 1752
+    },
+    {
+      "epoch": 0.740084388185654,
+      "grad_norm": 0.4812224805355072,
+      "learning_rate": 0.00019891136539769416,
+      "loss": 0.9000363945960999,
+      "step": 1754
+    },
+    {
+      "epoch": 0.7409282700421941,
+      "grad_norm": 0.4078022837638855,
+      "learning_rate": 0.00019890428631417595,
+      "loss": 0.7719792723655701,
+      "step": 1756
+    },
+    {
+      "epoch": 0.7417721518987341,
+      "grad_norm": 0.40169858932495117,
+      "learning_rate": 0.00019889718441533838,
+      "loss": 0.7681847214698792,
+      "step": 1758
+    },
+    {
+      "epoch": 0.7426160337552743,
+      "grad_norm": 0.3713594377040863,
+      "learning_rate": 0.00019889005970281972,
+      "loss": 0.7536827325820923,
+      "step": 1760
+    },
+    {
+      "epoch": 0.7434599156118143,
+      "grad_norm": 0.4183627665042877,
+      "learning_rate": 0.0001988829121782635,
+      "loss": 0.9000160694122314,
+      "step": 1762
+    },
+    {
+      "epoch": 0.7443037974683544,
+      "grad_norm": 0.4241594672203064,
+      "learning_rate": 0.00019887574184331851,
+      "loss": 0.8047307133674622,
+      "step": 1764
+    },
+    {
+      "epoch": 0.7451476793248946,
+      "grad_norm": 0.4249233603477478,
+      "learning_rate": 0.00019886854869963883,
+      "loss": 0.8932583928108215,
+      "step": 1766
+    },
+    {
+      "epoch": 0.7459915611814346,
+      "grad_norm": 0.35815340280532837,
+      "learning_rate": 0.00019886133274888378,
+      "loss": 0.7684977054595947,
+      "step": 1768
+    },
+    {
+      "epoch": 0.7468354430379747,
+      "grad_norm": 0.41035401821136475,
+      "learning_rate": 0.00019885409399271795,
+      "loss": 0.8644338250160217,
+      "step": 1770
+    },
+    {
+      "epoch": 0.7476793248945147,
+      "grad_norm": 0.5450286269187927,
+      "learning_rate": 0.00019884683243281116,
+      "loss": 0.8035860657691956,
+      "step": 1772
+    },
+    {
+      "epoch": 0.7485232067510549,
+      "grad_norm": 0.5280999541282654,
+      "learning_rate": 0.0001988395480708385,
+      "loss": 0.8861207962036133,
+      "step": 1774
+    },
+    {
+      "epoch": 0.7493670886075949,
+      "grad_norm": 0.42665231227874756,
+      "learning_rate": 0.00019883224090848036,
+      "loss": 0.7506847977638245,
+      "step": 1776
+    },
+    {
+      "epoch": 0.750210970464135,
+      "grad_norm": 0.39029282331466675,
+      "learning_rate": 0.00019882491094742232,
+      "loss": 0.7662046551704407,
+      "step": 1778
+    },
+    {
+      "epoch": 0.7510548523206751,
+      "grad_norm": 0.43870019912719727,
+      "learning_rate": 0.00019881755818935534,
+      "loss": 0.7590143084526062,
+      "step": 1780
+    },
+    {
+      "epoch": 0.7518987341772152,
+      "grad_norm": 0.4195050001144409,
+      "learning_rate": 0.0001988101826359755,
+      "loss": 0.9340365529060364,
+      "step": 1782
+    },
+    {
+      "epoch": 0.7527426160337553,
+      "grad_norm": 0.4236123263835907,
+      "learning_rate": 0.0001988027842889842,
+      "loss": 0.7969209551811218,
+      "step": 1784
+    },
+    {
+      "epoch": 0.7535864978902953,
+      "grad_norm": 0.4274357259273529,
+      "learning_rate": 0.00019879536315008808,
+      "loss": 0.7892382144927979,
+      "step": 1786
+    },
+    {
+      "epoch": 0.7544303797468355,
+      "grad_norm": 0.4622916579246521,
+      "learning_rate": 0.00019878791922099912,
+      "loss": 0.8752562403678894,
+      "step": 1788
+    },
+    {
+      "epoch": 0.7552742616033755,
+      "grad_norm": 0.3648734986782074,
+      "learning_rate": 0.00019878045250343445,
+      "loss": 0.813011884689331,
+      "step": 1790
+    },
+    {
+      "epoch": 0.7561181434599156,
+      "grad_norm": 0.41361239552497864,
+      "learning_rate": 0.00019877296299911648,
+      "loss": 0.7310198545455933,
+      "step": 1792
+    },
+    {
+      "epoch": 0.7569620253164557,
+      "grad_norm": 0.4113222062587738,
+      "learning_rate": 0.00019876545070977294,
+      "loss": 0.858386754989624,
+      "step": 1794
+    },
+    {
+      "epoch": 0.7578059071729958,
+      "grad_norm": 0.4295370876789093,
+      "learning_rate": 0.00019875791563713676,
+      "loss": 0.8377325534820557,
+      "step": 1796
+    },
+    {
+      "epoch": 0.7586497890295358,
+      "grad_norm": 0.4227522015571594,
+      "learning_rate": 0.0001987503577829461,
+      "loss": 0.9139418005943298,
+      "step": 1798
+    },
+    {
+      "epoch": 0.759493670886076,
+      "grad_norm": 0.3638151288032532,
+      "learning_rate": 0.00019874277714894442,
+      "loss": 0.6994872689247131,
+      "step": 1800
+    },
+    {
+      "epoch": 0.759493670886076,
+      "eval_loss": 0.8625519275665283,
+      "eval_runtime": 686.4271,
+      "eval_samples_per_second": 3.07,
+      "eval_steps_per_second": 3.07,
+      "step": 1800
+    },
+    {
+      "epoch": 0.760337552742616,
+      "grad_norm": 0.46954411268234253,
+      "learning_rate": 0.00019873517373688047,
+      "loss": 0.8681167960166931,
+      "step": 1802
+    },
+    {
+      "epoch": 0.7611814345991561,
+      "grad_norm": 0.41958674788475037,
+      "learning_rate": 0.00019872754754850819,
+      "loss": 0.7420852184295654,
+      "step": 1804
+    },
+    {
+      "epoch": 0.7620253164556962,
+      "grad_norm": 0.431226521730423,
+      "learning_rate": 0.00019871989858558678,
+      "loss": 0.8018608093261719,
+      "step": 1806
+    },
+    {
+      "epoch": 0.7628691983122363,
+      "grad_norm": 0.45129454135894775,
+      "learning_rate": 0.0001987122268498807,
+      "loss": 0.8793904781341553,
+      "step": 1808
+    },
+    {
+      "epoch": 0.7637130801687764,
+      "grad_norm": 0.44278961420059204,
+      "learning_rate": 0.00019870453234315972,
+      "loss": 0.8416730761528015,
+      "step": 1810
+    },
+    {
+      "epoch": 0.7645569620253164,
+      "grad_norm": 0.4689098000526428,
+      "learning_rate": 0.00019869681506719876,
+      "loss": 0.8225743174552917,
+      "step": 1812
+    },
+    {
+      "epoch": 0.7654008438818566,
+      "grad_norm": 0.4413386881351471,
+      "learning_rate": 0.00019868907502377806,
+      "loss": 0.7871913313865662,
+      "step": 1814
+    },
+    {
+      "epoch": 0.7662447257383966,
+      "grad_norm": 0.48395273089408875,
+      "learning_rate": 0.00019868131221468315,
+      "loss": 0.8616237044334412,
+      "step": 1816
+    },
+    {
+      "epoch": 0.7670886075949367,
+      "grad_norm": 0.370263934135437,
+      "learning_rate": 0.0001986735266417047,
+      "loss": 0.7243452668190002,
+      "step": 1818
+    },
+    {
+      "epoch": 0.7679324894514767,
+      "grad_norm": 0.4522445797920227,
+      "learning_rate": 0.00019866571830663875,
+      "loss": 0.8623812198638916,
+      "step": 1820
+    },
+    {
+      "epoch": 0.7687763713080169,
+      "grad_norm": 0.43821004033088684,
+      "learning_rate": 0.00019865788721128643,
+      "loss": 0.776618480682373,
+      "step": 1822
+    },
+    {
+      "epoch": 0.769620253164557,
+      "grad_norm": 0.3825650215148926,
+      "learning_rate": 0.00019865003335745436,
+      "loss": 0.8722774982452393,
+      "step": 1824
+    },
+    {
+      "epoch": 0.770464135021097,
+      "grad_norm": 0.43684205412864685,
+      "learning_rate": 0.00019864215674695416,
+      "loss": 0.907802939414978,
+      "step": 1826
+    },
+    {
+      "epoch": 0.7713080168776372,
+      "grad_norm": 0.4128975570201874,
+      "learning_rate": 0.00019863425738160287,
+      "loss": 0.8505539298057556,
+      "step": 1828
+    },
+    {
+      "epoch": 0.7721518987341772,
+      "grad_norm": 0.4254065454006195,
+      "learning_rate": 0.0001986263352632227,
+      "loss": 0.83274906873703,
+      "step": 1830
+    },
+    {
+      "epoch": 0.7729957805907173,
+      "grad_norm": 0.4137375056743622,
+      "learning_rate": 0.00019861839039364118,
+      "loss": 0.8172078728675842,
+      "step": 1832
+    },
+    {
+      "epoch": 0.7738396624472574,
+      "grad_norm": 0.4387703239917755,
+      "learning_rate": 0.00019861042277469096,
+      "loss": 0.8610842227935791,
+      "step": 1834
+    },
+    {
+      "epoch": 0.7746835443037975,
+      "grad_norm": 0.398952454328537,
+      "learning_rate": 0.00019860243240821005,
+      "loss": 0.7961188554763794,
+      "step": 1836
+    },
+    {
+      "epoch": 0.7755274261603375,
+      "grad_norm": 0.4079921543598175,
+      "learning_rate": 0.0001985944192960417,
+      "loss": 0.8078321218490601,
+      "step": 1838
+    },
+    {
+      "epoch": 0.7763713080168776,
+      "grad_norm": 0.4701489508152008,
+      "learning_rate": 0.00019858638344003433,
+      "loss": 0.8743909001350403,
+      "step": 1840
+    },
+    {
+      "epoch": 0.7772151898734178,
+      "grad_norm": 0.5391269326210022,
+      "learning_rate": 0.00019857832484204173,
+      "loss": 0.7681707143783569,
+      "step": 1842
+    },
+    {
+      "epoch": 0.7780590717299578,
+      "grad_norm": 0.36250734329223633,
+      "learning_rate": 0.00019857024350392277,
+      "loss": 0.7817525863647461,
+      "step": 1844
+    },
+    {
+      "epoch": 0.7789029535864979,
+      "grad_norm": 0.41237136721611023,
+      "learning_rate": 0.00019856213942754175,
+      "loss": 0.7614551782608032,
+      "step": 1846
+    },
+    {
+      "epoch": 0.779746835443038,
+      "grad_norm": 0.3968869745731354,
+      "learning_rate": 0.00019855401261476807,
+      "loss": 0.7492313981056213,
+      "step": 1848
+    },
+    {
+      "epoch": 0.7805907172995781,
+      "grad_norm": 0.441476434469223,
+      "learning_rate": 0.00019854586306747646,
+      "loss": 0.8615695238113403,
+      "step": 1850
+    },
+    {
+      "epoch": 0.7814345991561181,
+      "grad_norm": 0.4486154317855835,
+      "learning_rate": 0.00019853769078754686,
+      "loss": 0.8416724801063538,
+      "step": 1852
+    },
+    {
+      "epoch": 0.7822784810126582,
+      "grad_norm": 0.45148879289627075,
+      "learning_rate": 0.0001985294957768644,
+      "loss": 0.8987806439399719,
+      "step": 1854
+    },
+    {
+      "epoch": 0.7831223628691983,
+      "grad_norm": 0.39330095052719116,
+      "learning_rate": 0.0001985212780373196,
+      "loss": 0.7597590088844299,
+      "step": 1856
+    },
+    {
+      "epoch": 0.7839662447257384,
+      "grad_norm": 0.4260408580303192,
+      "learning_rate": 0.00019851303757080808,
+      "loss": 0.8429927229881287,
+      "step": 1858
+    },
+    {
+      "epoch": 0.7848101265822784,
+      "grad_norm": 0.37519171833992004,
+      "learning_rate": 0.00019850477437923075,
+      "loss": 0.7443564534187317,
+      "step": 1860
+    },
+    {
+      "epoch": 0.7856540084388186,
+      "grad_norm": 0.4991084337234497,
+      "learning_rate": 0.00019849648846449383,
+      "loss": 0.934238612651825,
+      "step": 1862
+    },
+    {
+      "epoch": 0.7864978902953587,
+      "grad_norm": 0.4711335003376007,
+      "learning_rate": 0.00019848817982850867,
+      "loss": 0.8938905000686646,
+      "step": 1864
+    },
+    {
+      "epoch": 0.7873417721518987,
+      "grad_norm": 0.4145370423793793,
+      "learning_rate": 0.00019847984847319193,
+      "loss": 0.8745643496513367,
+      "step": 1866
+    },
+    {
+      "epoch": 0.7881856540084389,
+      "grad_norm": 0.3444044888019562,
+      "learning_rate": 0.00019847149440046548,
+      "loss": 0.7178786396980286,
+      "step": 1868
+    },
+    {
+      "epoch": 0.7890295358649789,
+      "grad_norm": 0.34632962942123413,
+      "learning_rate": 0.00019846311761225646,
+      "loss": 0.9114303588867188,
+      "step": 1870
+    },
+    {
+      "epoch": 0.789873417721519,
+      "grad_norm": 0.38630741834640503,
+      "learning_rate": 0.0001984547181104972,
+      "loss": 0.836281955242157,
+      "step": 1872
+    },
+    {
+      "epoch": 0.790717299578059,
+      "grad_norm": 0.40947434306144714,
+      "learning_rate": 0.00019844629589712534,
+      "loss": 0.8247858881950378,
+      "step": 1874
+    },
+    {
+      "epoch": 0.7915611814345992,
+      "grad_norm": 0.45878684520721436,
+      "learning_rate": 0.00019843785097408372,
+      "loss": 0.8446351289749146,
+      "step": 1876
+    },
+    {
+      "epoch": 0.7924050632911392,
+      "grad_norm": 0.37202852964401245,
+      "learning_rate": 0.0001984293833433204,
+      "loss": 0.8567686676979065,
+      "step": 1878
+    },
+    {
+      "epoch": 0.7932489451476793,
+      "grad_norm": 0.3839069902896881,
+      "learning_rate": 0.00019842089300678876,
+      "loss": 0.8134047985076904,
+      "step": 1880
+    },
+    {
+      "epoch": 0.7940928270042195,
+      "grad_norm": 0.4018687605857849,
+      "learning_rate": 0.00019841237996644727,
+      "loss": 0.8670110106468201,
+      "step": 1882
+    },
+    {
+      "epoch": 0.7949367088607595,
+      "grad_norm": 0.372089684009552,
+      "learning_rate": 0.0001984038442242598,
+      "loss": 0.7975355386734009,
+      "step": 1884
+    },
+    {
+      "epoch": 0.7957805907172996,
+      "grad_norm": 0.4492705166339874,
+      "learning_rate": 0.0001983952857821953,
+      "loss": 0.8755611777305603,
+      "step": 1886
+    },
+    {
+      "epoch": 0.7966244725738396,
+      "grad_norm": 0.451054185628891,
+      "learning_rate": 0.00019838670464222813,
+      "loss": 0.8680241107940674,
+      "step": 1888
+    },
+    {
+      "epoch": 0.7974683544303798,
+      "grad_norm": 0.4036900997161865,
+      "learning_rate": 0.00019837810080633773,
+      "loss": 0.7990990877151489,
+      "step": 1890
+    },
+    {
+      "epoch": 0.7983122362869198,
+      "grad_norm": 0.46125802397727966,
+      "learning_rate": 0.0001983694742765089,
+      "loss": 0.8598953485488892,
+      "step": 1892
+    },
+    {
+      "epoch": 0.7991561181434599,
+      "grad_norm": 0.4458785951137543,
+      "learning_rate": 0.00019836082505473153,
+      "loss": 0.7707474827766418,
+      "step": 1894
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.44501590728759766,
+      "learning_rate": 0.0001983521531430009,
+      "loss": 0.8069534301757812,
+      "step": 1896
+    },
+    {
+      "epoch": 0.8008438818565401,
+      "grad_norm": 0.400771826505661,
+      "learning_rate": 0.0001983434585433174,
+      "loss": 0.862334668636322,
+      "step": 1898
+    },
+    {
+      "epoch": 0.8016877637130801,
+      "grad_norm": 0.39193347096443176,
+      "learning_rate": 0.00019833474125768676,
+      "loss": 0.8125433325767517,
+      "step": 1900
+    },
+    {
+      "epoch": 0.8016877637130801,
+      "eval_loss": 0.8546335697174072,
+      "eval_runtime": 688.5301,
+      "eval_samples_per_second": 3.06,
+      "eval_steps_per_second": 3.06,
+      "step": 1900
+    },
+    {
+      "epoch": 0.8025316455696202,
+      "grad_norm": 0.441129207611084,
+      "learning_rate": 0.00019832600128811986,
+      "loss": 0.8246012330055237,
+      "step": 1902
+    },
+    {
+      "epoch": 0.8033755274261604,
+      "grad_norm": 0.4089467227458954,
+      "learning_rate": 0.00019831723863663285,
+      "loss": 0.8387641310691833,
+      "step": 1904
+    },
+    {
+      "epoch": 0.8042194092827004,
+      "grad_norm": 0.4871654510498047,
+      "learning_rate": 0.0001983084533052471,
+      "loss": 0.7891429662704468,
+      "step": 1906
+    },
+    {
+      "epoch": 0.8050632911392405,
+      "grad_norm": 0.40085604786872864,
+      "learning_rate": 0.00019829964529598923,
+      "loss": 0.8008774518966675,
+      "step": 1908
+    },
+    {
+      "epoch": 0.8059071729957806,
+      "grad_norm": 0.38775014877319336,
+      "learning_rate": 0.00019829081461089109,
+      "loss": 0.7921834588050842,
+      "step": 1910
+    },
+    {
+      "epoch": 0.8067510548523207,
+      "grad_norm": 0.37830594182014465,
+      "learning_rate": 0.0001982819612519897,
+      "loss": 0.7752519845962524,
+      "step": 1912
+    },
+    {
+      "epoch": 0.8075949367088607,
+      "grad_norm": 0.41137194633483887,
+      "learning_rate": 0.0001982730852213274,
+      "loss": 0.7958255410194397,
+      "step": 1914
+    },
+    {
+      "epoch": 0.8084388185654009,
+      "grad_norm": 0.4871830940246582,
+      "learning_rate": 0.0001982641865209517,
+      "loss": 0.8177281618118286,
+      "step": 1916
+    },
+    {
+      "epoch": 0.809282700421941,
+      "grad_norm": 0.4157388210296631,
+      "learning_rate": 0.0001982552651529154,
+      "loss": 0.7726616859436035,
+      "step": 1918
+    },
+    {
+      "epoch": 0.810126582278481,
+      "grad_norm": 0.32338014245033264,
+      "learning_rate": 0.00019824632111927645,
+      "loss": 0.6601400375366211,
+      "step": 1920
+    },
+    {
+      "epoch": 0.810970464135021,
+      "grad_norm": 0.4642949104309082,
+      "learning_rate": 0.00019823735442209804,
+      "loss": 0.7866622805595398,
+      "step": 1922
+    },
+    {
+      "epoch": 0.8118143459915612,
+      "grad_norm": 0.43558797240257263,
+      "learning_rate": 0.00019822836506344865,
+      "loss": 0.7857986092567444,
+      "step": 1924
+    },
+    {
+      "epoch": 0.8126582278481013,
+      "grad_norm": 0.3996953070163727,
+      "learning_rate": 0.000198219353045402,
+      "loss": 0.8407763838768005,
+      "step": 1926
+    },
+    {
+      "epoch": 0.8135021097046413,
+      "grad_norm": 0.4415414333343506,
+      "learning_rate": 0.00019821031837003686,
+      "loss": 0.8433752655982971,
+      "step": 1928
+    },
+    {
+      "epoch": 0.8143459915611815,
+      "grad_norm": 0.37329310178756714,
+      "learning_rate": 0.00019820126103943752,
+      "loss": 0.6941158175468445,
+      "step": 1930
+    },
+    {
+      "epoch": 0.8151898734177215,
+      "grad_norm": 0.4482601284980774,
+      "learning_rate": 0.00019819218105569323,
+      "loss": 0.775604248046875,
+      "step": 1932
+    },
+    {
+      "epoch": 0.8160337552742616,
+      "grad_norm": 0.426194429397583,
+      "learning_rate": 0.00019818307842089854,
+      "loss": 0.8496418595314026,
+      "step": 1934
+    },
+    {
+      "epoch": 0.8168776371308016,
+      "grad_norm": 0.42379093170166016,
+      "learning_rate": 0.00019817395313715335,
+      "loss": 0.7767958045005798,
+      "step": 1936
+    },
+    {
+      "epoch": 0.8177215189873418,
+      "grad_norm": 0.4149782061576843,
+      "learning_rate": 0.0001981648052065626,
+      "loss": 0.7347666025161743,
+      "step": 1938
+    },
+    {
+      "epoch": 0.8185654008438819,
+      "grad_norm": 0.3927431106567383,
+      "learning_rate": 0.0001981556346312366,
+      "loss": 0.7839647531509399,
+      "step": 1940
+    },
+    {
+      "epoch": 0.8194092827004219,
+      "grad_norm": 0.4483136236667633,
+      "learning_rate": 0.00019814644141329083,
+      "loss": 0.8870531916618347,
+      "step": 1942
+    },
+    {
+      "epoch": 0.8202531645569621,
+      "grad_norm": 0.4400147497653961,
+      "learning_rate": 0.00019813722555484594,
+      "loss": 0.8646620512008667,
+      "step": 1944
+    },
+    {
+      "epoch": 0.8210970464135021,
+      "grad_norm": 0.469971626996994,
+      "learning_rate": 0.00019812798705802785,
+      "loss": 0.9576541781425476,
+      "step": 1946
+    },
+    {
+      "epoch": 0.8219409282700422,
+      "grad_norm": 0.388113409280777,
+      "learning_rate": 0.00019811872592496776,
+      "loss": 0.7956477403640747,
+      "step": 1948
+    },
+    {
+      "epoch": 0.8227848101265823,
+      "grad_norm": 0.4271424114704132,
+      "learning_rate": 0.00019810944215780201,
+      "loss": 0.8440352082252502,
+      "step": 1950
+    },
+    {
+      "epoch": 0.8236286919831224,
+      "grad_norm": 0.4097966253757477,
+      "learning_rate": 0.00019810013575867217,
+      "loss": 0.8589065074920654,
+      "step": 1952
+    },
+    {
+      "epoch": 0.8244725738396624,
+      "grad_norm": 0.41550710797309875,
+      "learning_rate": 0.00019809080672972504,
+      "loss": 0.8606626987457275,
+      "step": 1954
+    },
+    {
+      "epoch": 0.8253164556962025,
+      "grad_norm": 0.3680916130542755,
+      "learning_rate": 0.0001980814550731127,
+      "loss": 0.7582436800003052,
+      "step": 1956
+    },
+    {
+      "epoch": 0.8261603375527427,
+      "grad_norm": 0.573297917842865,
+      "learning_rate": 0.00019807208079099234,
+      "loss": 0.8345186710357666,
+      "step": 1958
+    },
+    {
+      "epoch": 0.8270042194092827,
+      "grad_norm": 0.43631303310394287,
+      "learning_rate": 0.00019806268388552646,
+      "loss": 0.889683723449707,
+      "step": 1960
+    },
+    {
+      "epoch": 0.8278481012658228,
+      "grad_norm": 0.442531019449234,
+      "learning_rate": 0.00019805326435888275,
+      "loss": 0.8092973232269287,
+      "step": 1962
+    },
+    {
+      "epoch": 0.8286919831223629,
+      "grad_norm": 0.45020824670791626,
+      "learning_rate": 0.0001980438222132341,
+      "loss": 0.8919535279273987,
+      "step": 1964
+    },
+    {
+      "epoch": 0.829535864978903,
+      "grad_norm": 0.43131786584854126,
+      "learning_rate": 0.0001980343574507586,
+      "loss": 0.7652381658554077,
+      "step": 1966
+    },
+    {
+      "epoch": 0.830379746835443,
+      "grad_norm": 0.4461326599121094,
+      "learning_rate": 0.00019802487007363967,
+      "loss": 0.8512389659881592,
+      "step": 1968
+    },
+    {
+      "epoch": 0.8312236286919831,
+      "grad_norm": 0.4676894247531891,
+      "learning_rate": 0.0001980153600840658,
+      "loss": 0.8681327700614929,
+      "step": 1970
+    },
+    {
+      "epoch": 0.8320675105485232,
+      "grad_norm": 0.38376888632774353,
+      "learning_rate": 0.00019800582748423075,
+      "loss": 0.7476955652236938,
+      "step": 1972
+    },
+    {
+      "epoch": 0.8329113924050633,
+      "grad_norm": 0.37173229455947876,
+      "learning_rate": 0.00019799627227633354,
+      "loss": 0.814192533493042,
+      "step": 1974
+    },
+    {
+      "epoch": 0.8337552742616033,
+      "grad_norm": 0.4345237910747528,
+      "learning_rate": 0.00019798669446257844,
+      "loss": 0.7668994069099426,
+      "step": 1976
+    },
+    {
+      "epoch": 0.8345991561181435,
+      "grad_norm": 0.45872583985328674,
+      "learning_rate": 0.0001979770940451747,
+      "loss": 0.8321775197982788,
+      "step": 1978
+    },
+    {
+      "epoch": 0.8354430379746836,
+      "grad_norm": 0.40618205070495605,
+      "learning_rate": 0.0001979674710263371,
+      "loss": 0.7824342250823975,
+      "step": 1980
+    },
+    {
+      "epoch": 0.8362869198312236,
+      "grad_norm": 0.38377392292022705,
+      "learning_rate": 0.00019795782540828544,
+      "loss": 0.8752480745315552,
+      "step": 1982
+    },
+    {
+      "epoch": 0.8371308016877637,
+      "grad_norm": 0.36340072751045227,
+      "learning_rate": 0.00019794815719324476,
+      "loss": 0.66700679063797,
+      "step": 1984
+    },
+    {
+      "epoch": 0.8379746835443038,
+      "grad_norm": 0.4638384282588959,
+      "learning_rate": 0.00019793846638344536,
+      "loss": 0.8582209944725037,
+      "step": 1986
+    },
+    {
+      "epoch": 0.8388185654008439,
+      "grad_norm": 0.38364699482917786,
+      "learning_rate": 0.00019792875298112268,
+      "loss": 0.7724968791007996,
+      "step": 1988
+    },
+    {
+      "epoch": 0.8396624472573839,
+      "grad_norm": 0.439730703830719,
+      "learning_rate": 0.00019791901698851749,
+      "loss": 0.8542404174804688,
+      "step": 1990
+    },
+    {
+      "epoch": 0.8405063291139241,
+      "grad_norm": 0.4622722566127777,
+      "learning_rate": 0.00019790925840787563,
+      "loss": 0.8424296975135803,
+      "step": 1992
+    },
+    {
+      "epoch": 0.8413502109704641,
+      "grad_norm": 0.4491989314556122,
+      "learning_rate": 0.0001978994772414482,
+      "loss": 0.8400710225105286,
+      "step": 1994
+    },
+    {
+      "epoch": 0.8421940928270042,
+      "grad_norm": 0.372295081615448,
+      "learning_rate": 0.0001978896734914916,
+      "loss": 0.7265452146530151,
+      "step": 1996
+    },
+    {
+      "epoch": 0.8430379746835444,
+      "grad_norm": 0.4328666925430298,
+      "learning_rate": 0.00019787984716026732,
+      "loss": 0.8458228707313538,
+      "step": 1998
+    },
+    {
+      "epoch": 0.8438818565400844,
+      "grad_norm": 0.42246317863464355,
+      "learning_rate": 0.00019786999825004216,
+      "loss": 0.8517491817474365,
+      "step": 2000
+    },
+    {
+      "epoch": 0.8438818565400844,
+      "eval_loss": 0.8460908532142639,
+      "eval_runtime": 685.2518,
+      "eval_samples_per_second": 3.075,
+      "eval_steps_per_second": 3.075,
+      "step": 2000
+    },
+    {
+      "epoch": 0.8447257383966245,
+      "grad_norm": 0.3927323818206787,
+      "learning_rate": 0.00019786012676308802,
+      "loss": 0.8045481443405151,
+      "step": 2002
+    },
+    {
+      "epoch": 0.8455696202531645,
+      "grad_norm": 0.38189369440078735,
+      "learning_rate": 0.00019785023270168202,
+      "loss": 0.7509838938713074,
+      "step": 2004
+    },
+    {
+      "epoch": 0.8464135021097047,
+      "grad_norm": 0.40383535623550415,
+      "learning_rate": 0.00019784031606810663,
+      "loss": 0.7730163335800171,
+      "step": 2006
+    },
+    {
+      "epoch": 0.8472573839662447,
+      "grad_norm": 0.3665419816970825,
+      "learning_rate": 0.00019783037686464935,
+      "loss": 0.7096828818321228,
+      "step": 2008
+    },
+    {
+      "epoch": 0.8481012658227848,
+      "grad_norm": 0.36623379588127136,
+      "learning_rate": 0.00019782041509360302,
+      "loss": 0.8186401128768921,
+      "step": 2010
+    },
+    {
+      "epoch": 0.8489451476793249,
+      "grad_norm": 0.477255642414093,
+      "learning_rate": 0.0001978104307572656,
+      "loss": 0.8502879738807678,
+      "step": 2012
+    },
+    {
+      "epoch": 0.849789029535865,
+      "grad_norm": 0.44456636905670166,
+      "learning_rate": 0.00019780042385794032,
+      "loss": 0.9010884761810303,
+      "step": 2014
+    },
+    {
+      "epoch": 0.850632911392405,
+      "grad_norm": 0.42378175258636475,
+      "learning_rate": 0.00019779039439793552,
+      "loss": 0.774183452129364,
+      "step": 2016
+    },
+    {
+      "epoch": 0.8514767932489451,
+      "grad_norm": 0.43021997809410095,
+      "learning_rate": 0.00019778034237956482,
+      "loss": 0.8487694263458252,
+      "step": 2018
+    },
+    {
+      "epoch": 0.8523206751054853,
+      "grad_norm": 0.38188812136650085,
+      "learning_rate": 0.00019777026780514704,
+      "loss": 0.8004868030548096,
+      "step": 2020
+    },
+    {
+      "epoch": 0.8531645569620253,
+      "grad_norm": 0.3684130609035492,
+      "learning_rate": 0.0001977601706770062,
+      "loss": 0.7680953145027161,
+      "step": 2022
+    },
+    {
+      "epoch": 0.8540084388185654,
+      "grad_norm": 0.444844514131546,
+      "learning_rate": 0.00019775005099747153,
+      "loss": 0.8990264534950256,
+      "step": 2024
+    },
+    {
+      "epoch": 0.8548523206751055,
+      "grad_norm": 0.3432365655899048,
+      "learning_rate": 0.00019773990876887737,
+      "loss": 0.7676048874855042,
+      "step": 2026
+    },
+    {
+      "epoch": 0.8556962025316456,
+      "grad_norm": 0.3728618025779724,
+      "learning_rate": 0.00019772974399356343,
+      "loss": 0.8546769618988037,
+      "step": 2028
+    },
+    {
+      "epoch": 0.8565400843881856,
+      "grad_norm": 0.42767980694770813,
+      "learning_rate": 0.00019771955667387449,
+      "loss": 0.8009977340698242,
+      "step": 2030
+    },
+    {
+      "epoch": 0.8573839662447258,
+      "grad_norm": 0.4125710129737854,
+      "learning_rate": 0.00019770934681216055,
+      "loss": 0.8424299955368042,
+      "step": 2032
+    },
+    {
+      "epoch": 0.8582278481012658,
+      "grad_norm": 0.4399266839027405,
+      "learning_rate": 0.00019769911441077685,
+      "loss": 0.7940502762794495,
+      "step": 2034
+    },
+    {
+      "epoch": 0.8590717299578059,
+      "grad_norm": 0.41797709465026855,
+      "learning_rate": 0.0001976888594720838,
+      "loss": 0.9424314498901367,
+      "step": 2036
+    },
+    {
+      "epoch": 0.859915611814346,
+      "grad_norm": 0.5245009660720825,
+      "learning_rate": 0.00019767858199844698,
+      "loss": 0.9102896451950073,
+      "step": 2038
+    },
+    {
+      "epoch": 0.8607594936708861,
+      "grad_norm": 0.4092593193054199,
+      "learning_rate": 0.00019766828199223728,
+      "loss": 0.7434720396995544,
+      "step": 2040
+    },
+    {
+      "epoch": 0.8616033755274262,
+      "grad_norm": 0.40787473320961,
+      "learning_rate": 0.00019765795945583068,
+      "loss": 0.7441987991333008,
+      "step": 2042
+    },
+    {
+      "epoch": 0.8624472573839662,
+      "grad_norm": 0.42438021302223206,
+      "learning_rate": 0.0001976476143916084,
+      "loss": 0.8317872881889343,
+      "step": 2044
+    },
+    {
+      "epoch": 0.8632911392405064,
+      "grad_norm": 0.38261914253234863,
+      "learning_rate": 0.00019763724680195682,
+      "loss": 0.764509379863739,
+      "step": 2046
+    },
+    {
+      "epoch": 0.8641350210970464,
+      "grad_norm": 0.43295741081237793,
+      "learning_rate": 0.00019762685668926758,
+      "loss": 0.8666167855262756,
+      "step": 2048
+    },
+    {
+      "epoch": 0.8649789029535865,
+      "grad_norm": 0.36384403705596924,
+      "learning_rate": 0.00019761644405593743,
+      "loss": 0.7606754302978516,
+      "step": 2050
+    },
+    {
+      "epoch": 0.8658227848101265,
+      "grad_norm": 0.4296814501285553,
+      "learning_rate": 0.0001976060089043684,
+      "loss": 0.827495813369751,
+      "step": 2052
+    },
+    {
+      "epoch": 0.8666666666666667,
+      "grad_norm": 0.3831455111503601,
+      "learning_rate": 0.0001975955512369677,
+      "loss": 0.7418577075004578,
+      "step": 2054
+    },
+    {
+      "epoch": 0.8675105485232067,
+      "grad_norm": 0.44710344076156616,
+      "learning_rate": 0.0001975850710561477,
+      "loss": 0.8603045344352722,
+      "step": 2056
+    },
+    {
+      "epoch": 0.8683544303797468,
+      "grad_norm": 0.44012558460235596,
+      "learning_rate": 0.00019757456836432593,
+      "loss": 0.900658905506134,
+      "step": 2058
+    },
+    {
+      "epoch": 0.869198312236287,
+      "grad_norm": 0.42166298627853394,
+      "learning_rate": 0.0001975640431639252,
+      "loss": 0.8416497707366943,
+      "step": 2060
+    },
+    {
+      "epoch": 0.870042194092827,
+      "grad_norm": 0.40824490785598755,
+      "learning_rate": 0.00019755349545737345,
+      "loss": 0.8281066417694092,
+      "step": 2062
+    },
+    {
+      "epoch": 0.8708860759493671,
+      "grad_norm": 0.4272223711013794,
+      "learning_rate": 0.00019754292524710388,
+      "loss": 0.83869469165802,
+      "step": 2064
+    },
+    {
+      "epoch": 0.8717299578059071,
+      "grad_norm": 0.3732798397541046,
+      "learning_rate": 0.00019753233253555478,
+      "loss": 0.8337644934654236,
+      "step": 2066
+    },
+    {
+      "epoch": 0.8725738396624473,
+      "grad_norm": 0.4048069417476654,
+      "learning_rate": 0.00019752171732516973,
+      "loss": 0.7677452564239502,
+      "step": 2068
+    },
+    {
+      "epoch": 0.8734177215189873,
+      "grad_norm": 0.43628430366516113,
+      "learning_rate": 0.00019751107961839742,
+      "loss": 0.8320774435997009,
+      "step": 2070
+    },
+    {
+      "epoch": 0.8742616033755274,
+      "grad_norm": 0.43674105405807495,
+      "learning_rate": 0.00019750041941769174,
+      "loss": 0.8480262160301208,
+      "step": 2072
+    },
+    {
+      "epoch": 0.8751054852320675,
+      "grad_norm": 0.3678584098815918,
+      "learning_rate": 0.0001974897367255119,
+      "loss": 0.7598503828048706,
+      "step": 2074
+    },
+    {
+      "epoch": 0.8759493670886076,
+      "grad_norm": 0.4316411316394806,
+      "learning_rate": 0.0001974790315443221,
+      "loss": 0.7907084822654724,
+      "step": 2076
+    },
+    {
+      "epoch": 0.8767932489451477,
+      "grad_norm": 0.426465779542923,
+      "learning_rate": 0.00019746830387659182,
+      "loss": 0.8287727236747742,
+      "step": 2078
+    },
+    {
+      "epoch": 0.8776371308016878,
+      "grad_norm": 0.447337806224823,
+      "learning_rate": 0.00019745755372479578,
+      "loss": 0.8464577794075012,
+      "step": 2080
+    },
+    {
+      "epoch": 0.8784810126582279,
+      "grad_norm": 0.37877580523490906,
+      "learning_rate": 0.0001974467810914138,
+      "loss": 0.7383086681365967,
+      "step": 2082
+    },
+    {
+      "epoch": 0.8793248945147679,
+      "grad_norm": 0.3914664089679718,
+      "learning_rate": 0.0001974359859789309,
+      "loss": 0.729221522808075,
+      "step": 2084
+    },
+    {
+      "epoch": 0.880168776371308,
+      "grad_norm": 0.39767366647720337,
+      "learning_rate": 0.00019742516838983733,
+      "loss": 0.692332923412323,
+      "step": 2086
+    },
+    {
+      "epoch": 0.8810126582278481,
+      "grad_norm": 0.4396754205226898,
+      "learning_rate": 0.0001974143283266285,
+      "loss": 0.8540068864822388,
+      "step": 2088
+    },
+    {
+      "epoch": 0.8818565400843882,
+      "grad_norm": 0.4968172609806061,
+      "learning_rate": 0.00019740346579180502,
+      "loss": 0.8570998907089233,
+      "step": 2090
+    },
+    {
+      "epoch": 0.8827004219409282,
+      "grad_norm": 0.41686809062957764,
+      "learning_rate": 0.00019739258078787264,
+      "loss": 0.784139096736908,
+      "step": 2092
+    },
+    {
+      "epoch": 0.8835443037974684,
+      "grad_norm": 0.38788047432899475,
+      "learning_rate": 0.0001973816733173423,
+      "loss": 0.8521397113800049,
+      "step": 2094
+    },
+    {
+      "epoch": 0.8843881856540085,
+      "grad_norm": 0.38369178771972656,
+      "learning_rate": 0.00019737074338273023,
+      "loss": 0.8457935452461243,
+      "step": 2096
+    },
+    {
+      "epoch": 0.8852320675105485,
+      "grad_norm": 0.38869398832321167,
+      "learning_rate": 0.0001973597909865577,
+      "loss": 0.7425959706306458,
+      "step": 2098
+    },
+    {
+      "epoch": 0.8860759493670886,
+      "grad_norm": 0.4067917466163635,
+      "learning_rate": 0.00019734881613135122,
+      "loss": 0.7742393016815186,
+      "step": 2100
+    },
+    {
+      "epoch": 0.8860759493670886,
+      "eval_loss": 0.8401098847389221,
+      "eval_runtime": 669.1149,
+      "eval_samples_per_second": 3.149,
+      "eval_steps_per_second": 3.149,
+      "step": 2100
+    },
+    {
+      "epoch": 0.8869198312236287,
+      "grad_norm": 0.4282098412513733,
+      "learning_rate": 0.00019733781881964242,
+      "loss": 0.9099814295768738,
+      "step": 2102
+    },
+    {
+      "epoch": 0.8877637130801688,
+      "grad_norm": 0.33882173895835876,
+      "learning_rate": 0.00019732679905396826,
+      "loss": 0.7458856105804443,
+      "step": 2104
+    },
+    {
+      "epoch": 0.8886075949367088,
+      "grad_norm": 0.43959516286849976,
+      "learning_rate": 0.00019731575683687078,
+      "loss": 0.7829693555831909,
+      "step": 2106
+    },
+    {
+      "epoch": 0.889451476793249,
+      "grad_norm": 0.40046483278274536,
+      "learning_rate": 0.00019730469217089716,
+      "loss": 0.7898997068405151,
+      "step": 2108
+    },
+    {
+      "epoch": 0.890295358649789,
+      "grad_norm": 0.40572553873062134,
+      "learning_rate": 0.00019729360505859984,
+      "loss": 0.7874186635017395,
+      "step": 2110
+    },
+    {
+      "epoch": 0.8911392405063291,
+      "grad_norm": 0.42086347937583923,
+      "learning_rate": 0.0001972824955025364,
+      "loss": 0.8499898910522461,
+      "step": 2112
+    },
+    {
+      "epoch": 0.8919831223628693,
+      "grad_norm": 0.38845229148864746,
+      "learning_rate": 0.00019727136350526956,
+      "loss": 0.7940409779548645,
+      "step": 2114
+    },
+    {
+      "epoch": 0.8928270042194093,
+      "grad_norm": 0.41575267910957336,
+      "learning_rate": 0.00019726020906936728,
+      "loss": 0.8279082775115967,
+      "step": 2116
+    },
+    {
+      "epoch": 0.8936708860759494,
+      "grad_norm": 0.37615057826042175,
+      "learning_rate": 0.0001972490321974027,
+      "loss": 0.8217757940292358,
+      "step": 2118
+    },
+    {
+      "epoch": 0.8945147679324894,
+      "grad_norm": 0.9318385124206543,
+      "learning_rate": 0.00019723783289195415,
+      "loss": 0.7637750506401062,
+      "step": 2120
+    },
+    {
+      "epoch": 0.8953586497890296,
+      "grad_norm": 0.41948962211608887,
+      "learning_rate": 0.000197226611155605,
+      "loss": 0.8702366948127747,
+      "step": 2122
+    },
+    {
+      "epoch": 0.8962025316455696,
+      "grad_norm": 0.4718417525291443,
+      "learning_rate": 0.00019721536699094392,
+      "loss": 0.8146984577178955,
+      "step": 2124
+    },
+    {
+      "epoch": 0.8970464135021097,
+      "grad_norm": 0.4470657408237457,
+      "learning_rate": 0.00019720410040056474,
+      "loss": 0.8646075129508972,
+      "step": 2126
+    },
+    {
+      "epoch": 0.8978902953586498,
+      "grad_norm": 0.4176540970802307,
+      "learning_rate": 0.00019719281138706643,
+      "loss": 0.7925992012023926,
+      "step": 2128
+    },
+    {
+      "epoch": 0.8987341772151899,
+      "grad_norm": 0.4854719042778015,
+      "learning_rate": 0.00019718149995305315,
+      "loss": 0.8142143487930298,
+      "step": 2130
+    },
+    {
+      "epoch": 0.8995780590717299,
+      "grad_norm": 0.4270440340042114,
+      "learning_rate": 0.00019717016610113427,
+      "loss": 0.8050890564918518,
+      "step": 2132
+    },
+    {
+      "epoch": 0.90042194092827,
+      "grad_norm": 0.37710970640182495,
+      "learning_rate": 0.00019715880983392422,
+      "loss": 0.7888354659080505,
+      "step": 2134
+    },
+    {
+      "epoch": 0.9012658227848102,
+      "grad_norm": 0.4181252717971802,
+      "learning_rate": 0.00019714743115404273,
+      "loss": 0.7348225712776184,
+      "step": 2136
+    },
+    {
+      "epoch": 0.9021097046413502,
+      "grad_norm": 0.4475359618663788,
+      "learning_rate": 0.0001971360300641146,
+      "loss": 0.7729623913764954,
+      "step": 2138
+    },
+    {
+      "epoch": 0.9029535864978903,
+      "grad_norm": 0.4052332937717438,
+      "learning_rate": 0.00019712460656676992,
+      "loss": 0.8066384792327881,
+      "step": 2140
+    },
+    {
+      "epoch": 0.9037974683544304,
+      "grad_norm": 0.4567393362522125,
+      "learning_rate": 0.0001971131606646438,
+      "loss": 0.8053154349327087,
+      "step": 2142
+    },
+    {
+      "epoch": 0.9046413502109705,
+      "grad_norm": 0.4304409325122833,
+      "learning_rate": 0.00019710169236037656,
+      "loss": 0.7242560386657715,
+      "step": 2144
+    },
+    {
+      "epoch": 0.9054852320675105,
+      "grad_norm": 0.4785008132457733,
+      "learning_rate": 0.0001970902016566138,
+      "loss": 0.8475301265716553,
+      "step": 2146
+    },
+    {
+      "epoch": 0.9063291139240506,
+      "grad_norm": 0.45226508378982544,
+      "learning_rate": 0.00019707868855600618,
+      "loss": 0.928544282913208,
+      "step": 2148
+    },
+    {
+      "epoch": 0.9071729957805907,
+      "grad_norm": 0.3794468641281128,
+      "learning_rate": 0.00019706715306120956,
+      "loss": 0.8135722875595093,
+      "step": 2150
+    },
+    {
+      "epoch": 0.9080168776371308,
+      "grad_norm": 0.4005141258239746,
+      "learning_rate": 0.0001970555951748849,
+      "loss": 0.7808290719985962,
+      "step": 2152
+    },
+    {
+      "epoch": 0.9088607594936708,
+      "grad_norm": 0.37929919362068176,
+      "learning_rate": 0.00019704401489969842,
+      "loss": 0.8292917013168335,
+      "step": 2154
+    },
+    {
+      "epoch": 0.909704641350211,
+      "grad_norm": 0.4458678066730499,
+      "learning_rate": 0.0001970324122383215,
+      "loss": 0.772884726524353,
+      "step": 2156
+    },
+    {
+      "epoch": 0.9105485232067511,
+      "grad_norm": 0.41907885670661926,
+      "learning_rate": 0.0001970207871934306,
+      "loss": 0.8204470872879028,
+      "step": 2158
+    },
+    {
+      "epoch": 0.9113924050632911,
+      "grad_norm": 0.40610212087631226,
+      "learning_rate": 0.00019700913976770741,
+      "loss": 0.8361451625823975,
+      "step": 2160
+    },
+    {
+      "epoch": 0.9122362869198313,
+      "grad_norm": 0.4213273227214813,
+      "learning_rate": 0.00019699746996383878,
+      "loss": 0.8456075191497803,
+      "step": 2162
+    },
+    {
+      "epoch": 0.9130801687763713,
+      "grad_norm": 0.49264612793922424,
+      "learning_rate": 0.0001969857777845167,
+      "loss": 0.8731603622436523,
+      "step": 2164
+    },
+    {
+      "epoch": 0.9139240506329114,
+      "grad_norm": 0.42002132534980774,
+      "learning_rate": 0.00019697406323243834,
+      "loss": 0.8149666786193848,
+      "step": 2166
+    },
+    {
+      "epoch": 0.9147679324894514,
+      "grad_norm": 0.37501421570777893,
+      "learning_rate": 0.000196962326310306,
+      "loss": 0.7508546710014343,
+      "step": 2168
+    },
+    {
+      "epoch": 0.9156118143459916,
+      "grad_norm": 0.3706159293651581,
+      "learning_rate": 0.00019695056702082717,
+      "loss": 0.830280601978302,
+      "step": 2170
+    },
+    {
+      "epoch": 0.9164556962025316,
+      "grad_norm": 0.3873467445373535,
+      "learning_rate": 0.00019693878536671452,
+      "loss": 0.7297167778015137,
+      "step": 2172
+    },
+    {
+      "epoch": 0.9172995780590717,
+      "grad_norm": 0.42876651883125305,
+      "learning_rate": 0.00019692698135068582,
+      "loss": 0.7924292087554932,
+      "step": 2174
+    },
+    {
+      "epoch": 0.9181434599156119,
+      "grad_norm": 0.4259159564971924,
+      "learning_rate": 0.00019691515497546406,
+      "loss": 0.8098247051239014,
+      "step": 2176
+    },
+    {
+      "epoch": 0.9189873417721519,
+      "grad_norm": 0.38152721524238586,
+      "learning_rate": 0.00019690330624377728,
+      "loss": 0.6789063215255737,
+      "step": 2178
+    },
+    {
+      "epoch": 0.919831223628692,
+      "grad_norm": 0.4135437607765198,
+      "learning_rate": 0.00019689143515835887,
+      "loss": 0.8582620620727539,
+      "step": 2180
+    },
+    {
+      "epoch": 0.920675105485232,
+      "grad_norm": 0.3818798065185547,
+      "learning_rate": 0.0001968795417219472,
+      "loss": 0.8059743642807007,
+      "step": 2182
+    },
+    {
+      "epoch": 0.9215189873417722,
+      "grad_norm": 0.39021286368370056,
+      "learning_rate": 0.00019686762593728582,
+      "loss": 0.7432259917259216,
+      "step": 2184
+    },
+    {
+      "epoch": 0.9223628691983122,
+      "grad_norm": 0.37864068150520325,
+      "learning_rate": 0.00019685568780712355,
+      "loss": 0.7561086416244507,
+      "step": 2186
+    },
+    {
+      "epoch": 0.9232067510548523,
+      "grad_norm": 0.450847864151001,
+      "learning_rate": 0.00019684372733421425,
+      "loss": 0.7552722692489624,
+      "step": 2188
+    },
+    {
+      "epoch": 0.9240506329113924,
+      "grad_norm": 0.37909507751464844,
+      "learning_rate": 0.00019683174452131695,
+      "loss": 0.6905336380004883,
+      "step": 2190
+    },
+    {
+      "epoch": 0.9248945147679325,
+      "grad_norm": 0.4439173638820648,
+      "learning_rate": 0.0001968197393711959,
+      "loss": 0.7710593938827515,
+      "step": 2192
+    },
+    {
+      "epoch": 0.9257383966244725,
+      "grad_norm": 0.3908376395702362,
+      "learning_rate": 0.00019680771188662044,
+      "loss": 0.8212814331054688,
+      "step": 2194
+    },
+    {
+      "epoch": 0.9265822784810127,
+      "grad_norm": 0.3528299033641815,
+      "learning_rate": 0.00019679566207036507,
+      "loss": 0.751355767250061,
+      "step": 2196
+    },
+    {
+      "epoch": 0.9274261603375528,
+      "grad_norm": 0.4351397156715393,
+      "learning_rate": 0.00019678358992520945,
+      "loss": 0.8204057812690735,
+      "step": 2198
+    },
+    {
+      "epoch": 0.9282700421940928,
+      "grad_norm": 0.45566943287849426,
+      "learning_rate": 0.0001967714954539384,
+      "loss": 0.763841450214386,
+      "step": 2200
+    },
+    {
+      "epoch": 0.9282700421940928,
+      "eval_loss": 0.8336610198020935,
+      "eval_runtime": 674.5134,
+      "eval_samples_per_second": 3.124,
+      "eval_steps_per_second": 3.124,
+      "step": 2200
+    },
+    {
+      "epoch": 0.9291139240506329,
+      "grad_norm": 0.4516221284866333,
+      "learning_rate": 0.00019675937865934188,
+      "loss": 0.861711323261261,
+      "step": 2202
+    },
+    {
+      "epoch": 0.929957805907173,
+      "grad_norm": 0.43421855568885803,
+      "learning_rate": 0.00019674723954421502,
+      "loss": 0.8062718510627747,
+      "step": 2204
+    },
+    {
+      "epoch": 0.9308016877637131,
+      "grad_norm": 0.42442476749420166,
+      "learning_rate": 0.00019673507811135803,
+      "loss": 0.8069695234298706,
+      "step": 2206
+    },
+    {
+      "epoch": 0.9316455696202531,
+      "grad_norm": 0.3976045250892639,
+      "learning_rate": 0.00019672289436357637,
+      "loss": 0.7336108684539795,
+      "step": 2208
+    },
+    {
+      "epoch": 0.9324894514767933,
+      "grad_norm": 0.4131270945072174,
+      "learning_rate": 0.00019671068830368058,
+      "loss": 0.7559890151023865,
+      "step": 2210
+    },
+    {
+      "epoch": 0.9333333333333333,
+      "grad_norm": 0.4599294364452362,
+      "learning_rate": 0.00019669845993448634,
+      "loss": 0.8447075486183167,
+      "step": 2212
+    },
+    {
+      "epoch": 0.9341772151898734,
+      "grad_norm": 0.4291204810142517,
+      "learning_rate": 0.00019668620925881455,
+      "loss": 0.8329233527183533,
+      "step": 2214
+    },
+    {
+      "epoch": 0.9350210970464135,
+      "grad_norm": 0.4226539731025696,
+      "learning_rate": 0.00019667393627949116,
+      "loss": 0.749232292175293,
+      "step": 2216
+    },
+    {
+      "epoch": 0.9358649789029536,
+      "grad_norm": 0.4159580171108246,
+      "learning_rate": 0.00019666164099934732,
+      "loss": 0.7470011115074158,
+      "step": 2218
+    },
+    {
+      "epoch": 0.9367088607594937,
+      "grad_norm": 0.4160766303539276,
+      "learning_rate": 0.00019664932342121935,
+      "loss": 0.8094202280044556,
+      "step": 2220
+    },
+    {
+      "epoch": 0.9375527426160337,
+      "grad_norm": 0.45284298062324524,
+      "learning_rate": 0.00019663698354794863,
+      "loss": 0.8398169875144958,
+      "step": 2222
+    },
+    {
+      "epoch": 0.9383966244725739,
+      "grad_norm": 0.3960540294647217,
+      "learning_rate": 0.00019662462138238178,
+      "loss": 0.7652443051338196,
+      "step": 2224
+    },
+    {
+      "epoch": 0.9392405063291139,
+      "grad_norm": 0.4269275367259979,
+      "learning_rate": 0.00019661223692737047,
+      "loss": 0.7909489870071411,
+      "step": 2226
+    },
+    {
+      "epoch": 0.940084388185654,
+      "grad_norm": 0.38187873363494873,
+      "learning_rate": 0.00019659983018577157,
+      "loss": 0.7821935415267944,
+      "step": 2228
+    },
+    {
+      "epoch": 0.9409282700421941,
+      "grad_norm": 0.45898866653442383,
+      "learning_rate": 0.0001965874011604471,
+      "loss": 0.7792721390724182,
+      "step": 2230
+    },
+    {
+      "epoch": 0.9417721518987342,
+      "grad_norm": 0.4740699827671051,
+      "learning_rate": 0.00019657494985426418,
+      "loss": 0.885725200176239,
+      "step": 2232
+    },
+    {
+      "epoch": 0.9426160337552743,
+      "grad_norm": 0.3912813365459442,
+      "learning_rate": 0.00019656247627009506,
+      "loss": 0.7898101806640625,
+      "step": 2234
+    },
+    {
+      "epoch": 0.9434599156118143,
+      "grad_norm": 0.4552418291568756,
+      "learning_rate": 0.0001965499804108172,
+      "loss": 0.8646330833435059,
+      "step": 2236
+    },
+    {
+      "epoch": 0.9443037974683545,
+      "grad_norm": 0.4178548753261566,
+      "learning_rate": 0.0001965374622793131,
+      "loss": 0.8068296313285828,
+      "step": 2238
+    },
+    {
+      "epoch": 0.9451476793248945,
+      "grad_norm": 0.38649076223373413,
+      "learning_rate": 0.00019652492187847057,
+      "loss": 0.7953271865844727,
+      "step": 2240
+    },
+    {
+      "epoch": 0.9459915611814346,
+      "grad_norm": 0.39785075187683105,
+      "learning_rate": 0.0001965123592111823,
+      "loss": 0.8423402309417725,
+      "step": 2242
+    },
+    {
+      "epoch": 0.9468354430379747,
+      "grad_norm": 0.47455263137817383,
+      "learning_rate": 0.00019649977428034632,
+      "loss": 0.8883589506149292,
+      "step": 2244
+    },
+    {
+      "epoch": 0.9476793248945148,
+      "grad_norm": 0.3866841197013855,
+      "learning_rate": 0.0001964871670888657,
+      "loss": 0.7232863903045654,
+      "step": 2246
+    },
+    {
+      "epoch": 0.9485232067510548,
+      "grad_norm": 0.3950609564781189,
+      "learning_rate": 0.00019647453763964876,
+      "loss": 0.7265347838401794,
+      "step": 2248
+    },
+    {
+      "epoch": 0.9493670886075949,
+      "grad_norm": 0.4315829575061798,
+      "learning_rate": 0.0001964618859356088,
+      "loss": 0.8000131249427795,
+      "step": 2250
+    },
+    {
+      "epoch": 0.950210970464135,
+      "grad_norm": 0.403984934091568,
+      "learning_rate": 0.00019644921197966433,
+      "loss": 0.823086678981781,
+      "step": 2252
+    },
+    {
+      "epoch": 0.9510548523206751,
+      "grad_norm": 0.4311080574989319,
+      "learning_rate": 0.00019643651577473897,
+      "loss": 0.8457553386688232,
+      "step": 2254
+    },
+    {
+      "epoch": 0.9518987341772152,
+      "grad_norm": 0.3884856104850769,
+      "learning_rate": 0.00019642379732376158,
+      "loss": 0.7866189479827881,
+      "step": 2256
+    },
+    {
+      "epoch": 0.9527426160337553,
+      "grad_norm": 0.3384067714214325,
+      "learning_rate": 0.00019641105662966595,
+      "loss": 0.7095813751220703,
+      "step": 2258
+    },
+    {
+      "epoch": 0.9535864978902954,
+      "grad_norm": 0.40426746010780334,
+      "learning_rate": 0.00019639829369539117,
+      "loss": 0.8271632194519043,
+      "step": 2260
+    },
+    {
+      "epoch": 0.9544303797468354,
+      "grad_norm": 0.45495131611824036,
+      "learning_rate": 0.00019638550852388144,
+      "loss": 0.8204861283302307,
+      "step": 2262
+    },
+    {
+      "epoch": 0.9552742616033755,
+      "grad_norm": 0.47201067209243774,
+      "learning_rate": 0.00019637270111808598,
+      "loss": 0.884763777256012,
+      "step": 2264
+    },
+    {
+      "epoch": 0.9561181434599156,
+      "grad_norm": 0.3959396779537201,
+      "learning_rate": 0.0001963598714809592,
+      "loss": 0.7630926370620728,
+      "step": 2266
+    },
+    {
+      "epoch": 0.9569620253164557,
+      "grad_norm": 0.4197136461734772,
+      "learning_rate": 0.00019634701961546077,
+      "loss": 0.797870397567749,
+      "step": 2268
+    },
+    {
+      "epoch": 0.9578059071729957,
+      "grad_norm": 0.381290078163147,
+      "learning_rate": 0.00019633414552455526,
+      "loss": 0.7755846977233887,
+      "step": 2270
+    },
+    {
+      "epoch": 0.9586497890295359,
+      "grad_norm": 0.3868124783039093,
+      "learning_rate": 0.00019632124921121255,
+      "loss": 0.7553264498710632,
+      "step": 2272
+    },
+    {
+      "epoch": 0.959493670886076,
+      "grad_norm": 0.4350631833076477,
+      "learning_rate": 0.00019630833067840747,
+      "loss": 0.7792160511016846,
+      "step": 2274
+    },
+    {
+      "epoch": 0.960337552742616,
+      "grad_norm": 0.4717157781124115,
+      "learning_rate": 0.00019629538992912016,
+      "loss": 0.878220796585083,
+      "step": 2276
+    },
+    {
+      "epoch": 0.9611814345991562,
+      "grad_norm": 0.45545122027397156,
+      "learning_rate": 0.00019628242696633584,
+      "loss": 0.8891014456748962,
+      "step": 2278
+    },
+    {
+      "epoch": 0.9620253164556962,
+      "grad_norm": 0.4421025514602661,
+      "learning_rate": 0.00019626944179304467,
+      "loss": 0.8024723529815674,
+      "step": 2280
+    },
+    {
+      "epoch": 0.9628691983122363,
+      "grad_norm": 0.40385812520980835,
+      "learning_rate": 0.00019625643441224222,
+      "loss": 0.8036280274391174,
+      "step": 2282
+    },
+    {
+      "epoch": 0.9637130801687763,
+      "grad_norm": 0.3866552412509918,
+      "learning_rate": 0.00019624340482692898,
+      "loss": 0.8072395324707031,
+      "step": 2284
+    },
+    {
+      "epoch": 0.9645569620253165,
+      "grad_norm": 0.4113233685493469,
+      "learning_rate": 0.0001962303530401106,
+      "loss": 0.8046114444732666,
+      "step": 2286
+    },
+    {
+      "epoch": 0.9654008438818565,
+      "grad_norm": 0.40720227360725403,
+      "learning_rate": 0.00019621727905479798,
+      "loss": 0.8434037566184998,
+      "step": 2288
+    },
+    {
+      "epoch": 0.9662447257383966,
+      "grad_norm": 0.35759273171424866,
+      "learning_rate": 0.00019620418287400693,
+      "loss": 0.7832419276237488,
+      "step": 2290
+    },
+    {
+      "epoch": 0.9670886075949368,
+      "grad_norm": 0.4170442223548889,
+      "learning_rate": 0.0001961910645007585,
+      "loss": 0.7975693345069885,
+      "step": 2292
+    },
+    {
+      "epoch": 0.9679324894514768,
+      "grad_norm": 0.4126700162887573,
+      "learning_rate": 0.00019617792393807895,
+      "loss": 0.8554494976997375,
+      "step": 2294
+    },
+    {
+      "epoch": 0.9687763713080169,
+      "grad_norm": 0.4231695830821991,
+      "learning_rate": 0.00019616476118899942,
+      "loss": 0.824539065361023,
+      "step": 2296
+    },
+    {
+      "epoch": 0.9696202531645569,
+      "grad_norm": 0.3556209206581116,
+      "learning_rate": 0.0001961515762565564,
+      "loss": 0.7008040547370911,
+      "step": 2298
+    },
+    {
+      "epoch": 0.9704641350210971,
+      "grad_norm": 0.4370400309562683,
+      "learning_rate": 0.00019613836914379132,
+      "loss": 0.8192558288574219,
+      "step": 2300
+    },
+    {
+      "epoch": 0.9704641350210971,
+      "eval_loss": 0.8281980156898499,
+      "eval_runtime": 688.6136,
+      "eval_samples_per_second": 3.06,
+      "eval_steps_per_second": 3.06,
+      "step": 2300
+    },
+    {
+      "epoch": 0.9713080168776371,
+      "grad_norm": 0.4207614064216614,
+      "learning_rate": 0.00019612513985375088,
+      "loss": 0.878928005695343,
+      "step": 2302
+    },
+    {
+      "epoch": 0.9721518987341772,
+      "grad_norm": 0.37607815861701965,
+      "learning_rate": 0.0001961118883894868,
+      "loss": 0.736182689666748,
+      "step": 2304
+    },
+    {
+      "epoch": 0.9729957805907173,
+      "grad_norm": 0.43311476707458496,
+      "learning_rate": 0.00019609861475405592,
+      "loss": 0.8403618335723877,
+      "step": 2306
+    },
+    {
+      "epoch": 0.9738396624472574,
+      "grad_norm": 0.4533176124095917,
+      "learning_rate": 0.00019608531895052021,
+      "loss": 0.8414564728736877,
+      "step": 2308
+    },
+    {
+      "epoch": 0.9746835443037974,
+      "grad_norm": 0.4025697410106659,
+      "learning_rate": 0.00019607200098194677,
+      "loss": 0.8105775117874146,
+      "step": 2310
+    },
+    {
+      "epoch": 0.9755274261603376,
+      "grad_norm": 0.37894198298454285,
+      "learning_rate": 0.00019605866085140784,
+      "loss": 0.8053821325302124,
+      "step": 2312
+    },
+    {
+      "epoch": 0.9763713080168777,
+      "grad_norm": 0.46265891194343567,
+      "learning_rate": 0.0001960452985619807,
+      "loss": 0.8092884421348572,
+      "step": 2314
+    },
+    {
+      "epoch": 0.9772151898734177,
+      "grad_norm": 0.3770054280757904,
+      "learning_rate": 0.0001960319141167477,
+      "loss": 0.7456095218658447,
+      "step": 2316
+    },
+    {
+      "epoch": 0.9780590717299578,
+      "grad_norm": 0.44084450602531433,
+      "learning_rate": 0.0001960185075187965,
+      "loss": 0.7739883661270142,
+      "step": 2318
+    },
+    {
+      "epoch": 0.9789029535864979,
+      "grad_norm": 0.3880341351032257,
+      "learning_rate": 0.00019600507877121965,
+      "loss": 0.6961302757263184,
+      "step": 2320
+    },
+    {
+      "epoch": 0.979746835443038,
+      "grad_norm": 0.468037486076355,
+      "learning_rate": 0.00019599162787711495,
+      "loss": 0.8447984457015991,
+      "step": 2322
+    },
+    {
+      "epoch": 0.980590717299578,
+      "grad_norm": 0.4123631417751312,
+      "learning_rate": 0.00019597815483958527,
+      "loss": 0.8155615329742432,
+      "step": 2324
+    },
+    {
+      "epoch": 0.9814345991561182,
+      "grad_norm": 0.3827875554561615,
+      "learning_rate": 0.00019596465966173853,
+      "loss": 0.7990952730178833,
+      "step": 2326
+    },
+    {
+      "epoch": 0.9822784810126582,
+      "grad_norm": 0.4644222557544708,
+      "learning_rate": 0.00019595114234668788,
+      "loss": 0.8899750709533691,
+      "step": 2328
+    },
+    {
+      "epoch": 0.9831223628691983,
+      "grad_norm": 0.43699249625205994,
+      "learning_rate": 0.00019593760289755144,
+      "loss": 0.7863160371780396,
+      "step": 2330
+    },
+    {
+      "epoch": 0.9839662447257383,
+      "grad_norm": 0.40713244676589966,
+      "learning_rate": 0.00019592404131745254,
+      "loss": 0.7960637807846069,
+      "step": 2332
+    },
+    {
+      "epoch": 0.9848101265822785,
+      "grad_norm": 0.37364068627357483,
+      "learning_rate": 0.00019591045760951957,
+      "loss": 0.714144229888916,
+      "step": 2334
+    },
+    {
+      "epoch": 0.9856540084388186,
+      "grad_norm": 0.4094736576080322,
+      "learning_rate": 0.00019589685177688599,
+      "loss": 0.7311791777610779,
+      "step": 2336
+    },
+    {
+      "epoch": 0.9864978902953586,
+      "grad_norm": 0.4275180697441101,
+      "learning_rate": 0.0001958832238226905,
+      "loss": 0.8191342353820801,
+      "step": 2338
+    },
+    {
+      "epoch": 0.9873417721518988,
+      "grad_norm": 0.4627651274204254,
+      "learning_rate": 0.00019586957375007668,
+      "loss": 0.7476106286048889,
+      "step": 2340
+    },
+    {
+      "epoch": 0.9881856540084388,
+      "grad_norm": 0.44676414132118225,
+      "learning_rate": 0.00019585590156219346,
+      "loss": 0.8257967233657837,
+      "step": 2342
+    },
+    {
+      "epoch": 0.9890295358649789,
+      "grad_norm": 0.40667134523391724,
+      "learning_rate": 0.00019584220726219466,
+      "loss": 0.7922079563140869,
+      "step": 2344
+    },
+    {
+      "epoch": 0.9898734177215189,
+      "grad_norm": 0.40865346789360046,
+      "learning_rate": 0.00019582849085323934,
+      "loss": 0.7892434597015381,
+      "step": 2346
+    },
+    {
+      "epoch": 0.9907172995780591,
+      "grad_norm": 0.43309709429740906,
+      "learning_rate": 0.00019581475233849163,
+      "loss": 0.8766644597053528,
+      "step": 2348
+    },
+    {
+      "epoch": 0.9915611814345991,
+      "grad_norm": 0.41558781266212463,
+      "learning_rate": 0.00019580099172112069,
+      "loss": 0.9183312058448792,
+      "step": 2350
+    },
+    {
+      "epoch": 0.9924050632911392,
+      "grad_norm": 0.432748019695282,
+      "learning_rate": 0.00019578720900430082,
+      "loss": 0.7905668020248413,
+      "step": 2352
+    },
+    {
+      "epoch": 0.9932489451476794,
+      "grad_norm": 0.4116985499858856,
+      "learning_rate": 0.0001957734041912115,
+      "loss": 0.8279712200164795,
+      "step": 2354
+    },
+    {
+      "epoch": 0.9940928270042194,
+      "grad_norm": 0.3927275240421295,
+      "learning_rate": 0.00019575957728503718,
+      "loss": 0.7451362609863281,
+      "step": 2356
+    },
+    {
+      "epoch": 0.9949367088607595,
+      "grad_norm": 0.3389509320259094,
+      "learning_rate": 0.0001957457282889675,
+      "loss": 0.6769390106201172,
+      "step": 2358
+    },
+    {
+      "epoch": 0.9957805907172996,
+      "grad_norm": 0.4220774471759796,
+      "learning_rate": 0.00019573185720619712,
+      "loss": 0.7590897083282471,
+      "step": 2360
+    },
+    {
+      "epoch": 0.9966244725738397,
+      "grad_norm": 0.4299897849559784,
+      "learning_rate": 0.00019571796403992583,
+      "loss": 0.8301284313201904,
+      "step": 2362
+    },
+    {
+      "epoch": 0.9974683544303797,
+      "grad_norm": 0.4031789302825928,
+      "learning_rate": 0.00019570404879335856,
+      "loss": 0.8019471764564514,
+      "step": 2364
+    },
+    {
+      "epoch": 0.9983122362869198,
+      "grad_norm": 0.3957567811012268,
+      "learning_rate": 0.00019569011146970526,
+      "loss": 0.7427027821540833,
+      "step": 2366
+    },
+    {
+      "epoch": 0.99915611814346,
+      "grad_norm": 0.41741347312927246,
+      "learning_rate": 0.000195676152072181,
+      "loss": 0.8221617341041565,
+      "step": 2368
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.47121551632881165,
+      "learning_rate": 0.00019566217060400595,
+      "loss": 0.8143455982208252,
+      "step": 2370
+    },
+    {
+      "epoch": 1.0008438818565402,
+      "grad_norm": 0.3854716122150421,
+      "learning_rate": 0.0001956481670684054,
+      "loss": 0.6971694231033325,
+      "step": 2372
+    },
+    {
+      "epoch": 1.00168776371308,
+      "grad_norm": 0.4029138386249542,
+      "learning_rate": 0.00019563414146860964,
+      "loss": 0.6719155311584473,
+      "step": 2374
+    },
+    {
+      "epoch": 1.0025316455696203,
+      "grad_norm": 0.40136656165122986,
+      "learning_rate": 0.00019562009380785412,
+      "loss": 0.7161872386932373,
+      "step": 2376
+    },
+    {
+      "epoch": 1.0033755274261604,
+      "grad_norm": 0.40685850381851196,
+      "learning_rate": 0.0001956060240893794,
+      "loss": 0.71699059009552,
+      "step": 2378
+    },
+    {
+      "epoch": 1.0042194092827004,
+      "grad_norm": 0.46153366565704346,
+      "learning_rate": 0.00019559193231643104,
+      "loss": 0.8030059337615967,
+      "step": 2380
+    },
+    {
+      "epoch": 1.0050632911392405,
+      "grad_norm": 0.45048823952674866,
+      "learning_rate": 0.0001955778184922598,
+      "loss": 0.7666479349136353,
+      "step": 2382
+    },
+    {
+      "epoch": 1.0059071729957807,
+      "grad_norm": 0.4260413348674774,
+      "learning_rate": 0.00019556368262012142,
+      "loss": 0.7481234073638916,
+      "step": 2384
+    },
+    {
+      "epoch": 1.0067510548523206,
+      "grad_norm": 0.4819575548171997,
+      "learning_rate": 0.0001955495247032768,
+      "loss": 0.7943820953369141,
+      "step": 2386
+    },
+    {
+      "epoch": 1.0075949367088608,
+      "grad_norm": 0.45789605379104614,
+      "learning_rate": 0.0001955353447449919,
+      "loss": 0.6401634812355042,
+      "step": 2388
+    },
+    {
+      "epoch": 1.0084388185654007,
+      "grad_norm": 0.37019097805023193,
+      "learning_rate": 0.00019552114274853778,
+      "loss": 0.7058886885643005,
+      "step": 2390
+    },
+    {
+      "epoch": 1.009282700421941,
+      "grad_norm": 0.44416263699531555,
+      "learning_rate": 0.0001955069187171905,
+      "loss": 0.7631242871284485,
+      "step": 2392
+    },
+    {
+      "epoch": 1.010126582278481,
+      "grad_norm": 0.4478379786014557,
+      "learning_rate": 0.00019549267265423138,
+      "loss": 0.7873892784118652,
+      "step": 2394
+    },
+    {
+      "epoch": 1.010970464135021,
+      "grad_norm": 0.4596130847930908,
+      "learning_rate": 0.00019547840456294658,
+      "loss": 0.7882059216499329,
+      "step": 2396
+    },
+    {
+      "epoch": 1.0118143459915612,
+      "grad_norm": 0.4062999188899994,
+      "learning_rate": 0.0001954641144466276,
+      "loss": 0.6358574628829956,
+      "step": 2398
+    },
+    {
+      "epoch": 1.0126582278481013,
+      "grad_norm": 0.47627049684524536,
+      "learning_rate": 0.0001954498023085708,
+      "loss": 0.7803886532783508,
+      "step": 2400
+    },
+    {
+      "epoch": 1.0126582278481013,
+      "eval_loss": 0.8250564932823181,
+      "eval_runtime": 691.5833,
+      "eval_samples_per_second": 3.047,
+      "eval_steps_per_second": 3.047,
+      "step": 2400
+    },
+    {
+      "epoch": 1.0135021097046413,
+      "grad_norm": 0.40690696239471436,
+      "learning_rate": 0.00019543546815207777,
+      "loss": 0.7394700050354004,
+      "step": 2402
+    },
+    {
+      "epoch": 1.0143459915611814,
+      "grad_norm": 0.45543980598449707,
+      "learning_rate": 0.0001954211119804551,
+      "loss": 0.7096555829048157,
+      "step": 2404
+    },
+    {
+      "epoch": 1.0151898734177216,
+      "grad_norm": 0.4367314279079437,
+      "learning_rate": 0.00019540673379701454,
+      "loss": 0.7961843013763428,
+      "step": 2406
+    },
+    {
+      "epoch": 1.0160337552742615,
+      "grad_norm": 0.4458340108394623,
+      "learning_rate": 0.00019539233360507278,
+      "loss": 0.743843138217926,
+      "step": 2408
+    },
+    {
+      "epoch": 1.0168776371308017,
+      "grad_norm": 0.3741893470287323,
+      "learning_rate": 0.0001953779114079517,
+      "loss": 0.7273756265640259,
+      "step": 2410
+    },
+    {
+      "epoch": 1.0177215189873419,
+      "grad_norm": 0.4340079426765442,
+      "learning_rate": 0.00019536346720897823,
+      "loss": 0.7922506928443909,
+      "step": 2412
+    },
+    {
+      "epoch": 1.0185654008438818,
+      "grad_norm": 0.4518018662929535,
+      "learning_rate": 0.00019534900101148437,
+      "loss": 0.6492037773132324,
+      "step": 2414
+    },
+    {
+      "epoch": 1.019409282700422,
+      "grad_norm": 0.4160727560520172,
+      "learning_rate": 0.00019533451281880716,
+      "loss": 0.7422950863838196,
+      "step": 2416
+    },
+    {
+      "epoch": 1.0202531645569621,
+      "grad_norm": 0.35953211784362793,
+      "learning_rate": 0.00019532000263428883,
+      "loss": 0.6489207148551941,
+      "step": 2418
+    },
+    {
+      "epoch": 1.021097046413502,
+      "grad_norm": 0.38774433732032776,
+      "learning_rate": 0.0001953054704612765,
+      "loss": 0.7449417114257812,
+      "step": 2420
+    },
+    {
+      "epoch": 1.0219409282700422,
+      "grad_norm": 0.48185858130455017,
+      "learning_rate": 0.0001952909163031225,
+      "loss": 0.8481494188308716,
+      "step": 2422
+    },
+    {
+      "epoch": 1.0227848101265822,
+      "grad_norm": 0.4098491072654724,
+      "learning_rate": 0.00019527634016318425,
+      "loss": 0.7442291378974915,
+      "step": 2424
+    },
+    {
+      "epoch": 1.0236286919831223,
+      "grad_norm": 0.41284480690956116,
+      "learning_rate": 0.00019526174204482412,
+      "loss": 0.7564480900764465,
+      "step": 2426
+    },
+    {
+      "epoch": 1.0244725738396625,
+      "grad_norm": 0.44948479533195496,
+      "learning_rate": 0.00019524712195140965,
+      "loss": 0.7234854102134705,
+      "step": 2428
+    },
+    {
+      "epoch": 1.0253164556962024,
+      "grad_norm": 0.39939233660697937,
+      "learning_rate": 0.0001952324798863134,
+      "loss": 0.7638729810714722,
+      "step": 2430
+    },
+    {
+      "epoch": 1.0261603375527426,
+      "grad_norm": 0.40807482600212097,
+      "learning_rate": 0.00019521781585291298,
+      "loss": 0.6922630071640015,
+      "step": 2432
+    },
+    {
+      "epoch": 1.0270042194092828,
+      "grad_norm": 0.45735791325569153,
+      "learning_rate": 0.00019520312985459116,
+      "loss": 0.762996256351471,
+      "step": 2434
+    },
+    {
+      "epoch": 1.0278481012658227,
+      "grad_norm": 0.4280409812927246,
+      "learning_rate": 0.0001951884218947357,
+      "loss": 0.7811596989631653,
+      "step": 2436
+    },
+    {
+      "epoch": 1.0286919831223629,
+      "grad_norm": 0.41957560181617737,
+      "learning_rate": 0.00019517369197673943,
+      "loss": 0.7278305292129517,
+      "step": 2438
+    },
+    {
+      "epoch": 1.029535864978903,
+      "grad_norm": 0.4832586646080017,
+      "learning_rate": 0.00019515894010400028,
+      "loss": 0.807456910610199,
+      "step": 2440
+    },
+    {
+      "epoch": 1.030379746835443,
+      "grad_norm": 0.46448788046836853,
+      "learning_rate": 0.00019514416627992122,
+      "loss": 0.7421302795410156,
+      "step": 2442
+    },
+    {
+      "epoch": 1.0312236286919831,
+      "grad_norm": 0.41362619400024414,
+      "learning_rate": 0.00019512937050791024,
+      "loss": 0.6892213225364685,
+      "step": 2444
+    },
+    {
+      "epoch": 1.0320675105485233,
+      "grad_norm": 0.4158109128475189,
+      "learning_rate": 0.0001951145527913805,
+      "loss": 0.7305331826210022,
+      "step": 2446
+    },
+    {
+      "epoch": 1.0329113924050632,
+      "grad_norm": 0.4725087881088257,
+      "learning_rate": 0.00019509971313375011,
+      "loss": 0.7682753205299377,
+      "step": 2448
+    },
+    {
+      "epoch": 1.0337552742616034,
+      "grad_norm": 0.4341438114643097,
+      "learning_rate": 0.00019508485153844237,
+      "loss": 0.747041642665863,
+      "step": 2450
+    },
+    {
+      "epoch": 1.0345991561181433,
+      "grad_norm": 0.44515377283096313,
+      "learning_rate": 0.00019506996800888548,
+      "loss": 0.7312062382698059,
+      "step": 2452
+    },
+    {
+      "epoch": 1.0354430379746835,
+      "grad_norm": 0.5313915014266968,
+      "learning_rate": 0.00019505506254851283,
+      "loss": 0.7861107587814331,
+      "step": 2454
+    },
+    {
+      "epoch": 1.0362869198312237,
+      "grad_norm": 0.6094246506690979,
+      "learning_rate": 0.00019504013516076285,
+      "loss": 0.8458115458488464,
+      "step": 2456
+    },
+    {
+      "epoch": 1.0371308016877636,
+      "grad_norm": 0.36419153213500977,
+      "learning_rate": 0.00019502518584907895,
+      "loss": 0.6091256141662598,
+      "step": 2458
+    },
+    {
+      "epoch": 1.0379746835443038,
+      "grad_norm": 0.399271696805954,
+      "learning_rate": 0.00019501021461690966,
+      "loss": 0.6703370809555054,
+      "step": 2460
+    },
+    {
+      "epoch": 1.038818565400844,
+      "grad_norm": 0.38560932874679565,
+      "learning_rate": 0.00019499522146770856,
+      "loss": 0.7220943570137024,
+      "step": 2462
+    },
+    {
+      "epoch": 1.0396624472573839,
+      "grad_norm": 0.44199803471565247,
+      "learning_rate": 0.00019498020640493428,
+      "loss": 0.7501406073570251,
+      "step": 2464
+    },
+    {
+      "epoch": 1.040506329113924,
+      "grad_norm": 0.47394421696662903,
+      "learning_rate": 0.00019496516943205054,
+      "loss": 0.7957925796508789,
+      "step": 2466
+    },
+    {
+      "epoch": 1.0413502109704642,
+      "grad_norm": 0.5274425745010376,
+      "learning_rate": 0.00019495011055252603,
+      "loss": 0.7964824438095093,
+      "step": 2468
+    },
+    {
+      "epoch": 1.0421940928270041,
+      "grad_norm": 0.44004881381988525,
+      "learning_rate": 0.0001949350297698346,
+      "loss": 0.7478582262992859,
+      "step": 2470
+    },
+    {
+      "epoch": 1.0430379746835443,
+      "grad_norm": 0.40662670135498047,
+      "learning_rate": 0.000194919927087455,
+      "loss": 0.738339900970459,
+      "step": 2472
+    },
+    {
+      "epoch": 1.0438818565400845,
+      "grad_norm": 0.43934181332588196,
+      "learning_rate": 0.00019490480250887123,
+      "loss": 0.817171037197113,
+      "step": 2474
+    },
+    {
+      "epoch": 1.0447257383966244,
+      "grad_norm": 0.4111221432685852,
+      "learning_rate": 0.00019488965603757223,
+      "loss": 0.7413169145584106,
+      "step": 2476
+    },
+    {
+      "epoch": 1.0455696202531646,
+      "grad_norm": 0.3770996630191803,
+      "learning_rate": 0.00019487448767705194,
+      "loss": 0.7893102169036865,
+      "step": 2478
+    },
+    {
+      "epoch": 1.0464135021097047,
+      "grad_norm": 0.4524461627006531,
+      "learning_rate": 0.00019485929743080945,
+      "loss": 0.7037841081619263,
+      "step": 2480
+    },
+    {
+      "epoch": 1.0472573839662447,
+      "grad_norm": 0.4232378900051117,
+      "learning_rate": 0.00019484408530234885,
+      "loss": 0.7456525564193726,
+      "step": 2482
+    },
+    {
+      "epoch": 1.0481012658227848,
+      "grad_norm": 0.3842810392379761,
+      "learning_rate": 0.00019482885129517929,
+      "loss": 0.7263808250427246,
+      "step": 2484
+    },
+    {
+      "epoch": 1.048945147679325,
+      "grad_norm": 0.43203726410865784,
+      "learning_rate": 0.00019481359541281495,
+      "loss": 0.7101474404335022,
+      "step": 2486
+    },
+    {
+      "epoch": 1.049789029535865,
+      "grad_norm": 0.46944454312324524,
+      "learning_rate": 0.00019479831765877509,
+      "loss": 0.8561921715736389,
+      "step": 2488
+    },
+    {
+      "epoch": 1.0506329113924051,
+      "grad_norm": 0.4455019533634186,
+      "learning_rate": 0.00019478301803658397,
+      "loss": 0.7206679582595825,
+      "step": 2490
+    },
+    {
+      "epoch": 1.051476793248945,
+      "grad_norm": 0.4253813326358795,
+      "learning_rate": 0.00019476769654977095,
+      "loss": 0.7271745800971985,
+      "step": 2492
+    },
+    {
+      "epoch": 1.0523206751054852,
+      "grad_norm": 0.42758429050445557,
+      "learning_rate": 0.00019475235320187033,
+      "loss": 0.7358576059341431,
+      "step": 2494
+    },
+    {
+      "epoch": 1.0531645569620254,
+      "grad_norm": 0.400634765625,
+      "learning_rate": 0.00019473698799642164,
+      "loss": 0.6617647409439087,
+      "step": 2496
+    },
+    {
+      "epoch": 1.0540084388185653,
+      "grad_norm": 0.4132104814052582,
+      "learning_rate": 0.00019472160093696926,
+      "loss": 0.6900970935821533,
+      "step": 2498
+    },
+    {
+      "epoch": 1.0548523206751055,
+      "grad_norm": 0.5116495490074158,
+      "learning_rate": 0.00019470619202706268,
+      "loss": 0.7352861762046814,
+      "step": 2500
+    },
+    {
+      "epoch": 1.0548523206751055,
+      "eval_loss": 0.8249453902244568,
+      "eval_runtime": 679.4446,
+      "eval_samples_per_second": 3.101,
+      "eval_steps_per_second": 3.101,
+      "step": 2500
+    },
+    {
+      "epoch": 1.0556962025316456,
+      "grad_norm": 0.42822444438934326,
+      "learning_rate": 0.0001946907612702565,
+      "loss": 0.7583479881286621,
+      "step": 2502
+    },
+    {
+      "epoch": 1.0565400843881856,
+      "grad_norm": 0.4788980782032013,
+      "learning_rate": 0.00019467530867011025,
+      "loss": 0.777511477470398,
+      "step": 2504
+    },
+    {
+      "epoch": 1.0573839662447257,
+      "grad_norm": 0.4210498034954071,
+      "learning_rate": 0.00019465983423018856,
+      "loss": 0.7452852129936218,
+      "step": 2506
+    },
+    {
+      "epoch": 1.058227848101266,
+      "grad_norm": 0.43219906091690063,
+      "learning_rate": 0.0001946443379540611,
+      "loss": 0.7540958523750305,
+      "step": 2508
+    },
+    {
+      "epoch": 1.0590717299578059,
+      "grad_norm": 0.4361002445220947,
+      "learning_rate": 0.00019462881984530253,
+      "loss": 0.7175624966621399,
+      "step": 2510
+    },
+    {
+      "epoch": 1.059915611814346,
+      "grad_norm": 0.4575091600418091,
+      "learning_rate": 0.00019461327990749264,
+      "loss": 0.7342076301574707,
+      "step": 2512
+    },
+    {
+      "epoch": 1.0607594936708862,
+      "grad_norm": 0.4823324978351593,
+      "learning_rate": 0.00019459771814421615,
+      "loss": 0.8255488872528076,
+      "step": 2514
+    },
+    {
+      "epoch": 1.0616033755274261,
+      "grad_norm": 0.41693785786628723,
+      "learning_rate": 0.00019458213455906284,
+      "loss": 0.6624729633331299,
+      "step": 2516
+    },
+    {
+      "epoch": 1.0624472573839663,
+      "grad_norm": 0.41319167613983154,
+      "learning_rate": 0.00019456652915562762,
+      "loss": 0.8224266767501831,
+      "step": 2518
+    },
+    {
+      "epoch": 1.0632911392405062,
+      "grad_norm": 0.4405190050601959,
+      "learning_rate": 0.00019455090193751023,
+      "loss": 0.7582900524139404,
+      "step": 2520
+    },
+    {
+      "epoch": 1.0641350210970464,
+      "grad_norm": 0.4564387798309326,
+      "learning_rate": 0.0001945352529083157,
+      "loss": 0.7948101162910461,
+      "step": 2522
+    },
+    {
+      "epoch": 1.0649789029535865,
+      "grad_norm": 0.4247225224971771,
+      "learning_rate": 0.00019451958207165394,
+      "loss": 0.7670508027076721,
+      "step": 2524
+    },
+    {
+      "epoch": 1.0658227848101265,
+      "grad_norm": 0.42070862650871277,
+      "learning_rate": 0.00019450388943113983,
+      "loss": 0.7753698229789734,
+      "step": 2526
+    },
+    {
+      "epoch": 1.0666666666666667,
+      "grad_norm": 0.3729179799556732,
+      "learning_rate": 0.00019448817499039342,
+      "loss": 0.710904598236084,
+      "step": 2528
+    },
+    {
+      "epoch": 1.0675105485232068,
+      "grad_norm": 0.40901753306388855,
+      "learning_rate": 0.0001944724387530397,
+      "loss": 0.7000213265419006,
+      "step": 2530
+    },
+    {
+      "epoch": 1.0683544303797468,
+      "grad_norm": 0.4481421113014221,
+      "learning_rate": 0.00019445668072270877,
+      "loss": 0.7374410629272461,
+      "step": 2532
+    },
+    {
+      "epoch": 1.069198312236287,
+      "grad_norm": 0.4839915335178375,
+      "learning_rate": 0.00019444090090303567,
+      "loss": 0.746309220790863,
+      "step": 2534
+    },
+    {
+      "epoch": 1.070042194092827,
+      "grad_norm": 0.4982132613658905,
+      "learning_rate": 0.00019442509929766048,
+      "loss": 0.7448083162307739,
+      "step": 2536
+    },
+    {
+      "epoch": 1.070886075949367,
+      "grad_norm": 0.3987475037574768,
+      "learning_rate": 0.00019440927591022838,
+      "loss": 0.7052555680274963,
+      "step": 2538
+    },
+    {
+      "epoch": 1.0717299578059072,
+      "grad_norm": 0.4089896082878113,
+      "learning_rate": 0.00019439343074438944,
+      "loss": 0.7090281248092651,
+      "step": 2540
+    },
+    {
+      "epoch": 1.0725738396624473,
+      "grad_norm": 0.3703136146068573,
+      "learning_rate": 0.00019437756380379895,
+      "loss": 0.7530601024627686,
+      "step": 2542
+    },
+    {
+      "epoch": 1.0734177215189873,
+      "grad_norm": 0.4015701413154602,
+      "learning_rate": 0.00019436167509211702,
+      "loss": 0.7597941160202026,
+      "step": 2544
+    },
+    {
+      "epoch": 1.0742616033755275,
+      "grad_norm": 0.3915148675441742,
+      "learning_rate": 0.00019434576461300888,
+      "loss": 0.7168124318122864,
+      "step": 2546
+    },
+    {
+      "epoch": 1.0751054852320676,
+      "grad_norm": 0.44643113017082214,
+      "learning_rate": 0.00019432983237014482,
+      "loss": 0.7762725949287415,
+      "step": 2548
+    },
+    {
+      "epoch": 1.0759493670886076,
+      "grad_norm": 0.46874719858169556,
+      "learning_rate": 0.00019431387836720004,
+      "loss": 0.7879725098609924,
+      "step": 2550
+    },
+    {
+      "epoch": 1.0767932489451477,
+      "grad_norm": 0.4143001437187195,
+      "learning_rate": 0.00019429790260785487,
+      "loss": 0.7351391315460205,
+      "step": 2552
+    },
+    {
+      "epoch": 1.0776371308016879,
+      "grad_norm": 0.4569888412952423,
+      "learning_rate": 0.0001942819050957946,
+      "loss": 0.86613929271698,
+      "step": 2554
+    },
+    {
+      "epoch": 1.0784810126582278,
+      "grad_norm": 0.41729897260665894,
+      "learning_rate": 0.00019426588583470954,
+      "loss": 0.7122719287872314,
+      "step": 2556
+    },
+    {
+      "epoch": 1.079324894514768,
+      "grad_norm": 0.42295604944229126,
+      "learning_rate": 0.000194249844828295,
+      "loss": 0.6605332493782043,
+      "step": 2558
+    },
+    {
+      "epoch": 1.080168776371308,
+      "grad_norm": 0.45787641406059265,
+      "learning_rate": 0.00019423378208025137,
+      "loss": 0.7839207649230957,
+      "step": 2560
+    },
+    {
+      "epoch": 1.081012658227848,
+      "grad_norm": 0.43370622396469116,
+      "learning_rate": 0.00019421769759428404,
+      "loss": 0.7557987570762634,
+      "step": 2562
+    },
+    {
+      "epoch": 1.0818565400843883,
+      "grad_norm": 0.4371975362300873,
+      "learning_rate": 0.00019420159137410335,
+      "loss": 0.7349345088005066,
+      "step": 2564
+    },
+    {
+      "epoch": 1.0827004219409282,
+      "grad_norm": 0.468730628490448,
+      "learning_rate": 0.0001941854634234247,
+      "loss": 0.7024962902069092,
+      "step": 2566
+    },
+    {
+      "epoch": 1.0835443037974684,
+      "grad_norm": 0.4257284998893738,
+      "learning_rate": 0.0001941693137459685,
+      "loss": 0.6981064081192017,
+      "step": 2568
+    },
+    {
+      "epoch": 1.0843881856540085,
+      "grad_norm": 0.4322902262210846,
+      "learning_rate": 0.00019415314234546015,
+      "loss": 0.7899637818336487,
+      "step": 2570
+    },
+    {
+      "epoch": 1.0852320675105485,
+      "grad_norm": 0.44384217262268066,
+      "learning_rate": 0.00019413694922563014,
+      "loss": 0.7964152693748474,
+      "step": 2572
+    },
+    {
+      "epoch": 1.0860759493670886,
+      "grad_norm": 0.47391360998153687,
+      "learning_rate": 0.0001941207343902139,
+      "loss": 0.7902941107749939,
+      "step": 2574
+    },
+    {
+      "epoch": 1.0869198312236288,
+      "grad_norm": 0.4575883746147156,
+      "learning_rate": 0.00019410449784295182,
+      "loss": 0.7818595170974731,
+      "step": 2576
+    },
+    {
+      "epoch": 1.0877637130801687,
+      "grad_norm": 0.41236087679862976,
+      "learning_rate": 0.0001940882395875894,
+      "loss": 0.7516021728515625,
+      "step": 2578
+    },
+    {
+      "epoch": 1.0886075949367089,
+      "grad_norm": 0.3817494213581085,
+      "learning_rate": 0.0001940719596278771,
+      "loss": 0.6870401501655579,
+      "step": 2580
+    },
+    {
+      "epoch": 1.0894514767932488,
+      "grad_norm": 0.4630168378353119,
+      "learning_rate": 0.00019405565796757043,
+      "loss": 0.7437341809272766,
+      "step": 2582
+    },
+    {
+      "epoch": 1.090295358649789,
+      "grad_norm": 0.4063820540904999,
+      "learning_rate": 0.00019403933461042982,
+      "loss": 0.7101680040359497,
+      "step": 2584
+    },
+    {
+      "epoch": 1.0911392405063292,
+      "grad_norm": 0.44423359632492065,
+      "learning_rate": 0.0001940229895602208,
+      "loss": 0.6894780397415161,
+      "step": 2586
+    },
+    {
+      "epoch": 1.091983122362869,
+      "grad_norm": 0.4267353117465973,
+      "learning_rate": 0.0001940066228207138,
+      "loss": 0.7277120351791382,
+      "step": 2588
+    },
+    {
+      "epoch": 1.0928270042194093,
+      "grad_norm": 0.44747647643089294,
+      "learning_rate": 0.00019399023439568434,
+      "loss": 0.6512353420257568,
+      "step": 2590
+    },
+    {
+      "epoch": 1.0936708860759494,
+      "grad_norm": 0.4557100832462311,
+      "learning_rate": 0.00019397382428891296,
+      "loss": 0.7850068807601929,
+      "step": 2592
+    },
+    {
+      "epoch": 1.0945147679324894,
+      "grad_norm": 0.44980114698410034,
+      "learning_rate": 0.0001939573925041851,
+      "loss": 0.786868691444397,
+      "step": 2594
+    },
+    {
+      "epoch": 1.0953586497890295,
+      "grad_norm": 0.45374399423599243,
+      "learning_rate": 0.00019394093904529126,
+      "loss": 0.8260899186134338,
+      "step": 2596
+    },
+    {
+      "epoch": 1.0962025316455697,
+      "grad_norm": 0.40681353211402893,
+      "learning_rate": 0.00019392446391602695,
+      "loss": 0.7056689262390137,
+      "step": 2598
+    },
+    {
+      "epoch": 1.0970464135021096,
+      "grad_norm": 0.39278075098991394,
+      "learning_rate": 0.0001939079671201927,
+      "loss": 0.7557252645492554,
+      "step": 2600
+    },
+    {
+      "epoch": 1.0970464135021096,
+      "eval_loss": 0.8211485743522644,
+      "eval_runtime": 670.2276,
+      "eval_samples_per_second": 3.144,
+      "eval_steps_per_second": 3.144,
+      "step": 2600
+    },
+    {
+      "epoch": 1.0978902953586498,
+      "grad_norm": 0.4261230230331421,
+      "learning_rate": 0.00019389144866159398,
+      "loss": 0.7139819264411926,
+      "step": 2602
+    },
+    {
+      "epoch": 1.09873417721519,
+      "grad_norm": 0.4664439558982849,
+      "learning_rate": 0.0001938749085440412,
+      "loss": 0.7274705767631531,
+      "step": 2604
+    },
+    {
+      "epoch": 1.09957805907173,
+      "grad_norm": 0.46306154131889343,
+      "learning_rate": 0.00019385834677134998,
+      "loss": 0.7547051906585693,
+      "step": 2606
+    },
+    {
+      "epoch": 1.10042194092827,
+      "grad_norm": 0.39753687381744385,
+      "learning_rate": 0.00019384176334734073,
+      "loss": 0.6942694187164307,
+      "step": 2608
+    },
+    {
+      "epoch": 1.1012658227848102,
+      "grad_norm": 0.4556522071361542,
+      "learning_rate": 0.00019382515827583898,
+      "loss": 0.7161701321601868,
+      "step": 2610
+    },
+    {
+      "epoch": 1.1021097046413502,
+      "grad_norm": 0.3863290250301361,
+      "learning_rate": 0.0001938085315606751,
+      "loss": 0.7165710926055908,
+      "step": 2612
+    },
+    {
+      "epoch": 1.1029535864978903,
+      "grad_norm": 0.49601292610168457,
+      "learning_rate": 0.00019379188320568466,
+      "loss": 0.8080679774284363,
+      "step": 2614
+    },
+    {
+      "epoch": 1.1037974683544305,
+      "grad_norm": 0.4498465955257416,
+      "learning_rate": 0.00019377521321470805,
+      "loss": 0.7032502889633179,
+      "step": 2616
+    },
+    {
+      "epoch": 1.1046413502109704,
+      "grad_norm": 0.4537963569164276,
+      "learning_rate": 0.00019375852159159073,
+      "loss": 0.7076731324195862,
+      "step": 2618
+    },
+    {
+      "epoch": 1.1054852320675106,
+      "grad_norm": 0.4042968153953552,
+      "learning_rate": 0.0001937418083401832,
+      "loss": 0.7020825743675232,
+      "step": 2620
+    },
+    {
+      "epoch": 1.1063291139240505,
+      "grad_norm": 0.44306516647338867,
+      "learning_rate": 0.00019372507346434076,
+      "loss": 0.7222874760627747,
+      "step": 2622
+    },
+    {
+      "epoch": 1.1071729957805907,
+      "grad_norm": 0.3792516887187958,
+      "learning_rate": 0.00019370831696792392,
+      "loss": 0.6136211156845093,
+      "step": 2624
+    },
+    {
+      "epoch": 1.1080168776371309,
+      "grad_norm": 0.46655353903770447,
+      "learning_rate": 0.00019369153885479806,
+      "loss": 0.7358977198600769,
+      "step": 2626
+    },
+    {
+      "epoch": 1.1088607594936708,
+      "grad_norm": 0.4416999816894531,
+      "learning_rate": 0.0001936747391288336,
+      "loss": 0.7484551072120667,
+      "step": 2628
+    },
+    {
+      "epoch": 1.109704641350211,
+      "grad_norm": 0.4620150327682495,
+      "learning_rate": 0.00019365791779390584,
+      "loss": 0.6874802708625793,
+      "step": 2630
+    },
+    {
+      "epoch": 1.1105485232067511,
+      "grad_norm": 0.4357883334159851,
+      "learning_rate": 0.00019364107485389518,
+      "loss": 0.7168865203857422,
+      "step": 2632
+    },
+    {
+      "epoch": 1.111392405063291,
+      "grad_norm": 0.4581143856048584,
+      "learning_rate": 0.00019362421031268699,
+      "loss": 0.7533737421035767,
+      "step": 2634
+    },
+    {
+      "epoch": 1.1122362869198312,
+      "grad_norm": 0.46408653259277344,
+      "learning_rate": 0.00019360732417417152,
+      "loss": 0.8245398998260498,
+      "step": 2636
+    },
+    {
+      "epoch": 1.1130801687763714,
+      "grad_norm": 0.4639282524585724,
+      "learning_rate": 0.00019359041644224415,
+      "loss": 0.7095663547515869,
+      "step": 2638
+    },
+    {
+      "epoch": 1.1139240506329113,
+      "grad_norm": 0.4598273038864136,
+      "learning_rate": 0.00019357348712080517,
+      "loss": 0.7030285596847534,
+      "step": 2640
+    },
+    {
+      "epoch": 1.1147679324894515,
+      "grad_norm": 0.42993804812431335,
+      "learning_rate": 0.00019355653621375978,
+      "loss": 0.7158969640731812,
+      "step": 2642
+    },
+    {
+      "epoch": 1.1156118143459917,
+      "grad_norm": 0.36132168769836426,
+      "learning_rate": 0.0001935395637250183,
+      "loss": 0.6800820231437683,
+      "step": 2644
+    },
+    {
+      "epoch": 1.1164556962025316,
+      "grad_norm": 0.5004428029060364,
+      "learning_rate": 0.00019352256965849592,
+      "loss": 0.694972813129425,
+      "step": 2646
+    },
+    {
+      "epoch": 1.1172995780590718,
+      "grad_norm": 0.43715354800224304,
+      "learning_rate": 0.00019350555401811286,
+      "loss": 0.7164262533187866,
+      "step": 2648
+    },
+    {
+      "epoch": 1.1181434599156117,
+      "grad_norm": 0.42316755652427673,
+      "learning_rate": 0.00019348851680779433,
+      "loss": 0.7026737928390503,
+      "step": 2650
+    },
+    {
+      "epoch": 1.1189873417721519,
+      "grad_norm": 0.35848063230514526,
+      "learning_rate": 0.0001934714580314704,
+      "loss": 0.6774232387542725,
+      "step": 2652
+    },
+    {
+      "epoch": 1.119831223628692,
+      "grad_norm": 0.46613892912864685,
+      "learning_rate": 0.00019345437769307627,
+      "loss": 0.7789244055747986,
+      "step": 2654
+    },
+    {
+      "epoch": 1.120675105485232,
+      "grad_norm": 0.4298146367073059,
+      "learning_rate": 0.00019343727579655204,
+      "loss": 0.8647177815437317,
+      "step": 2656
+    },
+    {
+      "epoch": 1.1215189873417721,
+      "grad_norm": 0.4638161361217499,
+      "learning_rate": 0.00019342015234584276,
+      "loss": 0.6949124336242676,
+      "step": 2658
+    },
+    {
+      "epoch": 1.1223628691983123,
+      "grad_norm": 0.4079605042934418,
+      "learning_rate": 0.00019340300734489854,
+      "loss": 0.6677125096321106,
+      "step": 2660
+    },
+    {
+      "epoch": 1.1232067510548522,
+      "grad_norm": 0.46079063415527344,
+      "learning_rate": 0.00019338584079767434,
+      "loss": 0.7521067261695862,
+      "step": 2662
+    },
+    {
+      "epoch": 1.1240506329113924,
+      "grad_norm": 0.3935202956199646,
+      "learning_rate": 0.0001933686527081302,
+      "loss": 0.6829184889793396,
+      "step": 2664
+    },
+    {
+      "epoch": 1.1248945147679326,
+      "grad_norm": 0.45696479082107544,
+      "learning_rate": 0.000193351443080231,
+      "loss": 0.7122842669487,
+      "step": 2666
+    },
+    {
+      "epoch": 1.1257383966244725,
+      "grad_norm": 0.4078282415866852,
+      "learning_rate": 0.0001933342119179468,
+      "loss": 0.7316977977752686,
+      "step": 2668
+    },
+    {
+      "epoch": 1.1265822784810127,
+      "grad_norm": 0.4282447099685669,
+      "learning_rate": 0.0001933169592252524,
+      "loss": 0.7323395013809204,
+      "step": 2670
+    },
+    {
+      "epoch": 1.1274261603375528,
+      "grad_norm": 0.46398648619651794,
+      "learning_rate": 0.00019329968500612766,
+      "loss": 0.7177348732948303,
+      "step": 2672
+    },
+    {
+      "epoch": 1.1282700421940928,
+      "grad_norm": 0.4560202658176422,
+      "learning_rate": 0.0001932823892645575,
+      "loss": 0.7957183122634888,
+      "step": 2674
+    },
+    {
+      "epoch": 1.129113924050633,
+      "grad_norm": 0.4790743887424469,
+      "learning_rate": 0.00019326507200453162,
+      "loss": 0.762060284614563,
+      "step": 2676
+    },
+    {
+      "epoch": 1.129957805907173,
+      "grad_norm": 0.46524202823638916,
+      "learning_rate": 0.00019324773323004484,
+      "loss": 0.7227895259857178,
+      "step": 2678
+    },
+    {
+      "epoch": 1.130801687763713,
+      "grad_norm": 0.47912222146987915,
+      "learning_rate": 0.00019323037294509683,
+      "loss": 0.6968042254447937,
+      "step": 2680
+    },
+    {
+      "epoch": 1.1316455696202532,
+      "grad_norm": 0.4412849247455597,
+      "learning_rate": 0.00019321299115369233,
+      "loss": 0.6682016253471375,
+      "step": 2682
+    },
+    {
+      "epoch": 1.1324894514767934,
+      "grad_norm": 0.4138370752334595,
+      "learning_rate": 0.00019319558785984095,
+      "loss": 0.7287812829017639,
+      "step": 2684
+    },
+    {
+      "epoch": 1.1333333333333333,
+      "grad_norm": 0.391170859336853,
+      "learning_rate": 0.00019317816306755733,
+      "loss": 0.6547526121139526,
+      "step": 2686
+    },
+    {
+      "epoch": 1.1341772151898735,
+      "grad_norm": 0.45472806692123413,
+      "learning_rate": 0.00019316071678086098,
+      "loss": 0.6942042708396912,
+      "step": 2688
+    },
+    {
+      "epoch": 1.1350210970464134,
+      "grad_norm": 0.40321481227874756,
+      "learning_rate": 0.00019314324900377648,
+      "loss": 0.7417841553688049,
+      "step": 2690
+    },
+    {
+      "epoch": 1.1358649789029536,
+      "grad_norm": 0.3806784152984619,
+      "learning_rate": 0.00019312575974033327,
+      "loss": 0.7780525088310242,
+      "step": 2692
+    },
+    {
+      "epoch": 1.1367088607594937,
+      "grad_norm": 0.43763765692710876,
+      "learning_rate": 0.0001931082489945658,
+      "loss": 0.661472499370575,
+      "step": 2694
+    },
+    {
+      "epoch": 1.1375527426160337,
+      "grad_norm": 0.45410826802253723,
+      "learning_rate": 0.00019309071677051348,
+      "loss": 0.7443127632141113,
+      "step": 2696
+    },
+    {
+      "epoch": 1.1383966244725738,
+      "grad_norm": 0.47907447814941406,
+      "learning_rate": 0.00019307316307222062,
+      "loss": 0.744393527507782,
+      "step": 2698
+    },
+    {
+      "epoch": 1.139240506329114,
+      "grad_norm": 0.41723814606666565,
+      "learning_rate": 0.00019305558790373654,
+      "loss": 0.7042794227600098,
+      "step": 2700
+    },
+    {
+      "epoch": 1.139240506329114,
+      "eval_loss": 0.8155058026313782,
+      "eval_runtime": 678.284,
+      "eval_samples_per_second": 3.106,
+      "eval_steps_per_second": 3.106,
+      "step": 2700
+    },
+    {
+      "epoch": 1.140084388185654,
+      "grad_norm": 0.3863268196582794,
+      "learning_rate": 0.0001930379912691155,
+      "loss": 0.6715745329856873,
+      "step": 2702
+    },
+    {
+      "epoch": 1.140928270042194,
+      "grad_norm": 0.49994897842407227,
+      "learning_rate": 0.00019302037317241673,
+      "loss": 0.6959030628204346,
+      "step": 2704
+    },
+    {
+      "epoch": 1.1417721518987343,
+      "grad_norm": 0.46377572417259216,
+      "learning_rate": 0.00019300273361770432,
+      "loss": 0.6534222364425659,
+      "step": 2706
+    },
+    {
+      "epoch": 1.1426160337552742,
+      "grad_norm": 0.3841579556465149,
+      "learning_rate": 0.00019298507260904745,
+      "loss": 0.6750475764274597,
+      "step": 2708
+    },
+    {
+      "epoch": 1.1434599156118144,
+      "grad_norm": 0.4171544313430786,
+      "learning_rate": 0.00019296739015052015,
+      "loss": 0.7714900374412537,
+      "step": 2710
+    },
+    {
+      "epoch": 1.1443037974683543,
+      "grad_norm": 0.44384199380874634,
+      "learning_rate": 0.00019294968624620137,
+      "loss": 0.7281306982040405,
+      "step": 2712
+    },
+    {
+      "epoch": 1.1451476793248945,
+      "grad_norm": 0.43345192074775696,
+      "learning_rate": 0.00019293196090017513,
+      "loss": 0.7407262325286865,
+      "step": 2714
+    },
+    {
+      "epoch": 1.1459915611814346,
+      "grad_norm": 0.3897020220756531,
+      "learning_rate": 0.00019291421411653034,
+      "loss": 0.6794347167015076,
+      "step": 2716
+    },
+    {
+      "epoch": 1.1468354430379746,
+      "grad_norm": 0.4289143681526184,
+      "learning_rate": 0.00019289644589936075,
+      "loss": 0.7026289105415344,
+      "step": 2718
+    },
+    {
+      "epoch": 1.1476793248945147,
+      "grad_norm": 0.4783557057380676,
+      "learning_rate": 0.0001928786562527652,
+      "loss": 0.7613762617111206,
+      "step": 2720
+    },
+    {
+      "epoch": 1.148523206751055,
+      "grad_norm": 0.4152112901210785,
+      "learning_rate": 0.00019286084518084745,
+      "loss": 0.7123059630393982,
+      "step": 2722
+    },
+    {
+      "epoch": 1.1493670886075948,
+      "grad_norm": 0.4204908013343811,
+      "learning_rate": 0.0001928430126877161,
+      "loss": 0.7198041677474976,
+      "step": 2724
+    },
+    {
+      "epoch": 1.150210970464135,
+      "grad_norm": 0.41333863139152527,
+      "learning_rate": 0.00019282515877748485,
+      "loss": 0.6859839558601379,
+      "step": 2726
+    },
+    {
+      "epoch": 1.1510548523206752,
+      "grad_norm": 0.4041498005390167,
+      "learning_rate": 0.00019280728345427218,
+      "loss": 0.7185758948326111,
+      "step": 2728
+    },
+    {
+      "epoch": 1.1518987341772151,
+      "grad_norm": 0.4487358629703522,
+      "learning_rate": 0.00019278938672220167,
+      "loss": 0.7899971008300781,
+      "step": 2730
+    },
+    {
+      "epoch": 1.1527426160337553,
+      "grad_norm": 0.3510114252567291,
+      "learning_rate": 0.00019277146858540166,
+      "loss": 0.6266586780548096,
+      "step": 2732
+    },
+    {
+      "epoch": 1.1535864978902954,
+      "grad_norm": 0.4295555055141449,
+      "learning_rate": 0.00019275352904800553,
+      "loss": 0.7450110912322998,
+      "step": 2734
+    },
+    {
+      "epoch": 1.1544303797468354,
+      "grad_norm": 0.40629687905311584,
+      "learning_rate": 0.00019273556811415165,
+      "loss": 0.6875549554824829,
+      "step": 2736
+    },
+    {
+      "epoch": 1.1552742616033755,
+      "grad_norm": 0.4243362247943878,
+      "learning_rate": 0.00019271758578798322,
+      "loss": 0.8141522407531738,
+      "step": 2738
+    },
+    {
+      "epoch": 1.1561181434599157,
+      "grad_norm": 0.483306884765625,
+      "learning_rate": 0.00019269958207364843,
+      "loss": 0.6905186176300049,
+      "step": 2740
+    },
+    {
+      "epoch": 1.1569620253164556,
+      "grad_norm": 0.4303925335407257,
+      "learning_rate": 0.00019268155697530037,
+      "loss": 0.6709047555923462,
+      "step": 2742
+    },
+    {
+      "epoch": 1.1578059071729958,
+      "grad_norm": 0.4533381760120392,
+      "learning_rate": 0.0001926635104970971,
+      "loss": 0.7107465863227844,
+      "step": 2744
+    },
+    {
+      "epoch": 1.158649789029536,
+      "grad_norm": 0.4823286533355713,
+      "learning_rate": 0.0001926454426432016,
+      "loss": 0.8043986558914185,
+      "step": 2746
+    },
+    {
+      "epoch": 1.159493670886076,
+      "grad_norm": 0.3892867863178253,
+      "learning_rate": 0.00019262735341778179,
+      "loss": 0.7812165021896362,
+      "step": 2748
+    },
+    {
+      "epoch": 1.160337552742616,
+      "grad_norm": 0.46737003326416016,
+      "learning_rate": 0.00019260924282501045,
+      "loss": 0.7141886353492737,
+      "step": 2750
+    },
+    {
+      "epoch": 1.1611814345991562,
+      "grad_norm": 0.4335328936576843,
+      "learning_rate": 0.0001925911108690654,
+      "loss": 0.7528460621833801,
+      "step": 2752
+    },
+    {
+      "epoch": 1.1620253164556962,
+      "grad_norm": 0.3792484700679779,
+      "learning_rate": 0.00019257295755412932,
+      "loss": 0.7817611694335938,
+      "step": 2754
+    },
+    {
+      "epoch": 1.1628691983122363,
+      "grad_norm": 0.41246429085731506,
+      "learning_rate": 0.00019255478288438983,
+      "loss": 0.6904958486557007,
+      "step": 2756
+    },
+    {
+      "epoch": 1.1637130801687763,
+      "grad_norm": 0.44368481636047363,
+      "learning_rate": 0.00019253658686403948,
+      "loss": 0.7571630477905273,
+      "step": 2758
+    },
+    {
+      "epoch": 1.1645569620253164,
+      "grad_norm": 0.40121781826019287,
+      "learning_rate": 0.00019251836949727574,
+      "loss": 0.6747061014175415,
+      "step": 2760
+    },
+    {
+      "epoch": 1.1654008438818566,
+      "grad_norm": 0.4660188853740692,
+      "learning_rate": 0.000192500130788301,
+      "loss": 0.7722339034080505,
+      "step": 2762
+    },
+    {
+      "epoch": 1.1662447257383965,
+      "grad_norm": 0.41693782806396484,
+      "learning_rate": 0.0001924818707413226,
+      "loss": 0.8081457018852234,
+      "step": 2764
+    },
+    {
+      "epoch": 1.1670886075949367,
+      "grad_norm": 0.4274570345878601,
+      "learning_rate": 0.00019246358936055274,
+      "loss": 0.7745853066444397,
+      "step": 2766
+    },
+    {
+      "epoch": 1.1679324894514769,
+      "grad_norm": 0.4385969638824463,
+      "learning_rate": 0.00019244528665020863,
+      "loss": 0.7918146252632141,
+      "step": 2768
+    },
+    {
+      "epoch": 1.1687763713080168,
+      "grad_norm": 0.42832866311073303,
+      "learning_rate": 0.00019242696261451234,
+      "loss": 0.7382830381393433,
+      "step": 2770
+    },
+    {
+      "epoch": 1.169620253164557,
+      "grad_norm": 0.44177982211112976,
+      "learning_rate": 0.00019240861725769085,
+      "loss": 0.7501721978187561,
+      "step": 2772
+    },
+    {
+      "epoch": 1.1704641350210971,
+      "grad_norm": 0.4193457365036011,
+      "learning_rate": 0.00019239025058397611,
+      "loss": 0.6495388150215149,
+      "step": 2774
+    },
+    {
+      "epoch": 1.171308016877637,
+      "grad_norm": 0.5145925283432007,
+      "learning_rate": 0.000192371862597605,
+      "loss": 0.8291991353034973,
+      "step": 2776
+    },
+    {
+      "epoch": 1.1721518987341772,
+      "grad_norm": 0.40911251306533813,
+      "learning_rate": 0.00019235345330281914,
+      "loss": 0.7491340637207031,
+      "step": 2778
+    },
+    {
+      "epoch": 1.1729957805907172,
+      "grad_norm": 0.3920276165008545,
+      "learning_rate": 0.0001923350227038653,
+      "loss": 0.6315004825592041,
+      "step": 2780
+    },
+    {
+      "epoch": 1.1738396624472573,
+      "grad_norm": 0.442303329706192,
+      "learning_rate": 0.0001923165708049951,
+      "loss": 0.7538990378379822,
+      "step": 2782
+    },
+    {
+      "epoch": 1.1746835443037975,
+      "grad_norm": 0.4693523049354553,
+      "learning_rate": 0.00019229809761046497,
+      "loss": 0.7440794110298157,
+      "step": 2784
+    },
+    {
+      "epoch": 1.1755274261603375,
+      "grad_norm": 0.4948059618473053,
+      "learning_rate": 0.0001922796031245363,
+      "loss": 0.7342568635940552,
+      "step": 2786
+    },
+    {
+      "epoch": 1.1763713080168776,
+      "grad_norm": 0.3900465965270996,
+      "learning_rate": 0.00019226108735147547,
+      "loss": 0.6969661116600037,
+      "step": 2788
+    },
+    {
+      "epoch": 1.1772151898734178,
+      "grad_norm": 0.36153602600097656,
+      "learning_rate": 0.00019224255029555372,
+      "loss": 0.7097541689872742,
+      "step": 2790
+    },
+    {
+      "epoch": 1.1780590717299577,
+      "grad_norm": 0.41168832778930664,
+      "learning_rate": 0.00019222399196104715,
+      "loss": 0.8141831755638123,
+      "step": 2792
+    },
+    {
+      "epoch": 1.1789029535864979,
+      "grad_norm": 0.4542821943759918,
+      "learning_rate": 0.0001922054123522368,
+      "loss": 0.7458863258361816,
+      "step": 2794
+    },
+    {
+      "epoch": 1.179746835443038,
+      "grad_norm": 0.4026047885417938,
+      "learning_rate": 0.0001921868114734087,
+      "loss": 0.6535642147064209,
+      "step": 2796
+    },
+    {
+      "epoch": 1.180590717299578,
+      "grad_norm": 0.4417034387588501,
+      "learning_rate": 0.0001921681893288537,
+      "loss": 0.6248135566711426,
+      "step": 2798
+    },
+    {
+      "epoch": 1.1814345991561181,
+      "grad_norm": 0.45027074217796326,
+      "learning_rate": 0.00019214954592286748,
+      "loss": 0.6777785420417786,
+      "step": 2800
+    },
+    {
+      "epoch": 1.1814345991561181,
+      "eval_loss": 0.8124309182167053,
+      "eval_runtime": 688.4759,
+      "eval_samples_per_second": 3.06,
+      "eval_steps_per_second": 3.06,
+      "step": 2800
+    },
+    {
+      "epoch": 1.1822784810126583,
+      "grad_norm": 0.406865656375885,
+      "learning_rate": 0.00019213088125975082,
+      "loss": 0.6460301876068115,
+      "step": 2802
+    },
+    {
+      "epoch": 1.1831223628691983,
+      "grad_norm": 0.5052749514579773,
+      "learning_rate": 0.00019211219534380928,
+      "loss": 0.7292830944061279,
+      "step": 2804
+    },
+    {
+      "epoch": 1.1839662447257384,
+      "grad_norm": 0.4028380513191223,
+      "learning_rate": 0.00019209348817935328,
+      "loss": 0.7399409413337708,
+      "step": 2806
+    },
+    {
+      "epoch": 1.1848101265822786,
+      "grad_norm": 0.4522664248943329,
+      "learning_rate": 0.00019207475977069826,
+      "loss": 0.6885398030281067,
+      "step": 2808
+    },
+    {
+      "epoch": 1.1856540084388185,
+      "grad_norm": 0.419546514749527,
+      "learning_rate": 0.0001920560101221645,
+      "loss": 0.6962382197380066,
+      "step": 2810
+    },
+    {
+      "epoch": 1.1864978902953587,
+      "grad_norm": 0.48912352323532104,
+      "learning_rate": 0.00019203723923807715,
+      "loss": 0.7380965352058411,
+      "step": 2812
+    },
+    {
+      "epoch": 1.1873417721518988,
+      "grad_norm": 0.40905487537384033,
+      "learning_rate": 0.00019201844712276635,
+      "loss": 0.755382776260376,
+      "step": 2814
+    },
+    {
+      "epoch": 1.1881856540084388,
+      "grad_norm": 0.385623037815094,
+      "learning_rate": 0.000191999633780567,
+      "loss": 0.7066168189048767,
+      "step": 2816
+    },
+    {
+      "epoch": 1.189029535864979,
+      "grad_norm": 0.4488723576068878,
+      "learning_rate": 0.00019198079921581908,
+      "loss": 0.7172947525978088,
+      "step": 2818
+    },
+    {
+      "epoch": 1.189873417721519,
+      "grad_norm": 0.385277658700943,
+      "learning_rate": 0.00019196194343286727,
+      "loss": 0.6714195013046265,
+      "step": 2820
+    },
+    {
+      "epoch": 1.190717299578059,
+      "grad_norm": 0.4280855655670166,
+      "learning_rate": 0.00019194306643606127,
+      "loss": 0.7076966166496277,
+      "step": 2822
+    },
+    {
+      "epoch": 1.1915611814345992,
+      "grad_norm": 0.4626067578792572,
+      "learning_rate": 0.00019192416822975567,
+      "loss": 0.7926747798919678,
+      "step": 2824
+    },
+    {
+      "epoch": 1.1924050632911392,
+      "grad_norm": 0.3994208574295044,
+      "learning_rate": 0.00019190524881830987,
+      "loss": 0.7654290199279785,
+      "step": 2826
+    },
+    {
+      "epoch": 1.1932489451476793,
+      "grad_norm": 0.47346508502960205,
+      "learning_rate": 0.00019188630820608825,
+      "loss": 0.7857998609542847,
+      "step": 2828
+    },
+    {
+      "epoch": 1.1940928270042195,
+      "grad_norm": 0.446077823638916,
+      "learning_rate": 0.00019186734639746003,
+      "loss": 0.7603738903999329,
+      "step": 2830
+    },
+    {
+      "epoch": 1.1949367088607594,
+      "grad_norm": 0.41503727436065674,
+      "learning_rate": 0.00019184836339679936,
+      "loss": 0.790582537651062,
+      "step": 2832
+    },
+    {
+      "epoch": 1.1957805907172996,
+      "grad_norm": 0.43651700019836426,
+      "learning_rate": 0.00019182935920848522,
+      "loss": 0.7229864597320557,
+      "step": 2834
+    },
+    {
+      "epoch": 1.1966244725738397,
+      "grad_norm": 0.4409031867980957,
+      "learning_rate": 0.00019181033383690153,
+      "loss": 0.7817287445068359,
+      "step": 2836
+    },
+    {
+      "epoch": 1.1974683544303797,
+      "grad_norm": 0.3562015891075134,
+      "learning_rate": 0.0001917912872864371,
+      "loss": 0.6342297792434692,
+      "step": 2838
+    },
+    {
+      "epoch": 1.1983122362869199,
+      "grad_norm": 0.4767192602157593,
+      "learning_rate": 0.00019177221956148555,
+      "loss": 0.7065272927284241,
+      "step": 2840
+    },
+    {
+      "epoch": 1.1991561181434598,
+      "grad_norm": 0.45131945610046387,
+      "learning_rate": 0.00019175313066644546,
+      "loss": 0.777110755443573,
+      "step": 2842
+    },
+    {
+      "epoch": 1.2,
+      "grad_norm": 0.47566652297973633,
+      "learning_rate": 0.00019173402060572027,
+      "loss": 0.7456457018852234,
+      "step": 2844
+    },
+    {
+      "epoch": 1.2008438818565401,
+      "grad_norm": 0.48909857869148254,
+      "learning_rate": 0.00019171488938371834,
+      "loss": 0.8041547536849976,
+      "step": 2846
+    },
+    {
+      "epoch": 1.20168776371308,
+      "grad_norm": 0.436591774225235,
+      "learning_rate": 0.00019169573700485284,
+      "loss": 0.6786676049232483,
+      "step": 2848
+    },
+    {
+      "epoch": 1.2025316455696202,
+      "grad_norm": 0.44230160117149353,
+      "learning_rate": 0.00019167656347354183,
+      "loss": 0.7661880850791931,
+      "step": 2850
+    },
+    {
+      "epoch": 1.2033755274261604,
+      "grad_norm": 0.4844048321247101,
+      "learning_rate": 0.00019165736879420836,
+      "loss": 0.7369645833969116,
+      "step": 2852
+    },
+    {
+      "epoch": 1.2042194092827003,
+      "grad_norm": 0.4272955060005188,
+      "learning_rate": 0.0001916381529712802,
+      "loss": 0.7776681184768677,
+      "step": 2854
+    },
+    {
+      "epoch": 1.2050632911392405,
+      "grad_norm": 0.42392730712890625,
+      "learning_rate": 0.00019161891600919008,
+      "loss": 0.8019667267799377,
+      "step": 2856
+    },
+    {
+      "epoch": 1.2059071729957807,
+      "grad_norm": 0.4528326988220215,
+      "learning_rate": 0.00019159965791237565,
+      "loss": 0.7811952233314514,
+      "step": 2858
+    },
+    {
+      "epoch": 1.2067510548523206,
+      "grad_norm": 0.42840948700904846,
+      "learning_rate": 0.00019158037868527932,
+      "loss": 0.7244279980659485,
+      "step": 2860
+    },
+    {
+      "epoch": 1.2075949367088608,
+      "grad_norm": 0.4291226863861084,
+      "learning_rate": 0.00019156107833234845,
+      "loss": 0.7675760984420776,
+      "step": 2862
+    },
+    {
+      "epoch": 1.208438818565401,
+      "grad_norm": 0.4455093443393707,
+      "learning_rate": 0.00019154175685803527,
+      "loss": 0.7071483135223389,
+      "step": 2864
+    },
+    {
+      "epoch": 1.2092827004219409,
+      "grad_norm": 0.41356268525123596,
+      "learning_rate": 0.00019152241426679688,
+      "loss": 0.7554603815078735,
+      "step": 2866
+    },
+    {
+      "epoch": 1.210126582278481,
+      "grad_norm": 0.43875646591186523,
+      "learning_rate": 0.00019150305056309525,
+      "loss": 0.7198356986045837,
+      "step": 2868
+    },
+    {
+      "epoch": 1.2109704641350212,
+      "grad_norm": 0.454356849193573,
+      "learning_rate": 0.0001914836657513972,
+      "loss": 0.7393195629119873,
+      "step": 2870
+    },
+    {
+      "epoch": 1.2118143459915611,
+      "grad_norm": 0.37668535113334656,
+      "learning_rate": 0.00019146425983617444,
+      "loss": 0.7277671694755554,
+      "step": 2872
+    },
+    {
+      "epoch": 1.2126582278481013,
+      "grad_norm": 0.37869662046432495,
+      "learning_rate": 0.00019144483282190354,
+      "loss": 0.6783158183097839,
+      "step": 2874
+    },
+    {
+      "epoch": 1.2135021097046415,
+      "grad_norm": 0.4333525002002716,
+      "learning_rate": 0.00019142538471306596,
+      "loss": 0.7024772763252258,
+      "step": 2876
+    },
+    {
+      "epoch": 1.2143459915611814,
+      "grad_norm": 0.4693842828273773,
+      "learning_rate": 0.00019140591551414796,
+      "loss": 0.7387984395027161,
+      "step": 2878
+    },
+    {
+      "epoch": 1.2151898734177216,
+      "grad_norm": 0.432640939950943,
+      "learning_rate": 0.00019138642522964074,
+      "loss": 0.7597174644470215,
+      "step": 2880
+    },
+    {
+      "epoch": 1.2160337552742617,
+      "grad_norm": 0.432477205991745,
+      "learning_rate": 0.00019136691386404032,
+      "loss": 0.7884587049484253,
+      "step": 2882
+    },
+    {
+      "epoch": 1.2168776371308017,
+      "grad_norm": 0.39305904507637024,
+      "learning_rate": 0.00019134738142184763,
+      "loss": 0.6873269081115723,
+      "step": 2884
+    },
+    {
+      "epoch": 1.2177215189873418,
+      "grad_norm": 0.4485352635383606,
+      "learning_rate": 0.00019132782790756838,
+      "loss": 0.6523491144180298,
+      "step": 2886
+    },
+    {
+      "epoch": 1.2185654008438818,
+      "grad_norm": 0.4480868875980377,
+      "learning_rate": 0.00019130825332571322,
+      "loss": 0.7383941411972046,
+      "step": 2888
+    },
+    {
+      "epoch": 1.219409282700422,
+      "grad_norm": 0.44085049629211426,
+      "learning_rate": 0.00019128865768079764,
+      "loss": 0.7391335964202881,
+      "step": 2890
+    },
+    {
+      "epoch": 1.220253164556962,
+      "grad_norm": 0.4621105492115021,
+      "learning_rate": 0.00019126904097734198,
+      "loss": 0.7282017469406128,
+      "step": 2892
+    },
+    {
+      "epoch": 1.221097046413502,
+      "grad_norm": 0.41478198766708374,
+      "learning_rate": 0.00019124940321987136,
+      "loss": 0.7554038763046265,
+      "step": 2894
+    },
+    {
+      "epoch": 1.2219409282700422,
+      "grad_norm": 0.4143751263618469,
+      "learning_rate": 0.00019122974441291595,
+      "loss": 0.6826799511909485,
+      "step": 2896
+    },
+    {
+      "epoch": 1.2227848101265824,
+      "grad_norm": 0.45392894744873047,
+      "learning_rate": 0.00019121006456101057,
+      "loss": 0.751000702381134,
+      "step": 2898
+    },
+    {
+      "epoch": 1.2236286919831223,
+      "grad_norm": 0.46946296095848083,
+      "learning_rate": 0.00019119036366869506,
+      "loss": 0.7598516941070557,
+      "step": 2900
+    },
+    {
+      "epoch": 1.2236286919831223,
+      "eval_loss": 0.8077136278152466,
+      "eval_runtime": 685.6042,
+      "eval_samples_per_second": 3.073,
+      "eval_steps_per_second": 3.073,
+      "step": 2900
+    },
+    {
+      "epoch": 1.2244725738396625,
+      "grad_norm": 0.4316656291484833,
+      "learning_rate": 0.00019117064174051395,
+      "loss": 0.7518358826637268,
+      "step": 2902
+    },
+    {
+      "epoch": 1.2253164556962026,
+      "grad_norm": 0.47737377882003784,
+      "learning_rate": 0.0001911508987810168,
+      "loss": 0.7308215498924255,
+      "step": 2904
+    },
+    {
+      "epoch": 1.2261603375527426,
+      "grad_norm": 0.40906062722206116,
+      "learning_rate": 0.00019113113479475783,
+      "loss": 0.6996226906776428,
+      "step": 2906
+    },
+    {
+      "epoch": 1.2270042194092827,
+      "grad_norm": 0.517085075378418,
+      "learning_rate": 0.00019111134978629633,
+      "loss": 0.8364415168762207,
+      "step": 2908
+    },
+    {
+      "epoch": 1.2278481012658227,
+      "grad_norm": 0.425581693649292,
+      "learning_rate": 0.00019109154376019624,
+      "loss": 0.722011923789978,
+      "step": 2910
+    },
+    {
+      "epoch": 1.2286919831223628,
+      "grad_norm": 0.4651593565940857,
+      "learning_rate": 0.00019107171672102643,
+      "loss": 0.7586172223091125,
+      "step": 2912
+    },
+    {
+      "epoch": 1.229535864978903,
+      "grad_norm": 0.4578891396522522,
+      "learning_rate": 0.00019105186867336067,
+      "loss": 0.7387955188751221,
+      "step": 2914
+    },
+    {
+      "epoch": 1.230379746835443,
+      "grad_norm": 0.4312961995601654,
+      "learning_rate": 0.00019103199962177748,
+      "loss": 0.8036285638809204,
+      "step": 2916
+    },
+    {
+      "epoch": 1.231223628691983,
+      "grad_norm": 0.4355425536632538,
+      "learning_rate": 0.00019101210957086026,
+      "loss": 0.7356550097465515,
+      "step": 2918
+    },
+    {
+      "epoch": 1.2320675105485233,
+      "grad_norm": 0.424675852060318,
+      "learning_rate": 0.00019099219852519733,
+      "loss": 0.6730139255523682,
+      "step": 2920
+    },
+    {
+      "epoch": 1.2329113924050632,
+      "grad_norm": 0.41102367639541626,
+      "learning_rate": 0.0001909722664893817,
+      "loss": 0.6983892321586609,
+      "step": 2922
+    },
+    {
+      "epoch": 1.2337552742616034,
+      "grad_norm": 0.4088250398635864,
+      "learning_rate": 0.00019095231346801132,
+      "loss": 0.6297144889831543,
+      "step": 2924
+    },
+    {
+      "epoch": 1.2345991561181435,
+      "grad_norm": 0.3685890734195709,
+      "learning_rate": 0.00019093233946568906,
+      "loss": 0.7035690546035767,
+      "step": 2926
+    },
+    {
+      "epoch": 1.2354430379746835,
+      "grad_norm": 0.39438629150390625,
+      "learning_rate": 0.0001909123444870224,
+      "loss": 0.820853054523468,
+      "step": 2928
+    },
+    {
+      "epoch": 1.2362869198312236,
+      "grad_norm": 0.5041676163673401,
+      "learning_rate": 0.00019089232853662392,
+      "loss": 0.7884082794189453,
+      "step": 2930
+    },
+    {
+      "epoch": 1.2371308016877638,
+      "grad_norm": 0.40592801570892334,
+      "learning_rate": 0.00019087229161911086,
+      "loss": 0.773978590965271,
+      "step": 2932
+    },
+    {
+      "epoch": 1.2379746835443037,
+      "grad_norm": 0.41828030347824097,
+      "learning_rate": 0.00019085223373910536,
+      "loss": 0.6742141842842102,
+      "step": 2934
+    },
+    {
+      "epoch": 1.238818565400844,
+      "grad_norm": 0.39984026551246643,
+      "learning_rate": 0.0001908321549012344,
+      "loss": 0.7106125950813293,
+      "step": 2936
+    },
+    {
+      "epoch": 1.239662447257384,
+      "grad_norm": 0.4841296970844269,
+      "learning_rate": 0.00019081205511012973,
+      "loss": 0.7390915751457214,
+      "step": 2938
+    },
+    {
+      "epoch": 1.240506329113924,
+      "grad_norm": 0.39837637543678284,
+      "learning_rate": 0.00019079193437042806,
+      "loss": 0.6527198553085327,
+      "step": 2940
+    },
+    {
+      "epoch": 1.2413502109704642,
+      "grad_norm": 0.40916863083839417,
+      "learning_rate": 0.0001907717926867708,
+      "loss": 0.6939712762832642,
+      "step": 2942
+    },
+    {
+      "epoch": 1.2421940928270043,
+      "grad_norm": 0.4329306483268738,
+      "learning_rate": 0.0001907516300638043,
+      "loss": 0.7566131353378296,
+      "step": 2944
+    },
+    {
+      "epoch": 1.2430379746835443,
+      "grad_norm": 0.47983601689338684,
+      "learning_rate": 0.00019073144650617963,
+      "loss": 0.7893748879432678,
+      "step": 2946
+    },
+    {
+      "epoch": 1.2438818565400844,
+      "grad_norm": 0.4368246793746948,
+      "learning_rate": 0.0001907112420185528,
+      "loss": 0.8259525299072266,
+      "step": 2948
+    },
+    {
+      "epoch": 1.2447257383966246,
+      "grad_norm": 0.41272372007369995,
+      "learning_rate": 0.0001906910166055846,
+      "loss": 0.734692394733429,
+      "step": 2950
+    },
+    {
+      "epoch": 1.2455696202531645,
+      "grad_norm": 0.36929839849472046,
+      "learning_rate": 0.00019067077027194057,
+      "loss": 0.7091750502586365,
+      "step": 2952
+    },
+    {
+      "epoch": 1.2464135021097047,
+      "grad_norm": 0.4395252466201782,
+      "learning_rate": 0.00019065050302229125,
+      "loss": 0.6960210204124451,
+      "step": 2954
+    },
+    {
+      "epoch": 1.2472573839662446,
+      "grad_norm": 0.46784868836402893,
+      "learning_rate": 0.00019063021486131186,
+      "loss": 0.7817115187644958,
+      "step": 2956
+    },
+    {
+      "epoch": 1.2481012658227848,
+      "grad_norm": 0.4191588759422302,
+      "learning_rate": 0.00019060990579368244,
+      "loss": 0.7933676242828369,
+      "step": 2958
+    },
+    {
+      "epoch": 1.248945147679325,
+      "grad_norm": 0.3967524468898773,
+      "learning_rate": 0.000190589575824088,
+      "loss": 0.6856688857078552,
+      "step": 2960
+    },
+    {
+      "epoch": 1.249789029535865,
+      "grad_norm": 0.40346091985702515,
+      "learning_rate": 0.0001905692249572182,
+      "loss": 0.6783818602561951,
+      "step": 2962
+    },
+    {
+      "epoch": 1.250632911392405,
+      "grad_norm": 0.38349980115890503,
+      "learning_rate": 0.0001905488531977676,
+      "loss": 0.6997019052505493,
+      "step": 2964
+    },
+    {
+      "epoch": 1.2514767932489452,
+      "grad_norm": 0.44021397829055786,
+      "learning_rate": 0.0001905284605504356,
+      "loss": 0.6978800892829895,
+      "step": 2966
+    },
+    {
+      "epoch": 1.2523206751054852,
+      "grad_norm": 0.47491583228111267,
+      "learning_rate": 0.0001905080470199264,
+      "loss": 0.7007693648338318,
+      "step": 2968
+    },
+    {
+      "epoch": 1.2531645569620253,
+      "grad_norm": 0.4011451005935669,
+      "learning_rate": 0.00019048761261094898,
+      "loss": 0.7258896231651306,
+      "step": 2970
+    },
+    {
+      "epoch": 1.2540084388185653,
+      "grad_norm": 0.44206640124320984,
+      "learning_rate": 0.0001904671573282172,
+      "loss": 0.6746299266815186,
+      "step": 2972
+    },
+    {
+      "epoch": 1.2548523206751054,
+      "grad_norm": 0.4532456696033478,
+      "learning_rate": 0.00019044668117644965,
+      "loss": 0.7638527154922485,
+      "step": 2974
+    },
+    {
+      "epoch": 1.2556962025316456,
+      "grad_norm": 0.44301095604896545,
+      "learning_rate": 0.00019042618416036984,
+      "loss": 0.659598171710968,
+      "step": 2976
+    },
+    {
+      "epoch": 1.2565400843881855,
+      "grad_norm": 0.37427952885627747,
+      "learning_rate": 0.00019040566628470597,
+      "loss": 0.7132709622383118,
+      "step": 2978
+    },
+    {
+      "epoch": 1.2573839662447257,
+      "grad_norm": 0.4195408821105957,
+      "learning_rate": 0.0001903851275541912,
+      "loss": 0.8163022398948669,
+      "step": 2980
+    },
+    {
+      "epoch": 1.2582278481012659,
+      "grad_norm": 0.3951880633831024,
+      "learning_rate": 0.00019036456797356336,
+      "loss": 0.7481505870819092,
+      "step": 2982
+    },
+    {
+      "epoch": 1.2590717299578058,
+      "grad_norm": 0.4239100217819214,
+      "learning_rate": 0.00019034398754756522,
+      "loss": 0.750126838684082,
+      "step": 2984
+    },
+    {
+      "epoch": 1.259915611814346,
+      "grad_norm": 0.4209354817867279,
+      "learning_rate": 0.0001903233862809442,
+      "loss": 0.746433436870575,
+      "step": 2986
+    },
+    {
+      "epoch": 1.2607594936708861,
+      "grad_norm": 0.4188138544559479,
+      "learning_rate": 0.00019030276417845266,
+      "loss": 0.7104454040527344,
+      "step": 2988
+    },
+    {
+      "epoch": 1.261603375527426,
+      "grad_norm": 0.4531371593475342,
+      "learning_rate": 0.00019028212124484772,
+      "loss": 0.650738000869751,
+      "step": 2990
+    },
+    {
+      "epoch": 1.2624472573839662,
+      "grad_norm": 0.4672204256057739,
+      "learning_rate": 0.00019026145748489133,
+      "loss": 0.6870551109313965,
+      "step": 2992
+    },
+    {
+      "epoch": 1.2632911392405064,
+      "grad_norm": 0.40681278705596924,
+      "learning_rate": 0.00019024077290335018,
+      "loss": 0.6879403591156006,
+      "step": 2994
+    },
+    {
+      "epoch": 1.2641350210970463,
+      "grad_norm": 0.4491691291332245,
+      "learning_rate": 0.00019022006750499585,
+      "loss": 0.6886372566223145,
+      "step": 2996
+    },
+    {
+      "epoch": 1.2649789029535865,
+      "grad_norm": 0.4144807457923889,
+      "learning_rate": 0.0001901993412946046,
+      "loss": 0.7199324369430542,
+      "step": 2998
+    },
+    {
+      "epoch": 1.2658227848101267,
+      "grad_norm": 0.40370482206344604,
+      "learning_rate": 0.0001901785942769577,
+      "loss": 0.7105490565299988,
+      "step": 3000
+    },
+    {
+      "epoch": 1.2658227848101267,
+      "eval_loss": 0.8033165335655212,
+      "eval_runtime": 678.7554,
+      "eval_samples_per_second": 3.104,
+      "eval_steps_per_second": 3.104,
+      "step": 3000
+    },
+    {
+      "epoch": 1.2666666666666666,
+      "grad_norm": 0.42637890577316284,
+      "learning_rate": 0.00019015782645684097,
+      "loss": 0.7385403513908386,
+      "step": 3002
+    },
+    {
+      "epoch": 1.2675105485232068,
+      "grad_norm": 0.3709809482097626,
+      "learning_rate": 0.00019013703783904518,
+      "loss": 0.6395129561424255,
+      "step": 3004
+    },
+    {
+      "epoch": 1.268354430379747,
+      "grad_norm": 0.4642501175403595,
+      "learning_rate": 0.00019011622842836592,
+      "loss": 0.7123657464981079,
+      "step": 3006
+    },
+    {
+      "epoch": 1.2691983122362869,
+      "grad_norm": 0.4553911089897156,
+      "learning_rate": 0.00019009539822960342,
+      "loss": 0.744048535823822,
+      "step": 3008
+    },
+    {
+      "epoch": 1.270042194092827,
+      "grad_norm": 0.39334386587142944,
+      "learning_rate": 0.0001900745472475629,
+      "loss": 0.7181914448738098,
+      "step": 3010
+    },
+    {
+      "epoch": 1.2708860759493672,
+      "grad_norm": 0.40902766585350037,
+      "learning_rate": 0.00019005367548705426,
+      "loss": 0.7238107919692993,
+      "step": 3012
+    },
+    {
+      "epoch": 1.2717299578059071,
+      "grad_norm": 0.48296019434928894,
+      "learning_rate": 0.00019003278295289216,
+      "loss": 0.7479744553565979,
+      "step": 3014
+    },
+    {
+      "epoch": 1.2725738396624473,
+      "grad_norm": 0.4807029962539673,
+      "learning_rate": 0.00019001186964989617,
+      "loss": 0.829924464225769,
+      "step": 3016
+    },
+    {
+      "epoch": 1.2734177215189875,
+      "grad_norm": 0.4239083528518677,
+      "learning_rate": 0.00018999093558289055,
+      "loss": 0.7204707264900208,
+      "step": 3018
+    },
+    {
+      "epoch": 1.2742616033755274,
+      "grad_norm": 0.4687928259372711,
+      "learning_rate": 0.0001899699807567044,
+      "loss": 0.7668164968490601,
+      "step": 3020
+    },
+    {
+      "epoch": 1.2751054852320676,
+      "grad_norm": 0.41724732518196106,
+      "learning_rate": 0.0001899490051761716,
+      "loss": 0.766342043876648,
+      "step": 3022
+    },
+    {
+      "epoch": 1.2759493670886077,
+      "grad_norm": 0.4131613075733185,
+      "learning_rate": 0.00018992800884613082,
+      "loss": 0.6595507860183716,
+      "step": 3024
+    },
+    {
+      "epoch": 1.2767932489451477,
+      "grad_norm": 0.4156053364276886,
+      "learning_rate": 0.00018990699177142548,
+      "loss": 0.7419850826263428,
+      "step": 3026
+    },
+    {
+      "epoch": 1.2776371308016878,
+      "grad_norm": 0.41845977306365967,
+      "learning_rate": 0.00018988595395690388,
+      "loss": 0.67264324426651,
+      "step": 3028
+    },
+    {
+      "epoch": 1.2784810126582278,
+      "grad_norm": 0.3826757073402405,
+      "learning_rate": 0.00018986489540741893,
+      "loss": 0.6269281506538391,
+      "step": 3030
+    },
+    {
+      "epoch": 1.279324894514768,
+      "grad_norm": 0.4531428813934326,
+      "learning_rate": 0.00018984381612782854,
+      "loss": 0.71211177110672,
+      "step": 3032
+    },
+    {
+      "epoch": 1.2801687763713079,
+      "grad_norm": 0.3747255802154541,
+      "learning_rate": 0.00018982271612299524,
+      "loss": 0.635804295539856,
+      "step": 3034
+    },
+    {
+      "epoch": 1.281012658227848,
+      "grad_norm": 0.4079552888870239,
+      "learning_rate": 0.00018980159539778638,
+      "loss": 0.7269464731216431,
+      "step": 3036
+    },
+    {
+      "epoch": 1.2818565400843882,
+      "grad_norm": 0.4193854033946991,
+      "learning_rate": 0.00018978045395707418,
+      "loss": 0.7737077474594116,
+      "step": 3038
+    },
+    {
+      "epoch": 1.2827004219409281,
+      "grad_norm": 0.4182516038417816,
+      "learning_rate": 0.0001897592918057355,
+      "loss": 0.7458174824714661,
+      "step": 3040
+    },
+    {
+      "epoch": 1.2835443037974683,
+      "grad_norm": 0.43343567848205566,
+      "learning_rate": 0.00018973810894865205,
+      "loss": 0.6565808057785034,
+      "step": 3042
+    },
+    {
+      "epoch": 1.2843881856540085,
+      "grad_norm": 0.4294503331184387,
+      "learning_rate": 0.00018971690539071033,
+      "loss": 0.736944317817688,
+      "step": 3044
+    },
+    {
+      "epoch": 1.2852320675105484,
+      "grad_norm": 0.41330763697624207,
+      "learning_rate": 0.0001896956811368016,
+      "loss": 0.6921742558479309,
+      "step": 3046
+    },
+    {
+      "epoch": 1.2860759493670886,
+      "grad_norm": 0.44992703199386597,
+      "learning_rate": 0.0001896744361918218,
+      "loss": 0.7004357576370239,
+      "step": 3048
+    },
+    {
+      "epoch": 1.2869198312236287,
+      "grad_norm": 0.43748870491981506,
+      "learning_rate": 0.0001896531705606719,
+      "loss": 0.6767720580101013,
+      "step": 3050
+    },
+    {
+      "epoch": 1.2877637130801687,
+      "grad_norm": 0.47581973671913147,
+      "learning_rate": 0.0001896318842482573,
+      "loss": 0.7596892714500427,
+      "step": 3052
+    },
+    {
+      "epoch": 1.2886075949367088,
+      "grad_norm": 0.45806005597114563,
+      "learning_rate": 0.00018961057725948844,
+      "loss": 0.7650044560432434,
+      "step": 3054
+    },
+    {
+      "epoch": 1.289451476793249,
+      "grad_norm": 0.4291481375694275,
+      "learning_rate": 0.00018958924959928042,
+      "loss": 0.6654610633850098,
+      "step": 3056
+    },
+    {
+      "epoch": 1.290295358649789,
+      "grad_norm": 0.42454418540000916,
+      "learning_rate": 0.00018956790127255309,
+      "loss": 0.7079797387123108,
+      "step": 3058
+    },
+    {
+      "epoch": 1.2911392405063291,
+      "grad_norm": 0.46820637583732605,
+      "learning_rate": 0.00018954653228423114,
+      "loss": 0.7724313735961914,
+      "step": 3060
+    },
+    {
+      "epoch": 1.2919831223628693,
+      "grad_norm": 0.4440016746520996,
+      "learning_rate": 0.00018952514263924396,
+      "loss": 0.7746338248252869,
+      "step": 3062
+    },
+    {
+      "epoch": 1.2928270042194092,
+      "grad_norm": 0.4296419620513916,
+      "learning_rate": 0.00018950373234252572,
+      "loss": 0.7570929527282715,
+      "step": 3064
+    },
+    {
+      "epoch": 1.2936708860759494,
+      "grad_norm": 0.428780734539032,
+      "learning_rate": 0.0001894823013990154,
+      "loss": 0.7274802923202515,
+      "step": 3066
+    },
+    {
+      "epoch": 1.2945147679324895,
+      "grad_norm": 0.4111585021018982,
+      "learning_rate": 0.0001894608498136567,
+      "loss": 0.6595024466514587,
+      "step": 3068
+    },
+    {
+      "epoch": 1.2953586497890295,
+      "grad_norm": 0.4373013973236084,
+      "learning_rate": 0.00018943937759139802,
+      "loss": 0.7316440939903259,
+      "step": 3070
+    },
+    {
+      "epoch": 1.2962025316455696,
+      "grad_norm": 0.37705400586128235,
+      "learning_rate": 0.0001894178847371927,
+      "loss": 0.7072293162345886,
+      "step": 3072
+    },
+    {
+      "epoch": 1.2970464135021098,
+      "grad_norm": 0.4171125888824463,
+      "learning_rate": 0.00018939637125599864,
+      "loss": 0.7218101024627686,
+      "step": 3074
+    },
+    {
+      "epoch": 1.2978902953586497,
+      "grad_norm": 0.4264235198497772,
+      "learning_rate": 0.00018937483715277865,
+      "loss": 0.725140392780304,
+      "step": 3076
+    },
+    {
+      "epoch": 1.29873417721519,
+      "grad_norm": 0.4251643121242523,
+      "learning_rate": 0.0001893532824325002,
+      "loss": 0.7423256635665894,
+      "step": 3078
+    },
+    {
+      "epoch": 1.29957805907173,
+      "grad_norm": 0.43096452951431274,
+      "learning_rate": 0.00018933170710013554,
+      "loss": 0.7058811187744141,
+      "step": 3080
+    },
+    {
+      "epoch": 1.30042194092827,
+      "grad_norm": 0.466804176568985,
+      "learning_rate": 0.00018931011116066172,
+      "loss": 0.7334373593330383,
+      "step": 3082
+    },
+    {
+      "epoch": 1.3012658227848102,
+      "grad_norm": 0.36896654963493347,
+      "learning_rate": 0.00018928849461906047,
+      "loss": 0.6033462285995483,
+      "step": 3084
+    },
+    {
+      "epoch": 1.3021097046413503,
+      "grad_norm": 0.42875027656555176,
+      "learning_rate": 0.00018926685748031833,
+      "loss": 0.690375566482544,
+      "step": 3086
+    },
+    {
+      "epoch": 1.3029535864978903,
+      "grad_norm": 0.4041171967983246,
+      "learning_rate": 0.00018924519974942658,
+      "loss": 0.7468435168266296,
+      "step": 3088
+    },
+    {
+      "epoch": 1.3037974683544304,
+      "grad_norm": 0.4704097807407379,
+      "learning_rate": 0.00018922352143138126,
+      "loss": 0.7574344873428345,
+      "step": 3090
+    },
+    {
+      "epoch": 1.3046413502109704,
+      "grad_norm": 0.4391418397426605,
+      "learning_rate": 0.0001892018225311831,
+      "loss": 0.756452739238739,
+      "step": 3092
+    },
+    {
+      "epoch": 1.3054852320675105,
+      "grad_norm": 0.42684388160705566,
+      "learning_rate": 0.0001891801030538376,
+      "loss": 0.6620991230010986,
+      "step": 3094
+    },
+    {
+      "epoch": 1.3063291139240507,
+      "grad_norm": 0.4972277283668518,
+      "learning_rate": 0.00018915836300435507,
+      "loss": 0.7764125466346741,
+      "step": 3096
+    },
+    {
+      "epoch": 1.3071729957805907,
+      "grad_norm": 0.3905445337295532,
+      "learning_rate": 0.00018913660238775053,
+      "loss": 0.6557663083076477,
+      "step": 3098
+    },
+    {
+      "epoch": 1.3080168776371308,
+      "grad_norm": 0.4553062319755554,
+      "learning_rate": 0.0001891148212090437,
+      "loss": 0.6746718883514404,
+      "step": 3100
+    },
+    {
+      "epoch": 1.3080168776371308,
+      "eval_loss": 0.8010181784629822,
+      "eval_runtime": 668.1688,
+      "eval_samples_per_second": 3.153,
+      "eval_steps_per_second": 3.153,
+      "step": 3100
+    },
+    {
+      "epoch": 1.3088607594936708,
+      "grad_norm": 0.4254429340362549,
+      "learning_rate": 0.00018909301947325915,
+      "loss": 0.7916419506072998,
+      "step": 3102
+    },
+    {
+      "epoch": 1.309704641350211,
+      "grad_norm": 0.6118508577346802,
+      "learning_rate": 0.00018907119718542602,
+      "loss": 0.883936882019043,
+      "step": 3104
+    },
+    {
+      "epoch": 1.310548523206751,
+      "grad_norm": 0.43206971883773804,
+      "learning_rate": 0.00018904935435057837,
+      "loss": 0.7145928740501404,
+      "step": 3106
+    },
+    {
+      "epoch": 1.311392405063291,
+      "grad_norm": 0.39657703042030334,
+      "learning_rate": 0.0001890274909737549,
+      "loss": 0.7442842125892639,
+      "step": 3108
+    },
+    {
+      "epoch": 1.3122362869198312,
+      "grad_norm": 0.4087927043437958,
+      "learning_rate": 0.00018900560705999904,
+      "loss": 0.6815276145935059,
+      "step": 3110
+    },
+    {
+      "epoch": 1.3130801687763713,
+      "grad_norm": 0.430259108543396,
+      "learning_rate": 0.00018898370261435904,
+      "loss": 0.8153640627861023,
+      "step": 3112
+    },
+    {
+      "epoch": 1.3139240506329113,
+      "grad_norm": 0.47486868500709534,
+      "learning_rate": 0.0001889617776418878,
+      "loss": 0.7828266024589539,
+      "step": 3114
+    },
+    {
+      "epoch": 1.3147679324894515,
+      "grad_norm": 0.4929087162017822,
+      "learning_rate": 0.000188939832147643,
+      "loss": 0.7849065661430359,
+      "step": 3116
+    },
+    {
+      "epoch": 1.3156118143459916,
+      "grad_norm": 0.39038679003715515,
+      "learning_rate": 0.00018891786613668704,
+      "loss": 0.7262985706329346,
+      "step": 3118
+    },
+    {
+      "epoch": 1.3164556962025316,
+      "grad_norm": 0.4827214479446411,
+      "learning_rate": 0.0001888958796140871,
+      "loss": 0.8121856451034546,
+      "step": 3120
+    },
+    {
+      "epoch": 1.3172995780590717,
+      "grad_norm": 0.47739601135253906,
+      "learning_rate": 0.000188873872584915,
+      "loss": 0.7165566086769104,
+      "step": 3122
+    },
+    {
+      "epoch": 1.3181434599156119,
+      "grad_norm": 0.4274037778377533,
+      "learning_rate": 0.0001888518450542473,
+      "loss": 0.6698412299156189,
+      "step": 3124
+    },
+    {
+      "epoch": 1.3189873417721518,
+      "grad_norm": 0.44825276732444763,
+      "learning_rate": 0.00018882979702716544,
+      "loss": 0.7271128296852112,
+      "step": 3126
+    },
+    {
+      "epoch": 1.319831223628692,
+      "grad_norm": 0.42956918478012085,
+      "learning_rate": 0.00018880772850875537,
+      "loss": 0.7924301624298096,
+      "step": 3128
+    },
+    {
+      "epoch": 1.3206751054852321,
+      "grad_norm": 0.4848136305809021,
+      "learning_rate": 0.00018878563950410792,
+      "loss": 0.7846497893333435,
+      "step": 3130
+    },
+    {
+      "epoch": 1.321518987341772,
+      "grad_norm": 0.46651726961135864,
+      "learning_rate": 0.00018876353001831864,
+      "loss": 0.7262214422225952,
+      "step": 3132
+    },
+    {
+      "epoch": 1.3223628691983123,
+      "grad_norm": 0.44598284363746643,
+      "learning_rate": 0.0001887414000564877,
+      "loss": 0.7640091180801392,
+      "step": 3134
+    },
+    {
+      "epoch": 1.3232067510548524,
+      "grad_norm": 0.43744903802871704,
+      "learning_rate": 0.00018871924962372006,
+      "loss": 0.6960080862045288,
+      "step": 3136
+    },
+    {
+      "epoch": 1.3240506329113924,
+      "grad_norm": 0.4737822413444519,
+      "learning_rate": 0.00018869707872512544,
+      "loss": 0.7121244072914124,
+      "step": 3138
+    },
+    {
+      "epoch": 1.3248945147679325,
+      "grad_norm": 0.44452038407325745,
+      "learning_rate": 0.00018867488736581817,
+      "loss": 0.713148295879364,
+      "step": 3140
+    },
+    {
+      "epoch": 1.3257383966244727,
+      "grad_norm": 0.47142693400382996,
+      "learning_rate": 0.0001886526755509175,
+      "loss": 0.6904974579811096,
+      "step": 3142
+    },
+    {
+      "epoch": 1.3265822784810126,
+      "grad_norm": 0.40572887659072876,
+      "learning_rate": 0.00018863044328554712,
+      "loss": 0.7319386005401611,
+      "step": 3144
+    },
+    {
+      "epoch": 1.3274261603375528,
+      "grad_norm": 0.44173532724380493,
+      "learning_rate": 0.00018860819057483566,
+      "loss": 0.7268120646476746,
+      "step": 3146
+    },
+    {
+      "epoch": 1.328270042194093,
+      "grad_norm": 0.43025046586990356,
+      "learning_rate": 0.00018858591742391642,
+      "loss": 0.7129524946212769,
+      "step": 3148
+    },
+    {
+      "epoch": 1.3291139240506329,
+      "grad_norm": 0.41418036818504333,
+      "learning_rate": 0.00018856362383792732,
+      "loss": 0.6917402148246765,
+      "step": 3150
+    },
+    {
+      "epoch": 1.329957805907173,
+      "grad_norm": 0.46275854110717773,
+      "learning_rate": 0.00018854130982201111,
+      "loss": 0.7653918862342834,
+      "step": 3152
+    },
+    {
+      "epoch": 1.3308016877637132,
+      "grad_norm": 0.38927653431892395,
+      "learning_rate": 0.0001885189753813152,
+      "loss": 0.7311477065086365,
+      "step": 3154
+    },
+    {
+      "epoch": 1.3316455696202532,
+      "grad_norm": 0.44294974207878113,
+      "learning_rate": 0.0001884966205209917,
+      "loss": 0.7978816628456116,
+      "step": 3156
+    },
+    {
+      "epoch": 1.3324894514767933,
+      "grad_norm": 0.3772844970226288,
+      "learning_rate": 0.00018847424524619745,
+      "loss": 0.7497738599777222,
+      "step": 3158
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.4538787603378296,
+      "learning_rate": 0.000188451849562094,
+      "loss": 0.712806761264801,
+      "step": 3160
+    },
+    {
+      "epoch": 1.3341772151898734,
+      "grad_norm": 0.4194336533546448,
+      "learning_rate": 0.00018842943347384763,
+      "loss": 0.7282580137252808,
+      "step": 3162
+    },
+    {
+      "epoch": 1.3350210970464136,
+      "grad_norm": 0.4186438024044037,
+      "learning_rate": 0.00018840699698662925,
+      "loss": 0.797752320766449,
+      "step": 3164
+    },
+    {
+      "epoch": 1.3358649789029535,
+      "grad_norm": 0.47711506485939026,
+      "learning_rate": 0.00018838454010561457,
+      "loss": 0.8175379633903503,
+      "step": 3166
+    },
+    {
+      "epoch": 1.3367088607594937,
+      "grad_norm": 0.4380176067352295,
+      "learning_rate": 0.00018836206283598394,
+      "loss": 0.659176766872406,
+      "step": 3168
+    },
+    {
+      "epoch": 1.3375527426160336,
+      "grad_norm": 0.4505249559879303,
+      "learning_rate": 0.00018833956518292245,
+      "loss": 0.732314944267273,
+      "step": 3170
+    },
+    {
+      "epoch": 1.3383966244725738,
+      "grad_norm": 0.3832913935184479,
+      "learning_rate": 0.00018831704715161984,
+      "loss": 0.6791377067565918,
+      "step": 3172
+    },
+    {
+      "epoch": 1.339240506329114,
+      "grad_norm": 0.42996498942375183,
+      "learning_rate": 0.00018829450874727064,
+      "loss": 0.7395796775817871,
+      "step": 3174
+    },
+    {
+      "epoch": 1.340084388185654,
+      "grad_norm": 0.3926190435886383,
+      "learning_rate": 0.000188271949975074,
+      "loss": 0.6893631219863892,
+      "step": 3176
+    },
+    {
+      "epoch": 1.340928270042194,
+      "grad_norm": 0.4353938102722168,
+      "learning_rate": 0.0001882493708402338,
+      "loss": 0.6708382964134216,
+      "step": 3178
+    },
+    {
+      "epoch": 1.3417721518987342,
+      "grad_norm": 0.43043214082717896,
+      "learning_rate": 0.0001882267713479586,
+      "loss": 0.8447695374488831,
+      "step": 3180
+    },
+    {
+      "epoch": 1.3426160337552742,
+      "grad_norm": 0.40693536400794983,
+      "learning_rate": 0.00018820415150346163,
+      "loss": 0.7081091403961182,
+      "step": 3182
+    },
+    {
+      "epoch": 1.3434599156118143,
+      "grad_norm": 0.45037057995796204,
+      "learning_rate": 0.00018818151131196097,
+      "loss": 0.6822336316108704,
+      "step": 3184
+    },
+    {
+      "epoch": 1.3443037974683545,
+      "grad_norm": 0.3731948137283325,
+      "learning_rate": 0.00018815885077867915,
+      "loss": 0.6316909790039062,
+      "step": 3186
+    },
+    {
+      "epoch": 1.3451476793248944,
+      "grad_norm": 0.4248095750808716,
+      "learning_rate": 0.00018813616990884363,
+      "loss": 0.7452375292778015,
+      "step": 3188
+    },
+    {
+      "epoch": 1.3459915611814346,
+      "grad_norm": 0.39229142665863037,
+      "learning_rate": 0.00018811346870768638,
+      "loss": 0.6493782997131348,
+      "step": 3190
+    },
+    {
+      "epoch": 1.3468354430379748,
+      "grad_norm": 0.45015549659729004,
+      "learning_rate": 0.00018809074718044414,
+      "loss": 0.735163688659668,
+      "step": 3192
+    },
+    {
+      "epoch": 1.3476793248945147,
+      "grad_norm": 0.39756354689598083,
+      "learning_rate": 0.00018806800533235836,
+      "loss": 0.6656006574630737,
+      "step": 3194
+    },
+    {
+      "epoch": 1.3485232067510549,
+      "grad_norm": 0.4410010576248169,
+      "learning_rate": 0.0001880452431686751,
+      "loss": 0.7040987014770508,
+      "step": 3196
+    },
+    {
+      "epoch": 1.349367088607595,
+      "grad_norm": 0.48144352436065674,
+      "learning_rate": 0.0001880224606946452,
+      "loss": 0.7722331285476685,
+      "step": 3198
+    },
+    {
+      "epoch": 1.350210970464135,
+      "grad_norm": 0.38644957542419434,
+      "learning_rate": 0.00018799965791552414,
+      "loss": 0.6681689023971558,
+      "step": 3200
+    },
+    {
+      "epoch": 1.350210970464135,
+      "eval_loss": 0.797160804271698,
+      "eval_runtime": 680.976,
+      "eval_samples_per_second": 3.094,
+      "eval_steps_per_second": 3.094,
+      "step": 3200
+    },
+    {
+      "epoch": 1.3510548523206751,
+      "grad_norm": 0.4471014440059662,
+      "learning_rate": 0.00018797683483657201,
+      "loss": 0.6995527744293213,
+      "step": 3202
+    },
+    {
+      "epoch": 1.3518987341772153,
+      "grad_norm": 0.4065002202987671,
+      "learning_rate": 0.00018795399146305378,
+      "loss": 0.659622848033905,
+      "step": 3204
+    },
+    {
+      "epoch": 1.3527426160337552,
+      "grad_norm": 0.4787910580635071,
+      "learning_rate": 0.00018793112780023888,
+      "loss": 0.7927116751670837,
+      "step": 3206
+    },
+    {
+      "epoch": 1.3535864978902954,
+      "grad_norm": 0.43351876735687256,
+      "learning_rate": 0.0001879082438534015,
+      "loss": 0.6738835573196411,
+      "step": 3208
+    },
+    {
+      "epoch": 1.3544303797468356,
+      "grad_norm": 0.4275745749473572,
+      "learning_rate": 0.00018788533962782063,
+      "loss": 0.7785354852676392,
+      "step": 3210
+    },
+    {
+      "epoch": 1.3552742616033755,
+      "grad_norm": 0.4506145119667053,
+      "learning_rate": 0.00018786241512877975,
+      "loss": 0.7325424551963806,
+      "step": 3212
+    },
+    {
+      "epoch": 1.3561181434599157,
+      "grad_norm": 0.42040735483169556,
+      "learning_rate": 0.00018783947036156713,
+      "loss": 0.6898744702339172,
+      "step": 3214
+    },
+    {
+      "epoch": 1.3569620253164558,
+      "grad_norm": 0.4650438129901886,
+      "learning_rate": 0.0001878165053314757,
+      "loss": 0.7498815655708313,
+      "step": 3216
+    },
+    {
+      "epoch": 1.3578059071729958,
+      "grad_norm": 0.4194130599498749,
+      "learning_rate": 0.000187793520043803,
+      "loss": 0.6891883611679077,
+      "step": 3218
+    },
+    {
+      "epoch": 1.358649789029536,
+      "grad_norm": 0.4447539746761322,
+      "learning_rate": 0.0001877705145038514,
+      "loss": 0.753805935382843,
+      "step": 3220
+    },
+    {
+      "epoch": 1.3594936708860759,
+      "grad_norm": 0.4047110974788666,
+      "learning_rate": 0.00018774748871692773,
+      "loss": 0.6476293802261353,
+      "step": 3222
+    },
+    {
+      "epoch": 1.360337552742616,
+      "grad_norm": 0.39818957448005676,
+      "learning_rate": 0.00018772444268834364,
+      "loss": 0.6765931844711304,
+      "step": 3224
+    },
+    {
+      "epoch": 1.3611814345991562,
+      "grad_norm": 0.41843733191490173,
+      "learning_rate": 0.0001877013764234154,
+      "loss": 0.7360382676124573,
+      "step": 3226
+    },
+    {
+      "epoch": 1.3620253164556961,
+      "grad_norm": 0.41397789120674133,
+      "learning_rate": 0.00018767828992746394,
+      "loss": 0.7149524688720703,
+      "step": 3228
+    },
+    {
+      "epoch": 1.3628691983122363,
+      "grad_norm": 0.43585047125816345,
+      "learning_rate": 0.00018765518320581492,
+      "loss": 0.6932925581932068,
+      "step": 3230
+    },
+    {
+      "epoch": 1.3637130801687762,
+      "grad_norm": 0.3651057481765747,
+      "learning_rate": 0.00018763205626379857,
+      "loss": 0.6561182141304016,
+      "step": 3232
+    },
+    {
+      "epoch": 1.3645569620253164,
+      "grad_norm": 0.44937077164649963,
+      "learning_rate": 0.00018760890910674984,
+      "loss": 0.7436831593513489,
+      "step": 3234
+    },
+    {
+      "epoch": 1.3654008438818566,
+      "grad_norm": 0.5045920014381409,
+      "learning_rate": 0.00018758574174000832,
+      "loss": 0.7849522233009338,
+      "step": 3236
+    },
+    {
+      "epoch": 1.3662447257383965,
+      "grad_norm": 0.4625580906867981,
+      "learning_rate": 0.00018756255416891833,
+      "loss": 0.7501618266105652,
+      "step": 3238
+    },
+    {
+      "epoch": 1.3670886075949367,
+      "grad_norm": 0.3721269369125366,
+      "learning_rate": 0.00018753934639882875,
+      "loss": 0.7656083106994629,
+      "step": 3240
+    },
+    {
+      "epoch": 1.3679324894514768,
+      "grad_norm": 0.41721680760383606,
+      "learning_rate": 0.00018751611843509318,
+      "loss": 0.7816168069839478,
+      "step": 3242
+    },
+    {
+      "epoch": 1.3687763713080168,
+      "grad_norm": 0.8452199101448059,
+      "learning_rate": 0.00018749287028306988,
+      "loss": 0.7002755403518677,
+      "step": 3244
+    },
+    {
+      "epoch": 1.369620253164557,
+      "grad_norm": 0.4320507347583771,
+      "learning_rate": 0.00018746960194812172,
+      "loss": 0.7358270883560181,
+      "step": 3246
+    },
+    {
+      "epoch": 1.370464135021097,
+      "grad_norm": 0.43323323130607605,
+      "learning_rate": 0.00018744631343561627,
+      "loss": 0.6528644561767578,
+      "step": 3248
+    },
+    {
+      "epoch": 1.371308016877637,
+      "grad_norm": 0.41487041115760803,
+      "learning_rate": 0.0001874230047509258,
+      "loss": 0.7317517995834351,
+      "step": 3250
+    },
+    {
+      "epoch": 1.3721518987341772,
+      "grad_norm": 0.5085384249687195,
+      "learning_rate": 0.00018739967589942707,
+      "loss": 0.8236643671989441,
+      "step": 3252
+    },
+    {
+      "epoch": 1.3729957805907174,
+      "grad_norm": 0.4126952290534973,
+      "learning_rate": 0.00018737632688650168,
+      "loss": 0.730254054069519,
+      "step": 3254
+    },
+    {
+      "epoch": 1.3738396624472573,
+      "grad_norm": 0.5272142291069031,
+      "learning_rate": 0.00018735295771753573,
+      "loss": 0.7637890577316284,
+      "step": 3256
+    },
+    {
+      "epoch": 1.3746835443037975,
+      "grad_norm": 0.4319530725479126,
+      "learning_rate": 0.00018732956839792012,
+      "loss": 0.7906988263130188,
+      "step": 3258
+    },
+    {
+      "epoch": 1.3755274261603376,
+      "grad_norm": 0.4393143355846405,
+      "learning_rate": 0.00018730615893305025,
+      "loss": 0.7261547446250916,
+      "step": 3260
+    },
+    {
+      "epoch": 1.3763713080168776,
+      "grad_norm": 0.4161636233329773,
+      "learning_rate": 0.00018728272932832632,
+      "loss": 0.6857497692108154,
+      "step": 3262
+    },
+    {
+      "epoch": 1.3772151898734177,
+      "grad_norm": 0.48862120509147644,
+      "learning_rate": 0.00018725927958915297,
+      "loss": 0.7488857507705688,
+      "step": 3264
+    },
+    {
+      "epoch": 1.378059071729958,
+      "grad_norm": 0.4422304928302765,
+      "learning_rate": 0.0001872358097209397,
+      "loss": 0.7257721424102783,
+      "step": 3266
+    },
+    {
+      "epoch": 1.3789029535864978,
+      "grad_norm": 0.4577067494392395,
+      "learning_rate": 0.00018721231972910048,
+      "loss": 0.746697187423706,
+      "step": 3268
+    },
+    {
+      "epoch": 1.379746835443038,
+      "grad_norm": 0.42382001876831055,
+      "learning_rate": 0.00018718880961905406,
+      "loss": 0.768519401550293,
+      "step": 3270
+    },
+    {
+      "epoch": 1.3805907172995782,
+      "grad_norm": 0.43368014693260193,
+      "learning_rate": 0.00018716527939622378,
+      "loss": 0.7426102757453918,
+      "step": 3272
+    },
+    {
+      "epoch": 1.381434599156118,
+      "grad_norm": 0.3777220547199249,
+      "learning_rate": 0.00018714172906603757,
+      "loss": 0.7104197144508362,
+      "step": 3274
+    },
+    {
+      "epoch": 1.3822784810126583,
+      "grad_norm": 0.46071696281433105,
+      "learning_rate": 0.00018711815863392802,
+      "loss": 0.7502968907356262,
+      "step": 3276
+    },
+    {
+      "epoch": 1.3831223628691984,
+      "grad_norm": 0.42510855197906494,
+      "learning_rate": 0.00018709456810533245,
+      "loss": 0.6400540471076965,
+      "step": 3278
+    },
+    {
+      "epoch": 1.3839662447257384,
+      "grad_norm": 0.4217228293418884,
+      "learning_rate": 0.00018707095748569268,
+      "loss": 0.6648687720298767,
+      "step": 3280
+    },
+    {
+      "epoch": 1.3848101265822785,
+      "grad_norm": 0.48267799615859985,
+      "learning_rate": 0.00018704732678045526,
+      "loss": 0.7176171541213989,
+      "step": 3282
+    },
+    {
+      "epoch": 1.3856540084388187,
+      "grad_norm": 0.47114965319633484,
+      "learning_rate": 0.00018702367599507134,
+      "loss": 0.8402457237243652,
+      "step": 3284
+    },
+    {
+      "epoch": 1.3864978902953586,
+      "grad_norm": 0.5054492354393005,
+      "learning_rate": 0.00018700000513499667,
+      "loss": 0.8130836486816406,
+      "step": 3286
+    },
+    {
+      "epoch": 1.3873417721518988,
+      "grad_norm": 0.39435645937919617,
+      "learning_rate": 0.00018697631420569168,
+      "loss": 0.6432579755783081,
+      "step": 3288
+    },
+    {
+      "epoch": 1.3881856540084387,
+      "grad_norm": 0.3997228443622589,
+      "learning_rate": 0.00018695260321262142,
+      "loss": 0.6921643614768982,
+      "step": 3290
+    },
+    {
+      "epoch": 1.389029535864979,
+      "grad_norm": 0.37304237484931946,
+      "learning_rate": 0.00018692887216125557,
+      "loss": 0.7162153124809265,
+      "step": 3292
+    },
+    {
+      "epoch": 1.389873417721519,
+      "grad_norm": 0.7997981309890747,
+      "learning_rate": 0.00018690512105706838,
+      "loss": 0.7194803953170776,
+      "step": 3294
+    },
+    {
+      "epoch": 1.390717299578059,
+      "grad_norm": 0.41268959641456604,
+      "learning_rate": 0.00018688134990553883,
+      "loss": 0.6888322830200195,
+      "step": 3296
+    },
+    {
+      "epoch": 1.3915611814345992,
+      "grad_norm": 0.4485826790332794,
+      "learning_rate": 0.00018685755871215043,
+      "loss": 0.7387221455574036,
+      "step": 3298
+    },
+    {
+      "epoch": 1.3924050632911391,
+      "grad_norm": 0.47835296392440796,
+      "learning_rate": 0.0001868337474823914,
+      "loss": 0.8061952590942383,
+      "step": 3300
+    },
+    {
+      "epoch": 1.3924050632911391,
+      "eval_loss": 0.795619547367096,
+      "eval_runtime": 692.7157,
+      "eval_samples_per_second": 3.042,
+      "eval_steps_per_second": 3.042,
+      "step": 3300
+    },
+    {
+      "epoch": 1.3932489451476793,
+      "grad_norm": 0.4024396240711212,
+      "learning_rate": 0.00018680991622175446,
+      "loss": 0.7317537069320679,
+      "step": 3302
+    },
+    {
+      "epoch": 1.3940928270042194,
+      "grad_norm": 0.4526490271091461,
+      "learning_rate": 0.00018678606493573708,
+      "loss": 0.7470992207527161,
+      "step": 3304
+    },
+    {
+      "epoch": 1.3949367088607594,
+      "grad_norm": 0.5055129528045654,
+      "learning_rate": 0.00018676219362984126,
+      "loss": 0.7627943754196167,
+      "step": 3306
+    },
+    {
+      "epoch": 1.3957805907172995,
+      "grad_norm": 0.4030216634273529,
+      "learning_rate": 0.0001867383023095737,
+      "loss": 0.684033989906311,
+      "step": 3308
+    },
+    {
+      "epoch": 1.3966244725738397,
+      "grad_norm": 0.4353308379650116,
+      "learning_rate": 0.0001867143909804456,
+      "loss": 0.7366722822189331,
+      "step": 3310
+    },
+    {
+      "epoch": 1.3974683544303796,
+      "grad_norm": 0.5360553860664368,
+      "learning_rate": 0.0001866904596479729,
+      "loss": 0.8050951957702637,
+      "step": 3312
+    },
+    {
+      "epoch": 1.3983122362869198,
+      "grad_norm": 0.46081677079200745,
+      "learning_rate": 0.00018666650831767608,
+      "loss": 0.7227963805198669,
+      "step": 3314
+    },
+    {
+      "epoch": 1.39915611814346,
+      "grad_norm": 0.4325845539569855,
+      "learning_rate": 0.00018664253699508029,
+      "loss": 0.714216411113739,
+      "step": 3316
+    },
+    {
+      "epoch": 1.4,
+      "grad_norm": 0.4373100996017456,
+      "learning_rate": 0.0001866185456857152,
+      "loss": 0.7378593683242798,
+      "step": 3318
+    },
+    {
+      "epoch": 1.40084388185654,
+      "grad_norm": 0.43221279978752136,
+      "learning_rate": 0.00018659453439511512,
+      "loss": 0.7565981149673462,
+      "step": 3320
+    },
+    {
+      "epoch": 1.4016877637130802,
+      "grad_norm": 0.3938370943069458,
+      "learning_rate": 0.00018657050312881904,
+      "loss": 0.6409488916397095,
+      "step": 3322
+    },
+    {
+      "epoch": 1.4025316455696202,
+      "grad_norm": 0.43609440326690674,
+      "learning_rate": 0.00018654645189237055,
+      "loss": 0.6911652088165283,
+      "step": 3324
+    },
+    {
+      "epoch": 1.4033755274261603,
+      "grad_norm": 0.4526350200176239,
+      "learning_rate": 0.00018652238069131774,
+      "loss": 0.7275710701942444,
+      "step": 3326
+    },
+    {
+      "epoch": 1.4042194092827005,
+      "grad_norm": 0.3990516662597656,
+      "learning_rate": 0.0001864982895312134,
+      "loss": 0.6466895341873169,
+      "step": 3328
+    },
+    {
+      "epoch": 1.4050632911392404,
+      "grad_norm": 0.44321122765541077,
+      "learning_rate": 0.0001864741784176149,
+      "loss": 0.6520540714263916,
+      "step": 3330
+    },
+    {
+      "epoch": 1.4059071729957806,
+      "grad_norm": 0.44733941555023193,
+      "learning_rate": 0.0001864500473560842,
+      "loss": 0.7289671301841736,
+      "step": 3332
+    },
+    {
+      "epoch": 1.4067510548523208,
+      "grad_norm": 0.4569414556026459,
+      "learning_rate": 0.00018642589635218789,
+      "loss": 0.7664391398429871,
+      "step": 3334
+    },
+    {
+      "epoch": 1.4075949367088607,
+      "grad_norm": 0.4172186851501465,
+      "learning_rate": 0.00018640172541149709,
+      "loss": 0.7747462391853333,
+      "step": 3336
+    },
+    {
+      "epoch": 1.4084388185654009,
+      "grad_norm": 0.4540993869304657,
+      "learning_rate": 0.0001863775345395877,
+      "loss": 0.8337607383728027,
+      "step": 3338
+    },
+    {
+      "epoch": 1.409282700421941,
+      "grad_norm": 0.4306154251098633,
+      "learning_rate": 0.00018635332374203993,
+      "loss": 0.6851577758789062,
+      "step": 3340
+    },
+    {
+      "epoch": 1.410126582278481,
+      "grad_norm": 0.45615455508232117,
+      "learning_rate": 0.00018632909302443884,
+      "loss": 0.6841398477554321,
+      "step": 3342
+    },
+    {
+      "epoch": 1.4109704641350211,
+      "grad_norm": 0.35296666622161865,
+      "learning_rate": 0.00018630484239237397,
+      "loss": 0.6672312021255493,
+      "step": 3344
+    },
+    {
+      "epoch": 1.4118143459915613,
+      "grad_norm": 0.4424770176410675,
+      "learning_rate": 0.00018628057185143945,
+      "loss": 0.7027004361152649,
+      "step": 3346
+    },
+    {
+      "epoch": 1.4126582278481012,
+      "grad_norm": 0.429817259311676,
+      "learning_rate": 0.0001862562814072341,
+      "loss": 0.7397928237915039,
+      "step": 3348
+    },
+    {
+      "epoch": 1.4135021097046414,
+      "grad_norm": 0.4720051884651184,
+      "learning_rate": 0.00018623197106536118,
+      "loss": 0.8009215593338013,
+      "step": 3350
+    },
+    {
+      "epoch": 1.4143459915611816,
+      "grad_norm": 0.42197802662849426,
+      "learning_rate": 0.0001862076408314287,
+      "loss": 0.6875046491622925,
+      "step": 3352
+    },
+    {
+      "epoch": 1.4151898734177215,
+      "grad_norm": 0.3674047589302063,
+      "learning_rate": 0.00018618329071104907,
+      "loss": 0.7202261090278625,
+      "step": 3354
+    },
+    {
+      "epoch": 1.4160337552742617,
+      "grad_norm": 0.42327889800071716,
+      "learning_rate": 0.0001861589207098395,
+      "loss": 0.6981248259544373,
+      "step": 3356
+    },
+    {
+      "epoch": 1.4168776371308016,
+      "grad_norm": 0.40644150972366333,
+      "learning_rate": 0.00018613453083342162,
+      "loss": 0.7081356048583984,
+      "step": 3358
+    },
+    {
+      "epoch": 1.4177215189873418,
+      "grad_norm": 0.3948482275009155,
+      "learning_rate": 0.00018611012108742168,
+      "loss": 0.6420091986656189,
+      "step": 3360
+    },
+    {
+      "epoch": 1.4185654008438817,
+      "grad_norm": 0.44622674584388733,
+      "learning_rate": 0.00018608569147747064,
+      "loss": 0.7662023305892944,
+      "step": 3362
+    },
+    {
+      "epoch": 1.4194092827004219,
+      "grad_norm": 0.39772045612335205,
+      "learning_rate": 0.00018606124200920386,
+      "loss": 0.7243877649307251,
+      "step": 3364
+    },
+    {
+      "epoch": 1.420253164556962,
+      "grad_norm": 0.4284326136112213,
+      "learning_rate": 0.00018603677268826138,
+      "loss": 0.7399296164512634,
+      "step": 3366
+    },
+    {
+      "epoch": 1.421097046413502,
+      "grad_norm": 0.3749450445175171,
+      "learning_rate": 0.0001860122835202878,
+      "loss": 0.6978716850280762,
+      "step": 3368
+    },
+    {
+      "epoch": 1.4219409282700421,
+      "grad_norm": 0.46721795201301575,
+      "learning_rate": 0.00018598777451093234,
+      "loss": 0.7733059525489807,
+      "step": 3370
+    },
+    {
+      "epoch": 1.4227848101265823,
+      "grad_norm": 0.39606085419654846,
+      "learning_rate": 0.0001859632456658487,
+      "loss": 0.6721867918968201,
+      "step": 3372
+    },
+    {
+      "epoch": 1.4236286919831223,
+      "grad_norm": 0.3879879117012024,
+      "learning_rate": 0.00018593869699069527,
+      "loss": 0.715825080871582,
+      "step": 3374
+    },
+    {
+      "epoch": 1.4244725738396624,
+      "grad_norm": 0.3641676902770996,
+      "learning_rate": 0.0001859141284911349,
+      "loss": 0.6226941347122192,
+      "step": 3376
+    },
+    {
+      "epoch": 1.4253164556962026,
+      "grad_norm": 0.4612879753112793,
+      "learning_rate": 0.0001858895401728351,
+      "loss": 0.7972121238708496,
+      "step": 3378
+    },
+    {
+      "epoch": 1.4261603375527425,
+      "grad_norm": 0.4406103193759918,
+      "learning_rate": 0.00018586493204146795,
+      "loss": 0.6663658022880554,
+      "step": 3380
+    },
+    {
+      "epoch": 1.4270042194092827,
+      "grad_norm": 0.420205295085907,
+      "learning_rate": 0.00018584030410271,
+      "loss": 0.6867321133613586,
+      "step": 3382
+    },
+    {
+      "epoch": 1.4278481012658228,
+      "grad_norm": 0.4499560594558716,
+      "learning_rate": 0.00018581565636224253,
+      "loss": 0.66688072681427,
+      "step": 3384
+    },
+    {
+      "epoch": 1.4286919831223628,
+      "grad_norm": 0.42507919669151306,
+      "learning_rate": 0.00018579098882575123,
+      "loss": 0.6714702844619751,
+      "step": 3386
+    },
+    {
+      "epoch": 1.429535864978903,
+      "grad_norm": 0.43856513500213623,
+      "learning_rate": 0.0001857663014989265,
+      "loss": 0.7219351530075073,
+      "step": 3388
+    },
+    {
+      "epoch": 1.4303797468354431,
+      "grad_norm": 0.5167419910430908,
+      "learning_rate": 0.00018574159438746314,
+      "loss": 0.7157991528511047,
+      "step": 3390
+    },
+    {
+      "epoch": 1.431223628691983,
+      "grad_norm": 0.4443167746067047,
+      "learning_rate": 0.00018571686749706067,
+      "loss": 0.7509840726852417,
+      "step": 3392
+    },
+    {
+      "epoch": 1.4320675105485232,
+      "grad_norm": 0.41321757435798645,
+      "learning_rate": 0.0001856921208334231,
+      "loss": 0.756000816822052,
+      "step": 3394
+    },
+    {
+      "epoch": 1.4329113924050634,
+      "grad_norm": 0.4094153046607971,
+      "learning_rate": 0.000185667354402259,
+      "loss": 0.7047854065895081,
+      "step": 3396
+    },
+    {
+      "epoch": 1.4337552742616033,
+      "grad_norm": 0.4248482286930084,
+      "learning_rate": 0.0001856425682092815,
+      "loss": 0.7620865106582642,
+      "step": 3398
+    },
+    {
+      "epoch": 1.4345991561181435,
+      "grad_norm": 0.4190550148487091,
+      "learning_rate": 0.0001856177622602083,
+      "loss": 0.7441845536231995,
+      "step": 3400
+    },
+    {
+      "epoch": 1.4345991561181435,
+      "eval_loss": 0.7917885780334473,
+      "eval_runtime": 686.1689,
+      "eval_samples_per_second": 3.071,
+      "eval_steps_per_second": 3.071,
+      "step": 3400
+    },
+    {
+      "epoch": 1.4354430379746836,
+      "grad_norm": 0.45602327585220337,
+      "learning_rate": 0.00018559293656076166,
+      "loss": 0.7123498916625977,
+      "step": 3402
+    },
+    {
+      "epoch": 1.4362869198312236,
+      "grad_norm": 0.5209269523620605,
+      "learning_rate": 0.00018556809111666843,
+      "loss": 0.7096527814865112,
+      "step": 3404
+    },
+    {
+      "epoch": 1.4371308016877637,
+      "grad_norm": 0.44521182775497437,
+      "learning_rate": 0.00018554322593365995,
+      "loss": 0.7116838693618774,
+      "step": 3406
+    },
+    {
+      "epoch": 1.437974683544304,
+      "grad_norm": 0.44571003317832947,
+      "learning_rate": 0.00018551834101747212,
+      "loss": 0.7756352424621582,
+      "step": 3408
+    },
+    {
+      "epoch": 1.4388185654008439,
+      "grad_norm": 0.4030776023864746,
+      "learning_rate": 0.00018549343637384543,
+      "loss": 0.689633309841156,
+      "step": 3410
+    },
+    {
+      "epoch": 1.439662447257384,
+      "grad_norm": 0.4025530517101288,
+      "learning_rate": 0.0001854685120085249,
+      "loss": 0.7173827290534973,
+      "step": 3412
+    },
+    {
+      "epoch": 1.4405063291139242,
+      "grad_norm": 0.4376870095729828,
+      "learning_rate": 0.0001854435679272601,
+      "loss": 0.7212072610855103,
+      "step": 3414
+    },
+    {
+      "epoch": 1.4413502109704641,
+      "grad_norm": 0.4469084143638611,
+      "learning_rate": 0.00018541860413580515,
+      "loss": 0.8260660171508789,
+      "step": 3416
+    },
+    {
+      "epoch": 1.4421940928270043,
+      "grad_norm": 0.48033684492111206,
+      "learning_rate": 0.00018539362063991872,
+      "loss": 0.764930248260498,
+      "step": 3418
+    },
+    {
+      "epoch": 1.4430379746835442,
+      "grad_norm": 0.4428006708621979,
+      "learning_rate": 0.00018536861744536403,
+      "loss": 0.7436494827270508,
+      "step": 3420
+    },
+    {
+      "epoch": 1.4438818565400844,
+      "grad_norm": 0.41812989115715027,
+      "learning_rate": 0.00018534359455790885,
+      "loss": 0.7695552706718445,
+      "step": 3422
+    },
+    {
+      "epoch": 1.4447257383966245,
+      "grad_norm": 0.4243115186691284,
+      "learning_rate": 0.00018531855198332544,
+      "loss": 0.7430856823921204,
+      "step": 3424
+    },
+    {
+      "epoch": 1.4455696202531645,
+      "grad_norm": 0.44387027621269226,
+      "learning_rate": 0.00018529348972739067,
+      "loss": 0.7015851140022278,
+      "step": 3426
+    },
+    {
+      "epoch": 1.4464135021097047,
+      "grad_norm": 0.5105122923851013,
+      "learning_rate": 0.0001852684077958859,
+      "loss": 0.7073075771331787,
+      "step": 3428
+    },
+    {
+      "epoch": 1.4472573839662446,
+      "grad_norm": 0.40193185210227966,
+      "learning_rate": 0.00018524330619459708,
+      "loss": 0.7478930950164795,
+      "step": 3430
+    },
+    {
+      "epoch": 1.4481012658227848,
+      "grad_norm": 0.47935256361961365,
+      "learning_rate": 0.00018521818492931463,
+      "loss": 0.6987152099609375,
+      "step": 3432
+    },
+    {
+      "epoch": 1.448945147679325,
+      "grad_norm": 0.42298686504364014,
+      "learning_rate": 0.00018519304400583356,
+      "loss": 0.6677015423774719,
+      "step": 3434
+    },
+    {
+      "epoch": 1.4497890295358649,
+      "grad_norm": 0.37220221757888794,
+      "learning_rate": 0.0001851678834299534,
+      "loss": 0.5727065205574036,
+      "step": 3436
+    },
+    {
+      "epoch": 1.450632911392405,
+      "grad_norm": 0.4990854561328888,
+      "learning_rate": 0.00018514270320747823,
+      "loss": 0.7245276570320129,
+      "step": 3438
+    },
+    {
+      "epoch": 1.4514767932489452,
+      "grad_norm": 0.4366965889930725,
+      "learning_rate": 0.00018511750334421665,
+      "loss": 0.5907649993896484,
+      "step": 3440
+    },
+    {
+      "epoch": 1.4523206751054851,
+      "grad_norm": 0.4355267286300659,
+      "learning_rate": 0.00018509228384598172,
+      "loss": 0.6898093223571777,
+      "step": 3442
+    },
+    {
+      "epoch": 1.4531645569620253,
+      "grad_norm": 0.4055832028388977,
+      "learning_rate": 0.00018506704471859116,
+      "loss": 0.6242932081222534,
+      "step": 3444
+    },
+    {
+      "epoch": 1.4540084388185655,
+      "grad_norm": 0.42599600553512573,
+      "learning_rate": 0.00018504178596786712,
+      "loss": 0.7214970588684082,
+      "step": 3446
+    },
+    {
+      "epoch": 1.4548523206751054,
+      "grad_norm": 0.41228237748146057,
+      "learning_rate": 0.0001850165075996363,
+      "loss": 0.7120789885520935,
+      "step": 3448
+    },
+    {
+      "epoch": 1.4556962025316456,
+      "grad_norm": 0.4529595375061035,
+      "learning_rate": 0.00018499120961972996,
+      "loss": 0.7376183867454529,
+      "step": 3450
+    },
+    {
+      "epoch": 1.4565400843881857,
+      "grad_norm": 0.4260895550251007,
+      "learning_rate": 0.00018496589203398386,
+      "loss": 0.7321018576622009,
+      "step": 3452
+    },
+    {
+      "epoch": 1.4573839662447257,
+      "grad_norm": 0.46325981616973877,
+      "learning_rate": 0.00018494055484823824,
+      "loss": 0.7099342346191406,
+      "step": 3454
+    },
+    {
+      "epoch": 1.4582278481012658,
+      "grad_norm": 0.43322354555130005,
+      "learning_rate": 0.00018491519806833795,
+      "loss": 0.6663834452629089,
+      "step": 3456
+    },
+    {
+      "epoch": 1.459071729957806,
+      "grad_norm": 0.421683132648468,
+      "learning_rate": 0.0001848898217001323,
+      "loss": 0.6634764671325684,
+      "step": 3458
+    },
+    {
+      "epoch": 1.459915611814346,
+      "grad_norm": 0.44365957379341125,
+      "learning_rate": 0.00018486442574947511,
+      "loss": 0.697095513343811,
+      "step": 3460
+    },
+    {
+      "epoch": 1.460759493670886,
+      "grad_norm": 0.4892813265323639,
+      "learning_rate": 0.00018483901022222474,
+      "loss": 0.8009715676307678,
+      "step": 3462
+    },
+    {
+      "epoch": 1.4616033755274263,
+      "grad_norm": 0.4230288863182068,
+      "learning_rate": 0.00018481357512424408,
+      "loss": 0.631499707698822,
+      "step": 3464
+    },
+    {
+      "epoch": 1.4624472573839662,
+      "grad_norm": 0.49490758776664734,
+      "learning_rate": 0.00018478812046140056,
+      "loss": 0.8045525550842285,
+      "step": 3466
+    },
+    {
+      "epoch": 1.4632911392405064,
+      "grad_norm": 0.5035375952720642,
+      "learning_rate": 0.00018476264623956597,
+      "loss": 0.7299133539199829,
+      "step": 3468
+    },
+    {
+      "epoch": 1.4641350210970465,
+      "grad_norm": 0.45048409700393677,
+      "learning_rate": 0.00018473715246461684,
+      "loss": 0.7313228845596313,
+      "step": 3470
+    },
+    {
+      "epoch": 1.4649789029535865,
+      "grad_norm": 0.41414541006088257,
+      "learning_rate": 0.00018471163914243405,
+      "loss": 0.7515901923179626,
+      "step": 3472
+    },
+    {
+      "epoch": 1.4658227848101266,
+      "grad_norm": 0.4440646767616272,
+      "learning_rate": 0.00018468610627890298,
+      "loss": 0.7425532341003418,
+      "step": 3474
+    },
+    {
+      "epoch": 1.4666666666666668,
+      "grad_norm": 0.4749740660190582,
+      "learning_rate": 0.00018466055387991362,
+      "loss": 0.7509122490882874,
+      "step": 3476
+    },
+    {
+      "epoch": 1.4675105485232067,
+      "grad_norm": 0.498951256275177,
+      "learning_rate": 0.00018463498195136046,
+      "loss": 0.7395774126052856,
+      "step": 3478
+    },
+    {
+      "epoch": 1.4683544303797469,
+      "grad_norm": 0.45996859669685364,
+      "learning_rate": 0.00018460939049914237,
+      "loss": 0.7860153317451477,
+      "step": 3480
+    },
+    {
+      "epoch": 1.469198312236287,
+      "grad_norm": 0.5989604592323303,
+      "learning_rate": 0.0001845837795291629,
+      "loss": 0.8193565607070923,
+      "step": 3482
+    },
+    {
+      "epoch": 1.470042194092827,
+      "grad_norm": 0.3903721272945404,
+      "learning_rate": 0.0001845581490473299,
+      "loss": 0.6569210290908813,
+      "step": 3484
+    },
+    {
+      "epoch": 1.4708860759493672,
+      "grad_norm": 0.4316142201423645,
+      "learning_rate": 0.00018453249905955592,
+      "loss": 0.7226642370223999,
+      "step": 3486
+    },
+    {
+      "epoch": 1.471729957805907,
+      "grad_norm": 0.4447585344314575,
+      "learning_rate": 0.00018450682957175786,
+      "loss": 0.8134398460388184,
+      "step": 3488
+    },
+    {
+      "epoch": 1.4725738396624473,
+      "grad_norm": 0.42515450716018677,
+      "learning_rate": 0.0001844811405898572,
+      "loss": 0.7668985724449158,
+      "step": 3490
+    },
+    {
+      "epoch": 1.4734177215189874,
+      "grad_norm": 0.39883363246917725,
+      "learning_rate": 0.0001844554321197799,
+      "loss": 0.7289790511131287,
+      "step": 3492
+    },
+    {
+      "epoch": 1.4742616033755274,
+      "grad_norm": 0.4419080317020416,
+      "learning_rate": 0.00018442970416745645,
+      "loss": 0.7611824870109558,
+      "step": 3494
+    },
+    {
+      "epoch": 1.4751054852320675,
+      "grad_norm": 0.37460315227508545,
+      "learning_rate": 0.0001844039567388217,
+      "loss": 0.6926230788230896,
+      "step": 3496
+    },
+    {
+      "epoch": 1.4759493670886075,
+      "grad_norm": 0.47706854343414307,
+      "learning_rate": 0.00018437818983981513,
+      "loss": 0.7889325618743896,
+      "step": 3498
+    },
+    {
+      "epoch": 1.4767932489451476,
+      "grad_norm": 0.449028879404068,
+      "learning_rate": 0.00018435240347638074,
+      "loss": 0.7231656312942505,
+      "step": 3500
+    },
+    {
+      "epoch": 1.4767932489451476,
+      "eval_loss": 0.7902651429176331,
+      "eval_runtime": 672.312,
+      "eval_samples_per_second": 3.134,
+      "eval_steps_per_second": 3.134,
+      "step": 3500
+    },
+    {
+      "epoch": 1.4776371308016878,
+      "grad_norm": 0.44085755944252014,
+      "learning_rate": 0.00018432659765446686,
+      "loss": 0.7464927434921265,
+      "step": 3502
+    },
+    {
+      "epoch": 1.4784810126582277,
+      "grad_norm": 0.4243546724319458,
+      "learning_rate": 0.0001843007723800264,
+      "loss": 0.7312393188476562,
+      "step": 3504
+    },
+    {
+      "epoch": 1.479324894514768,
+      "grad_norm": 0.4437529444694519,
+      "learning_rate": 0.00018427492765901677,
+      "loss": 0.7169944643974304,
+      "step": 3506
+    },
+    {
+      "epoch": 1.480168776371308,
+      "grad_norm": 0.3639313280582428,
+      "learning_rate": 0.0001842490634973999,
+      "loss": 0.6169725060462952,
+      "step": 3508
+    },
+    {
+      "epoch": 1.481012658227848,
+      "grad_norm": 0.4452592730522156,
+      "learning_rate": 0.0001842231799011421,
+      "loss": 0.7034649848937988,
+      "step": 3510
+    },
+    {
+      "epoch": 1.4818565400843882,
+      "grad_norm": 0.5229591131210327,
+      "learning_rate": 0.00018419727687621421,
+      "loss": 0.7535615563392639,
+      "step": 3512
+    },
+    {
+      "epoch": 1.4827004219409283,
+      "grad_norm": 0.40748512744903564,
+      "learning_rate": 0.00018417135442859163,
+      "loss": 0.6858918070793152,
+      "step": 3514
+    },
+    {
+      "epoch": 1.4835443037974683,
+      "grad_norm": 0.4532773792743683,
+      "learning_rate": 0.0001841454125642541,
+      "loss": 0.7120109796524048,
+      "step": 3516
+    },
+    {
+      "epoch": 1.4843881856540084,
+      "grad_norm": 0.4281020760536194,
+      "learning_rate": 0.00018411945128918593,
+      "loss": 0.6837015748023987,
+      "step": 3518
+    },
+    {
+      "epoch": 1.4852320675105486,
+      "grad_norm": 0.4891088306903839,
+      "learning_rate": 0.0001840934706093759,
+      "loss": 0.6639618873596191,
+      "step": 3520
+    },
+    {
+      "epoch": 1.4860759493670885,
+      "grad_norm": 0.44358906149864197,
+      "learning_rate": 0.00018406747053081724,
+      "loss": 0.693483829498291,
+      "step": 3522
+    },
+    {
+      "epoch": 1.4869198312236287,
+      "grad_norm": 0.40134626626968384,
+      "learning_rate": 0.00018404145105950767,
+      "loss": 0.7046290636062622,
+      "step": 3524
+    },
+    {
+      "epoch": 1.4877637130801689,
+      "grad_norm": 0.4601934552192688,
+      "learning_rate": 0.0001840154122014494,
+      "loss": 0.6991957426071167,
+      "step": 3526
+    },
+    {
+      "epoch": 1.4886075949367088,
+      "grad_norm": 0.39107242226600647,
+      "learning_rate": 0.00018398935396264906,
+      "loss": 0.667023241519928,
+      "step": 3528
+    },
+    {
+      "epoch": 1.489451476793249,
+      "grad_norm": 0.4498196542263031,
+      "learning_rate": 0.00018396327634911784,
+      "loss": 0.7187332510948181,
+      "step": 3530
+    },
+    {
+      "epoch": 1.4902953586497891,
+      "grad_norm": 0.41769179701805115,
+      "learning_rate": 0.0001839371793668713,
+      "loss": 0.6854014992713928,
+      "step": 3532
+    },
+    {
+      "epoch": 1.491139240506329,
+      "grad_norm": 0.40170612931251526,
+      "learning_rate": 0.0001839110630219295,
+      "loss": 0.6643120050430298,
+      "step": 3534
+    },
+    {
+      "epoch": 1.4919831223628692,
+      "grad_norm": 0.43697425723075867,
+      "learning_rate": 0.00018388492732031701,
+      "loss": 0.6822295188903809,
+      "step": 3536
+    },
+    {
+      "epoch": 1.4928270042194094,
+      "grad_norm": 0.4400080442428589,
+      "learning_rate": 0.00018385877226806288,
+      "loss": 0.7500558495521545,
+      "step": 3538
+    },
+    {
+      "epoch": 1.4936708860759493,
+      "grad_norm": 0.5162575244903564,
+      "learning_rate": 0.00018383259787120048,
+      "loss": 0.7742161750793457,
+      "step": 3540
+    },
+    {
+      "epoch": 1.4945147679324895,
+      "grad_norm": 0.4234789311885834,
+      "learning_rate": 0.0001838064041357678,
+      "loss": 0.7334147095680237,
+      "step": 3542
+    },
+    {
+      "epoch": 1.4953586497890297,
+      "grad_norm": 0.40562525391578674,
+      "learning_rate": 0.0001837801910678072,
+      "loss": 0.691303014755249,
+      "step": 3544
+    },
+    {
+      "epoch": 1.4962025316455696,
+      "grad_norm": 0.3651523292064667,
+      "learning_rate": 0.00018375395867336556,
+      "loss": 0.6478174328804016,
+      "step": 3546
+    },
+    {
+      "epoch": 1.4970464135021098,
+      "grad_norm": 0.45428016781806946,
+      "learning_rate": 0.00018372770695849417,
+      "loss": 0.6800187826156616,
+      "step": 3548
+    },
+    {
+      "epoch": 1.49789029535865,
+      "grad_norm": 0.4258574843406677,
+      "learning_rate": 0.00018370143592924882,
+      "loss": 0.7386951446533203,
+      "step": 3550
+    },
+    {
+      "epoch": 1.4987341772151899,
+      "grad_norm": 0.4260908365249634,
+      "learning_rate": 0.00018367514559168972,
+      "loss": 0.757290780544281,
+      "step": 3552
+    },
+    {
+      "epoch": 1.49957805907173,
+      "grad_norm": 0.4037017226219177,
+      "learning_rate": 0.00018364883595188152,
+      "loss": 0.7254101037979126,
+      "step": 3554
+    },
+    {
+      "epoch": 1.5004219409282702,
+      "grad_norm": 0.4212391674518585,
+      "learning_rate": 0.0001836225070158934,
+      "loss": 0.7342239618301392,
+      "step": 3556
+    },
+    {
+      "epoch": 1.5012658227848101,
+      "grad_norm": 0.4902292490005493,
+      "learning_rate": 0.0001835961587897989,
+      "loss": 0.7147436738014221,
+      "step": 3558
+    },
+    {
+      "epoch": 1.50210970464135,
+      "grad_norm": 0.39230459928512573,
+      "learning_rate": 0.00018356979127967605,
+      "loss": 0.7473883628845215,
+      "step": 3560
+    },
+    {
+      "epoch": 1.5029535864978905,
+      "grad_norm": 0.4203200340270996,
+      "learning_rate": 0.00018354340449160735,
+      "loss": 0.7026845812797546,
+      "step": 3562
+    },
+    {
+      "epoch": 1.5037974683544304,
+      "grad_norm": 0.44363775849342346,
+      "learning_rate": 0.00018351699843167972,
+      "loss": 0.7426526546478271,
+      "step": 3564
+    },
+    {
+      "epoch": 1.5046413502109703,
+      "grad_norm": 0.4479754865169525,
+      "learning_rate": 0.0001834905731059845,
+      "loss": 0.6800932884216309,
+      "step": 3566
+    },
+    {
+      "epoch": 1.5054852320675105,
+      "grad_norm": 0.43898341059684753,
+      "learning_rate": 0.0001834641285206176,
+      "loss": 0.6681315898895264,
+      "step": 3568
+    },
+    {
+      "epoch": 1.5063291139240507,
+      "grad_norm": 0.44789746403694153,
+      "learning_rate": 0.00018343766468167916,
+      "loss": 0.7319603562355042,
+      "step": 3570
+    },
+    {
+      "epoch": 1.5071729957805906,
+      "grad_norm": 0.44450023770332336,
+      "learning_rate": 0.00018341118159527392,
+      "loss": 0.7205765843391418,
+      "step": 3572
+    },
+    {
+      "epoch": 1.5080168776371308,
+      "grad_norm": 0.4461601674556732,
+      "learning_rate": 0.00018338467926751103,
+      "loss": 0.8053180575370789,
+      "step": 3574
+    },
+    {
+      "epoch": 1.508860759493671,
+      "grad_norm": 0.415216863155365,
+      "learning_rate": 0.00018335815770450408,
+      "loss": 0.6944835186004639,
+      "step": 3576
+    },
+    {
+      "epoch": 1.5097046413502109,
+      "grad_norm": 0.48354068398475647,
+      "learning_rate": 0.00018333161691237105,
+      "loss": 0.7333522439002991,
+      "step": 3578
+    },
+    {
+      "epoch": 1.510548523206751,
+      "grad_norm": 0.44450995326042175,
+      "learning_rate": 0.00018330505689723446,
+      "loss": 0.7036639451980591,
+      "step": 3580
+    },
+    {
+      "epoch": 1.5113924050632912,
+      "grad_norm": 0.355945348739624,
+      "learning_rate": 0.00018327847766522111,
+      "loss": 0.6589286923408508,
+      "step": 3582
+    },
+    {
+      "epoch": 1.5122362869198311,
+      "grad_norm": 0.4532116651535034,
+      "learning_rate": 0.00018325187922246236,
+      "loss": 0.7284913063049316,
+      "step": 3584
+    },
+    {
+      "epoch": 1.5130801687763713,
+      "grad_norm": 0.45910802483558655,
+      "learning_rate": 0.00018322526157509396,
+      "loss": 0.7253853678703308,
+      "step": 3586
+    },
+    {
+      "epoch": 1.5139240506329115,
+      "grad_norm": 0.4729263186454773,
+      "learning_rate": 0.0001831986247292561,
+      "loss": 0.7626010179519653,
+      "step": 3588
+    },
+    {
+      "epoch": 1.5147679324894514,
+      "grad_norm": 0.42071840167045593,
+      "learning_rate": 0.00018317196869109336,
+      "loss": 0.705466091632843,
+      "step": 3590
+    },
+    {
+      "epoch": 1.5156118143459916,
+      "grad_norm": 0.40275856852531433,
+      "learning_rate": 0.00018314529346675479,
+      "loss": 0.6900550723075867,
+      "step": 3592
+    },
+    {
+      "epoch": 1.5164556962025317,
+      "grad_norm": 0.43557414412498474,
+      "learning_rate": 0.00018311859906239387,
+      "loss": 0.6951512694358826,
+      "step": 3594
+    },
+    {
+      "epoch": 1.5172995780590717,
+      "grad_norm": 0.4007355570793152,
+      "learning_rate": 0.00018309188548416844,
+      "loss": 0.7453750967979431,
+      "step": 3596
+    },
+    {
+      "epoch": 1.5181434599156118,
+      "grad_norm": 0.4686451256275177,
+      "learning_rate": 0.00018306515273824086,
+      "loss": 0.7549434900283813,
+      "step": 3598
+    },
+    {
+      "epoch": 1.518987341772152,
+      "grad_norm": 0.47308704257011414,
+      "learning_rate": 0.0001830384008307778,
+      "loss": 0.7708864212036133,
+      "step": 3600
+    },
+    {
+      "epoch": 1.518987341772152,
+      "eval_loss": 0.785450279712677,
+      "eval_runtime": 675.2312,
+      "eval_samples_per_second": 3.12,
+      "eval_steps_per_second": 3.12,
+      "step": 3600
+    },
+    {
+      "epoch": 1.519831223628692,
+      "grad_norm": 0.4343389570713043,
+      "learning_rate": 0.0001830116297679505,
+      "loss": 0.7501338124275208,
+      "step": 3602
+    },
+    {
+      "epoch": 1.520675105485232,
+      "grad_norm": 0.4971345067024231,
+      "learning_rate": 0.00018298483955593445,
+      "loss": 0.7358015179634094,
+      "step": 3604
+    },
+    {
+      "epoch": 1.5215189873417723,
+      "grad_norm": 0.42013493180274963,
+      "learning_rate": 0.0001829580302009097,
+      "loss": 0.7972728610038757,
+      "step": 3606
+    },
+    {
+      "epoch": 1.5223628691983122,
+      "grad_norm": 0.44829806685447693,
+      "learning_rate": 0.00018293120170906062,
+      "loss": 0.5908644199371338,
+      "step": 3608
+    },
+    {
+      "epoch": 1.5232067510548524,
+      "grad_norm": 0.47412917017936707,
+      "learning_rate": 0.000182904354086576,
+      "loss": 0.7872797846794128,
+      "step": 3610
+    },
+    {
+      "epoch": 1.5240506329113925,
+      "grad_norm": 0.4224211275577545,
+      "learning_rate": 0.0001828774873396491,
+      "loss": 0.7327616214752197,
+      "step": 3612
+    },
+    {
+      "epoch": 1.5248945147679325,
+      "grad_norm": 0.4055147171020508,
+      "learning_rate": 0.00018285060147447756,
+      "loss": 0.6649892330169678,
+      "step": 3614
+    },
+    {
+      "epoch": 1.5257383966244724,
+      "grad_norm": 0.4058035910129547,
+      "learning_rate": 0.00018282369649726346,
+      "loss": 0.6939427852630615,
+      "step": 3616
+    },
+    {
+      "epoch": 1.5265822784810128,
+      "grad_norm": 0.4192068874835968,
+      "learning_rate": 0.00018279677241421325,
+      "loss": 0.6351069211959839,
+      "step": 3618
+    },
+    {
+      "epoch": 1.5274261603375527,
+      "grad_norm": 0.4659916162490845,
+      "learning_rate": 0.00018276982923153776,
+      "loss": 0.6963213682174683,
+      "step": 3620
+    },
+    {
+      "epoch": 1.5282700421940927,
+      "grad_norm": 0.4993051290512085,
+      "learning_rate": 0.00018274286695545229,
+      "loss": 0.7463255524635315,
+      "step": 3622
+    },
+    {
+      "epoch": 1.529113924050633,
+      "grad_norm": 0.39232251048088074,
+      "learning_rate": 0.00018271588559217654,
+      "loss": 0.6566499471664429,
+      "step": 3624
+    },
+    {
+      "epoch": 1.529957805907173,
+      "grad_norm": 0.4482063353061676,
+      "learning_rate": 0.00018268888514793457,
+      "loss": 0.7695494890213013,
+      "step": 3626
+    },
+    {
+      "epoch": 1.530801687763713,
+      "grad_norm": 0.4000336825847626,
+      "learning_rate": 0.00018266186562895483,
+      "loss": 0.7357239723205566,
+      "step": 3628
+    },
+    {
+      "epoch": 1.5316455696202531,
+      "grad_norm": 0.4220462441444397,
+      "learning_rate": 0.0001826348270414703,
+      "loss": 0.7234170436859131,
+      "step": 3630
+    },
+    {
+      "epoch": 1.5324894514767933,
+      "grad_norm": 0.4787490665912628,
+      "learning_rate": 0.00018260776939171814,
+      "loss": 0.7864061594009399,
+      "step": 3632
+    },
+    {
+      "epoch": 1.5333333333333332,
+      "grad_norm": 0.3670683205127716,
+      "learning_rate": 0.00018258069268594013,
+      "loss": 0.670537531375885,
+      "step": 3634
+    },
+    {
+      "epoch": 1.5341772151898734,
+      "grad_norm": 0.41599538922309875,
+      "learning_rate": 0.0001825535969303823,
+      "loss": 0.7535691261291504,
+      "step": 3636
+    },
+    {
+      "epoch": 1.5350210970464135,
+      "grad_norm": 0.5033245086669922,
+      "learning_rate": 0.00018252648213129514,
+      "loss": 0.7553712725639343,
+      "step": 3638
+    },
+    {
+      "epoch": 1.5358649789029535,
+      "grad_norm": 0.4235823452472687,
+      "learning_rate": 0.0001824993482949335,
+      "loss": 0.7149441242218018,
+      "step": 3640
+    },
+    {
+      "epoch": 1.5367088607594936,
+      "grad_norm": 0.37993329763412476,
+      "learning_rate": 0.00018247219542755664,
+      "loss": 0.735568642616272,
+      "step": 3642
+    },
+    {
+      "epoch": 1.5375527426160338,
+      "grad_norm": 0.39534425735473633,
+      "learning_rate": 0.00018244502353542818,
+      "loss": 0.693510115146637,
+      "step": 3644
+    },
+    {
+      "epoch": 1.5383966244725737,
+      "grad_norm": 0.45447441935539246,
+      "learning_rate": 0.00018241783262481622,
+      "loss": 0.7261984348297119,
+      "step": 3646
+    },
+    {
+      "epoch": 1.539240506329114,
+      "grad_norm": 0.4131409823894501,
+      "learning_rate": 0.0001823906227019931,
+      "loss": 0.7518824338912964,
+      "step": 3648
+    },
+    {
+      "epoch": 1.540084388185654,
+      "grad_norm": 0.46603885293006897,
+      "learning_rate": 0.00018236339377323567,
+      "loss": 0.7650659680366516,
+      "step": 3650
+    },
+    {
+      "epoch": 1.540928270042194,
+      "grad_norm": 0.3998747766017914,
+      "learning_rate": 0.00018233614584482513,
+      "loss": 0.6436064839363098,
+      "step": 3652
+    },
+    {
+      "epoch": 1.5417721518987342,
+      "grad_norm": 0.4314839839935303,
+      "learning_rate": 0.00018230887892304703,
+      "loss": 0.7068451642990112,
+      "step": 3654
+    },
+    {
+      "epoch": 1.5426160337552743,
+      "grad_norm": 0.4324610233306885,
+      "learning_rate": 0.0001822815930141913,
+      "loss": 0.7342977523803711,
+      "step": 3656
+    },
+    {
+      "epoch": 1.5434599156118143,
+      "grad_norm": 0.4385458827018738,
+      "learning_rate": 0.0001822542881245524,
+      "loss": 0.6482785940170288,
+      "step": 3658
+    },
+    {
+      "epoch": 1.5443037974683544,
+      "grad_norm": 0.3736587166786194,
+      "learning_rate": 0.0001822269642604289,
+      "loss": 0.6319621205329895,
+      "step": 3660
+    },
+    {
+      "epoch": 1.5451476793248946,
+      "grad_norm": 0.42113035917282104,
+      "learning_rate": 0.00018219962142812393,
+      "loss": 0.6603307723999023,
+      "step": 3662
+    },
+    {
+      "epoch": 1.5459915611814345,
+      "grad_norm": 0.422158420085907,
+      "learning_rate": 0.00018217225963394496,
+      "loss": 0.6874772906303406,
+      "step": 3664
+    },
+    {
+      "epoch": 1.5468354430379747,
+      "grad_norm": 0.42724940180778503,
+      "learning_rate": 0.00018214487888420388,
+      "loss": 0.7231854796409607,
+      "step": 3666
+    },
+    {
+      "epoch": 1.5476793248945149,
+      "grad_norm": 0.3747844398021698,
+      "learning_rate": 0.00018211747918521685,
+      "loss": 0.6418231129646301,
+      "step": 3668
+    },
+    {
+      "epoch": 1.5485232067510548,
+      "grad_norm": 0.37007269263267517,
+      "learning_rate": 0.00018209006054330446,
+      "loss": 0.7019823789596558,
+      "step": 3670
+    },
+    {
+      "epoch": 1.549367088607595,
+      "grad_norm": 0.409701943397522,
+      "learning_rate": 0.00018206262296479168,
+      "loss": 0.6632764339447021,
+      "step": 3672
+    },
+    {
+      "epoch": 1.5502109704641351,
+      "grad_norm": 0.49566221237182617,
+      "learning_rate": 0.00018203516645600782,
+      "loss": 0.7327075600624084,
+      "step": 3674
+    },
+    {
+      "epoch": 1.551054852320675,
+      "grad_norm": 0.4549940526485443,
+      "learning_rate": 0.0001820076910232866,
+      "loss": 0.6535149812698364,
+      "step": 3676
+    },
+    {
+      "epoch": 1.5518987341772152,
+      "grad_norm": 0.42619478702545166,
+      "learning_rate": 0.000181980196672966,
+      "loss": 0.7761320471763611,
+      "step": 3678
+    },
+    {
+      "epoch": 1.5527426160337554,
+      "grad_norm": 0.4332887828350067,
+      "learning_rate": 0.00018195268341138852,
+      "loss": 0.6728626489639282,
+      "step": 3680
+    },
+    {
+      "epoch": 1.5535864978902953,
+      "grad_norm": 0.4972975552082062,
+      "learning_rate": 0.0001819251512449009,
+      "loss": 0.7369232177734375,
+      "step": 3682
+    },
+    {
+      "epoch": 1.5544303797468353,
+      "grad_norm": 0.3619556725025177,
+      "learning_rate": 0.00018189760017985428,
+      "loss": 0.6703663468360901,
+      "step": 3684
+    },
+    {
+      "epoch": 1.5552742616033757,
+      "grad_norm": 0.46435829997062683,
+      "learning_rate": 0.00018187003022260417,
+      "loss": 0.7183756828308105,
+      "step": 3686
+    },
+    {
+      "epoch": 1.5561181434599156,
+      "grad_norm": 0.4605063498020172,
+      "learning_rate": 0.00018184244137951045,
+      "loss": 0.680042564868927,
+      "step": 3688
+    },
+    {
+      "epoch": 1.5569620253164556,
+      "grad_norm": 0.4023124873638153,
+      "learning_rate": 0.00018181483365693732,
+      "loss": 0.679821789264679,
+      "step": 3690
+    },
+    {
+      "epoch": 1.557805907172996,
+      "grad_norm": 0.459499716758728,
+      "learning_rate": 0.00018178720706125333,
+      "loss": 0.7408921718597412,
+      "step": 3692
+    },
+    {
+      "epoch": 1.5586497890295359,
+      "grad_norm": 0.4397135078907013,
+      "learning_rate": 0.00018175956159883146,
+      "loss": 0.6926741003990173,
+      "step": 3694
+    },
+    {
+      "epoch": 1.5594936708860758,
+      "grad_norm": 0.4476170837879181,
+      "learning_rate": 0.0001817318972760489,
+      "loss": 0.752724826335907,
+      "step": 3696
+    },
+    {
+      "epoch": 1.560337552742616,
+      "grad_norm": 0.43684154748916626,
+      "learning_rate": 0.00018170421409928735,
+      "loss": 0.7217722535133362,
+      "step": 3698
+    },
+    {
+      "epoch": 1.5611814345991561,
+      "grad_norm": 0.4118131995201111,
+      "learning_rate": 0.00018167651207493278,
+      "loss": 0.7264936566352844,
+      "step": 3700
+    },
+    {
+      "epoch": 1.5611814345991561,
+      "eval_loss": 0.7854447960853577,
+      "eval_runtime": 687.7907,
+      "eval_samples_per_second": 3.063,
+      "eval_steps_per_second": 3.063,
+      "step": 3700
+    },
+    {
+      "epoch": 1.562025316455696,
+      "grad_norm": 0.40639030933380127,
+      "learning_rate": 0.0001816487912093755,
+      "loss": 0.701478123664856,
+      "step": 3702
+    },
+    {
+      "epoch": 1.5628691983122363,
+      "grad_norm": 0.4205699563026428,
+      "learning_rate": 0.00018162105150901017,
+      "loss": 0.696711003780365,
+      "step": 3704
+    },
+    {
+      "epoch": 1.5637130801687764,
+      "grad_norm": 0.46437183022499084,
+      "learning_rate": 0.0001815932929802358,
+      "loss": 0.7130293250083923,
+      "step": 3706
+    },
+    {
+      "epoch": 1.5645569620253164,
+      "grad_norm": 0.4413023591041565,
+      "learning_rate": 0.0001815655156294558,
+      "loss": 0.7180310487747192,
+      "step": 3708
+    },
+    {
+      "epoch": 1.5654008438818565,
+      "grad_norm": 0.3964478075504303,
+      "learning_rate": 0.00018153771946307783,
+      "loss": 0.6883325576782227,
+      "step": 3710
+    },
+    {
+      "epoch": 1.5662447257383967,
+      "grad_norm": 0.34486109018325806,
+      "learning_rate": 0.00018150990448751394,
+      "loss": 0.6683162450790405,
+      "step": 3712
+    },
+    {
+      "epoch": 1.5670886075949366,
+      "grad_norm": 0.4209035336971283,
+      "learning_rate": 0.00018148207070918052,
+      "loss": 0.6969789862632751,
+      "step": 3714
+    },
+    {
+      "epoch": 1.5679324894514768,
+      "grad_norm": 0.41674816608428955,
+      "learning_rate": 0.00018145421813449829,
+      "loss": 0.714320182800293,
+      "step": 3716
+    },
+    {
+      "epoch": 1.568776371308017,
+      "grad_norm": 0.4365934729576111,
+      "learning_rate": 0.00018142634676989228,
+      "loss": 0.7125279307365417,
+      "step": 3718
+    },
+    {
+      "epoch": 1.5696202531645569,
+      "grad_norm": 0.43599721789360046,
+      "learning_rate": 0.00018139845662179188,
+      "loss": 0.6787398457527161,
+      "step": 3720
+    },
+    {
+      "epoch": 1.570464135021097,
+      "grad_norm": 0.36320003867149353,
+      "learning_rate": 0.0001813705476966308,
+      "loss": 0.6401328444480896,
+      "step": 3722
+    },
+    {
+      "epoch": 1.5713080168776372,
+      "grad_norm": 0.43504419922828674,
+      "learning_rate": 0.00018134262000084718,
+      "loss": 0.6875129342079163,
+      "step": 3724
+    },
+    {
+      "epoch": 1.5721518987341772,
+      "grad_norm": 0.476723313331604,
+      "learning_rate": 0.00018131467354088332,
+      "loss": 0.6903918385505676,
+      "step": 3726
+    },
+    {
+      "epoch": 1.5729957805907173,
+      "grad_norm": 0.4465550184249878,
+      "learning_rate": 0.00018128670832318592,
+      "loss": 0.690018892288208,
+      "step": 3728
+    },
+    {
+      "epoch": 1.5738396624472575,
+      "grad_norm": 0.43816477060317993,
+      "learning_rate": 0.0001812587243542061,
+      "loss": 0.6912158727645874,
+      "step": 3730
+    },
+    {
+      "epoch": 1.5746835443037974,
+      "grad_norm": 0.40546250343322754,
+      "learning_rate": 0.00018123072164039913,
+      "loss": 0.7415492534637451,
+      "step": 3732
+    },
+    {
+      "epoch": 1.5755274261603376,
+      "grad_norm": 0.4253178536891937,
+      "learning_rate": 0.00018120270018822477,
+      "loss": 0.7664727568626404,
+      "step": 3734
+    },
+    {
+      "epoch": 1.5763713080168777,
+      "grad_norm": 0.503524661064148,
+      "learning_rate": 0.000181174660004147,
+      "loss": 0.782779335975647,
+      "step": 3736
+    },
+    {
+      "epoch": 1.5772151898734177,
+      "grad_norm": 0.5260395407676697,
+      "learning_rate": 0.00018114660109463414,
+      "loss": 0.806460976600647,
+      "step": 3738
+    },
+    {
+      "epoch": 1.5780590717299579,
+      "grad_norm": 0.3641359806060791,
+      "learning_rate": 0.0001811185234661589,
+      "loss": 0.6859704852104187,
+      "step": 3740
+    },
+    {
+      "epoch": 1.578902953586498,
+      "grad_norm": 0.4468916058540344,
+      "learning_rate": 0.00018109042712519818,
+      "loss": 0.7798553109169006,
+      "step": 3742
+    },
+    {
+      "epoch": 1.579746835443038,
+      "grad_norm": 0.4078207015991211,
+      "learning_rate": 0.00018106231207823327,
+      "loss": 0.7086471915245056,
+      "step": 3744
+    },
+    {
+      "epoch": 1.580590717299578,
+      "grad_norm": 0.39144742488861084,
+      "learning_rate": 0.00018103417833174981,
+      "loss": 0.6669760346412659,
+      "step": 3746
+    },
+    {
+      "epoch": 1.5814345991561183,
+      "grad_norm": 0.4366845190525055,
+      "learning_rate": 0.0001810060258922377,
+      "loss": 0.7296954989433289,
+      "step": 3748
+    },
+    {
+      "epoch": 1.5822784810126582,
+      "grad_norm": 0.46795082092285156,
+      "learning_rate": 0.00018097785476619118,
+      "loss": 0.7663856148719788,
+      "step": 3750
+    },
+    {
+      "epoch": 1.5831223628691982,
+      "grad_norm": 0.40331873297691345,
+      "learning_rate": 0.00018094966496010878,
+      "loss": 0.7105234861373901,
+      "step": 3752
+    },
+    {
+      "epoch": 1.5839662447257385,
+      "grad_norm": 0.38210469484329224,
+      "learning_rate": 0.00018092145648049334,
+      "loss": 0.666635274887085,
+      "step": 3754
+    },
+    {
+      "epoch": 1.5848101265822785,
+      "grad_norm": 0.45946040749549866,
+      "learning_rate": 0.000180893229333852,
+      "loss": 0.7287175059318542,
+      "step": 3756
+    },
+    {
+      "epoch": 1.5856540084388184,
+      "grad_norm": 0.4035777747631073,
+      "learning_rate": 0.00018086498352669624,
+      "loss": 0.7145618796348572,
+      "step": 3758
+    },
+    {
+      "epoch": 1.5864978902953588,
+      "grad_norm": 0.42843320965766907,
+      "learning_rate": 0.0001808367190655418,
+      "loss": 0.6550979018211365,
+      "step": 3760
+    },
+    {
+      "epoch": 1.5873417721518988,
+      "grad_norm": 0.4016468822956085,
+      "learning_rate": 0.0001808084359569088,
+      "loss": 0.7040430307388306,
+      "step": 3762
+    },
+    {
+      "epoch": 1.5881856540084387,
+      "grad_norm": 0.4126848578453064,
+      "learning_rate": 0.00018078013420732157,
+      "loss": 0.7523179054260254,
+      "step": 3764
+    },
+    {
+      "epoch": 1.5890295358649789,
+      "grad_norm": 0.45878592133522034,
+      "learning_rate": 0.00018075181382330878,
+      "loss": 0.7497967481613159,
+      "step": 3766
+    },
+    {
+      "epoch": 1.589873417721519,
+      "grad_norm": 0.44179272651672363,
+      "learning_rate": 0.00018072347481140345,
+      "loss": 0.7694459557533264,
+      "step": 3768
+    },
+    {
+      "epoch": 1.590717299578059,
+      "grad_norm": 0.3706412613391876,
+      "learning_rate": 0.00018069511717814276,
+      "loss": 0.6766634583473206,
+      "step": 3770
+    },
+    {
+      "epoch": 1.5915611814345991,
+      "grad_norm": 0.40555453300476074,
+      "learning_rate": 0.0001806667409300683,
+      "loss": 0.6394028067588806,
+      "step": 3772
+    },
+    {
+      "epoch": 1.5924050632911393,
+      "grad_norm": 0.3752219080924988,
+      "learning_rate": 0.000180638346073726,
+      "loss": 0.6775637865066528,
+      "step": 3774
+    },
+    {
+      "epoch": 1.5932489451476792,
+      "grad_norm": 0.40960127115249634,
+      "learning_rate": 0.00018060993261566593,
+      "loss": 0.7200632095336914,
+      "step": 3776
+    },
+    {
+      "epoch": 1.5940928270042194,
+      "grad_norm": 0.4215535521507263,
+      "learning_rate": 0.00018058150056244254,
+      "loss": 0.7314276695251465,
+      "step": 3778
+    },
+    {
+      "epoch": 1.5949367088607596,
+      "grad_norm": 0.42734605073928833,
+      "learning_rate": 0.00018055304992061456,
+      "loss": 0.7954897284507751,
+      "step": 3780
+    },
+    {
+      "epoch": 1.5957805907172995,
+      "grad_norm": 0.3980397582054138,
+      "learning_rate": 0.00018052458069674505,
+      "loss": 0.6552245020866394,
+      "step": 3782
+    },
+    {
+      "epoch": 1.5966244725738397,
+      "grad_norm": 0.44602933526039124,
+      "learning_rate": 0.00018049609289740124,
+      "loss": 0.7638092041015625,
+      "step": 3784
+    },
+    {
+      "epoch": 1.5974683544303798,
+      "grad_norm": 0.43835216760635376,
+      "learning_rate": 0.00018046758652915474,
+      "loss": 0.6886481046676636,
+      "step": 3786
+    },
+    {
+      "epoch": 1.5983122362869198,
+      "grad_norm": 0.38229402899742126,
+      "learning_rate": 0.00018043906159858147,
+      "loss": 0.6726123690605164,
+      "step": 3788
+    },
+    {
+      "epoch": 1.59915611814346,
+      "grad_norm": 0.43049392104148865,
+      "learning_rate": 0.00018041051811226155,
+      "loss": 0.7416896224021912,
+      "step": 3790
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 0.4027816951274872,
+      "learning_rate": 0.00018038195607677943,
+      "loss": 0.682691752910614,
+      "step": 3792
+    },
+    {
+      "epoch": 1.60084388185654,
+      "grad_norm": 0.4143599569797516,
+      "learning_rate": 0.00018035337549872376,
+      "loss": 0.7199558615684509,
+      "step": 3794
+    },
+    {
+      "epoch": 1.6016877637130802,
+      "grad_norm": 0.41221415996551514,
+      "learning_rate": 0.00018032477638468762,
+      "loss": 0.6586267948150635,
+      "step": 3796
+    },
+    {
+      "epoch": 1.6025316455696204,
+      "grad_norm": 0.40655994415283203,
+      "learning_rate": 0.0001802961587412682,
+      "loss": 0.7579189538955688,
+      "step": 3798
+    },
+    {
+      "epoch": 1.6033755274261603,
+      "grad_norm": 0.4163990914821625,
+      "learning_rate": 0.00018026752257506708,
+      "loss": 0.7246071100234985,
+      "step": 3800
+    },
+    {
+      "epoch": 1.6033755274261603,
+      "eval_loss": 0.778353214263916,
+      "eval_runtime": 692.5522,
+      "eval_samples_per_second": 3.042,
+      "eval_steps_per_second": 3.042,
+      "step": 3800
+    },
+    {
+      "epoch": 1.6042194092827005,
+      "grad_norm": 0.378628671169281,
+      "learning_rate": 0.0001802388678926901,
+      "loss": 0.7114223837852478,
+      "step": 3802
+    },
+    {
+      "epoch": 1.6050632911392406,
+      "grad_norm": 0.5050750374794006,
+      "learning_rate": 0.0001802101947007473,
+      "loss": 0.7519527077674866,
+      "step": 3804
+    },
+    {
+      "epoch": 1.6059071729957806,
+      "grad_norm": 0.5003950595855713,
+      "learning_rate": 0.00018018150300585305,
+      "loss": 0.7553771138191223,
+      "step": 3806
+    },
+    {
+      "epoch": 1.6067510548523207,
+      "grad_norm": 0.3851728141307831,
+      "learning_rate": 0.00018015279281462595,
+      "loss": 0.6584326028823853,
+      "step": 3808
+    },
+    {
+      "epoch": 1.6075949367088609,
+      "grad_norm": 0.41256240010261536,
+      "learning_rate": 0.00018012406413368893,
+      "loss": 0.7215287089347839,
+      "step": 3810
+    },
+    {
+      "epoch": 1.6084388185654008,
+      "grad_norm": 0.41479718685150146,
+      "learning_rate": 0.00018009531696966912,
+      "loss": 0.7753872275352478,
+      "step": 3812
+    },
+    {
+      "epoch": 1.6092827004219408,
+      "grad_norm": 0.398509681224823,
+      "learning_rate": 0.00018006655132919797,
+      "loss": 0.6984832882881165,
+      "step": 3814
+    },
+    {
+      "epoch": 1.6101265822784812,
+      "grad_norm": 0.42210307717323303,
+      "learning_rate": 0.0001800377672189111,
+      "loss": 0.8117644786834717,
+      "step": 3816
+    },
+    {
+      "epoch": 1.610970464135021,
+      "grad_norm": 0.3714556396007538,
+      "learning_rate": 0.0001800089646454485,
+      "loss": 0.6302977204322815,
+      "step": 3818
+    },
+    {
+      "epoch": 1.611814345991561,
+      "grad_norm": 0.4478890299797058,
+      "learning_rate": 0.00017998014361545438,
+      "loss": 0.7327325940132141,
+      "step": 3820
+    },
+    {
+      "epoch": 1.6126582278481014,
+      "grad_norm": 0.43516477942466736,
+      "learning_rate": 0.00017995130413557713,
+      "loss": 0.6877713203430176,
+      "step": 3822
+    },
+    {
+      "epoch": 1.6135021097046414,
+      "grad_norm": 0.42912542819976807,
+      "learning_rate": 0.0001799224462124695,
+      "loss": 0.6797093152999878,
+      "step": 3824
+    },
+    {
+      "epoch": 1.6143459915611813,
+      "grad_norm": 0.40646228194236755,
+      "learning_rate": 0.00017989356985278853,
+      "loss": 0.7402470111846924,
+      "step": 3826
+    },
+    {
+      "epoch": 1.6151898734177215,
+      "grad_norm": 0.42576655745506287,
+      "learning_rate": 0.00017986467506319535,
+      "loss": 0.6698583364486694,
+      "step": 3828
+    },
+    {
+      "epoch": 1.6160337552742616,
+      "grad_norm": 0.4298654794692993,
+      "learning_rate": 0.0001798357618503555,
+      "loss": 0.6736086010932922,
+      "step": 3830
+    },
+    {
+      "epoch": 1.6168776371308016,
+      "grad_norm": 0.4205976128578186,
+      "learning_rate": 0.0001798068302209386,
+      "loss": 0.7724249958992004,
+      "step": 3832
+    },
+    {
+      "epoch": 1.6177215189873417,
+      "grad_norm": 0.39669108390808105,
+      "learning_rate": 0.00017977788018161872,
+      "loss": 0.6468521952629089,
+      "step": 3834
+    },
+    {
+      "epoch": 1.618565400843882,
+      "grad_norm": 0.428551584482193,
+      "learning_rate": 0.00017974891173907405,
+      "loss": 0.7086918950080872,
+      "step": 3836
+    },
+    {
+      "epoch": 1.6194092827004218,
+      "grad_norm": 0.43773409724235535,
+      "learning_rate": 0.00017971992489998703,
+      "loss": 0.6483711004257202,
+      "step": 3838
+    },
+    {
+      "epoch": 1.620253164556962,
+      "grad_norm": 0.40540891885757446,
+      "learning_rate": 0.00017969091967104438,
+      "loss": 0.6947301030158997,
+      "step": 3840
+    },
+    {
+      "epoch": 1.6210970464135022,
+      "grad_norm": 0.3685019016265869,
+      "learning_rate": 0.00017966189605893708,
+      "loss": 0.667517900466919,
+      "step": 3842
+    },
+    {
+      "epoch": 1.621940928270042,
+      "grad_norm": 0.44813284277915955,
+      "learning_rate": 0.0001796328540703603,
+      "loss": 0.6957448124885559,
+      "step": 3844
+    },
+    {
+      "epoch": 1.6227848101265823,
+      "grad_norm": 0.4524349868297577,
+      "learning_rate": 0.00017960379371201345,
+      "loss": 0.7735851407051086,
+      "step": 3846
+    },
+    {
+      "epoch": 1.6236286919831224,
+      "grad_norm": 0.42349138855934143,
+      "learning_rate": 0.0001795747149906002,
+      "loss": 0.7052056193351746,
+      "step": 3848
+    },
+    {
+      "epoch": 1.6244725738396624,
+      "grad_norm": 0.47317537665367126,
+      "learning_rate": 0.0001795456179128285,
+      "loss": 0.7402311563491821,
+      "step": 3850
+    },
+    {
+      "epoch": 1.6253164556962025,
+      "grad_norm": 0.38915711641311646,
+      "learning_rate": 0.0001795165024854104,
+      "loss": 0.6743243932723999,
+      "step": 3852
+    },
+    {
+      "epoch": 1.6261603375527427,
+      "grad_norm": 0.4430672526359558,
+      "learning_rate": 0.00017948736871506235,
+      "loss": 0.6983860731124878,
+      "step": 3854
+    },
+    {
+      "epoch": 1.6270042194092826,
+      "grad_norm": 0.44286689162254333,
+      "learning_rate": 0.00017945821660850494,
+      "loss": 0.8012514710426331,
+      "step": 3856
+    },
+    {
+      "epoch": 1.6278481012658228,
+      "grad_norm": 0.43364086747169495,
+      "learning_rate": 0.00017942904617246295,
+      "loss": 0.8064966201782227,
+      "step": 3858
+    },
+    {
+      "epoch": 1.628691983122363,
+      "grad_norm": 0.45848122239112854,
+      "learning_rate": 0.00017939985741366547,
+      "loss": 0.7069969177246094,
+      "step": 3860
+    },
+    {
+      "epoch": 1.629535864978903,
+      "grad_norm": 0.4199717044830322,
+      "learning_rate": 0.0001793706503388458,
+      "loss": 0.6722052097320557,
+      "step": 3862
+    },
+    {
+      "epoch": 1.630379746835443,
+      "grad_norm": 0.3820365071296692,
+      "learning_rate": 0.00017934142495474142,
+      "loss": 0.663442075252533,
+      "step": 3864
+    },
+    {
+      "epoch": 1.6312236286919832,
+      "grad_norm": 0.4470944106578827,
+      "learning_rate": 0.00017931218126809412,
+      "loss": 0.70677250623703,
+      "step": 3866
+    },
+    {
+      "epoch": 1.6320675105485232,
+      "grad_norm": 0.3941600024700165,
+      "learning_rate": 0.0001792829192856498,
+      "loss": 0.7125341892242432,
+      "step": 3868
+    },
+    {
+      "epoch": 1.6329113924050633,
+      "grad_norm": 0.44924396276474,
+      "learning_rate": 0.00017925363901415863,
+      "loss": 0.7245699763298035,
+      "step": 3870
+    },
+    {
+      "epoch": 1.6337552742616035,
+      "grad_norm": 0.4070027470588684,
+      "learning_rate": 0.00017922434046037508,
+      "loss": 0.6977120637893677,
+      "step": 3872
+    },
+    {
+      "epoch": 1.6345991561181434,
+      "grad_norm": 0.45040908455848694,
+      "learning_rate": 0.00017919502363105772,
+      "loss": 0.7631482481956482,
+      "step": 3874
+    },
+    {
+      "epoch": 1.6354430379746834,
+      "grad_norm": 0.4083724617958069,
+      "learning_rate": 0.00017916568853296934,
+      "loss": 0.6656452417373657,
+      "step": 3876
+    },
+    {
+      "epoch": 1.6362869198312238,
+      "grad_norm": 0.3762308359146118,
+      "learning_rate": 0.00017913633517287704,
+      "loss": 0.6302213072776794,
+      "step": 3878
+    },
+    {
+      "epoch": 1.6371308016877637,
+      "grad_norm": 0.36366912722587585,
+      "learning_rate": 0.00017910696355755205,
+      "loss": 0.7119566202163696,
+      "step": 3880
+    },
+    {
+      "epoch": 1.6379746835443036,
+      "grad_norm": 0.43168532848358154,
+      "learning_rate": 0.00017907757369376985,
+      "loss": 0.7271678447723389,
+      "step": 3882
+    },
+    {
+      "epoch": 1.638818565400844,
+      "grad_norm": 0.37720009684562683,
+      "learning_rate": 0.00017904816558831014,
+      "loss": 0.6340095400810242,
+      "step": 3884
+    },
+    {
+      "epoch": 1.639662447257384,
+      "grad_norm": 0.36539995670318604,
+      "learning_rate": 0.00017901873924795677,
+      "loss": 0.6910572052001953,
+      "step": 3886
+    },
+    {
+      "epoch": 1.640506329113924,
+      "grad_norm": 0.42437905073165894,
+      "learning_rate": 0.00017898929467949782,
+      "loss": 0.7570282220840454,
+      "step": 3888
+    },
+    {
+      "epoch": 1.6413502109704643,
+      "grad_norm": 0.43482181429862976,
+      "learning_rate": 0.00017895983188972558,
+      "loss": 0.7835264205932617,
+      "step": 3890
+    },
+    {
+      "epoch": 1.6421940928270042,
+      "grad_norm": 0.40460237860679626,
+      "learning_rate": 0.00017893035088543662,
+      "loss": 0.7058968544006348,
+      "step": 3892
+    },
+    {
+      "epoch": 1.6430379746835442,
+      "grad_norm": 0.39779701828956604,
+      "learning_rate": 0.00017890085167343158,
+      "loss": 0.723969042301178,
+      "step": 3894
+    },
+    {
+      "epoch": 1.6438818565400843,
+      "grad_norm": 0.46131494641304016,
+      "learning_rate": 0.00017887133426051536,
+      "loss": 0.7773922681808472,
+      "step": 3896
+    },
+    {
+      "epoch": 1.6447257383966245,
+      "grad_norm": 0.48412877321243286,
+      "learning_rate": 0.0001788417986534971,
+      "loss": 0.7742162942886353,
+      "step": 3898
+    },
+    {
+      "epoch": 1.6455696202531644,
+      "grad_norm": 0.40083935856819153,
+      "learning_rate": 0.00017881224485919006,
+      "loss": 0.6674869656562805,
+      "step": 3900
+    },
+    {
+      "epoch": 1.6455696202531644,
+      "eval_loss": 0.7763351202011108,
+      "eval_runtime": 682.0824,
+      "eval_samples_per_second": 3.089,
+      "eval_steps_per_second": 3.089,
+      "step": 3900
+    },
+    {
+      "epoch": 1.6464135021097046,
+      "grad_norm": 0.3944476842880249,
+      "learning_rate": 0.00017878267288441176,
+      "loss": 0.6777821183204651,
+      "step": 3902
+    },
+    {
+      "epoch": 1.6472573839662448,
+      "grad_norm": 0.44863224029541016,
+      "learning_rate": 0.0001787530827359839,
+      "loss": 0.7398322224617004,
+      "step": 3904
+    },
+    {
+      "epoch": 1.6481012658227847,
+      "grad_norm": 0.4665529131889343,
+      "learning_rate": 0.00017872347442073232,
+      "loss": 0.7203068733215332,
+      "step": 3906
+    },
+    {
+      "epoch": 1.6489451476793249,
+      "grad_norm": 0.3603050410747528,
+      "learning_rate": 0.00017869384794548708,
+      "loss": 0.6232115030288696,
+      "step": 3908
+    },
+    {
+      "epoch": 1.649789029535865,
+      "grad_norm": 0.3545325696468353,
+      "learning_rate": 0.0001786642033170825,
+      "loss": 0.6120040416717529,
+      "step": 3910
+    },
+    {
+      "epoch": 1.650632911392405,
+      "grad_norm": 0.3947487473487854,
+      "learning_rate": 0.00017863454054235697,
+      "loss": 0.7285997867584229,
+      "step": 3912
+    },
+    {
+      "epoch": 1.6514767932489451,
+      "grad_norm": 0.33153054118156433,
+      "learning_rate": 0.00017860485962815312,
+      "loss": 0.6486109495162964,
+      "step": 3914
+    },
+    {
+      "epoch": 1.6523206751054853,
+      "grad_norm": 0.3886185884475708,
+      "learning_rate": 0.00017857516058131782,
+      "loss": 0.6870376467704773,
+      "step": 3916
+    },
+    {
+      "epoch": 1.6531645569620252,
+      "grad_norm": 0.46399059891700745,
+      "learning_rate": 0.00017854544340870201,
+      "loss": 0.7351298928260803,
+      "step": 3918
+    },
+    {
+      "epoch": 1.6540084388185654,
+      "grad_norm": 0.49280229210853577,
+      "learning_rate": 0.00017851570811716093,
+      "loss": 0.7283915281295776,
+      "step": 3920
+    },
+    {
+      "epoch": 1.6548523206751056,
+      "grad_norm": 0.41570642590522766,
+      "learning_rate": 0.00017848595471355388,
+      "loss": 0.7436906099319458,
+      "step": 3922
+    },
+    {
+      "epoch": 1.6556962025316455,
+      "grad_norm": 0.44591766595840454,
+      "learning_rate": 0.00017845618320474444,
+      "loss": 0.7427505254745483,
+      "step": 3924
+    },
+    {
+      "epoch": 1.6565400843881857,
+      "grad_norm": 0.4804318845272064,
+      "learning_rate": 0.00017842639359760032,
+      "loss": 0.7642725706100464,
+      "step": 3926
+    },
+    {
+      "epoch": 1.6573839662447258,
+      "grad_norm": 0.4164349138736725,
+      "learning_rate": 0.00017839658589899343,
+      "loss": 0.7097790837287903,
+      "step": 3928
+    },
+    {
+      "epoch": 1.6582278481012658,
+      "grad_norm": 0.5307883620262146,
+      "learning_rate": 0.00017836676011579976,
+      "loss": 0.8346033096313477,
+      "step": 3930
+    },
+    {
+      "epoch": 1.659071729957806,
+      "grad_norm": 0.37231504917144775,
+      "learning_rate": 0.0001783369162548996,
+      "loss": 0.6327160596847534,
+      "step": 3932
+    },
+    {
+      "epoch": 1.659915611814346,
+      "grad_norm": 0.4178757965564728,
+      "learning_rate": 0.00017830705432317738,
+      "loss": 0.6441237330436707,
+      "step": 3934
+    },
+    {
+      "epoch": 1.660759493670886,
+      "grad_norm": 0.42698073387145996,
+      "learning_rate": 0.00017827717432752162,
+      "loss": 0.7267951369285583,
+      "step": 3936
+    },
+    {
+      "epoch": 1.6616033755274262,
+      "grad_norm": 0.48763108253479004,
+      "learning_rate": 0.0001782472762748251,
+      "loss": 0.7248380780220032,
+      "step": 3938
+    },
+    {
+      "epoch": 1.6624472573839664,
+      "grad_norm": 0.38882505893707275,
+      "learning_rate": 0.0001782173601719847,
+      "loss": 0.6665275692939758,
+      "step": 3940
+    },
+    {
+      "epoch": 1.6632911392405063,
+      "grad_norm": 0.4684160649776459,
+      "learning_rate": 0.0001781874260259015,
+      "loss": 0.8017846941947937,
+      "step": 3942
+    },
+    {
+      "epoch": 1.6641350210970463,
+      "grad_norm": 0.4200204908847809,
+      "learning_rate": 0.00017815747384348075,
+      "loss": 0.6810902953147888,
+      "step": 3944
+    },
+    {
+      "epoch": 1.6649789029535866,
+      "grad_norm": 0.45843279361724854,
+      "learning_rate": 0.0001781275036316318,
+      "loss": 0.714082658290863,
+      "step": 3946
+    },
+    {
+      "epoch": 1.6658227848101266,
+      "grad_norm": 0.522654116153717,
+      "learning_rate": 0.00017809751539726824,
+      "loss": 0.842085599899292,
+      "step": 3948
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 0.43468809127807617,
+      "learning_rate": 0.00017806750914730777,
+      "loss": 0.6917972564697266,
+      "step": 3950
+    },
+    {
+      "epoch": 1.667510548523207,
+      "grad_norm": 0.4071558117866516,
+      "learning_rate": 0.00017803748488867224,
+      "loss": 0.6676424741744995,
+      "step": 3952
+    },
+    {
+      "epoch": 1.6683544303797468,
+      "grad_norm": 0.4478702247142792,
+      "learning_rate": 0.00017800744262828772,
+      "loss": 0.7563748359680176,
+      "step": 3954
+    },
+    {
+      "epoch": 1.6691983122362868,
+      "grad_norm": 0.42986053228378296,
+      "learning_rate": 0.00017797738237308432,
+      "loss": 0.685714066028595,
+      "step": 3956
+    },
+    {
+      "epoch": 1.6700421940928272,
+      "grad_norm": 0.4410569667816162,
+      "learning_rate": 0.0001779473041299964,
+      "loss": 0.7753605842590332,
+      "step": 3958
+    },
+    {
+      "epoch": 1.6708860759493671,
+      "grad_norm": 0.44055965542793274,
+      "learning_rate": 0.00017791720790596242,
+      "loss": 0.7826018333435059,
+      "step": 3960
+    },
+    {
+      "epoch": 1.671729957805907,
+      "grad_norm": 0.4660222828388214,
+      "learning_rate": 0.00017788709370792502,
+      "loss": 0.7012629508972168,
+      "step": 3962
+    },
+    {
+      "epoch": 1.6725738396624472,
+      "grad_norm": 0.3655732274055481,
+      "learning_rate": 0.00017785696154283093,
+      "loss": 0.6647007465362549,
+      "step": 3964
+    },
+    {
+      "epoch": 1.6734177215189874,
+      "grad_norm": 0.45735979080200195,
+      "learning_rate": 0.0001778268114176311,
+      "loss": 0.7238346338272095,
+      "step": 3966
+    },
+    {
+      "epoch": 1.6742616033755273,
+      "grad_norm": 0.4894043207168579,
+      "learning_rate": 0.00017779664333928053,
+      "loss": 0.7595300078392029,
+      "step": 3968
+    },
+    {
+      "epoch": 1.6751054852320675,
+      "grad_norm": 0.3613647222518921,
+      "learning_rate": 0.00017776645731473848,
+      "loss": 0.6534504294395447,
+      "step": 3970
+    },
+    {
+      "epoch": 1.6759493670886076,
+      "grad_norm": 0.4402320683002472,
+      "learning_rate": 0.00017773625335096828,
+      "loss": 0.7699148058891296,
+      "step": 3972
+    },
+    {
+      "epoch": 1.6767932489451476,
+      "grad_norm": 0.4391469955444336,
+      "learning_rate": 0.0001777060314549374,
+      "loss": 0.771800696849823,
+      "step": 3974
+    },
+    {
+      "epoch": 1.6776371308016877,
+      "grad_norm": 0.41348180174827576,
+      "learning_rate": 0.00017767579163361735,
+      "loss": 0.6957393884658813,
+      "step": 3976
+    },
+    {
+      "epoch": 1.678481012658228,
+      "grad_norm": 0.4115046560764313,
+      "learning_rate": 0.00017764553389398407,
+      "loss": 0.6666664481163025,
+      "step": 3978
+    },
+    {
+      "epoch": 1.6793248945147679,
+      "grad_norm": 0.44647306203842163,
+      "learning_rate": 0.00017761525824301725,
+      "loss": 0.7364352345466614,
+      "step": 3980
+    },
+    {
+      "epoch": 1.680168776371308,
+      "grad_norm": 0.424234002828598,
+      "learning_rate": 0.00017758496468770103,
+      "loss": 0.7229774594306946,
+      "step": 3982
+    },
+    {
+      "epoch": 1.6810126582278482,
+      "grad_norm": 0.4850127696990967,
+      "learning_rate": 0.00017755465323502346,
+      "loss": 0.6992678642272949,
+      "step": 3984
+    },
+    {
+      "epoch": 1.6818565400843881,
+      "grad_norm": 0.44637173414230347,
+      "learning_rate": 0.00017752432389197688,
+      "loss": 0.6957634091377258,
+      "step": 3986
+    },
+    {
+      "epoch": 1.6827004219409283,
+      "grad_norm": 0.477393239736557,
+      "learning_rate": 0.00017749397666555768,
+      "loss": 0.6842859983444214,
+      "step": 3988
+    },
+    {
+      "epoch": 1.6835443037974684,
+      "grad_norm": 0.3634146451950073,
+      "learning_rate": 0.00017746361156276632,
+      "loss": 0.6478199362754822,
+      "step": 3990
+    },
+    {
+      "epoch": 1.6843881856540084,
+      "grad_norm": 0.4806389808654785,
+      "learning_rate": 0.00017743322859060752,
+      "loss": 0.7166973352432251,
+      "step": 3992
+    },
+    {
+      "epoch": 1.6852320675105485,
+      "grad_norm": 0.4273380935192108,
+      "learning_rate": 0.00017740282775609,
+      "loss": 0.6624643206596375,
+      "step": 3994
+    },
+    {
+      "epoch": 1.6860759493670887,
+      "grad_norm": 0.4340716600418091,
+      "learning_rate": 0.0001773724090662267,
+      "loss": 0.6408489346504211,
+      "step": 3996
+    },
+    {
+      "epoch": 1.6869198312236287,
+      "grad_norm": 0.50480717420578,
+      "learning_rate": 0.00017734197252803458,
+      "loss": 0.8096797466278076,
+      "step": 3998
+    },
+    {
+      "epoch": 1.6877637130801688,
+      "grad_norm": 0.4549327790737152,
+      "learning_rate": 0.00017731151814853475,
+      "loss": 0.6902434825897217,
+      "step": 4000
+    },
+    {
+      "epoch": 1.6877637130801688,
+      "eval_loss": 0.7721371650695801,
+      "eval_runtime": 668.395,
+      "eval_samples_per_second": 3.152,
+      "eval_steps_per_second": 3.152,
+      "step": 4000
+    },
+    {
+      "epoch": 1.688607594936709,
+      "grad_norm": 0.38909077644348145,
+      "learning_rate": 0.00017728104593475247,
+      "loss": 0.6510542035102844,
+      "step": 4002
+    },
+    {
+      "epoch": 1.689451476793249,
+      "grad_norm": 0.444608598947525,
+      "learning_rate": 0.00017725055589371715,
+      "loss": 0.7331066727638245,
+      "step": 4004
+    },
+    {
+      "epoch": 1.690295358649789,
+      "grad_norm": 0.3364178538322449,
+      "learning_rate": 0.0001772200480324622,
+      "loss": 0.5950609445571899,
+      "step": 4006
+    },
+    {
+      "epoch": 1.6911392405063292,
+      "grad_norm": 0.37264227867126465,
+      "learning_rate": 0.0001771895223580252,
+      "loss": 0.7024590969085693,
+      "step": 4008
+    },
+    {
+      "epoch": 1.6919831223628692,
+      "grad_norm": 0.3576899766921997,
+      "learning_rate": 0.00017715897887744784,
+      "loss": 0.6620895266532898,
+      "step": 4010
+    },
+    {
+      "epoch": 1.6928270042194091,
+      "grad_norm": 0.45855066180229187,
+      "learning_rate": 0.00017712841759777591,
+      "loss": 0.6726759076118469,
+      "step": 4012
+    },
+    {
+      "epoch": 1.6936708860759495,
+      "grad_norm": 0.44610634446144104,
+      "learning_rate": 0.00017709783852605933,
+      "loss": 0.6550953984260559,
+      "step": 4014
+    },
+    {
+      "epoch": 1.6945147679324895,
+      "grad_norm": 0.4262036681175232,
+      "learning_rate": 0.00017706724166935208,
+      "loss": 0.7338476777076721,
+      "step": 4016
+    },
+    {
+      "epoch": 1.6953586497890294,
+      "grad_norm": 0.44584009051322937,
+      "learning_rate": 0.0001770366270347123,
+      "loss": 0.6785868406295776,
+      "step": 4018
+    },
+    {
+      "epoch": 1.6962025316455698,
+      "grad_norm": 0.44828134775161743,
+      "learning_rate": 0.0001770059946292022,
+      "loss": 0.7113003134727478,
+      "step": 4020
+    },
+    {
+      "epoch": 1.6970464135021097,
+      "grad_norm": 0.39716270565986633,
+      "learning_rate": 0.00017697534445988803,
+      "loss": 0.6420866847038269,
+      "step": 4022
+    },
+    {
+      "epoch": 1.6978902953586497,
+      "grad_norm": 0.43546608090400696,
+      "learning_rate": 0.00017694467653384024,
+      "loss": 0.8087385892868042,
+      "step": 4024
+    },
+    {
+      "epoch": 1.6987341772151898,
+      "grad_norm": 0.3698647618293762,
+      "learning_rate": 0.00017691399085813334,
+      "loss": 0.6378461718559265,
+      "step": 4026
+    },
+    {
+      "epoch": 1.69957805907173,
+      "grad_norm": 0.4256688058376312,
+      "learning_rate": 0.0001768832874398459,
+      "loss": 0.6504215002059937,
+      "step": 4028
+    },
+    {
+      "epoch": 1.70042194092827,
+      "grad_norm": 0.4014805555343628,
+      "learning_rate": 0.00017685256628606063,
+      "loss": 0.6648795008659363,
+      "step": 4030
+    },
+    {
+      "epoch": 1.70126582278481,
+      "grad_norm": 0.43924233317375183,
+      "learning_rate": 0.00017682182740386427,
+      "loss": 0.6905859112739563,
+      "step": 4032
+    },
+    {
+      "epoch": 1.7021097046413503,
+      "grad_norm": 0.43565094470977783,
+      "learning_rate": 0.00017679107080034774,
+      "loss": 0.7800544500350952,
+      "step": 4034
+    },
+    {
+      "epoch": 1.7029535864978902,
+      "grad_norm": 0.4681089520454407,
+      "learning_rate": 0.00017676029648260597,
+      "loss": 0.7487648725509644,
+      "step": 4036
+    },
+    {
+      "epoch": 1.7037974683544304,
+      "grad_norm": 0.41400185227394104,
+      "learning_rate": 0.00017672950445773804,
+      "loss": 0.7374032735824585,
+      "step": 4038
+    },
+    {
+      "epoch": 1.7046413502109705,
+      "grad_norm": 0.5120159983634949,
+      "learning_rate": 0.00017669869473284702,
+      "loss": 0.7539847493171692,
+      "step": 4040
+    },
+    {
+      "epoch": 1.7054852320675105,
+      "grad_norm": 0.4043315052986145,
+      "learning_rate": 0.00017666786731504013,
+      "loss": 0.7029932737350464,
+      "step": 4042
+    },
+    {
+      "epoch": 1.7063291139240506,
+      "grad_norm": 0.462801992893219,
+      "learning_rate": 0.00017663702221142874,
+      "loss": 0.6464276313781738,
+      "step": 4044
+    },
+    {
+      "epoch": 1.7071729957805908,
+      "grad_norm": 0.4751932621002197,
+      "learning_rate": 0.00017660615942912813,
+      "loss": 0.7646113634109497,
+      "step": 4046
+    },
+    {
+      "epoch": 1.7080168776371307,
+      "grad_norm": 0.41720855236053467,
+      "learning_rate": 0.0001765752789752578,
+      "loss": 0.7191748023033142,
+      "step": 4048
+    },
+    {
+      "epoch": 1.7088607594936709,
+      "grad_norm": 0.40235280990600586,
+      "learning_rate": 0.00017654438085694128,
+      "loss": 0.6946488618850708,
+      "step": 4050
+    },
+    {
+      "epoch": 1.709704641350211,
+      "grad_norm": 0.4385916292667389,
+      "learning_rate": 0.00017651346508130612,
+      "loss": 0.7655195593833923,
+      "step": 4052
+    },
+    {
+      "epoch": 1.710548523206751,
+      "grad_norm": 0.4192262887954712,
+      "learning_rate": 0.00017648253165548406,
+      "loss": 0.7192418575286865,
+      "step": 4054
+    },
+    {
+      "epoch": 1.7113924050632912,
+      "grad_norm": 0.4348951280117035,
+      "learning_rate": 0.00017645158058661082,
+      "loss": 0.6501081585884094,
+      "step": 4056
+    },
+    {
+      "epoch": 1.7122362869198313,
+      "grad_norm": 0.4178772270679474,
+      "learning_rate": 0.0001764206118818262,
+      "loss": 0.741439163684845,
+      "step": 4058
+    },
+    {
+      "epoch": 1.7130801687763713,
+      "grad_norm": 0.46996670961380005,
+      "learning_rate": 0.00017638962554827414,
+      "loss": 0.7087798714637756,
+      "step": 4060
+    },
+    {
+      "epoch": 1.7139240506329114,
+      "grad_norm": 0.4775761365890503,
+      "learning_rate": 0.00017635862159310253,
+      "loss": 0.6716644167900085,
+      "step": 4062
+    },
+    {
+      "epoch": 1.7147679324894516,
+      "grad_norm": 0.4094069004058838,
+      "learning_rate": 0.00017632760002346345,
+      "loss": 0.6630500555038452,
+      "step": 4064
+    },
+    {
+      "epoch": 1.7156118143459915,
+      "grad_norm": 0.3968847393989563,
+      "learning_rate": 0.0001762965608465129,
+      "loss": 0.7678145170211792,
+      "step": 4066
+    },
+    {
+      "epoch": 1.7164556962025317,
+      "grad_norm": 0.43751946091651917,
+      "learning_rate": 0.0001762655040694111,
+      "loss": 0.6981875896453857,
+      "step": 4068
+    },
+    {
+      "epoch": 1.7172995780590719,
+      "grad_norm": 0.44504284858703613,
+      "learning_rate": 0.00017623442969932218,
+      "loss": 0.6919799447059631,
+      "step": 4070
+    },
+    {
+      "epoch": 1.7181434599156118,
+      "grad_norm": 0.36895114183425903,
+      "learning_rate": 0.00017620333774341447,
+      "loss": 0.6310772895812988,
+      "step": 4072
+    },
+    {
+      "epoch": 1.7189873417721517,
+      "grad_norm": 0.4312762916088104,
+      "learning_rate": 0.00017617222820886022,
+      "loss": 0.6954092383384705,
+      "step": 4074
+    },
+    {
+      "epoch": 1.7198312236286921,
+      "grad_norm": 0.40909886360168457,
+      "learning_rate": 0.00017614110110283587,
+      "loss": 0.6973698139190674,
+      "step": 4076
+    },
+    {
+      "epoch": 1.720675105485232,
+      "grad_norm": 0.43111470341682434,
+      "learning_rate": 0.0001761099564325218,
+      "loss": 0.6775513291358948,
+      "step": 4078
+    },
+    {
+      "epoch": 1.721518987341772,
+      "grad_norm": 0.45573189854621887,
+      "learning_rate": 0.0001760787942051025,
+      "loss": 0.6889033913612366,
+      "step": 4080
+    },
+    {
+      "epoch": 1.7223628691983124,
+      "grad_norm": 0.44934743642807007,
+      "learning_rate": 0.0001760476144277665,
+      "loss": 0.6894892454147339,
+      "step": 4082
+    },
+    {
+      "epoch": 1.7232067510548523,
+      "grad_norm": 0.39716506004333496,
+      "learning_rate": 0.0001760164171077064,
+      "loss": 0.6841691732406616,
+      "step": 4084
+    },
+    {
+      "epoch": 1.7240506329113923,
+      "grad_norm": 0.4486255645751953,
+      "learning_rate": 0.00017598520225211876,
+      "loss": 0.7197495102882385,
+      "step": 4086
+    },
+    {
+      "epoch": 1.7248945147679327,
+      "grad_norm": 0.40130025148391724,
+      "learning_rate": 0.00017595396986820432,
+      "loss": 0.7388272881507874,
+      "step": 4088
+    },
+    {
+      "epoch": 1.7257383966244726,
+      "grad_norm": 0.40419334173202515,
+      "learning_rate": 0.0001759227199631677,
+      "loss": 0.7447546720504761,
+      "step": 4090
+    },
+    {
+      "epoch": 1.7265822784810125,
+      "grad_norm": 0.39778804779052734,
+      "learning_rate": 0.00017589145254421774,
+      "loss": 0.6348387598991394,
+      "step": 4092
+    },
+    {
+      "epoch": 1.7274261603375527,
+      "grad_norm": 0.3934648036956787,
+      "learning_rate": 0.0001758601676185672,
+      "loss": 0.6813296675682068,
+      "step": 4094
+    },
+    {
+      "epoch": 1.7282700421940929,
+      "grad_norm": 0.4080725908279419,
+      "learning_rate": 0.0001758288651934329,
+      "loss": 0.6369423270225525,
+      "step": 4096
+    },
+    {
+      "epoch": 1.7291139240506328,
+      "grad_norm": 0.5186997652053833,
+      "learning_rate": 0.00017579754527603575,
+      "loss": 0.6664180159568787,
+      "step": 4098
+    },
+    {
+      "epoch": 1.729957805907173,
+      "grad_norm": 0.45290765166282654,
+      "learning_rate": 0.0001757662078736006,
+      "loss": 0.7181394696235657,
+      "step": 4100
+    },
+    {
+      "epoch": 1.729957805907173,
+      "eval_loss": 0.7690847516059875,
+      "eval_runtime": 673.6323,
+      "eval_samples_per_second": 3.128,
+      "eval_steps_per_second": 3.128,
+      "step": 4100
+    },
+    {
+      "epoch": 1.7308016877637131,
+      "grad_norm": 0.45677125453948975,
+      "learning_rate": 0.00017573485299335643,
+      "loss": 0.6850123405456543,
+      "step": 4102
+    },
+    {
+      "epoch": 1.731645569620253,
+      "grad_norm": 0.4854084253311157,
+      "learning_rate": 0.0001757034806425362,
+      "loss": 0.7225809693336487,
+      "step": 4104
+    },
+    {
+      "epoch": 1.7324894514767932,
+      "grad_norm": 0.486583411693573,
+      "learning_rate": 0.0001756720908283769,
+      "loss": 0.7303882837295532,
+      "step": 4106
+    },
+    {
+      "epoch": 1.7333333333333334,
+      "grad_norm": 0.4609545171260834,
+      "learning_rate": 0.00017564068355811952,
+      "loss": 0.6789878010749817,
+      "step": 4108
+    },
+    {
+      "epoch": 1.7341772151898733,
+      "grad_norm": 0.3853047788143158,
+      "learning_rate": 0.00017560925883900914,
+      "loss": 0.7311062812805176,
+      "step": 4110
+    },
+    {
+      "epoch": 1.7350210970464135,
+      "grad_norm": 0.41957035660743713,
+      "learning_rate": 0.00017557781667829487,
+      "loss": 0.7106070518493652,
+      "step": 4112
+    },
+    {
+      "epoch": 1.7358649789029537,
+      "grad_norm": 0.37419262528419495,
+      "learning_rate": 0.0001755463570832298,
+      "loss": 0.6345783472061157,
+      "step": 4114
+    },
+    {
+      "epoch": 1.7367088607594936,
+      "grad_norm": 0.5050614476203918,
+      "learning_rate": 0.00017551488006107104,
+      "loss": 0.7923876047134399,
+      "step": 4116
+    },
+    {
+      "epoch": 1.7375527426160338,
+      "grad_norm": 0.49726125597953796,
+      "learning_rate": 0.00017548338561907976,
+      "loss": 0.7502455115318298,
+      "step": 4118
+    },
+    {
+      "epoch": 1.738396624472574,
+      "grad_norm": 0.38956591486930847,
+      "learning_rate": 0.00017545187376452105,
+      "loss": 0.6333624720573425,
+      "step": 4120
+    },
+    {
+      "epoch": 1.7392405063291139,
+      "grad_norm": 0.430318146944046,
+      "learning_rate": 0.0001754203445046642,
+      "loss": 0.6884907484054565,
+      "step": 4122
+    },
+    {
+      "epoch": 1.740084388185654,
+      "grad_norm": 0.39826440811157227,
+      "learning_rate": 0.0001753887978467823,
+      "loss": 0.7345985174179077,
+      "step": 4124
+    },
+    {
+      "epoch": 1.7409282700421942,
+      "grad_norm": 0.4263492822647095,
+      "learning_rate": 0.00017535723379815267,
+      "loss": 0.6537752151489258,
+      "step": 4126
+    },
+    {
+      "epoch": 1.7417721518987341,
+      "grad_norm": 0.4012296199798584,
+      "learning_rate": 0.0001753256523660564,
+      "loss": 0.6368000507354736,
+      "step": 4128
+    },
+    {
+      "epoch": 1.7426160337552743,
+      "grad_norm": 0.3881204426288605,
+      "learning_rate": 0.0001752940535577788,
+      "loss": 0.6038334369659424,
+      "step": 4130
+    },
+    {
+      "epoch": 1.7434599156118145,
+      "grad_norm": 0.38410526514053345,
+      "learning_rate": 0.0001752624373806091,
+      "loss": 0.6657052636146545,
+      "step": 4132
+    },
+    {
+      "epoch": 1.7443037974683544,
+      "grad_norm": 0.3970552980899811,
+      "learning_rate": 0.00017523080384184057,
+      "loss": 0.6518114805221558,
+      "step": 4134
+    },
+    {
+      "epoch": 1.7451476793248946,
+      "grad_norm": 0.37542012333869934,
+      "learning_rate": 0.0001751991529487704,
+      "loss": 0.609853208065033,
+      "step": 4136
+    },
+    {
+      "epoch": 1.7459915611814347,
+      "grad_norm": 0.49812132120132446,
+      "learning_rate": 0.00017516748470869988,
+      "loss": 0.7356346845626831,
+      "step": 4138
+    },
+    {
+      "epoch": 1.7468354430379747,
+      "grad_norm": 0.48197710514068604,
+      "learning_rate": 0.00017513579912893426,
+      "loss": 0.7198674082756042,
+      "step": 4140
+    },
+    {
+      "epoch": 1.7476793248945146,
+      "grad_norm": 0.38266444206237793,
+      "learning_rate": 0.0001751040962167828,
+      "loss": 0.6841102242469788,
+      "step": 4142
+    },
+    {
+      "epoch": 1.748523206751055,
+      "grad_norm": 0.4165695309638977,
+      "learning_rate": 0.0001750723759795587,
+      "loss": 0.7348533272743225,
+      "step": 4144
+    },
+    {
+      "epoch": 1.749367088607595,
+      "grad_norm": 0.4414960443973541,
+      "learning_rate": 0.0001750406384245793,
+      "loss": 0.7344526648521423,
+      "step": 4146
+    },
+    {
+      "epoch": 1.7502109704641349,
+      "grad_norm": 0.4324318468570709,
+      "learning_rate": 0.00017500888355916576,
+      "loss": 0.7047654986381531,
+      "step": 4148
+    },
+    {
+      "epoch": 1.7510548523206753,
+      "grad_norm": 0.44670364260673523,
+      "learning_rate": 0.00017497711139064336,
+      "loss": 0.7317671179771423,
+      "step": 4150
+    },
+    {
+      "epoch": 1.7518987341772152,
+      "grad_norm": 0.43074679374694824,
+      "learning_rate": 0.00017494532192634138,
+      "loss": 0.7447407245635986,
+      "step": 4152
+    },
+    {
+      "epoch": 1.7527426160337551,
+      "grad_norm": 0.43057894706726074,
+      "learning_rate": 0.00017491351517359292,
+      "loss": 0.7352449893951416,
+      "step": 4154
+    },
+    {
+      "epoch": 1.7535864978902953,
+      "grad_norm": 0.49488556385040283,
+      "learning_rate": 0.00017488169113973528,
+      "loss": 0.7244091033935547,
+      "step": 4156
+    },
+    {
+      "epoch": 1.7544303797468355,
+      "grad_norm": 0.4565380811691284,
+      "learning_rate": 0.0001748498498321096,
+      "loss": 0.7515795230865479,
+      "step": 4158
+    },
+    {
+      "epoch": 1.7552742616033754,
+      "grad_norm": 0.45722341537475586,
+      "learning_rate": 0.0001748179912580611,
+      "loss": 0.7451042532920837,
+      "step": 4160
+    },
+    {
+      "epoch": 1.7561181434599156,
+      "grad_norm": 0.47247326374053955,
+      "learning_rate": 0.0001747861154249389,
+      "loss": 0.7133217453956604,
+      "step": 4162
+    },
+    {
+      "epoch": 1.7569620253164557,
+      "grad_norm": 0.48307132720947266,
+      "learning_rate": 0.00017475422234009623,
+      "loss": 0.643742561340332,
+      "step": 4164
+    },
+    {
+      "epoch": 1.7578059071729957,
+      "grad_norm": 0.49163761734962463,
+      "learning_rate": 0.00017472231201089012,
+      "loss": 0.7587857842445374,
+      "step": 4166
+    },
+    {
+      "epoch": 1.7586497890295358,
+      "grad_norm": 0.4726962149143219,
+      "learning_rate": 0.00017469038444468165,
+      "loss": 0.6980416774749756,
+      "step": 4168
+    },
+    {
+      "epoch": 1.759493670886076,
+      "grad_norm": 0.4192392826080322,
+      "learning_rate": 0.00017465843964883598,
+      "loss": 0.6929903626441956,
+      "step": 4170
+    },
+    {
+      "epoch": 1.760337552742616,
+      "grad_norm": 0.43000563979148865,
+      "learning_rate": 0.00017462647763072216,
+      "loss": 0.7076222896575928,
+      "step": 4172
+    },
+    {
+      "epoch": 1.761181434599156,
+      "grad_norm": 0.45402681827545166,
+      "learning_rate": 0.00017459449839771313,
+      "loss": 0.7260748147964478,
+      "step": 4174
+    },
+    {
+      "epoch": 1.7620253164556963,
+      "grad_norm": 0.4077165126800537,
+      "learning_rate": 0.00017456250195718595,
+      "loss": 0.6901930570602417,
+      "step": 4176
+    },
+    {
+      "epoch": 1.7628691983122362,
+      "grad_norm": 0.4167994558811188,
+      "learning_rate": 0.00017453048831652157,
+      "loss": 0.7085300087928772,
+      "step": 4178
+    },
+    {
+      "epoch": 1.7637130801687764,
+      "grad_norm": 0.45262131094932556,
+      "learning_rate": 0.00017449845748310493,
+      "loss": 0.7646799087524414,
+      "step": 4180
+    },
+    {
+      "epoch": 1.7645569620253165,
+      "grad_norm": 0.4639943540096283,
+      "learning_rate": 0.0001744664094643249,
+      "loss": 0.7847923040390015,
+      "step": 4182
+    },
+    {
+      "epoch": 1.7654008438818565,
+      "grad_norm": 0.33605465292930603,
+      "learning_rate": 0.00017443434426757439,
+      "loss": 0.7192286252975464,
+      "step": 4184
+    },
+    {
+      "epoch": 1.7662447257383966,
+      "grad_norm": 0.4036388099193573,
+      "learning_rate": 0.00017440226190025015,
+      "loss": 0.6554746627807617,
+      "step": 4186
+    },
+    {
+      "epoch": 1.7670886075949368,
+      "grad_norm": 0.45648789405822754,
+      "learning_rate": 0.00017437016236975303,
+      "loss": 0.6433794498443604,
+      "step": 4188
+    },
+    {
+      "epoch": 1.7679324894514767,
+      "grad_norm": 0.38929006457328796,
+      "learning_rate": 0.00017433804568348776,
+      "loss": 0.6608775854110718,
+      "step": 4190
+    },
+    {
+      "epoch": 1.768776371308017,
+      "grad_norm": 0.37158843874931335,
+      "learning_rate": 0.00017430591184886308,
+      "loss": 0.6470816135406494,
+      "step": 4192
+    },
+    {
+      "epoch": 1.769620253164557,
+      "grad_norm": 0.39842307567596436,
+      "learning_rate": 0.00017427376087329158,
+      "loss": 0.6825124025344849,
+      "step": 4194
+    },
+    {
+      "epoch": 1.770464135021097,
+      "grad_norm": 0.44220465421676636,
+      "learning_rate": 0.00017424159276418987,
+      "loss": 0.7641696333885193,
+      "step": 4196
+    },
+    {
+      "epoch": 1.7713080168776372,
+      "grad_norm": 0.44530072808265686,
+      "learning_rate": 0.0001742094075289786,
+      "loss": 0.7783681154251099,
+      "step": 4198
+    },
+    {
+      "epoch": 1.7721518987341773,
+      "grad_norm": 0.4497349262237549,
+      "learning_rate": 0.00017417720517508216,
+      "loss": 0.6718224883079529,
+      "step": 4200
+    },
+    {
+      "epoch": 1.7721518987341773,
+      "eval_loss": 0.7676366567611694,
+      "eval_runtime": 687.9619,
+      "eval_samples_per_second": 3.063,
+      "eval_steps_per_second": 3.063,
+      "step": 4200
+    },
+    {
+      "epoch": 1.7729957805907173,
+      "grad_norm": 0.3857066333293915,
+      "learning_rate": 0.00017414498570992914,
+      "loss": 0.7050814032554626,
+      "step": 4202
+    },
+    {
+      "epoch": 1.7738396624472574,
+      "grad_norm": 0.4533625543117523,
+      "learning_rate": 0.00017411274914095188,
+      "loss": 0.7222742438316345,
+      "step": 4204
+    },
+    {
+      "epoch": 1.7746835443037976,
+      "grad_norm": 0.40805840492248535,
+      "learning_rate": 0.00017408049547558676,
+      "loss": 0.6434906125068665,
+      "step": 4206
+    },
+    {
+      "epoch": 1.7755274261603375,
+      "grad_norm": 0.5041740536689758,
+      "learning_rate": 0.00017404822472127406,
+      "loss": 0.686283528804779,
+      "step": 4208
+    },
+    {
+      "epoch": 1.7763713080168775,
+      "grad_norm": 0.41846802830696106,
+      "learning_rate": 0.00017401593688545807,
+      "loss": 0.6785043478012085,
+      "step": 4210
+    },
+    {
+      "epoch": 1.7772151898734179,
+      "grad_norm": 0.4696303606033325,
+      "learning_rate": 0.00017398363197558694,
+      "loss": 0.7566665410995483,
+      "step": 4212
+    },
+    {
+      "epoch": 1.7780590717299578,
+      "grad_norm": 0.3731461763381958,
+      "learning_rate": 0.0001739513099991128,
+      "loss": 0.6178455948829651,
+      "step": 4214
+    },
+    {
+      "epoch": 1.7789029535864977,
+      "grad_norm": 0.48098450899124146,
+      "learning_rate": 0.00017391897096349167,
+      "loss": 0.771265983581543,
+      "step": 4216
+    },
+    {
+      "epoch": 1.7797468354430381,
+      "grad_norm": 0.4164000451564789,
+      "learning_rate": 0.0001738866148761836,
+      "loss": 0.6372597813606262,
+      "step": 4218
+    },
+    {
+      "epoch": 1.780590717299578,
+      "grad_norm": 0.4490415155887604,
+      "learning_rate": 0.00017385424174465251,
+      "loss": 0.7213301658630371,
+      "step": 4220
+    },
+    {
+      "epoch": 1.781434599156118,
+      "grad_norm": 0.44518229365348816,
+      "learning_rate": 0.00017382185157636623,
+      "loss": 0.7575726509094238,
+      "step": 4222
+    },
+    {
+      "epoch": 1.7822784810126582,
+      "grad_norm": 0.34899434447288513,
+      "learning_rate": 0.00017378944437879658,
+      "loss": 0.6473780274391174,
+      "step": 4224
+    },
+    {
+      "epoch": 1.7831223628691983,
+      "grad_norm": 0.4068683087825775,
+      "learning_rate": 0.00017375702015941924,
+      "loss": 0.7238336205482483,
+      "step": 4226
+    },
+    {
+      "epoch": 1.7839662447257383,
+      "grad_norm": 0.4025551974773407,
+      "learning_rate": 0.0001737245789257139,
+      "loss": 0.7797979116439819,
+      "step": 4228
+    },
+    {
+      "epoch": 1.7848101265822784,
+      "grad_norm": 0.35764896869659424,
+      "learning_rate": 0.00017369212068516412,
+      "loss": 0.6304885149002075,
+      "step": 4230
+    },
+    {
+      "epoch": 1.7856540084388186,
+      "grad_norm": 0.43337100744247437,
+      "learning_rate": 0.00017365964544525738,
+      "loss": 0.6606637835502625,
+      "step": 4232
+    },
+    {
+      "epoch": 1.7864978902953585,
+      "grad_norm": 0.4630814492702484,
+      "learning_rate": 0.0001736271532134851,
+      "loss": 0.712163507938385,
+      "step": 4234
+    },
+    {
+      "epoch": 1.7873417721518987,
+      "grad_norm": 0.38714799284935,
+      "learning_rate": 0.00017359464399734255,
+      "loss": 0.6649466753005981,
+      "step": 4236
+    },
+    {
+      "epoch": 1.7881856540084389,
+      "grad_norm": 0.38999977707862854,
+      "learning_rate": 0.0001735621178043291,
+      "loss": 0.6223326921463013,
+      "step": 4238
+    },
+    {
+      "epoch": 1.7890295358649788,
+      "grad_norm": 0.4736211597919464,
+      "learning_rate": 0.00017352957464194786,
+      "loss": 0.7207517027854919,
+      "step": 4240
+    },
+    {
+      "epoch": 1.789873417721519,
+      "grad_norm": 0.423476904630661,
+      "learning_rate": 0.00017349701451770588,
+      "loss": 0.7203661203384399,
+      "step": 4242
+    },
+    {
+      "epoch": 1.7907172995780591,
+      "grad_norm": 0.4487479627132416,
+      "learning_rate": 0.0001734644374391142,
+      "loss": 0.709998607635498,
+      "step": 4244
+    },
+    {
+      "epoch": 1.791561181434599,
+      "grad_norm": 0.43721267580986023,
+      "learning_rate": 0.00017343184341368773,
+      "loss": 0.7079582214355469,
+      "step": 4246
+    },
+    {
+      "epoch": 1.7924050632911392,
+      "grad_norm": 0.426864892244339,
+      "learning_rate": 0.00017339923244894524,
+      "loss": 0.6772971749305725,
+      "step": 4248
+    },
+    {
+      "epoch": 1.7932489451476794,
+      "grad_norm": 0.3580208420753479,
+      "learning_rate": 0.0001733666045524095,
+      "loss": 0.6882370710372925,
+      "step": 4250
+    },
+    {
+      "epoch": 1.7940928270042193,
+      "grad_norm": 0.487248957157135,
+      "learning_rate": 0.00017333395973160714,
+      "loss": 0.7534610629081726,
+      "step": 4252
+    },
+    {
+      "epoch": 1.7949367088607595,
+      "grad_norm": 0.44908860325813293,
+      "learning_rate": 0.00017330129799406866,
+      "loss": 0.6897269487380981,
+      "step": 4254
+    },
+    {
+      "epoch": 1.7957805907172997,
+      "grad_norm": 0.43617817759513855,
+      "learning_rate": 0.00017326861934732852,
+      "loss": 0.7081538438796997,
+      "step": 4256
+    },
+    {
+      "epoch": 1.7966244725738396,
+      "grad_norm": 0.428666353225708,
+      "learning_rate": 0.00017323592379892504,
+      "loss": 0.6852753758430481,
+      "step": 4258
+    },
+    {
+      "epoch": 1.7974683544303798,
+      "grad_norm": 0.4522199034690857,
+      "learning_rate": 0.00017320321135640045,
+      "loss": 0.7463182210922241,
+      "step": 4260
+    },
+    {
+      "epoch": 1.79831223628692,
+      "grad_norm": 0.38632410764694214,
+      "learning_rate": 0.00017317048202730088,
+      "loss": 0.6643774509429932,
+      "step": 4262
+    },
+    {
+      "epoch": 1.7991561181434599,
+      "grad_norm": 0.41171637177467346,
+      "learning_rate": 0.00017313773581917642,
+      "loss": 0.6726668477058411,
+      "step": 4264
+    },
+    {
+      "epoch": 1.8,
+      "grad_norm": 0.398520290851593,
+      "learning_rate": 0.0001731049727395809,
+      "loss": 0.6278629302978516,
+      "step": 4266
+    },
+    {
+      "epoch": 1.8008438818565402,
+      "grad_norm": 0.4000356197357178,
+      "learning_rate": 0.0001730721927960722,
+      "loss": 0.5823352336883545,
+      "step": 4268
+    },
+    {
+      "epoch": 1.8016877637130801,
+      "grad_norm": 0.47376710176467896,
+      "learning_rate": 0.000173039395996212,
+      "loss": 0.7492824196815491,
+      "step": 4270
+    },
+    {
+      "epoch": 1.80253164556962,
+      "grad_norm": 0.4624378979206085,
+      "learning_rate": 0.00017300658234756588,
+      "loss": 0.7630324363708496,
+      "step": 4272
+    },
+    {
+      "epoch": 1.8033755274261605,
+      "grad_norm": 0.4178096354007721,
+      "learning_rate": 0.00017297375185770336,
+      "loss": 0.6484803557395935,
+      "step": 4274
+    },
+    {
+      "epoch": 1.8042194092827004,
+      "grad_norm": 0.511215329170227,
+      "learning_rate": 0.00017294090453419774,
+      "loss": 0.776732325553894,
+      "step": 4276
+    },
+    {
+      "epoch": 1.8050632911392404,
+      "grad_norm": 0.44643548130989075,
+      "learning_rate": 0.00017290804038462633,
+      "loss": 0.6801077723503113,
+      "step": 4278
+    },
+    {
+      "epoch": 1.8059071729957807,
+      "grad_norm": 0.5090142488479614,
+      "learning_rate": 0.0001728751594165702,
+      "loss": 0.8024200797080994,
+      "step": 4280
+    },
+    {
+      "epoch": 1.8067510548523207,
+      "grad_norm": 0.45888981223106384,
+      "learning_rate": 0.00017284226163761436,
+      "loss": 0.7810231447219849,
+      "step": 4282
+    },
+    {
+      "epoch": 1.8075949367088606,
+      "grad_norm": 0.3985820710659027,
+      "learning_rate": 0.0001728093470553478,
+      "loss": 0.6768189072608948,
+      "step": 4284
+    },
+    {
+      "epoch": 1.808438818565401,
+      "grad_norm": 0.3988003730773926,
+      "learning_rate": 0.00017277641567736316,
+      "loss": 0.6093943119049072,
+      "step": 4286
+    },
+    {
+      "epoch": 1.809282700421941,
+      "grad_norm": 0.46131962537765503,
+      "learning_rate": 0.0001727434675112571,
+      "loss": 0.6836619973182678,
+      "step": 4288
+    },
+    {
+      "epoch": 1.810126582278481,
+      "grad_norm": 0.42959487438201904,
+      "learning_rate": 0.0001727105025646302,
+      "loss": 0.7144147157669067,
+      "step": 4290
+    },
+    {
+      "epoch": 1.810970464135021,
+      "grad_norm": 0.4110259413719177,
+      "learning_rate": 0.00017267752084508675,
+      "loss": 0.7190625667572021,
+      "step": 4292
+    },
+    {
+      "epoch": 1.8118143459915612,
+      "grad_norm": 0.4501377046108246,
+      "learning_rate": 0.00017264452236023505,
+      "loss": 0.7272495031356812,
+      "step": 4294
+    },
+    {
+      "epoch": 1.8126582278481012,
+      "grad_norm": 0.467817097902298,
+      "learning_rate": 0.0001726115071176872,
+      "loss": 0.758166491985321,
+      "step": 4296
+    },
+    {
+      "epoch": 1.8135021097046413,
+      "grad_norm": 0.42394939064979553,
+      "learning_rate": 0.0001725784751250592,
+      "loss": 0.6712231040000916,
+      "step": 4298
+    },
+    {
+      "epoch": 1.8143459915611815,
+      "grad_norm": 0.37668925523757935,
+      "learning_rate": 0.00017254542638997088,
+      "loss": 0.5999719500541687,
+      "step": 4300
+    },
+    {
+      "epoch": 1.8143459915611815,
+      "eval_loss": 0.7672964930534363,
+      "eval_runtime": 688.4249,
+      "eval_samples_per_second": 3.061,
+      "eval_steps_per_second": 3.061,
+      "step": 4300
+    },
+    {
+      "epoch": 1.8151898734177214,
+      "grad_norm": 0.44964465498924255,
+      "learning_rate": 0.00017251236092004596,
+      "loss": 0.7188607454299927,
+      "step": 4302
+    },
+    {
+      "epoch": 1.8160337552742616,
+      "grad_norm": 0.4108492434024811,
+      "learning_rate": 0.000172479278722912,
+      "loss": 0.707834005355835,
+      "step": 4304
+    },
+    {
+      "epoch": 1.8168776371308017,
+      "grad_norm": 0.4455305337905884,
+      "learning_rate": 0.00017244617980620043,
+      "loss": 0.6915075182914734,
+      "step": 4306
+    },
+    {
+      "epoch": 1.8177215189873417,
+      "grad_norm": 0.3977123200893402,
+      "learning_rate": 0.00017241306417754655,
+      "loss": 0.6591874361038208,
+      "step": 4308
+    },
+    {
+      "epoch": 1.8185654008438819,
+      "grad_norm": 0.4028434157371521,
+      "learning_rate": 0.00017237993184458946,
+      "loss": 0.6254785060882568,
+      "step": 4310
+    },
+    {
+      "epoch": 1.819409282700422,
+      "grad_norm": 0.41024482250213623,
+      "learning_rate": 0.00017234678281497217,
+      "loss": 0.6968662142753601,
+      "step": 4312
+    },
+    {
+      "epoch": 1.820253164556962,
+      "grad_norm": 0.45941361784935,
+      "learning_rate": 0.00017231361709634153,
+      "loss": 0.6535647511482239,
+      "step": 4314
+    },
+    {
+      "epoch": 1.8210970464135021,
+      "grad_norm": 0.39477071166038513,
+      "learning_rate": 0.00017228043469634825,
+      "loss": 0.6378563642501831,
+      "step": 4316
+    },
+    {
+      "epoch": 1.8219409282700423,
+      "grad_norm": 0.403826504945755,
+      "learning_rate": 0.00017224723562264687,
+      "loss": 0.7021255493164062,
+      "step": 4318
+    },
+    {
+      "epoch": 1.8227848101265822,
+      "grad_norm": 0.4179201126098633,
+      "learning_rate": 0.00017221401988289575,
+      "loss": 0.7411118745803833,
+      "step": 4320
+    },
+    {
+      "epoch": 1.8236286919831224,
+      "grad_norm": 0.4719425141811371,
+      "learning_rate": 0.0001721807874847571,
+      "loss": 0.7435210943222046,
+      "step": 4322
+    },
+    {
+      "epoch": 1.8244725738396625,
+      "grad_norm": 0.39946448802948,
+      "learning_rate": 0.0001721475384358971,
+      "loss": 0.6813794374465942,
+      "step": 4324
+    },
+    {
+      "epoch": 1.8253164556962025,
+      "grad_norm": 0.4367005527019501,
+      "learning_rate": 0.00017211427274398559,
+      "loss": 0.6468828320503235,
+      "step": 4326
+    },
+    {
+      "epoch": 1.8261603375527427,
+      "grad_norm": 0.4078138470649719,
+      "learning_rate": 0.00017208099041669632,
+      "loss": 0.6628729701042175,
+      "step": 4328
+    },
+    {
+      "epoch": 1.8270042194092828,
+      "grad_norm": 0.4141235649585724,
+      "learning_rate": 0.0001720476914617069,
+      "loss": 0.7028043866157532,
+      "step": 4330
+    },
+    {
+      "epoch": 1.8278481012658228,
+      "grad_norm": 0.448733925819397,
+      "learning_rate": 0.00017201437588669878,
+      "loss": 0.6710682511329651,
+      "step": 4332
+    },
+    {
+      "epoch": 1.828691983122363,
+      "grad_norm": 0.5084340572357178,
+      "learning_rate": 0.00017198104369935718,
+      "loss": 0.792453408241272,
+      "step": 4334
+    },
+    {
+      "epoch": 1.829535864978903,
+      "grad_norm": 0.43463611602783203,
+      "learning_rate": 0.00017194769490737124,
+      "loss": 0.7536583542823792,
+      "step": 4336
+    },
+    {
+      "epoch": 1.830379746835443,
+      "grad_norm": 0.46039044857025146,
+      "learning_rate": 0.00017191432951843387,
+      "loss": 0.7445555925369263,
+      "step": 4338
+    },
+    {
+      "epoch": 1.831223628691983,
+      "grad_norm": 0.3894030451774597,
+      "learning_rate": 0.0001718809475402418,
+      "loss": 0.7003044486045837,
+      "step": 4340
+    },
+    {
+      "epoch": 1.8320675105485233,
+      "grad_norm": 0.5851747393608093,
+      "learning_rate": 0.00017184754898049568,
+      "loss": 0.6948950886726379,
+      "step": 4342
+    },
+    {
+      "epoch": 1.8329113924050633,
+      "grad_norm": 0.40906789898872375,
+      "learning_rate": 0.00017181413384689982,
+      "loss": 0.675811231136322,
+      "step": 4344
+    },
+    {
+      "epoch": 1.8337552742616032,
+      "grad_norm": 0.4235765039920807,
+      "learning_rate": 0.0001717807021471625,
+      "loss": 0.7422580122947693,
+      "step": 4346
+    },
+    {
+      "epoch": 1.8345991561181436,
+      "grad_norm": 0.3898368775844574,
+      "learning_rate": 0.0001717472538889958,
+      "loss": 0.7014731168746948,
+      "step": 4348
+    },
+    {
+      "epoch": 1.8354430379746836,
+      "grad_norm": 0.4122314751148224,
+      "learning_rate": 0.00017171378908011553,
+      "loss": 0.6952776312828064,
+      "step": 4350
+    },
+    {
+      "epoch": 1.8362869198312235,
+      "grad_norm": 0.38085612654685974,
+      "learning_rate": 0.00017168030772824144,
+      "loss": 0.6758598685264587,
+      "step": 4352
+    },
+    {
+      "epoch": 1.8371308016877637,
+      "grad_norm": 0.43519729375839233,
+      "learning_rate": 0.00017164680984109695,
+      "loss": 0.7139424085617065,
+      "step": 4354
+    },
+    {
+      "epoch": 1.8379746835443038,
+      "grad_norm": 0.4404325783252716,
+      "learning_rate": 0.0001716132954264095,
+      "loss": 0.7125198841094971,
+      "step": 4356
+    },
+    {
+      "epoch": 1.8388185654008438,
+      "grad_norm": 0.3965512812137604,
+      "learning_rate": 0.00017157976449191012,
+      "loss": 0.7011314630508423,
+      "step": 4358
+    },
+    {
+      "epoch": 1.839662447257384,
+      "grad_norm": 0.4273630976676941,
+      "learning_rate": 0.0001715462170453338,
+      "loss": 0.6485645174980164,
+      "step": 4360
+    },
+    {
+      "epoch": 1.840506329113924,
+      "grad_norm": 0.4639208912849426,
+      "learning_rate": 0.00017151265309441925,
+      "loss": 0.6669548153877258,
+      "step": 4362
+    },
+    {
+      "epoch": 1.841350210970464,
+      "grad_norm": 0.4190906286239624,
+      "learning_rate": 0.00017147907264690907,
+      "loss": 0.6276374459266663,
+      "step": 4364
+    },
+    {
+      "epoch": 1.8421940928270042,
+      "grad_norm": 0.4146515130996704,
+      "learning_rate": 0.00017144547571054962,
+      "loss": 0.5975915193557739,
+      "step": 4366
+    },
+    {
+      "epoch": 1.8430379746835444,
+      "grad_norm": 0.38717755675315857,
+      "learning_rate": 0.00017141186229309105,
+      "loss": 0.6536507606506348,
+      "step": 4368
+    },
+    {
+      "epoch": 1.8438818565400843,
+      "grad_norm": 0.3988541066646576,
+      "learning_rate": 0.00017137823240228735,
+      "loss": 0.7078962326049805,
+      "step": 4370
+    },
+    {
+      "epoch": 1.8447257383966245,
+      "grad_norm": 0.43783870339393616,
+      "learning_rate": 0.00017134458604589628,
+      "loss": 0.6795231699943542,
+      "step": 4372
+    },
+    {
+      "epoch": 1.8455696202531646,
+      "grad_norm": 0.439636766910553,
+      "learning_rate": 0.00017131092323167938,
+      "loss": 0.7030490040779114,
+      "step": 4374
+    },
+    {
+      "epoch": 1.8464135021097046,
+      "grad_norm": 0.4426078498363495,
+      "learning_rate": 0.00017127724396740206,
+      "loss": 0.7611671686172485,
+      "step": 4376
+    },
+    {
+      "epoch": 1.8472573839662447,
+      "grad_norm": 0.40058812499046326,
+      "learning_rate": 0.00017124354826083348,
+      "loss": 0.737910270690918,
+      "step": 4378
+    },
+    {
+      "epoch": 1.8481012658227849,
+      "grad_norm": 0.43037283420562744,
+      "learning_rate": 0.00017120983611974654,
+      "loss": 0.683937132358551,
+      "step": 4380
+    },
+    {
+      "epoch": 1.8489451476793248,
+      "grad_norm": 0.4479113221168518,
+      "learning_rate": 0.00017117610755191807,
+      "loss": 0.6987335681915283,
+      "step": 4382
+    },
+    {
+      "epoch": 1.849789029535865,
+      "grad_norm": 0.4218229651451111,
+      "learning_rate": 0.0001711423625651285,
+      "loss": 0.6964257955551147,
+      "step": 4384
+    },
+    {
+      "epoch": 1.8506329113924052,
+      "grad_norm": 0.397733211517334,
+      "learning_rate": 0.00017110860116716222,
+      "loss": 0.6983811855316162,
+      "step": 4386
+    },
+    {
+      "epoch": 1.851476793248945,
+      "grad_norm": 0.4844493865966797,
+      "learning_rate": 0.00017107482336580735,
+      "loss": 0.7579324245452881,
+      "step": 4388
+    },
+    {
+      "epoch": 1.8523206751054853,
+      "grad_norm": 0.4494120478630066,
+      "learning_rate": 0.0001710410291688557,
+      "loss": 0.703558623790741,
+      "step": 4390
+    },
+    {
+      "epoch": 1.8531645569620254,
+      "grad_norm": 0.4721727967262268,
+      "learning_rate": 0.000171007218584103,
+      "loss": 0.7386981248855591,
+      "step": 4392
+    },
+    {
+      "epoch": 1.8540084388185654,
+      "grad_norm": 0.4816451370716095,
+      "learning_rate": 0.0001709733916193487,
+      "loss": 0.7769017219543457,
+      "step": 4394
+    },
+    {
+      "epoch": 1.8548523206751055,
+      "grad_norm": 0.43638876080513,
+      "learning_rate": 0.00017093954828239602,
+      "loss": 0.7315034866333008,
+      "step": 4396
+    },
+    {
+      "epoch": 1.8556962025316457,
+      "grad_norm": 0.4035055935382843,
+      "learning_rate": 0.000170905688581052,
+      "loss": 0.7289890050888062,
+      "step": 4398
+    },
+    {
+      "epoch": 1.8565400843881856,
+      "grad_norm": 0.44265708327293396,
+      "learning_rate": 0.00017087181252312735,
+      "loss": 0.7190408110618591,
+      "step": 4400
+    },
+    {
+      "epoch": 1.8565400843881856,
+      "eval_loss": 0.7635221481323242,
+      "eval_runtime": 678.243,
+      "eval_samples_per_second": 3.107,
+      "eval_steps_per_second": 3.107,
+      "step": 4400
+    },
+    {
+      "epoch": 1.8573839662447258,
+      "grad_norm": 0.39691466093063354,
+      "learning_rate": 0.0001708379201164367,
+      "loss": 0.6318185925483704,
+      "step": 4402
+    },
+    {
+      "epoch": 1.858227848101266,
+      "grad_norm": 0.47801393270492554,
+      "learning_rate": 0.0001708040113687983,
+      "loss": 0.7040913105010986,
+      "step": 4404
+    },
+    {
+      "epoch": 1.859071729957806,
+      "grad_norm": 0.47457239031791687,
+      "learning_rate": 0.00017077008628803436,
+      "loss": 0.7834659218788147,
+      "step": 4406
+    },
+    {
+      "epoch": 1.8599156118143458,
+      "grad_norm": 0.41362807154655457,
+      "learning_rate": 0.00017073614488197065,
+      "loss": 0.7027983665466309,
+      "step": 4408
+    },
+    {
+      "epoch": 1.8607594936708862,
+      "grad_norm": 0.457927942276001,
+      "learning_rate": 0.0001707021871584368,
+      "loss": 0.7143591642379761,
+      "step": 4410
+    },
+    {
+      "epoch": 1.8616033755274262,
+      "grad_norm": 0.42931967973709106,
+      "learning_rate": 0.00017066821312526625,
+      "loss": 0.7328225374221802,
+      "step": 4412
+    },
+    {
+      "epoch": 1.862447257383966,
+      "grad_norm": 0.3821638524532318,
+      "learning_rate": 0.00017063422279029616,
+      "loss": 0.7199252843856812,
+      "step": 4414
+    },
+    {
+      "epoch": 1.8632911392405065,
+      "grad_norm": 0.3954671621322632,
+      "learning_rate": 0.00017060021616136742,
+      "loss": 0.7345157265663147,
+      "step": 4416
+    },
+    {
+      "epoch": 1.8641350210970464,
+      "grad_norm": 0.38824960589408875,
+      "learning_rate": 0.0001705661932463247,
+      "loss": 0.587895393371582,
+      "step": 4418
+    },
+    {
+      "epoch": 1.8649789029535864,
+      "grad_norm": 0.510923445224762,
+      "learning_rate": 0.00017053215405301648,
+      "loss": 0.7227945327758789,
+      "step": 4420
+    },
+    {
+      "epoch": 1.8658227848101265,
+      "grad_norm": 0.38688936829566956,
+      "learning_rate": 0.0001704980985892949,
+      "loss": 0.6605637073516846,
+      "step": 4422
+    },
+    {
+      "epoch": 1.8666666666666667,
+      "grad_norm": 0.4886745512485504,
+      "learning_rate": 0.0001704640268630159,
+      "loss": 0.7581667304039001,
+      "step": 4424
+    },
+    {
+      "epoch": 1.8675105485232066,
+      "grad_norm": 0.4214446246623993,
+      "learning_rate": 0.0001704299388820392,
+      "loss": 0.7051677107810974,
+      "step": 4426
+    },
+    {
+      "epoch": 1.8683544303797468,
+      "grad_norm": 0.4189293384552002,
+      "learning_rate": 0.00017039583465422824,
+      "loss": 0.6489346623420715,
+      "step": 4428
+    },
+    {
+      "epoch": 1.869198312236287,
+      "grad_norm": 0.4264903962612152,
+      "learning_rate": 0.0001703617141874502,
+      "loss": 0.6896629929542542,
+      "step": 4430
+    },
+    {
+      "epoch": 1.870042194092827,
+      "grad_norm": 0.5351413488388062,
+      "learning_rate": 0.00017032757748957603,
+      "loss": 0.7468283772468567,
+      "step": 4432
+    },
+    {
+      "epoch": 1.870886075949367,
+      "grad_norm": 0.40163540840148926,
+      "learning_rate": 0.00017029342456848035,
+      "loss": 0.6029976010322571,
+      "step": 4434
+    },
+    {
+      "epoch": 1.8717299578059072,
+      "grad_norm": 0.3786795139312744,
+      "learning_rate": 0.0001702592554320417,
+      "loss": 0.6891162991523743,
+      "step": 4436
+    },
+    {
+      "epoch": 1.8725738396624472,
+      "grad_norm": 0.4266849458217621,
+      "learning_rate": 0.00017022507008814214,
+      "loss": 0.6189853549003601,
+      "step": 4438
+    },
+    {
+      "epoch": 1.8734177215189873,
+      "grad_norm": 0.4076935052871704,
+      "learning_rate": 0.0001701908685446676,
+      "loss": 0.6858769655227661,
+      "step": 4440
+    },
+    {
+      "epoch": 1.8742616033755275,
+      "grad_norm": 0.4209558665752411,
+      "learning_rate": 0.00017015665080950774,
+      "loss": 0.6650485396385193,
+      "step": 4442
+    },
+    {
+      "epoch": 1.8751054852320674,
+      "grad_norm": 0.4114036560058594,
+      "learning_rate": 0.0001701224168905559,
+      "loss": 0.6887878179550171,
+      "step": 4444
+    },
+    {
+      "epoch": 1.8759493670886076,
+      "grad_norm": 0.34596455097198486,
+      "learning_rate": 0.00017008816679570926,
+      "loss": 0.6950712203979492,
+      "step": 4446
+    },
+    {
+      "epoch": 1.8767932489451478,
+      "grad_norm": 0.45132747292518616,
+      "learning_rate": 0.0001700539005328686,
+      "loss": 0.7360581159591675,
+      "step": 4448
+    },
+    {
+      "epoch": 1.8776371308016877,
+      "grad_norm": 0.4258243441581726,
+      "learning_rate": 0.00017001961810993846,
+      "loss": 0.6672278046607971,
+      "step": 4450
+    },
+    {
+      "epoch": 1.8784810126582279,
+      "grad_norm": 0.39144474267959595,
+      "learning_rate": 0.0001699853195348272,
+      "loss": 0.6740089654922485,
+      "step": 4452
+    },
+    {
+      "epoch": 1.879324894514768,
+      "grad_norm": 0.42183035612106323,
+      "learning_rate": 0.00016995100481544681,
+      "loss": 0.7747141122817993,
+      "step": 4454
+    },
+    {
+      "epoch": 1.880168776371308,
+      "grad_norm": 0.4110005795955658,
+      "learning_rate": 0.00016991667395971303,
+      "loss": 0.7189701199531555,
+      "step": 4456
+    },
+    {
+      "epoch": 1.8810126582278481,
+      "grad_norm": 0.4073067307472229,
+      "learning_rate": 0.00016988232697554538,
+      "loss": 0.6880105137825012,
+      "step": 4458
+    },
+    {
+      "epoch": 1.8818565400843883,
+      "grad_norm": 0.467443585395813,
+      "learning_rate": 0.00016984796387086698,
+      "loss": 0.6994736790657043,
+      "step": 4460
+    },
+    {
+      "epoch": 1.8827004219409282,
+      "grad_norm": 0.4769018292427063,
+      "learning_rate": 0.00016981358465360482,
+      "loss": 0.7389267683029175,
+      "step": 4462
+    },
+    {
+      "epoch": 1.8835443037974684,
+      "grad_norm": 0.3834097981452942,
+      "learning_rate": 0.00016977918933168938,
+      "loss": 0.6375740170478821,
+      "step": 4464
+    },
+    {
+      "epoch": 1.8843881856540086,
+      "grad_norm": 0.44548940658569336,
+      "learning_rate": 0.00016974477791305517,
+      "loss": 0.6845629811286926,
+      "step": 4466
+    },
+    {
+      "epoch": 1.8852320675105485,
+      "grad_norm": 0.39344266057014465,
+      "learning_rate": 0.00016971035040564016,
+      "loss": 0.7307721376419067,
+      "step": 4468
+    },
+    {
+      "epoch": 1.8860759493670884,
+      "grad_norm": 0.37901216745376587,
+      "learning_rate": 0.0001696759068173861,
+      "loss": 0.6733962297439575,
+      "step": 4470
+    },
+    {
+      "epoch": 1.8869198312236288,
+      "grad_norm": 0.44465333223342896,
+      "learning_rate": 0.00016964144715623852,
+      "loss": 0.7294619679450989,
+      "step": 4472
+    },
+    {
+      "epoch": 1.8877637130801688,
+      "grad_norm": 0.39566636085510254,
+      "learning_rate": 0.00016960697143014657,
+      "loss": 0.6537288427352905,
+      "step": 4474
+    },
+    {
+      "epoch": 1.8886075949367087,
+      "grad_norm": 0.4047028124332428,
+      "learning_rate": 0.0001695724796470631,
+      "loss": 0.623903751373291,
+      "step": 4476
+    },
+    {
+      "epoch": 1.889451476793249,
+      "grad_norm": 0.4559679329395294,
+      "learning_rate": 0.00016953797181494475,
+      "loss": 0.661283552646637,
+      "step": 4478
+    },
+    {
+      "epoch": 1.890295358649789,
+      "grad_norm": 0.3980371952056885,
+      "learning_rate": 0.00016950344794175183,
+      "loss": 0.6587702035903931,
+      "step": 4480
+    },
+    {
+      "epoch": 1.891139240506329,
+      "grad_norm": 0.41418227553367615,
+      "learning_rate": 0.0001694689080354483,
+      "loss": 0.7898983955383301,
+      "step": 4482
+    },
+    {
+      "epoch": 1.8919831223628694,
+      "grad_norm": 0.41184720396995544,
+      "learning_rate": 0.00016943435210400184,
+      "loss": 0.6213424205780029,
+      "step": 4484
+    },
+    {
+      "epoch": 1.8928270042194093,
+      "grad_norm": 0.39218711853027344,
+      "learning_rate": 0.00016939978015538388,
+      "loss": 0.6624190807342529,
+      "step": 4486
+    },
+    {
+      "epoch": 1.8936708860759492,
+      "grad_norm": 0.4875730872154236,
+      "learning_rate": 0.0001693651921975695,
+      "loss": 0.7063044309616089,
+      "step": 4488
+    },
+    {
+      "epoch": 1.8945147679324894,
+      "grad_norm": 0.38058432936668396,
+      "learning_rate": 0.00016933058823853748,
+      "loss": 0.6471660137176514,
+      "step": 4490
+    },
+    {
+      "epoch": 1.8953586497890296,
+      "grad_norm": 0.5368654131889343,
+      "learning_rate": 0.0001692959682862702,
+      "loss": 0.7694545388221741,
+      "step": 4492
+    },
+    {
+      "epoch": 1.8962025316455695,
+      "grad_norm": 0.41736599802970886,
+      "learning_rate": 0.00016926133234875396,
+      "loss": 0.700016438961029,
+      "step": 4494
+    },
+    {
+      "epoch": 1.8970464135021097,
+      "grad_norm": 0.4086865484714508,
+      "learning_rate": 0.0001692266804339785,
+      "loss": 0.6926955580711365,
+      "step": 4496
+    },
+    {
+      "epoch": 1.8978902953586498,
+      "grad_norm": 0.447516530752182,
+      "learning_rate": 0.00016919201254993738,
+      "loss": 0.6776167154312134,
+      "step": 4498
+    },
+    {
+      "epoch": 1.8987341772151898,
+      "grad_norm": 0.43951377272605896,
+      "learning_rate": 0.00016915732870462783,
+      "loss": 0.6860454678535461,
+      "step": 4500
+    },
+    {
+      "epoch": 1.8987341772151898,
+      "eval_loss": 0.7600579857826233,
+      "eval_runtime": 674.2593,
+      "eval_samples_per_second": 3.125,
+      "eval_steps_per_second": 3.125,
+      "step": 4500
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 14220,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 6,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 5,
+        "early_stopping_threshold": 0.001
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 0
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.690249164222458e+18,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-4500/training_args.bin b/sft_qwen_14B_v2/checkpoints/checkpoint-4500/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..0bd5f45757bdc6c7fc2d67718b18675b51983e0e
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-4500/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:afae4a8ce1391f149eb00b14eed8f891c715e892ea546bf754d22db2c2bc5969
+size 4792
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-500/README.md b/sft_qwen_14B_v2/checkpoints/checkpoint-500/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..134a3a8531b68adda3fed2a98db824f7c1b7dcfd
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-500/README.md
@@ -0,0 +1,207 @@
+---
+base_model: Models/Qwen2.5-Coder-14B-CPT
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Models/Qwen2.5-Coder-14B-CPT
+- lora
+- transformers
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.0
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-500/adapter_config.json b/sft_qwen_14B_v2/checkpoints/checkpoint-500/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..d6316742bc71d30704e2ccdb9080ed0b8062e454
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-500/adapter_config.json
@@ -0,0 +1,43 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Models/Qwen2.5-Coder-14B-CPT",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.0",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "v_proj",
+    "k_proj",
+    "q_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-500/adapter_model.safetensors b/sft_qwen_14B_v2/checkpoints/checkpoint-500/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..9c2b9d4c3ca1ff91ac6516d2990fefabe57bc571
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-500/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:091040c7ce6fb505a7d3e39fafc8046dc915dbcb4f1c3176b8c2d35c69386847
+size 201378736
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-500/optimizer.pt b/sft_qwen_14B_v2/checkpoints/checkpoint-500/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..9a9acac1c145d10bf8e654dddb9d3b27bf8bc63b
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-500/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4bf99e0bddcd16fcd5aa4dacbeb664c0e2b144711c917dd805561d1b7f6e0ef
+size 402976786
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-500/rng_state.pth b/sft_qwen_14B_v2/checkpoints/checkpoint-500/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..550896d282d41fa39d529fc7560e5b45982ec083
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-500/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7554aa7bc01964a4c12187b3026a2cce7c932c5c0fd522183b3a9f9414ab4d39
+size 14244
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-500/scheduler.pt b/sft_qwen_14B_v2/checkpoints/checkpoint-500/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..b7ab0fa93800d30f80dae3909114434dedea1740
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-500/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b9ab227797448d6a0b690540001676d252024d08b5dd438fce4fd8a3b96a64c5
+size 1064
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-500/trainer_state.json b/sft_qwen_14B_v2/checkpoints/checkpoint-500/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..f8925ebaf3f3d2069061357d83b31f3f6932801a
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-500/trainer_state.json
@@ -0,0 +1,1833 @@
+{
+  "best_global_step": 500,
+  "best_metric": 1.042096495628357,
+  "best_model_checkpoint": "task2file/sft_qwen_14B_v2/checkpoints/checkpoint-500",
+  "epoch": 0.2109704641350211,
+  "eval_steps": 100,
+  "global_step": 500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0008438818565400844,
+      "grad_norm": 0.5386583805084229,
+      "learning_rate": 1.7574692442882248e-07,
+      "loss": 1.6941628456115723,
+      "step": 2
+    },
+    {
+      "epoch": 0.0016877637130801688,
+      "grad_norm": 0.5477277636528015,
+      "learning_rate": 5.272407732864675e-07,
+      "loss": 1.7132279872894287,
+      "step": 4
+    },
+    {
+      "epoch": 0.002531645569620253,
+      "grad_norm": 0.5390765070915222,
+      "learning_rate": 8.787346221441126e-07,
+      "loss": 1.641180396080017,
+      "step": 6
+    },
+    {
+      "epoch": 0.0033755274261603376,
+      "grad_norm": 0.5023683905601501,
+      "learning_rate": 1.2302284710017575e-06,
+      "loss": 1.5616240501403809,
+      "step": 8
+    },
+    {
+      "epoch": 0.004219409282700422,
+      "grad_norm": 0.4899154603481293,
+      "learning_rate": 1.5817223198594026e-06,
+      "loss": 1.572033405303955,
+      "step": 10
+    },
+    {
+      "epoch": 0.005063291139240506,
+      "grad_norm": 0.5239788293838501,
+      "learning_rate": 1.9332161687170474e-06,
+      "loss": 1.6242921352386475,
+      "step": 12
+    },
+    {
+      "epoch": 0.00590717299578059,
+      "grad_norm": 0.5172926783561707,
+      "learning_rate": 2.2847100175746925e-06,
+      "loss": 1.6800041198730469,
+      "step": 14
+    },
+    {
+      "epoch": 0.006751054852320675,
+      "grad_norm": 0.5539224743843079,
+      "learning_rate": 2.6362038664323376e-06,
+      "loss": 1.6450834274291992,
+      "step": 16
+    },
+    {
+      "epoch": 0.007594936708860759,
+      "grad_norm": 0.5255337953567505,
+      "learning_rate": 2.9876977152899827e-06,
+      "loss": 1.6673263311386108,
+      "step": 18
+    },
+    {
+      "epoch": 0.008438818565400843,
+      "grad_norm": 0.5074548721313477,
+      "learning_rate": 3.3391915641476277e-06,
+      "loss": 1.531802773475647,
+      "step": 20
+    },
+    {
+      "epoch": 0.009282700421940928,
+      "grad_norm": 0.4160279333591461,
+      "learning_rate": 3.6906854130052724e-06,
+      "loss": 1.599354863166809,
+      "step": 22
+    },
+    {
+      "epoch": 0.010126582278481013,
+      "grad_norm": 0.5716474652290344,
+      "learning_rate": 4.0421792618629174e-06,
+      "loss": 1.6700962781906128,
+      "step": 24
+    },
+    {
+      "epoch": 0.010970464135021098,
+      "grad_norm": 0.5148899555206299,
+      "learning_rate": 4.3936731107205625e-06,
+      "loss": 1.66217839717865,
+      "step": 26
+    },
+    {
+      "epoch": 0.01181434599156118,
+      "grad_norm": 0.575722336769104,
+      "learning_rate": 4.7451669595782076e-06,
+      "loss": 1.6692266464233398,
+      "step": 28
+    },
+    {
+      "epoch": 0.012658227848101266,
+      "grad_norm": 0.5345953106880188,
+      "learning_rate": 5.096660808435853e-06,
+      "loss": 1.5518689155578613,
+      "step": 30
+    },
+    {
+      "epoch": 0.01350210970464135,
+      "grad_norm": 0.4462043344974518,
+      "learning_rate": 5.448154657293498e-06,
+      "loss": 1.5930007696151733,
+      "step": 32
+    },
+    {
+      "epoch": 0.014345991561181435,
+      "grad_norm": 0.5119605660438538,
+      "learning_rate": 5.799648506151143e-06,
+      "loss": 1.6069684028625488,
+      "step": 34
+    },
+    {
+      "epoch": 0.015189873417721518,
+      "grad_norm": 0.5328608751296997,
+      "learning_rate": 6.151142355008788e-06,
+      "loss": 1.5838109254837036,
+      "step": 36
+    },
+    {
+      "epoch": 0.016033755274261603,
+      "grad_norm": 0.5065920352935791,
+      "learning_rate": 6.502636203866433e-06,
+      "loss": 1.608130931854248,
+      "step": 38
+    },
+    {
+      "epoch": 0.016877637130801686,
+      "grad_norm": 0.4479359984397888,
+      "learning_rate": 6.854130052724078e-06,
+      "loss": 1.5942182540893555,
+      "step": 40
+    },
+    {
+      "epoch": 0.017721518987341773,
+      "grad_norm": 0.42844903469085693,
+      "learning_rate": 7.205623901581722e-06,
+      "loss": 1.6441553831100464,
+      "step": 42
+    },
+    {
+      "epoch": 0.018565400843881856,
+      "grad_norm": 0.476630836725235,
+      "learning_rate": 7.557117750439367e-06,
+      "loss": 1.6068111658096313,
+      "step": 44
+    },
+    {
+      "epoch": 0.019409282700421943,
+      "grad_norm": 0.4532654881477356,
+      "learning_rate": 7.908611599297012e-06,
+      "loss": 1.6618021726608276,
+      "step": 46
+    },
+    {
+      "epoch": 0.020253164556962026,
+      "grad_norm": 0.3701118230819702,
+      "learning_rate": 8.260105448154657e-06,
+      "loss": 1.4730033874511719,
+      "step": 48
+    },
+    {
+      "epoch": 0.02109704641350211,
+      "grad_norm": 0.38471561670303345,
+      "learning_rate": 8.611599297012302e-06,
+      "loss": 1.4828267097473145,
+      "step": 50
+    },
+    {
+      "epoch": 0.021940928270042195,
+      "grad_norm": 0.3602336347103119,
+      "learning_rate": 8.963093145869948e-06,
+      "loss": 1.3877452611923218,
+      "step": 52
+    },
+    {
+      "epoch": 0.02278481012658228,
+      "grad_norm": 0.40318572521209717,
+      "learning_rate": 9.314586994727593e-06,
+      "loss": 1.49052894115448,
+      "step": 54
+    },
+    {
+      "epoch": 0.02362869198312236,
+      "grad_norm": 0.3223826587200165,
+      "learning_rate": 9.666080843585238e-06,
+      "loss": 1.4912524223327637,
+      "step": 56
+    },
+    {
+      "epoch": 0.024472573839662448,
+      "grad_norm": 0.3873065114021301,
+      "learning_rate": 1.0017574692442883e-05,
+      "loss": 1.526674509048462,
+      "step": 58
+    },
+    {
+      "epoch": 0.02531645569620253,
+      "grad_norm": 0.410159707069397,
+      "learning_rate": 1.0369068541300528e-05,
+      "loss": 1.4480271339416504,
+      "step": 60
+    },
+    {
+      "epoch": 0.026160337552742614,
+      "grad_norm": 0.3632003962993622,
+      "learning_rate": 1.0720562390158173e-05,
+      "loss": 1.4222990274429321,
+      "step": 62
+    },
+    {
+      "epoch": 0.0270042194092827,
+      "grad_norm": 0.33118435740470886,
+      "learning_rate": 1.1072056239015818e-05,
+      "loss": 1.387171745300293,
+      "step": 64
+    },
+    {
+      "epoch": 0.027848101265822784,
+      "grad_norm": 0.3301764726638794,
+      "learning_rate": 1.1423550087873463e-05,
+      "loss": 1.3523777723312378,
+      "step": 66
+    },
+    {
+      "epoch": 0.02869198312236287,
+      "grad_norm": 0.34342435002326965,
+      "learning_rate": 1.1775043936731108e-05,
+      "loss": 1.4515162706375122,
+      "step": 68
+    },
+    {
+      "epoch": 0.029535864978902954,
+      "grad_norm": 0.3243122100830078,
+      "learning_rate": 1.2126537785588753e-05,
+      "loss": 1.3509243726730347,
+      "step": 70
+    },
+    {
+      "epoch": 0.030379746835443037,
+      "grad_norm": 0.3450150787830353,
+      "learning_rate": 1.2478031634446398e-05,
+      "loss": 1.4936245679855347,
+      "step": 72
+    },
+    {
+      "epoch": 0.031223628691983123,
+      "grad_norm": 0.38912028074264526,
+      "learning_rate": 1.2829525483304042e-05,
+      "loss": 1.3419109582901,
+      "step": 74
+    },
+    {
+      "epoch": 0.032067510548523206,
+      "grad_norm": 0.3019310235977173,
+      "learning_rate": 1.3181019332161687e-05,
+      "loss": 1.4284154176712036,
+      "step": 76
+    },
+    {
+      "epoch": 0.03291139240506329,
+      "grad_norm": 0.37803682684898376,
+      "learning_rate": 1.3532513181019332e-05,
+      "loss": 1.4256561994552612,
+      "step": 78
+    },
+    {
+      "epoch": 0.03375527426160337,
+      "grad_norm": 0.34191736578941345,
+      "learning_rate": 1.3884007029876977e-05,
+      "loss": 1.3256909847259521,
+      "step": 80
+    },
+    {
+      "epoch": 0.03459915611814346,
+      "grad_norm": 0.35242700576782227,
+      "learning_rate": 1.4235500878734624e-05,
+      "loss": 1.2710685729980469,
+      "step": 82
+    },
+    {
+      "epoch": 0.035443037974683546,
+      "grad_norm": 0.38094228506088257,
+      "learning_rate": 1.4586994727592269e-05,
+      "loss": 1.253411889076233,
+      "step": 84
+    },
+    {
+      "epoch": 0.036286919831223625,
+      "grad_norm": 0.36837366223335266,
+      "learning_rate": 1.4938488576449914e-05,
+      "loss": 1.3064342737197876,
+      "step": 86
+    },
+    {
+      "epoch": 0.03713080168776371,
+      "grad_norm": 0.3443569242954254,
+      "learning_rate": 1.5289982425307557e-05,
+      "loss": 1.293562412261963,
+      "step": 88
+    },
+    {
+      "epoch": 0.0379746835443038,
+      "grad_norm": 0.3799338936805725,
+      "learning_rate": 1.5641476274165202e-05,
+      "loss": 1.3382648229599,
+      "step": 90
+    },
+    {
+      "epoch": 0.038818565400843885,
+      "grad_norm": 0.40501922369003296,
+      "learning_rate": 1.599297012302285e-05,
+      "loss": 1.3925724029541016,
+      "step": 92
+    },
+    {
+      "epoch": 0.039662447257383965,
+      "grad_norm": 0.4419630467891693,
+      "learning_rate": 1.6344463971880492e-05,
+      "loss": 1.357171893119812,
+      "step": 94
+    },
+    {
+      "epoch": 0.04050632911392405,
+      "grad_norm": 0.3619817793369293,
+      "learning_rate": 1.6695957820738137e-05,
+      "loss": 1.3029985427856445,
+      "step": 96
+    },
+    {
+      "epoch": 0.04135021097046414,
+      "grad_norm": 0.4851357340812683,
+      "learning_rate": 1.7047451669595782e-05,
+      "loss": 1.3498191833496094,
+      "step": 98
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "grad_norm": 0.418658584356308,
+      "learning_rate": 1.7398945518453427e-05,
+      "loss": 1.185287356376648,
+      "step": 100
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "eval_loss": 1.2979938983917236,
+      "eval_runtime": 682.1979,
+      "eval_samples_per_second": 3.089,
+      "eval_steps_per_second": 3.089,
+      "step": 100
+    },
+    {
+      "epoch": 0.043037974683544304,
+      "grad_norm": 0.4464418888092041,
+      "learning_rate": 1.7750439367311073e-05,
+      "loss": 1.2217272520065308,
+      "step": 102
+    },
+    {
+      "epoch": 0.04388185654008439,
+      "grad_norm": 0.4706237316131592,
+      "learning_rate": 1.8101933216168718e-05,
+      "loss": 1.2052050828933716,
+      "step": 104
+    },
+    {
+      "epoch": 0.04472573839662447,
+      "grad_norm": 0.46394404768943787,
+      "learning_rate": 1.8453427065026363e-05,
+      "loss": 1.221343994140625,
+      "step": 106
+    },
+    {
+      "epoch": 0.04556962025316456,
+      "grad_norm": 0.4726889431476593,
+      "learning_rate": 1.8804920913884008e-05,
+      "loss": 1.2387475967407227,
+      "step": 108
+    },
+    {
+      "epoch": 0.046413502109704644,
+      "grad_norm": 0.42130985856056213,
+      "learning_rate": 1.9156414762741653e-05,
+      "loss": 1.2851309776306152,
+      "step": 110
+    },
+    {
+      "epoch": 0.04725738396624472,
+      "grad_norm": 0.4504576623439789,
+      "learning_rate": 1.9507908611599298e-05,
+      "loss": 1.2753145694732666,
+      "step": 112
+    },
+    {
+      "epoch": 0.04810126582278481,
+      "grad_norm": 0.396085262298584,
+      "learning_rate": 1.9859402460456943e-05,
+      "loss": 1.2427717447280884,
+      "step": 114
+    },
+    {
+      "epoch": 0.048945147679324896,
+      "grad_norm": 0.5106491446495056,
+      "learning_rate": 2.0210896309314588e-05,
+      "loss": 1.2943825721740723,
+      "step": 116
+    },
+    {
+      "epoch": 0.049789029535864976,
+      "grad_norm": 0.42351317405700684,
+      "learning_rate": 2.0562390158172233e-05,
+      "loss": 1.263301134109497,
+      "step": 118
+    },
+    {
+      "epoch": 0.05063291139240506,
+      "grad_norm": 0.4403539299964905,
+      "learning_rate": 2.0913884007029878e-05,
+      "loss": 1.2647849321365356,
+      "step": 120
+    },
+    {
+      "epoch": 0.05147679324894515,
+      "grad_norm": 0.5260752439498901,
+      "learning_rate": 2.1265377855887523e-05,
+      "loss": 1.2351393699645996,
+      "step": 122
+    },
+    {
+      "epoch": 0.05232067510548523,
+      "grad_norm": 0.44978851079940796,
+      "learning_rate": 2.1616871704745168e-05,
+      "loss": 1.0384471416473389,
+      "step": 124
+    },
+    {
+      "epoch": 0.053164556962025315,
+      "grad_norm": 0.47732362151145935,
+      "learning_rate": 2.1968365553602813e-05,
+      "loss": 1.1518068313598633,
+      "step": 126
+    },
+    {
+      "epoch": 0.0540084388185654,
+      "grad_norm": 0.5473551750183105,
+      "learning_rate": 2.231985940246046e-05,
+      "loss": 1.2264912128448486,
+      "step": 128
+    },
+    {
+      "epoch": 0.05485232067510549,
+      "grad_norm": 0.4473855197429657,
+      "learning_rate": 2.2671353251318103e-05,
+      "loss": 1.1615246534347534,
+      "step": 130
+    },
+    {
+      "epoch": 0.05569620253164557,
+      "grad_norm": 0.5980377197265625,
+      "learning_rate": 2.302284710017575e-05,
+      "loss": 1.1334880590438843,
+      "step": 132
+    },
+    {
+      "epoch": 0.056540084388185655,
+      "grad_norm": 0.5987792015075684,
+      "learning_rate": 2.3374340949033394e-05,
+      "loss": 1.1546804904937744,
+      "step": 134
+    },
+    {
+      "epoch": 0.05738396624472574,
+      "grad_norm": 0.45355498790740967,
+      "learning_rate": 2.372583479789104e-05,
+      "loss": 1.194953441619873,
+      "step": 136
+    },
+    {
+      "epoch": 0.05822784810126582,
+      "grad_norm": 0.5373698472976685,
+      "learning_rate": 2.4077328646748684e-05,
+      "loss": 1.1067466735839844,
+      "step": 138
+    },
+    {
+      "epoch": 0.05907172995780591,
+      "grad_norm": 0.48734328150749207,
+      "learning_rate": 2.442882249560633e-05,
+      "loss": 1.188468098640442,
+      "step": 140
+    },
+    {
+      "epoch": 0.059915611814345994,
+      "grad_norm": 0.4692173898220062,
+      "learning_rate": 2.478031634446397e-05,
+      "loss": 1.1624362468719482,
+      "step": 142
+    },
+    {
+      "epoch": 0.060759493670886074,
+      "grad_norm": 0.532554030418396,
+      "learning_rate": 2.513181019332162e-05,
+      "loss": 1.0978907346725464,
+      "step": 144
+    },
+    {
+      "epoch": 0.06160337552742616,
+      "grad_norm": 0.5853802561759949,
+      "learning_rate": 2.5483304042179264e-05,
+      "loss": 1.2030781507492065,
+      "step": 146
+    },
+    {
+      "epoch": 0.06244725738396625,
+      "grad_norm": 0.5061611533164978,
+      "learning_rate": 2.583479789103691e-05,
+      "loss": 1.082366943359375,
+      "step": 148
+    },
+    {
+      "epoch": 0.06329113924050633,
+      "grad_norm": 0.49426141381263733,
+      "learning_rate": 2.6186291739894554e-05,
+      "loss": 1.10564386844635,
+      "step": 150
+    },
+    {
+      "epoch": 0.06413502109704641,
+      "grad_norm": 0.5846618413925171,
+      "learning_rate": 2.6537785588752196e-05,
+      "loss": 1.1992807388305664,
+      "step": 152
+    },
+    {
+      "epoch": 0.06497890295358649,
+      "grad_norm": 0.5517552495002747,
+      "learning_rate": 2.6889279437609844e-05,
+      "loss": 1.1757566928863525,
+      "step": 154
+    },
+    {
+      "epoch": 0.06582278481012659,
+      "grad_norm": 0.5667305588722229,
+      "learning_rate": 2.7240773286467486e-05,
+      "loss": 1.0548783540725708,
+      "step": 156
+    },
+    {
+      "epoch": 0.06666666666666667,
+      "grad_norm": 0.6760414242744446,
+      "learning_rate": 2.7592267135325134e-05,
+      "loss": 1.184364914894104,
+      "step": 158
+    },
+    {
+      "epoch": 0.06751054852320675,
+      "grad_norm": 0.5261430740356445,
+      "learning_rate": 2.7943760984182776e-05,
+      "loss": 1.1945042610168457,
+      "step": 160
+    },
+    {
+      "epoch": 0.06835443037974684,
+      "grad_norm": 0.6155015230178833,
+      "learning_rate": 2.8295254833040425e-05,
+      "loss": 1.2021973133087158,
+      "step": 162
+    },
+    {
+      "epoch": 0.06919831223628692,
+      "grad_norm": 0.6131619215011597,
+      "learning_rate": 2.8646748681898066e-05,
+      "loss": 1.144123911857605,
+      "step": 164
+    },
+    {
+      "epoch": 0.070042194092827,
+      "grad_norm": 0.5749185681343079,
+      "learning_rate": 2.8998242530755715e-05,
+      "loss": 1.1329256296157837,
+      "step": 166
+    },
+    {
+      "epoch": 0.07088607594936709,
+      "grad_norm": 0.5243118405342102,
+      "learning_rate": 2.9349736379613356e-05,
+      "loss": 1.0892387628555298,
+      "step": 168
+    },
+    {
+      "epoch": 0.07172995780590717,
+      "grad_norm": 0.7190104722976685,
+      "learning_rate": 2.9701230228471005e-05,
+      "loss": 1.163260817527771,
+      "step": 170
+    },
+    {
+      "epoch": 0.07257383966244725,
+      "grad_norm": 0.5486982464790344,
+      "learning_rate": 3.0052724077328647e-05,
+      "loss": 1.0880777835845947,
+      "step": 172
+    },
+    {
+      "epoch": 0.07341772151898734,
+      "grad_norm": 0.5020889043807983,
+      "learning_rate": 3.0404217926186295e-05,
+      "loss": 1.0433368682861328,
+      "step": 174
+    },
+    {
+      "epoch": 0.07426160337552742,
+      "grad_norm": 0.47329774498939514,
+      "learning_rate": 3.075571177504394e-05,
+      "loss": 1.0528991222381592,
+      "step": 176
+    },
+    {
+      "epoch": 0.0751054852320675,
+      "grad_norm": 0.6635547876358032,
+      "learning_rate": 3.110720562390158e-05,
+      "loss": 1.1627811193466187,
+      "step": 178
+    },
+    {
+      "epoch": 0.0759493670886076,
+      "grad_norm": 0.5624618530273438,
+      "learning_rate": 3.145869947275923e-05,
+      "loss": 1.084869384765625,
+      "step": 180
+    },
+    {
+      "epoch": 0.07679324894514768,
+      "grad_norm": 0.6029536724090576,
+      "learning_rate": 3.181019332161687e-05,
+      "loss": 1.2227671146392822,
+      "step": 182
+    },
+    {
+      "epoch": 0.07763713080168777,
+      "grad_norm": 0.930959939956665,
+      "learning_rate": 3.216168717047452e-05,
+      "loss": 1.0955452919006348,
+      "step": 184
+    },
+    {
+      "epoch": 0.07848101265822785,
+      "grad_norm": 0.5326952338218689,
+      "learning_rate": 3.251318101933216e-05,
+      "loss": 1.0640798807144165,
+      "step": 186
+    },
+    {
+      "epoch": 0.07932489451476793,
+      "grad_norm": 0.5484727621078491,
+      "learning_rate": 3.286467486818981e-05,
+      "loss": 1.0700589418411255,
+      "step": 188
+    },
+    {
+      "epoch": 0.08016877637130802,
+      "grad_norm": 0.605273425579071,
+      "learning_rate": 3.3216168717047456e-05,
+      "loss": 1.1593081951141357,
+      "step": 190
+    },
+    {
+      "epoch": 0.0810126582278481,
+      "grad_norm": 0.5704394578933716,
+      "learning_rate": 3.35676625659051e-05,
+      "loss": 1.1617076396942139,
+      "step": 192
+    },
+    {
+      "epoch": 0.08185654008438818,
+      "grad_norm": 0.5929452180862427,
+      "learning_rate": 3.3919156414762746e-05,
+      "loss": 1.1346839666366577,
+      "step": 194
+    },
+    {
+      "epoch": 0.08270042194092828,
+      "grad_norm": 0.5624077916145325,
+      "learning_rate": 3.427065026362039e-05,
+      "loss": 1.0934710502624512,
+      "step": 196
+    },
+    {
+      "epoch": 0.08354430379746836,
+      "grad_norm": 0.6717425584793091,
+      "learning_rate": 3.4622144112478036e-05,
+      "loss": 1.1810534000396729,
+      "step": 198
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "grad_norm": 0.5120199918746948,
+      "learning_rate": 3.4973637961335674e-05,
+      "loss": 1.1525514125823975,
+      "step": 200
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "eval_loss": 1.142486810684204,
+      "eval_runtime": 668.2356,
+      "eval_samples_per_second": 3.153,
+      "eval_steps_per_second": 3.153,
+      "step": 200
+    },
+    {
+      "epoch": 0.08523206751054853,
+      "grad_norm": 0.5144487023353577,
+      "learning_rate": 3.5325131810193326e-05,
+      "loss": 1.0243735313415527,
+      "step": 202
+    },
+    {
+      "epoch": 0.08607594936708861,
+      "grad_norm": 0.6325069069862366,
+      "learning_rate": 3.5676625659050964e-05,
+      "loss": 1.118743896484375,
+      "step": 204
+    },
+    {
+      "epoch": 0.08691983122362869,
+      "grad_norm": 0.5501633882522583,
+      "learning_rate": 3.6028119507908616e-05,
+      "loss": 1.0380504131317139,
+      "step": 206
+    },
+    {
+      "epoch": 0.08776371308016878,
+      "grad_norm": 0.6133899688720703,
+      "learning_rate": 3.6379613356766254e-05,
+      "loss": 0.9837555885314941,
+      "step": 208
+    },
+    {
+      "epoch": 0.08860759493670886,
+      "grad_norm": 0.5799810886383057,
+      "learning_rate": 3.6731107205623906e-05,
+      "loss": 1.090720295906067,
+      "step": 210
+    },
+    {
+      "epoch": 0.08945147679324894,
+      "grad_norm": 0.6039511561393738,
+      "learning_rate": 3.7082601054481544e-05,
+      "loss": 1.120232343673706,
+      "step": 212
+    },
+    {
+      "epoch": 0.09029535864978903,
+      "grad_norm": 0.5983024835586548,
+      "learning_rate": 3.7434094903339196e-05,
+      "loss": 1.096949815750122,
+      "step": 214
+    },
+    {
+      "epoch": 0.09113924050632911,
+      "grad_norm": 0.5641079545021057,
+      "learning_rate": 3.7785588752196835e-05,
+      "loss": 1.1226298809051514,
+      "step": 216
+    },
+    {
+      "epoch": 0.0919831223628692,
+      "grad_norm": 0.655717134475708,
+      "learning_rate": 3.8137082601054486e-05,
+      "loss": 1.1260643005371094,
+      "step": 218
+    },
+    {
+      "epoch": 0.09282700421940929,
+      "grad_norm": 0.6111898422241211,
+      "learning_rate": 3.848857644991213e-05,
+      "loss": 1.0777709484100342,
+      "step": 220
+    },
+    {
+      "epoch": 0.09367088607594937,
+      "grad_norm": 0.6821302771568298,
+      "learning_rate": 3.884007029876977e-05,
+      "loss": 1.10588800907135,
+      "step": 222
+    },
+    {
+      "epoch": 0.09451476793248945,
+      "grad_norm": 0.693175733089447,
+      "learning_rate": 3.919156414762742e-05,
+      "loss": 1.1498671770095825,
+      "step": 224
+    },
+    {
+      "epoch": 0.09535864978902954,
+      "grad_norm": 0.5288166403770447,
+      "learning_rate": 3.954305799648506e-05,
+      "loss": 1.0587562322616577,
+      "step": 226
+    },
+    {
+      "epoch": 0.09620253164556962,
+      "grad_norm": 0.6882867813110352,
+      "learning_rate": 3.989455184534271e-05,
+      "loss": 1.1107512712478638,
+      "step": 228
+    },
+    {
+      "epoch": 0.0970464135021097,
+      "grad_norm": 0.5834154486656189,
+      "learning_rate": 4.024604569420035e-05,
+      "loss": 1.020510196685791,
+      "step": 230
+    },
+    {
+      "epoch": 0.09789029535864979,
+      "grad_norm": 0.7157064080238342,
+      "learning_rate": 4.0597539543058e-05,
+      "loss": 1.0642449855804443,
+      "step": 232
+    },
+    {
+      "epoch": 0.09873417721518987,
+      "grad_norm": 0.6530708074569702,
+      "learning_rate": 4.094903339191564e-05,
+      "loss": 1.0359872579574585,
+      "step": 234
+    },
+    {
+      "epoch": 0.09957805907172995,
+      "grad_norm": 0.6329686045646667,
+      "learning_rate": 4.130052724077329e-05,
+      "loss": 1.050504446029663,
+      "step": 236
+    },
+    {
+      "epoch": 0.10042194092827005,
+      "grad_norm": 0.6597026586532593,
+      "learning_rate": 4.165202108963093e-05,
+      "loss": 1.2621175050735474,
+      "step": 238
+    },
+    {
+      "epoch": 0.10126582278481013,
+      "grad_norm": 0.6195225119590759,
+      "learning_rate": 4.200351493848858e-05,
+      "loss": 1.1218310594558716,
+      "step": 240
+    },
+    {
+      "epoch": 0.1021097046413502,
+      "grad_norm": 0.6764137744903564,
+      "learning_rate": 4.235500878734622e-05,
+      "loss": 1.1250728368759155,
+      "step": 242
+    },
+    {
+      "epoch": 0.1029535864978903,
+      "grad_norm": 0.552363395690918,
+      "learning_rate": 4.270650263620387e-05,
+      "loss": 1.028212308883667,
+      "step": 244
+    },
+    {
+      "epoch": 0.10379746835443038,
+      "grad_norm": 0.5620495676994324,
+      "learning_rate": 4.305799648506151e-05,
+      "loss": 1.0425450801849365,
+      "step": 246
+    },
+    {
+      "epoch": 0.10464135021097046,
+      "grad_norm": 0.6860032081604004,
+      "learning_rate": 4.3409490333919156e-05,
+      "loss": 1.144278883934021,
+      "step": 248
+    },
+    {
+      "epoch": 0.10548523206751055,
+      "grad_norm": 0.6033259034156799,
+      "learning_rate": 4.37609841827768e-05,
+      "loss": 1.1223982572555542,
+      "step": 250
+    },
+    {
+      "epoch": 0.10632911392405063,
+      "grad_norm": 0.6292146444320679,
+      "learning_rate": 4.4112478031634446e-05,
+      "loss": 1.1609960794448853,
+      "step": 252
+    },
+    {
+      "epoch": 0.10717299578059072,
+      "grad_norm": 0.7982883453369141,
+      "learning_rate": 4.44639718804921e-05,
+      "loss": 1.063547968864441,
+      "step": 254
+    },
+    {
+      "epoch": 0.1080168776371308,
+      "grad_norm": 0.7719110250473022,
+      "learning_rate": 4.4815465729349736e-05,
+      "loss": 1.0719804763793945,
+      "step": 256
+    },
+    {
+      "epoch": 0.10886075949367088,
+      "grad_norm": 0.6101011633872986,
+      "learning_rate": 4.516695957820739e-05,
+      "loss": 1.0778400897979736,
+      "step": 258
+    },
+    {
+      "epoch": 0.10970464135021098,
+      "grad_norm": 0.7300994396209717,
+      "learning_rate": 4.5518453427065026e-05,
+      "loss": 1.2129558324813843,
+      "step": 260
+    },
+    {
+      "epoch": 0.11054852320675106,
+      "grad_norm": 0.8348747491836548,
+      "learning_rate": 4.586994727592268e-05,
+      "loss": 1.221714735031128,
+      "step": 262
+    },
+    {
+      "epoch": 0.11139240506329114,
+      "grad_norm": 0.5445612072944641,
+      "learning_rate": 4.6221441124780316e-05,
+      "loss": 1.0187978744506836,
+      "step": 264
+    },
+    {
+      "epoch": 0.11223628691983123,
+      "grad_norm": 0.6230319738388062,
+      "learning_rate": 4.657293497363797e-05,
+      "loss": 1.096561312675476,
+      "step": 266
+    },
+    {
+      "epoch": 0.11308016877637131,
+      "grad_norm": 0.6231237649917603,
+      "learning_rate": 4.6924428822495606e-05,
+      "loss": 1.089842438697815,
+      "step": 268
+    },
+    {
+      "epoch": 0.11392405063291139,
+      "grad_norm": 0.7178627252578735,
+      "learning_rate": 4.727592267135325e-05,
+      "loss": 1.0696645975112915,
+      "step": 270
+    },
+    {
+      "epoch": 0.11476793248945148,
+      "grad_norm": 0.6895854473114014,
+      "learning_rate": 4.7627416520210896e-05,
+      "loss": 1.0511361360549927,
+      "step": 272
+    },
+    {
+      "epoch": 0.11561181434599156,
+      "grad_norm": 0.6046878695487976,
+      "learning_rate": 4.797891036906854e-05,
+      "loss": 1.1373958587646484,
+      "step": 274
+    },
+    {
+      "epoch": 0.11645569620253164,
+      "grad_norm": 0.6524552702903748,
+      "learning_rate": 4.833040421792619e-05,
+      "loss": 1.0734186172485352,
+      "step": 276
+    },
+    {
+      "epoch": 0.11729957805907174,
+      "grad_norm": 0.6331019997596741,
+      "learning_rate": 4.868189806678383e-05,
+      "loss": 1.123913049697876,
+      "step": 278
+    },
+    {
+      "epoch": 0.11814345991561181,
+      "grad_norm": 0.5919018983840942,
+      "learning_rate": 4.903339191564148e-05,
+      "loss": 1.0635710954666138,
+      "step": 280
+    },
+    {
+      "epoch": 0.1189873417721519,
+      "grad_norm": 0.6067633032798767,
+      "learning_rate": 4.938488576449912e-05,
+      "loss": 1.0429247617721558,
+      "step": 282
+    },
+    {
+      "epoch": 0.11983122362869199,
+      "grad_norm": 0.6583750247955322,
+      "learning_rate": 4.9736379613356774e-05,
+      "loss": 1.1397464275360107,
+      "step": 284
+    },
+    {
+      "epoch": 0.12067510548523207,
+      "grad_norm": 0.6200069785118103,
+      "learning_rate": 5.008787346221442e-05,
+      "loss": 1.0590803623199463,
+      "step": 286
+    },
+    {
+      "epoch": 0.12151898734177215,
+      "grad_norm": 0.6798665523529053,
+      "learning_rate": 5.0439367311072064e-05,
+      "loss": 1.1318789720535278,
+      "step": 288
+    },
+    {
+      "epoch": 0.12236286919831224,
+      "grad_norm": 0.7508794069290161,
+      "learning_rate": 5.07908611599297e-05,
+      "loss": 1.0934956073760986,
+      "step": 290
+    },
+    {
+      "epoch": 0.12320675105485232,
+      "grad_norm": 0.6901452541351318,
+      "learning_rate": 5.114235500878735e-05,
+      "loss": 1.163407802581787,
+      "step": 292
+    },
+    {
+      "epoch": 0.1240506329113924,
+      "grad_norm": 0.6423285603523254,
+      "learning_rate": 5.1493848857644985e-05,
+      "loss": 1.09059476852417,
+      "step": 294
+    },
+    {
+      "epoch": 0.1248945147679325,
+      "grad_norm": 0.6839275360107422,
+      "learning_rate": 5.1845342706502644e-05,
+      "loss": 1.0690211057662964,
+      "step": 296
+    },
+    {
+      "epoch": 0.1257383966244726,
+      "grad_norm": 0.6350128054618835,
+      "learning_rate": 5.219683655536028e-05,
+      "loss": 0.982322096824646,
+      "step": 298
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "grad_norm": 0.7136530876159668,
+      "learning_rate": 5.254833040421793e-05,
+      "loss": 1.1132930517196655,
+      "step": 300
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "eval_loss": 1.0952109098434448,
+      "eval_runtime": 677.0652,
+      "eval_samples_per_second": 3.112,
+      "eval_steps_per_second": 3.112,
+      "step": 300
+    },
+    {
+      "epoch": 0.12742616033755275,
+      "grad_norm": 0.7339721322059631,
+      "learning_rate": 5.289982425307557e-05,
+      "loss": 0.973595917224884,
+      "step": 302
+    },
+    {
+      "epoch": 0.12827004219409283,
+      "grad_norm": 0.5941481590270996,
+      "learning_rate": 5.3251318101933224e-05,
+      "loss": 0.9819849729537964,
+      "step": 304
+    },
+    {
+      "epoch": 0.1291139240506329,
+      "grad_norm": 0.7153938412666321,
+      "learning_rate": 5.360281195079086e-05,
+      "loss": 1.0315470695495605,
+      "step": 306
+    },
+    {
+      "epoch": 0.12995780590717299,
+      "grad_norm": 0.5167180299758911,
+      "learning_rate": 5.395430579964851e-05,
+      "loss": 0.9492001533508301,
+      "step": 308
+    },
+    {
+      "epoch": 0.1308016877637131,
+      "grad_norm": 0.6055944561958313,
+      "learning_rate": 5.430579964850615e-05,
+      "loss": 1.0156209468841553,
+      "step": 310
+    },
+    {
+      "epoch": 0.13164556962025317,
+      "grad_norm": 0.7662386298179626,
+      "learning_rate": 5.4657293497363805e-05,
+      "loss": 1.1791651248931885,
+      "step": 312
+    },
+    {
+      "epoch": 0.13248945147679325,
+      "grad_norm": 0.6065546274185181,
+      "learning_rate": 5.500878734622145e-05,
+      "loss": 1.0009297132492065,
+      "step": 314
+    },
+    {
+      "epoch": 0.13333333333333333,
+      "grad_norm": 0.604225754737854,
+      "learning_rate": 5.536028119507909e-05,
+      "loss": 1.0208244323730469,
+      "step": 316
+    },
+    {
+      "epoch": 0.1341772151898734,
+      "grad_norm": 0.6186763048171997,
+      "learning_rate": 5.571177504393673e-05,
+      "loss": 0.9968416690826416,
+      "step": 318
+    },
+    {
+      "epoch": 0.1350210970464135,
+      "grad_norm": 0.7100363969802856,
+      "learning_rate": 5.606326889279437e-05,
+      "loss": 0.9540256857872009,
+      "step": 320
+    },
+    {
+      "epoch": 0.1358649789029536,
+      "grad_norm": 0.6979711055755615,
+      "learning_rate": 5.641476274165203e-05,
+      "loss": 1.0631953477859497,
+      "step": 322
+    },
+    {
+      "epoch": 0.13670886075949368,
+      "grad_norm": 0.6237109303474426,
+      "learning_rate": 5.676625659050967e-05,
+      "loss": 1.0170501470565796,
+      "step": 324
+    },
+    {
+      "epoch": 0.13755274261603376,
+      "grad_norm": 0.6525548696517944,
+      "learning_rate": 5.711775043936731e-05,
+      "loss": 1.0715603828430176,
+      "step": 326
+    },
+    {
+      "epoch": 0.13839662447257384,
+      "grad_norm": 0.6869221329689026,
+      "learning_rate": 5.746924428822495e-05,
+      "loss": 1.0111541748046875,
+      "step": 328
+    },
+    {
+      "epoch": 0.13924050632911392,
+      "grad_norm": 0.553188145160675,
+      "learning_rate": 5.782073813708261e-05,
+      "loss": 1.0311682224273682,
+      "step": 330
+    },
+    {
+      "epoch": 0.140084388185654,
+      "grad_norm": 0.6760852932929993,
+      "learning_rate": 5.817223198594025e-05,
+      "loss": 1.0213634967803955,
+      "step": 332
+    },
+    {
+      "epoch": 0.1409282700421941,
+      "grad_norm": 0.5907419919967651,
+      "learning_rate": 5.8523725834797894e-05,
+      "loss": 0.9748594164848328,
+      "step": 334
+    },
+    {
+      "epoch": 0.14177215189873418,
+      "grad_norm": 0.7044920921325684,
+      "learning_rate": 5.887521968365554e-05,
+      "loss": 1.05863356590271,
+      "step": 336
+    },
+    {
+      "epoch": 0.14261603375527426,
+      "grad_norm": 0.679073691368103,
+      "learning_rate": 5.922671353251318e-05,
+      "loss": 1.1341127157211304,
+      "step": 338
+    },
+    {
+      "epoch": 0.14345991561181434,
+      "grad_norm": 0.7676237225532532,
+      "learning_rate": 5.957820738137083e-05,
+      "loss": 0.9540836215019226,
+      "step": 340
+    },
+    {
+      "epoch": 0.14430379746835442,
+      "grad_norm": 0.6313899755477905,
+      "learning_rate": 5.9929701230228474e-05,
+      "loss": 1.0585911273956299,
+      "step": 342
+    },
+    {
+      "epoch": 0.1451476793248945,
+      "grad_norm": 0.7123099565505981,
+      "learning_rate": 6.028119507908612e-05,
+      "loss": 1.0760118961334229,
+      "step": 344
+    },
+    {
+      "epoch": 0.1459915611814346,
+      "grad_norm": 0.585935652256012,
+      "learning_rate": 6.063268892794376e-05,
+      "loss": 1.036866307258606,
+      "step": 346
+    },
+    {
+      "epoch": 0.1468354430379747,
+      "grad_norm": 0.5643263459205627,
+      "learning_rate": 6.0984182776801416e-05,
+      "loss": 1.0242938995361328,
+      "step": 348
+    },
+    {
+      "epoch": 0.14767932489451477,
+      "grad_norm": 0.626761794090271,
+      "learning_rate": 6.133567662565906e-05,
+      "loss": 1.0497376918792725,
+      "step": 350
+    },
+    {
+      "epoch": 0.14852320675105485,
+      "grad_norm": 0.5106956958770752,
+      "learning_rate": 6.16871704745167e-05,
+      "loss": 0.9811885356903076,
+      "step": 352
+    },
+    {
+      "epoch": 0.14936708860759493,
+      "grad_norm": 0.6948089003562927,
+      "learning_rate": 6.203866432337434e-05,
+      "loss": 1.0715330839157104,
+      "step": 354
+    },
+    {
+      "epoch": 0.150210970464135,
+      "grad_norm": 0.699713945388794,
+      "learning_rate": 6.239015817223199e-05,
+      "loss": 1.0405226945877075,
+      "step": 356
+    },
+    {
+      "epoch": 0.15105485232067511,
+      "grad_norm": 0.6437667012214661,
+      "learning_rate": 6.274165202108964e-05,
+      "loss": 1.0490930080413818,
+      "step": 358
+    },
+    {
+      "epoch": 0.1518987341772152,
+      "grad_norm": 0.6952699422836304,
+      "learning_rate": 6.309314586994728e-05,
+      "loss": 0.9267548322677612,
+      "step": 360
+    },
+    {
+      "epoch": 0.15274261603375527,
+      "grad_norm": 0.6713186502456665,
+      "learning_rate": 6.344463971880492e-05,
+      "loss": 1.0427420139312744,
+      "step": 362
+    },
+    {
+      "epoch": 0.15358649789029535,
+      "grad_norm": 0.6750379800796509,
+      "learning_rate": 6.379613356766257e-05,
+      "loss": 1.048950433731079,
+      "step": 364
+    },
+    {
+      "epoch": 0.15443037974683543,
+      "grad_norm": 0.6053379774093628,
+      "learning_rate": 6.414762741652022e-05,
+      "loss": 1.0156004428863525,
+      "step": 366
+    },
+    {
+      "epoch": 0.15527426160337554,
+      "grad_norm": 0.8063633441925049,
+      "learning_rate": 6.449912126537786e-05,
+      "loss": 1.0020819902420044,
+      "step": 368
+    },
+    {
+      "epoch": 0.15611814345991562,
+      "grad_norm": 0.8027494549751282,
+      "learning_rate": 6.48506151142355e-05,
+      "loss": 1.055633783340454,
+      "step": 370
+    },
+    {
+      "epoch": 0.1569620253164557,
+      "grad_norm": 0.6580121517181396,
+      "learning_rate": 6.520210896309315e-05,
+      "loss": 1.0149940252304077,
+      "step": 372
+    },
+    {
+      "epoch": 0.15780590717299578,
+      "grad_norm": 0.6561233997344971,
+      "learning_rate": 6.55536028119508e-05,
+      "loss": 0.9769611954689026,
+      "step": 374
+    },
+    {
+      "epoch": 0.15864978902953586,
+      "grad_norm": 0.6444346308708191,
+      "learning_rate": 6.590509666080844e-05,
+      "loss": 0.9099349975585938,
+      "step": 376
+    },
+    {
+      "epoch": 0.15949367088607594,
+      "grad_norm": 0.5879359245300293,
+      "learning_rate": 6.625659050966608e-05,
+      "loss": 1.0797548294067383,
+      "step": 378
+    },
+    {
+      "epoch": 0.16033755274261605,
+      "grad_norm": 0.6994144916534424,
+      "learning_rate": 6.660808435852373e-05,
+      "loss": 1.0336791276931763,
+      "step": 380
+    },
+    {
+      "epoch": 0.16118143459915613,
+      "grad_norm": 0.6128669381141663,
+      "learning_rate": 6.695957820738138e-05,
+      "loss": 1.018118143081665,
+      "step": 382
+    },
+    {
+      "epoch": 0.1620253164556962,
+      "grad_norm": 1.0237540006637573,
+      "learning_rate": 6.731107205623902e-05,
+      "loss": 1.1405497789382935,
+      "step": 384
+    },
+    {
+      "epoch": 0.16286919831223629,
+      "grad_norm": 0.6091578006744385,
+      "learning_rate": 6.766256590509666e-05,
+      "loss": 1.0314189195632935,
+      "step": 386
+    },
+    {
+      "epoch": 0.16371308016877636,
+      "grad_norm": 0.5916037559509277,
+      "learning_rate": 6.801405975395431e-05,
+      "loss": 0.9564052820205688,
+      "step": 388
+    },
+    {
+      "epoch": 0.16455696202531644,
+      "grad_norm": 0.771653950214386,
+      "learning_rate": 6.836555360281195e-05,
+      "loss": 1.0023859739303589,
+      "step": 390
+    },
+    {
+      "epoch": 0.16540084388185655,
+      "grad_norm": 0.654658317565918,
+      "learning_rate": 6.87170474516696e-05,
+      "loss": 1.07024085521698,
+      "step": 392
+    },
+    {
+      "epoch": 0.16624472573839663,
+      "grad_norm": 0.6611968874931335,
+      "learning_rate": 6.906854130052724e-05,
+      "loss": 1.0552500486373901,
+      "step": 394
+    },
+    {
+      "epoch": 0.1670886075949367,
+      "grad_norm": 0.6955893039703369,
+      "learning_rate": 6.942003514938489e-05,
+      "loss": 1.0562875270843506,
+      "step": 396
+    },
+    {
+      "epoch": 0.1679324894514768,
+      "grad_norm": 0.6666058301925659,
+      "learning_rate": 6.977152899824253e-05,
+      "loss": 0.9850592017173767,
+      "step": 398
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "grad_norm": 0.6131711006164551,
+      "learning_rate": 7.012302284710018e-05,
+      "loss": 1.0077755451202393,
+      "step": 400
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "eval_loss": 1.0625108480453491,
+      "eval_runtime": 691.0068,
+      "eval_samples_per_second": 3.049,
+      "eval_steps_per_second": 3.049,
+      "step": 400
+    },
+    {
+      "epoch": 0.16962025316455695,
+      "grad_norm": 0.6286499500274658,
+      "learning_rate": 7.047451669595783e-05,
+      "loss": 1.1012427806854248,
+      "step": 402
+    },
+    {
+      "epoch": 0.17046413502109706,
+      "grad_norm": 0.6639351844787598,
+      "learning_rate": 7.082601054481547e-05,
+      "loss": 1.0379719734191895,
+      "step": 404
+    },
+    {
+      "epoch": 0.17130801687763714,
+      "grad_norm": 0.750401496887207,
+      "learning_rate": 7.117750439367311e-05,
+      "loss": 1.031856656074524,
+      "step": 406
+    },
+    {
+      "epoch": 0.17215189873417722,
+      "grad_norm": 0.8084847331047058,
+      "learning_rate": 7.152899824253075e-05,
+      "loss": 1.0493193864822388,
+      "step": 408
+    },
+    {
+      "epoch": 0.1729957805907173,
+      "grad_norm": 0.7448462247848511,
+      "learning_rate": 7.188049209138841e-05,
+      "loss": 1.1012418270111084,
+      "step": 410
+    },
+    {
+      "epoch": 0.17383966244725738,
+      "grad_norm": 0.5841867923736572,
+      "learning_rate": 7.223198594024605e-05,
+      "loss": 0.9926692247390747,
+      "step": 412
+    },
+    {
+      "epoch": 0.17468354430379746,
+      "grad_norm": 0.7125606536865234,
+      "learning_rate": 7.258347978910369e-05,
+      "loss": 1.0588877201080322,
+      "step": 414
+    },
+    {
+      "epoch": 0.17552742616033756,
+      "grad_norm": 0.5750942230224609,
+      "learning_rate": 7.293497363796134e-05,
+      "loss": 1.038270354270935,
+      "step": 416
+    },
+    {
+      "epoch": 0.17637130801687764,
+      "grad_norm": 0.565444827079773,
+      "learning_rate": 7.328646748681899e-05,
+      "loss": 0.9843021035194397,
+      "step": 418
+    },
+    {
+      "epoch": 0.17721518987341772,
+      "grad_norm": 0.5825693011283875,
+      "learning_rate": 7.363796133567663e-05,
+      "loss": 1.0731632709503174,
+      "step": 420
+    },
+    {
+      "epoch": 0.1780590717299578,
+      "grad_norm": 0.6267391443252563,
+      "learning_rate": 7.398945518453427e-05,
+      "loss": 1.0061273574829102,
+      "step": 422
+    },
+    {
+      "epoch": 0.17890295358649788,
+      "grad_norm": 0.6621372103691101,
+      "learning_rate": 7.434094903339192e-05,
+      "loss": 1.0461612939834595,
+      "step": 424
+    },
+    {
+      "epoch": 0.17974683544303796,
+      "grad_norm": 0.6635435223579407,
+      "learning_rate": 7.469244288224957e-05,
+      "loss": 0.9789207577705383,
+      "step": 426
+    },
+    {
+      "epoch": 0.18059071729957807,
+      "grad_norm": 0.6342346668243408,
+      "learning_rate": 7.504393673110721e-05,
+      "loss": 1.0527069568634033,
+      "step": 428
+    },
+    {
+      "epoch": 0.18143459915611815,
+      "grad_norm": 0.6762149930000305,
+      "learning_rate": 7.539543057996485e-05,
+      "loss": 0.9708702564239502,
+      "step": 430
+    },
+    {
+      "epoch": 0.18227848101265823,
+      "grad_norm": 0.7073282599449158,
+      "learning_rate": 7.57469244288225e-05,
+      "loss": 1.0509834289550781,
+      "step": 432
+    },
+    {
+      "epoch": 0.1831223628691983,
+      "grad_norm": 0.6917856931686401,
+      "learning_rate": 7.609841827768014e-05,
+      "loss": 1.0128819942474365,
+      "step": 434
+    },
+    {
+      "epoch": 0.1839662447257384,
+      "grad_norm": 0.5574942231178284,
+      "learning_rate": 7.644991212653779e-05,
+      "loss": 0.989395797252655,
+      "step": 436
+    },
+    {
+      "epoch": 0.1848101265822785,
+      "grad_norm": 0.640765905380249,
+      "learning_rate": 7.680140597539543e-05,
+      "loss": 0.9846042990684509,
+      "step": 438
+    },
+    {
+      "epoch": 0.18565400843881857,
+      "grad_norm": 0.6699127554893494,
+      "learning_rate": 7.715289982425308e-05,
+      "loss": 1.0344442129135132,
+      "step": 440
+    },
+    {
+      "epoch": 0.18649789029535865,
+      "grad_norm": 0.6164930462837219,
+      "learning_rate": 7.750439367311072e-05,
+      "loss": 1.0179373025894165,
+      "step": 442
+    },
+    {
+      "epoch": 0.18734177215189873,
+      "grad_norm": 0.6880720853805542,
+      "learning_rate": 7.785588752196837e-05,
+      "loss": 1.0518895387649536,
+      "step": 444
+    },
+    {
+      "epoch": 0.1881856540084388,
+      "grad_norm": 0.6501413583755493,
+      "learning_rate": 7.820738137082601e-05,
+      "loss": 1.0442606210708618,
+      "step": 446
+    },
+    {
+      "epoch": 0.1890295358649789,
+      "grad_norm": 0.6076085567474365,
+      "learning_rate": 7.855887521968366e-05,
+      "loss": 0.9828442335128784,
+      "step": 448
+    },
+    {
+      "epoch": 0.189873417721519,
+      "grad_norm": 0.6418202519416809,
+      "learning_rate": 7.89103690685413e-05,
+      "loss": 1.0573710203170776,
+      "step": 450
+    },
+    {
+      "epoch": 0.19071729957805908,
+      "grad_norm": 0.7055076360702515,
+      "learning_rate": 7.926186291739895e-05,
+      "loss": 1.0216103792190552,
+      "step": 452
+    },
+    {
+      "epoch": 0.19156118143459916,
+      "grad_norm": 0.5668330192565918,
+      "learning_rate": 7.961335676625659e-05,
+      "loss": 0.9837722778320312,
+      "step": 454
+    },
+    {
+      "epoch": 0.19240506329113924,
+      "grad_norm": 0.6419380307197571,
+      "learning_rate": 7.996485061511424e-05,
+      "loss": 1.0003894567489624,
+      "step": 456
+    },
+    {
+      "epoch": 0.19324894514767932,
+      "grad_norm": 0.5949198007583618,
+      "learning_rate": 8.031634446397188e-05,
+      "loss": 1.0609031915664673,
+      "step": 458
+    },
+    {
+      "epoch": 0.1940928270042194,
+      "grad_norm": 0.7032039761543274,
+      "learning_rate": 8.066783831282952e-05,
+      "loss": 1.0543403625488281,
+      "step": 460
+    },
+    {
+      "epoch": 0.1949367088607595,
+      "grad_norm": 0.5775868892669678,
+      "learning_rate": 8.101933216168718e-05,
+      "loss": 0.9819303154945374,
+      "step": 462
+    },
+    {
+      "epoch": 0.19578059071729959,
+      "grad_norm": 0.9301062226295471,
+      "learning_rate": 8.137082601054482e-05,
+      "loss": 1.0542067289352417,
+      "step": 464
+    },
+    {
+      "epoch": 0.19662447257383966,
+      "grad_norm": 0.6193217039108276,
+      "learning_rate": 8.172231985940246e-05,
+      "loss": 0.9966341257095337,
+      "step": 466
+    },
+    {
+      "epoch": 0.19746835443037974,
+      "grad_norm": 0.6286146640777588,
+      "learning_rate": 8.20738137082601e-05,
+      "loss": 1.0474121570587158,
+      "step": 468
+    },
+    {
+      "epoch": 0.19831223628691982,
+      "grad_norm": 0.7418972253799438,
+      "learning_rate": 8.242530755711776e-05,
+      "loss": 0.9549239277839661,
+      "step": 470
+    },
+    {
+      "epoch": 0.1991561181434599,
+      "grad_norm": 0.6122808456420898,
+      "learning_rate": 8.27768014059754e-05,
+      "loss": 1.0191338062286377,
+      "step": 472
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.6375362277030945,
+      "learning_rate": 8.312829525483304e-05,
+      "loss": 1.0987539291381836,
+      "step": 474
+    },
+    {
+      "epoch": 0.2008438818565401,
+      "grad_norm": 0.6459513306617737,
+      "learning_rate": 8.347978910369068e-05,
+      "loss": 1.0369136333465576,
+      "step": 476
+    },
+    {
+      "epoch": 0.20168776371308017,
+      "grad_norm": 0.7029640674591064,
+      "learning_rate": 8.383128295254833e-05,
+      "loss": 1.0582096576690674,
+      "step": 478
+    },
+    {
+      "epoch": 0.20253164556962025,
+      "grad_norm": 0.6345387697219849,
+      "learning_rate": 8.418277680140598e-05,
+      "loss": 1.022916316986084,
+      "step": 480
+    },
+    {
+      "epoch": 0.20337552742616033,
+      "grad_norm": 0.5764590501785278,
+      "learning_rate": 8.453427065026362e-05,
+      "loss": 0.973024308681488,
+      "step": 482
+    },
+    {
+      "epoch": 0.2042194092827004,
+      "grad_norm": 0.5884482860565186,
+      "learning_rate": 8.488576449912127e-05,
+      "loss": 1.0292812585830688,
+      "step": 484
+    },
+    {
+      "epoch": 0.20506329113924052,
+      "grad_norm": 0.616357147693634,
+      "learning_rate": 8.523725834797891e-05,
+      "loss": 1.0083447694778442,
+      "step": 486
+    },
+    {
+      "epoch": 0.2059071729957806,
+      "grad_norm": 0.7671196460723877,
+      "learning_rate": 8.558875219683656e-05,
+      "loss": 0.9936985373497009,
+      "step": 488
+    },
+    {
+      "epoch": 0.20675105485232068,
+      "grad_norm": 0.6197299957275391,
+      "learning_rate": 8.59402460456942e-05,
+      "loss": 1.051513910293579,
+      "step": 490
+    },
+    {
+      "epoch": 0.20759493670886076,
+      "grad_norm": 0.6912890672683716,
+      "learning_rate": 8.629173989455185e-05,
+      "loss": 0.9474978446960449,
+      "step": 492
+    },
+    {
+      "epoch": 0.20843881856540084,
+      "grad_norm": 0.6941592693328857,
+      "learning_rate": 8.664323374340949e-05,
+      "loss": 1.0671660900115967,
+      "step": 494
+    },
+    {
+      "epoch": 0.20928270042194091,
+      "grad_norm": 0.5889528393745422,
+      "learning_rate": 8.699472759226714e-05,
+      "loss": 1.0020159482955933,
+      "step": 496
+    },
+    {
+      "epoch": 0.21012658227848102,
+      "grad_norm": 0.6478549838066101,
+      "learning_rate": 8.734622144112478e-05,
+      "loss": 1.0165860652923584,
+      "step": 498
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "grad_norm": 0.6018255949020386,
+      "learning_rate": 8.769771528998243e-05,
+      "loss": 0.8798263072967529,
+      "step": 500
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "eval_loss": 1.042096495628357,
+      "eval_runtime": 692.4361,
+      "eval_samples_per_second": 3.043,
+      "eval_steps_per_second": 3.043,
+      "step": 500
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 14220,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 6,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 5,
+        "early_stopping_threshold": 0.001
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 0
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.9886635097296486e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-500/training_args.bin b/sft_qwen_14B_v2/checkpoints/checkpoint-500/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..0bd5f45757bdc6c7fc2d67718b18675b51983e0e
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-500/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:afae4a8ce1391f149eb00b14eed8f891c715e892ea546bf754d22db2c2bc5969
+size 4792
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-5000/README.md b/sft_qwen_14B_v2/checkpoints/checkpoint-5000/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..134a3a8531b68adda3fed2a98db824f7c1b7dcfd
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-5000/README.md
@@ -0,0 +1,207 @@
+---
+base_model: Models/Qwen2.5-Coder-14B-CPT
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Models/Qwen2.5-Coder-14B-CPT
+- lora
+- transformers
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.0
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-5000/adapter_config.json b/sft_qwen_14B_v2/checkpoints/checkpoint-5000/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..d6316742bc71d30704e2ccdb9080ed0b8062e454
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-5000/adapter_config.json
@@ -0,0 +1,43 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Models/Qwen2.5-Coder-14B-CPT",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.0",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "v_proj",
+    "k_proj",
+    "q_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-5000/adapter_model.safetensors b/sft_qwen_14B_v2/checkpoints/checkpoint-5000/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..cf7b739c3338723d9e7c93715c4fb9da8acc6788
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-5000/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91374177c110742104990e0b3e67a689c6def75d7cd2392cccf4408711361353
+size 201378736
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-5000/optimizer.pt b/sft_qwen_14B_v2/checkpoints/checkpoint-5000/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..4fe7369c92280ff331956f0c14374edd875ff237
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-5000/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f1287edd609e83dd8fdfb7e5e11c1234f29de14279fb1962e92a51ab1bc0cdd4
+size 402976786
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-5000/rng_state.pth b/sft_qwen_14B_v2/checkpoints/checkpoint-5000/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..7600dd2a76937678254e486844a40a1d23dcbd70
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-5000/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:272788126099f44bf6e2ad111d0fa03e729dc3a5173be84593a81bb280c50c89
+size 14244
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-5000/scheduler.pt b/sft_qwen_14B_v2/checkpoints/checkpoint-5000/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..47f1c854cdcd885c5a195f58051f31bb5b982023
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-5000/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0472665e9206c71b6537f62269216f026bad30c8e7d8a98059554058a9199d76
+size 1064
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-5000/trainer_state.json b/sft_qwen_14B_v2/checkpoints/checkpoint-5000/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..a9bc5e90ad07e327f007176c207c40705f202e7e
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-5000/trainer_state.json
@@ -0,0 +1,17943 @@
+{
+  "best_global_step": 4700,
+  "best_metric": 0.7582268714904785,
+  "best_model_checkpoint": "task2file/sft_qwen_14B_v2/checkpoints/checkpoint-4500",
+  "epoch": 2.109704641350211,
+  "eval_steps": 100,
+  "global_step": 5000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0008438818565400844,
+      "grad_norm": 0.5386583805084229,
+      "learning_rate": 1.7574692442882248e-07,
+      "loss": 1.6941628456115723,
+      "step": 2
+    },
+    {
+      "epoch": 0.0016877637130801688,
+      "grad_norm": 0.5477277636528015,
+      "learning_rate": 5.272407732864675e-07,
+      "loss": 1.7132279872894287,
+      "step": 4
+    },
+    {
+      "epoch": 0.002531645569620253,
+      "grad_norm": 0.5390765070915222,
+      "learning_rate": 8.787346221441126e-07,
+      "loss": 1.641180396080017,
+      "step": 6
+    },
+    {
+      "epoch": 0.0033755274261603376,
+      "grad_norm": 0.5023683905601501,
+      "learning_rate": 1.2302284710017575e-06,
+      "loss": 1.5616240501403809,
+      "step": 8
+    },
+    {
+      "epoch": 0.004219409282700422,
+      "grad_norm": 0.4899154603481293,
+      "learning_rate": 1.5817223198594026e-06,
+      "loss": 1.572033405303955,
+      "step": 10
+    },
+    {
+      "epoch": 0.005063291139240506,
+      "grad_norm": 0.5239788293838501,
+      "learning_rate": 1.9332161687170474e-06,
+      "loss": 1.6242921352386475,
+      "step": 12
+    },
+    {
+      "epoch": 0.00590717299578059,
+      "grad_norm": 0.5172926783561707,
+      "learning_rate": 2.2847100175746925e-06,
+      "loss": 1.6800041198730469,
+      "step": 14
+    },
+    {
+      "epoch": 0.006751054852320675,
+      "grad_norm": 0.5539224743843079,
+      "learning_rate": 2.6362038664323376e-06,
+      "loss": 1.6450834274291992,
+      "step": 16
+    },
+    {
+      "epoch": 0.007594936708860759,
+      "grad_norm": 0.5255337953567505,
+      "learning_rate": 2.9876977152899827e-06,
+      "loss": 1.6673263311386108,
+      "step": 18
+    },
+    {
+      "epoch": 0.008438818565400843,
+      "grad_norm": 0.5074548721313477,
+      "learning_rate": 3.3391915641476277e-06,
+      "loss": 1.531802773475647,
+      "step": 20
+    },
+    {
+      "epoch": 0.009282700421940928,
+      "grad_norm": 0.4160279333591461,
+      "learning_rate": 3.6906854130052724e-06,
+      "loss": 1.599354863166809,
+      "step": 22
+    },
+    {
+      "epoch": 0.010126582278481013,
+      "grad_norm": 0.5716474652290344,
+      "learning_rate": 4.0421792618629174e-06,
+      "loss": 1.6700962781906128,
+      "step": 24
+    },
+    {
+      "epoch": 0.010970464135021098,
+      "grad_norm": 0.5148899555206299,
+      "learning_rate": 4.3936731107205625e-06,
+      "loss": 1.66217839717865,
+      "step": 26
+    },
+    {
+      "epoch": 0.01181434599156118,
+      "grad_norm": 0.575722336769104,
+      "learning_rate": 4.7451669595782076e-06,
+      "loss": 1.6692266464233398,
+      "step": 28
+    },
+    {
+      "epoch": 0.012658227848101266,
+      "grad_norm": 0.5345953106880188,
+      "learning_rate": 5.096660808435853e-06,
+      "loss": 1.5518689155578613,
+      "step": 30
+    },
+    {
+      "epoch": 0.01350210970464135,
+      "grad_norm": 0.4462043344974518,
+      "learning_rate": 5.448154657293498e-06,
+      "loss": 1.5930007696151733,
+      "step": 32
+    },
+    {
+      "epoch": 0.014345991561181435,
+      "grad_norm": 0.5119605660438538,
+      "learning_rate": 5.799648506151143e-06,
+      "loss": 1.6069684028625488,
+      "step": 34
+    },
+    {
+      "epoch": 0.015189873417721518,
+      "grad_norm": 0.5328608751296997,
+      "learning_rate": 6.151142355008788e-06,
+      "loss": 1.5838109254837036,
+      "step": 36
+    },
+    {
+      "epoch": 0.016033755274261603,
+      "grad_norm": 0.5065920352935791,
+      "learning_rate": 6.502636203866433e-06,
+      "loss": 1.608130931854248,
+      "step": 38
+    },
+    {
+      "epoch": 0.016877637130801686,
+      "grad_norm": 0.4479359984397888,
+      "learning_rate": 6.854130052724078e-06,
+      "loss": 1.5942182540893555,
+      "step": 40
+    },
+    {
+      "epoch": 0.017721518987341773,
+      "grad_norm": 0.42844903469085693,
+      "learning_rate": 7.205623901581722e-06,
+      "loss": 1.6441553831100464,
+      "step": 42
+    },
+    {
+      "epoch": 0.018565400843881856,
+      "grad_norm": 0.476630836725235,
+      "learning_rate": 7.557117750439367e-06,
+      "loss": 1.6068111658096313,
+      "step": 44
+    },
+    {
+      "epoch": 0.019409282700421943,
+      "grad_norm": 0.4532654881477356,
+      "learning_rate": 7.908611599297012e-06,
+      "loss": 1.6618021726608276,
+      "step": 46
+    },
+    {
+      "epoch": 0.020253164556962026,
+      "grad_norm": 0.3701118230819702,
+      "learning_rate": 8.260105448154657e-06,
+      "loss": 1.4730033874511719,
+      "step": 48
+    },
+    {
+      "epoch": 0.02109704641350211,
+      "grad_norm": 0.38471561670303345,
+      "learning_rate": 8.611599297012302e-06,
+      "loss": 1.4828267097473145,
+      "step": 50
+    },
+    {
+      "epoch": 0.021940928270042195,
+      "grad_norm": 0.3602336347103119,
+      "learning_rate": 8.963093145869948e-06,
+      "loss": 1.3877452611923218,
+      "step": 52
+    },
+    {
+      "epoch": 0.02278481012658228,
+      "grad_norm": 0.40318572521209717,
+      "learning_rate": 9.314586994727593e-06,
+      "loss": 1.49052894115448,
+      "step": 54
+    },
+    {
+      "epoch": 0.02362869198312236,
+      "grad_norm": 0.3223826587200165,
+      "learning_rate": 9.666080843585238e-06,
+      "loss": 1.4912524223327637,
+      "step": 56
+    },
+    {
+      "epoch": 0.024472573839662448,
+      "grad_norm": 0.3873065114021301,
+      "learning_rate": 1.0017574692442883e-05,
+      "loss": 1.526674509048462,
+      "step": 58
+    },
+    {
+      "epoch": 0.02531645569620253,
+      "grad_norm": 0.410159707069397,
+      "learning_rate": 1.0369068541300528e-05,
+      "loss": 1.4480271339416504,
+      "step": 60
+    },
+    {
+      "epoch": 0.026160337552742614,
+      "grad_norm": 0.3632003962993622,
+      "learning_rate": 1.0720562390158173e-05,
+      "loss": 1.4222990274429321,
+      "step": 62
+    },
+    {
+      "epoch": 0.0270042194092827,
+      "grad_norm": 0.33118435740470886,
+      "learning_rate": 1.1072056239015818e-05,
+      "loss": 1.387171745300293,
+      "step": 64
+    },
+    {
+      "epoch": 0.027848101265822784,
+      "grad_norm": 0.3301764726638794,
+      "learning_rate": 1.1423550087873463e-05,
+      "loss": 1.3523777723312378,
+      "step": 66
+    },
+    {
+      "epoch": 0.02869198312236287,
+      "grad_norm": 0.34342435002326965,
+      "learning_rate": 1.1775043936731108e-05,
+      "loss": 1.4515162706375122,
+      "step": 68
+    },
+    {
+      "epoch": 0.029535864978902954,
+      "grad_norm": 0.3243122100830078,
+      "learning_rate": 1.2126537785588753e-05,
+      "loss": 1.3509243726730347,
+      "step": 70
+    },
+    {
+      "epoch": 0.030379746835443037,
+      "grad_norm": 0.3450150787830353,
+      "learning_rate": 1.2478031634446398e-05,
+      "loss": 1.4936245679855347,
+      "step": 72
+    },
+    {
+      "epoch": 0.031223628691983123,
+      "grad_norm": 0.38912028074264526,
+      "learning_rate": 1.2829525483304042e-05,
+      "loss": 1.3419109582901,
+      "step": 74
+    },
+    {
+      "epoch": 0.032067510548523206,
+      "grad_norm": 0.3019310235977173,
+      "learning_rate": 1.3181019332161687e-05,
+      "loss": 1.4284154176712036,
+      "step": 76
+    },
+    {
+      "epoch": 0.03291139240506329,
+      "grad_norm": 0.37803682684898376,
+      "learning_rate": 1.3532513181019332e-05,
+      "loss": 1.4256561994552612,
+      "step": 78
+    },
+    {
+      "epoch": 0.03375527426160337,
+      "grad_norm": 0.34191736578941345,
+      "learning_rate": 1.3884007029876977e-05,
+      "loss": 1.3256909847259521,
+      "step": 80
+    },
+    {
+      "epoch": 0.03459915611814346,
+      "grad_norm": 0.35242700576782227,
+      "learning_rate": 1.4235500878734624e-05,
+      "loss": 1.2710685729980469,
+      "step": 82
+    },
+    {
+      "epoch": 0.035443037974683546,
+      "grad_norm": 0.38094228506088257,
+      "learning_rate": 1.4586994727592269e-05,
+      "loss": 1.253411889076233,
+      "step": 84
+    },
+    {
+      "epoch": 0.036286919831223625,
+      "grad_norm": 0.36837366223335266,
+      "learning_rate": 1.4938488576449914e-05,
+      "loss": 1.3064342737197876,
+      "step": 86
+    },
+    {
+      "epoch": 0.03713080168776371,
+      "grad_norm": 0.3443569242954254,
+      "learning_rate": 1.5289982425307557e-05,
+      "loss": 1.293562412261963,
+      "step": 88
+    },
+    {
+      "epoch": 0.0379746835443038,
+      "grad_norm": 0.3799338936805725,
+      "learning_rate": 1.5641476274165202e-05,
+      "loss": 1.3382648229599,
+      "step": 90
+    },
+    {
+      "epoch": 0.038818565400843885,
+      "grad_norm": 0.40501922369003296,
+      "learning_rate": 1.599297012302285e-05,
+      "loss": 1.3925724029541016,
+      "step": 92
+    },
+    {
+      "epoch": 0.039662447257383965,
+      "grad_norm": 0.4419630467891693,
+      "learning_rate": 1.6344463971880492e-05,
+      "loss": 1.357171893119812,
+      "step": 94
+    },
+    {
+      "epoch": 0.04050632911392405,
+      "grad_norm": 0.3619817793369293,
+      "learning_rate": 1.6695957820738137e-05,
+      "loss": 1.3029985427856445,
+      "step": 96
+    },
+    {
+      "epoch": 0.04135021097046414,
+      "grad_norm": 0.4851357340812683,
+      "learning_rate": 1.7047451669595782e-05,
+      "loss": 1.3498191833496094,
+      "step": 98
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "grad_norm": 0.418658584356308,
+      "learning_rate": 1.7398945518453427e-05,
+      "loss": 1.185287356376648,
+      "step": 100
+    },
+    {
+      "epoch": 0.04219409282700422,
+      "eval_loss": 1.2979938983917236,
+      "eval_runtime": 682.1979,
+      "eval_samples_per_second": 3.089,
+      "eval_steps_per_second": 3.089,
+      "step": 100
+    },
+    {
+      "epoch": 0.043037974683544304,
+      "grad_norm": 0.4464418888092041,
+      "learning_rate": 1.7750439367311073e-05,
+      "loss": 1.2217272520065308,
+      "step": 102
+    },
+    {
+      "epoch": 0.04388185654008439,
+      "grad_norm": 0.4706237316131592,
+      "learning_rate": 1.8101933216168718e-05,
+      "loss": 1.2052050828933716,
+      "step": 104
+    },
+    {
+      "epoch": 0.04472573839662447,
+      "grad_norm": 0.46394404768943787,
+      "learning_rate": 1.8453427065026363e-05,
+      "loss": 1.221343994140625,
+      "step": 106
+    },
+    {
+      "epoch": 0.04556962025316456,
+      "grad_norm": 0.4726889431476593,
+      "learning_rate": 1.8804920913884008e-05,
+      "loss": 1.2387475967407227,
+      "step": 108
+    },
+    {
+      "epoch": 0.046413502109704644,
+      "grad_norm": 0.42130985856056213,
+      "learning_rate": 1.9156414762741653e-05,
+      "loss": 1.2851309776306152,
+      "step": 110
+    },
+    {
+      "epoch": 0.04725738396624472,
+      "grad_norm": 0.4504576623439789,
+      "learning_rate": 1.9507908611599298e-05,
+      "loss": 1.2753145694732666,
+      "step": 112
+    },
+    {
+      "epoch": 0.04810126582278481,
+      "grad_norm": 0.396085262298584,
+      "learning_rate": 1.9859402460456943e-05,
+      "loss": 1.2427717447280884,
+      "step": 114
+    },
+    {
+      "epoch": 0.048945147679324896,
+      "grad_norm": 0.5106491446495056,
+      "learning_rate": 2.0210896309314588e-05,
+      "loss": 1.2943825721740723,
+      "step": 116
+    },
+    {
+      "epoch": 0.049789029535864976,
+      "grad_norm": 0.42351317405700684,
+      "learning_rate": 2.0562390158172233e-05,
+      "loss": 1.263301134109497,
+      "step": 118
+    },
+    {
+      "epoch": 0.05063291139240506,
+      "grad_norm": 0.4403539299964905,
+      "learning_rate": 2.0913884007029878e-05,
+      "loss": 1.2647849321365356,
+      "step": 120
+    },
+    {
+      "epoch": 0.05147679324894515,
+      "grad_norm": 0.5260752439498901,
+      "learning_rate": 2.1265377855887523e-05,
+      "loss": 1.2351393699645996,
+      "step": 122
+    },
+    {
+      "epoch": 0.05232067510548523,
+      "grad_norm": 0.44978851079940796,
+      "learning_rate": 2.1616871704745168e-05,
+      "loss": 1.0384471416473389,
+      "step": 124
+    },
+    {
+      "epoch": 0.053164556962025315,
+      "grad_norm": 0.47732362151145935,
+      "learning_rate": 2.1968365553602813e-05,
+      "loss": 1.1518068313598633,
+      "step": 126
+    },
+    {
+      "epoch": 0.0540084388185654,
+      "grad_norm": 0.5473551750183105,
+      "learning_rate": 2.231985940246046e-05,
+      "loss": 1.2264912128448486,
+      "step": 128
+    },
+    {
+      "epoch": 0.05485232067510549,
+      "grad_norm": 0.4473855197429657,
+      "learning_rate": 2.2671353251318103e-05,
+      "loss": 1.1615246534347534,
+      "step": 130
+    },
+    {
+      "epoch": 0.05569620253164557,
+      "grad_norm": 0.5980377197265625,
+      "learning_rate": 2.302284710017575e-05,
+      "loss": 1.1334880590438843,
+      "step": 132
+    },
+    {
+      "epoch": 0.056540084388185655,
+      "grad_norm": 0.5987792015075684,
+      "learning_rate": 2.3374340949033394e-05,
+      "loss": 1.1546804904937744,
+      "step": 134
+    },
+    {
+      "epoch": 0.05738396624472574,
+      "grad_norm": 0.45355498790740967,
+      "learning_rate": 2.372583479789104e-05,
+      "loss": 1.194953441619873,
+      "step": 136
+    },
+    {
+      "epoch": 0.05822784810126582,
+      "grad_norm": 0.5373698472976685,
+      "learning_rate": 2.4077328646748684e-05,
+      "loss": 1.1067466735839844,
+      "step": 138
+    },
+    {
+      "epoch": 0.05907172995780591,
+      "grad_norm": 0.48734328150749207,
+      "learning_rate": 2.442882249560633e-05,
+      "loss": 1.188468098640442,
+      "step": 140
+    },
+    {
+      "epoch": 0.059915611814345994,
+      "grad_norm": 0.4692173898220062,
+      "learning_rate": 2.478031634446397e-05,
+      "loss": 1.1624362468719482,
+      "step": 142
+    },
+    {
+      "epoch": 0.060759493670886074,
+      "grad_norm": 0.532554030418396,
+      "learning_rate": 2.513181019332162e-05,
+      "loss": 1.0978907346725464,
+      "step": 144
+    },
+    {
+      "epoch": 0.06160337552742616,
+      "grad_norm": 0.5853802561759949,
+      "learning_rate": 2.5483304042179264e-05,
+      "loss": 1.2030781507492065,
+      "step": 146
+    },
+    {
+      "epoch": 0.06244725738396625,
+      "grad_norm": 0.5061611533164978,
+      "learning_rate": 2.583479789103691e-05,
+      "loss": 1.082366943359375,
+      "step": 148
+    },
+    {
+      "epoch": 0.06329113924050633,
+      "grad_norm": 0.49426141381263733,
+      "learning_rate": 2.6186291739894554e-05,
+      "loss": 1.10564386844635,
+      "step": 150
+    },
+    {
+      "epoch": 0.06413502109704641,
+      "grad_norm": 0.5846618413925171,
+      "learning_rate": 2.6537785588752196e-05,
+      "loss": 1.1992807388305664,
+      "step": 152
+    },
+    {
+      "epoch": 0.06497890295358649,
+      "grad_norm": 0.5517552495002747,
+      "learning_rate": 2.6889279437609844e-05,
+      "loss": 1.1757566928863525,
+      "step": 154
+    },
+    {
+      "epoch": 0.06582278481012659,
+      "grad_norm": 0.5667305588722229,
+      "learning_rate": 2.7240773286467486e-05,
+      "loss": 1.0548783540725708,
+      "step": 156
+    },
+    {
+      "epoch": 0.06666666666666667,
+      "grad_norm": 0.6760414242744446,
+      "learning_rate": 2.7592267135325134e-05,
+      "loss": 1.184364914894104,
+      "step": 158
+    },
+    {
+      "epoch": 0.06751054852320675,
+      "grad_norm": 0.5261430740356445,
+      "learning_rate": 2.7943760984182776e-05,
+      "loss": 1.1945042610168457,
+      "step": 160
+    },
+    {
+      "epoch": 0.06835443037974684,
+      "grad_norm": 0.6155015230178833,
+      "learning_rate": 2.8295254833040425e-05,
+      "loss": 1.2021973133087158,
+      "step": 162
+    },
+    {
+      "epoch": 0.06919831223628692,
+      "grad_norm": 0.6131619215011597,
+      "learning_rate": 2.8646748681898066e-05,
+      "loss": 1.144123911857605,
+      "step": 164
+    },
+    {
+      "epoch": 0.070042194092827,
+      "grad_norm": 0.5749185681343079,
+      "learning_rate": 2.8998242530755715e-05,
+      "loss": 1.1329256296157837,
+      "step": 166
+    },
+    {
+      "epoch": 0.07088607594936709,
+      "grad_norm": 0.5243118405342102,
+      "learning_rate": 2.9349736379613356e-05,
+      "loss": 1.0892387628555298,
+      "step": 168
+    },
+    {
+      "epoch": 0.07172995780590717,
+      "grad_norm": 0.7190104722976685,
+      "learning_rate": 2.9701230228471005e-05,
+      "loss": 1.163260817527771,
+      "step": 170
+    },
+    {
+      "epoch": 0.07257383966244725,
+      "grad_norm": 0.5486982464790344,
+      "learning_rate": 3.0052724077328647e-05,
+      "loss": 1.0880777835845947,
+      "step": 172
+    },
+    {
+      "epoch": 0.07341772151898734,
+      "grad_norm": 0.5020889043807983,
+      "learning_rate": 3.0404217926186295e-05,
+      "loss": 1.0433368682861328,
+      "step": 174
+    },
+    {
+      "epoch": 0.07426160337552742,
+      "grad_norm": 0.47329774498939514,
+      "learning_rate": 3.075571177504394e-05,
+      "loss": 1.0528991222381592,
+      "step": 176
+    },
+    {
+      "epoch": 0.0751054852320675,
+      "grad_norm": 0.6635547876358032,
+      "learning_rate": 3.110720562390158e-05,
+      "loss": 1.1627811193466187,
+      "step": 178
+    },
+    {
+      "epoch": 0.0759493670886076,
+      "grad_norm": 0.5624618530273438,
+      "learning_rate": 3.145869947275923e-05,
+      "loss": 1.084869384765625,
+      "step": 180
+    },
+    {
+      "epoch": 0.07679324894514768,
+      "grad_norm": 0.6029536724090576,
+      "learning_rate": 3.181019332161687e-05,
+      "loss": 1.2227671146392822,
+      "step": 182
+    },
+    {
+      "epoch": 0.07763713080168777,
+      "grad_norm": 0.930959939956665,
+      "learning_rate": 3.216168717047452e-05,
+      "loss": 1.0955452919006348,
+      "step": 184
+    },
+    {
+      "epoch": 0.07848101265822785,
+      "grad_norm": 0.5326952338218689,
+      "learning_rate": 3.251318101933216e-05,
+      "loss": 1.0640798807144165,
+      "step": 186
+    },
+    {
+      "epoch": 0.07932489451476793,
+      "grad_norm": 0.5484727621078491,
+      "learning_rate": 3.286467486818981e-05,
+      "loss": 1.0700589418411255,
+      "step": 188
+    },
+    {
+      "epoch": 0.08016877637130802,
+      "grad_norm": 0.605273425579071,
+      "learning_rate": 3.3216168717047456e-05,
+      "loss": 1.1593081951141357,
+      "step": 190
+    },
+    {
+      "epoch": 0.0810126582278481,
+      "grad_norm": 0.5704394578933716,
+      "learning_rate": 3.35676625659051e-05,
+      "loss": 1.1617076396942139,
+      "step": 192
+    },
+    {
+      "epoch": 0.08185654008438818,
+      "grad_norm": 0.5929452180862427,
+      "learning_rate": 3.3919156414762746e-05,
+      "loss": 1.1346839666366577,
+      "step": 194
+    },
+    {
+      "epoch": 0.08270042194092828,
+      "grad_norm": 0.5624077916145325,
+      "learning_rate": 3.427065026362039e-05,
+      "loss": 1.0934710502624512,
+      "step": 196
+    },
+    {
+      "epoch": 0.08354430379746836,
+      "grad_norm": 0.6717425584793091,
+      "learning_rate": 3.4622144112478036e-05,
+      "loss": 1.1810534000396729,
+      "step": 198
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "grad_norm": 0.5120199918746948,
+      "learning_rate": 3.4973637961335674e-05,
+      "loss": 1.1525514125823975,
+      "step": 200
+    },
+    {
+      "epoch": 0.08438818565400844,
+      "eval_loss": 1.142486810684204,
+      "eval_runtime": 668.2356,
+      "eval_samples_per_second": 3.153,
+      "eval_steps_per_second": 3.153,
+      "step": 200
+    },
+    {
+      "epoch": 0.08523206751054853,
+      "grad_norm": 0.5144487023353577,
+      "learning_rate": 3.5325131810193326e-05,
+      "loss": 1.0243735313415527,
+      "step": 202
+    },
+    {
+      "epoch": 0.08607594936708861,
+      "grad_norm": 0.6325069069862366,
+      "learning_rate": 3.5676625659050964e-05,
+      "loss": 1.118743896484375,
+      "step": 204
+    },
+    {
+      "epoch": 0.08691983122362869,
+      "grad_norm": 0.5501633882522583,
+      "learning_rate": 3.6028119507908616e-05,
+      "loss": 1.0380504131317139,
+      "step": 206
+    },
+    {
+      "epoch": 0.08776371308016878,
+      "grad_norm": 0.6133899688720703,
+      "learning_rate": 3.6379613356766254e-05,
+      "loss": 0.9837555885314941,
+      "step": 208
+    },
+    {
+      "epoch": 0.08860759493670886,
+      "grad_norm": 0.5799810886383057,
+      "learning_rate": 3.6731107205623906e-05,
+      "loss": 1.090720295906067,
+      "step": 210
+    },
+    {
+      "epoch": 0.08945147679324894,
+      "grad_norm": 0.6039511561393738,
+      "learning_rate": 3.7082601054481544e-05,
+      "loss": 1.120232343673706,
+      "step": 212
+    },
+    {
+      "epoch": 0.09029535864978903,
+      "grad_norm": 0.5983024835586548,
+      "learning_rate": 3.7434094903339196e-05,
+      "loss": 1.096949815750122,
+      "step": 214
+    },
+    {
+      "epoch": 0.09113924050632911,
+      "grad_norm": 0.5641079545021057,
+      "learning_rate": 3.7785588752196835e-05,
+      "loss": 1.1226298809051514,
+      "step": 216
+    },
+    {
+      "epoch": 0.0919831223628692,
+      "grad_norm": 0.655717134475708,
+      "learning_rate": 3.8137082601054486e-05,
+      "loss": 1.1260643005371094,
+      "step": 218
+    },
+    {
+      "epoch": 0.09282700421940929,
+      "grad_norm": 0.6111898422241211,
+      "learning_rate": 3.848857644991213e-05,
+      "loss": 1.0777709484100342,
+      "step": 220
+    },
+    {
+      "epoch": 0.09367088607594937,
+      "grad_norm": 0.6821302771568298,
+      "learning_rate": 3.884007029876977e-05,
+      "loss": 1.10588800907135,
+      "step": 222
+    },
+    {
+      "epoch": 0.09451476793248945,
+      "grad_norm": 0.693175733089447,
+      "learning_rate": 3.919156414762742e-05,
+      "loss": 1.1498671770095825,
+      "step": 224
+    },
+    {
+      "epoch": 0.09535864978902954,
+      "grad_norm": 0.5288166403770447,
+      "learning_rate": 3.954305799648506e-05,
+      "loss": 1.0587562322616577,
+      "step": 226
+    },
+    {
+      "epoch": 0.09620253164556962,
+      "grad_norm": 0.6882867813110352,
+      "learning_rate": 3.989455184534271e-05,
+      "loss": 1.1107512712478638,
+      "step": 228
+    },
+    {
+      "epoch": 0.0970464135021097,
+      "grad_norm": 0.5834154486656189,
+      "learning_rate": 4.024604569420035e-05,
+      "loss": 1.020510196685791,
+      "step": 230
+    },
+    {
+      "epoch": 0.09789029535864979,
+      "grad_norm": 0.7157064080238342,
+      "learning_rate": 4.0597539543058e-05,
+      "loss": 1.0642449855804443,
+      "step": 232
+    },
+    {
+      "epoch": 0.09873417721518987,
+      "grad_norm": 0.6530708074569702,
+      "learning_rate": 4.094903339191564e-05,
+      "loss": 1.0359872579574585,
+      "step": 234
+    },
+    {
+      "epoch": 0.09957805907172995,
+      "grad_norm": 0.6329686045646667,
+      "learning_rate": 4.130052724077329e-05,
+      "loss": 1.050504446029663,
+      "step": 236
+    },
+    {
+      "epoch": 0.10042194092827005,
+      "grad_norm": 0.6597026586532593,
+      "learning_rate": 4.165202108963093e-05,
+      "loss": 1.2621175050735474,
+      "step": 238
+    },
+    {
+      "epoch": 0.10126582278481013,
+      "grad_norm": 0.6195225119590759,
+      "learning_rate": 4.200351493848858e-05,
+      "loss": 1.1218310594558716,
+      "step": 240
+    },
+    {
+      "epoch": 0.1021097046413502,
+      "grad_norm": 0.6764137744903564,
+      "learning_rate": 4.235500878734622e-05,
+      "loss": 1.1250728368759155,
+      "step": 242
+    },
+    {
+      "epoch": 0.1029535864978903,
+      "grad_norm": 0.552363395690918,
+      "learning_rate": 4.270650263620387e-05,
+      "loss": 1.028212308883667,
+      "step": 244
+    },
+    {
+      "epoch": 0.10379746835443038,
+      "grad_norm": 0.5620495676994324,
+      "learning_rate": 4.305799648506151e-05,
+      "loss": 1.0425450801849365,
+      "step": 246
+    },
+    {
+      "epoch": 0.10464135021097046,
+      "grad_norm": 0.6860032081604004,
+      "learning_rate": 4.3409490333919156e-05,
+      "loss": 1.144278883934021,
+      "step": 248
+    },
+    {
+      "epoch": 0.10548523206751055,
+      "grad_norm": 0.6033259034156799,
+      "learning_rate": 4.37609841827768e-05,
+      "loss": 1.1223982572555542,
+      "step": 250
+    },
+    {
+      "epoch": 0.10632911392405063,
+      "grad_norm": 0.6292146444320679,
+      "learning_rate": 4.4112478031634446e-05,
+      "loss": 1.1609960794448853,
+      "step": 252
+    },
+    {
+      "epoch": 0.10717299578059072,
+      "grad_norm": 0.7982883453369141,
+      "learning_rate": 4.44639718804921e-05,
+      "loss": 1.063547968864441,
+      "step": 254
+    },
+    {
+      "epoch": 0.1080168776371308,
+      "grad_norm": 0.7719110250473022,
+      "learning_rate": 4.4815465729349736e-05,
+      "loss": 1.0719804763793945,
+      "step": 256
+    },
+    {
+      "epoch": 0.10886075949367088,
+      "grad_norm": 0.6101011633872986,
+      "learning_rate": 4.516695957820739e-05,
+      "loss": 1.0778400897979736,
+      "step": 258
+    },
+    {
+      "epoch": 0.10970464135021098,
+      "grad_norm": 0.7300994396209717,
+      "learning_rate": 4.5518453427065026e-05,
+      "loss": 1.2129558324813843,
+      "step": 260
+    },
+    {
+      "epoch": 0.11054852320675106,
+      "grad_norm": 0.8348747491836548,
+      "learning_rate": 4.586994727592268e-05,
+      "loss": 1.221714735031128,
+      "step": 262
+    },
+    {
+      "epoch": 0.11139240506329114,
+      "grad_norm": 0.5445612072944641,
+      "learning_rate": 4.6221441124780316e-05,
+      "loss": 1.0187978744506836,
+      "step": 264
+    },
+    {
+      "epoch": 0.11223628691983123,
+      "grad_norm": 0.6230319738388062,
+      "learning_rate": 4.657293497363797e-05,
+      "loss": 1.096561312675476,
+      "step": 266
+    },
+    {
+      "epoch": 0.11308016877637131,
+      "grad_norm": 0.6231237649917603,
+      "learning_rate": 4.6924428822495606e-05,
+      "loss": 1.089842438697815,
+      "step": 268
+    },
+    {
+      "epoch": 0.11392405063291139,
+      "grad_norm": 0.7178627252578735,
+      "learning_rate": 4.727592267135325e-05,
+      "loss": 1.0696645975112915,
+      "step": 270
+    },
+    {
+      "epoch": 0.11476793248945148,
+      "grad_norm": 0.6895854473114014,
+      "learning_rate": 4.7627416520210896e-05,
+      "loss": 1.0511361360549927,
+      "step": 272
+    },
+    {
+      "epoch": 0.11561181434599156,
+      "grad_norm": 0.6046878695487976,
+      "learning_rate": 4.797891036906854e-05,
+      "loss": 1.1373958587646484,
+      "step": 274
+    },
+    {
+      "epoch": 0.11645569620253164,
+      "grad_norm": 0.6524552702903748,
+      "learning_rate": 4.833040421792619e-05,
+      "loss": 1.0734186172485352,
+      "step": 276
+    },
+    {
+      "epoch": 0.11729957805907174,
+      "grad_norm": 0.6331019997596741,
+      "learning_rate": 4.868189806678383e-05,
+      "loss": 1.123913049697876,
+      "step": 278
+    },
+    {
+      "epoch": 0.11814345991561181,
+      "grad_norm": 0.5919018983840942,
+      "learning_rate": 4.903339191564148e-05,
+      "loss": 1.0635710954666138,
+      "step": 280
+    },
+    {
+      "epoch": 0.1189873417721519,
+      "grad_norm": 0.6067633032798767,
+      "learning_rate": 4.938488576449912e-05,
+      "loss": 1.0429247617721558,
+      "step": 282
+    },
+    {
+      "epoch": 0.11983122362869199,
+      "grad_norm": 0.6583750247955322,
+      "learning_rate": 4.9736379613356774e-05,
+      "loss": 1.1397464275360107,
+      "step": 284
+    },
+    {
+      "epoch": 0.12067510548523207,
+      "grad_norm": 0.6200069785118103,
+      "learning_rate": 5.008787346221442e-05,
+      "loss": 1.0590803623199463,
+      "step": 286
+    },
+    {
+      "epoch": 0.12151898734177215,
+      "grad_norm": 0.6798665523529053,
+      "learning_rate": 5.0439367311072064e-05,
+      "loss": 1.1318789720535278,
+      "step": 288
+    },
+    {
+      "epoch": 0.12236286919831224,
+      "grad_norm": 0.7508794069290161,
+      "learning_rate": 5.07908611599297e-05,
+      "loss": 1.0934956073760986,
+      "step": 290
+    },
+    {
+      "epoch": 0.12320675105485232,
+      "grad_norm": 0.6901452541351318,
+      "learning_rate": 5.114235500878735e-05,
+      "loss": 1.163407802581787,
+      "step": 292
+    },
+    {
+      "epoch": 0.1240506329113924,
+      "grad_norm": 0.6423285603523254,
+      "learning_rate": 5.1493848857644985e-05,
+      "loss": 1.09059476852417,
+      "step": 294
+    },
+    {
+      "epoch": 0.1248945147679325,
+      "grad_norm": 0.6839275360107422,
+      "learning_rate": 5.1845342706502644e-05,
+      "loss": 1.0690211057662964,
+      "step": 296
+    },
+    {
+      "epoch": 0.1257383966244726,
+      "grad_norm": 0.6350128054618835,
+      "learning_rate": 5.219683655536028e-05,
+      "loss": 0.982322096824646,
+      "step": 298
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "grad_norm": 0.7136530876159668,
+      "learning_rate": 5.254833040421793e-05,
+      "loss": 1.1132930517196655,
+      "step": 300
+    },
+    {
+      "epoch": 0.12658227848101267,
+      "eval_loss": 1.0952109098434448,
+      "eval_runtime": 677.0652,
+      "eval_samples_per_second": 3.112,
+      "eval_steps_per_second": 3.112,
+      "step": 300
+    },
+    {
+      "epoch": 0.12742616033755275,
+      "grad_norm": 0.7339721322059631,
+      "learning_rate": 5.289982425307557e-05,
+      "loss": 0.973595917224884,
+      "step": 302
+    },
+    {
+      "epoch": 0.12827004219409283,
+      "grad_norm": 0.5941481590270996,
+      "learning_rate": 5.3251318101933224e-05,
+      "loss": 0.9819849729537964,
+      "step": 304
+    },
+    {
+      "epoch": 0.1291139240506329,
+      "grad_norm": 0.7153938412666321,
+      "learning_rate": 5.360281195079086e-05,
+      "loss": 1.0315470695495605,
+      "step": 306
+    },
+    {
+      "epoch": 0.12995780590717299,
+      "grad_norm": 0.5167180299758911,
+      "learning_rate": 5.395430579964851e-05,
+      "loss": 0.9492001533508301,
+      "step": 308
+    },
+    {
+      "epoch": 0.1308016877637131,
+      "grad_norm": 0.6055944561958313,
+      "learning_rate": 5.430579964850615e-05,
+      "loss": 1.0156209468841553,
+      "step": 310
+    },
+    {
+      "epoch": 0.13164556962025317,
+      "grad_norm": 0.7662386298179626,
+      "learning_rate": 5.4657293497363805e-05,
+      "loss": 1.1791651248931885,
+      "step": 312
+    },
+    {
+      "epoch": 0.13248945147679325,
+      "grad_norm": 0.6065546274185181,
+      "learning_rate": 5.500878734622145e-05,
+      "loss": 1.0009297132492065,
+      "step": 314
+    },
+    {
+      "epoch": 0.13333333333333333,
+      "grad_norm": 0.604225754737854,
+      "learning_rate": 5.536028119507909e-05,
+      "loss": 1.0208244323730469,
+      "step": 316
+    },
+    {
+      "epoch": 0.1341772151898734,
+      "grad_norm": 0.6186763048171997,
+      "learning_rate": 5.571177504393673e-05,
+      "loss": 0.9968416690826416,
+      "step": 318
+    },
+    {
+      "epoch": 0.1350210970464135,
+      "grad_norm": 0.7100363969802856,
+      "learning_rate": 5.606326889279437e-05,
+      "loss": 0.9540256857872009,
+      "step": 320
+    },
+    {
+      "epoch": 0.1358649789029536,
+      "grad_norm": 0.6979711055755615,
+      "learning_rate": 5.641476274165203e-05,
+      "loss": 1.0631953477859497,
+      "step": 322
+    },
+    {
+      "epoch": 0.13670886075949368,
+      "grad_norm": 0.6237109303474426,
+      "learning_rate": 5.676625659050967e-05,
+      "loss": 1.0170501470565796,
+      "step": 324
+    },
+    {
+      "epoch": 0.13755274261603376,
+      "grad_norm": 0.6525548696517944,
+      "learning_rate": 5.711775043936731e-05,
+      "loss": 1.0715603828430176,
+      "step": 326
+    },
+    {
+      "epoch": 0.13839662447257384,
+      "grad_norm": 0.6869221329689026,
+      "learning_rate": 5.746924428822495e-05,
+      "loss": 1.0111541748046875,
+      "step": 328
+    },
+    {
+      "epoch": 0.13924050632911392,
+      "grad_norm": 0.553188145160675,
+      "learning_rate": 5.782073813708261e-05,
+      "loss": 1.0311682224273682,
+      "step": 330
+    },
+    {
+      "epoch": 0.140084388185654,
+      "grad_norm": 0.6760852932929993,
+      "learning_rate": 5.817223198594025e-05,
+      "loss": 1.0213634967803955,
+      "step": 332
+    },
+    {
+      "epoch": 0.1409282700421941,
+      "grad_norm": 0.5907419919967651,
+      "learning_rate": 5.8523725834797894e-05,
+      "loss": 0.9748594164848328,
+      "step": 334
+    },
+    {
+      "epoch": 0.14177215189873418,
+      "grad_norm": 0.7044920921325684,
+      "learning_rate": 5.887521968365554e-05,
+      "loss": 1.05863356590271,
+      "step": 336
+    },
+    {
+      "epoch": 0.14261603375527426,
+      "grad_norm": 0.679073691368103,
+      "learning_rate": 5.922671353251318e-05,
+      "loss": 1.1341127157211304,
+      "step": 338
+    },
+    {
+      "epoch": 0.14345991561181434,
+      "grad_norm": 0.7676237225532532,
+      "learning_rate": 5.957820738137083e-05,
+      "loss": 0.9540836215019226,
+      "step": 340
+    },
+    {
+      "epoch": 0.14430379746835442,
+      "grad_norm": 0.6313899755477905,
+      "learning_rate": 5.9929701230228474e-05,
+      "loss": 1.0585911273956299,
+      "step": 342
+    },
+    {
+      "epoch": 0.1451476793248945,
+      "grad_norm": 0.7123099565505981,
+      "learning_rate": 6.028119507908612e-05,
+      "loss": 1.0760118961334229,
+      "step": 344
+    },
+    {
+      "epoch": 0.1459915611814346,
+      "grad_norm": 0.585935652256012,
+      "learning_rate": 6.063268892794376e-05,
+      "loss": 1.036866307258606,
+      "step": 346
+    },
+    {
+      "epoch": 0.1468354430379747,
+      "grad_norm": 0.5643263459205627,
+      "learning_rate": 6.0984182776801416e-05,
+      "loss": 1.0242938995361328,
+      "step": 348
+    },
+    {
+      "epoch": 0.14767932489451477,
+      "grad_norm": 0.626761794090271,
+      "learning_rate": 6.133567662565906e-05,
+      "loss": 1.0497376918792725,
+      "step": 350
+    },
+    {
+      "epoch": 0.14852320675105485,
+      "grad_norm": 0.5106956958770752,
+      "learning_rate": 6.16871704745167e-05,
+      "loss": 0.9811885356903076,
+      "step": 352
+    },
+    {
+      "epoch": 0.14936708860759493,
+      "grad_norm": 0.6948089003562927,
+      "learning_rate": 6.203866432337434e-05,
+      "loss": 1.0715330839157104,
+      "step": 354
+    },
+    {
+      "epoch": 0.150210970464135,
+      "grad_norm": 0.699713945388794,
+      "learning_rate": 6.239015817223199e-05,
+      "loss": 1.0405226945877075,
+      "step": 356
+    },
+    {
+      "epoch": 0.15105485232067511,
+      "grad_norm": 0.6437667012214661,
+      "learning_rate": 6.274165202108964e-05,
+      "loss": 1.0490930080413818,
+      "step": 358
+    },
+    {
+      "epoch": 0.1518987341772152,
+      "grad_norm": 0.6952699422836304,
+      "learning_rate": 6.309314586994728e-05,
+      "loss": 0.9267548322677612,
+      "step": 360
+    },
+    {
+      "epoch": 0.15274261603375527,
+      "grad_norm": 0.6713186502456665,
+      "learning_rate": 6.344463971880492e-05,
+      "loss": 1.0427420139312744,
+      "step": 362
+    },
+    {
+      "epoch": 0.15358649789029535,
+      "grad_norm": 0.6750379800796509,
+      "learning_rate": 6.379613356766257e-05,
+      "loss": 1.048950433731079,
+      "step": 364
+    },
+    {
+      "epoch": 0.15443037974683543,
+      "grad_norm": 0.6053379774093628,
+      "learning_rate": 6.414762741652022e-05,
+      "loss": 1.0156004428863525,
+      "step": 366
+    },
+    {
+      "epoch": 0.15527426160337554,
+      "grad_norm": 0.8063633441925049,
+      "learning_rate": 6.449912126537786e-05,
+      "loss": 1.0020819902420044,
+      "step": 368
+    },
+    {
+      "epoch": 0.15611814345991562,
+      "grad_norm": 0.8027494549751282,
+      "learning_rate": 6.48506151142355e-05,
+      "loss": 1.055633783340454,
+      "step": 370
+    },
+    {
+      "epoch": 0.1569620253164557,
+      "grad_norm": 0.6580121517181396,
+      "learning_rate": 6.520210896309315e-05,
+      "loss": 1.0149940252304077,
+      "step": 372
+    },
+    {
+      "epoch": 0.15780590717299578,
+      "grad_norm": 0.6561233997344971,
+      "learning_rate": 6.55536028119508e-05,
+      "loss": 0.9769611954689026,
+      "step": 374
+    },
+    {
+      "epoch": 0.15864978902953586,
+      "grad_norm": 0.6444346308708191,
+      "learning_rate": 6.590509666080844e-05,
+      "loss": 0.9099349975585938,
+      "step": 376
+    },
+    {
+      "epoch": 0.15949367088607594,
+      "grad_norm": 0.5879359245300293,
+      "learning_rate": 6.625659050966608e-05,
+      "loss": 1.0797548294067383,
+      "step": 378
+    },
+    {
+      "epoch": 0.16033755274261605,
+      "grad_norm": 0.6994144916534424,
+      "learning_rate": 6.660808435852373e-05,
+      "loss": 1.0336791276931763,
+      "step": 380
+    },
+    {
+      "epoch": 0.16118143459915613,
+      "grad_norm": 0.6128669381141663,
+      "learning_rate": 6.695957820738138e-05,
+      "loss": 1.018118143081665,
+      "step": 382
+    },
+    {
+      "epoch": 0.1620253164556962,
+      "grad_norm": 1.0237540006637573,
+      "learning_rate": 6.731107205623902e-05,
+      "loss": 1.1405497789382935,
+      "step": 384
+    },
+    {
+      "epoch": 0.16286919831223629,
+      "grad_norm": 0.6091578006744385,
+      "learning_rate": 6.766256590509666e-05,
+      "loss": 1.0314189195632935,
+      "step": 386
+    },
+    {
+      "epoch": 0.16371308016877636,
+      "grad_norm": 0.5916037559509277,
+      "learning_rate": 6.801405975395431e-05,
+      "loss": 0.9564052820205688,
+      "step": 388
+    },
+    {
+      "epoch": 0.16455696202531644,
+      "grad_norm": 0.771653950214386,
+      "learning_rate": 6.836555360281195e-05,
+      "loss": 1.0023859739303589,
+      "step": 390
+    },
+    {
+      "epoch": 0.16540084388185655,
+      "grad_norm": 0.654658317565918,
+      "learning_rate": 6.87170474516696e-05,
+      "loss": 1.07024085521698,
+      "step": 392
+    },
+    {
+      "epoch": 0.16624472573839663,
+      "grad_norm": 0.6611968874931335,
+      "learning_rate": 6.906854130052724e-05,
+      "loss": 1.0552500486373901,
+      "step": 394
+    },
+    {
+      "epoch": 0.1670886075949367,
+      "grad_norm": 0.6955893039703369,
+      "learning_rate": 6.942003514938489e-05,
+      "loss": 1.0562875270843506,
+      "step": 396
+    },
+    {
+      "epoch": 0.1679324894514768,
+      "grad_norm": 0.6666058301925659,
+      "learning_rate": 6.977152899824253e-05,
+      "loss": 0.9850592017173767,
+      "step": 398
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "grad_norm": 0.6131711006164551,
+      "learning_rate": 7.012302284710018e-05,
+      "loss": 1.0077755451202393,
+      "step": 400
+    },
+    {
+      "epoch": 0.16877637130801687,
+      "eval_loss": 1.0625108480453491,
+      "eval_runtime": 691.0068,
+      "eval_samples_per_second": 3.049,
+      "eval_steps_per_second": 3.049,
+      "step": 400
+    },
+    {
+      "epoch": 0.16962025316455695,
+      "grad_norm": 0.6286499500274658,
+      "learning_rate": 7.047451669595783e-05,
+      "loss": 1.1012427806854248,
+      "step": 402
+    },
+    {
+      "epoch": 0.17046413502109706,
+      "grad_norm": 0.6639351844787598,
+      "learning_rate": 7.082601054481547e-05,
+      "loss": 1.0379719734191895,
+      "step": 404
+    },
+    {
+      "epoch": 0.17130801687763714,
+      "grad_norm": 0.750401496887207,
+      "learning_rate": 7.117750439367311e-05,
+      "loss": 1.031856656074524,
+      "step": 406
+    },
+    {
+      "epoch": 0.17215189873417722,
+      "grad_norm": 0.8084847331047058,
+      "learning_rate": 7.152899824253075e-05,
+      "loss": 1.0493193864822388,
+      "step": 408
+    },
+    {
+      "epoch": 0.1729957805907173,
+      "grad_norm": 0.7448462247848511,
+      "learning_rate": 7.188049209138841e-05,
+      "loss": 1.1012418270111084,
+      "step": 410
+    },
+    {
+      "epoch": 0.17383966244725738,
+      "grad_norm": 0.5841867923736572,
+      "learning_rate": 7.223198594024605e-05,
+      "loss": 0.9926692247390747,
+      "step": 412
+    },
+    {
+      "epoch": 0.17468354430379746,
+      "grad_norm": 0.7125606536865234,
+      "learning_rate": 7.258347978910369e-05,
+      "loss": 1.0588877201080322,
+      "step": 414
+    },
+    {
+      "epoch": 0.17552742616033756,
+      "grad_norm": 0.5750942230224609,
+      "learning_rate": 7.293497363796134e-05,
+      "loss": 1.038270354270935,
+      "step": 416
+    },
+    {
+      "epoch": 0.17637130801687764,
+      "grad_norm": 0.565444827079773,
+      "learning_rate": 7.328646748681899e-05,
+      "loss": 0.9843021035194397,
+      "step": 418
+    },
+    {
+      "epoch": 0.17721518987341772,
+      "grad_norm": 0.5825693011283875,
+      "learning_rate": 7.363796133567663e-05,
+      "loss": 1.0731632709503174,
+      "step": 420
+    },
+    {
+      "epoch": 0.1780590717299578,
+      "grad_norm": 0.6267391443252563,
+      "learning_rate": 7.398945518453427e-05,
+      "loss": 1.0061273574829102,
+      "step": 422
+    },
+    {
+      "epoch": 0.17890295358649788,
+      "grad_norm": 0.6621372103691101,
+      "learning_rate": 7.434094903339192e-05,
+      "loss": 1.0461612939834595,
+      "step": 424
+    },
+    {
+      "epoch": 0.17974683544303796,
+      "grad_norm": 0.6635435223579407,
+      "learning_rate": 7.469244288224957e-05,
+      "loss": 0.9789207577705383,
+      "step": 426
+    },
+    {
+      "epoch": 0.18059071729957807,
+      "grad_norm": 0.6342346668243408,
+      "learning_rate": 7.504393673110721e-05,
+      "loss": 1.0527069568634033,
+      "step": 428
+    },
+    {
+      "epoch": 0.18143459915611815,
+      "grad_norm": 0.6762149930000305,
+      "learning_rate": 7.539543057996485e-05,
+      "loss": 0.9708702564239502,
+      "step": 430
+    },
+    {
+      "epoch": 0.18227848101265823,
+      "grad_norm": 0.7073282599449158,
+      "learning_rate": 7.57469244288225e-05,
+      "loss": 1.0509834289550781,
+      "step": 432
+    },
+    {
+      "epoch": 0.1831223628691983,
+      "grad_norm": 0.6917856931686401,
+      "learning_rate": 7.609841827768014e-05,
+      "loss": 1.0128819942474365,
+      "step": 434
+    },
+    {
+      "epoch": 0.1839662447257384,
+      "grad_norm": 0.5574942231178284,
+      "learning_rate": 7.644991212653779e-05,
+      "loss": 0.989395797252655,
+      "step": 436
+    },
+    {
+      "epoch": 0.1848101265822785,
+      "grad_norm": 0.640765905380249,
+      "learning_rate": 7.680140597539543e-05,
+      "loss": 0.9846042990684509,
+      "step": 438
+    },
+    {
+      "epoch": 0.18565400843881857,
+      "grad_norm": 0.6699127554893494,
+      "learning_rate": 7.715289982425308e-05,
+      "loss": 1.0344442129135132,
+      "step": 440
+    },
+    {
+      "epoch": 0.18649789029535865,
+      "grad_norm": 0.6164930462837219,
+      "learning_rate": 7.750439367311072e-05,
+      "loss": 1.0179373025894165,
+      "step": 442
+    },
+    {
+      "epoch": 0.18734177215189873,
+      "grad_norm": 0.6880720853805542,
+      "learning_rate": 7.785588752196837e-05,
+      "loss": 1.0518895387649536,
+      "step": 444
+    },
+    {
+      "epoch": 0.1881856540084388,
+      "grad_norm": 0.6501413583755493,
+      "learning_rate": 7.820738137082601e-05,
+      "loss": 1.0442606210708618,
+      "step": 446
+    },
+    {
+      "epoch": 0.1890295358649789,
+      "grad_norm": 0.6076085567474365,
+      "learning_rate": 7.855887521968366e-05,
+      "loss": 0.9828442335128784,
+      "step": 448
+    },
+    {
+      "epoch": 0.189873417721519,
+      "grad_norm": 0.6418202519416809,
+      "learning_rate": 7.89103690685413e-05,
+      "loss": 1.0573710203170776,
+      "step": 450
+    },
+    {
+      "epoch": 0.19071729957805908,
+      "grad_norm": 0.7055076360702515,
+      "learning_rate": 7.926186291739895e-05,
+      "loss": 1.0216103792190552,
+      "step": 452
+    },
+    {
+      "epoch": 0.19156118143459916,
+      "grad_norm": 0.5668330192565918,
+      "learning_rate": 7.961335676625659e-05,
+      "loss": 0.9837722778320312,
+      "step": 454
+    },
+    {
+      "epoch": 0.19240506329113924,
+      "grad_norm": 0.6419380307197571,
+      "learning_rate": 7.996485061511424e-05,
+      "loss": 1.0003894567489624,
+      "step": 456
+    },
+    {
+      "epoch": 0.19324894514767932,
+      "grad_norm": 0.5949198007583618,
+      "learning_rate": 8.031634446397188e-05,
+      "loss": 1.0609031915664673,
+      "step": 458
+    },
+    {
+      "epoch": 0.1940928270042194,
+      "grad_norm": 0.7032039761543274,
+      "learning_rate": 8.066783831282952e-05,
+      "loss": 1.0543403625488281,
+      "step": 460
+    },
+    {
+      "epoch": 0.1949367088607595,
+      "grad_norm": 0.5775868892669678,
+      "learning_rate": 8.101933216168718e-05,
+      "loss": 0.9819303154945374,
+      "step": 462
+    },
+    {
+      "epoch": 0.19578059071729959,
+      "grad_norm": 0.9301062226295471,
+      "learning_rate": 8.137082601054482e-05,
+      "loss": 1.0542067289352417,
+      "step": 464
+    },
+    {
+      "epoch": 0.19662447257383966,
+      "grad_norm": 0.6193217039108276,
+      "learning_rate": 8.172231985940246e-05,
+      "loss": 0.9966341257095337,
+      "step": 466
+    },
+    {
+      "epoch": 0.19746835443037974,
+      "grad_norm": 0.6286146640777588,
+      "learning_rate": 8.20738137082601e-05,
+      "loss": 1.0474121570587158,
+      "step": 468
+    },
+    {
+      "epoch": 0.19831223628691982,
+      "grad_norm": 0.7418972253799438,
+      "learning_rate": 8.242530755711776e-05,
+      "loss": 0.9549239277839661,
+      "step": 470
+    },
+    {
+      "epoch": 0.1991561181434599,
+      "grad_norm": 0.6122808456420898,
+      "learning_rate": 8.27768014059754e-05,
+      "loss": 1.0191338062286377,
+      "step": 472
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.6375362277030945,
+      "learning_rate": 8.312829525483304e-05,
+      "loss": 1.0987539291381836,
+      "step": 474
+    },
+    {
+      "epoch": 0.2008438818565401,
+      "grad_norm": 0.6459513306617737,
+      "learning_rate": 8.347978910369068e-05,
+      "loss": 1.0369136333465576,
+      "step": 476
+    },
+    {
+      "epoch": 0.20168776371308017,
+      "grad_norm": 0.7029640674591064,
+      "learning_rate": 8.383128295254833e-05,
+      "loss": 1.0582096576690674,
+      "step": 478
+    },
+    {
+      "epoch": 0.20253164556962025,
+      "grad_norm": 0.6345387697219849,
+      "learning_rate": 8.418277680140598e-05,
+      "loss": 1.022916316986084,
+      "step": 480
+    },
+    {
+      "epoch": 0.20337552742616033,
+      "grad_norm": 0.5764590501785278,
+      "learning_rate": 8.453427065026362e-05,
+      "loss": 0.973024308681488,
+      "step": 482
+    },
+    {
+      "epoch": 0.2042194092827004,
+      "grad_norm": 0.5884482860565186,
+      "learning_rate": 8.488576449912127e-05,
+      "loss": 1.0292812585830688,
+      "step": 484
+    },
+    {
+      "epoch": 0.20506329113924052,
+      "grad_norm": 0.616357147693634,
+      "learning_rate": 8.523725834797891e-05,
+      "loss": 1.0083447694778442,
+      "step": 486
+    },
+    {
+      "epoch": 0.2059071729957806,
+      "grad_norm": 0.7671196460723877,
+      "learning_rate": 8.558875219683656e-05,
+      "loss": 0.9936985373497009,
+      "step": 488
+    },
+    {
+      "epoch": 0.20675105485232068,
+      "grad_norm": 0.6197299957275391,
+      "learning_rate": 8.59402460456942e-05,
+      "loss": 1.051513910293579,
+      "step": 490
+    },
+    {
+      "epoch": 0.20759493670886076,
+      "grad_norm": 0.6912890672683716,
+      "learning_rate": 8.629173989455185e-05,
+      "loss": 0.9474978446960449,
+      "step": 492
+    },
+    {
+      "epoch": 0.20843881856540084,
+      "grad_norm": 0.6941592693328857,
+      "learning_rate": 8.664323374340949e-05,
+      "loss": 1.0671660900115967,
+      "step": 494
+    },
+    {
+      "epoch": 0.20928270042194091,
+      "grad_norm": 0.5889528393745422,
+      "learning_rate": 8.699472759226714e-05,
+      "loss": 1.0020159482955933,
+      "step": 496
+    },
+    {
+      "epoch": 0.21012658227848102,
+      "grad_norm": 0.6478549838066101,
+      "learning_rate": 8.734622144112478e-05,
+      "loss": 1.0165860652923584,
+      "step": 498
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "grad_norm": 0.6018255949020386,
+      "learning_rate": 8.769771528998243e-05,
+      "loss": 0.8798263072967529,
+      "step": 500
+    },
+    {
+      "epoch": 0.2109704641350211,
+      "eval_loss": 1.042096495628357,
+      "eval_runtime": 692.4361,
+      "eval_samples_per_second": 3.043,
+      "eval_steps_per_second": 3.043,
+      "step": 500
+    },
+    {
+      "epoch": 0.21181434599156118,
+      "grad_norm": 0.578990638256073,
+      "learning_rate": 8.804920913884007e-05,
+      "loss": 1.092096209526062,
+      "step": 502
+    },
+    {
+      "epoch": 0.21265822784810126,
+      "grad_norm": 0.6597883701324463,
+      "learning_rate": 8.840070298769771e-05,
+      "loss": 1.0413451194763184,
+      "step": 504
+    },
+    {
+      "epoch": 0.21350210970464134,
+      "grad_norm": 0.6660305261611938,
+      "learning_rate": 8.875219683655536e-05,
+      "loss": 1.0073142051696777,
+      "step": 506
+    },
+    {
+      "epoch": 0.21434599156118145,
+      "grad_norm": 0.6283115148544312,
+      "learning_rate": 8.910369068541301e-05,
+      "loss": 1.0319768190383911,
+      "step": 508
+    },
+    {
+      "epoch": 0.21518987341772153,
+      "grad_norm": 0.6257343292236328,
+      "learning_rate": 8.945518453427065e-05,
+      "loss": 1.0046353340148926,
+      "step": 510
+    },
+    {
+      "epoch": 0.2160337552742616,
+      "grad_norm": 0.5530875325202942,
+      "learning_rate": 8.980667838312829e-05,
+      "loss": 0.9169099926948547,
+      "step": 512
+    },
+    {
+      "epoch": 0.2168776371308017,
+      "grad_norm": 0.5369633436203003,
+      "learning_rate": 9.015817223198594e-05,
+      "loss": 0.9081505537033081,
+      "step": 514
+    },
+    {
+      "epoch": 0.21772151898734177,
+      "grad_norm": 0.6618232131004333,
+      "learning_rate": 9.05096660808436e-05,
+      "loss": 0.9165045022964478,
+      "step": 516
+    },
+    {
+      "epoch": 0.21856540084388185,
+      "grad_norm": 0.600666344165802,
+      "learning_rate": 9.086115992970123e-05,
+      "loss": 0.91348797082901,
+      "step": 518
+    },
+    {
+      "epoch": 0.21940928270042195,
+      "grad_norm": 0.5919831991195679,
+      "learning_rate": 9.121265377855887e-05,
+      "loss": 1.006508469581604,
+      "step": 520
+    },
+    {
+      "epoch": 0.22025316455696203,
+      "grad_norm": 0.688058614730835,
+      "learning_rate": 9.156414762741654e-05,
+      "loss": 1.0013236999511719,
+      "step": 522
+    },
+    {
+      "epoch": 0.2210970464135021,
+      "grad_norm": 0.6721227765083313,
+      "learning_rate": 9.191564147627418e-05,
+      "loss": 1.0909923315048218,
+      "step": 524
+    },
+    {
+      "epoch": 0.2219409282700422,
+      "grad_norm": 0.5987313389778137,
+      "learning_rate": 9.226713532513181e-05,
+      "loss": 1.0117096900939941,
+      "step": 526
+    },
+    {
+      "epoch": 0.22278481012658227,
+      "grad_norm": 0.6191489696502686,
+      "learning_rate": 9.261862917398945e-05,
+      "loss": 1.0153647661209106,
+      "step": 528
+    },
+    {
+      "epoch": 0.22362869198312235,
+      "grad_norm": 0.6821563243865967,
+      "learning_rate": 9.29701230228471e-05,
+      "loss": 0.9649755954742432,
+      "step": 530
+    },
+    {
+      "epoch": 0.22447257383966246,
+      "grad_norm": 1.760398268699646,
+      "learning_rate": 9.332161687170476e-05,
+      "loss": 0.8673232197761536,
+      "step": 532
+    },
+    {
+      "epoch": 0.22531645569620254,
+      "grad_norm": 0.6670058369636536,
+      "learning_rate": 9.36731107205624e-05,
+      "loss": 0.9942440986633301,
+      "step": 534
+    },
+    {
+      "epoch": 0.22616033755274262,
+      "grad_norm": 0.7345916032791138,
+      "learning_rate": 9.402460456942003e-05,
+      "loss": 1.0364389419555664,
+      "step": 536
+    },
+    {
+      "epoch": 0.2270042194092827,
+      "grad_norm": 0.5946128964424133,
+      "learning_rate": 9.437609841827768e-05,
+      "loss": 0.9314924478530884,
+      "step": 538
+    },
+    {
+      "epoch": 0.22784810126582278,
+      "grad_norm": 0.5800848603248596,
+      "learning_rate": 9.472759226713534e-05,
+      "loss": 1.0694862604141235,
+      "step": 540
+    },
+    {
+      "epoch": 0.22869198312236286,
+      "grad_norm": 0.6712192893028259,
+      "learning_rate": 9.507908611599297e-05,
+      "loss": 1.03531014919281,
+      "step": 542
+    },
+    {
+      "epoch": 0.22953586497890296,
+      "grad_norm": 0.5641416311264038,
+      "learning_rate": 9.543057996485063e-05,
+      "loss": 0.9795235991477966,
+      "step": 544
+    },
+    {
+      "epoch": 0.23037974683544304,
+      "grad_norm": 0.50412517786026,
+      "learning_rate": 9.578207381370826e-05,
+      "loss": 0.9641494750976562,
+      "step": 546
+    },
+    {
+      "epoch": 0.23122362869198312,
+      "grad_norm": 0.579118549823761,
+      "learning_rate": 9.61335676625659e-05,
+      "loss": 0.9375281929969788,
+      "step": 548
+    },
+    {
+      "epoch": 0.2320675105485232,
+      "grad_norm": 0.5888341665267944,
+      "learning_rate": 9.648506151142355e-05,
+      "loss": 0.9414046406745911,
+      "step": 550
+    },
+    {
+      "epoch": 0.23291139240506328,
+      "grad_norm": 0.5595056414604187,
+      "learning_rate": 9.68365553602812e-05,
+      "loss": 0.9005617499351501,
+      "step": 552
+    },
+    {
+      "epoch": 0.23375527426160336,
+      "grad_norm": 0.6605326533317566,
+      "learning_rate": 9.718804920913884e-05,
+      "loss": 1.0283968448638916,
+      "step": 554
+    },
+    {
+      "epoch": 0.23459915611814347,
+      "grad_norm": 0.5657313466072083,
+      "learning_rate": 9.753954305799648e-05,
+      "loss": 1.0058249235153198,
+      "step": 556
+    },
+    {
+      "epoch": 0.23544303797468355,
+      "grad_norm": 0.5433364510536194,
+      "learning_rate": 9.789103690685413e-05,
+      "loss": 0.9835494756698608,
+      "step": 558
+    },
+    {
+      "epoch": 0.23628691983122363,
+      "grad_norm": 0.6129802465438843,
+      "learning_rate": 9.824253075571179e-05,
+      "loss": 1.054532527923584,
+      "step": 560
+    },
+    {
+      "epoch": 0.2371308016877637,
+      "grad_norm": 0.6496239304542542,
+      "learning_rate": 9.859402460456942e-05,
+      "loss": 1.0240973234176636,
+      "step": 562
+    },
+    {
+      "epoch": 0.2379746835443038,
+      "grad_norm": 0.6380873918533325,
+      "learning_rate": 9.894551845342706e-05,
+      "loss": 1.0229179859161377,
+      "step": 564
+    },
+    {
+      "epoch": 0.23881856540084387,
+      "grad_norm": 0.6151993870735168,
+      "learning_rate": 9.929701230228471e-05,
+      "loss": 1.0111570358276367,
+      "step": 566
+    },
+    {
+      "epoch": 0.23966244725738398,
+      "grad_norm": 0.5727584958076477,
+      "learning_rate": 9.964850615114237e-05,
+      "loss": 0.9450829029083252,
+      "step": 568
+    },
+    {
+      "epoch": 0.24050632911392406,
+      "grad_norm": 0.6620725989341736,
+      "learning_rate": 0.0001,
+      "loss": 0.9800319075584412,
+      "step": 570
+    },
+    {
+      "epoch": 0.24135021097046414,
+      "grad_norm": 0.6151163578033447,
+      "learning_rate": 0.00010035149384885764,
+      "loss": 0.9757438898086548,
+      "step": 572
+    },
+    {
+      "epoch": 0.24219409282700421,
+      "grad_norm": 0.5672140717506409,
+      "learning_rate": 0.0001007029876977153,
+      "loss": 0.9104921817779541,
+      "step": 574
+    },
+    {
+      "epoch": 0.2430379746835443,
+      "grad_norm": 0.5697256326675415,
+      "learning_rate": 0.00010105448154657293,
+      "loss": 1.1027376651763916,
+      "step": 576
+    },
+    {
+      "epoch": 0.2438818565400844,
+      "grad_norm": 0.5590381622314453,
+      "learning_rate": 0.00010140597539543057,
+      "loss": 1.0055404901504517,
+      "step": 578
+    },
+    {
+      "epoch": 0.24472573839662448,
+      "grad_norm": 0.5518567562103271,
+      "learning_rate": 0.00010175746924428824,
+      "loss": 1.020835518836975,
+      "step": 580
+    },
+    {
+      "epoch": 0.24556962025316456,
+      "grad_norm": 0.6338496208190918,
+      "learning_rate": 0.00010210896309314588,
+      "loss": 0.9528344869613647,
+      "step": 582
+    },
+    {
+      "epoch": 0.24641350210970464,
+      "grad_norm": 0.6497329473495483,
+      "learning_rate": 0.00010246045694200353,
+      "loss": 1.0088670253753662,
+      "step": 584
+    },
+    {
+      "epoch": 0.24725738396624472,
+      "grad_norm": 0.49888095259666443,
+      "learning_rate": 0.00010281195079086117,
+      "loss": 0.9961200952529907,
+      "step": 586
+    },
+    {
+      "epoch": 0.2481012658227848,
+      "grad_norm": 0.5680158734321594,
+      "learning_rate": 0.0001031634446397188,
+      "loss": 0.9635610580444336,
+      "step": 588
+    },
+    {
+      "epoch": 0.2489451476793249,
+      "grad_norm": 0.658168375492096,
+      "learning_rate": 0.00010351493848857646,
+      "loss": 0.9392287135124207,
+      "step": 590
+    },
+    {
+      "epoch": 0.249789029535865,
+      "grad_norm": 0.618262767791748,
+      "learning_rate": 0.0001038664323374341,
+      "loss": 0.9600516557693481,
+      "step": 592
+    },
+    {
+      "epoch": 0.25063291139240507,
+      "grad_norm": 0.6003909111022949,
+      "learning_rate": 0.00010421792618629173,
+      "loss": 1.005476713180542,
+      "step": 594
+    },
+    {
+      "epoch": 0.2514767932489452,
+      "grad_norm": 0.5437078475952148,
+      "learning_rate": 0.00010456942003514938,
+      "loss": 0.9523017406463623,
+      "step": 596
+    },
+    {
+      "epoch": 0.2523206751054852,
+      "grad_norm": 0.5524541735649109,
+      "learning_rate": 0.00010492091388400705,
+      "loss": 0.9526668787002563,
+      "step": 598
+    },
+    {
+      "epoch": 0.25316455696202533,
+      "grad_norm": 0.679504930973053,
+      "learning_rate": 0.00010527240773286469,
+      "loss": 1.019660472869873,
+      "step": 600
+    },
+    {
+      "epoch": 0.25316455696202533,
+      "eval_loss": 1.0193854570388794,
+      "eval_runtime": 677.9523,
+      "eval_samples_per_second": 3.108,
+      "eval_steps_per_second": 3.108,
+      "step": 600
+    },
+    {
+      "epoch": 0.2540084388185654,
+      "grad_norm": 0.5646136999130249,
+      "learning_rate": 0.00010562390158172233,
+      "loss": 0.9910882711410522,
+      "step": 602
+    },
+    {
+      "epoch": 0.2548523206751055,
+      "grad_norm": 0.5238093137741089,
+      "learning_rate": 0.00010597539543057998,
+      "loss": 0.9616432785987854,
+      "step": 604
+    },
+    {
+      "epoch": 0.25569620253164554,
+      "grad_norm": 0.7483857274055481,
+      "learning_rate": 0.00010632688927943762,
+      "loss": 1.0078275203704834,
+      "step": 606
+    },
+    {
+      "epoch": 0.25654008438818565,
+      "grad_norm": 0.578948974609375,
+      "learning_rate": 0.00010667838312829525,
+      "loss": 0.9827103018760681,
+      "step": 608
+    },
+    {
+      "epoch": 0.25738396624472576,
+      "grad_norm": 0.5525906085968018,
+      "learning_rate": 0.00010702987697715289,
+      "loss": 1.0423277616500854,
+      "step": 610
+    },
+    {
+      "epoch": 0.2582278481012658,
+      "grad_norm": 0.6721326112747192,
+      "learning_rate": 0.00010738137082601054,
+      "loss": 0.9561693072319031,
+      "step": 612
+    },
+    {
+      "epoch": 0.2590717299578059,
+      "grad_norm": 0.5701051354408264,
+      "learning_rate": 0.00010773286467486821,
+      "loss": 0.9602992534637451,
+      "step": 614
+    },
+    {
+      "epoch": 0.25991561181434597,
+      "grad_norm": 0.6349860429763794,
+      "learning_rate": 0.00010808435852372585,
+      "loss": 1.1422650814056396,
+      "step": 616
+    },
+    {
+      "epoch": 0.2607594936708861,
+      "grad_norm": 0.5496085286140442,
+      "learning_rate": 0.00010843585237258349,
+      "loss": 0.9762773513793945,
+      "step": 618
+    },
+    {
+      "epoch": 0.2616033755274262,
+      "grad_norm": 0.6080722808837891,
+      "learning_rate": 0.00010878734622144114,
+      "loss": 1.0133616924285889,
+      "step": 620
+    },
+    {
+      "epoch": 0.26244725738396624,
+      "grad_norm": 0.5450218915939331,
+      "learning_rate": 0.00010913884007029878,
+      "loss": 0.9385587573051453,
+      "step": 622
+    },
+    {
+      "epoch": 0.26329113924050634,
+      "grad_norm": 0.592106819152832,
+      "learning_rate": 0.00010949033391915641,
+      "loss": 0.9359989762306213,
+      "step": 624
+    },
+    {
+      "epoch": 0.2641350210970464,
+      "grad_norm": 0.6449427604675293,
+      "learning_rate": 0.00010984182776801407,
+      "loss": 1.0266027450561523,
+      "step": 626
+    },
+    {
+      "epoch": 0.2649789029535865,
+      "grad_norm": 0.538299560546875,
+      "learning_rate": 0.0001101933216168717,
+      "loss": 0.9303187131881714,
+      "step": 628
+    },
+    {
+      "epoch": 0.26582278481012656,
+      "grad_norm": 0.546316921710968,
+      "learning_rate": 0.00011054481546572934,
+      "loss": 0.9368857145309448,
+      "step": 630
+    },
+    {
+      "epoch": 0.26666666666666666,
+      "grad_norm": 0.5818730592727661,
+      "learning_rate": 0.00011089630931458701,
+      "loss": 0.9573145508766174,
+      "step": 632
+    },
+    {
+      "epoch": 0.26751054852320677,
+      "grad_norm": 0.5958262085914612,
+      "learning_rate": 0.00011124780316344465,
+      "loss": 0.9345449805259705,
+      "step": 634
+    },
+    {
+      "epoch": 0.2683544303797468,
+      "grad_norm": 0.6259077787399292,
+      "learning_rate": 0.0001115992970123023,
+      "loss": 1.0906590223312378,
+      "step": 636
+    },
+    {
+      "epoch": 0.26919831223628693,
+      "grad_norm": 0.589672863483429,
+      "learning_rate": 0.00011195079086115994,
+      "loss": 1.0757447481155396,
+      "step": 638
+    },
+    {
+      "epoch": 0.270042194092827,
+      "grad_norm": 0.5714080333709717,
+      "learning_rate": 0.00011230228471001758,
+      "loss": 0.9310855269432068,
+      "step": 640
+    },
+    {
+      "epoch": 0.2708860759493671,
+      "grad_norm": 0.45342639088630676,
+      "learning_rate": 0.00011265377855887523,
+      "loss": 0.9276360273361206,
+      "step": 642
+    },
+    {
+      "epoch": 0.2717299578059072,
+      "grad_norm": 0.6386750340461731,
+      "learning_rate": 0.00011300527240773287,
+      "loss": 1.084719181060791,
+      "step": 644
+    },
+    {
+      "epoch": 0.27257383966244725,
+      "grad_norm": 0.6446163654327393,
+      "learning_rate": 0.0001133567662565905,
+      "loss": 0.9763918519020081,
+      "step": 646
+    },
+    {
+      "epoch": 0.27341772151898736,
+      "grad_norm": 0.5925686359405518,
+      "learning_rate": 0.00011370826010544816,
+      "loss": 0.9517921805381775,
+      "step": 648
+    },
+    {
+      "epoch": 0.2742616033755274,
+      "grad_norm": 0.5399773716926575,
+      "learning_rate": 0.00011405975395430582,
+      "loss": 1.0587927103042603,
+      "step": 650
+    },
+    {
+      "epoch": 0.2751054852320675,
+      "grad_norm": 0.5872456431388855,
+      "learning_rate": 0.00011441124780316346,
+      "loss": 0.883341908454895,
+      "step": 652
+    },
+    {
+      "epoch": 0.2759493670886076,
+      "grad_norm": 0.5574564337730408,
+      "learning_rate": 0.0001147627416520211,
+      "loss": 1.0306891202926636,
+      "step": 654
+    },
+    {
+      "epoch": 0.2767932489451477,
+      "grad_norm": 0.47789013385772705,
+      "learning_rate": 0.00011511423550087874,
+      "loss": 0.8814032077789307,
+      "step": 656
+    },
+    {
+      "epoch": 0.2776371308016878,
+      "grad_norm": 0.5565530061721802,
+      "learning_rate": 0.00011546572934973639,
+      "loss": 0.9460552334785461,
+      "step": 658
+    },
+    {
+      "epoch": 0.27848101265822783,
+      "grad_norm": 0.5299761295318604,
+      "learning_rate": 0.00011581722319859403,
+      "loss": 0.9475110769271851,
+      "step": 660
+    },
+    {
+      "epoch": 0.27932489451476794,
+      "grad_norm": 0.6503344178199768,
+      "learning_rate": 0.00011616871704745166,
+      "loss": 1.0630913972854614,
+      "step": 662
+    },
+    {
+      "epoch": 0.280168776371308,
+      "grad_norm": 0.5794585943222046,
+      "learning_rate": 0.00011652021089630932,
+      "loss": 0.9389138221740723,
+      "step": 664
+    },
+    {
+      "epoch": 0.2810126582278481,
+      "grad_norm": 0.5762867331504822,
+      "learning_rate": 0.00011687170474516695,
+      "loss": 0.8934136033058167,
+      "step": 666
+    },
+    {
+      "epoch": 0.2818565400843882,
+      "grad_norm": 0.6565435528755188,
+      "learning_rate": 0.00011722319859402462,
+      "loss": 1.1072614192962646,
+      "step": 668
+    },
+    {
+      "epoch": 0.28270042194092826,
+      "grad_norm": 0.5819830298423767,
+      "learning_rate": 0.00011757469244288226,
+      "loss": 1.0501434803009033,
+      "step": 670
+    },
+    {
+      "epoch": 0.28354430379746837,
+      "grad_norm": 0.6071487069129944,
+      "learning_rate": 0.00011792618629173991,
+      "loss": 0.9880793690681458,
+      "step": 672
+    },
+    {
+      "epoch": 0.2843881856540084,
+      "grad_norm": 0.5765058398246765,
+      "learning_rate": 0.00011827768014059755,
+      "loss": 0.9670693874359131,
+      "step": 674
+    },
+    {
+      "epoch": 0.2852320675105485,
+      "grad_norm": 0.5245351791381836,
+      "learning_rate": 0.00011862917398945519,
+      "loss": 0.9602360725402832,
+      "step": 676
+    },
+    {
+      "epoch": 0.28607594936708863,
+      "grad_norm": 0.6189922094345093,
+      "learning_rate": 0.00011898066783831282,
+      "loss": 0.9684560894966125,
+      "step": 678
+    },
+    {
+      "epoch": 0.2869198312236287,
+      "grad_norm": 0.6138690710067749,
+      "learning_rate": 0.00011933216168717048,
+      "loss": 0.9465792775154114,
+      "step": 680
+    },
+    {
+      "epoch": 0.2877637130801688,
+      "grad_norm": 0.5371595621109009,
+      "learning_rate": 0.00011968365553602812,
+      "loss": 0.8495944738388062,
+      "step": 682
+    },
+    {
+      "epoch": 0.28860759493670884,
+      "grad_norm": 0.5549944639205933,
+      "learning_rate": 0.00012003514938488578,
+      "loss": 0.9663267135620117,
+      "step": 684
+    },
+    {
+      "epoch": 0.28945147679324895,
+      "grad_norm": 0.6484189033508301,
+      "learning_rate": 0.00012038664323374342,
+      "loss": 0.9736058712005615,
+      "step": 686
+    },
+    {
+      "epoch": 0.290295358649789,
+      "grad_norm": 0.540351390838623,
+      "learning_rate": 0.00012073813708260107,
+      "loss": 1.0591845512390137,
+      "step": 688
+    },
+    {
+      "epoch": 0.2911392405063291,
+      "grad_norm": 0.5657922029495239,
+      "learning_rate": 0.00012108963093145871,
+      "loss": 0.944908618927002,
+      "step": 690
+    },
+    {
+      "epoch": 0.2919831223628692,
+      "grad_norm": 0.6040505170822144,
+      "learning_rate": 0.00012144112478031635,
+      "loss": 1.0018219947814941,
+      "step": 692
+    },
+    {
+      "epoch": 0.29282700421940927,
+      "grad_norm": 0.5435477495193481,
+      "learning_rate": 0.000121792618629174,
+      "loss": 1.0351502895355225,
+      "step": 694
+    },
+    {
+      "epoch": 0.2936708860759494,
+      "grad_norm": 0.5712518692016602,
+      "learning_rate": 0.00012214411247803164,
+      "loss": 0.9935672283172607,
+      "step": 696
+    },
+    {
+      "epoch": 0.29451476793248943,
+      "grad_norm": 0.6138222813606262,
+      "learning_rate": 0.00012249560632688928,
+      "loss": 1.0165108442306519,
+      "step": 698
+    },
+    {
+      "epoch": 0.29535864978902954,
+      "grad_norm": 0.4495212435722351,
+      "learning_rate": 0.00012284710017574691,
+      "loss": 0.9334425926208496,
+      "step": 700
+    },
+    {
+      "epoch": 0.29535864978902954,
+      "eval_loss": 0.996929407119751,
+      "eval_runtime": 668.6398,
+      "eval_samples_per_second": 3.151,
+      "eval_steps_per_second": 3.151,
+      "step": 700
+    },
+    {
+      "epoch": 0.29620253164556964,
+      "grad_norm": 0.5321539044380188,
+      "learning_rate": 0.00012319859402460458,
+      "loss": 1.0516537427902222,
+      "step": 702
+    },
+    {
+      "epoch": 0.2970464135021097,
+      "grad_norm": 0.5716516971588135,
+      "learning_rate": 0.00012355008787346222,
+      "loss": 0.9387198686599731,
+      "step": 704
+    },
+    {
+      "epoch": 0.2978902953586498,
+      "grad_norm": 0.5617920160293579,
+      "learning_rate": 0.00012390158172231988,
+      "loss": 0.99737948179245,
+      "step": 706
+    },
+    {
+      "epoch": 0.29873417721518986,
+      "grad_norm": 0.4922899007797241,
+      "learning_rate": 0.00012425307557117752,
+      "loss": 0.9955025911331177,
+      "step": 708
+    },
+    {
+      "epoch": 0.29957805907172996,
+      "grad_norm": 0.543501615524292,
+      "learning_rate": 0.00012460456942003516,
+      "loss": 0.9124280214309692,
+      "step": 710
+    },
+    {
+      "epoch": 0.30042194092827,
+      "grad_norm": 0.49590054154396057,
+      "learning_rate": 0.0001249560632688928,
+      "loss": 0.9820216298103333,
+      "step": 712
+    },
+    {
+      "epoch": 0.3012658227848101,
+      "grad_norm": 0.5984305739402771,
+      "learning_rate": 0.00012530755711775044,
+      "loss": 1.0152074098587036,
+      "step": 714
+    },
+    {
+      "epoch": 0.30210970464135023,
+      "grad_norm": 0.9343504905700684,
+      "learning_rate": 0.00012565905096660807,
+      "loss": 1.0577725172042847,
+      "step": 716
+    },
+    {
+      "epoch": 0.3029535864978903,
+      "grad_norm": 0.5118702054023743,
+      "learning_rate": 0.0001260105448154657,
+      "loss": 0.9830358028411865,
+      "step": 718
+    },
+    {
+      "epoch": 0.3037974683544304,
+      "grad_norm": 0.4940392076969147,
+      "learning_rate": 0.00012636203866432338,
+      "loss": 0.9466043710708618,
+      "step": 720
+    },
+    {
+      "epoch": 0.30464135021097044,
+      "grad_norm": 0.5965693593025208,
+      "learning_rate": 0.00012671353251318104,
+      "loss": 1.015270709991455,
+      "step": 722
+    },
+    {
+      "epoch": 0.30548523206751055,
+      "grad_norm": 0.5020529627799988,
+      "learning_rate": 0.00012706502636203868,
+      "loss": 0.9703927636146545,
+      "step": 724
+    },
+    {
+      "epoch": 0.30632911392405066,
+      "grad_norm": 0.6067010164260864,
+      "learning_rate": 0.00012741652021089632,
+      "loss": 1.0255526304244995,
+      "step": 726
+    },
+    {
+      "epoch": 0.3071729957805907,
+      "grad_norm": 0.5931884050369263,
+      "learning_rate": 0.00012776801405975396,
+      "loss": 0.9335633516311646,
+      "step": 728
+    },
+    {
+      "epoch": 0.3080168776371308,
+      "grad_norm": 0.5938752293586731,
+      "learning_rate": 0.0001281195079086116,
+      "loss": 1.0921578407287598,
+      "step": 730
+    },
+    {
+      "epoch": 0.30886075949367087,
+      "grad_norm": 0.49728086590766907,
+      "learning_rate": 0.00012847100175746923,
+      "loss": 0.963066041469574,
+      "step": 732
+    },
+    {
+      "epoch": 0.309704641350211,
+      "grad_norm": 0.5452080965042114,
+      "learning_rate": 0.0001288224956063269,
+      "loss": 0.9513075351715088,
+      "step": 734
+    },
+    {
+      "epoch": 0.3105485232067511,
+      "grad_norm": 0.5497731566429138,
+      "learning_rate": 0.00012917398945518454,
+      "loss": 0.8576077222824097,
+      "step": 736
+    },
+    {
+      "epoch": 0.31139240506329113,
+      "grad_norm": 0.5580397248268127,
+      "learning_rate": 0.0001295254833040422,
+      "loss": 0.9542577862739563,
+      "step": 738
+    },
+    {
+      "epoch": 0.31223628691983124,
+      "grad_norm": 0.5890427827835083,
+      "learning_rate": 0.00012987697715289984,
+      "loss": 0.8992732167243958,
+      "step": 740
+    },
+    {
+      "epoch": 0.3130801687763713,
+      "grad_norm": 0.5942965745925903,
+      "learning_rate": 0.00013022847100175748,
+      "loss": 1.0322896242141724,
+      "step": 742
+    },
+    {
+      "epoch": 0.3139240506329114,
+      "grad_norm": 0.6341713070869446,
+      "learning_rate": 0.00013057996485061512,
+      "loss": 0.9217103719711304,
+      "step": 744
+    },
+    {
+      "epoch": 0.31476793248945145,
+      "grad_norm": 0.5294105410575867,
+      "learning_rate": 0.00013093145869947276,
+      "loss": 0.951789915561676,
+      "step": 746
+    },
+    {
+      "epoch": 0.31561181434599156,
+      "grad_norm": 0.6372058391571045,
+      "learning_rate": 0.0001312829525483304,
+      "loss": 0.9459875226020813,
+      "step": 748
+    },
+    {
+      "epoch": 0.31645569620253167,
+      "grad_norm": 0.5979796648025513,
+      "learning_rate": 0.00013163444639718806,
+      "loss": 0.9626097679138184,
+      "step": 750
+    },
+    {
+      "epoch": 0.3172995780590717,
+      "grad_norm": 0.5682399868965149,
+      "learning_rate": 0.0001319859402460457,
+      "loss": 1.0261781215667725,
+      "step": 752
+    },
+    {
+      "epoch": 0.3181434599156118,
+      "grad_norm": 0.5349125266075134,
+      "learning_rate": 0.00013233743409490336,
+      "loss": 0.9319828152656555,
+      "step": 754
+    },
+    {
+      "epoch": 0.3189873417721519,
+      "grad_norm": 0.6093934178352356,
+      "learning_rate": 0.000132688927943761,
+      "loss": 0.9216550588607788,
+      "step": 756
+    },
+    {
+      "epoch": 0.319831223628692,
+      "grad_norm": 0.5188612341880798,
+      "learning_rate": 0.00013304042179261864,
+      "loss": 0.901739776134491,
+      "step": 758
+    },
+    {
+      "epoch": 0.3206751054852321,
+      "grad_norm": 0.5877130627632141,
+      "learning_rate": 0.00013339191564147628,
+      "loss": 1.0362589359283447,
+      "step": 760
+    },
+    {
+      "epoch": 0.32151898734177214,
+      "grad_norm": 0.5542771816253662,
+      "learning_rate": 0.00013374340949033392,
+      "loss": 0.8787116408348083,
+      "step": 762
+    },
+    {
+      "epoch": 0.32236286919831225,
+      "grad_norm": 0.5084902048110962,
+      "learning_rate": 0.00013409490333919156,
+      "loss": 0.9237037301063538,
+      "step": 764
+    },
+    {
+      "epoch": 0.3232067510548523,
+      "grad_norm": 0.5461528301239014,
+      "learning_rate": 0.00013444639718804922,
+      "loss": 1.0150731801986694,
+      "step": 766
+    },
+    {
+      "epoch": 0.3240506329113924,
+      "grad_norm": 0.53483647108078,
+      "learning_rate": 0.00013479789103690686,
+      "loss": 0.8985214829444885,
+      "step": 768
+    },
+    {
+      "epoch": 0.32489451476793246,
+      "grad_norm": 0.5580531358718872,
+      "learning_rate": 0.0001351493848857645,
+      "loss": 1.0225775241851807,
+      "step": 770
+    },
+    {
+      "epoch": 0.32573839662447257,
+      "grad_norm": 0.5203377604484558,
+      "learning_rate": 0.00013550087873462216,
+      "loss": 0.9571293592453003,
+      "step": 772
+    },
+    {
+      "epoch": 0.3265822784810127,
+      "grad_norm": 0.5049671530723572,
+      "learning_rate": 0.0001358523725834798,
+      "loss": 1.0468909740447998,
+      "step": 774
+    },
+    {
+      "epoch": 0.32742616033755273,
+      "grad_norm": 0.4723063111305237,
+      "learning_rate": 0.00013620386643233744,
+      "loss": 0.7743215560913086,
+      "step": 776
+    },
+    {
+      "epoch": 0.32827004219409284,
+      "grad_norm": 0.6310980916023254,
+      "learning_rate": 0.00013655536028119508,
+      "loss": 1.021510362625122,
+      "step": 778
+    },
+    {
+      "epoch": 0.3291139240506329,
+      "grad_norm": 0.47066664695739746,
+      "learning_rate": 0.00013690685413005274,
+      "loss": 0.9134382605552673,
+      "step": 780
+    },
+    {
+      "epoch": 0.329957805907173,
+      "grad_norm": 0.5725092887878418,
+      "learning_rate": 0.00013725834797891038,
+      "loss": 0.9797834753990173,
+      "step": 782
+    },
+    {
+      "epoch": 0.3308016877637131,
+      "grad_norm": 0.5139563083648682,
+      "learning_rate": 0.00013760984182776802,
+      "loss": 0.9372621178627014,
+      "step": 784
+    },
+    {
+      "epoch": 0.33164556962025316,
+      "grad_norm": 0.5275821685791016,
+      "learning_rate": 0.00013796133567662566,
+      "loss": 0.9528245329856873,
+      "step": 786
+    },
+    {
+      "epoch": 0.33248945147679326,
+      "grad_norm": 0.5702582001686096,
+      "learning_rate": 0.0001383128295254833,
+      "loss": 0.9750176072120667,
+      "step": 788
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.5281293392181396,
+      "learning_rate": 0.00013866432337434096,
+      "loss": 0.9412306547164917,
+      "step": 790
+    },
+    {
+      "epoch": 0.3341772151898734,
+      "grad_norm": 0.5578986406326294,
+      "learning_rate": 0.0001390158172231986,
+      "loss": 0.997580885887146,
+      "step": 792
+    },
+    {
+      "epoch": 0.33502109704641353,
+      "grad_norm": 0.50461345911026,
+      "learning_rate": 0.00013936731107205624,
+      "loss": 0.9082320928573608,
+      "step": 794
+    },
+    {
+      "epoch": 0.3358649789029536,
+      "grad_norm": 0.5258530378341675,
+      "learning_rate": 0.0001397188049209139,
+      "loss": 1.0082844495773315,
+      "step": 796
+    },
+    {
+      "epoch": 0.3367088607594937,
+      "grad_norm": 0.5548169016838074,
+      "learning_rate": 0.00014007029876977154,
+      "loss": 0.9729003310203552,
+      "step": 798
+    },
+    {
+      "epoch": 0.33755274261603374,
+      "grad_norm": 0.48601076006889343,
+      "learning_rate": 0.00014042179261862918,
+      "loss": 0.9099526405334473,
+      "step": 800
+    },
+    {
+      "epoch": 0.33755274261603374,
+      "eval_loss": 0.9800403714179993,
+      "eval_runtime": 678.8306,
+      "eval_samples_per_second": 3.104,
+      "eval_steps_per_second": 3.104,
+      "step": 800
+    },
+    {
+      "epoch": 0.33839662447257385,
+      "grad_norm": 0.5413158535957336,
+      "learning_rate": 0.00014077328646748682,
+      "loss": 0.8610644936561584,
+      "step": 802
+    },
+    {
+      "epoch": 0.3392405063291139,
+      "grad_norm": 0.5147035717964172,
+      "learning_rate": 0.00014112478031634446,
+      "loss": 0.9584825038909912,
+      "step": 804
+    },
+    {
+      "epoch": 0.340084388185654,
+      "grad_norm": 0.5931771397590637,
+      "learning_rate": 0.0001414762741652021,
+      "loss": 1.0142558813095093,
+      "step": 806
+    },
+    {
+      "epoch": 0.3409282700421941,
+      "grad_norm": 0.5178377032279968,
+      "learning_rate": 0.00014182776801405976,
+      "loss": 1.0078763961791992,
+      "step": 808
+    },
+    {
+      "epoch": 0.34177215189873417,
+      "grad_norm": 0.5453237295150757,
+      "learning_rate": 0.0001421792618629174,
+      "loss": 0.9107215404510498,
+      "step": 810
+    },
+    {
+      "epoch": 0.3426160337552743,
+      "grad_norm": 0.5886152982711792,
+      "learning_rate": 0.00014253075571177506,
+      "loss": 0.9981362819671631,
+      "step": 812
+    },
+    {
+      "epoch": 0.3434599156118143,
+      "grad_norm": 0.48040178418159485,
+      "learning_rate": 0.0001428822495606327,
+      "loss": 0.9636131525039673,
+      "step": 814
+    },
+    {
+      "epoch": 0.34430379746835443,
+      "grad_norm": 0.5011753439903259,
+      "learning_rate": 0.00014323374340949034,
+      "loss": 0.9590586423873901,
+      "step": 816
+    },
+    {
+      "epoch": 0.34514767932489454,
+      "grad_norm": 0.57858806848526,
+      "learning_rate": 0.00014358523725834798,
+      "loss": 0.978246808052063,
+      "step": 818
+    },
+    {
+      "epoch": 0.3459915611814346,
+      "grad_norm": 0.46092939376831055,
+      "learning_rate": 0.00014393673110720562,
+      "loss": 0.8549934029579163,
+      "step": 820
+    },
+    {
+      "epoch": 0.3468354430379747,
+      "grad_norm": 0.5756489038467407,
+      "learning_rate": 0.00014428822495606326,
+      "loss": 0.9771265387535095,
+      "step": 822
+    },
+    {
+      "epoch": 0.34767932489451475,
+      "grad_norm": 0.5501731634140015,
+      "learning_rate": 0.00014463971880492092,
+      "loss": 0.8739748001098633,
+      "step": 824
+    },
+    {
+      "epoch": 0.34852320675105486,
+      "grad_norm": 0.5451868176460266,
+      "learning_rate": 0.0001449912126537786,
+      "loss": 0.9129468202590942,
+      "step": 826
+    },
+    {
+      "epoch": 0.3493670886075949,
+      "grad_norm": 0.4624619781970978,
+      "learning_rate": 0.00014534270650263623,
+      "loss": 0.9196704030036926,
+      "step": 828
+    },
+    {
+      "epoch": 0.350210970464135,
+      "grad_norm": 0.520878791809082,
+      "learning_rate": 0.00014569420035149386,
+      "loss": 0.9976527690887451,
+      "step": 830
+    },
+    {
+      "epoch": 0.3510548523206751,
+      "grad_norm": 0.4469171464443207,
+      "learning_rate": 0.0001460456942003515,
+      "loss": 0.7753443717956543,
+      "step": 832
+    },
+    {
+      "epoch": 0.3518987341772152,
+      "grad_norm": 0.5105249881744385,
+      "learning_rate": 0.00014639718804920914,
+      "loss": 0.9584846496582031,
+      "step": 834
+    },
+    {
+      "epoch": 0.3527426160337553,
+      "grad_norm": 0.5043913125991821,
+      "learning_rate": 0.00014674868189806678,
+      "loss": 0.9013500213623047,
+      "step": 836
+    },
+    {
+      "epoch": 0.35358649789029534,
+      "grad_norm": 0.575850784778595,
+      "learning_rate": 0.00014710017574692442,
+      "loss": 0.9775562286376953,
+      "step": 838
+    },
+    {
+      "epoch": 0.35443037974683544,
+      "grad_norm": 0.5128876566886902,
+      "learning_rate": 0.00014745166959578208,
+      "loss": 0.9278940558433533,
+      "step": 840
+    },
+    {
+      "epoch": 0.35527426160337555,
+      "grad_norm": 0.5757885575294495,
+      "learning_rate": 0.00014780316344463975,
+      "loss": 1.0091488361358643,
+      "step": 842
+    },
+    {
+      "epoch": 0.3561181434599156,
+      "grad_norm": 0.500934898853302,
+      "learning_rate": 0.00014815465729349739,
+      "loss": 0.9286836981773376,
+      "step": 844
+    },
+    {
+      "epoch": 0.3569620253164557,
+      "grad_norm": 0.5220686197280884,
+      "learning_rate": 0.00014850615114235502,
+      "loss": 0.9484171867370605,
+      "step": 846
+    },
+    {
+      "epoch": 0.35780590717299576,
+      "grad_norm": 0.5494697690010071,
+      "learning_rate": 0.00014885764499121266,
+      "loss": 1.0556397438049316,
+      "step": 848
+    },
+    {
+      "epoch": 0.35864978902953587,
+      "grad_norm": 0.46633943915367126,
+      "learning_rate": 0.0001492091388400703,
+      "loss": 0.933089017868042,
+      "step": 850
+    },
+    {
+      "epoch": 0.3594936708860759,
+      "grad_norm": 0.4864962697029114,
+      "learning_rate": 0.00014956063268892794,
+      "loss": 1.0016963481903076,
+      "step": 852
+    },
+    {
+      "epoch": 0.36033755274261603,
+      "grad_norm": 0.5032764673233032,
+      "learning_rate": 0.0001499121265377856,
+      "loss": 0.9085348844528198,
+      "step": 854
+    },
+    {
+      "epoch": 0.36118143459915614,
+      "grad_norm": 0.5424998998641968,
+      "learning_rate": 0.00015026362038664324,
+      "loss": 0.923502504825592,
+      "step": 856
+    },
+    {
+      "epoch": 0.3620253164556962,
+      "grad_norm": 0.5181655287742615,
+      "learning_rate": 0.00015061511423550088,
+      "loss": 0.919174313545227,
+      "step": 858
+    },
+    {
+      "epoch": 0.3628691983122363,
+      "grad_norm": 0.5129443407058716,
+      "learning_rate": 0.00015096660808435855,
+      "loss": 0.851981520652771,
+      "step": 860
+    },
+    {
+      "epoch": 0.36371308016877635,
+      "grad_norm": 0.49540698528289795,
+      "learning_rate": 0.00015131810193321618,
+      "loss": 0.8633858561515808,
+      "step": 862
+    },
+    {
+      "epoch": 0.36455696202531646,
+      "grad_norm": 0.4706701934337616,
+      "learning_rate": 0.00015166959578207382,
+      "loss": 0.9473677277565002,
+      "step": 864
+    },
+    {
+      "epoch": 0.36540084388185656,
+      "grad_norm": 0.587704598903656,
+      "learning_rate": 0.00015202108963093146,
+      "loss": 0.942383885383606,
+      "step": 866
+    },
+    {
+      "epoch": 0.3662447257383966,
+      "grad_norm": 0.5851273536682129,
+      "learning_rate": 0.0001523725834797891,
+      "loss": 0.9811251163482666,
+      "step": 868
+    },
+    {
+      "epoch": 0.3670886075949367,
+      "grad_norm": 0.46357613801956177,
+      "learning_rate": 0.00015272407732864676,
+      "loss": 0.8151084184646606,
+      "step": 870
+    },
+    {
+      "epoch": 0.3679324894514768,
+      "grad_norm": 0.47250670194625854,
+      "learning_rate": 0.0001530755711775044,
+      "loss": 0.9023333191871643,
+      "step": 872
+    },
+    {
+      "epoch": 0.3687763713080169,
+      "grad_norm": 0.47510042786598206,
+      "learning_rate": 0.00015342706502636204,
+      "loss": 0.8758499622344971,
+      "step": 874
+    },
+    {
+      "epoch": 0.369620253164557,
+      "grad_norm": 0.5687124729156494,
+      "learning_rate": 0.00015377855887521968,
+      "loss": 0.9837421774864197,
+      "step": 876
+    },
+    {
+      "epoch": 0.37046413502109704,
+      "grad_norm": 0.49064236879348755,
+      "learning_rate": 0.00015413005272407735,
+      "loss": 0.9084216356277466,
+      "step": 878
+    },
+    {
+      "epoch": 0.37130801687763715,
+      "grad_norm": 0.5288164615631104,
+      "learning_rate": 0.00015448154657293498,
+      "loss": 0.8425542712211609,
+      "step": 880
+    },
+    {
+      "epoch": 0.3721518987341772,
+      "grad_norm": 0.5446951985359192,
+      "learning_rate": 0.00015483304042179262,
+      "loss": 1.006197452545166,
+      "step": 882
+    },
+    {
+      "epoch": 0.3729957805907173,
+      "grad_norm": 0.46872642636299133,
+      "learning_rate": 0.00015518453427065026,
+      "loss": 0.9779444932937622,
+      "step": 884
+    },
+    {
+      "epoch": 0.37383966244725736,
+      "grad_norm": 0.4852714240550995,
+      "learning_rate": 0.00015553602811950793,
+      "loss": 0.9099963307380676,
+      "step": 886
+    },
+    {
+      "epoch": 0.37468354430379747,
+      "grad_norm": 0.5219841003417969,
+      "learning_rate": 0.00015588752196836556,
+      "loss": 0.9730570316314697,
+      "step": 888
+    },
+    {
+      "epoch": 0.3755274261603376,
+      "grad_norm": 0.5258626341819763,
+      "learning_rate": 0.0001562390158172232,
+      "loss": 0.9304586052894592,
+      "step": 890
+    },
+    {
+      "epoch": 0.3763713080168776,
+      "grad_norm": 0.4266716241836548,
+      "learning_rate": 0.00015659050966608084,
+      "loss": 0.8680716156959534,
+      "step": 892
+    },
+    {
+      "epoch": 0.37721518987341773,
+      "grad_norm": 0.46361327171325684,
+      "learning_rate": 0.0001569420035149385,
+      "loss": 0.8428018093109131,
+      "step": 894
+    },
+    {
+      "epoch": 0.3780590717299578,
+      "grad_norm": 0.5313687920570374,
+      "learning_rate": 0.00015729349736379614,
+      "loss": 0.8465750217437744,
+      "step": 896
+    },
+    {
+      "epoch": 0.3789029535864979,
+      "grad_norm": 0.47166210412979126,
+      "learning_rate": 0.00015764499121265378,
+      "loss": 0.9120327234268188,
+      "step": 898
+    },
+    {
+      "epoch": 0.379746835443038,
+      "grad_norm": 0.4647318720817566,
+      "learning_rate": 0.00015799648506151145,
+      "loss": 0.8950425982475281,
+      "step": 900
+    },
+    {
+      "epoch": 0.379746835443038,
+      "eval_loss": 0.9643027186393738,
+      "eval_runtime": 691.7929,
+      "eval_samples_per_second": 3.046,
+      "eval_steps_per_second": 3.046,
+      "step": 900
+    },
+    {
+      "epoch": 0.38059071729957805,
+      "grad_norm": 0.5445119738578796,
+      "learning_rate": 0.00015834797891036909,
+      "loss": 0.908163845539093,
+      "step": 902
+    },
+    {
+      "epoch": 0.38143459915611816,
+      "grad_norm": 0.4311858117580414,
+      "learning_rate": 0.00015869947275922672,
+      "loss": 0.8945821523666382,
+      "step": 904
+    },
+    {
+      "epoch": 0.3822784810126582,
+      "grad_norm": 0.5590984225273132,
+      "learning_rate": 0.00015905096660808436,
+      "loss": 0.9478458762168884,
+      "step": 906
+    },
+    {
+      "epoch": 0.3831223628691983,
+      "grad_norm": 0.5470241904258728,
+      "learning_rate": 0.000159402460456942,
+      "loss": 0.9259957671165466,
+      "step": 908
+    },
+    {
+      "epoch": 0.38396624472573837,
+      "grad_norm": 0.5498791337013245,
+      "learning_rate": 0.00015975395430579964,
+      "loss": 0.8824930787086487,
+      "step": 910
+    },
+    {
+      "epoch": 0.3848101265822785,
+      "grad_norm": 0.4779198467731476,
+      "learning_rate": 0.0001601054481546573,
+      "loss": 0.8842340707778931,
+      "step": 912
+    },
+    {
+      "epoch": 0.3856540084388186,
+      "grad_norm": 0.5390620827674866,
+      "learning_rate": 0.00016045694200351494,
+      "loss": 0.92950040102005,
+      "step": 914
+    },
+    {
+      "epoch": 0.38649789029535864,
+      "grad_norm": 0.505519688129425,
+      "learning_rate": 0.0001608084358523726,
+      "loss": 0.8420897126197815,
+      "step": 916
+    },
+    {
+      "epoch": 0.38734177215189874,
+      "grad_norm": 0.4463907778263092,
+      "learning_rate": 0.00016115992970123025,
+      "loss": 0.8256624341011047,
+      "step": 918
+    },
+    {
+      "epoch": 0.3881856540084388,
+      "grad_norm": 0.5321422219276428,
+      "learning_rate": 0.00016151142355008788,
+      "loss": 0.8701168298721313,
+      "step": 920
+    },
+    {
+      "epoch": 0.3890295358649789,
+      "grad_norm": 0.4343073070049286,
+      "learning_rate": 0.00016186291739894552,
+      "loss": 0.9019309878349304,
+      "step": 922
+    },
+    {
+      "epoch": 0.389873417721519,
+      "grad_norm": 0.5311984419822693,
+      "learning_rate": 0.00016221441124780316,
+      "loss": 0.8560551404953003,
+      "step": 924
+    },
+    {
+      "epoch": 0.39071729957805906,
+      "grad_norm": 0.588691771030426,
+      "learning_rate": 0.0001625659050966608,
+      "loss": 0.9700050354003906,
+      "step": 926
+    },
+    {
+      "epoch": 0.39156118143459917,
+      "grad_norm": 0.5425586104393005,
+      "learning_rate": 0.00016291739894551844,
+      "loss": 1.043768048286438,
+      "step": 928
+    },
+    {
+      "epoch": 0.3924050632911392,
+      "grad_norm": 0.5228736996650696,
+      "learning_rate": 0.0001632688927943761,
+      "loss": 0.9501712918281555,
+      "step": 930
+    },
+    {
+      "epoch": 0.39324894514767933,
+      "grad_norm": 0.48960360884666443,
+      "learning_rate": 0.00016362038664323377,
+      "loss": 0.9223058223724365,
+      "step": 932
+    },
+    {
+      "epoch": 0.39409282700421944,
+      "grad_norm": 0.45204755663871765,
+      "learning_rate": 0.0001639718804920914,
+      "loss": 0.9692960977554321,
+      "step": 934
+    },
+    {
+      "epoch": 0.3949367088607595,
+      "grad_norm": 0.5299274921417236,
+      "learning_rate": 0.00016432337434094905,
+      "loss": 0.9467466473579407,
+      "step": 936
+    },
+    {
+      "epoch": 0.3957805907172996,
+      "grad_norm": 0.5607715249061584,
+      "learning_rate": 0.00016467486818980668,
+      "loss": 0.9118053317070007,
+      "step": 938
+    },
+    {
+      "epoch": 0.39662447257383965,
+      "grad_norm": 0.5271831154823303,
+      "learning_rate": 0.00016502636203866432,
+      "loss": 0.9131460189819336,
+      "step": 940
+    },
+    {
+      "epoch": 0.39746835443037976,
+      "grad_norm": 0.5075286030769348,
+      "learning_rate": 0.00016537785588752196,
+      "loss": 0.9358300566673279,
+      "step": 942
+    },
+    {
+      "epoch": 0.3983122362869198,
+      "grad_norm": 0.515731155872345,
+      "learning_rate": 0.00016572934973637963,
+      "loss": 0.8908210396766663,
+      "step": 944
+    },
+    {
+      "epoch": 0.3991561181434599,
+      "grad_norm": 0.4856977164745331,
+      "learning_rate": 0.00016608084358523726,
+      "loss": 0.9775290489196777,
+      "step": 946
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.48846355080604553,
+      "learning_rate": 0.00016643233743409493,
+      "loss": 0.8957490921020508,
+      "step": 948
+    },
+    {
+      "epoch": 0.4008438818565401,
+      "grad_norm": 0.42990800738334656,
+      "learning_rate": 0.00016678383128295257,
+      "loss": 0.9036174416542053,
+      "step": 950
+    },
+    {
+      "epoch": 0.4016877637130802,
+      "grad_norm": 0.49552062153816223,
+      "learning_rate": 0.0001671353251318102,
+      "loss": 0.991032600402832,
+      "step": 952
+    },
+    {
+      "epoch": 0.40253164556962023,
+      "grad_norm": 0.4565040171146393,
+      "learning_rate": 0.00016748681898066784,
+      "loss": 0.823063313961029,
+      "step": 954
+    },
+    {
+      "epoch": 0.40337552742616034,
+      "grad_norm": 0.4290153682231903,
+      "learning_rate": 0.00016783831282952548,
+      "loss": 0.8785063624382019,
+      "step": 956
+    },
+    {
+      "epoch": 0.40421940928270045,
+      "grad_norm": 0.5419702529907227,
+      "learning_rate": 0.00016818980667838312,
+      "loss": 0.8763971924781799,
+      "step": 958
+    },
+    {
+      "epoch": 0.4050632911392405,
+      "grad_norm": 0.5177501440048218,
+      "learning_rate": 0.00016854130052724079,
+      "loss": 0.9470553398132324,
+      "step": 960
+    },
+    {
+      "epoch": 0.4059071729957806,
+      "grad_norm": 0.539725661277771,
+      "learning_rate": 0.00016889279437609842,
+      "loss": 0.9235025644302368,
+      "step": 962
+    },
+    {
+      "epoch": 0.40675105485232066,
+      "grad_norm": 0.5324983596801758,
+      "learning_rate": 0.0001692442882249561,
+      "loss": 1.0248996019363403,
+      "step": 964
+    },
+    {
+      "epoch": 0.40759493670886077,
+      "grad_norm": 0.4936407506465912,
+      "learning_rate": 0.00016959578207381373,
+      "loss": 0.9076873660087585,
+      "step": 966
+    },
+    {
+      "epoch": 0.4084388185654008,
+      "grad_norm": 0.4960501194000244,
+      "learning_rate": 0.00016994727592267137,
+      "loss": 0.9162673950195312,
+      "step": 968
+    },
+    {
+      "epoch": 0.4092827004219409,
+      "grad_norm": 0.45093682408332825,
+      "learning_rate": 0.000170298769771529,
+      "loss": 0.904100775718689,
+      "step": 970
+    },
+    {
+      "epoch": 0.41012658227848103,
+      "grad_norm": 0.4560275077819824,
+      "learning_rate": 0.00017065026362038664,
+      "loss": 0.8633337020874023,
+      "step": 972
+    },
+    {
+      "epoch": 0.4109704641350211,
+      "grad_norm": 0.44885000586509705,
+      "learning_rate": 0.00017100175746924428,
+      "loss": 0.8454209566116333,
+      "step": 974
+    },
+    {
+      "epoch": 0.4118143459915612,
+      "grad_norm": 0.47251659631729126,
+      "learning_rate": 0.00017135325131810195,
+      "loss": 0.824730396270752,
+      "step": 976
+    },
+    {
+      "epoch": 0.41265822784810124,
+      "grad_norm": 0.6597666144371033,
+      "learning_rate": 0.00017170474516695959,
+      "loss": 0.9496501684188843,
+      "step": 978
+    },
+    {
+      "epoch": 0.41350210970464135,
+      "grad_norm": 0.39806297421455383,
+      "learning_rate": 0.00017205623901581722,
+      "loss": 0.9419087171554565,
+      "step": 980
+    },
+    {
+      "epoch": 0.41434599156118146,
+      "grad_norm": 0.48231109976768494,
+      "learning_rate": 0.0001724077328646749,
+      "loss": 0.9182976484298706,
+      "step": 982
+    },
+    {
+      "epoch": 0.4151898734177215,
+      "grad_norm": 0.5438776612281799,
+      "learning_rate": 0.00017275922671353253,
+      "loss": 0.9386967420578003,
+      "step": 984
+    },
+    {
+      "epoch": 0.4160337552742616,
+      "grad_norm": 0.4959667921066284,
+      "learning_rate": 0.00017311072056239017,
+      "loss": 0.897849440574646,
+      "step": 986
+    },
+    {
+      "epoch": 0.41687763713080167,
+      "grad_norm": 0.43533357977867126,
+      "learning_rate": 0.0001734622144112478,
+      "loss": 0.8776953816413879,
+      "step": 988
+    },
+    {
+      "epoch": 0.4177215189873418,
+      "grad_norm": 0.47513946890830994,
+      "learning_rate": 0.00017381370826010547,
+      "loss": 0.9162989854812622,
+      "step": 990
+    },
+    {
+      "epoch": 0.41856540084388183,
+      "grad_norm": 0.4907188415527344,
+      "learning_rate": 0.0001741652021089631,
+      "loss": 0.9482660889625549,
+      "step": 992
+    },
+    {
+      "epoch": 0.41940928270042194,
+      "grad_norm": 0.44499966502189636,
+      "learning_rate": 0.00017451669595782075,
+      "loss": 0.8812930583953857,
+      "step": 994
+    },
+    {
+      "epoch": 0.42025316455696204,
+      "grad_norm": 0.4535730481147766,
+      "learning_rate": 0.00017486818980667838,
+      "loss": 0.9439874887466431,
+      "step": 996
+    },
+    {
+      "epoch": 0.4210970464135021,
+      "grad_norm": 0.5240745544433594,
+      "learning_rate": 0.00017521968365553602,
+      "loss": 0.8818395137786865,
+      "step": 998
+    },
+    {
+      "epoch": 0.4219409282700422,
+      "grad_norm": 0.5301211476325989,
+      "learning_rate": 0.0001755711775043937,
+      "loss": 0.886186957359314,
+      "step": 1000
+    },
+    {
+      "epoch": 0.4219409282700422,
+      "eval_loss": 0.9487298727035522,
+      "eval_runtime": 689.4288,
+      "eval_samples_per_second": 3.056,
+      "eval_steps_per_second": 3.056,
+      "step": 1000
+    },
+    {
+      "epoch": 0.42278481012658226,
+      "grad_norm": 0.47876957058906555,
+      "learning_rate": 0.00017592267135325133,
+      "loss": 0.8814021348953247,
+      "step": 1002
+    },
+    {
+      "epoch": 0.42362869198312236,
+      "grad_norm": 0.4929780960083008,
+      "learning_rate": 0.00017627416520210896,
+      "loss": 0.8295068740844727,
+      "step": 1004
+    },
+    {
+      "epoch": 0.42447257383966247,
+      "grad_norm": 0.41888436675071716,
+      "learning_rate": 0.00017662565905096663,
+      "loss": 0.8364827036857605,
+      "step": 1006
+    },
+    {
+      "epoch": 0.4253164556962025,
+      "grad_norm": 0.5175151824951172,
+      "learning_rate": 0.00017697715289982427,
+      "loss": 1.0167189836502075,
+      "step": 1008
+    },
+    {
+      "epoch": 0.42616033755274263,
+      "grad_norm": 0.4815356135368347,
+      "learning_rate": 0.0001773286467486819,
+      "loss": 0.8460752964019775,
+      "step": 1010
+    },
+    {
+      "epoch": 0.4270042194092827,
+      "grad_norm": 0.5210875272750854,
+      "learning_rate": 0.00017768014059753954,
+      "loss": 0.8595574498176575,
+      "step": 1012
+    },
+    {
+      "epoch": 0.4278481012658228,
+      "grad_norm": 0.4400486946105957,
+      "learning_rate": 0.00017803163444639718,
+      "loss": 0.8764723539352417,
+      "step": 1014
+    },
+    {
+      "epoch": 0.4286919831223629,
+      "grad_norm": 0.5282127857208252,
+      "learning_rate": 0.00017838312829525482,
+      "loss": 0.9706798791885376,
+      "step": 1016
+    },
+    {
+      "epoch": 0.42953586497890295,
+      "grad_norm": 0.47523441910743713,
+      "learning_rate": 0.00017873462214411249,
+      "loss": 0.8912002444267273,
+      "step": 1018
+    },
+    {
+      "epoch": 0.43037974683544306,
+      "grad_norm": 0.49640706181526184,
+      "learning_rate": 0.00017908611599297012,
+      "loss": 0.8835636973381042,
+      "step": 1020
+    },
+    {
+      "epoch": 0.4312236286919831,
+      "grad_norm": 0.5253039002418518,
+      "learning_rate": 0.0001794376098418278,
+      "loss": 0.8711735606193542,
+      "step": 1022
+    },
+    {
+      "epoch": 0.4320675105485232,
+      "grad_norm": 0.49285009503364563,
+      "learning_rate": 0.00017978910369068543,
+      "loss": 0.9064869284629822,
+      "step": 1024
+    },
+    {
+      "epoch": 0.43291139240506327,
+      "grad_norm": 0.48758041858673096,
+      "learning_rate": 0.00018014059753954307,
+      "loss": 0.8993359208106995,
+      "step": 1026
+    },
+    {
+      "epoch": 0.4337552742616034,
+      "grad_norm": 0.4815461039543152,
+      "learning_rate": 0.0001804920913884007,
+      "loss": 0.8747937679290771,
+      "step": 1028
+    },
+    {
+      "epoch": 0.4345991561181435,
+      "grad_norm": 0.4930349886417389,
+      "learning_rate": 0.00018084358523725834,
+      "loss": 0.8879084587097168,
+      "step": 1030
+    },
+    {
+      "epoch": 0.43544303797468353,
+      "grad_norm": 0.49303027987480164,
+      "learning_rate": 0.00018119507908611598,
+      "loss": 1.0023083686828613,
+      "step": 1032
+    },
+    {
+      "epoch": 0.43628691983122364,
+      "grad_norm": 0.5312249064445496,
+      "learning_rate": 0.00018154657293497365,
+      "loss": 0.938680112361908,
+      "step": 1034
+    },
+    {
+      "epoch": 0.4371308016877637,
+      "grad_norm": 0.5174582004547119,
+      "learning_rate": 0.0001818980667838313,
+      "loss": 0.854195773601532,
+      "step": 1036
+    },
+    {
+      "epoch": 0.4379746835443038,
+      "grad_norm": 0.5452545881271362,
+      "learning_rate": 0.00018224956063268895,
+      "loss": 0.9059375524520874,
+      "step": 1038
+    },
+    {
+      "epoch": 0.4388185654008439,
+      "grad_norm": 0.5480839014053345,
+      "learning_rate": 0.0001826010544815466,
+      "loss": 0.9708920121192932,
+      "step": 1040
+    },
+    {
+      "epoch": 0.43966244725738396,
+      "grad_norm": 0.44693151116371155,
+      "learning_rate": 0.00018295254833040423,
+      "loss": 0.8561046123504639,
+      "step": 1042
+    },
+    {
+      "epoch": 0.44050632911392407,
+      "grad_norm": 0.4024234116077423,
+      "learning_rate": 0.00018330404217926187,
+      "loss": 0.8811968564987183,
+      "step": 1044
+    },
+    {
+      "epoch": 0.4413502109704641,
+      "grad_norm": 0.5586408376693726,
+      "learning_rate": 0.0001836555360281195,
+      "loss": 0.978068470954895,
+      "step": 1046
+    },
+    {
+      "epoch": 0.4421940928270042,
+      "grad_norm": 0.4918624758720398,
+      "learning_rate": 0.00018400702987697714,
+      "loss": 0.8640981316566467,
+      "step": 1048
+    },
+    {
+      "epoch": 0.4430379746835443,
+      "grad_norm": 0.4335230588912964,
+      "learning_rate": 0.0001843585237258348,
+      "loss": 0.9077964425086975,
+      "step": 1050
+    },
+    {
+      "epoch": 0.4438818565400844,
+      "grad_norm": 0.5275123715400696,
+      "learning_rate": 0.00018471001757469247,
+      "loss": 0.960682213306427,
+      "step": 1052
+    },
+    {
+      "epoch": 0.4447257383966245,
+      "grad_norm": 0.5144415497779846,
+      "learning_rate": 0.0001850615114235501,
+      "loss": 0.8549577593803406,
+      "step": 1054
+    },
+    {
+      "epoch": 0.44556962025316454,
+      "grad_norm": 0.5166662931442261,
+      "learning_rate": 0.00018541300527240775,
+      "loss": 0.9320827126502991,
+      "step": 1056
+    },
+    {
+      "epoch": 0.44641350210970465,
+      "grad_norm": 0.5011980533599854,
+      "learning_rate": 0.0001857644991212654,
+      "loss": 0.9632431864738464,
+      "step": 1058
+    },
+    {
+      "epoch": 0.4472573839662447,
+      "grad_norm": 0.49312469363212585,
+      "learning_rate": 0.00018611599297012303,
+      "loss": 0.9442946910858154,
+      "step": 1060
+    },
+    {
+      "epoch": 0.4481012658227848,
+      "grad_norm": 0.49958568811416626,
+      "learning_rate": 0.00018646748681898066,
+      "loss": 0.952802300453186,
+      "step": 1062
+    },
+    {
+      "epoch": 0.4489451476793249,
+      "grad_norm": 0.49827462434768677,
+      "learning_rate": 0.00018681898066783833,
+      "loss": 0.9630650877952576,
+      "step": 1064
+    },
+    {
+      "epoch": 0.44978902953586497,
+      "grad_norm": 0.523980438709259,
+      "learning_rate": 0.00018717047451669597,
+      "loss": 0.9081395268440247,
+      "step": 1066
+    },
+    {
+      "epoch": 0.4506329113924051,
+      "grad_norm": 0.5108568668365479,
+      "learning_rate": 0.0001875219683655536,
+      "loss": 0.9398958683013916,
+      "step": 1068
+    },
+    {
+      "epoch": 0.45147679324894513,
+      "grad_norm": 0.4453965127468109,
+      "learning_rate": 0.00018787346221441127,
+      "loss": 0.93592768907547,
+      "step": 1070
+    },
+    {
+      "epoch": 0.45232067510548524,
+      "grad_norm": 0.4675683081150055,
+      "learning_rate": 0.0001882249560632689,
+      "loss": 0.8879633545875549,
+      "step": 1072
+    },
+    {
+      "epoch": 0.4531645569620253,
+      "grad_norm": 0.47398847341537476,
+      "learning_rate": 0.00018857644991212655,
+      "loss": 0.9966004490852356,
+      "step": 1074
+    },
+    {
+      "epoch": 0.4540084388185654,
+      "grad_norm": 0.46053192019462585,
+      "learning_rate": 0.0001889279437609842,
+      "loss": 0.9184179902076721,
+      "step": 1076
+    },
+    {
+      "epoch": 0.4548523206751055,
+      "grad_norm": 0.5601398348808289,
+      "learning_rate": 0.00018927943760984182,
+      "loss": 0.9434974193572998,
+      "step": 1078
+    },
+    {
+      "epoch": 0.45569620253164556,
+      "grad_norm": 0.48422637581825256,
+      "learning_rate": 0.0001896309314586995,
+      "loss": 0.9522465467453003,
+      "step": 1080
+    },
+    {
+      "epoch": 0.45654008438818566,
+      "grad_norm": 0.46280911564826965,
+      "learning_rate": 0.00018998242530755713,
+      "loss": 0.9315434694290161,
+      "step": 1082
+    },
+    {
+      "epoch": 0.4573839662447257,
+      "grad_norm": 0.4658683240413666,
+      "learning_rate": 0.00019033391915641477,
+      "loss": 0.8165783286094666,
+      "step": 1084
+    },
+    {
+      "epoch": 0.4582278481012658,
+      "grad_norm": 0.4192182123661041,
+      "learning_rate": 0.0001906854130052724,
+      "loss": 0.8024274110794067,
+      "step": 1086
+    },
+    {
+      "epoch": 0.45907172995780593,
+      "grad_norm": 0.4178735613822937,
+      "learning_rate": 0.00019103690685413007,
+      "loss": 0.8620653748512268,
+      "step": 1088
+    },
+    {
+      "epoch": 0.459915611814346,
+      "grad_norm": 0.4488574266433716,
+      "learning_rate": 0.0001913884007029877,
+      "loss": 0.9121530055999756,
+      "step": 1090
+    },
+    {
+      "epoch": 0.4607594936708861,
+      "grad_norm": 0.5164965987205505,
+      "learning_rate": 0.00019173989455184535,
+      "loss": 0.9496700763702393,
+      "step": 1092
+    },
+    {
+      "epoch": 0.46160337552742614,
+      "grad_norm": 0.41563132405281067,
+      "learning_rate": 0.00019209138840070299,
+      "loss": 0.8757708668708801,
+      "step": 1094
+    },
+    {
+      "epoch": 0.46244725738396625,
+      "grad_norm": 2.3222429752349854,
+      "learning_rate": 0.00019244288224956065,
+      "loss": 0.8406533002853394,
+      "step": 1096
+    },
+    {
+      "epoch": 0.46329113924050636,
+      "grad_norm": 0.4813845157623291,
+      "learning_rate": 0.0001927943760984183,
+      "loss": 0.9459465742111206,
+      "step": 1098
+    },
+    {
+      "epoch": 0.4641350210970464,
+      "grad_norm": 0.6233882308006287,
+      "learning_rate": 0.00019314586994727593,
+      "loss": 0.9376904368400574,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4641350210970464,
+      "eval_loss": 0.9357889294624329,
+      "eval_runtime": 676.9573,
+      "eval_samples_per_second": 3.112,
+      "eval_steps_per_second": 3.112,
+      "step": 1100
+    },
+    {
+      "epoch": 0.4649789029535865,
+      "grad_norm": 0.5125579237937927,
+      "learning_rate": 0.00019349736379613357,
+      "loss": 0.8998825550079346,
+      "step": 1102
+    },
+    {
+      "epoch": 0.46582278481012657,
+      "grad_norm": 0.4534320831298828,
+      "learning_rate": 0.00019384885764499123,
+      "loss": 0.9213768839836121,
+      "step": 1104
+    },
+    {
+      "epoch": 0.4666666666666667,
+      "grad_norm": 0.4715143144130707,
+      "learning_rate": 0.00019420035149384887,
+      "loss": 0.9739661812782288,
+      "step": 1106
+    },
+    {
+      "epoch": 0.4675105485232067,
+      "grad_norm": 0.45529672503471375,
+      "learning_rate": 0.0001945518453427065,
+      "loss": 0.9113216400146484,
+      "step": 1108
+    },
+    {
+      "epoch": 0.46835443037974683,
+      "grad_norm": 0.48199015855789185,
+      "learning_rate": 0.00019490333919156417,
+      "loss": 0.92528235912323,
+      "step": 1110
+    },
+    {
+      "epoch": 0.46919831223628694,
+      "grad_norm": 0.4425188899040222,
+      "learning_rate": 0.0001952548330404218,
+      "loss": 0.8612716197967529,
+      "step": 1112
+    },
+    {
+      "epoch": 0.470042194092827,
+      "grad_norm": 0.4311593770980835,
+      "learning_rate": 0.00019560632688927945,
+      "loss": 0.8901699185371399,
+      "step": 1114
+    },
+    {
+      "epoch": 0.4708860759493671,
+      "grad_norm": 0.497806578874588,
+      "learning_rate": 0.0001959578207381371,
+      "loss": 0.9271994829177856,
+      "step": 1116
+    },
+    {
+      "epoch": 0.47172995780590715,
+      "grad_norm": 0.47149473428726196,
+      "learning_rate": 0.00019630931458699473,
+      "loss": 0.8740925788879395,
+      "step": 1118
+    },
+    {
+      "epoch": 0.47257383966244726,
+      "grad_norm": 0.5029966235160828,
+      "learning_rate": 0.00019666080843585236,
+      "loss": 0.9190115928649902,
+      "step": 1120
+    },
+    {
+      "epoch": 0.47341772151898737,
+      "grad_norm": 0.47972601652145386,
+      "learning_rate": 0.00019701230228471003,
+      "loss": 1.0043057203292847,
+      "step": 1122
+    },
+    {
+      "epoch": 0.4742616033755274,
+      "grad_norm": 0.45096471905708313,
+      "learning_rate": 0.00019736379613356767,
+      "loss": 0.9472925066947937,
+      "step": 1124
+    },
+    {
+      "epoch": 0.4751054852320675,
+      "grad_norm": 0.459852933883667,
+      "learning_rate": 0.00019771528998242533,
+      "loss": 0.8043124079704285,
+      "step": 1126
+    },
+    {
+      "epoch": 0.4759493670886076,
+      "grad_norm": 0.5313422679901123,
+      "learning_rate": 0.00019806678383128297,
+      "loss": 0.9662142395973206,
+      "step": 1128
+    },
+    {
+      "epoch": 0.4767932489451477,
+      "grad_norm": 0.43474531173706055,
+      "learning_rate": 0.0001984182776801406,
+      "loss": 0.8691151738166809,
+      "step": 1130
+    },
+    {
+      "epoch": 0.47763713080168774,
+      "grad_norm": 0.5232312083244324,
+      "learning_rate": 0.00019876977152899825,
+      "loss": 0.9024254083633423,
+      "step": 1132
+    },
+    {
+      "epoch": 0.47848101265822784,
+      "grad_norm": 0.47671905159950256,
+      "learning_rate": 0.0001991212653778559,
+      "loss": 0.8996873497962952,
+      "step": 1134
+    },
+    {
+      "epoch": 0.47932489451476795,
+      "grad_norm": 0.42289480566978455,
+      "learning_rate": 0.00019947275922671353,
+      "loss": 0.8336917757987976,
+      "step": 1136
+    },
+    {
+      "epoch": 0.480168776371308,
+      "grad_norm": 0.4700844883918762,
+      "learning_rate": 0.0001998242530755712,
+      "loss": 0.8943206071853638,
+      "step": 1138
+    },
+    {
+      "epoch": 0.4810126582278481,
+      "grad_norm": 0.5341399312019348,
+      "learning_rate": 0.00019999999711649004,
+      "loss": 0.9340365529060364,
+      "step": 1140
+    },
+    {
+      "epoch": 0.48185654008438816,
+      "grad_norm": 0.46169522404670715,
+      "learning_rate": 0.00019999997404841123,
+      "loss": 0.8241778016090393,
+      "step": 1142
+    },
+    {
+      "epoch": 0.48270042194092827,
+      "grad_norm": 0.5475223064422607,
+      "learning_rate": 0.00019999992791225896,
+      "loss": 1.0096158981323242,
+      "step": 1144
+    },
+    {
+      "epoch": 0.4835443037974684,
+      "grad_norm": 0.524641215801239,
+      "learning_rate": 0.00019999985870804385,
+      "loss": 0.9650378227233887,
+      "step": 1146
+    },
+    {
+      "epoch": 0.48438818565400843,
+      "grad_norm": 0.4326174855232239,
+      "learning_rate": 0.00019999976643578186,
+      "loss": 0.9003009796142578,
+      "step": 1148
+    },
+    {
+      "epoch": 0.48523206751054854,
+      "grad_norm": 0.49034059047698975,
+      "learning_rate": 0.0001999996510954943,
+      "loss": 0.9793432950973511,
+      "step": 1150
+    },
+    {
+      "epoch": 0.4860759493670886,
+      "grad_norm": 0.4760092496871948,
+      "learning_rate": 0.00019999951268720776,
+      "loss": 0.8793007731437683,
+      "step": 1152
+    },
+    {
+      "epoch": 0.4869198312236287,
+      "grad_norm": 0.44606879353523254,
+      "learning_rate": 0.00019999935121095417,
+      "loss": 0.888630747795105,
+      "step": 1154
+    },
+    {
+      "epoch": 0.4877637130801688,
+      "grad_norm": 0.5030332207679749,
+      "learning_rate": 0.0001999991666667708,
+      "loss": 0.9660000801086426,
+      "step": 1156
+    },
+    {
+      "epoch": 0.48860759493670886,
+      "grad_norm": 0.4295555055141449,
+      "learning_rate": 0.00019999895905470014,
+      "loss": 0.824654757976532,
+      "step": 1158
+    },
+    {
+      "epoch": 0.48945147679324896,
+      "grad_norm": 0.392167866230011,
+      "learning_rate": 0.0001999987283747902,
+      "loss": 0.8159562349319458,
+      "step": 1160
+    },
+    {
+      "epoch": 0.490295358649789,
+      "grad_norm": 0.484611839056015,
+      "learning_rate": 0.00019999847462709412,
+      "loss": 0.9630686044692993,
+      "step": 1162
+    },
+    {
+      "epoch": 0.4911392405063291,
+      "grad_norm": 0.45891445875167847,
+      "learning_rate": 0.00019999819781167042,
+      "loss": 0.8396129608154297,
+      "step": 1164
+    },
+    {
+      "epoch": 0.4919831223628692,
+      "grad_norm": 0.5553452968597412,
+      "learning_rate": 0.00019999789792858304,
+      "loss": 0.8194513916969299,
+      "step": 1166
+    },
+    {
+      "epoch": 0.4928270042194093,
+      "grad_norm": 0.38998672366142273,
+      "learning_rate": 0.00019999757497790106,
+      "loss": 0.9268721342086792,
+      "step": 1168
+    },
+    {
+      "epoch": 0.4936708860759494,
+      "grad_norm": 0.45445355772972107,
+      "learning_rate": 0.00019999722895969904,
+      "loss": 0.9204684495925903,
+      "step": 1170
+    },
+    {
+      "epoch": 0.49451476793248944,
+      "grad_norm": 0.484225332736969,
+      "learning_rate": 0.00019999685987405678,
+      "loss": 0.9270301461219788,
+      "step": 1172
+    },
+    {
+      "epoch": 0.49535864978902955,
+      "grad_norm": 0.45215415954589844,
+      "learning_rate": 0.00019999646772105942,
+      "loss": 0.8782645463943481,
+      "step": 1174
+    },
+    {
+      "epoch": 0.4962025316455696,
+      "grad_norm": 0.41578832268714905,
+      "learning_rate": 0.00019999605250079744,
+      "loss": 0.873112678527832,
+      "step": 1176
+    },
+    {
+      "epoch": 0.4970464135021097,
+      "grad_norm": 0.4632788300514221,
+      "learning_rate": 0.0001999956142133666,
+      "loss": 0.9593189358711243,
+      "step": 1178
+    },
+    {
+      "epoch": 0.4978902953586498,
+      "grad_norm": 0.5007622838020325,
+      "learning_rate": 0.000199995152858868,
+      "loss": 0.9215621948242188,
+      "step": 1180
+    },
+    {
+      "epoch": 0.49873417721518987,
+      "grad_norm": 0.5201935768127441,
+      "learning_rate": 0.0001999946684374081,
+      "loss": 0.8964219689369202,
+      "step": 1182
+    },
+    {
+      "epoch": 0.49957805907173,
+      "grad_norm": 0.5053967237472534,
+      "learning_rate": 0.0001999941609490986,
+      "loss": 0.8528663516044617,
+      "step": 1184
+    },
+    {
+      "epoch": 0.5004219409282701,
+      "grad_norm": 0.44397974014282227,
+      "learning_rate": 0.0001999936303940566,
+      "loss": 0.8214734792709351,
+      "step": 1186
+    },
+    {
+      "epoch": 0.5012658227848101,
+      "grad_norm": 0.4563101530075073,
+      "learning_rate": 0.0001999930767724045,
+      "loss": 0.8882166743278503,
+      "step": 1188
+    },
+    {
+      "epoch": 0.5021097046413502,
+      "grad_norm": 0.4666728675365448,
+      "learning_rate": 0.00019999250008426997,
+      "loss": 0.8882588148117065,
+      "step": 1190
+    },
+    {
+      "epoch": 0.5029535864978903,
+      "grad_norm": 0.4423629939556122,
+      "learning_rate": 0.00019999190032978607,
+      "loss": 0.8353691697120667,
+      "step": 1192
+    },
+    {
+      "epoch": 0.5037974683544304,
+      "grad_norm": 0.4514595866203308,
+      "learning_rate": 0.00019999127750909118,
+      "loss": 0.8721219301223755,
+      "step": 1194
+    },
+    {
+      "epoch": 0.5046413502109705,
+      "grad_norm": 0.48469996452331543,
+      "learning_rate": 0.00019999063162232886,
+      "loss": 0.902795135974884,
+      "step": 1196
+    },
+    {
+      "epoch": 0.5054852320675105,
+      "grad_norm": 0.4931983947753906,
+      "learning_rate": 0.00019998996266964823,
+      "loss": 0.8843175768852234,
+      "step": 1198
+    },
+    {
+      "epoch": 0.5063291139240507,
+      "grad_norm": 0.507408082485199,
+      "learning_rate": 0.00019998927065120357,
+      "loss": 0.9179208278656006,
+      "step": 1200
+    },
+    {
+      "epoch": 0.5063291139240507,
+      "eval_loss": 0.9224098324775696,
+      "eval_runtime": 669.7542,
+      "eval_samples_per_second": 3.146,
+      "eval_steps_per_second": 3.146,
+      "step": 1200
+    },
+    {
+      "epoch": 0.5071729957805907,
+      "grad_norm": 0.4024188220500946,
+      "learning_rate": 0.00019998855556715447,
+      "loss": 0.9078981876373291,
+      "step": 1202
+    },
+    {
+      "epoch": 0.5080168776371308,
+      "grad_norm": 0.6558667421340942,
+      "learning_rate": 0.00019998781741766594,
+      "loss": 0.8089252710342407,
+      "step": 1204
+    },
+    {
+      "epoch": 0.5088607594936709,
+      "grad_norm": 0.45128145813941956,
+      "learning_rate": 0.00019998705620290823,
+      "loss": 0.9316248297691345,
+      "step": 1206
+    },
+    {
+      "epoch": 0.509704641350211,
+      "grad_norm": 0.4733511507511139,
+      "learning_rate": 0.00019998627192305694,
+      "loss": 0.9047867655754089,
+      "step": 1208
+    },
+    {
+      "epoch": 0.510548523206751,
+      "grad_norm": 0.4026021659374237,
+      "learning_rate": 0.00019998546457829298,
+      "loss": 0.8507166504859924,
+      "step": 1210
+    },
+    {
+      "epoch": 0.5113924050632911,
+      "grad_norm": 0.43416184186935425,
+      "learning_rate": 0.00019998463416880262,
+      "loss": 0.8490090370178223,
+      "step": 1212
+    },
+    {
+      "epoch": 0.5122362869198313,
+      "grad_norm": 0.3678364157676697,
+      "learning_rate": 0.0001999837806947774,
+      "loss": 0.7730492949485779,
+      "step": 1214
+    },
+    {
+      "epoch": 0.5130801687763713,
+      "grad_norm": 0.49042677879333496,
+      "learning_rate": 0.00019998290415641415,
+      "loss": 0.9625269174575806,
+      "step": 1216
+    },
+    {
+      "epoch": 0.5139240506329114,
+      "grad_norm": 0.4576701819896698,
+      "learning_rate": 0.00019998200455391516,
+      "loss": 0.8961732983589172,
+      "step": 1218
+    },
+    {
+      "epoch": 0.5147679324894515,
+      "grad_norm": 0.4465518593788147,
+      "learning_rate": 0.00019998108188748793,
+      "loss": 0.8508996367454529,
+      "step": 1220
+    },
+    {
+      "epoch": 0.5156118143459916,
+      "grad_norm": 0.4267960488796234,
+      "learning_rate": 0.00019998013615734524,
+      "loss": 0.8664930462837219,
+      "step": 1222
+    },
+    {
+      "epoch": 0.5164556962025316,
+      "grad_norm": 0.39936602115631104,
+      "learning_rate": 0.0001999791673637053,
+      "loss": 0.8841317892074585,
+      "step": 1224
+    },
+    {
+      "epoch": 0.5172995780590718,
+      "grad_norm": 0.4442414343357086,
+      "learning_rate": 0.0001999781755067916,
+      "loss": 0.8601276874542236,
+      "step": 1226
+    },
+    {
+      "epoch": 0.5181434599156118,
+      "grad_norm": 0.4397028088569641,
+      "learning_rate": 0.00019997716058683292,
+      "loss": 0.8377046585083008,
+      "step": 1228
+    },
+    {
+      "epoch": 0.5189873417721519,
+      "grad_norm": 0.5327648520469666,
+      "learning_rate": 0.0001999761226040634,
+      "loss": 0.9853615164756775,
+      "step": 1230
+    },
+    {
+      "epoch": 0.5198312236286919,
+      "grad_norm": 0.40931421518325806,
+      "learning_rate": 0.00019997506155872244,
+      "loss": 0.9235715866088867,
+      "step": 1232
+    },
+    {
+      "epoch": 0.5206751054852321,
+      "grad_norm": 0.4599161148071289,
+      "learning_rate": 0.00019997397745105487,
+      "loss": 0.868396520614624,
+      "step": 1234
+    },
+    {
+      "epoch": 0.5215189873417722,
+      "grad_norm": 0.4813833236694336,
+      "learning_rate": 0.0001999728702813107,
+      "loss": 0.9946733117103577,
+      "step": 1236
+    },
+    {
+      "epoch": 0.5223628691983122,
+      "grad_norm": 0.43518269062042236,
+      "learning_rate": 0.00019997174004974543,
+      "loss": 0.8608635067939758,
+      "step": 1238
+    },
+    {
+      "epoch": 0.5232067510548524,
+      "grad_norm": 0.45646214485168457,
+      "learning_rate": 0.0001999705867566197,
+      "loss": 0.819354772567749,
+      "step": 1240
+    },
+    {
+      "epoch": 0.5240506329113924,
+      "grad_norm": 0.46834197640419006,
+      "learning_rate": 0.00019996941040219954,
+      "loss": 0.8934658765792847,
+      "step": 1242
+    },
+    {
+      "epoch": 0.5248945147679325,
+      "grad_norm": 0.46065032482147217,
+      "learning_rate": 0.0001999682109867564,
+      "loss": 0.8624778985977173,
+      "step": 1244
+    },
+    {
+      "epoch": 0.5257383966244725,
+      "grad_norm": 0.4506741166114807,
+      "learning_rate": 0.00019996698851056688,
+      "loss": 0.8540882468223572,
+      "step": 1246
+    },
+    {
+      "epoch": 0.5265822784810127,
+      "grad_norm": 0.4550519287586212,
+      "learning_rate": 0.00019996574297391302,
+      "loss": 0.9003006219863892,
+      "step": 1248
+    },
+    {
+      "epoch": 0.5274261603375527,
+      "grad_norm": 0.4097813367843628,
+      "learning_rate": 0.00019996447437708214,
+      "loss": 0.8490248918533325,
+      "step": 1250
+    },
+    {
+      "epoch": 0.5282700421940928,
+      "grad_norm": 0.4375711679458618,
+      "learning_rate": 0.00019996318272036688,
+      "loss": 0.9204569458961487,
+      "step": 1252
+    },
+    {
+      "epoch": 0.529113924050633,
+      "grad_norm": 0.4335198998451233,
+      "learning_rate": 0.00019996186800406518,
+      "loss": 0.8831789493560791,
+      "step": 1254
+    },
+    {
+      "epoch": 0.529957805907173,
+      "grad_norm": 0.4840933084487915,
+      "learning_rate": 0.00019996053022848035,
+      "loss": 0.8578064441680908,
+      "step": 1256
+    },
+    {
+      "epoch": 0.5308016877637131,
+      "grad_norm": 0.48911216855049133,
+      "learning_rate": 0.00019995916939392097,
+      "loss": 0.9235416650772095,
+      "step": 1258
+    },
+    {
+      "epoch": 0.5316455696202531,
+      "grad_norm": 0.41106730699539185,
+      "learning_rate": 0.00019995778550070098,
+      "loss": 0.7897764444351196,
+      "step": 1260
+    },
+    {
+      "epoch": 0.5324894514767933,
+      "grad_norm": 0.42233356833457947,
+      "learning_rate": 0.00019995637854913957,
+      "loss": 0.9392989873886108,
+      "step": 1262
+    },
+    {
+      "epoch": 0.5333333333333333,
+      "grad_norm": 0.39145180583000183,
+      "learning_rate": 0.0001999549485395613,
+      "loss": 0.9130091667175293,
+      "step": 1264
+    },
+    {
+      "epoch": 0.5341772151898734,
+      "grad_norm": 0.42780739068984985,
+      "learning_rate": 0.00019995349547229614,
+      "loss": 0.9690561890602112,
+      "step": 1266
+    },
+    {
+      "epoch": 0.5350210970464135,
+      "grad_norm": 0.4823327362537384,
+      "learning_rate": 0.0001999520193476792,
+      "loss": 0.9990131258964539,
+      "step": 1268
+    },
+    {
+      "epoch": 0.5358649789029536,
+      "grad_norm": 0.4669715166091919,
+      "learning_rate": 0.00019995052016605097,
+      "loss": 0.8343052864074707,
+      "step": 1270
+    },
+    {
+      "epoch": 0.5367088607594936,
+      "grad_norm": 0.4024743437767029,
+      "learning_rate": 0.0001999489979277574,
+      "loss": 0.9605970978736877,
+      "step": 1272
+    },
+    {
+      "epoch": 0.5375527426160338,
+      "grad_norm": 0.42063432931900024,
+      "learning_rate": 0.0001999474526331495,
+      "loss": 0.8164438009262085,
+      "step": 1274
+    },
+    {
+      "epoch": 0.5383966244725739,
+      "grad_norm": 0.47953495383262634,
+      "learning_rate": 0.00019994588428258383,
+      "loss": 0.879767656326294,
+      "step": 1276
+    },
+    {
+      "epoch": 0.5392405063291139,
+      "grad_norm": 0.39668557047843933,
+      "learning_rate": 0.00019994429287642214,
+      "loss": 0.7658726572990417,
+      "step": 1278
+    },
+    {
+      "epoch": 0.540084388185654,
+      "grad_norm": 0.5136121511459351,
+      "learning_rate": 0.00019994267841503155,
+      "loss": 0.9477384686470032,
+      "step": 1280
+    },
+    {
+      "epoch": 0.5409282700421941,
+      "grad_norm": 0.4039286673069,
+      "learning_rate": 0.00019994104089878452,
+      "loss": 0.7883896827697754,
+      "step": 1282
+    },
+    {
+      "epoch": 0.5417721518987342,
+      "grad_norm": 0.4425487816333771,
+      "learning_rate": 0.0001999393803280587,
+      "loss": 0.8526129126548767,
+      "step": 1284
+    },
+    {
+      "epoch": 0.5426160337552742,
+      "grad_norm": 0.4544559121131897,
+      "learning_rate": 0.00019993769670323725,
+      "loss": 0.8694683909416199,
+      "step": 1286
+    },
+    {
+      "epoch": 0.5434599156118144,
+      "grad_norm": 0.42872917652130127,
+      "learning_rate": 0.0001999359900247085,
+      "loss": 0.8211527466773987,
+      "step": 1288
+    },
+    {
+      "epoch": 0.5443037974683544,
+      "grad_norm": 0.41549673676490784,
+      "learning_rate": 0.00019993426029286617,
+      "loss": 0.7967932224273682,
+      "step": 1290
+    },
+    {
+      "epoch": 0.5451476793248945,
+      "grad_norm": 0.42888355255126953,
+      "learning_rate": 0.00019993250750810926,
+      "loss": 0.8291563987731934,
+      "step": 1292
+    },
+    {
+      "epoch": 0.5459915611814345,
+      "grad_norm": 0.44985485076904297,
+      "learning_rate": 0.0001999307316708421,
+      "loss": 0.8787803649902344,
+      "step": 1294
+    },
+    {
+      "epoch": 0.5468354430379747,
+      "grad_norm": 0.43326008319854736,
+      "learning_rate": 0.00019992893278147436,
+      "loss": 0.8375519514083862,
+      "step": 1296
+    },
+    {
+      "epoch": 0.5476793248945148,
+      "grad_norm": 0.406556099653244,
+      "learning_rate": 0.00019992711084042096,
+      "loss": 0.8538051247596741,
+      "step": 1298
+    },
+    {
+      "epoch": 0.5485232067510548,
+      "grad_norm": 0.46090877056121826,
+      "learning_rate": 0.00019992526584810226,
+      "loss": 0.9722132086753845,
+      "step": 1300
+    },
+    {
+      "epoch": 0.5485232067510548,
+      "eval_loss": 0.9068717360496521,
+      "eval_runtime": 680.7718,
+      "eval_samples_per_second": 3.095,
+      "eval_steps_per_second": 3.095,
+      "step": 1300
+    },
+    {
+      "epoch": 0.549367088607595,
+      "grad_norm": 0.44932207465171814,
+      "learning_rate": 0.00019992339780494382,
+      "loss": 0.8474152684211731,
+      "step": 1302
+    },
+    {
+      "epoch": 0.550210970464135,
+      "grad_norm": 0.5185276865959167,
+      "learning_rate": 0.0001999215067113766,
+      "loss": 0.8440775871276855,
+      "step": 1304
+    },
+    {
+      "epoch": 0.5510548523206751,
+      "grad_norm": 0.43811365962028503,
+      "learning_rate": 0.0001999195925678368,
+      "loss": 0.8848010897636414,
+      "step": 1306
+    },
+    {
+      "epoch": 0.5518987341772152,
+      "grad_norm": 0.4861660301685333,
+      "learning_rate": 0.00019991765537476596,
+      "loss": 0.8265100121498108,
+      "step": 1308
+    },
+    {
+      "epoch": 0.5527426160337553,
+      "grad_norm": 0.5068721771240234,
+      "learning_rate": 0.000199915695132611,
+      "loss": 0.9327963590621948,
+      "step": 1310
+    },
+    {
+      "epoch": 0.5535864978902953,
+      "grad_norm": 0.4902805685997009,
+      "learning_rate": 0.0001999137118418241,
+      "loss": 0.9315155744552612,
+      "step": 1312
+    },
+    {
+      "epoch": 0.5544303797468354,
+      "grad_norm": 0.4530964493751526,
+      "learning_rate": 0.00019991170550286272,
+      "loss": 0.9344852566719055,
+      "step": 1314
+    },
+    {
+      "epoch": 0.5552742616033756,
+      "grad_norm": 0.4106673300266266,
+      "learning_rate": 0.00019990967611618974,
+      "loss": 0.8326173424720764,
+      "step": 1316
+    },
+    {
+      "epoch": 0.5561181434599156,
+      "grad_norm": 0.4439711570739746,
+      "learning_rate": 0.0001999076236822733,
+      "loss": 0.8934909105300903,
+      "step": 1318
+    },
+    {
+      "epoch": 0.5569620253164557,
+      "grad_norm": 0.40245553851127625,
+      "learning_rate": 0.0001999055482015868,
+      "loss": 0.6691107153892517,
+      "step": 1320
+    },
+    {
+      "epoch": 0.5578059071729958,
+      "grad_norm": 0.43904435634613037,
+      "learning_rate": 0.0001999034496746091,
+      "loss": 0.8560307621955872,
+      "step": 1322
+    },
+    {
+      "epoch": 0.5586497890295359,
+      "grad_norm": 0.5131705403327942,
+      "learning_rate": 0.00019990132810182422,
+      "loss": 0.9024442434310913,
+      "step": 1324
+    },
+    {
+      "epoch": 0.5594936708860759,
+      "grad_norm": 0.4539531171321869,
+      "learning_rate": 0.00019989918348372156,
+      "loss": 0.8675428628921509,
+      "step": 1326
+    },
+    {
+      "epoch": 0.560337552742616,
+      "grad_norm": 0.44469985365867615,
+      "learning_rate": 0.00019989701582079591,
+      "loss": 0.8853429555892944,
+      "step": 1328
+    },
+    {
+      "epoch": 0.5611814345991561,
+      "grad_norm": 0.4485546946525574,
+      "learning_rate": 0.00019989482511354725,
+      "loss": 0.9675378799438477,
+      "step": 1330
+    },
+    {
+      "epoch": 0.5620253164556962,
+      "grad_norm": 0.43316468596458435,
+      "learning_rate": 0.00019989261136248097,
+      "loss": 0.8256655335426331,
+      "step": 1332
+    },
+    {
+      "epoch": 0.5628691983122363,
+      "grad_norm": 0.4315780699253082,
+      "learning_rate": 0.00019989037456810772,
+      "loss": 0.7727690935134888,
+      "step": 1334
+    },
+    {
+      "epoch": 0.5637130801687764,
+      "grad_norm": 0.45147502422332764,
+      "learning_rate": 0.00019988811473094348,
+      "loss": 0.9240618348121643,
+      "step": 1336
+    },
+    {
+      "epoch": 0.5645569620253165,
+      "grad_norm": 0.4606908857822418,
+      "learning_rate": 0.00019988583185150957,
+      "loss": 0.8005949258804321,
+      "step": 1338
+    },
+    {
+      "epoch": 0.5654008438818565,
+      "grad_norm": 0.48932451009750366,
+      "learning_rate": 0.00019988352593033255,
+      "loss": 0.9042718410491943,
+      "step": 1340
+    },
+    {
+      "epoch": 0.5662447257383966,
+      "grad_norm": 0.46022218465805054,
+      "learning_rate": 0.00019988119696794443,
+      "loss": 0.8724613785743713,
+      "step": 1342
+    },
+    {
+      "epoch": 0.5670886075949367,
+      "grad_norm": 0.4414621889591217,
+      "learning_rate": 0.00019987884496488242,
+      "loss": 0.8653066158294678,
+      "step": 1344
+    },
+    {
+      "epoch": 0.5679324894514768,
+      "grad_norm": 0.4490342140197754,
+      "learning_rate": 0.0001998764699216891,
+      "loss": 0.8843849301338196,
+      "step": 1346
+    },
+    {
+      "epoch": 0.5687763713080168,
+      "grad_norm": 0.47690069675445557,
+      "learning_rate": 0.00019987407183891232,
+      "loss": 0.8385393023490906,
+      "step": 1348
+    },
+    {
+      "epoch": 0.569620253164557,
+      "grad_norm": 0.41409361362457275,
+      "learning_rate": 0.00019987165071710527,
+      "loss": 0.77492755651474,
+      "step": 1350
+    },
+    {
+      "epoch": 0.570464135021097,
+      "grad_norm": 0.4151647984981537,
+      "learning_rate": 0.0001998692065568265,
+      "loss": 0.8368680477142334,
+      "step": 1352
+    },
+    {
+      "epoch": 0.5713080168776371,
+      "grad_norm": 0.4105067849159241,
+      "learning_rate": 0.0001998667393586398,
+      "loss": 0.833220362663269,
+      "step": 1354
+    },
+    {
+      "epoch": 0.5721518987341773,
+      "grad_norm": 0.44701820611953735,
+      "learning_rate": 0.0001998642491231143,
+      "loss": 0.8041552305221558,
+      "step": 1356
+    },
+    {
+      "epoch": 0.5729957805907173,
+      "grad_norm": 0.4621582329273224,
+      "learning_rate": 0.00019986173585082444,
+      "loss": 0.9052709937095642,
+      "step": 1358
+    },
+    {
+      "epoch": 0.5738396624472574,
+      "grad_norm": 0.4934713840484619,
+      "learning_rate": 0.00019985919954235005,
+      "loss": 0.8036054372787476,
+      "step": 1360
+    },
+    {
+      "epoch": 0.5746835443037974,
+      "grad_norm": 0.4557384252548218,
+      "learning_rate": 0.0001998566401982761,
+      "loss": 0.8340095281600952,
+      "step": 1362
+    },
+    {
+      "epoch": 0.5755274261603376,
+      "grad_norm": 0.5048403739929199,
+      "learning_rate": 0.00019985405781919305,
+      "loss": 0.8703633546829224,
+      "step": 1364
+    },
+    {
+      "epoch": 0.5763713080168776,
+      "grad_norm": 0.4707714319229126,
+      "learning_rate": 0.0001998514524056966,
+      "loss": 0.9115830659866333,
+      "step": 1366
+    },
+    {
+      "epoch": 0.5772151898734177,
+      "grad_norm": 0.4502186179161072,
+      "learning_rate": 0.00019984882395838777,
+      "loss": 0.8603149056434631,
+      "step": 1368
+    },
+    {
+      "epoch": 0.5780590717299579,
+      "grad_norm": 0.4229136109352112,
+      "learning_rate": 0.00019984617247787288,
+      "loss": 0.7482197284698486,
+      "step": 1370
+    },
+    {
+      "epoch": 0.5789029535864979,
+      "grad_norm": 0.49208030104637146,
+      "learning_rate": 0.0001998434979647636,
+      "loss": 0.9071239233016968,
+      "step": 1372
+    },
+    {
+      "epoch": 0.579746835443038,
+      "grad_norm": 0.47393161058425903,
+      "learning_rate": 0.00019984080041967687,
+      "loss": 0.9370644092559814,
+      "step": 1374
+    },
+    {
+      "epoch": 0.580590717299578,
+      "grad_norm": 0.4483324885368347,
+      "learning_rate": 0.00019983807984323492,
+      "loss": 0.8536180257797241,
+      "step": 1376
+    },
+    {
+      "epoch": 0.5814345991561182,
+      "grad_norm": 0.4163796007633209,
+      "learning_rate": 0.00019983533623606543,
+      "loss": 0.7360405325889587,
+      "step": 1378
+    },
+    {
+      "epoch": 0.5822784810126582,
+      "grad_norm": 0.44147446751594543,
+      "learning_rate": 0.00019983256959880122,
+      "loss": 0.8162824511528015,
+      "step": 1380
+    },
+    {
+      "epoch": 0.5831223628691983,
+      "grad_norm": 0.44443491101264954,
+      "learning_rate": 0.0001998297799320805,
+      "loss": 0.9169327616691589,
+      "step": 1382
+    },
+    {
+      "epoch": 0.5839662447257384,
+      "grad_norm": 0.4641949236392975,
+      "learning_rate": 0.00019982696723654686,
+      "loss": 0.8875693678855896,
+      "step": 1384
+    },
+    {
+      "epoch": 0.5848101265822785,
+      "grad_norm": 0.4327974319458008,
+      "learning_rate": 0.00019982413151284906,
+      "loss": 0.7747344970703125,
+      "step": 1386
+    },
+    {
+      "epoch": 0.5856540084388185,
+      "grad_norm": 0.44200772047042847,
+      "learning_rate": 0.0001998212727616413,
+      "loss": 0.8490481972694397,
+      "step": 1388
+    },
+    {
+      "epoch": 0.5864978902953587,
+      "grad_norm": 0.39084959030151367,
+      "learning_rate": 0.000199818390983583,
+      "loss": 0.8014808297157288,
+      "step": 1390
+    },
+    {
+      "epoch": 0.5873417721518988,
+      "grad_norm": 0.3691277801990509,
+      "learning_rate": 0.00019981548617933897,
+      "loss": 0.8650928139686584,
+      "step": 1392
+    },
+    {
+      "epoch": 0.5881856540084388,
+      "grad_norm": 0.480002760887146,
+      "learning_rate": 0.00019981255834957925,
+      "loss": 0.7713267803192139,
+      "step": 1394
+    },
+    {
+      "epoch": 0.5890295358649789,
+      "grad_norm": 0.4397226572036743,
+      "learning_rate": 0.00019980960749497927,
+      "loss": 0.9053656458854675,
+      "step": 1396
+    },
+    {
+      "epoch": 0.589873417721519,
+      "grad_norm": 0.4263727366924286,
+      "learning_rate": 0.00019980663361621973,
+      "loss": 0.8737669587135315,
+      "step": 1398
+    },
+    {
+      "epoch": 0.5907172995780591,
+      "grad_norm": 0.4468817710876465,
+      "learning_rate": 0.0001998036367139866,
+      "loss": 0.9396650791168213,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5907172995780591,
+      "eval_loss": 0.8971880674362183,
+      "eval_runtime": 692.8046,
+      "eval_samples_per_second": 3.041,
+      "eval_steps_per_second": 3.041,
+      "step": 1400
+    },
+    {
+      "epoch": 0.5915611814345991,
+      "grad_norm": 0.4128144383430481,
+      "learning_rate": 0.00019980061678897127,
+      "loss": 0.926134467124939,
+      "step": 1402
+    },
+    {
+      "epoch": 0.5924050632911393,
+      "grad_norm": 0.45524629950523376,
+      "learning_rate": 0.00019979757384187035,
+      "loss": 0.826360285282135,
+      "step": 1404
+    },
+    {
+      "epoch": 0.5932489451476793,
+      "grad_norm": 0.42521336674690247,
+      "learning_rate": 0.0001997945078733858,
+      "loss": 0.8147702813148499,
+      "step": 1406
+    },
+    {
+      "epoch": 0.5940928270042194,
+      "grad_norm": 0.4172525107860565,
+      "learning_rate": 0.00019979141888422489,
+      "loss": 0.7711596488952637,
+      "step": 1408
+    },
+    {
+      "epoch": 0.5949367088607594,
+      "grad_norm": 0.46193623542785645,
+      "learning_rate": 0.00019978830687510013,
+      "loss": 0.7875980138778687,
+      "step": 1410
+    },
+    {
+      "epoch": 0.5957805907172996,
+      "grad_norm": 0.4455825686454773,
+      "learning_rate": 0.00019978517184672946,
+      "loss": 0.8590195178985596,
+      "step": 1412
+    },
+    {
+      "epoch": 0.5966244725738397,
+      "grad_norm": 0.4266505539417267,
+      "learning_rate": 0.00019978201379983608,
+      "loss": 0.9318227767944336,
+      "step": 1414
+    },
+    {
+      "epoch": 0.5974683544303797,
+      "grad_norm": 0.42627251148223877,
+      "learning_rate": 0.00019977883273514843,
+      "loss": 0.8644474744796753,
+      "step": 1416
+    },
+    {
+      "epoch": 0.5983122362869199,
+      "grad_norm": 0.393873393535614,
+      "learning_rate": 0.00019977562865340038,
+      "loss": 0.7760446071624756,
+      "step": 1418
+    },
+    {
+      "epoch": 0.5991561181434599,
+      "grad_norm": 0.44334208965301514,
+      "learning_rate": 0.000199772401555331,
+      "loss": 0.8071584701538086,
+      "step": 1420
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.39390063285827637,
+      "learning_rate": 0.0001997691514416848,
+      "loss": 0.9076889753341675,
+      "step": 1422
+    },
+    {
+      "epoch": 0.60084388185654,
+      "grad_norm": 0.46753421425819397,
+      "learning_rate": 0.00019976587831321144,
+      "loss": 0.8534318804740906,
+      "step": 1424
+    },
+    {
+      "epoch": 0.6016877637130802,
+      "grad_norm": 0.4840681254863739,
+      "learning_rate": 0.00019976258217066598,
+      "loss": 0.9049596190452576,
+      "step": 1426
+    },
+    {
+      "epoch": 0.6025316455696202,
+      "grad_norm": 0.4740557372570038,
+      "learning_rate": 0.00019975926301480878,
+      "loss": 0.8520918488502502,
+      "step": 1428
+    },
+    {
+      "epoch": 0.6033755274261603,
+      "grad_norm": 0.46498584747314453,
+      "learning_rate": 0.00019975592084640553,
+      "loss": 0.8221305012702942,
+      "step": 1430
+    },
+    {
+      "epoch": 0.6042194092827005,
+      "grad_norm": 0.40885692834854126,
+      "learning_rate": 0.0001997525556662272,
+      "loss": 0.8132198452949524,
+      "step": 1432
+    },
+    {
+      "epoch": 0.6050632911392405,
+      "grad_norm": 0.4375821053981781,
+      "learning_rate": 0.00019974916747505006,
+      "loss": 0.870534360408783,
+      "step": 1434
+    },
+    {
+      "epoch": 0.6059071729957806,
+      "grad_norm": 0.4780935049057007,
+      "learning_rate": 0.00019974575627365569,
+      "loss": 0.8825759291648865,
+      "step": 1436
+    },
+    {
+      "epoch": 0.6067510548523207,
+      "grad_norm": 0.4623970687389374,
+      "learning_rate": 0.00019974232206283098,
+      "loss": 0.8187641501426697,
+      "step": 1438
+    },
+    {
+      "epoch": 0.6075949367088608,
+      "grad_norm": 0.42172595858573914,
+      "learning_rate": 0.0001997388648433682,
+      "loss": 0.8569821119308472,
+      "step": 1440
+    },
+    {
+      "epoch": 0.6084388185654008,
+      "grad_norm": 0.49386659264564514,
+      "learning_rate": 0.0001997353846160648,
+      "loss": 0.9173614382743835,
+      "step": 1442
+    },
+    {
+      "epoch": 0.6092827004219409,
+      "grad_norm": 0.47122615575790405,
+      "learning_rate": 0.00019973188138172363,
+      "loss": 0.9444975256919861,
+      "step": 1444
+    },
+    {
+      "epoch": 0.610126582278481,
+      "grad_norm": 0.4831530451774597,
+      "learning_rate": 0.0001997283551411528,
+      "loss": 0.9537245631217957,
+      "step": 1446
+    },
+    {
+      "epoch": 0.6109704641350211,
+      "grad_norm": 0.44460317492485046,
+      "learning_rate": 0.0001997248058951658,
+      "loss": 0.8481367230415344,
+      "step": 1448
+    },
+    {
+      "epoch": 0.6118143459915611,
+      "grad_norm": 0.4315263628959656,
+      "learning_rate": 0.00019972123364458128,
+      "loss": 0.8572371006011963,
+      "step": 1450
+    },
+    {
+      "epoch": 0.6126582278481013,
+      "grad_norm": 0.38559582829475403,
+      "learning_rate": 0.00019971763839022336,
+      "loss": 0.762590765953064,
+      "step": 1452
+    },
+    {
+      "epoch": 0.6135021097046414,
+      "grad_norm": 0.3820290267467499,
+      "learning_rate": 0.00019971402013292138,
+      "loss": 0.7879500389099121,
+      "step": 1454
+    },
+    {
+      "epoch": 0.6143459915611814,
+      "grad_norm": 0.42509570717811584,
+      "learning_rate": 0.00019971037887351,
+      "loss": 0.8696863055229187,
+      "step": 1456
+    },
+    {
+      "epoch": 0.6151898734177215,
+      "grad_norm": 0.46750548481941223,
+      "learning_rate": 0.00019970671461282916,
+      "loss": 0.8425862789154053,
+      "step": 1458
+    },
+    {
+      "epoch": 0.6160337552742616,
+      "grad_norm": 0.3937121033668518,
+      "learning_rate": 0.0001997030273517242,
+      "loss": 0.8028931021690369,
+      "step": 1460
+    },
+    {
+      "epoch": 0.6168776371308017,
+      "grad_norm": 0.4361220896244049,
+      "learning_rate": 0.00019969931709104565,
+      "loss": 0.8662509918212891,
+      "step": 1462
+    },
+    {
+      "epoch": 0.6177215189873417,
+      "grad_norm": 0.4737963378429413,
+      "learning_rate": 0.00019969558383164943,
+      "loss": 0.8962596654891968,
+      "step": 1464
+    },
+    {
+      "epoch": 0.6185654008438819,
+      "grad_norm": 0.3952867388725281,
+      "learning_rate": 0.00019969182757439668,
+      "loss": 0.8630867600440979,
+      "step": 1466
+    },
+    {
+      "epoch": 0.619409282700422,
+      "grad_norm": 0.41688743233680725,
+      "learning_rate": 0.00019968804832015393,
+      "loss": 0.8759240508079529,
+      "step": 1468
+    },
+    {
+      "epoch": 0.620253164556962,
+      "grad_norm": 0.4599224627017975,
+      "learning_rate": 0.00019968424606979298,
+      "loss": 0.8459385633468628,
+      "step": 1470
+    },
+    {
+      "epoch": 0.6210970464135022,
+      "grad_norm": 0.5476765632629395,
+      "learning_rate": 0.00019968042082419094,
+      "loss": 0.8844659328460693,
+      "step": 1472
+    },
+    {
+      "epoch": 0.6219409282700422,
+      "grad_norm": 0.46202269196510315,
+      "learning_rate": 0.00019967657258423022,
+      "loss": 0.860946536064148,
+      "step": 1474
+    },
+    {
+      "epoch": 0.6227848101265823,
+      "grad_norm": 0.39446666836738586,
+      "learning_rate": 0.00019967270135079853,
+      "loss": 0.7587860822677612,
+      "step": 1476
+    },
+    {
+      "epoch": 0.6236286919831223,
+      "grad_norm": 0.4500375986099243,
+      "learning_rate": 0.0001996688071247889,
+      "loss": 0.9143000245094299,
+      "step": 1478
+    },
+    {
+      "epoch": 0.6244725738396625,
+      "grad_norm": 0.4203545153141022,
+      "learning_rate": 0.0001996648899070996,
+      "loss": 0.7865519523620605,
+      "step": 1480
+    },
+    {
+      "epoch": 0.6253164556962025,
+      "grad_norm": 0.45623889565467834,
+      "learning_rate": 0.00019966094969863432,
+      "loss": 0.9192912578582764,
+      "step": 1482
+    },
+    {
+      "epoch": 0.6261603375527426,
+      "grad_norm": 0.4139637053012848,
+      "learning_rate": 0.00019965698650030195,
+      "loss": 0.8459161520004272,
+      "step": 1484
+    },
+    {
+      "epoch": 0.6270042194092827,
+      "grad_norm": 0.4410824179649353,
+      "learning_rate": 0.00019965300031301678,
+      "loss": 0.8159077763557434,
+      "step": 1486
+    },
+    {
+      "epoch": 0.6278481012658228,
+      "grad_norm": 0.43821197748184204,
+      "learning_rate": 0.00019964899113769828,
+      "loss": 0.884467363357544,
+      "step": 1488
+    },
+    {
+      "epoch": 0.6286919831223629,
+      "grad_norm": 0.4692091643810272,
+      "learning_rate": 0.00019964495897527133,
+      "loss": 0.8693601489067078,
+      "step": 1490
+    },
+    {
+      "epoch": 0.6295358649789029,
+      "grad_norm": 0.40715447068214417,
+      "learning_rate": 0.00019964090382666608,
+      "loss": 0.8897743225097656,
+      "step": 1492
+    },
+    {
+      "epoch": 0.6303797468354431,
+      "grad_norm": 0.4123334586620331,
+      "learning_rate": 0.0001996368256928179,
+      "loss": 0.8274733424186707,
+      "step": 1494
+    },
+    {
+      "epoch": 0.6312236286919831,
+      "grad_norm": 0.44347211718559265,
+      "learning_rate": 0.00019963272457466767,
+      "loss": 0.8158749938011169,
+      "step": 1496
+    },
+    {
+      "epoch": 0.6320675105485232,
+      "grad_norm": 0.4700329601764679,
+      "learning_rate": 0.00019962860047316135,
+      "loss": 0.8439569473266602,
+      "step": 1498
+    },
+    {
+      "epoch": 0.6329113924050633,
+      "grad_norm": 0.46415844559669495,
+      "learning_rate": 0.00019962445338925027,
+      "loss": 0.8316822648048401,
+      "step": 1500
+    },
+    {
+      "epoch": 0.6329113924050633,
+      "eval_loss": 0.887488842010498,
+      "eval_runtime": 686.2804,
+      "eval_samples_per_second": 3.07,
+      "eval_steps_per_second": 3.07,
+      "step": 1500
+    },
+    {
+      "epoch": 0.6337552742616034,
+      "grad_norm": 0.44183167815208435,
+      "learning_rate": 0.0001996202833238911,
+      "loss": 0.9121994972229004,
+      "step": 1502
+    },
+    {
+      "epoch": 0.6345991561181434,
+      "grad_norm": 0.4279208779335022,
+      "learning_rate": 0.00019961609027804587,
+      "loss": 0.9160211086273193,
+      "step": 1504
+    },
+    {
+      "epoch": 0.6354430379746835,
+      "grad_norm": 0.3967169523239136,
+      "learning_rate": 0.00019961187425268176,
+      "loss": 0.861677348613739,
+      "step": 1506
+    },
+    {
+      "epoch": 0.6362869198312237,
+      "grad_norm": 0.4796451926231384,
+      "learning_rate": 0.0001996076352487713,
+      "loss": 0.8478423357009888,
+      "step": 1508
+    },
+    {
+      "epoch": 0.6371308016877637,
+      "grad_norm": 0.3861426115036011,
+      "learning_rate": 0.00019960337326729245,
+      "loss": 0.8739159107208252,
+      "step": 1510
+    },
+    {
+      "epoch": 0.6379746835443038,
+      "grad_norm": 0.4567820429801941,
+      "learning_rate": 0.00019959908830922824,
+      "loss": 0.7693920731544495,
+      "step": 1512
+    },
+    {
+      "epoch": 0.6388185654008439,
+      "grad_norm": 0.4457108676433563,
+      "learning_rate": 0.00019959478037556724,
+      "loss": 0.8482301831245422,
+      "step": 1514
+    },
+    {
+      "epoch": 0.639662447257384,
+      "grad_norm": 0.43918928503990173,
+      "learning_rate": 0.00019959044946730314,
+      "loss": 0.8626812696456909,
+      "step": 1516
+    },
+    {
+      "epoch": 0.640506329113924,
+      "grad_norm": 0.4095900058746338,
+      "learning_rate": 0.00019958609558543504,
+      "loss": 0.8342230916023254,
+      "step": 1518
+    },
+    {
+      "epoch": 0.6413502109704642,
+      "grad_norm": 0.4181270897388458,
+      "learning_rate": 0.00019958171873096724,
+      "loss": 0.8794118165969849,
+      "step": 1520
+    },
+    {
+      "epoch": 0.6421940928270042,
+      "grad_norm": 0.4750959277153015,
+      "learning_rate": 0.00019957731890490947,
+      "loss": 0.7787677049636841,
+      "step": 1522
+    },
+    {
+      "epoch": 0.6430379746835443,
+      "grad_norm": 0.5129296779632568,
+      "learning_rate": 0.00019957289610827663,
+      "loss": 0.9084368348121643,
+      "step": 1524
+    },
+    {
+      "epoch": 0.6438818565400843,
+      "grad_norm": 0.4829029142856598,
+      "learning_rate": 0.000199568450342089,
+      "loss": 0.9066952466964722,
+      "step": 1526
+    },
+    {
+      "epoch": 0.6447257383966245,
+      "grad_norm": 0.4225057363510132,
+      "learning_rate": 0.00019956398160737214,
+      "loss": 0.8127874732017517,
+      "step": 1528
+    },
+    {
+      "epoch": 0.6455696202531646,
+      "grad_norm": 0.4231826364994049,
+      "learning_rate": 0.00019955948990515684,
+      "loss": 0.8781921863555908,
+      "step": 1530
+    },
+    {
+      "epoch": 0.6464135021097046,
+      "grad_norm": 0.4436216950416565,
+      "learning_rate": 0.00019955497523647933,
+      "loss": 0.8136996626853943,
+      "step": 1532
+    },
+    {
+      "epoch": 0.6472573839662448,
+      "grad_norm": 0.4018244743347168,
+      "learning_rate": 0.00019955043760238098,
+      "loss": 0.8003877401351929,
+      "step": 1534
+    },
+    {
+      "epoch": 0.6481012658227848,
+      "grad_norm": 0.40024611353874207,
+      "learning_rate": 0.0001995458770039086,
+      "loss": 0.8996267914772034,
+      "step": 1536
+    },
+    {
+      "epoch": 0.6489451476793249,
+      "grad_norm": 0.42845702171325684,
+      "learning_rate": 0.00019954129344211424,
+      "loss": 0.8610522150993347,
+      "step": 1538
+    },
+    {
+      "epoch": 0.6497890295358649,
+      "grad_norm": 0.4258182644844055,
+      "learning_rate": 0.00019953668691805517,
+      "loss": 0.8991837501525879,
+      "step": 1540
+    },
+    {
+      "epoch": 0.6506329113924051,
+      "grad_norm": 0.4483257830142975,
+      "learning_rate": 0.0001995320574327941,
+      "loss": 0.8150189518928528,
+      "step": 1542
+    },
+    {
+      "epoch": 0.6514767932489451,
+      "grad_norm": 0.4320627748966217,
+      "learning_rate": 0.0001995274049873989,
+      "loss": 0.8572840690612793,
+      "step": 1544
+    },
+    {
+      "epoch": 0.6523206751054852,
+      "grad_norm": 0.39950594305992126,
+      "learning_rate": 0.00019952272958294288,
+      "loss": 0.7766129374504089,
+      "step": 1546
+    },
+    {
+      "epoch": 0.6531645569620254,
+      "grad_norm": 0.4130117893218994,
+      "learning_rate": 0.00019951803122050446,
+      "loss": 0.9032199382781982,
+      "step": 1548
+    },
+    {
+      "epoch": 0.6540084388185654,
+      "grad_norm": 0.4436919391155243,
+      "learning_rate": 0.00019951330990116754,
+      "loss": 0.8361048102378845,
+      "step": 1550
+    },
+    {
+      "epoch": 0.6548523206751055,
+      "grad_norm": 0.4229227900505066,
+      "learning_rate": 0.00019950856562602121,
+      "loss": 0.8661768436431885,
+      "step": 1552
+    },
+    {
+      "epoch": 0.6556962025316456,
+      "grad_norm": 0.4839727580547333,
+      "learning_rate": 0.00019950379839615988,
+      "loss": 0.9737826585769653,
+      "step": 1554
+    },
+    {
+      "epoch": 0.6565400843881857,
+      "grad_norm": 0.3859386742115021,
+      "learning_rate": 0.00019949900821268328,
+      "loss": 0.7542453408241272,
+      "step": 1556
+    },
+    {
+      "epoch": 0.6573839662447257,
+      "grad_norm": 0.39643988013267517,
+      "learning_rate": 0.0001994941950766964,
+      "loss": 0.9226290583610535,
+      "step": 1558
+    },
+    {
+      "epoch": 0.6582278481012658,
+      "grad_norm": 0.4261355400085449,
+      "learning_rate": 0.00019948935898930954,
+      "loss": 0.8375602960586548,
+      "step": 1560
+    },
+    {
+      "epoch": 0.6590717299578059,
+      "grad_norm": 0.445931077003479,
+      "learning_rate": 0.00019948449995163826,
+      "loss": 0.7847310304641724,
+      "step": 1562
+    },
+    {
+      "epoch": 0.659915611814346,
+      "grad_norm": 0.4680459797382355,
+      "learning_rate": 0.00019947961796480353,
+      "loss": 0.851150631904602,
+      "step": 1564
+    },
+    {
+      "epoch": 0.660759493670886,
+      "grad_norm": 0.6826074719429016,
+      "learning_rate": 0.00019947471302993143,
+      "loss": 0.8259562253952026,
+      "step": 1566
+    },
+    {
+      "epoch": 0.6616033755274262,
+      "grad_norm": 0.47369205951690674,
+      "learning_rate": 0.00019946978514815352,
+      "loss": 0.879021406173706,
+      "step": 1568
+    },
+    {
+      "epoch": 0.6624472573839663,
+      "grad_norm": 0.41111937165260315,
+      "learning_rate": 0.0001994648343206065,
+      "loss": 0.9296056032180786,
+      "step": 1570
+    },
+    {
+      "epoch": 0.6632911392405063,
+      "grad_norm": 0.4297783076763153,
+      "learning_rate": 0.00019945986054843248,
+      "loss": 0.873822808265686,
+      "step": 1572
+    },
+    {
+      "epoch": 0.6641350210970464,
+      "grad_norm": 0.48124316334724426,
+      "learning_rate": 0.00019945486383277878,
+      "loss": 0.937151312828064,
+      "step": 1574
+    },
+    {
+      "epoch": 0.6649789029535865,
+      "grad_norm": 0.4048190116882324,
+      "learning_rate": 0.00019944984417479805,
+      "loss": 0.8398929834365845,
+      "step": 1576
+    },
+    {
+      "epoch": 0.6658227848101266,
+      "grad_norm": 0.4023774266242981,
+      "learning_rate": 0.00019944480157564826,
+      "loss": 0.8269520998001099,
+      "step": 1578
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.3784184157848358,
+      "learning_rate": 0.00019943973603649262,
+      "loss": 0.8237608671188354,
+      "step": 1580
+    },
+    {
+      "epoch": 0.6675105485232068,
+      "grad_norm": 0.4448573887348175,
+      "learning_rate": 0.00019943464755849965,
+      "loss": 0.8469099998474121,
+      "step": 1582
+    },
+    {
+      "epoch": 0.6683544303797468,
+      "grad_norm": 0.4690774083137512,
+      "learning_rate": 0.00019942953614284317,
+      "loss": 0.8630822896957397,
+      "step": 1584
+    },
+    {
+      "epoch": 0.6691983122362869,
+      "grad_norm": 0.43670853972435,
+      "learning_rate": 0.00019942440179070227,
+      "loss": 0.8904082179069519,
+      "step": 1586
+    },
+    {
+      "epoch": 0.6700421940928271,
+      "grad_norm": 0.4236849844455719,
+      "learning_rate": 0.00019941924450326138,
+      "loss": 0.8355640172958374,
+      "step": 1588
+    },
+    {
+      "epoch": 0.6708860759493671,
+      "grad_norm": 0.48624539375305176,
+      "learning_rate": 0.00019941406428171013,
+      "loss": 0.8898435831069946,
+      "step": 1590
+    },
+    {
+      "epoch": 0.6717299578059072,
+      "grad_norm": 0.4453780949115753,
+      "learning_rate": 0.00019940886112724355,
+      "loss": 0.9000012278556824,
+      "step": 1592
+    },
+    {
+      "epoch": 0.6725738396624472,
+      "grad_norm": 0.4417199194431305,
+      "learning_rate": 0.00019940363504106193,
+      "loss": 0.8171058893203735,
+      "step": 1594
+    },
+    {
+      "epoch": 0.6734177215189874,
+      "grad_norm": 0.45055362582206726,
+      "learning_rate": 0.00019939838602437074,
+      "loss": 0.8703798055648804,
+      "step": 1596
+    },
+    {
+      "epoch": 0.6742616033755274,
+      "grad_norm": 0.42316532135009766,
+      "learning_rate": 0.00019939311407838088,
+      "loss": 0.8677281737327576,
+      "step": 1598
+    },
+    {
+      "epoch": 0.6751054852320675,
+      "grad_norm": 0.3935781717300415,
+      "learning_rate": 0.0001993878192043085,
+      "loss": 0.7314013242721558,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6751054852320675,
+      "eval_loss": 0.8769772052764893,
+      "eval_runtime": 677.9338,
+      "eval_samples_per_second": 3.108,
+      "eval_steps_per_second": 3.108,
+      "step": 1600
+    },
+    {
+      "epoch": 0.6759493670886076,
+      "grad_norm": 0.4197012186050415,
+      "learning_rate": 0.000199382501403375,
+      "loss": 0.8075380325317383,
+      "step": 1602
+    },
+    {
+      "epoch": 0.6767932489451477,
+      "grad_norm": 0.45227327942848206,
+      "learning_rate": 0.00019937716067680713,
+      "loss": 0.7885794639587402,
+      "step": 1604
+    },
+    {
+      "epoch": 0.6776371308016877,
+      "grad_norm": 0.45660385489463806,
+      "learning_rate": 0.00019937179702583682,
+      "loss": 0.868261992931366,
+      "step": 1606
+    },
+    {
+      "epoch": 0.6784810126582278,
+      "grad_norm": 0.3955288231372833,
+      "learning_rate": 0.00019936641045170144,
+      "loss": 0.8306655287742615,
+      "step": 1608
+    },
+    {
+      "epoch": 0.679324894514768,
+      "grad_norm": 0.3651324510574341,
+      "learning_rate": 0.00019936100095564353,
+      "loss": 0.7505315542221069,
+      "step": 1610
+    },
+    {
+      "epoch": 0.680168776371308,
+      "grad_norm": 0.4240955412387848,
+      "learning_rate": 0.00019935556853891096,
+      "loss": 0.8792756795883179,
+      "step": 1612
+    },
+    {
+      "epoch": 0.6810126582278481,
+      "grad_norm": 0.4142507314682007,
+      "learning_rate": 0.00019935011320275687,
+      "loss": 0.7328272461891174,
+      "step": 1614
+    },
+    {
+      "epoch": 0.6818565400843882,
+      "grad_norm": 0.48339322209358215,
+      "learning_rate": 0.00019934463494843975,
+      "loss": 0.8414849042892456,
+      "step": 1616
+    },
+    {
+      "epoch": 0.6827004219409283,
+      "grad_norm": 0.4312443435192108,
+      "learning_rate": 0.00019933913377722328,
+      "loss": 0.8588716983795166,
+      "step": 1618
+    },
+    {
+      "epoch": 0.6835443037974683,
+      "grad_norm": 0.43849512934684753,
+      "learning_rate": 0.0001993336096903765,
+      "loss": 0.846511960029602,
+      "step": 1620
+    },
+    {
+      "epoch": 0.6843881856540084,
+      "grad_norm": 0.45767566561698914,
+      "learning_rate": 0.0001993280626891737,
+      "loss": 0.9352323412895203,
+      "step": 1622
+    },
+    {
+      "epoch": 0.6852320675105485,
+      "grad_norm": 0.438412606716156,
+      "learning_rate": 0.00019932249277489444,
+      "loss": 0.8884757161140442,
+      "step": 1624
+    },
+    {
+      "epoch": 0.6860759493670886,
+      "grad_norm": 0.45893922448158264,
+      "learning_rate": 0.0001993168999488237,
+      "loss": 0.7739649415016174,
+      "step": 1626
+    },
+    {
+      "epoch": 0.6869198312236287,
+      "grad_norm": 0.38631772994995117,
+      "learning_rate": 0.0001993112842122515,
+      "loss": 0.7948258519172668,
+      "step": 1628
+    },
+    {
+      "epoch": 0.6877637130801688,
+      "grad_norm": 0.3824027180671692,
+      "learning_rate": 0.00019930564556647334,
+      "loss": 0.750511884689331,
+      "step": 1630
+    },
+    {
+      "epoch": 0.6886075949367089,
+      "grad_norm": 0.3980776071548462,
+      "learning_rate": 0.00019929998401278996,
+      "loss": 0.9006738662719727,
+      "step": 1632
+    },
+    {
+      "epoch": 0.6894514767932489,
+      "grad_norm": 0.41419896483421326,
+      "learning_rate": 0.00019929429955250734,
+      "loss": 0.6996869444847107,
+      "step": 1634
+    },
+    {
+      "epoch": 0.6902953586497891,
+      "grad_norm": 0.43783676624298096,
+      "learning_rate": 0.00019928859218693682,
+      "loss": 0.8523393273353577,
+      "step": 1636
+    },
+    {
+      "epoch": 0.6911392405063291,
+      "grad_norm": 0.4378126561641693,
+      "learning_rate": 0.00019928286191739497,
+      "loss": 0.8885987401008606,
+      "step": 1638
+    },
+    {
+      "epoch": 0.6919831223628692,
+      "grad_norm": 0.4276133179664612,
+      "learning_rate": 0.00019927710874520363,
+      "loss": 0.8563809990882874,
+      "step": 1640
+    },
+    {
+      "epoch": 0.6928270042194092,
+      "grad_norm": 0.4600282311439514,
+      "learning_rate": 0.00019927133267168992,
+      "loss": 0.854114294052124,
+      "step": 1642
+    },
+    {
+      "epoch": 0.6936708860759494,
+      "grad_norm": 0.5058461427688599,
+      "learning_rate": 0.00019926553369818635,
+      "loss": 0.8557889461517334,
+      "step": 1644
+    },
+    {
+      "epoch": 0.6945147679324895,
+      "grad_norm": 0.4947412312030792,
+      "learning_rate": 0.00019925971182603057,
+      "loss": 0.8622767925262451,
+      "step": 1646
+    },
+    {
+      "epoch": 0.6953586497890295,
+      "grad_norm": 0.42555496096611023,
+      "learning_rate": 0.00019925386705656558,
+      "loss": 0.7836558818817139,
+      "step": 1648
+    },
+    {
+      "epoch": 0.6962025316455697,
+      "grad_norm": 0.47560831904411316,
+      "learning_rate": 0.00019924799939113967,
+      "loss": 0.7877846956253052,
+      "step": 1650
+    },
+    {
+      "epoch": 0.6970464135021097,
+      "grad_norm": 0.391702264547348,
+      "learning_rate": 0.0001992421088311064,
+      "loss": 0.775681734085083,
+      "step": 1652
+    },
+    {
+      "epoch": 0.6978902953586498,
+      "grad_norm": 0.463679701089859,
+      "learning_rate": 0.0001992361953778246,
+      "loss": 0.8809882402420044,
+      "step": 1654
+    },
+    {
+      "epoch": 0.6987341772151898,
+      "grad_norm": 0.44096484780311584,
+      "learning_rate": 0.00019923025903265836,
+      "loss": 0.8513299226760864,
+      "step": 1656
+    },
+    {
+      "epoch": 0.69957805907173,
+      "grad_norm": 0.47818854451179504,
+      "learning_rate": 0.0001992242997969772,
+      "loss": 0.8309667706489563,
+      "step": 1658
+    },
+    {
+      "epoch": 0.70042194092827,
+      "grad_norm": 0.4192790985107422,
+      "learning_rate": 0.00019921831767215565,
+      "loss": 0.8817558884620667,
+      "step": 1660
+    },
+    {
+      "epoch": 0.7012658227848101,
+      "grad_norm": 0.45285239815711975,
+      "learning_rate": 0.00019921231265957373,
+      "loss": 0.8315755724906921,
+      "step": 1662
+    },
+    {
+      "epoch": 0.7021097046413503,
+      "grad_norm": 0.47118625044822693,
+      "learning_rate": 0.00019920628476061673,
+      "loss": 0.8698058724403381,
+      "step": 1664
+    },
+    {
+      "epoch": 0.7029535864978903,
+      "grad_norm": 0.4769238829612732,
+      "learning_rate": 0.0001992002339766751,
+      "loss": 0.953620970249176,
+      "step": 1666
+    },
+    {
+      "epoch": 0.7037974683544304,
+      "grad_norm": 0.39397355914115906,
+      "learning_rate": 0.00019919416030914468,
+      "loss": 0.8536615371704102,
+      "step": 1668
+    },
+    {
+      "epoch": 0.7046413502109705,
+      "grad_norm": 0.3659648299217224,
+      "learning_rate": 0.00019918806375942655,
+      "loss": 0.8159828782081604,
+      "step": 1670
+    },
+    {
+      "epoch": 0.7054852320675106,
+      "grad_norm": 0.4381789565086365,
+      "learning_rate": 0.00019918194432892703,
+      "loss": 0.782463550567627,
+      "step": 1672
+    },
+    {
+      "epoch": 0.7063291139240506,
+      "grad_norm": 0.44745752215385437,
+      "learning_rate": 0.00019917580201905778,
+      "loss": 0.9314272999763489,
+      "step": 1674
+    },
+    {
+      "epoch": 0.7071729957805907,
+      "grad_norm": 0.4515164792537689,
+      "learning_rate": 0.00019916963683123574,
+      "loss": 0.860816240310669,
+      "step": 1676
+    },
+    {
+      "epoch": 0.7080168776371308,
+      "grad_norm": 0.42735886573791504,
+      "learning_rate": 0.00019916344876688303,
+      "loss": 0.8008456826210022,
+      "step": 1678
+    },
+    {
+      "epoch": 0.7088607594936709,
+      "grad_norm": 0.418275386095047,
+      "learning_rate": 0.00019915723782742718,
+      "loss": 0.7796595096588135,
+      "step": 1680
+    },
+    {
+      "epoch": 0.7097046413502109,
+      "grad_norm": 0.44347989559173584,
+      "learning_rate": 0.00019915100401430089,
+      "loss": 0.9006506204605103,
+      "step": 1682
+    },
+    {
+      "epoch": 0.7105485232067511,
+      "grad_norm": 0.3932702839374542,
+      "learning_rate": 0.00019914474732894225,
+      "loss": 0.7750146985054016,
+      "step": 1684
+    },
+    {
+      "epoch": 0.7113924050632912,
+      "grad_norm": 0.4810502231121063,
+      "learning_rate": 0.00019913846777279447,
+      "loss": 0.7508097290992737,
+      "step": 1686
+    },
+    {
+      "epoch": 0.7122362869198312,
+      "grad_norm": 0.4166778028011322,
+      "learning_rate": 0.00019913216534730622,
+      "loss": 0.7926796078681946,
+      "step": 1688
+    },
+    {
+      "epoch": 0.7130801687763713,
+      "grad_norm": 0.47900474071502686,
+      "learning_rate": 0.00019912584005393123,
+      "loss": 0.8363928198814392,
+      "step": 1690
+    },
+    {
+      "epoch": 0.7139240506329114,
+      "grad_norm": 0.435143381357193,
+      "learning_rate": 0.00019911949189412872,
+      "loss": 0.8649623990058899,
+      "step": 1692
+    },
+    {
+      "epoch": 0.7147679324894515,
+      "grad_norm": 0.4070208966732025,
+      "learning_rate": 0.00019911312086936305,
+      "loss": 0.8196499943733215,
+      "step": 1694
+    },
+    {
+      "epoch": 0.7156118143459915,
+      "grad_norm": 0.4717724621295929,
+      "learning_rate": 0.0001991067269811039,
+      "loss": 0.9157413840293884,
+      "step": 1696
+    },
+    {
+      "epoch": 0.7164556962025317,
+      "grad_norm": 0.3942880928516388,
+      "learning_rate": 0.0001991003102308262,
+      "loss": 0.7880456447601318,
+      "step": 1698
+    },
+    {
+      "epoch": 0.7172995780590717,
+      "grad_norm": 0.4037776589393616,
+      "learning_rate": 0.0001990938706200102,
+      "loss": 0.8088646531105042,
+      "step": 1700
+    },
+    {
+      "epoch": 0.7172995780590717,
+      "eval_loss": 0.8708170056343079,
+      "eval_runtime": 670.3019,
+      "eval_samples_per_second": 3.143,
+      "eval_steps_per_second": 3.143,
+      "step": 1700
+    },
+    {
+      "epoch": 0.7181434599156118,
+      "grad_norm": 0.4443536400794983,
+      "learning_rate": 0.0001990874081501414,
+      "loss": 0.8740324378013611,
+      "step": 1702
+    },
+    {
+      "epoch": 0.7189873417721518,
+      "grad_norm": 0.45089733600616455,
+      "learning_rate": 0.00019908092282271053,
+      "loss": 0.8305023908615112,
+      "step": 1704
+    },
+    {
+      "epoch": 0.719831223628692,
+      "grad_norm": 0.3903036415576935,
+      "learning_rate": 0.00019907441463921366,
+      "loss": 0.7718394994735718,
+      "step": 1706
+    },
+    {
+      "epoch": 0.7206751054852321,
+      "grad_norm": 0.42990729212760925,
+      "learning_rate": 0.00019906788360115208,
+      "loss": 0.843289315700531,
+      "step": 1708
+    },
+    {
+      "epoch": 0.7215189873417721,
+      "grad_norm": 0.4287833571434021,
+      "learning_rate": 0.0001990613297100324,
+      "loss": 0.8332282900810242,
+      "step": 1710
+    },
+    {
+      "epoch": 0.7223628691983123,
+      "grad_norm": 0.4557269513607025,
+      "learning_rate": 0.00019905475296736646,
+      "loss": 0.784180223941803,
+      "step": 1712
+    },
+    {
+      "epoch": 0.7232067510548523,
+      "grad_norm": 0.4001525640487671,
+      "learning_rate": 0.0001990481533746714,
+      "loss": 0.7807716131210327,
+      "step": 1714
+    },
+    {
+      "epoch": 0.7240506329113924,
+      "grad_norm": 0.5208709836006165,
+      "learning_rate": 0.0001990415309334696,
+      "loss": 0.7936414480209351,
+      "step": 1716
+    },
+    {
+      "epoch": 0.7248945147679325,
+      "grad_norm": 0.46469953656196594,
+      "learning_rate": 0.00019903488564528875,
+      "loss": 0.7964845895767212,
+      "step": 1718
+    },
+    {
+      "epoch": 0.7257383966244726,
+      "grad_norm": 0.405773401260376,
+      "learning_rate": 0.00019902821751166178,
+      "loss": 0.8302215337753296,
+      "step": 1720
+    },
+    {
+      "epoch": 0.7265822784810126,
+      "grad_norm": 0.3874973654747009,
+      "learning_rate": 0.00019902152653412692,
+      "loss": 0.8486512899398804,
+      "step": 1722
+    },
+    {
+      "epoch": 0.7274261603375527,
+      "grad_norm": 0.46865755319595337,
+      "learning_rate": 0.00019901481271422758,
+      "loss": 0.837841272354126,
+      "step": 1724
+    },
+    {
+      "epoch": 0.7282700421940929,
+      "grad_norm": 0.4139123558998108,
+      "learning_rate": 0.0001990080760535126,
+      "loss": 0.7640942335128784,
+      "step": 1726
+    },
+    {
+      "epoch": 0.7291139240506329,
+      "grad_norm": 0.5003024935722351,
+      "learning_rate": 0.000199001316553536,
+      "loss": 0.8236244320869446,
+      "step": 1728
+    },
+    {
+      "epoch": 0.729957805907173,
+      "grad_norm": 0.38495776057243347,
+      "learning_rate": 0.00019899453421585696,
+      "loss": 0.7792612314224243,
+      "step": 1730
+    },
+    {
+      "epoch": 0.7308016877637131,
+      "grad_norm": 0.4462733268737793,
+      "learning_rate": 0.00019898772904204014,
+      "loss": 0.7733872532844543,
+      "step": 1732
+    },
+    {
+      "epoch": 0.7316455696202532,
+      "grad_norm": 0.398248553276062,
+      "learning_rate": 0.00019898090103365532,
+      "loss": 0.8933147192001343,
+      "step": 1734
+    },
+    {
+      "epoch": 0.7324894514767932,
+      "grad_norm": 0.4231175482273102,
+      "learning_rate": 0.00019897405019227756,
+      "loss": 0.8198356628417969,
+      "step": 1736
+    },
+    {
+      "epoch": 0.7333333333333333,
+      "grad_norm": 0.43639084696769714,
+      "learning_rate": 0.0001989671765194873,
+      "loss": 0.8266391158103943,
+      "step": 1738
+    },
+    {
+      "epoch": 0.7341772151898734,
+      "grad_norm": 0.4235835671424866,
+      "learning_rate": 0.00019896028001687008,
+      "loss": 0.7839997410774231,
+      "step": 1740
+    },
+    {
+      "epoch": 0.7350210970464135,
+      "grad_norm": 0.4033111035823822,
+      "learning_rate": 0.00019895336068601687,
+      "loss": 0.7153518199920654,
+      "step": 1742
+    },
+    {
+      "epoch": 0.7358649789029535,
+      "grad_norm": 0.44281265139579773,
+      "learning_rate": 0.00019894641852852377,
+      "loss": 0.9079289436340332,
+      "step": 1744
+    },
+    {
+      "epoch": 0.7367088607594937,
+      "grad_norm": 0.4882747232913971,
+      "learning_rate": 0.0001989394535459922,
+      "loss": 0.8645254373550415,
+      "step": 1746
+    },
+    {
+      "epoch": 0.7375527426160338,
+      "grad_norm": 0.34858641028404236,
+      "learning_rate": 0.00019893246574002887,
+      "loss": 0.8134214878082275,
+      "step": 1748
+    },
+    {
+      "epoch": 0.7383966244725738,
+      "grad_norm": 0.4273303747177124,
+      "learning_rate": 0.00019892545511224574,
+      "loss": 0.8142663240432739,
+      "step": 1750
+    },
+    {
+      "epoch": 0.739240506329114,
+      "grad_norm": 0.4289957284927368,
+      "learning_rate": 0.00019891842166426004,
+      "loss": 0.7957643866539001,
+      "step": 1752
+    },
+    {
+      "epoch": 0.740084388185654,
+      "grad_norm": 0.4812224805355072,
+      "learning_rate": 0.00019891136539769416,
+      "loss": 0.9000363945960999,
+      "step": 1754
+    },
+    {
+      "epoch": 0.7409282700421941,
+      "grad_norm": 0.4078022837638855,
+      "learning_rate": 0.00019890428631417595,
+      "loss": 0.7719792723655701,
+      "step": 1756
+    },
+    {
+      "epoch": 0.7417721518987341,
+      "grad_norm": 0.40169858932495117,
+      "learning_rate": 0.00019889718441533838,
+      "loss": 0.7681847214698792,
+      "step": 1758
+    },
+    {
+      "epoch": 0.7426160337552743,
+      "grad_norm": 0.3713594377040863,
+      "learning_rate": 0.00019889005970281972,
+      "loss": 0.7536827325820923,
+      "step": 1760
+    },
+    {
+      "epoch": 0.7434599156118143,
+      "grad_norm": 0.4183627665042877,
+      "learning_rate": 0.0001988829121782635,
+      "loss": 0.9000160694122314,
+      "step": 1762
+    },
+    {
+      "epoch": 0.7443037974683544,
+      "grad_norm": 0.4241594672203064,
+      "learning_rate": 0.00019887574184331851,
+      "loss": 0.8047307133674622,
+      "step": 1764
+    },
+    {
+      "epoch": 0.7451476793248946,
+      "grad_norm": 0.4249233603477478,
+      "learning_rate": 0.00019886854869963883,
+      "loss": 0.8932583928108215,
+      "step": 1766
+    },
+    {
+      "epoch": 0.7459915611814346,
+      "grad_norm": 0.35815340280532837,
+      "learning_rate": 0.00019886133274888378,
+      "loss": 0.7684977054595947,
+      "step": 1768
+    },
+    {
+      "epoch": 0.7468354430379747,
+      "grad_norm": 0.41035401821136475,
+      "learning_rate": 0.00019885409399271795,
+      "loss": 0.8644338250160217,
+      "step": 1770
+    },
+    {
+      "epoch": 0.7476793248945147,
+      "grad_norm": 0.5450286269187927,
+      "learning_rate": 0.00019884683243281116,
+      "loss": 0.8035860657691956,
+      "step": 1772
+    },
+    {
+      "epoch": 0.7485232067510549,
+      "grad_norm": 0.5280999541282654,
+      "learning_rate": 0.0001988395480708385,
+      "loss": 0.8861207962036133,
+      "step": 1774
+    },
+    {
+      "epoch": 0.7493670886075949,
+      "grad_norm": 0.42665231227874756,
+      "learning_rate": 0.00019883224090848036,
+      "loss": 0.7506847977638245,
+      "step": 1776
+    },
+    {
+      "epoch": 0.750210970464135,
+      "grad_norm": 0.39029282331466675,
+      "learning_rate": 0.00019882491094742232,
+      "loss": 0.7662046551704407,
+      "step": 1778
+    },
+    {
+      "epoch": 0.7510548523206751,
+      "grad_norm": 0.43870019912719727,
+      "learning_rate": 0.00019881755818935534,
+      "loss": 0.7590143084526062,
+      "step": 1780
+    },
+    {
+      "epoch": 0.7518987341772152,
+      "grad_norm": 0.4195050001144409,
+      "learning_rate": 0.0001988101826359755,
+      "loss": 0.9340365529060364,
+      "step": 1782
+    },
+    {
+      "epoch": 0.7527426160337553,
+      "grad_norm": 0.4236123263835907,
+      "learning_rate": 0.0001988027842889842,
+      "loss": 0.7969209551811218,
+      "step": 1784
+    },
+    {
+      "epoch": 0.7535864978902953,
+      "grad_norm": 0.4274357259273529,
+      "learning_rate": 0.00019879536315008808,
+      "loss": 0.7892382144927979,
+      "step": 1786
+    },
+    {
+      "epoch": 0.7544303797468355,
+      "grad_norm": 0.4622916579246521,
+      "learning_rate": 0.00019878791922099912,
+      "loss": 0.8752562403678894,
+      "step": 1788
+    },
+    {
+      "epoch": 0.7552742616033755,
+      "grad_norm": 0.3648734986782074,
+      "learning_rate": 0.00019878045250343445,
+      "loss": 0.813011884689331,
+      "step": 1790
+    },
+    {
+      "epoch": 0.7561181434599156,
+      "grad_norm": 0.41361239552497864,
+      "learning_rate": 0.00019877296299911648,
+      "loss": 0.7310198545455933,
+      "step": 1792
+    },
+    {
+      "epoch": 0.7569620253164557,
+      "grad_norm": 0.4113222062587738,
+      "learning_rate": 0.00019876545070977294,
+      "loss": 0.858386754989624,
+      "step": 1794
+    },
+    {
+      "epoch": 0.7578059071729958,
+      "grad_norm": 0.4295370876789093,
+      "learning_rate": 0.00019875791563713676,
+      "loss": 0.8377325534820557,
+      "step": 1796
+    },
+    {
+      "epoch": 0.7586497890295358,
+      "grad_norm": 0.4227522015571594,
+      "learning_rate": 0.0001987503577829461,
+      "loss": 0.9139418005943298,
+      "step": 1798
+    },
+    {
+      "epoch": 0.759493670886076,
+      "grad_norm": 0.3638151288032532,
+      "learning_rate": 0.00019874277714894442,
+      "loss": 0.6994872689247131,
+      "step": 1800
+    },
+    {
+      "epoch": 0.759493670886076,
+      "eval_loss": 0.8625519275665283,
+      "eval_runtime": 686.4271,
+      "eval_samples_per_second": 3.07,
+      "eval_steps_per_second": 3.07,
+      "step": 1800
+    },
+    {
+      "epoch": 0.760337552742616,
+      "grad_norm": 0.46954411268234253,
+      "learning_rate": 0.00019873517373688047,
+      "loss": 0.8681167960166931,
+      "step": 1802
+    },
+    {
+      "epoch": 0.7611814345991561,
+      "grad_norm": 0.41958674788475037,
+      "learning_rate": 0.00019872754754850819,
+      "loss": 0.7420852184295654,
+      "step": 1804
+    },
+    {
+      "epoch": 0.7620253164556962,
+      "grad_norm": 0.431226521730423,
+      "learning_rate": 0.00019871989858558678,
+      "loss": 0.8018608093261719,
+      "step": 1806
+    },
+    {
+      "epoch": 0.7628691983122363,
+      "grad_norm": 0.45129454135894775,
+      "learning_rate": 0.0001987122268498807,
+      "loss": 0.8793904781341553,
+      "step": 1808
+    },
+    {
+      "epoch": 0.7637130801687764,
+      "grad_norm": 0.44278961420059204,
+      "learning_rate": 0.00019870453234315972,
+      "loss": 0.8416730761528015,
+      "step": 1810
+    },
+    {
+      "epoch": 0.7645569620253164,
+      "grad_norm": 0.4689098000526428,
+      "learning_rate": 0.00019869681506719876,
+      "loss": 0.8225743174552917,
+      "step": 1812
+    },
+    {
+      "epoch": 0.7654008438818566,
+      "grad_norm": 0.4413386881351471,
+      "learning_rate": 0.00019868907502377806,
+      "loss": 0.7871913313865662,
+      "step": 1814
+    },
+    {
+      "epoch": 0.7662447257383966,
+      "grad_norm": 0.48395273089408875,
+      "learning_rate": 0.00019868131221468315,
+      "loss": 0.8616237044334412,
+      "step": 1816
+    },
+    {
+      "epoch": 0.7670886075949367,
+      "grad_norm": 0.370263934135437,
+      "learning_rate": 0.0001986735266417047,
+      "loss": 0.7243452668190002,
+      "step": 1818
+    },
+    {
+      "epoch": 0.7679324894514767,
+      "grad_norm": 0.4522445797920227,
+      "learning_rate": 0.00019866571830663875,
+      "loss": 0.8623812198638916,
+      "step": 1820
+    },
+    {
+      "epoch": 0.7687763713080169,
+      "grad_norm": 0.43821004033088684,
+      "learning_rate": 0.00019865788721128643,
+      "loss": 0.776618480682373,
+      "step": 1822
+    },
+    {
+      "epoch": 0.769620253164557,
+      "grad_norm": 0.3825650215148926,
+      "learning_rate": 0.00019865003335745436,
+      "loss": 0.8722774982452393,
+      "step": 1824
+    },
+    {
+      "epoch": 0.770464135021097,
+      "grad_norm": 0.43684205412864685,
+      "learning_rate": 0.00019864215674695416,
+      "loss": 0.907802939414978,
+      "step": 1826
+    },
+    {
+      "epoch": 0.7713080168776372,
+      "grad_norm": 0.4128975570201874,
+      "learning_rate": 0.00019863425738160287,
+      "loss": 0.8505539298057556,
+      "step": 1828
+    },
+    {
+      "epoch": 0.7721518987341772,
+      "grad_norm": 0.4254065454006195,
+      "learning_rate": 0.0001986263352632227,
+      "loss": 0.83274906873703,
+      "step": 1830
+    },
+    {
+      "epoch": 0.7729957805907173,
+      "grad_norm": 0.4137375056743622,
+      "learning_rate": 0.00019861839039364118,
+      "loss": 0.8172078728675842,
+      "step": 1832
+    },
+    {
+      "epoch": 0.7738396624472574,
+      "grad_norm": 0.4387703239917755,
+      "learning_rate": 0.00019861042277469096,
+      "loss": 0.8610842227935791,
+      "step": 1834
+    },
+    {
+      "epoch": 0.7746835443037975,
+      "grad_norm": 0.398952454328537,
+      "learning_rate": 0.00019860243240821005,
+      "loss": 0.7961188554763794,
+      "step": 1836
+    },
+    {
+      "epoch": 0.7755274261603375,
+      "grad_norm": 0.4079921543598175,
+      "learning_rate": 0.0001985944192960417,
+      "loss": 0.8078321218490601,
+      "step": 1838
+    },
+    {
+      "epoch": 0.7763713080168776,
+      "grad_norm": 0.4701489508152008,
+      "learning_rate": 0.00019858638344003433,
+      "loss": 0.8743909001350403,
+      "step": 1840
+    },
+    {
+      "epoch": 0.7772151898734178,
+      "grad_norm": 0.5391269326210022,
+      "learning_rate": 0.00019857832484204173,
+      "loss": 0.7681707143783569,
+      "step": 1842
+    },
+    {
+      "epoch": 0.7780590717299578,
+      "grad_norm": 0.36250734329223633,
+      "learning_rate": 0.00019857024350392277,
+      "loss": 0.7817525863647461,
+      "step": 1844
+    },
+    {
+      "epoch": 0.7789029535864979,
+      "grad_norm": 0.41237136721611023,
+      "learning_rate": 0.00019856213942754175,
+      "loss": 0.7614551782608032,
+      "step": 1846
+    },
+    {
+      "epoch": 0.779746835443038,
+      "grad_norm": 0.3968869745731354,
+      "learning_rate": 0.00019855401261476807,
+      "loss": 0.7492313981056213,
+      "step": 1848
+    },
+    {
+      "epoch": 0.7805907172995781,
+      "grad_norm": 0.441476434469223,
+      "learning_rate": 0.00019854586306747646,
+      "loss": 0.8615695238113403,
+      "step": 1850
+    },
+    {
+      "epoch": 0.7814345991561181,
+      "grad_norm": 0.4486154317855835,
+      "learning_rate": 0.00019853769078754686,
+      "loss": 0.8416724801063538,
+      "step": 1852
+    },
+    {
+      "epoch": 0.7822784810126582,
+      "grad_norm": 0.45148879289627075,
+      "learning_rate": 0.0001985294957768644,
+      "loss": 0.8987806439399719,
+      "step": 1854
+    },
+    {
+      "epoch": 0.7831223628691983,
+      "grad_norm": 0.39330095052719116,
+      "learning_rate": 0.0001985212780373196,
+      "loss": 0.7597590088844299,
+      "step": 1856
+    },
+    {
+      "epoch": 0.7839662447257384,
+      "grad_norm": 0.4260408580303192,
+      "learning_rate": 0.00019851303757080808,
+      "loss": 0.8429927229881287,
+      "step": 1858
+    },
+    {
+      "epoch": 0.7848101265822784,
+      "grad_norm": 0.37519171833992004,
+      "learning_rate": 0.00019850477437923075,
+      "loss": 0.7443564534187317,
+      "step": 1860
+    },
+    {
+      "epoch": 0.7856540084388186,
+      "grad_norm": 0.4991084337234497,
+      "learning_rate": 0.00019849648846449383,
+      "loss": 0.934238612651825,
+      "step": 1862
+    },
+    {
+      "epoch": 0.7864978902953587,
+      "grad_norm": 0.4711335003376007,
+      "learning_rate": 0.00019848817982850867,
+      "loss": 0.8938905000686646,
+      "step": 1864
+    },
+    {
+      "epoch": 0.7873417721518987,
+      "grad_norm": 0.4145370423793793,
+      "learning_rate": 0.00019847984847319193,
+      "loss": 0.8745643496513367,
+      "step": 1866
+    },
+    {
+      "epoch": 0.7881856540084389,
+      "grad_norm": 0.3444044888019562,
+      "learning_rate": 0.00019847149440046548,
+      "loss": 0.7178786396980286,
+      "step": 1868
+    },
+    {
+      "epoch": 0.7890295358649789,
+      "grad_norm": 0.34632962942123413,
+      "learning_rate": 0.00019846311761225646,
+      "loss": 0.9114303588867188,
+      "step": 1870
+    },
+    {
+      "epoch": 0.789873417721519,
+      "grad_norm": 0.38630741834640503,
+      "learning_rate": 0.0001984547181104972,
+      "loss": 0.836281955242157,
+      "step": 1872
+    },
+    {
+      "epoch": 0.790717299578059,
+      "grad_norm": 0.40947434306144714,
+      "learning_rate": 0.00019844629589712534,
+      "loss": 0.8247858881950378,
+      "step": 1874
+    },
+    {
+      "epoch": 0.7915611814345992,
+      "grad_norm": 0.45878684520721436,
+      "learning_rate": 0.00019843785097408372,
+      "loss": 0.8446351289749146,
+      "step": 1876
+    },
+    {
+      "epoch": 0.7924050632911392,
+      "grad_norm": 0.37202852964401245,
+      "learning_rate": 0.0001984293833433204,
+      "loss": 0.8567686676979065,
+      "step": 1878
+    },
+    {
+      "epoch": 0.7932489451476793,
+      "grad_norm": 0.3839069902896881,
+      "learning_rate": 0.00019842089300678876,
+      "loss": 0.8134047985076904,
+      "step": 1880
+    },
+    {
+      "epoch": 0.7940928270042195,
+      "grad_norm": 0.4018687605857849,
+      "learning_rate": 0.00019841237996644727,
+      "loss": 0.8670110106468201,
+      "step": 1882
+    },
+    {
+      "epoch": 0.7949367088607595,
+      "grad_norm": 0.372089684009552,
+      "learning_rate": 0.0001984038442242598,
+      "loss": 0.7975355386734009,
+      "step": 1884
+    },
+    {
+      "epoch": 0.7957805907172996,
+      "grad_norm": 0.4492705166339874,
+      "learning_rate": 0.0001983952857821953,
+      "loss": 0.8755611777305603,
+      "step": 1886
+    },
+    {
+      "epoch": 0.7966244725738396,
+      "grad_norm": 0.451054185628891,
+      "learning_rate": 0.00019838670464222813,
+      "loss": 0.8680241107940674,
+      "step": 1888
+    },
+    {
+      "epoch": 0.7974683544303798,
+      "grad_norm": 0.4036900997161865,
+      "learning_rate": 0.00019837810080633773,
+      "loss": 0.7990990877151489,
+      "step": 1890
+    },
+    {
+      "epoch": 0.7983122362869198,
+      "grad_norm": 0.46125802397727966,
+      "learning_rate": 0.0001983694742765089,
+      "loss": 0.8598953485488892,
+      "step": 1892
+    },
+    {
+      "epoch": 0.7991561181434599,
+      "grad_norm": 0.4458785951137543,
+      "learning_rate": 0.00019836082505473153,
+      "loss": 0.7707474827766418,
+      "step": 1894
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.44501590728759766,
+      "learning_rate": 0.0001983521531430009,
+      "loss": 0.8069534301757812,
+      "step": 1896
+    },
+    {
+      "epoch": 0.8008438818565401,
+      "grad_norm": 0.400771826505661,
+      "learning_rate": 0.0001983434585433174,
+      "loss": 0.862334668636322,
+      "step": 1898
+    },
+    {
+      "epoch": 0.8016877637130801,
+      "grad_norm": 0.39193347096443176,
+      "learning_rate": 0.00019833474125768676,
+      "loss": 0.8125433325767517,
+      "step": 1900
+    },
+    {
+      "epoch": 0.8016877637130801,
+      "eval_loss": 0.8546335697174072,
+      "eval_runtime": 688.5301,
+      "eval_samples_per_second": 3.06,
+      "eval_steps_per_second": 3.06,
+      "step": 1900
+    },
+    {
+      "epoch": 0.8025316455696202,
+      "grad_norm": 0.441129207611084,
+      "learning_rate": 0.00019832600128811986,
+      "loss": 0.8246012330055237,
+      "step": 1902
+    },
+    {
+      "epoch": 0.8033755274261604,
+      "grad_norm": 0.4089467227458954,
+      "learning_rate": 0.00019831723863663285,
+      "loss": 0.8387641310691833,
+      "step": 1904
+    },
+    {
+      "epoch": 0.8042194092827004,
+      "grad_norm": 0.4871654510498047,
+      "learning_rate": 0.0001983084533052471,
+      "loss": 0.7891429662704468,
+      "step": 1906
+    },
+    {
+      "epoch": 0.8050632911392405,
+      "grad_norm": 0.40085604786872864,
+      "learning_rate": 0.00019829964529598923,
+      "loss": 0.8008774518966675,
+      "step": 1908
+    },
+    {
+      "epoch": 0.8059071729957806,
+      "grad_norm": 0.38775014877319336,
+      "learning_rate": 0.00019829081461089109,
+      "loss": 0.7921834588050842,
+      "step": 1910
+    },
+    {
+      "epoch": 0.8067510548523207,
+      "grad_norm": 0.37830594182014465,
+      "learning_rate": 0.0001982819612519897,
+      "loss": 0.7752519845962524,
+      "step": 1912
+    },
+    {
+      "epoch": 0.8075949367088607,
+      "grad_norm": 0.41137194633483887,
+      "learning_rate": 0.0001982730852213274,
+      "loss": 0.7958255410194397,
+      "step": 1914
+    },
+    {
+      "epoch": 0.8084388185654009,
+      "grad_norm": 0.4871830940246582,
+      "learning_rate": 0.0001982641865209517,
+      "loss": 0.8177281618118286,
+      "step": 1916
+    },
+    {
+      "epoch": 0.809282700421941,
+      "grad_norm": 0.4157388210296631,
+      "learning_rate": 0.0001982552651529154,
+      "loss": 0.7726616859436035,
+      "step": 1918
+    },
+    {
+      "epoch": 0.810126582278481,
+      "grad_norm": 0.32338014245033264,
+      "learning_rate": 0.00019824632111927645,
+      "loss": 0.6601400375366211,
+      "step": 1920
+    },
+    {
+      "epoch": 0.810970464135021,
+      "grad_norm": 0.4642949104309082,
+      "learning_rate": 0.00019823735442209804,
+      "loss": 0.7866622805595398,
+      "step": 1922
+    },
+    {
+      "epoch": 0.8118143459915612,
+      "grad_norm": 0.43558797240257263,
+      "learning_rate": 0.00019822836506344865,
+      "loss": 0.7857986092567444,
+      "step": 1924
+    },
+    {
+      "epoch": 0.8126582278481013,
+      "grad_norm": 0.3996953070163727,
+      "learning_rate": 0.000198219353045402,
+      "loss": 0.8407763838768005,
+      "step": 1926
+    },
+    {
+      "epoch": 0.8135021097046413,
+      "grad_norm": 0.4415414333343506,
+      "learning_rate": 0.00019821031837003686,
+      "loss": 0.8433752655982971,
+      "step": 1928
+    },
+    {
+      "epoch": 0.8143459915611815,
+      "grad_norm": 0.37329310178756714,
+      "learning_rate": 0.00019820126103943752,
+      "loss": 0.6941158175468445,
+      "step": 1930
+    },
+    {
+      "epoch": 0.8151898734177215,
+      "grad_norm": 0.4482601284980774,
+      "learning_rate": 0.00019819218105569323,
+      "loss": 0.775604248046875,
+      "step": 1932
+    },
+    {
+      "epoch": 0.8160337552742616,
+      "grad_norm": 0.426194429397583,
+      "learning_rate": 0.00019818307842089854,
+      "loss": 0.8496418595314026,
+      "step": 1934
+    },
+    {
+      "epoch": 0.8168776371308016,
+      "grad_norm": 0.42379093170166016,
+      "learning_rate": 0.00019817395313715335,
+      "loss": 0.7767958045005798,
+      "step": 1936
+    },
+    {
+      "epoch": 0.8177215189873418,
+      "grad_norm": 0.4149782061576843,
+      "learning_rate": 0.0001981648052065626,
+      "loss": 0.7347666025161743,
+      "step": 1938
+    },
+    {
+      "epoch": 0.8185654008438819,
+      "grad_norm": 0.3927431106567383,
+      "learning_rate": 0.0001981556346312366,
+      "loss": 0.7839647531509399,
+      "step": 1940
+    },
+    {
+      "epoch": 0.8194092827004219,
+      "grad_norm": 0.4483136236667633,
+      "learning_rate": 0.00019814644141329083,
+      "loss": 0.8870531916618347,
+      "step": 1942
+    },
+    {
+      "epoch": 0.8202531645569621,
+      "grad_norm": 0.4400147497653961,
+      "learning_rate": 0.00019813722555484594,
+      "loss": 0.8646620512008667,
+      "step": 1944
+    },
+    {
+      "epoch": 0.8210970464135021,
+      "grad_norm": 0.469971626996994,
+      "learning_rate": 0.00019812798705802785,
+      "loss": 0.9576541781425476,
+      "step": 1946
+    },
+    {
+      "epoch": 0.8219409282700422,
+      "grad_norm": 0.388113409280777,
+      "learning_rate": 0.00019811872592496776,
+      "loss": 0.7956477403640747,
+      "step": 1948
+    },
+    {
+      "epoch": 0.8227848101265823,
+      "grad_norm": 0.4271424114704132,
+      "learning_rate": 0.00019810944215780201,
+      "loss": 0.8440352082252502,
+      "step": 1950
+    },
+    {
+      "epoch": 0.8236286919831224,
+      "grad_norm": 0.4097966253757477,
+      "learning_rate": 0.00019810013575867217,
+      "loss": 0.8589065074920654,
+      "step": 1952
+    },
+    {
+      "epoch": 0.8244725738396624,
+      "grad_norm": 0.41550710797309875,
+      "learning_rate": 0.00019809080672972504,
+      "loss": 0.8606626987457275,
+      "step": 1954
+    },
+    {
+      "epoch": 0.8253164556962025,
+      "grad_norm": 0.3680916130542755,
+      "learning_rate": 0.0001980814550731127,
+      "loss": 0.7582436800003052,
+      "step": 1956
+    },
+    {
+      "epoch": 0.8261603375527427,
+      "grad_norm": 0.573297917842865,
+      "learning_rate": 0.00019807208079099234,
+      "loss": 0.8345186710357666,
+      "step": 1958
+    },
+    {
+      "epoch": 0.8270042194092827,
+      "grad_norm": 0.43631303310394287,
+      "learning_rate": 0.00019806268388552646,
+      "loss": 0.889683723449707,
+      "step": 1960
+    },
+    {
+      "epoch": 0.8278481012658228,
+      "grad_norm": 0.442531019449234,
+      "learning_rate": 0.00019805326435888275,
+      "loss": 0.8092973232269287,
+      "step": 1962
+    },
+    {
+      "epoch": 0.8286919831223629,
+      "grad_norm": 0.45020824670791626,
+      "learning_rate": 0.0001980438222132341,
+      "loss": 0.8919535279273987,
+      "step": 1964
+    },
+    {
+      "epoch": 0.829535864978903,
+      "grad_norm": 0.43131786584854126,
+      "learning_rate": 0.0001980343574507586,
+      "loss": 0.7652381658554077,
+      "step": 1966
+    },
+    {
+      "epoch": 0.830379746835443,
+      "grad_norm": 0.4461326599121094,
+      "learning_rate": 0.00019802487007363967,
+      "loss": 0.8512389659881592,
+      "step": 1968
+    },
+    {
+      "epoch": 0.8312236286919831,
+      "grad_norm": 0.4676894247531891,
+      "learning_rate": 0.0001980153600840658,
+      "loss": 0.8681327700614929,
+      "step": 1970
+    },
+    {
+      "epoch": 0.8320675105485232,
+      "grad_norm": 0.38376888632774353,
+      "learning_rate": 0.00019800582748423075,
+      "loss": 0.7476955652236938,
+      "step": 1972
+    },
+    {
+      "epoch": 0.8329113924050633,
+      "grad_norm": 0.37173229455947876,
+      "learning_rate": 0.00019799627227633354,
+      "loss": 0.814192533493042,
+      "step": 1974
+    },
+    {
+      "epoch": 0.8337552742616033,
+      "grad_norm": 0.4345237910747528,
+      "learning_rate": 0.00019798669446257844,
+      "loss": 0.7668994069099426,
+      "step": 1976
+    },
+    {
+      "epoch": 0.8345991561181435,
+      "grad_norm": 0.45872583985328674,
+      "learning_rate": 0.0001979770940451747,
+      "loss": 0.8321775197982788,
+      "step": 1978
+    },
+    {
+      "epoch": 0.8354430379746836,
+      "grad_norm": 0.40618205070495605,
+      "learning_rate": 0.0001979674710263371,
+      "loss": 0.7824342250823975,
+      "step": 1980
+    },
+    {
+      "epoch": 0.8362869198312236,
+      "grad_norm": 0.38377392292022705,
+      "learning_rate": 0.00019795782540828544,
+      "loss": 0.8752480745315552,
+      "step": 1982
+    },
+    {
+      "epoch": 0.8371308016877637,
+      "grad_norm": 0.36340072751045227,
+      "learning_rate": 0.00019794815719324476,
+      "loss": 0.66700679063797,
+      "step": 1984
+    },
+    {
+      "epoch": 0.8379746835443038,
+      "grad_norm": 0.4638384282588959,
+      "learning_rate": 0.00019793846638344536,
+      "loss": 0.8582209944725037,
+      "step": 1986
+    },
+    {
+      "epoch": 0.8388185654008439,
+      "grad_norm": 0.38364699482917786,
+      "learning_rate": 0.00019792875298112268,
+      "loss": 0.7724968791007996,
+      "step": 1988
+    },
+    {
+      "epoch": 0.8396624472573839,
+      "grad_norm": 0.439730703830719,
+      "learning_rate": 0.00019791901698851749,
+      "loss": 0.8542404174804688,
+      "step": 1990
+    },
+    {
+      "epoch": 0.8405063291139241,
+      "grad_norm": 0.4622722566127777,
+      "learning_rate": 0.00019790925840787563,
+      "loss": 0.8424296975135803,
+      "step": 1992
+    },
+    {
+      "epoch": 0.8413502109704641,
+      "grad_norm": 0.4491989314556122,
+      "learning_rate": 0.0001978994772414482,
+      "loss": 0.8400710225105286,
+      "step": 1994
+    },
+    {
+      "epoch": 0.8421940928270042,
+      "grad_norm": 0.372295081615448,
+      "learning_rate": 0.0001978896734914916,
+      "loss": 0.7265452146530151,
+      "step": 1996
+    },
+    {
+      "epoch": 0.8430379746835444,
+      "grad_norm": 0.4328666925430298,
+      "learning_rate": 0.00019787984716026732,
+      "loss": 0.8458228707313538,
+      "step": 1998
+    },
+    {
+      "epoch": 0.8438818565400844,
+      "grad_norm": 0.42246317863464355,
+      "learning_rate": 0.00019786999825004216,
+      "loss": 0.8517491817474365,
+      "step": 2000
+    },
+    {
+      "epoch": 0.8438818565400844,
+      "eval_loss": 0.8460908532142639,
+      "eval_runtime": 685.2518,
+      "eval_samples_per_second": 3.075,
+      "eval_steps_per_second": 3.075,
+      "step": 2000
+    },
+    {
+      "epoch": 0.8447257383966245,
+      "grad_norm": 0.3927323818206787,
+      "learning_rate": 0.00019786012676308802,
+      "loss": 0.8045481443405151,
+      "step": 2002
+    },
+    {
+      "epoch": 0.8455696202531645,
+      "grad_norm": 0.38189369440078735,
+      "learning_rate": 0.00019785023270168202,
+      "loss": 0.7509838938713074,
+      "step": 2004
+    },
+    {
+      "epoch": 0.8464135021097047,
+      "grad_norm": 0.40383535623550415,
+      "learning_rate": 0.00019784031606810663,
+      "loss": 0.7730163335800171,
+      "step": 2006
+    },
+    {
+      "epoch": 0.8472573839662447,
+      "grad_norm": 0.3665419816970825,
+      "learning_rate": 0.00019783037686464935,
+      "loss": 0.7096828818321228,
+      "step": 2008
+    },
+    {
+      "epoch": 0.8481012658227848,
+      "grad_norm": 0.36623379588127136,
+      "learning_rate": 0.00019782041509360302,
+      "loss": 0.8186401128768921,
+      "step": 2010
+    },
+    {
+      "epoch": 0.8489451476793249,
+      "grad_norm": 0.477255642414093,
+      "learning_rate": 0.0001978104307572656,
+      "loss": 0.8502879738807678,
+      "step": 2012
+    },
+    {
+      "epoch": 0.849789029535865,
+      "grad_norm": 0.44456636905670166,
+      "learning_rate": 0.00019780042385794032,
+      "loss": 0.9010884761810303,
+      "step": 2014
+    },
+    {
+      "epoch": 0.850632911392405,
+      "grad_norm": 0.42378175258636475,
+      "learning_rate": 0.00019779039439793552,
+      "loss": 0.774183452129364,
+      "step": 2016
+    },
+    {
+      "epoch": 0.8514767932489451,
+      "grad_norm": 0.43021997809410095,
+      "learning_rate": 0.00019778034237956482,
+      "loss": 0.8487694263458252,
+      "step": 2018
+    },
+    {
+      "epoch": 0.8523206751054853,
+      "grad_norm": 0.38188812136650085,
+      "learning_rate": 0.00019777026780514704,
+      "loss": 0.8004868030548096,
+      "step": 2020
+    },
+    {
+      "epoch": 0.8531645569620253,
+      "grad_norm": 0.3684130609035492,
+      "learning_rate": 0.0001977601706770062,
+      "loss": 0.7680953145027161,
+      "step": 2022
+    },
+    {
+      "epoch": 0.8540084388185654,
+      "grad_norm": 0.444844514131546,
+      "learning_rate": 0.00019775005099747153,
+      "loss": 0.8990264534950256,
+      "step": 2024
+    },
+    {
+      "epoch": 0.8548523206751055,
+      "grad_norm": 0.3432365655899048,
+      "learning_rate": 0.00019773990876887737,
+      "loss": 0.7676048874855042,
+      "step": 2026
+    },
+    {
+      "epoch": 0.8556962025316456,
+      "grad_norm": 0.3728618025779724,
+      "learning_rate": 0.00019772974399356343,
+      "loss": 0.8546769618988037,
+      "step": 2028
+    },
+    {
+      "epoch": 0.8565400843881856,
+      "grad_norm": 0.42767980694770813,
+      "learning_rate": 0.00019771955667387449,
+      "loss": 0.8009977340698242,
+      "step": 2030
+    },
+    {
+      "epoch": 0.8573839662447258,
+      "grad_norm": 0.4125710129737854,
+      "learning_rate": 0.00019770934681216055,
+      "loss": 0.8424299955368042,
+      "step": 2032
+    },
+    {
+      "epoch": 0.8582278481012658,
+      "grad_norm": 0.4399266839027405,
+      "learning_rate": 0.00019769911441077685,
+      "loss": 0.7940502762794495,
+      "step": 2034
+    },
+    {
+      "epoch": 0.8590717299578059,
+      "grad_norm": 0.41797709465026855,
+      "learning_rate": 0.0001976888594720838,
+      "loss": 0.9424314498901367,
+      "step": 2036
+    },
+    {
+      "epoch": 0.859915611814346,
+      "grad_norm": 0.5245009660720825,
+      "learning_rate": 0.00019767858199844698,
+      "loss": 0.9102896451950073,
+      "step": 2038
+    },
+    {
+      "epoch": 0.8607594936708861,
+      "grad_norm": 0.4092593193054199,
+      "learning_rate": 0.00019766828199223728,
+      "loss": 0.7434720396995544,
+      "step": 2040
+    },
+    {
+      "epoch": 0.8616033755274262,
+      "grad_norm": 0.40787473320961,
+      "learning_rate": 0.00019765795945583068,
+      "loss": 0.7441987991333008,
+      "step": 2042
+    },
+    {
+      "epoch": 0.8624472573839662,
+      "grad_norm": 0.42438021302223206,
+      "learning_rate": 0.0001976476143916084,
+      "loss": 0.8317872881889343,
+      "step": 2044
+    },
+    {
+      "epoch": 0.8632911392405064,
+      "grad_norm": 0.38261914253234863,
+      "learning_rate": 0.00019763724680195682,
+      "loss": 0.764509379863739,
+      "step": 2046
+    },
+    {
+      "epoch": 0.8641350210970464,
+      "grad_norm": 0.43295741081237793,
+      "learning_rate": 0.00019762685668926758,
+      "loss": 0.8666167855262756,
+      "step": 2048
+    },
+    {
+      "epoch": 0.8649789029535865,
+      "grad_norm": 0.36384403705596924,
+      "learning_rate": 0.00019761644405593743,
+      "loss": 0.7606754302978516,
+      "step": 2050
+    },
+    {
+      "epoch": 0.8658227848101265,
+      "grad_norm": 0.4296814501285553,
+      "learning_rate": 0.0001976060089043684,
+      "loss": 0.827495813369751,
+      "step": 2052
+    },
+    {
+      "epoch": 0.8666666666666667,
+      "grad_norm": 0.3831455111503601,
+      "learning_rate": 0.0001975955512369677,
+      "loss": 0.7418577075004578,
+      "step": 2054
+    },
+    {
+      "epoch": 0.8675105485232067,
+      "grad_norm": 0.44710344076156616,
+      "learning_rate": 0.0001975850710561477,
+      "loss": 0.8603045344352722,
+      "step": 2056
+    },
+    {
+      "epoch": 0.8683544303797468,
+      "grad_norm": 0.44012558460235596,
+      "learning_rate": 0.00019757456836432593,
+      "loss": 0.900658905506134,
+      "step": 2058
+    },
+    {
+      "epoch": 0.869198312236287,
+      "grad_norm": 0.42166298627853394,
+      "learning_rate": 0.0001975640431639252,
+      "loss": 0.8416497707366943,
+      "step": 2060
+    },
+    {
+      "epoch": 0.870042194092827,
+      "grad_norm": 0.40824490785598755,
+      "learning_rate": 0.00019755349545737345,
+      "loss": 0.8281066417694092,
+      "step": 2062
+    },
+    {
+      "epoch": 0.8708860759493671,
+      "grad_norm": 0.4272223711013794,
+      "learning_rate": 0.00019754292524710388,
+      "loss": 0.83869469165802,
+      "step": 2064
+    },
+    {
+      "epoch": 0.8717299578059071,
+      "grad_norm": 0.3732798397541046,
+      "learning_rate": 0.00019753233253555478,
+      "loss": 0.8337644934654236,
+      "step": 2066
+    },
+    {
+      "epoch": 0.8725738396624473,
+      "grad_norm": 0.4048069417476654,
+      "learning_rate": 0.00019752171732516973,
+      "loss": 0.7677452564239502,
+      "step": 2068
+    },
+    {
+      "epoch": 0.8734177215189873,
+      "grad_norm": 0.43628430366516113,
+      "learning_rate": 0.00019751107961839742,
+      "loss": 0.8320774435997009,
+      "step": 2070
+    },
+    {
+      "epoch": 0.8742616033755274,
+      "grad_norm": 0.43674105405807495,
+      "learning_rate": 0.00019750041941769174,
+      "loss": 0.8480262160301208,
+      "step": 2072
+    },
+    {
+      "epoch": 0.8751054852320675,
+      "grad_norm": 0.3678584098815918,
+      "learning_rate": 0.0001974897367255119,
+      "loss": 0.7598503828048706,
+      "step": 2074
+    },
+    {
+      "epoch": 0.8759493670886076,
+      "grad_norm": 0.4316411316394806,
+      "learning_rate": 0.0001974790315443221,
+      "loss": 0.7907084822654724,
+      "step": 2076
+    },
+    {
+      "epoch": 0.8767932489451477,
+      "grad_norm": 0.426465779542923,
+      "learning_rate": 0.00019746830387659182,
+      "loss": 0.8287727236747742,
+      "step": 2078
+    },
+    {
+      "epoch": 0.8776371308016878,
+      "grad_norm": 0.447337806224823,
+      "learning_rate": 0.00019745755372479578,
+      "loss": 0.8464577794075012,
+      "step": 2080
+    },
+    {
+      "epoch": 0.8784810126582279,
+      "grad_norm": 0.37877580523490906,
+      "learning_rate": 0.0001974467810914138,
+      "loss": 0.7383086681365967,
+      "step": 2082
+    },
+    {
+      "epoch": 0.8793248945147679,
+      "grad_norm": 0.3914664089679718,
+      "learning_rate": 0.0001974359859789309,
+      "loss": 0.729221522808075,
+      "step": 2084
+    },
+    {
+      "epoch": 0.880168776371308,
+      "grad_norm": 0.39767366647720337,
+      "learning_rate": 0.00019742516838983733,
+      "loss": 0.692332923412323,
+      "step": 2086
+    },
+    {
+      "epoch": 0.8810126582278481,
+      "grad_norm": 0.4396754205226898,
+      "learning_rate": 0.0001974143283266285,
+      "loss": 0.8540068864822388,
+      "step": 2088
+    },
+    {
+      "epoch": 0.8818565400843882,
+      "grad_norm": 0.4968172609806061,
+      "learning_rate": 0.00019740346579180502,
+      "loss": 0.8570998907089233,
+      "step": 2090
+    },
+    {
+      "epoch": 0.8827004219409282,
+      "grad_norm": 0.41686809062957764,
+      "learning_rate": 0.00019739258078787264,
+      "loss": 0.784139096736908,
+      "step": 2092
+    },
+    {
+      "epoch": 0.8835443037974684,
+      "grad_norm": 0.38788047432899475,
+      "learning_rate": 0.0001973816733173423,
+      "loss": 0.8521397113800049,
+      "step": 2094
+    },
+    {
+      "epoch": 0.8843881856540085,
+      "grad_norm": 0.38369178771972656,
+      "learning_rate": 0.00019737074338273023,
+      "loss": 0.8457935452461243,
+      "step": 2096
+    },
+    {
+      "epoch": 0.8852320675105485,
+      "grad_norm": 0.38869398832321167,
+      "learning_rate": 0.0001973597909865577,
+      "loss": 0.7425959706306458,
+      "step": 2098
+    },
+    {
+      "epoch": 0.8860759493670886,
+      "grad_norm": 0.4067917466163635,
+      "learning_rate": 0.00019734881613135122,
+      "loss": 0.7742393016815186,
+      "step": 2100
+    },
+    {
+      "epoch": 0.8860759493670886,
+      "eval_loss": 0.8401098847389221,
+      "eval_runtime": 669.1149,
+      "eval_samples_per_second": 3.149,
+      "eval_steps_per_second": 3.149,
+      "step": 2100
+    },
+    {
+      "epoch": 0.8869198312236287,
+      "grad_norm": 0.4282098412513733,
+      "learning_rate": 0.00019733781881964242,
+      "loss": 0.9099814295768738,
+      "step": 2102
+    },
+    {
+      "epoch": 0.8877637130801688,
+      "grad_norm": 0.33882173895835876,
+      "learning_rate": 0.00019732679905396826,
+      "loss": 0.7458856105804443,
+      "step": 2104
+    },
+    {
+      "epoch": 0.8886075949367088,
+      "grad_norm": 0.43959516286849976,
+      "learning_rate": 0.00019731575683687078,
+      "loss": 0.7829693555831909,
+      "step": 2106
+    },
+    {
+      "epoch": 0.889451476793249,
+      "grad_norm": 0.40046483278274536,
+      "learning_rate": 0.00019730469217089716,
+      "loss": 0.7898997068405151,
+      "step": 2108
+    },
+    {
+      "epoch": 0.890295358649789,
+      "grad_norm": 0.40572553873062134,
+      "learning_rate": 0.00019729360505859984,
+      "loss": 0.7874186635017395,
+      "step": 2110
+    },
+    {
+      "epoch": 0.8911392405063291,
+      "grad_norm": 0.42086347937583923,
+      "learning_rate": 0.0001972824955025364,
+      "loss": 0.8499898910522461,
+      "step": 2112
+    },
+    {
+      "epoch": 0.8919831223628693,
+      "grad_norm": 0.38845229148864746,
+      "learning_rate": 0.00019727136350526956,
+      "loss": 0.7940409779548645,
+      "step": 2114
+    },
+    {
+      "epoch": 0.8928270042194093,
+      "grad_norm": 0.41575267910957336,
+      "learning_rate": 0.00019726020906936728,
+      "loss": 0.8279082775115967,
+      "step": 2116
+    },
+    {
+      "epoch": 0.8936708860759494,
+      "grad_norm": 0.37615057826042175,
+      "learning_rate": 0.0001972490321974027,
+      "loss": 0.8217757940292358,
+      "step": 2118
+    },
+    {
+      "epoch": 0.8945147679324894,
+      "grad_norm": 0.9318385124206543,
+      "learning_rate": 0.00019723783289195415,
+      "loss": 0.7637750506401062,
+      "step": 2120
+    },
+    {
+      "epoch": 0.8953586497890296,
+      "grad_norm": 0.41948962211608887,
+      "learning_rate": 0.000197226611155605,
+      "loss": 0.8702366948127747,
+      "step": 2122
+    },
+    {
+      "epoch": 0.8962025316455696,
+      "grad_norm": 0.4718417525291443,
+      "learning_rate": 0.00019721536699094392,
+      "loss": 0.8146984577178955,
+      "step": 2124
+    },
+    {
+      "epoch": 0.8970464135021097,
+      "grad_norm": 0.4470657408237457,
+      "learning_rate": 0.00019720410040056474,
+      "loss": 0.8646075129508972,
+      "step": 2126
+    },
+    {
+      "epoch": 0.8978902953586498,
+      "grad_norm": 0.4176540970802307,
+      "learning_rate": 0.00019719281138706643,
+      "loss": 0.7925992012023926,
+      "step": 2128
+    },
+    {
+      "epoch": 0.8987341772151899,
+      "grad_norm": 0.4854719042778015,
+      "learning_rate": 0.00019718149995305315,
+      "loss": 0.8142143487930298,
+      "step": 2130
+    },
+    {
+      "epoch": 0.8995780590717299,
+      "grad_norm": 0.4270440340042114,
+      "learning_rate": 0.00019717016610113427,
+      "loss": 0.8050890564918518,
+      "step": 2132
+    },
+    {
+      "epoch": 0.90042194092827,
+      "grad_norm": 0.37710970640182495,
+      "learning_rate": 0.00019715880983392422,
+      "loss": 0.7888354659080505,
+      "step": 2134
+    },
+    {
+      "epoch": 0.9012658227848102,
+      "grad_norm": 0.4181252717971802,
+      "learning_rate": 0.00019714743115404273,
+      "loss": 0.7348225712776184,
+      "step": 2136
+    },
+    {
+      "epoch": 0.9021097046413502,
+      "grad_norm": 0.4475359618663788,
+      "learning_rate": 0.0001971360300641146,
+      "loss": 0.7729623913764954,
+      "step": 2138
+    },
+    {
+      "epoch": 0.9029535864978903,
+      "grad_norm": 0.4052332937717438,
+      "learning_rate": 0.00019712460656676992,
+      "loss": 0.8066384792327881,
+      "step": 2140
+    },
+    {
+      "epoch": 0.9037974683544304,
+      "grad_norm": 0.4567393362522125,
+      "learning_rate": 0.0001971131606646438,
+      "loss": 0.8053154349327087,
+      "step": 2142
+    },
+    {
+      "epoch": 0.9046413502109705,
+      "grad_norm": 0.4304409325122833,
+      "learning_rate": 0.00019710169236037656,
+      "loss": 0.7242560386657715,
+      "step": 2144
+    },
+    {
+      "epoch": 0.9054852320675105,
+      "grad_norm": 0.4785008132457733,
+      "learning_rate": 0.0001970902016566138,
+      "loss": 0.8475301265716553,
+      "step": 2146
+    },
+    {
+      "epoch": 0.9063291139240506,
+      "grad_norm": 0.45226508378982544,
+      "learning_rate": 0.00019707868855600618,
+      "loss": 0.928544282913208,
+      "step": 2148
+    },
+    {
+      "epoch": 0.9071729957805907,
+      "grad_norm": 0.3794468641281128,
+      "learning_rate": 0.00019706715306120956,
+      "loss": 0.8135722875595093,
+      "step": 2150
+    },
+    {
+      "epoch": 0.9080168776371308,
+      "grad_norm": 0.4005141258239746,
+      "learning_rate": 0.0001970555951748849,
+      "loss": 0.7808290719985962,
+      "step": 2152
+    },
+    {
+      "epoch": 0.9088607594936708,
+      "grad_norm": 0.37929919362068176,
+      "learning_rate": 0.00019704401489969842,
+      "loss": 0.8292917013168335,
+      "step": 2154
+    },
+    {
+      "epoch": 0.909704641350211,
+      "grad_norm": 0.4458678066730499,
+      "learning_rate": 0.0001970324122383215,
+      "loss": 0.772884726524353,
+      "step": 2156
+    },
+    {
+      "epoch": 0.9105485232067511,
+      "grad_norm": 0.41907885670661926,
+      "learning_rate": 0.0001970207871934306,
+      "loss": 0.8204470872879028,
+      "step": 2158
+    },
+    {
+      "epoch": 0.9113924050632911,
+      "grad_norm": 0.40610212087631226,
+      "learning_rate": 0.00019700913976770741,
+      "loss": 0.8361451625823975,
+      "step": 2160
+    },
+    {
+      "epoch": 0.9122362869198313,
+      "grad_norm": 0.4213273227214813,
+      "learning_rate": 0.00019699746996383878,
+      "loss": 0.8456075191497803,
+      "step": 2162
+    },
+    {
+      "epoch": 0.9130801687763713,
+      "grad_norm": 0.49264612793922424,
+      "learning_rate": 0.0001969857777845167,
+      "loss": 0.8731603622436523,
+      "step": 2164
+    },
+    {
+      "epoch": 0.9139240506329114,
+      "grad_norm": 0.42002132534980774,
+      "learning_rate": 0.00019697406323243834,
+      "loss": 0.8149666786193848,
+      "step": 2166
+    },
+    {
+      "epoch": 0.9147679324894514,
+      "grad_norm": 0.37501421570777893,
+      "learning_rate": 0.000196962326310306,
+      "loss": 0.7508546710014343,
+      "step": 2168
+    },
+    {
+      "epoch": 0.9156118143459916,
+      "grad_norm": 0.3706159293651581,
+      "learning_rate": 0.00019695056702082717,
+      "loss": 0.830280601978302,
+      "step": 2170
+    },
+    {
+      "epoch": 0.9164556962025316,
+      "grad_norm": 0.3873467445373535,
+      "learning_rate": 0.00019693878536671452,
+      "loss": 0.7297167778015137,
+      "step": 2172
+    },
+    {
+      "epoch": 0.9172995780590717,
+      "grad_norm": 0.42876651883125305,
+      "learning_rate": 0.00019692698135068582,
+      "loss": 0.7924292087554932,
+      "step": 2174
+    },
+    {
+      "epoch": 0.9181434599156119,
+      "grad_norm": 0.4259159564971924,
+      "learning_rate": 0.00019691515497546406,
+      "loss": 0.8098247051239014,
+      "step": 2176
+    },
+    {
+      "epoch": 0.9189873417721519,
+      "grad_norm": 0.38152721524238586,
+      "learning_rate": 0.00019690330624377728,
+      "loss": 0.6789063215255737,
+      "step": 2178
+    },
+    {
+      "epoch": 0.919831223628692,
+      "grad_norm": 0.4135437607765198,
+      "learning_rate": 0.00019689143515835887,
+      "loss": 0.8582620620727539,
+      "step": 2180
+    },
+    {
+      "epoch": 0.920675105485232,
+      "grad_norm": 0.3818798065185547,
+      "learning_rate": 0.0001968795417219472,
+      "loss": 0.8059743642807007,
+      "step": 2182
+    },
+    {
+      "epoch": 0.9215189873417722,
+      "grad_norm": 0.39021286368370056,
+      "learning_rate": 0.00019686762593728582,
+      "loss": 0.7432259917259216,
+      "step": 2184
+    },
+    {
+      "epoch": 0.9223628691983122,
+      "grad_norm": 0.37864068150520325,
+      "learning_rate": 0.00019685568780712355,
+      "loss": 0.7561086416244507,
+      "step": 2186
+    },
+    {
+      "epoch": 0.9232067510548523,
+      "grad_norm": 0.450847864151001,
+      "learning_rate": 0.00019684372733421425,
+      "loss": 0.7552722692489624,
+      "step": 2188
+    },
+    {
+      "epoch": 0.9240506329113924,
+      "grad_norm": 0.37909507751464844,
+      "learning_rate": 0.00019683174452131695,
+      "loss": 0.6905336380004883,
+      "step": 2190
+    },
+    {
+      "epoch": 0.9248945147679325,
+      "grad_norm": 0.4439173638820648,
+      "learning_rate": 0.0001968197393711959,
+      "loss": 0.7710593938827515,
+      "step": 2192
+    },
+    {
+      "epoch": 0.9257383966244725,
+      "grad_norm": 0.3908376395702362,
+      "learning_rate": 0.00019680771188662044,
+      "loss": 0.8212814331054688,
+      "step": 2194
+    },
+    {
+      "epoch": 0.9265822784810127,
+      "grad_norm": 0.3528299033641815,
+      "learning_rate": 0.00019679566207036507,
+      "loss": 0.751355767250061,
+      "step": 2196
+    },
+    {
+      "epoch": 0.9274261603375528,
+      "grad_norm": 0.4351397156715393,
+      "learning_rate": 0.00019678358992520945,
+      "loss": 0.8204057812690735,
+      "step": 2198
+    },
+    {
+      "epoch": 0.9282700421940928,
+      "grad_norm": 0.45566943287849426,
+      "learning_rate": 0.0001967714954539384,
+      "loss": 0.763841450214386,
+      "step": 2200
+    },
+    {
+      "epoch": 0.9282700421940928,
+      "eval_loss": 0.8336610198020935,
+      "eval_runtime": 674.5134,
+      "eval_samples_per_second": 3.124,
+      "eval_steps_per_second": 3.124,
+      "step": 2200
+    },
+    {
+      "epoch": 0.9291139240506329,
+      "grad_norm": 0.4516221284866333,
+      "learning_rate": 0.00019675937865934188,
+      "loss": 0.861711323261261,
+      "step": 2202
+    },
+    {
+      "epoch": 0.929957805907173,
+      "grad_norm": 0.43421855568885803,
+      "learning_rate": 0.00019674723954421502,
+      "loss": 0.8062718510627747,
+      "step": 2204
+    },
+    {
+      "epoch": 0.9308016877637131,
+      "grad_norm": 0.42442476749420166,
+      "learning_rate": 0.00019673507811135803,
+      "loss": 0.8069695234298706,
+      "step": 2206
+    },
+    {
+      "epoch": 0.9316455696202531,
+      "grad_norm": 0.3976045250892639,
+      "learning_rate": 0.00019672289436357637,
+      "loss": 0.7336108684539795,
+      "step": 2208
+    },
+    {
+      "epoch": 0.9324894514767933,
+      "grad_norm": 0.4131270945072174,
+      "learning_rate": 0.00019671068830368058,
+      "loss": 0.7559890151023865,
+      "step": 2210
+    },
+    {
+      "epoch": 0.9333333333333333,
+      "grad_norm": 0.4599294364452362,
+      "learning_rate": 0.00019669845993448634,
+      "loss": 0.8447075486183167,
+      "step": 2212
+    },
+    {
+      "epoch": 0.9341772151898734,
+      "grad_norm": 0.4291204810142517,
+      "learning_rate": 0.00019668620925881455,
+      "loss": 0.8329233527183533,
+      "step": 2214
+    },
+    {
+      "epoch": 0.9350210970464135,
+      "grad_norm": 0.4226539731025696,
+      "learning_rate": 0.00019667393627949116,
+      "loss": 0.749232292175293,
+      "step": 2216
+    },
+    {
+      "epoch": 0.9358649789029536,
+      "grad_norm": 0.4159580171108246,
+      "learning_rate": 0.00019666164099934732,
+      "loss": 0.7470011115074158,
+      "step": 2218
+    },
+    {
+      "epoch": 0.9367088607594937,
+      "grad_norm": 0.4160766303539276,
+      "learning_rate": 0.00019664932342121935,
+      "loss": 0.8094202280044556,
+      "step": 2220
+    },
+    {
+      "epoch": 0.9375527426160337,
+      "grad_norm": 0.45284298062324524,
+      "learning_rate": 0.00019663698354794863,
+      "loss": 0.8398169875144958,
+      "step": 2222
+    },
+    {
+      "epoch": 0.9383966244725739,
+      "grad_norm": 0.3960540294647217,
+      "learning_rate": 0.00019662462138238178,
+      "loss": 0.7652443051338196,
+      "step": 2224
+    },
+    {
+      "epoch": 0.9392405063291139,
+      "grad_norm": 0.4269275367259979,
+      "learning_rate": 0.00019661223692737047,
+      "loss": 0.7909489870071411,
+      "step": 2226
+    },
+    {
+      "epoch": 0.940084388185654,
+      "grad_norm": 0.38187873363494873,
+      "learning_rate": 0.00019659983018577157,
+      "loss": 0.7821935415267944,
+      "step": 2228
+    },
+    {
+      "epoch": 0.9409282700421941,
+      "grad_norm": 0.45898866653442383,
+      "learning_rate": 0.0001965874011604471,
+      "loss": 0.7792721390724182,
+      "step": 2230
+    },
+    {
+      "epoch": 0.9417721518987342,
+      "grad_norm": 0.4740699827671051,
+      "learning_rate": 0.00019657494985426418,
+      "loss": 0.885725200176239,
+      "step": 2232
+    },
+    {
+      "epoch": 0.9426160337552743,
+      "grad_norm": 0.3912813365459442,
+      "learning_rate": 0.00019656247627009506,
+      "loss": 0.7898101806640625,
+      "step": 2234
+    },
+    {
+      "epoch": 0.9434599156118143,
+      "grad_norm": 0.4552418291568756,
+      "learning_rate": 0.0001965499804108172,
+      "loss": 0.8646330833435059,
+      "step": 2236
+    },
+    {
+      "epoch": 0.9443037974683545,
+      "grad_norm": 0.4178548753261566,
+      "learning_rate": 0.0001965374622793131,
+      "loss": 0.8068296313285828,
+      "step": 2238
+    },
+    {
+      "epoch": 0.9451476793248945,
+      "grad_norm": 0.38649076223373413,
+      "learning_rate": 0.00019652492187847057,
+      "loss": 0.7953271865844727,
+      "step": 2240
+    },
+    {
+      "epoch": 0.9459915611814346,
+      "grad_norm": 0.39785075187683105,
+      "learning_rate": 0.0001965123592111823,
+      "loss": 0.8423402309417725,
+      "step": 2242
+    },
+    {
+      "epoch": 0.9468354430379747,
+      "grad_norm": 0.47455263137817383,
+      "learning_rate": 0.00019649977428034632,
+      "loss": 0.8883589506149292,
+      "step": 2244
+    },
+    {
+      "epoch": 0.9476793248945148,
+      "grad_norm": 0.3866841197013855,
+      "learning_rate": 0.0001964871670888657,
+      "loss": 0.7232863903045654,
+      "step": 2246
+    },
+    {
+      "epoch": 0.9485232067510548,
+      "grad_norm": 0.3950609564781189,
+      "learning_rate": 0.00019647453763964876,
+      "loss": 0.7265347838401794,
+      "step": 2248
+    },
+    {
+      "epoch": 0.9493670886075949,
+      "grad_norm": 0.4315829575061798,
+      "learning_rate": 0.0001964618859356088,
+      "loss": 0.8000131249427795,
+      "step": 2250
+    },
+    {
+      "epoch": 0.950210970464135,
+      "grad_norm": 0.403984934091568,
+      "learning_rate": 0.00019644921197966433,
+      "loss": 0.823086678981781,
+      "step": 2252
+    },
+    {
+      "epoch": 0.9510548523206751,
+      "grad_norm": 0.4311080574989319,
+      "learning_rate": 0.00019643651577473897,
+      "loss": 0.8457553386688232,
+      "step": 2254
+    },
+    {
+      "epoch": 0.9518987341772152,
+      "grad_norm": 0.3884856104850769,
+      "learning_rate": 0.00019642379732376158,
+      "loss": 0.7866189479827881,
+      "step": 2256
+    },
+    {
+      "epoch": 0.9527426160337553,
+      "grad_norm": 0.3384067714214325,
+      "learning_rate": 0.00019641105662966595,
+      "loss": 0.7095813751220703,
+      "step": 2258
+    },
+    {
+      "epoch": 0.9535864978902954,
+      "grad_norm": 0.40426746010780334,
+      "learning_rate": 0.00019639829369539117,
+      "loss": 0.8271632194519043,
+      "step": 2260
+    },
+    {
+      "epoch": 0.9544303797468354,
+      "grad_norm": 0.45495131611824036,
+      "learning_rate": 0.00019638550852388144,
+      "loss": 0.8204861283302307,
+      "step": 2262
+    },
+    {
+      "epoch": 0.9552742616033755,
+      "grad_norm": 0.47201067209243774,
+      "learning_rate": 0.00019637270111808598,
+      "loss": 0.884763777256012,
+      "step": 2264
+    },
+    {
+      "epoch": 0.9561181434599156,
+      "grad_norm": 0.3959396779537201,
+      "learning_rate": 0.0001963598714809592,
+      "loss": 0.7630926370620728,
+      "step": 2266
+    },
+    {
+      "epoch": 0.9569620253164557,
+      "grad_norm": 0.4197136461734772,
+      "learning_rate": 0.00019634701961546077,
+      "loss": 0.797870397567749,
+      "step": 2268
+    },
+    {
+      "epoch": 0.9578059071729957,
+      "grad_norm": 0.381290078163147,
+      "learning_rate": 0.00019633414552455526,
+      "loss": 0.7755846977233887,
+      "step": 2270
+    },
+    {
+      "epoch": 0.9586497890295359,
+      "grad_norm": 0.3868124783039093,
+      "learning_rate": 0.00019632124921121255,
+      "loss": 0.7553264498710632,
+      "step": 2272
+    },
+    {
+      "epoch": 0.959493670886076,
+      "grad_norm": 0.4350631833076477,
+      "learning_rate": 0.00019630833067840747,
+      "loss": 0.7792160511016846,
+      "step": 2274
+    },
+    {
+      "epoch": 0.960337552742616,
+      "grad_norm": 0.4717157781124115,
+      "learning_rate": 0.00019629538992912016,
+      "loss": 0.878220796585083,
+      "step": 2276
+    },
+    {
+      "epoch": 0.9611814345991562,
+      "grad_norm": 0.45545122027397156,
+      "learning_rate": 0.00019628242696633584,
+      "loss": 0.8891014456748962,
+      "step": 2278
+    },
+    {
+      "epoch": 0.9620253164556962,
+      "grad_norm": 0.4421025514602661,
+      "learning_rate": 0.00019626944179304467,
+      "loss": 0.8024723529815674,
+      "step": 2280
+    },
+    {
+      "epoch": 0.9628691983122363,
+      "grad_norm": 0.40385812520980835,
+      "learning_rate": 0.00019625643441224222,
+      "loss": 0.8036280274391174,
+      "step": 2282
+    },
+    {
+      "epoch": 0.9637130801687763,
+      "grad_norm": 0.3866552412509918,
+      "learning_rate": 0.00019624340482692898,
+      "loss": 0.8072395324707031,
+      "step": 2284
+    },
+    {
+      "epoch": 0.9645569620253165,
+      "grad_norm": 0.4113233685493469,
+      "learning_rate": 0.0001962303530401106,
+      "loss": 0.8046114444732666,
+      "step": 2286
+    },
+    {
+      "epoch": 0.9654008438818565,
+      "grad_norm": 0.40720227360725403,
+      "learning_rate": 0.00019621727905479798,
+      "loss": 0.8434037566184998,
+      "step": 2288
+    },
+    {
+      "epoch": 0.9662447257383966,
+      "grad_norm": 0.35759273171424866,
+      "learning_rate": 0.00019620418287400693,
+      "loss": 0.7832419276237488,
+      "step": 2290
+    },
+    {
+      "epoch": 0.9670886075949368,
+      "grad_norm": 0.4170442223548889,
+      "learning_rate": 0.0001961910645007585,
+      "loss": 0.7975693345069885,
+      "step": 2292
+    },
+    {
+      "epoch": 0.9679324894514768,
+      "grad_norm": 0.4126700162887573,
+      "learning_rate": 0.00019617792393807895,
+      "loss": 0.8554494976997375,
+      "step": 2294
+    },
+    {
+      "epoch": 0.9687763713080169,
+      "grad_norm": 0.4231695830821991,
+      "learning_rate": 0.00019616476118899942,
+      "loss": 0.824539065361023,
+      "step": 2296
+    },
+    {
+      "epoch": 0.9696202531645569,
+      "grad_norm": 0.3556209206581116,
+      "learning_rate": 0.0001961515762565564,
+      "loss": 0.7008040547370911,
+      "step": 2298
+    },
+    {
+      "epoch": 0.9704641350210971,
+      "grad_norm": 0.4370400309562683,
+      "learning_rate": 0.00019613836914379132,
+      "loss": 0.8192558288574219,
+      "step": 2300
+    },
+    {
+      "epoch": 0.9704641350210971,
+      "eval_loss": 0.8281980156898499,
+      "eval_runtime": 688.6136,
+      "eval_samples_per_second": 3.06,
+      "eval_steps_per_second": 3.06,
+      "step": 2300
+    },
+    {
+      "epoch": 0.9713080168776371,
+      "grad_norm": 0.4207614064216614,
+      "learning_rate": 0.00019612513985375088,
+      "loss": 0.878928005695343,
+      "step": 2302
+    },
+    {
+      "epoch": 0.9721518987341772,
+      "grad_norm": 0.37607815861701965,
+      "learning_rate": 0.0001961118883894868,
+      "loss": 0.736182689666748,
+      "step": 2304
+    },
+    {
+      "epoch": 0.9729957805907173,
+      "grad_norm": 0.43311476707458496,
+      "learning_rate": 0.00019609861475405592,
+      "loss": 0.8403618335723877,
+      "step": 2306
+    },
+    {
+      "epoch": 0.9738396624472574,
+      "grad_norm": 0.4533176124095917,
+      "learning_rate": 0.00019608531895052021,
+      "loss": 0.8414564728736877,
+      "step": 2308
+    },
+    {
+      "epoch": 0.9746835443037974,
+      "grad_norm": 0.4025697410106659,
+      "learning_rate": 0.00019607200098194677,
+      "loss": 0.8105775117874146,
+      "step": 2310
+    },
+    {
+      "epoch": 0.9755274261603376,
+      "grad_norm": 0.37894198298454285,
+      "learning_rate": 0.00019605866085140784,
+      "loss": 0.8053821325302124,
+      "step": 2312
+    },
+    {
+      "epoch": 0.9763713080168777,
+      "grad_norm": 0.46265891194343567,
+      "learning_rate": 0.0001960452985619807,
+      "loss": 0.8092884421348572,
+      "step": 2314
+    },
+    {
+      "epoch": 0.9772151898734177,
+      "grad_norm": 0.3770054280757904,
+      "learning_rate": 0.0001960319141167477,
+      "loss": 0.7456095218658447,
+      "step": 2316
+    },
+    {
+      "epoch": 0.9780590717299578,
+      "grad_norm": 0.44084450602531433,
+      "learning_rate": 0.0001960185075187965,
+      "loss": 0.7739883661270142,
+      "step": 2318
+    },
+    {
+      "epoch": 0.9789029535864979,
+      "grad_norm": 0.3880341351032257,
+      "learning_rate": 0.00019600507877121965,
+      "loss": 0.6961302757263184,
+      "step": 2320
+    },
+    {
+      "epoch": 0.979746835443038,
+      "grad_norm": 0.468037486076355,
+      "learning_rate": 0.00019599162787711495,
+      "loss": 0.8447984457015991,
+      "step": 2322
+    },
+    {
+      "epoch": 0.980590717299578,
+      "grad_norm": 0.4123631417751312,
+      "learning_rate": 0.00019597815483958527,
+      "loss": 0.8155615329742432,
+      "step": 2324
+    },
+    {
+      "epoch": 0.9814345991561182,
+      "grad_norm": 0.3827875554561615,
+      "learning_rate": 0.00019596465966173853,
+      "loss": 0.7990952730178833,
+      "step": 2326
+    },
+    {
+      "epoch": 0.9822784810126582,
+      "grad_norm": 0.4644222557544708,
+      "learning_rate": 0.00019595114234668788,
+      "loss": 0.8899750709533691,
+      "step": 2328
+    },
+    {
+      "epoch": 0.9831223628691983,
+      "grad_norm": 0.43699249625205994,
+      "learning_rate": 0.00019593760289755144,
+      "loss": 0.7863160371780396,
+      "step": 2330
+    },
+    {
+      "epoch": 0.9839662447257383,
+      "grad_norm": 0.40713244676589966,
+      "learning_rate": 0.00019592404131745254,
+      "loss": 0.7960637807846069,
+      "step": 2332
+    },
+    {
+      "epoch": 0.9848101265822785,
+      "grad_norm": 0.37364068627357483,
+      "learning_rate": 0.00019591045760951957,
+      "loss": 0.714144229888916,
+      "step": 2334
+    },
+    {
+      "epoch": 0.9856540084388186,
+      "grad_norm": 0.4094736576080322,
+      "learning_rate": 0.00019589685177688599,
+      "loss": 0.7311791777610779,
+      "step": 2336
+    },
+    {
+      "epoch": 0.9864978902953586,
+      "grad_norm": 0.4275180697441101,
+      "learning_rate": 0.0001958832238226905,
+      "loss": 0.8191342353820801,
+      "step": 2338
+    },
+    {
+      "epoch": 0.9873417721518988,
+      "grad_norm": 0.4627651274204254,
+      "learning_rate": 0.00019586957375007668,
+      "loss": 0.7476106286048889,
+      "step": 2340
+    },
+    {
+      "epoch": 0.9881856540084388,
+      "grad_norm": 0.44676414132118225,
+      "learning_rate": 0.00019585590156219346,
+      "loss": 0.8257967233657837,
+      "step": 2342
+    },
+    {
+      "epoch": 0.9890295358649789,
+      "grad_norm": 0.40667134523391724,
+      "learning_rate": 0.00019584220726219466,
+      "loss": 0.7922079563140869,
+      "step": 2344
+    },
+    {
+      "epoch": 0.9898734177215189,
+      "grad_norm": 0.40865346789360046,
+      "learning_rate": 0.00019582849085323934,
+      "loss": 0.7892434597015381,
+      "step": 2346
+    },
+    {
+      "epoch": 0.9907172995780591,
+      "grad_norm": 0.43309709429740906,
+      "learning_rate": 0.00019581475233849163,
+      "loss": 0.8766644597053528,
+      "step": 2348
+    },
+    {
+      "epoch": 0.9915611814345991,
+      "grad_norm": 0.41558781266212463,
+      "learning_rate": 0.00019580099172112069,
+      "loss": 0.9183312058448792,
+      "step": 2350
+    },
+    {
+      "epoch": 0.9924050632911392,
+      "grad_norm": 0.432748019695282,
+      "learning_rate": 0.00019578720900430082,
+      "loss": 0.7905668020248413,
+      "step": 2352
+    },
+    {
+      "epoch": 0.9932489451476794,
+      "grad_norm": 0.4116985499858856,
+      "learning_rate": 0.0001957734041912115,
+      "loss": 0.8279712200164795,
+      "step": 2354
+    },
+    {
+      "epoch": 0.9940928270042194,
+      "grad_norm": 0.3927275240421295,
+      "learning_rate": 0.00019575957728503718,
+      "loss": 0.7451362609863281,
+      "step": 2356
+    },
+    {
+      "epoch": 0.9949367088607595,
+      "grad_norm": 0.3389509320259094,
+      "learning_rate": 0.0001957457282889675,
+      "loss": 0.6769390106201172,
+      "step": 2358
+    },
+    {
+      "epoch": 0.9957805907172996,
+      "grad_norm": 0.4220774471759796,
+      "learning_rate": 0.00019573185720619712,
+      "loss": 0.7590897083282471,
+      "step": 2360
+    },
+    {
+      "epoch": 0.9966244725738397,
+      "grad_norm": 0.4299897849559784,
+      "learning_rate": 0.00019571796403992583,
+      "loss": 0.8301284313201904,
+      "step": 2362
+    },
+    {
+      "epoch": 0.9974683544303797,
+      "grad_norm": 0.4031789302825928,
+      "learning_rate": 0.00019570404879335856,
+      "loss": 0.8019471764564514,
+      "step": 2364
+    },
+    {
+      "epoch": 0.9983122362869198,
+      "grad_norm": 0.3957567811012268,
+      "learning_rate": 0.00019569011146970526,
+      "loss": 0.7427027821540833,
+      "step": 2366
+    },
+    {
+      "epoch": 0.99915611814346,
+      "grad_norm": 0.41741347312927246,
+      "learning_rate": 0.000195676152072181,
+      "loss": 0.8221617341041565,
+      "step": 2368
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.47121551632881165,
+      "learning_rate": 0.00019566217060400595,
+      "loss": 0.8143455982208252,
+      "step": 2370
+    },
+    {
+      "epoch": 1.0008438818565402,
+      "grad_norm": 0.3854716122150421,
+      "learning_rate": 0.0001956481670684054,
+      "loss": 0.6971694231033325,
+      "step": 2372
+    },
+    {
+      "epoch": 1.00168776371308,
+      "grad_norm": 0.4029138386249542,
+      "learning_rate": 0.00019563414146860964,
+      "loss": 0.6719155311584473,
+      "step": 2374
+    },
+    {
+      "epoch": 1.0025316455696203,
+      "grad_norm": 0.40136656165122986,
+      "learning_rate": 0.00019562009380785412,
+      "loss": 0.7161872386932373,
+      "step": 2376
+    },
+    {
+      "epoch": 1.0033755274261604,
+      "grad_norm": 0.40685850381851196,
+      "learning_rate": 0.0001956060240893794,
+      "loss": 0.71699059009552,
+      "step": 2378
+    },
+    {
+      "epoch": 1.0042194092827004,
+      "grad_norm": 0.46153366565704346,
+      "learning_rate": 0.00019559193231643104,
+      "loss": 0.8030059337615967,
+      "step": 2380
+    },
+    {
+      "epoch": 1.0050632911392405,
+      "grad_norm": 0.45048823952674866,
+      "learning_rate": 0.0001955778184922598,
+      "loss": 0.7666479349136353,
+      "step": 2382
+    },
+    {
+      "epoch": 1.0059071729957807,
+      "grad_norm": 0.4260413348674774,
+      "learning_rate": 0.00019556368262012142,
+      "loss": 0.7481234073638916,
+      "step": 2384
+    },
+    {
+      "epoch": 1.0067510548523206,
+      "grad_norm": 0.4819575548171997,
+      "learning_rate": 0.0001955495247032768,
+      "loss": 0.7943820953369141,
+      "step": 2386
+    },
+    {
+      "epoch": 1.0075949367088608,
+      "grad_norm": 0.45789605379104614,
+      "learning_rate": 0.0001955353447449919,
+      "loss": 0.6401634812355042,
+      "step": 2388
+    },
+    {
+      "epoch": 1.0084388185654007,
+      "grad_norm": 0.37019097805023193,
+      "learning_rate": 0.00019552114274853778,
+      "loss": 0.7058886885643005,
+      "step": 2390
+    },
+    {
+      "epoch": 1.009282700421941,
+      "grad_norm": 0.44416263699531555,
+      "learning_rate": 0.0001955069187171905,
+      "loss": 0.7631242871284485,
+      "step": 2392
+    },
+    {
+      "epoch": 1.010126582278481,
+      "grad_norm": 0.4478379786014557,
+      "learning_rate": 0.00019549267265423138,
+      "loss": 0.7873892784118652,
+      "step": 2394
+    },
+    {
+      "epoch": 1.010970464135021,
+      "grad_norm": 0.4596130847930908,
+      "learning_rate": 0.00019547840456294658,
+      "loss": 0.7882059216499329,
+      "step": 2396
+    },
+    {
+      "epoch": 1.0118143459915612,
+      "grad_norm": 0.4062999188899994,
+      "learning_rate": 0.0001954641144466276,
+      "loss": 0.6358574628829956,
+      "step": 2398
+    },
+    {
+      "epoch": 1.0126582278481013,
+      "grad_norm": 0.47627049684524536,
+      "learning_rate": 0.0001954498023085708,
+      "loss": 0.7803886532783508,
+      "step": 2400
+    },
+    {
+      "epoch": 1.0126582278481013,
+      "eval_loss": 0.8250564932823181,
+      "eval_runtime": 691.5833,
+      "eval_samples_per_second": 3.047,
+      "eval_steps_per_second": 3.047,
+      "step": 2400
+    },
+    {
+      "epoch": 1.0135021097046413,
+      "grad_norm": 0.40690696239471436,
+      "learning_rate": 0.00019543546815207777,
+      "loss": 0.7394700050354004,
+      "step": 2402
+    },
+    {
+      "epoch": 1.0143459915611814,
+      "grad_norm": 0.45543980598449707,
+      "learning_rate": 0.0001954211119804551,
+      "loss": 0.7096555829048157,
+      "step": 2404
+    },
+    {
+      "epoch": 1.0151898734177216,
+      "grad_norm": 0.4367314279079437,
+      "learning_rate": 0.00019540673379701454,
+      "loss": 0.7961843013763428,
+      "step": 2406
+    },
+    {
+      "epoch": 1.0160337552742615,
+      "grad_norm": 0.4458340108394623,
+      "learning_rate": 0.00019539233360507278,
+      "loss": 0.743843138217926,
+      "step": 2408
+    },
+    {
+      "epoch": 1.0168776371308017,
+      "grad_norm": 0.3741893470287323,
+      "learning_rate": 0.0001953779114079517,
+      "loss": 0.7273756265640259,
+      "step": 2410
+    },
+    {
+      "epoch": 1.0177215189873419,
+      "grad_norm": 0.4340079426765442,
+      "learning_rate": 0.00019536346720897823,
+      "loss": 0.7922506928443909,
+      "step": 2412
+    },
+    {
+      "epoch": 1.0185654008438818,
+      "grad_norm": 0.4518018662929535,
+      "learning_rate": 0.00019534900101148437,
+      "loss": 0.6492037773132324,
+      "step": 2414
+    },
+    {
+      "epoch": 1.019409282700422,
+      "grad_norm": 0.4160727560520172,
+      "learning_rate": 0.00019533451281880716,
+      "loss": 0.7422950863838196,
+      "step": 2416
+    },
+    {
+      "epoch": 1.0202531645569621,
+      "grad_norm": 0.35953211784362793,
+      "learning_rate": 0.00019532000263428883,
+      "loss": 0.6489207148551941,
+      "step": 2418
+    },
+    {
+      "epoch": 1.021097046413502,
+      "grad_norm": 0.38774433732032776,
+      "learning_rate": 0.0001953054704612765,
+      "loss": 0.7449417114257812,
+      "step": 2420
+    },
+    {
+      "epoch": 1.0219409282700422,
+      "grad_norm": 0.48185858130455017,
+      "learning_rate": 0.0001952909163031225,
+      "loss": 0.8481494188308716,
+      "step": 2422
+    },
+    {
+      "epoch": 1.0227848101265822,
+      "grad_norm": 0.4098491072654724,
+      "learning_rate": 0.00019527634016318425,
+      "loss": 0.7442291378974915,
+      "step": 2424
+    },
+    {
+      "epoch": 1.0236286919831223,
+      "grad_norm": 0.41284480690956116,
+      "learning_rate": 0.00019526174204482412,
+      "loss": 0.7564480900764465,
+      "step": 2426
+    },
+    {
+      "epoch": 1.0244725738396625,
+      "grad_norm": 0.44948479533195496,
+      "learning_rate": 0.00019524712195140965,
+      "loss": 0.7234854102134705,
+      "step": 2428
+    },
+    {
+      "epoch": 1.0253164556962024,
+      "grad_norm": 0.39939233660697937,
+      "learning_rate": 0.0001952324798863134,
+      "loss": 0.7638729810714722,
+      "step": 2430
+    },
+    {
+      "epoch": 1.0261603375527426,
+      "grad_norm": 0.40807482600212097,
+      "learning_rate": 0.00019521781585291298,
+      "loss": 0.6922630071640015,
+      "step": 2432
+    },
+    {
+      "epoch": 1.0270042194092828,
+      "grad_norm": 0.45735791325569153,
+      "learning_rate": 0.00019520312985459116,
+      "loss": 0.762996256351471,
+      "step": 2434
+    },
+    {
+      "epoch": 1.0278481012658227,
+      "grad_norm": 0.4280409812927246,
+      "learning_rate": 0.0001951884218947357,
+      "loss": 0.7811596989631653,
+      "step": 2436
+    },
+    {
+      "epoch": 1.0286919831223629,
+      "grad_norm": 0.41957560181617737,
+      "learning_rate": 0.00019517369197673943,
+      "loss": 0.7278305292129517,
+      "step": 2438
+    },
+    {
+      "epoch": 1.029535864978903,
+      "grad_norm": 0.4832586646080017,
+      "learning_rate": 0.00019515894010400028,
+      "loss": 0.807456910610199,
+      "step": 2440
+    },
+    {
+      "epoch": 1.030379746835443,
+      "grad_norm": 0.46448788046836853,
+      "learning_rate": 0.00019514416627992122,
+      "loss": 0.7421302795410156,
+      "step": 2442
+    },
+    {
+      "epoch": 1.0312236286919831,
+      "grad_norm": 0.41362619400024414,
+      "learning_rate": 0.00019512937050791024,
+      "loss": 0.6892213225364685,
+      "step": 2444
+    },
+    {
+      "epoch": 1.0320675105485233,
+      "grad_norm": 0.4158109128475189,
+      "learning_rate": 0.0001951145527913805,
+      "loss": 0.7305331826210022,
+      "step": 2446
+    },
+    {
+      "epoch": 1.0329113924050632,
+      "grad_norm": 0.4725087881088257,
+      "learning_rate": 0.00019509971313375011,
+      "loss": 0.7682753205299377,
+      "step": 2448
+    },
+    {
+      "epoch": 1.0337552742616034,
+      "grad_norm": 0.4341438114643097,
+      "learning_rate": 0.00019508485153844237,
+      "loss": 0.747041642665863,
+      "step": 2450
+    },
+    {
+      "epoch": 1.0345991561181433,
+      "grad_norm": 0.44515377283096313,
+      "learning_rate": 0.00019506996800888548,
+      "loss": 0.7312062382698059,
+      "step": 2452
+    },
+    {
+      "epoch": 1.0354430379746835,
+      "grad_norm": 0.5313915014266968,
+      "learning_rate": 0.00019505506254851283,
+      "loss": 0.7861107587814331,
+      "step": 2454
+    },
+    {
+      "epoch": 1.0362869198312237,
+      "grad_norm": 0.6094246506690979,
+      "learning_rate": 0.00019504013516076285,
+      "loss": 0.8458115458488464,
+      "step": 2456
+    },
+    {
+      "epoch": 1.0371308016877636,
+      "grad_norm": 0.36419153213500977,
+      "learning_rate": 0.00019502518584907895,
+      "loss": 0.6091256141662598,
+      "step": 2458
+    },
+    {
+      "epoch": 1.0379746835443038,
+      "grad_norm": 0.399271696805954,
+      "learning_rate": 0.00019501021461690966,
+      "loss": 0.6703370809555054,
+      "step": 2460
+    },
+    {
+      "epoch": 1.038818565400844,
+      "grad_norm": 0.38560932874679565,
+      "learning_rate": 0.00019499522146770856,
+      "loss": 0.7220943570137024,
+      "step": 2462
+    },
+    {
+      "epoch": 1.0396624472573839,
+      "grad_norm": 0.44199803471565247,
+      "learning_rate": 0.00019498020640493428,
+      "loss": 0.7501406073570251,
+      "step": 2464
+    },
+    {
+      "epoch": 1.040506329113924,
+      "grad_norm": 0.47394421696662903,
+      "learning_rate": 0.00019496516943205054,
+      "loss": 0.7957925796508789,
+      "step": 2466
+    },
+    {
+      "epoch": 1.0413502109704642,
+      "grad_norm": 0.5274425745010376,
+      "learning_rate": 0.00019495011055252603,
+      "loss": 0.7964824438095093,
+      "step": 2468
+    },
+    {
+      "epoch": 1.0421940928270041,
+      "grad_norm": 0.44004881381988525,
+      "learning_rate": 0.0001949350297698346,
+      "loss": 0.7478582262992859,
+      "step": 2470
+    },
+    {
+      "epoch": 1.0430379746835443,
+      "grad_norm": 0.40662670135498047,
+      "learning_rate": 0.000194919927087455,
+      "loss": 0.738339900970459,
+      "step": 2472
+    },
+    {
+      "epoch": 1.0438818565400845,
+      "grad_norm": 0.43934181332588196,
+      "learning_rate": 0.00019490480250887123,
+      "loss": 0.817171037197113,
+      "step": 2474
+    },
+    {
+      "epoch": 1.0447257383966244,
+      "grad_norm": 0.4111221432685852,
+      "learning_rate": 0.00019488965603757223,
+      "loss": 0.7413169145584106,
+      "step": 2476
+    },
+    {
+      "epoch": 1.0455696202531646,
+      "grad_norm": 0.3770996630191803,
+      "learning_rate": 0.00019487448767705194,
+      "loss": 0.7893102169036865,
+      "step": 2478
+    },
+    {
+      "epoch": 1.0464135021097047,
+      "grad_norm": 0.4524461627006531,
+      "learning_rate": 0.00019485929743080945,
+      "loss": 0.7037841081619263,
+      "step": 2480
+    },
+    {
+      "epoch": 1.0472573839662447,
+      "grad_norm": 0.4232378900051117,
+      "learning_rate": 0.00019484408530234885,
+      "loss": 0.7456525564193726,
+      "step": 2482
+    },
+    {
+      "epoch": 1.0481012658227848,
+      "grad_norm": 0.3842810392379761,
+      "learning_rate": 0.00019482885129517929,
+      "loss": 0.7263808250427246,
+      "step": 2484
+    },
+    {
+      "epoch": 1.048945147679325,
+      "grad_norm": 0.43203726410865784,
+      "learning_rate": 0.00019481359541281495,
+      "loss": 0.7101474404335022,
+      "step": 2486
+    },
+    {
+      "epoch": 1.049789029535865,
+      "grad_norm": 0.46944454312324524,
+      "learning_rate": 0.00019479831765877509,
+      "loss": 0.8561921715736389,
+      "step": 2488
+    },
+    {
+      "epoch": 1.0506329113924051,
+      "grad_norm": 0.4455019533634186,
+      "learning_rate": 0.00019478301803658397,
+      "loss": 0.7206679582595825,
+      "step": 2490
+    },
+    {
+      "epoch": 1.051476793248945,
+      "grad_norm": 0.4253813326358795,
+      "learning_rate": 0.00019476769654977095,
+      "loss": 0.7271745800971985,
+      "step": 2492
+    },
+    {
+      "epoch": 1.0523206751054852,
+      "grad_norm": 0.42758429050445557,
+      "learning_rate": 0.00019475235320187033,
+      "loss": 0.7358576059341431,
+      "step": 2494
+    },
+    {
+      "epoch": 1.0531645569620254,
+      "grad_norm": 0.400634765625,
+      "learning_rate": 0.00019473698799642164,
+      "loss": 0.6617647409439087,
+      "step": 2496
+    },
+    {
+      "epoch": 1.0540084388185653,
+      "grad_norm": 0.4132104814052582,
+      "learning_rate": 0.00019472160093696926,
+      "loss": 0.6900970935821533,
+      "step": 2498
+    },
+    {
+      "epoch": 1.0548523206751055,
+      "grad_norm": 0.5116495490074158,
+      "learning_rate": 0.00019470619202706268,
+      "loss": 0.7352861762046814,
+      "step": 2500
+    },
+    {
+      "epoch": 1.0548523206751055,
+      "eval_loss": 0.8249453902244568,
+      "eval_runtime": 679.4446,
+      "eval_samples_per_second": 3.101,
+      "eval_steps_per_second": 3.101,
+      "step": 2500
+    },
+    {
+      "epoch": 1.0556962025316456,
+      "grad_norm": 0.42822444438934326,
+      "learning_rate": 0.0001946907612702565,
+      "loss": 0.7583479881286621,
+      "step": 2502
+    },
+    {
+      "epoch": 1.0565400843881856,
+      "grad_norm": 0.4788980782032013,
+      "learning_rate": 0.00019467530867011025,
+      "loss": 0.777511477470398,
+      "step": 2504
+    },
+    {
+      "epoch": 1.0573839662447257,
+      "grad_norm": 0.4210498034954071,
+      "learning_rate": 0.00019465983423018856,
+      "loss": 0.7452852129936218,
+      "step": 2506
+    },
+    {
+      "epoch": 1.058227848101266,
+      "grad_norm": 0.43219906091690063,
+      "learning_rate": 0.0001946443379540611,
+      "loss": 0.7540958523750305,
+      "step": 2508
+    },
+    {
+      "epoch": 1.0590717299578059,
+      "grad_norm": 0.4361002445220947,
+      "learning_rate": 0.00019462881984530253,
+      "loss": 0.7175624966621399,
+      "step": 2510
+    },
+    {
+      "epoch": 1.059915611814346,
+      "grad_norm": 0.4575091600418091,
+      "learning_rate": 0.00019461327990749264,
+      "loss": 0.7342076301574707,
+      "step": 2512
+    },
+    {
+      "epoch": 1.0607594936708862,
+      "grad_norm": 0.4823324978351593,
+      "learning_rate": 0.00019459771814421615,
+      "loss": 0.8255488872528076,
+      "step": 2514
+    },
+    {
+      "epoch": 1.0616033755274261,
+      "grad_norm": 0.41693785786628723,
+      "learning_rate": 0.00019458213455906284,
+      "loss": 0.6624729633331299,
+      "step": 2516
+    },
+    {
+      "epoch": 1.0624472573839663,
+      "grad_norm": 0.41319167613983154,
+      "learning_rate": 0.00019456652915562762,
+      "loss": 0.8224266767501831,
+      "step": 2518
+    },
+    {
+      "epoch": 1.0632911392405062,
+      "grad_norm": 0.4405190050601959,
+      "learning_rate": 0.00019455090193751023,
+      "loss": 0.7582900524139404,
+      "step": 2520
+    },
+    {
+      "epoch": 1.0641350210970464,
+      "grad_norm": 0.4564387798309326,
+      "learning_rate": 0.0001945352529083157,
+      "loss": 0.7948101162910461,
+      "step": 2522
+    },
+    {
+      "epoch": 1.0649789029535865,
+      "grad_norm": 0.4247225224971771,
+      "learning_rate": 0.00019451958207165394,
+      "loss": 0.7670508027076721,
+      "step": 2524
+    },
+    {
+      "epoch": 1.0658227848101265,
+      "grad_norm": 0.42070862650871277,
+      "learning_rate": 0.00019450388943113983,
+      "loss": 0.7753698229789734,
+      "step": 2526
+    },
+    {
+      "epoch": 1.0666666666666667,
+      "grad_norm": 0.3729179799556732,
+      "learning_rate": 0.00019448817499039342,
+      "loss": 0.710904598236084,
+      "step": 2528
+    },
+    {
+      "epoch": 1.0675105485232068,
+      "grad_norm": 0.40901753306388855,
+      "learning_rate": 0.0001944724387530397,
+      "loss": 0.7000213265419006,
+      "step": 2530
+    },
+    {
+      "epoch": 1.0683544303797468,
+      "grad_norm": 0.4481421113014221,
+      "learning_rate": 0.00019445668072270877,
+      "loss": 0.7374410629272461,
+      "step": 2532
+    },
+    {
+      "epoch": 1.069198312236287,
+      "grad_norm": 0.4839915335178375,
+      "learning_rate": 0.00019444090090303567,
+      "loss": 0.746309220790863,
+      "step": 2534
+    },
+    {
+      "epoch": 1.070042194092827,
+      "grad_norm": 0.4982132613658905,
+      "learning_rate": 0.00019442509929766048,
+      "loss": 0.7448083162307739,
+      "step": 2536
+    },
+    {
+      "epoch": 1.070886075949367,
+      "grad_norm": 0.3987475037574768,
+      "learning_rate": 0.00019440927591022838,
+      "loss": 0.7052555680274963,
+      "step": 2538
+    },
+    {
+      "epoch": 1.0717299578059072,
+      "grad_norm": 0.4089896082878113,
+      "learning_rate": 0.00019439343074438944,
+      "loss": 0.7090281248092651,
+      "step": 2540
+    },
+    {
+      "epoch": 1.0725738396624473,
+      "grad_norm": 0.3703136146068573,
+      "learning_rate": 0.00019437756380379895,
+      "loss": 0.7530601024627686,
+      "step": 2542
+    },
+    {
+      "epoch": 1.0734177215189873,
+      "grad_norm": 0.4015701413154602,
+      "learning_rate": 0.00019436167509211702,
+      "loss": 0.7597941160202026,
+      "step": 2544
+    },
+    {
+      "epoch": 1.0742616033755275,
+      "grad_norm": 0.3915148675441742,
+      "learning_rate": 0.00019434576461300888,
+      "loss": 0.7168124318122864,
+      "step": 2546
+    },
+    {
+      "epoch": 1.0751054852320676,
+      "grad_norm": 0.44643113017082214,
+      "learning_rate": 0.00019432983237014482,
+      "loss": 0.7762725949287415,
+      "step": 2548
+    },
+    {
+      "epoch": 1.0759493670886076,
+      "grad_norm": 0.46874719858169556,
+      "learning_rate": 0.00019431387836720004,
+      "loss": 0.7879725098609924,
+      "step": 2550
+    },
+    {
+      "epoch": 1.0767932489451477,
+      "grad_norm": 0.4143001437187195,
+      "learning_rate": 0.00019429790260785487,
+      "loss": 0.7351391315460205,
+      "step": 2552
+    },
+    {
+      "epoch": 1.0776371308016879,
+      "grad_norm": 0.4569888412952423,
+      "learning_rate": 0.0001942819050957946,
+      "loss": 0.86613929271698,
+      "step": 2554
+    },
+    {
+      "epoch": 1.0784810126582278,
+      "grad_norm": 0.41729897260665894,
+      "learning_rate": 0.00019426588583470954,
+      "loss": 0.7122719287872314,
+      "step": 2556
+    },
+    {
+      "epoch": 1.079324894514768,
+      "grad_norm": 0.42295604944229126,
+      "learning_rate": 0.000194249844828295,
+      "loss": 0.6605332493782043,
+      "step": 2558
+    },
+    {
+      "epoch": 1.080168776371308,
+      "grad_norm": 0.45787641406059265,
+      "learning_rate": 0.00019423378208025137,
+      "loss": 0.7839207649230957,
+      "step": 2560
+    },
+    {
+      "epoch": 1.081012658227848,
+      "grad_norm": 0.43370622396469116,
+      "learning_rate": 0.00019421769759428404,
+      "loss": 0.7557987570762634,
+      "step": 2562
+    },
+    {
+      "epoch": 1.0818565400843883,
+      "grad_norm": 0.4371975362300873,
+      "learning_rate": 0.00019420159137410335,
+      "loss": 0.7349345088005066,
+      "step": 2564
+    },
+    {
+      "epoch": 1.0827004219409282,
+      "grad_norm": 0.468730628490448,
+      "learning_rate": 0.0001941854634234247,
+      "loss": 0.7024962902069092,
+      "step": 2566
+    },
+    {
+      "epoch": 1.0835443037974684,
+      "grad_norm": 0.4257284998893738,
+      "learning_rate": 0.0001941693137459685,
+      "loss": 0.6981064081192017,
+      "step": 2568
+    },
+    {
+      "epoch": 1.0843881856540085,
+      "grad_norm": 0.4322902262210846,
+      "learning_rate": 0.00019415314234546015,
+      "loss": 0.7899637818336487,
+      "step": 2570
+    },
+    {
+      "epoch": 1.0852320675105485,
+      "grad_norm": 0.44384217262268066,
+      "learning_rate": 0.00019413694922563014,
+      "loss": 0.7964152693748474,
+      "step": 2572
+    },
+    {
+      "epoch": 1.0860759493670886,
+      "grad_norm": 0.47391360998153687,
+      "learning_rate": 0.0001941207343902139,
+      "loss": 0.7902941107749939,
+      "step": 2574
+    },
+    {
+      "epoch": 1.0869198312236288,
+      "grad_norm": 0.4575883746147156,
+      "learning_rate": 0.00019410449784295182,
+      "loss": 0.7818595170974731,
+      "step": 2576
+    },
+    {
+      "epoch": 1.0877637130801687,
+      "grad_norm": 0.41236087679862976,
+      "learning_rate": 0.0001940882395875894,
+      "loss": 0.7516021728515625,
+      "step": 2578
+    },
+    {
+      "epoch": 1.0886075949367089,
+      "grad_norm": 0.3817494213581085,
+      "learning_rate": 0.0001940719596278771,
+      "loss": 0.6870401501655579,
+      "step": 2580
+    },
+    {
+      "epoch": 1.0894514767932488,
+      "grad_norm": 0.4630168378353119,
+      "learning_rate": 0.00019405565796757043,
+      "loss": 0.7437341809272766,
+      "step": 2582
+    },
+    {
+      "epoch": 1.090295358649789,
+      "grad_norm": 0.4063820540904999,
+      "learning_rate": 0.00019403933461042982,
+      "loss": 0.7101680040359497,
+      "step": 2584
+    },
+    {
+      "epoch": 1.0911392405063292,
+      "grad_norm": 0.44423359632492065,
+      "learning_rate": 0.0001940229895602208,
+      "loss": 0.6894780397415161,
+      "step": 2586
+    },
+    {
+      "epoch": 1.091983122362869,
+      "grad_norm": 0.4267353117465973,
+      "learning_rate": 0.0001940066228207138,
+      "loss": 0.7277120351791382,
+      "step": 2588
+    },
+    {
+      "epoch": 1.0928270042194093,
+      "grad_norm": 0.44747647643089294,
+      "learning_rate": 0.00019399023439568434,
+      "loss": 0.6512353420257568,
+      "step": 2590
+    },
+    {
+      "epoch": 1.0936708860759494,
+      "grad_norm": 0.4557100832462311,
+      "learning_rate": 0.00019397382428891296,
+      "loss": 0.7850068807601929,
+      "step": 2592
+    },
+    {
+      "epoch": 1.0945147679324894,
+      "grad_norm": 0.44980114698410034,
+      "learning_rate": 0.0001939573925041851,
+      "loss": 0.786868691444397,
+      "step": 2594
+    },
+    {
+      "epoch": 1.0953586497890295,
+      "grad_norm": 0.45374399423599243,
+      "learning_rate": 0.00019394093904529126,
+      "loss": 0.8260899186134338,
+      "step": 2596
+    },
+    {
+      "epoch": 1.0962025316455697,
+      "grad_norm": 0.40681353211402893,
+      "learning_rate": 0.00019392446391602695,
+      "loss": 0.7056689262390137,
+      "step": 2598
+    },
+    {
+      "epoch": 1.0970464135021096,
+      "grad_norm": 0.39278075098991394,
+      "learning_rate": 0.0001939079671201927,
+      "loss": 0.7557252645492554,
+      "step": 2600
+    },
+    {
+      "epoch": 1.0970464135021096,
+      "eval_loss": 0.8211485743522644,
+      "eval_runtime": 670.2276,
+      "eval_samples_per_second": 3.144,
+      "eval_steps_per_second": 3.144,
+      "step": 2600
+    },
+    {
+      "epoch": 1.0978902953586498,
+      "grad_norm": 0.4261230230331421,
+      "learning_rate": 0.00019389144866159398,
+      "loss": 0.7139819264411926,
+      "step": 2602
+    },
+    {
+      "epoch": 1.09873417721519,
+      "grad_norm": 0.4664439558982849,
+      "learning_rate": 0.0001938749085440412,
+      "loss": 0.7274705767631531,
+      "step": 2604
+    },
+    {
+      "epoch": 1.09957805907173,
+      "grad_norm": 0.46306154131889343,
+      "learning_rate": 0.00019385834677134998,
+      "loss": 0.7547051906585693,
+      "step": 2606
+    },
+    {
+      "epoch": 1.10042194092827,
+      "grad_norm": 0.39753687381744385,
+      "learning_rate": 0.00019384176334734073,
+      "loss": 0.6942694187164307,
+      "step": 2608
+    },
+    {
+      "epoch": 1.1012658227848102,
+      "grad_norm": 0.4556522071361542,
+      "learning_rate": 0.00019382515827583898,
+      "loss": 0.7161701321601868,
+      "step": 2610
+    },
+    {
+      "epoch": 1.1021097046413502,
+      "grad_norm": 0.3863290250301361,
+      "learning_rate": 0.0001938085315606751,
+      "loss": 0.7165710926055908,
+      "step": 2612
+    },
+    {
+      "epoch": 1.1029535864978903,
+      "grad_norm": 0.49601292610168457,
+      "learning_rate": 0.00019379188320568466,
+      "loss": 0.8080679774284363,
+      "step": 2614
+    },
+    {
+      "epoch": 1.1037974683544305,
+      "grad_norm": 0.4498465955257416,
+      "learning_rate": 0.00019377521321470805,
+      "loss": 0.7032502889633179,
+      "step": 2616
+    },
+    {
+      "epoch": 1.1046413502109704,
+      "grad_norm": 0.4537963569164276,
+      "learning_rate": 0.00019375852159159073,
+      "loss": 0.7076731324195862,
+      "step": 2618
+    },
+    {
+      "epoch": 1.1054852320675106,
+      "grad_norm": 0.4042968153953552,
+      "learning_rate": 0.0001937418083401832,
+      "loss": 0.7020825743675232,
+      "step": 2620
+    },
+    {
+      "epoch": 1.1063291139240505,
+      "grad_norm": 0.44306516647338867,
+      "learning_rate": 0.00019372507346434076,
+      "loss": 0.7222874760627747,
+      "step": 2622
+    },
+    {
+      "epoch": 1.1071729957805907,
+      "grad_norm": 0.3792516887187958,
+      "learning_rate": 0.00019370831696792392,
+      "loss": 0.6136211156845093,
+      "step": 2624
+    },
+    {
+      "epoch": 1.1080168776371309,
+      "grad_norm": 0.46655353903770447,
+      "learning_rate": 0.00019369153885479806,
+      "loss": 0.7358977198600769,
+      "step": 2626
+    },
+    {
+      "epoch": 1.1088607594936708,
+      "grad_norm": 0.4416999816894531,
+      "learning_rate": 0.0001936747391288336,
+      "loss": 0.7484551072120667,
+      "step": 2628
+    },
+    {
+      "epoch": 1.109704641350211,
+      "grad_norm": 0.4620150327682495,
+      "learning_rate": 0.00019365791779390584,
+      "loss": 0.6874802708625793,
+      "step": 2630
+    },
+    {
+      "epoch": 1.1105485232067511,
+      "grad_norm": 0.4357883334159851,
+      "learning_rate": 0.00019364107485389518,
+      "loss": 0.7168865203857422,
+      "step": 2632
+    },
+    {
+      "epoch": 1.111392405063291,
+      "grad_norm": 0.4581143856048584,
+      "learning_rate": 0.00019362421031268699,
+      "loss": 0.7533737421035767,
+      "step": 2634
+    },
+    {
+      "epoch": 1.1122362869198312,
+      "grad_norm": 0.46408653259277344,
+      "learning_rate": 0.00019360732417417152,
+      "loss": 0.8245398998260498,
+      "step": 2636
+    },
+    {
+      "epoch": 1.1130801687763714,
+      "grad_norm": 0.4639282524585724,
+      "learning_rate": 0.00019359041644224415,
+      "loss": 0.7095663547515869,
+      "step": 2638
+    },
+    {
+      "epoch": 1.1139240506329113,
+      "grad_norm": 0.4598273038864136,
+      "learning_rate": 0.00019357348712080517,
+      "loss": 0.7030285596847534,
+      "step": 2640
+    },
+    {
+      "epoch": 1.1147679324894515,
+      "grad_norm": 0.42993804812431335,
+      "learning_rate": 0.00019355653621375978,
+      "loss": 0.7158969640731812,
+      "step": 2642
+    },
+    {
+      "epoch": 1.1156118143459917,
+      "grad_norm": 0.36132168769836426,
+      "learning_rate": 0.0001935395637250183,
+      "loss": 0.6800820231437683,
+      "step": 2644
+    },
+    {
+      "epoch": 1.1164556962025316,
+      "grad_norm": 0.5004428029060364,
+      "learning_rate": 0.00019352256965849592,
+      "loss": 0.694972813129425,
+      "step": 2646
+    },
+    {
+      "epoch": 1.1172995780590718,
+      "grad_norm": 0.43715354800224304,
+      "learning_rate": 0.00019350555401811286,
+      "loss": 0.7164262533187866,
+      "step": 2648
+    },
+    {
+      "epoch": 1.1181434599156117,
+      "grad_norm": 0.42316755652427673,
+      "learning_rate": 0.00019348851680779433,
+      "loss": 0.7026737928390503,
+      "step": 2650
+    },
+    {
+      "epoch": 1.1189873417721519,
+      "grad_norm": 0.35848063230514526,
+      "learning_rate": 0.0001934714580314704,
+      "loss": 0.6774232387542725,
+      "step": 2652
+    },
+    {
+      "epoch": 1.119831223628692,
+      "grad_norm": 0.46613892912864685,
+      "learning_rate": 0.00019345437769307627,
+      "loss": 0.7789244055747986,
+      "step": 2654
+    },
+    {
+      "epoch": 1.120675105485232,
+      "grad_norm": 0.4298146367073059,
+      "learning_rate": 0.00019343727579655204,
+      "loss": 0.8647177815437317,
+      "step": 2656
+    },
+    {
+      "epoch": 1.1215189873417721,
+      "grad_norm": 0.4638161361217499,
+      "learning_rate": 0.00019342015234584276,
+      "loss": 0.6949124336242676,
+      "step": 2658
+    },
+    {
+      "epoch": 1.1223628691983123,
+      "grad_norm": 0.4079605042934418,
+      "learning_rate": 0.00019340300734489854,
+      "loss": 0.6677125096321106,
+      "step": 2660
+    },
+    {
+      "epoch": 1.1232067510548522,
+      "grad_norm": 0.46079063415527344,
+      "learning_rate": 0.00019338584079767434,
+      "loss": 0.7521067261695862,
+      "step": 2662
+    },
+    {
+      "epoch": 1.1240506329113924,
+      "grad_norm": 0.3935202956199646,
+      "learning_rate": 0.0001933686527081302,
+      "loss": 0.6829184889793396,
+      "step": 2664
+    },
+    {
+      "epoch": 1.1248945147679326,
+      "grad_norm": 0.45696479082107544,
+      "learning_rate": 0.000193351443080231,
+      "loss": 0.7122842669487,
+      "step": 2666
+    },
+    {
+      "epoch": 1.1257383966244725,
+      "grad_norm": 0.4078282415866852,
+      "learning_rate": 0.0001933342119179468,
+      "loss": 0.7316977977752686,
+      "step": 2668
+    },
+    {
+      "epoch": 1.1265822784810127,
+      "grad_norm": 0.4282447099685669,
+      "learning_rate": 0.0001933169592252524,
+      "loss": 0.7323395013809204,
+      "step": 2670
+    },
+    {
+      "epoch": 1.1274261603375528,
+      "grad_norm": 0.46398648619651794,
+      "learning_rate": 0.00019329968500612766,
+      "loss": 0.7177348732948303,
+      "step": 2672
+    },
+    {
+      "epoch": 1.1282700421940928,
+      "grad_norm": 0.4560202658176422,
+      "learning_rate": 0.0001932823892645575,
+      "loss": 0.7957183122634888,
+      "step": 2674
+    },
+    {
+      "epoch": 1.129113924050633,
+      "grad_norm": 0.4790743887424469,
+      "learning_rate": 0.00019326507200453162,
+      "loss": 0.762060284614563,
+      "step": 2676
+    },
+    {
+      "epoch": 1.129957805907173,
+      "grad_norm": 0.46524202823638916,
+      "learning_rate": 0.00019324773323004484,
+      "loss": 0.7227895259857178,
+      "step": 2678
+    },
+    {
+      "epoch": 1.130801687763713,
+      "grad_norm": 0.47912222146987915,
+      "learning_rate": 0.00019323037294509683,
+      "loss": 0.6968042254447937,
+      "step": 2680
+    },
+    {
+      "epoch": 1.1316455696202532,
+      "grad_norm": 0.4412849247455597,
+      "learning_rate": 0.00019321299115369233,
+      "loss": 0.6682016253471375,
+      "step": 2682
+    },
+    {
+      "epoch": 1.1324894514767934,
+      "grad_norm": 0.4138370752334595,
+      "learning_rate": 0.00019319558785984095,
+      "loss": 0.7287812829017639,
+      "step": 2684
+    },
+    {
+      "epoch": 1.1333333333333333,
+      "grad_norm": 0.391170859336853,
+      "learning_rate": 0.00019317816306755733,
+      "loss": 0.6547526121139526,
+      "step": 2686
+    },
+    {
+      "epoch": 1.1341772151898735,
+      "grad_norm": 0.45472806692123413,
+      "learning_rate": 0.00019316071678086098,
+      "loss": 0.6942042708396912,
+      "step": 2688
+    },
+    {
+      "epoch": 1.1350210970464134,
+      "grad_norm": 0.40321481227874756,
+      "learning_rate": 0.00019314324900377648,
+      "loss": 0.7417841553688049,
+      "step": 2690
+    },
+    {
+      "epoch": 1.1358649789029536,
+      "grad_norm": 0.3806784152984619,
+      "learning_rate": 0.00019312575974033327,
+      "loss": 0.7780525088310242,
+      "step": 2692
+    },
+    {
+      "epoch": 1.1367088607594937,
+      "grad_norm": 0.43763765692710876,
+      "learning_rate": 0.0001931082489945658,
+      "loss": 0.661472499370575,
+      "step": 2694
+    },
+    {
+      "epoch": 1.1375527426160337,
+      "grad_norm": 0.45410826802253723,
+      "learning_rate": 0.00019309071677051348,
+      "loss": 0.7443127632141113,
+      "step": 2696
+    },
+    {
+      "epoch": 1.1383966244725738,
+      "grad_norm": 0.47907447814941406,
+      "learning_rate": 0.00019307316307222062,
+      "loss": 0.744393527507782,
+      "step": 2698
+    },
+    {
+      "epoch": 1.139240506329114,
+      "grad_norm": 0.41723814606666565,
+      "learning_rate": 0.00019305558790373654,
+      "loss": 0.7042794227600098,
+      "step": 2700
+    },
+    {
+      "epoch": 1.139240506329114,
+      "eval_loss": 0.8155058026313782,
+      "eval_runtime": 678.284,
+      "eval_samples_per_second": 3.106,
+      "eval_steps_per_second": 3.106,
+      "step": 2700
+    },
+    {
+      "epoch": 1.140084388185654,
+      "grad_norm": 0.3863268196582794,
+      "learning_rate": 0.0001930379912691155,
+      "loss": 0.6715745329856873,
+      "step": 2702
+    },
+    {
+      "epoch": 1.140928270042194,
+      "grad_norm": 0.49994897842407227,
+      "learning_rate": 0.00019302037317241673,
+      "loss": 0.6959030628204346,
+      "step": 2704
+    },
+    {
+      "epoch": 1.1417721518987343,
+      "grad_norm": 0.46377572417259216,
+      "learning_rate": 0.00019300273361770432,
+      "loss": 0.6534222364425659,
+      "step": 2706
+    },
+    {
+      "epoch": 1.1426160337552742,
+      "grad_norm": 0.3841579556465149,
+      "learning_rate": 0.00019298507260904745,
+      "loss": 0.6750475764274597,
+      "step": 2708
+    },
+    {
+      "epoch": 1.1434599156118144,
+      "grad_norm": 0.4171544313430786,
+      "learning_rate": 0.00019296739015052015,
+      "loss": 0.7714900374412537,
+      "step": 2710
+    },
+    {
+      "epoch": 1.1443037974683543,
+      "grad_norm": 0.44384199380874634,
+      "learning_rate": 0.00019294968624620137,
+      "loss": 0.7281306982040405,
+      "step": 2712
+    },
+    {
+      "epoch": 1.1451476793248945,
+      "grad_norm": 0.43345192074775696,
+      "learning_rate": 0.00019293196090017513,
+      "loss": 0.7407262325286865,
+      "step": 2714
+    },
+    {
+      "epoch": 1.1459915611814346,
+      "grad_norm": 0.3897020220756531,
+      "learning_rate": 0.00019291421411653034,
+      "loss": 0.6794347167015076,
+      "step": 2716
+    },
+    {
+      "epoch": 1.1468354430379746,
+      "grad_norm": 0.4289143681526184,
+      "learning_rate": 0.00019289644589936075,
+      "loss": 0.7026289105415344,
+      "step": 2718
+    },
+    {
+      "epoch": 1.1476793248945147,
+      "grad_norm": 0.4783557057380676,
+      "learning_rate": 0.0001928786562527652,
+      "loss": 0.7613762617111206,
+      "step": 2720
+    },
+    {
+      "epoch": 1.148523206751055,
+      "grad_norm": 0.4152112901210785,
+      "learning_rate": 0.00019286084518084745,
+      "loss": 0.7123059630393982,
+      "step": 2722
+    },
+    {
+      "epoch": 1.1493670886075948,
+      "grad_norm": 0.4204908013343811,
+      "learning_rate": 0.0001928430126877161,
+      "loss": 0.7198041677474976,
+      "step": 2724
+    },
+    {
+      "epoch": 1.150210970464135,
+      "grad_norm": 0.41333863139152527,
+      "learning_rate": 0.00019282515877748485,
+      "loss": 0.6859839558601379,
+      "step": 2726
+    },
+    {
+      "epoch": 1.1510548523206752,
+      "grad_norm": 0.4041498005390167,
+      "learning_rate": 0.00019280728345427218,
+      "loss": 0.7185758948326111,
+      "step": 2728
+    },
+    {
+      "epoch": 1.1518987341772151,
+      "grad_norm": 0.4487358629703522,
+      "learning_rate": 0.00019278938672220167,
+      "loss": 0.7899971008300781,
+      "step": 2730
+    },
+    {
+      "epoch": 1.1527426160337553,
+      "grad_norm": 0.3510114252567291,
+      "learning_rate": 0.00019277146858540166,
+      "loss": 0.6266586780548096,
+      "step": 2732
+    },
+    {
+      "epoch": 1.1535864978902954,
+      "grad_norm": 0.4295555055141449,
+      "learning_rate": 0.00019275352904800553,
+      "loss": 0.7450110912322998,
+      "step": 2734
+    },
+    {
+      "epoch": 1.1544303797468354,
+      "grad_norm": 0.40629687905311584,
+      "learning_rate": 0.00019273556811415165,
+      "loss": 0.6875549554824829,
+      "step": 2736
+    },
+    {
+      "epoch": 1.1552742616033755,
+      "grad_norm": 0.4243362247943878,
+      "learning_rate": 0.00019271758578798322,
+      "loss": 0.8141522407531738,
+      "step": 2738
+    },
+    {
+      "epoch": 1.1561181434599157,
+      "grad_norm": 0.483306884765625,
+      "learning_rate": 0.00019269958207364843,
+      "loss": 0.6905186176300049,
+      "step": 2740
+    },
+    {
+      "epoch": 1.1569620253164556,
+      "grad_norm": 0.4303925335407257,
+      "learning_rate": 0.00019268155697530037,
+      "loss": 0.6709047555923462,
+      "step": 2742
+    },
+    {
+      "epoch": 1.1578059071729958,
+      "grad_norm": 0.4533381760120392,
+      "learning_rate": 0.0001926635104970971,
+      "loss": 0.7107465863227844,
+      "step": 2744
+    },
+    {
+      "epoch": 1.158649789029536,
+      "grad_norm": 0.4823286533355713,
+      "learning_rate": 0.0001926454426432016,
+      "loss": 0.8043986558914185,
+      "step": 2746
+    },
+    {
+      "epoch": 1.159493670886076,
+      "grad_norm": 0.3892867863178253,
+      "learning_rate": 0.00019262735341778179,
+      "loss": 0.7812165021896362,
+      "step": 2748
+    },
+    {
+      "epoch": 1.160337552742616,
+      "grad_norm": 0.46737003326416016,
+      "learning_rate": 0.00019260924282501045,
+      "loss": 0.7141886353492737,
+      "step": 2750
+    },
+    {
+      "epoch": 1.1611814345991562,
+      "grad_norm": 0.4335328936576843,
+      "learning_rate": 0.0001925911108690654,
+      "loss": 0.7528460621833801,
+      "step": 2752
+    },
+    {
+      "epoch": 1.1620253164556962,
+      "grad_norm": 0.3792484700679779,
+      "learning_rate": 0.00019257295755412932,
+      "loss": 0.7817611694335938,
+      "step": 2754
+    },
+    {
+      "epoch": 1.1628691983122363,
+      "grad_norm": 0.41246429085731506,
+      "learning_rate": 0.00019255478288438983,
+      "loss": 0.6904958486557007,
+      "step": 2756
+    },
+    {
+      "epoch": 1.1637130801687763,
+      "grad_norm": 0.44368481636047363,
+      "learning_rate": 0.00019253658686403948,
+      "loss": 0.7571630477905273,
+      "step": 2758
+    },
+    {
+      "epoch": 1.1645569620253164,
+      "grad_norm": 0.40121781826019287,
+      "learning_rate": 0.00019251836949727574,
+      "loss": 0.6747061014175415,
+      "step": 2760
+    },
+    {
+      "epoch": 1.1654008438818566,
+      "grad_norm": 0.4660188853740692,
+      "learning_rate": 0.000192500130788301,
+      "loss": 0.7722339034080505,
+      "step": 2762
+    },
+    {
+      "epoch": 1.1662447257383965,
+      "grad_norm": 0.41693782806396484,
+      "learning_rate": 0.0001924818707413226,
+      "loss": 0.8081457018852234,
+      "step": 2764
+    },
+    {
+      "epoch": 1.1670886075949367,
+      "grad_norm": 0.4274570345878601,
+      "learning_rate": 0.00019246358936055274,
+      "loss": 0.7745853066444397,
+      "step": 2766
+    },
+    {
+      "epoch": 1.1679324894514769,
+      "grad_norm": 0.4385969638824463,
+      "learning_rate": 0.00019244528665020863,
+      "loss": 0.7918146252632141,
+      "step": 2768
+    },
+    {
+      "epoch": 1.1687763713080168,
+      "grad_norm": 0.42832866311073303,
+      "learning_rate": 0.00019242696261451234,
+      "loss": 0.7382830381393433,
+      "step": 2770
+    },
+    {
+      "epoch": 1.169620253164557,
+      "grad_norm": 0.44177982211112976,
+      "learning_rate": 0.00019240861725769085,
+      "loss": 0.7501721978187561,
+      "step": 2772
+    },
+    {
+      "epoch": 1.1704641350210971,
+      "grad_norm": 0.4193457365036011,
+      "learning_rate": 0.00019239025058397611,
+      "loss": 0.6495388150215149,
+      "step": 2774
+    },
+    {
+      "epoch": 1.171308016877637,
+      "grad_norm": 0.5145925283432007,
+      "learning_rate": 0.000192371862597605,
+      "loss": 0.8291991353034973,
+      "step": 2776
+    },
+    {
+      "epoch": 1.1721518987341772,
+      "grad_norm": 0.40911251306533813,
+      "learning_rate": 0.00019235345330281914,
+      "loss": 0.7491340637207031,
+      "step": 2778
+    },
+    {
+      "epoch": 1.1729957805907172,
+      "grad_norm": 0.3920276165008545,
+      "learning_rate": 0.0001923350227038653,
+      "loss": 0.6315004825592041,
+      "step": 2780
+    },
+    {
+      "epoch": 1.1738396624472573,
+      "grad_norm": 0.442303329706192,
+      "learning_rate": 0.0001923165708049951,
+      "loss": 0.7538990378379822,
+      "step": 2782
+    },
+    {
+      "epoch": 1.1746835443037975,
+      "grad_norm": 0.4693523049354553,
+      "learning_rate": 0.00019229809761046497,
+      "loss": 0.7440794110298157,
+      "step": 2784
+    },
+    {
+      "epoch": 1.1755274261603375,
+      "grad_norm": 0.4948059618473053,
+      "learning_rate": 0.0001922796031245363,
+      "loss": 0.7342568635940552,
+      "step": 2786
+    },
+    {
+      "epoch": 1.1763713080168776,
+      "grad_norm": 0.3900465965270996,
+      "learning_rate": 0.00019226108735147547,
+      "loss": 0.6969661116600037,
+      "step": 2788
+    },
+    {
+      "epoch": 1.1772151898734178,
+      "grad_norm": 0.36153602600097656,
+      "learning_rate": 0.00019224255029555372,
+      "loss": 0.7097541689872742,
+      "step": 2790
+    },
+    {
+      "epoch": 1.1780590717299577,
+      "grad_norm": 0.41168832778930664,
+      "learning_rate": 0.00019222399196104715,
+      "loss": 0.8141831755638123,
+      "step": 2792
+    },
+    {
+      "epoch": 1.1789029535864979,
+      "grad_norm": 0.4542821943759918,
+      "learning_rate": 0.0001922054123522368,
+      "loss": 0.7458863258361816,
+      "step": 2794
+    },
+    {
+      "epoch": 1.179746835443038,
+      "grad_norm": 0.4026047885417938,
+      "learning_rate": 0.0001921868114734087,
+      "loss": 0.6535642147064209,
+      "step": 2796
+    },
+    {
+      "epoch": 1.180590717299578,
+      "grad_norm": 0.4417034387588501,
+      "learning_rate": 0.0001921681893288537,
+      "loss": 0.6248135566711426,
+      "step": 2798
+    },
+    {
+      "epoch": 1.1814345991561181,
+      "grad_norm": 0.45027074217796326,
+      "learning_rate": 0.00019214954592286748,
+      "loss": 0.6777785420417786,
+      "step": 2800
+    },
+    {
+      "epoch": 1.1814345991561181,
+      "eval_loss": 0.8124309182167053,
+      "eval_runtime": 688.4759,
+      "eval_samples_per_second": 3.06,
+      "eval_steps_per_second": 3.06,
+      "step": 2800
+    },
+    {
+      "epoch": 1.1822784810126583,
+      "grad_norm": 0.406865656375885,
+      "learning_rate": 0.00019213088125975082,
+      "loss": 0.6460301876068115,
+      "step": 2802
+    },
+    {
+      "epoch": 1.1831223628691983,
+      "grad_norm": 0.5052749514579773,
+      "learning_rate": 0.00019211219534380928,
+      "loss": 0.7292830944061279,
+      "step": 2804
+    },
+    {
+      "epoch": 1.1839662447257384,
+      "grad_norm": 0.4028380513191223,
+      "learning_rate": 0.00019209348817935328,
+      "loss": 0.7399409413337708,
+      "step": 2806
+    },
+    {
+      "epoch": 1.1848101265822786,
+      "grad_norm": 0.4522664248943329,
+      "learning_rate": 0.00019207475977069826,
+      "loss": 0.6885398030281067,
+      "step": 2808
+    },
+    {
+      "epoch": 1.1856540084388185,
+      "grad_norm": 0.419546514749527,
+      "learning_rate": 0.0001920560101221645,
+      "loss": 0.6962382197380066,
+      "step": 2810
+    },
+    {
+      "epoch": 1.1864978902953587,
+      "grad_norm": 0.48912352323532104,
+      "learning_rate": 0.00019203723923807715,
+      "loss": 0.7380965352058411,
+      "step": 2812
+    },
+    {
+      "epoch": 1.1873417721518988,
+      "grad_norm": 0.40905487537384033,
+      "learning_rate": 0.00019201844712276635,
+      "loss": 0.755382776260376,
+      "step": 2814
+    },
+    {
+      "epoch": 1.1881856540084388,
+      "grad_norm": 0.385623037815094,
+      "learning_rate": 0.000191999633780567,
+      "loss": 0.7066168189048767,
+      "step": 2816
+    },
+    {
+      "epoch": 1.189029535864979,
+      "grad_norm": 0.4488723576068878,
+      "learning_rate": 0.00019198079921581908,
+      "loss": 0.7172947525978088,
+      "step": 2818
+    },
+    {
+      "epoch": 1.189873417721519,
+      "grad_norm": 0.385277658700943,
+      "learning_rate": 0.00019196194343286727,
+      "loss": 0.6714195013046265,
+      "step": 2820
+    },
+    {
+      "epoch": 1.190717299578059,
+      "grad_norm": 0.4280855655670166,
+      "learning_rate": 0.00019194306643606127,
+      "loss": 0.7076966166496277,
+      "step": 2822
+    },
+    {
+      "epoch": 1.1915611814345992,
+      "grad_norm": 0.4626067578792572,
+      "learning_rate": 0.00019192416822975567,
+      "loss": 0.7926747798919678,
+      "step": 2824
+    },
+    {
+      "epoch": 1.1924050632911392,
+      "grad_norm": 0.3994208574295044,
+      "learning_rate": 0.00019190524881830987,
+      "loss": 0.7654290199279785,
+      "step": 2826
+    },
+    {
+      "epoch": 1.1932489451476793,
+      "grad_norm": 0.47346508502960205,
+      "learning_rate": 0.00019188630820608825,
+      "loss": 0.7857998609542847,
+      "step": 2828
+    },
+    {
+      "epoch": 1.1940928270042195,
+      "grad_norm": 0.446077823638916,
+      "learning_rate": 0.00019186734639746003,
+      "loss": 0.7603738903999329,
+      "step": 2830
+    },
+    {
+      "epoch": 1.1949367088607594,
+      "grad_norm": 0.41503727436065674,
+      "learning_rate": 0.00019184836339679936,
+      "loss": 0.790582537651062,
+      "step": 2832
+    },
+    {
+      "epoch": 1.1957805907172996,
+      "grad_norm": 0.43651700019836426,
+      "learning_rate": 0.00019182935920848522,
+      "loss": 0.7229864597320557,
+      "step": 2834
+    },
+    {
+      "epoch": 1.1966244725738397,
+      "grad_norm": 0.4409031867980957,
+      "learning_rate": 0.00019181033383690153,
+      "loss": 0.7817287445068359,
+      "step": 2836
+    },
+    {
+      "epoch": 1.1974683544303797,
+      "grad_norm": 0.3562015891075134,
+      "learning_rate": 0.0001917912872864371,
+      "loss": 0.6342297792434692,
+      "step": 2838
+    },
+    {
+      "epoch": 1.1983122362869199,
+      "grad_norm": 0.4767192602157593,
+      "learning_rate": 0.00019177221956148555,
+      "loss": 0.7065272927284241,
+      "step": 2840
+    },
+    {
+      "epoch": 1.1991561181434598,
+      "grad_norm": 0.45131945610046387,
+      "learning_rate": 0.00019175313066644546,
+      "loss": 0.777110755443573,
+      "step": 2842
+    },
+    {
+      "epoch": 1.2,
+      "grad_norm": 0.47566652297973633,
+      "learning_rate": 0.00019173402060572027,
+      "loss": 0.7456457018852234,
+      "step": 2844
+    },
+    {
+      "epoch": 1.2008438818565401,
+      "grad_norm": 0.48909857869148254,
+      "learning_rate": 0.00019171488938371834,
+      "loss": 0.8041547536849976,
+      "step": 2846
+    },
+    {
+      "epoch": 1.20168776371308,
+      "grad_norm": 0.436591774225235,
+      "learning_rate": 0.00019169573700485284,
+      "loss": 0.6786676049232483,
+      "step": 2848
+    },
+    {
+      "epoch": 1.2025316455696202,
+      "grad_norm": 0.44230160117149353,
+      "learning_rate": 0.00019167656347354183,
+      "loss": 0.7661880850791931,
+      "step": 2850
+    },
+    {
+      "epoch": 1.2033755274261604,
+      "grad_norm": 0.4844048321247101,
+      "learning_rate": 0.00019165736879420836,
+      "loss": 0.7369645833969116,
+      "step": 2852
+    },
+    {
+      "epoch": 1.2042194092827003,
+      "grad_norm": 0.4272955060005188,
+      "learning_rate": 0.0001916381529712802,
+      "loss": 0.7776681184768677,
+      "step": 2854
+    },
+    {
+      "epoch": 1.2050632911392405,
+      "grad_norm": 0.42392730712890625,
+      "learning_rate": 0.00019161891600919008,
+      "loss": 0.8019667267799377,
+      "step": 2856
+    },
+    {
+      "epoch": 1.2059071729957807,
+      "grad_norm": 0.4528326988220215,
+      "learning_rate": 0.00019159965791237565,
+      "loss": 0.7811952233314514,
+      "step": 2858
+    },
+    {
+      "epoch": 1.2067510548523206,
+      "grad_norm": 0.42840948700904846,
+      "learning_rate": 0.00019158037868527932,
+      "loss": 0.7244279980659485,
+      "step": 2860
+    },
+    {
+      "epoch": 1.2075949367088608,
+      "grad_norm": 0.4291226863861084,
+      "learning_rate": 0.00019156107833234845,
+      "loss": 0.7675760984420776,
+      "step": 2862
+    },
+    {
+      "epoch": 1.208438818565401,
+      "grad_norm": 0.4455093443393707,
+      "learning_rate": 0.00019154175685803527,
+      "loss": 0.7071483135223389,
+      "step": 2864
+    },
+    {
+      "epoch": 1.2092827004219409,
+      "grad_norm": 0.41356268525123596,
+      "learning_rate": 0.00019152241426679688,
+      "loss": 0.7554603815078735,
+      "step": 2866
+    },
+    {
+      "epoch": 1.210126582278481,
+      "grad_norm": 0.43875646591186523,
+      "learning_rate": 0.00019150305056309525,
+      "loss": 0.7198356986045837,
+      "step": 2868
+    },
+    {
+      "epoch": 1.2109704641350212,
+      "grad_norm": 0.454356849193573,
+      "learning_rate": 0.0001914836657513972,
+      "loss": 0.7393195629119873,
+      "step": 2870
+    },
+    {
+      "epoch": 1.2118143459915611,
+      "grad_norm": 0.37668535113334656,
+      "learning_rate": 0.00019146425983617444,
+      "loss": 0.7277671694755554,
+      "step": 2872
+    },
+    {
+      "epoch": 1.2126582278481013,
+      "grad_norm": 0.37869662046432495,
+      "learning_rate": 0.00019144483282190354,
+      "loss": 0.6783158183097839,
+      "step": 2874
+    },
+    {
+      "epoch": 1.2135021097046415,
+      "grad_norm": 0.4333525002002716,
+      "learning_rate": 0.00019142538471306596,
+      "loss": 0.7024772763252258,
+      "step": 2876
+    },
+    {
+      "epoch": 1.2143459915611814,
+      "grad_norm": 0.4693842828273773,
+      "learning_rate": 0.00019140591551414796,
+      "loss": 0.7387984395027161,
+      "step": 2878
+    },
+    {
+      "epoch": 1.2151898734177216,
+      "grad_norm": 0.432640939950943,
+      "learning_rate": 0.00019138642522964074,
+      "loss": 0.7597174644470215,
+      "step": 2880
+    },
+    {
+      "epoch": 1.2160337552742617,
+      "grad_norm": 0.432477205991745,
+      "learning_rate": 0.00019136691386404032,
+      "loss": 0.7884587049484253,
+      "step": 2882
+    },
+    {
+      "epoch": 1.2168776371308017,
+      "grad_norm": 0.39305904507637024,
+      "learning_rate": 0.00019134738142184763,
+      "loss": 0.6873269081115723,
+      "step": 2884
+    },
+    {
+      "epoch": 1.2177215189873418,
+      "grad_norm": 0.4485352635383606,
+      "learning_rate": 0.00019132782790756838,
+      "loss": 0.6523491144180298,
+      "step": 2886
+    },
+    {
+      "epoch": 1.2185654008438818,
+      "grad_norm": 0.4480868875980377,
+      "learning_rate": 0.00019130825332571322,
+      "loss": 0.7383941411972046,
+      "step": 2888
+    },
+    {
+      "epoch": 1.219409282700422,
+      "grad_norm": 0.44085049629211426,
+      "learning_rate": 0.00019128865768079764,
+      "loss": 0.7391335964202881,
+      "step": 2890
+    },
+    {
+      "epoch": 1.220253164556962,
+      "grad_norm": 0.4621105492115021,
+      "learning_rate": 0.00019126904097734198,
+      "loss": 0.7282017469406128,
+      "step": 2892
+    },
+    {
+      "epoch": 1.221097046413502,
+      "grad_norm": 0.41478198766708374,
+      "learning_rate": 0.00019124940321987136,
+      "loss": 0.7554038763046265,
+      "step": 2894
+    },
+    {
+      "epoch": 1.2219409282700422,
+      "grad_norm": 0.4143751263618469,
+      "learning_rate": 0.00019122974441291595,
+      "loss": 0.6826799511909485,
+      "step": 2896
+    },
+    {
+      "epoch": 1.2227848101265824,
+      "grad_norm": 0.45392894744873047,
+      "learning_rate": 0.00019121006456101057,
+      "loss": 0.751000702381134,
+      "step": 2898
+    },
+    {
+      "epoch": 1.2236286919831223,
+      "grad_norm": 0.46946296095848083,
+      "learning_rate": 0.00019119036366869506,
+      "loss": 0.7598516941070557,
+      "step": 2900
+    },
+    {
+      "epoch": 1.2236286919831223,
+      "eval_loss": 0.8077136278152466,
+      "eval_runtime": 685.6042,
+      "eval_samples_per_second": 3.073,
+      "eval_steps_per_second": 3.073,
+      "step": 2900
+    },
+    {
+      "epoch": 1.2244725738396625,
+      "grad_norm": 0.4316656291484833,
+      "learning_rate": 0.00019117064174051395,
+      "loss": 0.7518358826637268,
+      "step": 2902
+    },
+    {
+      "epoch": 1.2253164556962026,
+      "grad_norm": 0.47737377882003784,
+      "learning_rate": 0.0001911508987810168,
+      "loss": 0.7308215498924255,
+      "step": 2904
+    },
+    {
+      "epoch": 1.2261603375527426,
+      "grad_norm": 0.40906062722206116,
+      "learning_rate": 0.00019113113479475783,
+      "loss": 0.6996226906776428,
+      "step": 2906
+    },
+    {
+      "epoch": 1.2270042194092827,
+      "grad_norm": 0.517085075378418,
+      "learning_rate": 0.00019111134978629633,
+      "loss": 0.8364415168762207,
+      "step": 2908
+    },
+    {
+      "epoch": 1.2278481012658227,
+      "grad_norm": 0.425581693649292,
+      "learning_rate": 0.00019109154376019624,
+      "loss": 0.722011923789978,
+      "step": 2910
+    },
+    {
+      "epoch": 1.2286919831223628,
+      "grad_norm": 0.4651593565940857,
+      "learning_rate": 0.00019107171672102643,
+      "loss": 0.7586172223091125,
+      "step": 2912
+    },
+    {
+      "epoch": 1.229535864978903,
+      "grad_norm": 0.4578891396522522,
+      "learning_rate": 0.00019105186867336067,
+      "loss": 0.7387955188751221,
+      "step": 2914
+    },
+    {
+      "epoch": 1.230379746835443,
+      "grad_norm": 0.4312961995601654,
+      "learning_rate": 0.00019103199962177748,
+      "loss": 0.8036285638809204,
+      "step": 2916
+    },
+    {
+      "epoch": 1.231223628691983,
+      "grad_norm": 0.4355425536632538,
+      "learning_rate": 0.00019101210957086026,
+      "loss": 0.7356550097465515,
+      "step": 2918
+    },
+    {
+      "epoch": 1.2320675105485233,
+      "grad_norm": 0.424675852060318,
+      "learning_rate": 0.00019099219852519733,
+      "loss": 0.6730139255523682,
+      "step": 2920
+    },
+    {
+      "epoch": 1.2329113924050632,
+      "grad_norm": 0.41102367639541626,
+      "learning_rate": 0.0001909722664893817,
+      "loss": 0.6983892321586609,
+      "step": 2922
+    },
+    {
+      "epoch": 1.2337552742616034,
+      "grad_norm": 0.4088250398635864,
+      "learning_rate": 0.00019095231346801132,
+      "loss": 0.6297144889831543,
+      "step": 2924
+    },
+    {
+      "epoch": 1.2345991561181435,
+      "grad_norm": 0.3685890734195709,
+      "learning_rate": 0.00019093233946568906,
+      "loss": 0.7035690546035767,
+      "step": 2926
+    },
+    {
+      "epoch": 1.2354430379746835,
+      "grad_norm": 0.39438629150390625,
+      "learning_rate": 0.0001909123444870224,
+      "loss": 0.820853054523468,
+      "step": 2928
+    },
+    {
+      "epoch": 1.2362869198312236,
+      "grad_norm": 0.5041676163673401,
+      "learning_rate": 0.00019089232853662392,
+      "loss": 0.7884082794189453,
+      "step": 2930
+    },
+    {
+      "epoch": 1.2371308016877638,
+      "grad_norm": 0.40592801570892334,
+      "learning_rate": 0.00019087229161911086,
+      "loss": 0.773978590965271,
+      "step": 2932
+    },
+    {
+      "epoch": 1.2379746835443037,
+      "grad_norm": 0.41828030347824097,
+      "learning_rate": 0.00019085223373910536,
+      "loss": 0.6742141842842102,
+      "step": 2934
+    },
+    {
+      "epoch": 1.238818565400844,
+      "grad_norm": 0.39984026551246643,
+      "learning_rate": 0.0001908321549012344,
+      "loss": 0.7106125950813293,
+      "step": 2936
+    },
+    {
+      "epoch": 1.239662447257384,
+      "grad_norm": 0.4841296970844269,
+      "learning_rate": 0.00019081205511012973,
+      "loss": 0.7390915751457214,
+      "step": 2938
+    },
+    {
+      "epoch": 1.240506329113924,
+      "grad_norm": 0.39837637543678284,
+      "learning_rate": 0.00019079193437042806,
+      "loss": 0.6527198553085327,
+      "step": 2940
+    },
+    {
+      "epoch": 1.2413502109704642,
+      "grad_norm": 0.40916863083839417,
+      "learning_rate": 0.0001907717926867708,
+      "loss": 0.6939712762832642,
+      "step": 2942
+    },
+    {
+      "epoch": 1.2421940928270043,
+      "grad_norm": 0.4329306483268738,
+      "learning_rate": 0.0001907516300638043,
+      "loss": 0.7566131353378296,
+      "step": 2944
+    },
+    {
+      "epoch": 1.2430379746835443,
+      "grad_norm": 0.47983601689338684,
+      "learning_rate": 0.00019073144650617963,
+      "loss": 0.7893748879432678,
+      "step": 2946
+    },
+    {
+      "epoch": 1.2438818565400844,
+      "grad_norm": 0.4368246793746948,
+      "learning_rate": 0.0001907112420185528,
+      "loss": 0.8259525299072266,
+      "step": 2948
+    },
+    {
+      "epoch": 1.2447257383966246,
+      "grad_norm": 0.41272372007369995,
+      "learning_rate": 0.0001906910166055846,
+      "loss": 0.734692394733429,
+      "step": 2950
+    },
+    {
+      "epoch": 1.2455696202531645,
+      "grad_norm": 0.36929839849472046,
+      "learning_rate": 0.00019067077027194057,
+      "loss": 0.7091750502586365,
+      "step": 2952
+    },
+    {
+      "epoch": 1.2464135021097047,
+      "grad_norm": 0.4395252466201782,
+      "learning_rate": 0.00019065050302229125,
+      "loss": 0.6960210204124451,
+      "step": 2954
+    },
+    {
+      "epoch": 1.2472573839662446,
+      "grad_norm": 0.46784868836402893,
+      "learning_rate": 0.00019063021486131186,
+      "loss": 0.7817115187644958,
+      "step": 2956
+    },
+    {
+      "epoch": 1.2481012658227848,
+      "grad_norm": 0.4191588759422302,
+      "learning_rate": 0.00019060990579368244,
+      "loss": 0.7933676242828369,
+      "step": 2958
+    },
+    {
+      "epoch": 1.248945147679325,
+      "grad_norm": 0.3967524468898773,
+      "learning_rate": 0.000190589575824088,
+      "loss": 0.6856688857078552,
+      "step": 2960
+    },
+    {
+      "epoch": 1.249789029535865,
+      "grad_norm": 0.40346091985702515,
+      "learning_rate": 0.0001905692249572182,
+      "loss": 0.6783818602561951,
+      "step": 2962
+    },
+    {
+      "epoch": 1.250632911392405,
+      "grad_norm": 0.38349980115890503,
+      "learning_rate": 0.0001905488531977676,
+      "loss": 0.6997019052505493,
+      "step": 2964
+    },
+    {
+      "epoch": 1.2514767932489452,
+      "grad_norm": 0.44021397829055786,
+      "learning_rate": 0.0001905284605504356,
+      "loss": 0.6978800892829895,
+      "step": 2966
+    },
+    {
+      "epoch": 1.2523206751054852,
+      "grad_norm": 0.47491583228111267,
+      "learning_rate": 0.0001905080470199264,
+      "loss": 0.7007693648338318,
+      "step": 2968
+    },
+    {
+      "epoch": 1.2531645569620253,
+      "grad_norm": 0.4011451005935669,
+      "learning_rate": 0.00019048761261094898,
+      "loss": 0.7258896231651306,
+      "step": 2970
+    },
+    {
+      "epoch": 1.2540084388185653,
+      "grad_norm": 0.44206640124320984,
+      "learning_rate": 0.0001904671573282172,
+      "loss": 0.6746299266815186,
+      "step": 2972
+    },
+    {
+      "epoch": 1.2548523206751054,
+      "grad_norm": 0.4532456696033478,
+      "learning_rate": 0.00019044668117644965,
+      "loss": 0.7638527154922485,
+      "step": 2974
+    },
+    {
+      "epoch": 1.2556962025316456,
+      "grad_norm": 0.44301095604896545,
+      "learning_rate": 0.00019042618416036984,
+      "loss": 0.659598171710968,
+      "step": 2976
+    },
+    {
+      "epoch": 1.2565400843881855,
+      "grad_norm": 0.37427952885627747,
+      "learning_rate": 0.00019040566628470597,
+      "loss": 0.7132709622383118,
+      "step": 2978
+    },
+    {
+      "epoch": 1.2573839662447257,
+      "grad_norm": 0.4195408821105957,
+      "learning_rate": 0.0001903851275541912,
+      "loss": 0.8163022398948669,
+      "step": 2980
+    },
+    {
+      "epoch": 1.2582278481012659,
+      "grad_norm": 0.3951880633831024,
+      "learning_rate": 0.00019036456797356336,
+      "loss": 0.7481505870819092,
+      "step": 2982
+    },
+    {
+      "epoch": 1.2590717299578058,
+      "grad_norm": 0.4239100217819214,
+      "learning_rate": 0.00019034398754756522,
+      "loss": 0.750126838684082,
+      "step": 2984
+    },
+    {
+      "epoch": 1.259915611814346,
+      "grad_norm": 0.4209354817867279,
+      "learning_rate": 0.0001903233862809442,
+      "loss": 0.746433436870575,
+      "step": 2986
+    },
+    {
+      "epoch": 1.2607594936708861,
+      "grad_norm": 0.4188138544559479,
+      "learning_rate": 0.00019030276417845266,
+      "loss": 0.7104454040527344,
+      "step": 2988
+    },
+    {
+      "epoch": 1.261603375527426,
+      "grad_norm": 0.4531371593475342,
+      "learning_rate": 0.00019028212124484772,
+      "loss": 0.650738000869751,
+      "step": 2990
+    },
+    {
+      "epoch": 1.2624472573839662,
+      "grad_norm": 0.4672204256057739,
+      "learning_rate": 0.00019026145748489133,
+      "loss": 0.6870551109313965,
+      "step": 2992
+    },
+    {
+      "epoch": 1.2632911392405064,
+      "grad_norm": 0.40681278705596924,
+      "learning_rate": 0.00019024077290335018,
+      "loss": 0.6879403591156006,
+      "step": 2994
+    },
+    {
+      "epoch": 1.2641350210970463,
+      "grad_norm": 0.4491691291332245,
+      "learning_rate": 0.00019022006750499585,
+      "loss": 0.6886372566223145,
+      "step": 2996
+    },
+    {
+      "epoch": 1.2649789029535865,
+      "grad_norm": 0.4144807457923889,
+      "learning_rate": 0.0001901993412946046,
+      "loss": 0.7199324369430542,
+      "step": 2998
+    },
+    {
+      "epoch": 1.2658227848101267,
+      "grad_norm": 0.40370482206344604,
+      "learning_rate": 0.0001901785942769577,
+      "loss": 0.7105490565299988,
+      "step": 3000
+    },
+    {
+      "epoch": 1.2658227848101267,
+      "eval_loss": 0.8033165335655212,
+      "eval_runtime": 678.7554,
+      "eval_samples_per_second": 3.104,
+      "eval_steps_per_second": 3.104,
+      "step": 3000
+    },
+    {
+      "epoch": 1.2666666666666666,
+      "grad_norm": 0.42637890577316284,
+      "learning_rate": 0.00019015782645684097,
+      "loss": 0.7385403513908386,
+      "step": 3002
+    },
+    {
+      "epoch": 1.2675105485232068,
+      "grad_norm": 0.3709809482097626,
+      "learning_rate": 0.00019013703783904518,
+      "loss": 0.6395129561424255,
+      "step": 3004
+    },
+    {
+      "epoch": 1.268354430379747,
+      "grad_norm": 0.4642501175403595,
+      "learning_rate": 0.00019011622842836592,
+      "loss": 0.7123657464981079,
+      "step": 3006
+    },
+    {
+      "epoch": 1.2691983122362869,
+      "grad_norm": 0.4553911089897156,
+      "learning_rate": 0.00019009539822960342,
+      "loss": 0.744048535823822,
+      "step": 3008
+    },
+    {
+      "epoch": 1.270042194092827,
+      "grad_norm": 0.39334386587142944,
+      "learning_rate": 0.0001900745472475629,
+      "loss": 0.7181914448738098,
+      "step": 3010
+    },
+    {
+      "epoch": 1.2708860759493672,
+      "grad_norm": 0.40902766585350037,
+      "learning_rate": 0.00019005367548705426,
+      "loss": 0.7238107919692993,
+      "step": 3012
+    },
+    {
+      "epoch": 1.2717299578059071,
+      "grad_norm": 0.48296019434928894,
+      "learning_rate": 0.00019003278295289216,
+      "loss": 0.7479744553565979,
+      "step": 3014
+    },
+    {
+      "epoch": 1.2725738396624473,
+      "grad_norm": 0.4807029962539673,
+      "learning_rate": 0.00019001186964989617,
+      "loss": 0.829924464225769,
+      "step": 3016
+    },
+    {
+      "epoch": 1.2734177215189875,
+      "grad_norm": 0.4239083528518677,
+      "learning_rate": 0.00018999093558289055,
+      "loss": 0.7204707264900208,
+      "step": 3018
+    },
+    {
+      "epoch": 1.2742616033755274,
+      "grad_norm": 0.4687928259372711,
+      "learning_rate": 0.0001899699807567044,
+      "loss": 0.7668164968490601,
+      "step": 3020
+    },
+    {
+      "epoch": 1.2751054852320676,
+      "grad_norm": 0.41724732518196106,
+      "learning_rate": 0.0001899490051761716,
+      "loss": 0.766342043876648,
+      "step": 3022
+    },
+    {
+      "epoch": 1.2759493670886077,
+      "grad_norm": 0.4131613075733185,
+      "learning_rate": 0.00018992800884613082,
+      "loss": 0.6595507860183716,
+      "step": 3024
+    },
+    {
+      "epoch": 1.2767932489451477,
+      "grad_norm": 0.4156053364276886,
+      "learning_rate": 0.00018990699177142548,
+      "loss": 0.7419850826263428,
+      "step": 3026
+    },
+    {
+      "epoch": 1.2776371308016878,
+      "grad_norm": 0.41845977306365967,
+      "learning_rate": 0.00018988595395690388,
+      "loss": 0.67264324426651,
+      "step": 3028
+    },
+    {
+      "epoch": 1.2784810126582278,
+      "grad_norm": 0.3826757073402405,
+      "learning_rate": 0.00018986489540741893,
+      "loss": 0.6269281506538391,
+      "step": 3030
+    },
+    {
+      "epoch": 1.279324894514768,
+      "grad_norm": 0.4531428813934326,
+      "learning_rate": 0.00018984381612782854,
+      "loss": 0.71211177110672,
+      "step": 3032
+    },
+    {
+      "epoch": 1.2801687763713079,
+      "grad_norm": 0.3747255802154541,
+      "learning_rate": 0.00018982271612299524,
+      "loss": 0.635804295539856,
+      "step": 3034
+    },
+    {
+      "epoch": 1.281012658227848,
+      "grad_norm": 0.4079552888870239,
+      "learning_rate": 0.00018980159539778638,
+      "loss": 0.7269464731216431,
+      "step": 3036
+    },
+    {
+      "epoch": 1.2818565400843882,
+      "grad_norm": 0.4193854033946991,
+      "learning_rate": 0.00018978045395707418,
+      "loss": 0.7737077474594116,
+      "step": 3038
+    },
+    {
+      "epoch": 1.2827004219409281,
+      "grad_norm": 0.4182516038417816,
+      "learning_rate": 0.0001897592918057355,
+      "loss": 0.7458174824714661,
+      "step": 3040
+    },
+    {
+      "epoch": 1.2835443037974683,
+      "grad_norm": 0.43343567848205566,
+      "learning_rate": 0.00018973810894865205,
+      "loss": 0.6565808057785034,
+      "step": 3042
+    },
+    {
+      "epoch": 1.2843881856540085,
+      "grad_norm": 0.4294503331184387,
+      "learning_rate": 0.00018971690539071033,
+      "loss": 0.736944317817688,
+      "step": 3044
+    },
+    {
+      "epoch": 1.2852320675105484,
+      "grad_norm": 0.41330763697624207,
+      "learning_rate": 0.0001896956811368016,
+      "loss": 0.6921742558479309,
+      "step": 3046
+    },
+    {
+      "epoch": 1.2860759493670886,
+      "grad_norm": 0.44992703199386597,
+      "learning_rate": 0.0001896744361918218,
+      "loss": 0.7004357576370239,
+      "step": 3048
+    },
+    {
+      "epoch": 1.2869198312236287,
+      "grad_norm": 0.43748870491981506,
+      "learning_rate": 0.0001896531705606719,
+      "loss": 0.6767720580101013,
+      "step": 3050
+    },
+    {
+      "epoch": 1.2877637130801687,
+      "grad_norm": 0.47581973671913147,
+      "learning_rate": 0.0001896318842482573,
+      "loss": 0.7596892714500427,
+      "step": 3052
+    },
+    {
+      "epoch": 1.2886075949367088,
+      "grad_norm": 0.45806005597114563,
+      "learning_rate": 0.00018961057725948844,
+      "loss": 0.7650044560432434,
+      "step": 3054
+    },
+    {
+      "epoch": 1.289451476793249,
+      "grad_norm": 0.4291481375694275,
+      "learning_rate": 0.00018958924959928042,
+      "loss": 0.6654610633850098,
+      "step": 3056
+    },
+    {
+      "epoch": 1.290295358649789,
+      "grad_norm": 0.42454418540000916,
+      "learning_rate": 0.00018956790127255309,
+      "loss": 0.7079797387123108,
+      "step": 3058
+    },
+    {
+      "epoch": 1.2911392405063291,
+      "grad_norm": 0.46820637583732605,
+      "learning_rate": 0.00018954653228423114,
+      "loss": 0.7724313735961914,
+      "step": 3060
+    },
+    {
+      "epoch": 1.2919831223628693,
+      "grad_norm": 0.4440016746520996,
+      "learning_rate": 0.00018952514263924396,
+      "loss": 0.7746338248252869,
+      "step": 3062
+    },
+    {
+      "epoch": 1.2928270042194092,
+      "grad_norm": 0.4296419620513916,
+      "learning_rate": 0.00018950373234252572,
+      "loss": 0.7570929527282715,
+      "step": 3064
+    },
+    {
+      "epoch": 1.2936708860759494,
+      "grad_norm": 0.428780734539032,
+      "learning_rate": 0.0001894823013990154,
+      "loss": 0.7274802923202515,
+      "step": 3066
+    },
+    {
+      "epoch": 1.2945147679324895,
+      "grad_norm": 0.4111585021018982,
+      "learning_rate": 0.0001894608498136567,
+      "loss": 0.6595024466514587,
+      "step": 3068
+    },
+    {
+      "epoch": 1.2953586497890295,
+      "grad_norm": 0.4373013973236084,
+      "learning_rate": 0.00018943937759139802,
+      "loss": 0.7316440939903259,
+      "step": 3070
+    },
+    {
+      "epoch": 1.2962025316455696,
+      "grad_norm": 0.37705400586128235,
+      "learning_rate": 0.0001894178847371927,
+      "loss": 0.7072293162345886,
+      "step": 3072
+    },
+    {
+      "epoch": 1.2970464135021098,
+      "grad_norm": 0.4171125888824463,
+      "learning_rate": 0.00018939637125599864,
+      "loss": 0.7218101024627686,
+      "step": 3074
+    },
+    {
+      "epoch": 1.2978902953586497,
+      "grad_norm": 0.4264235198497772,
+      "learning_rate": 0.00018937483715277865,
+      "loss": 0.725140392780304,
+      "step": 3076
+    },
+    {
+      "epoch": 1.29873417721519,
+      "grad_norm": 0.4251643121242523,
+      "learning_rate": 0.0001893532824325002,
+      "loss": 0.7423256635665894,
+      "step": 3078
+    },
+    {
+      "epoch": 1.29957805907173,
+      "grad_norm": 0.43096452951431274,
+      "learning_rate": 0.00018933170710013554,
+      "loss": 0.7058811187744141,
+      "step": 3080
+    },
+    {
+      "epoch": 1.30042194092827,
+      "grad_norm": 0.466804176568985,
+      "learning_rate": 0.00018931011116066172,
+      "loss": 0.7334373593330383,
+      "step": 3082
+    },
+    {
+      "epoch": 1.3012658227848102,
+      "grad_norm": 0.36896654963493347,
+      "learning_rate": 0.00018928849461906047,
+      "loss": 0.6033462285995483,
+      "step": 3084
+    },
+    {
+      "epoch": 1.3021097046413503,
+      "grad_norm": 0.42875027656555176,
+      "learning_rate": 0.00018926685748031833,
+      "loss": 0.690375566482544,
+      "step": 3086
+    },
+    {
+      "epoch": 1.3029535864978903,
+      "grad_norm": 0.4041171967983246,
+      "learning_rate": 0.00018924519974942658,
+      "loss": 0.7468435168266296,
+      "step": 3088
+    },
+    {
+      "epoch": 1.3037974683544304,
+      "grad_norm": 0.4704097807407379,
+      "learning_rate": 0.00018922352143138126,
+      "loss": 0.7574344873428345,
+      "step": 3090
+    },
+    {
+      "epoch": 1.3046413502109704,
+      "grad_norm": 0.4391418397426605,
+      "learning_rate": 0.0001892018225311831,
+      "loss": 0.756452739238739,
+      "step": 3092
+    },
+    {
+      "epoch": 1.3054852320675105,
+      "grad_norm": 0.42684388160705566,
+      "learning_rate": 0.0001891801030538376,
+      "loss": 0.6620991230010986,
+      "step": 3094
+    },
+    {
+      "epoch": 1.3063291139240507,
+      "grad_norm": 0.4972277283668518,
+      "learning_rate": 0.00018915836300435507,
+      "loss": 0.7764125466346741,
+      "step": 3096
+    },
+    {
+      "epoch": 1.3071729957805907,
+      "grad_norm": 0.3905445337295532,
+      "learning_rate": 0.00018913660238775053,
+      "loss": 0.6557663083076477,
+      "step": 3098
+    },
+    {
+      "epoch": 1.3080168776371308,
+      "grad_norm": 0.4553062319755554,
+      "learning_rate": 0.0001891148212090437,
+      "loss": 0.6746718883514404,
+      "step": 3100
+    },
+    {
+      "epoch": 1.3080168776371308,
+      "eval_loss": 0.8010181784629822,
+      "eval_runtime": 668.1688,
+      "eval_samples_per_second": 3.153,
+      "eval_steps_per_second": 3.153,
+      "step": 3100
+    },
+    {
+      "epoch": 1.3088607594936708,
+      "grad_norm": 0.4254429340362549,
+      "learning_rate": 0.00018909301947325915,
+      "loss": 0.7916419506072998,
+      "step": 3102
+    },
+    {
+      "epoch": 1.309704641350211,
+      "grad_norm": 0.6118508577346802,
+      "learning_rate": 0.00018907119718542602,
+      "loss": 0.883936882019043,
+      "step": 3104
+    },
+    {
+      "epoch": 1.310548523206751,
+      "grad_norm": 0.43206971883773804,
+      "learning_rate": 0.00018904935435057837,
+      "loss": 0.7145928740501404,
+      "step": 3106
+    },
+    {
+      "epoch": 1.311392405063291,
+      "grad_norm": 0.39657703042030334,
+      "learning_rate": 0.0001890274909737549,
+      "loss": 0.7442842125892639,
+      "step": 3108
+    },
+    {
+      "epoch": 1.3122362869198312,
+      "grad_norm": 0.4087927043437958,
+      "learning_rate": 0.00018900560705999904,
+      "loss": 0.6815276145935059,
+      "step": 3110
+    },
+    {
+      "epoch": 1.3130801687763713,
+      "grad_norm": 0.430259108543396,
+      "learning_rate": 0.00018898370261435904,
+      "loss": 0.8153640627861023,
+      "step": 3112
+    },
+    {
+      "epoch": 1.3139240506329113,
+      "grad_norm": 0.47486868500709534,
+      "learning_rate": 0.0001889617776418878,
+      "loss": 0.7828266024589539,
+      "step": 3114
+    },
+    {
+      "epoch": 1.3147679324894515,
+      "grad_norm": 0.4929087162017822,
+      "learning_rate": 0.000188939832147643,
+      "loss": 0.7849065661430359,
+      "step": 3116
+    },
+    {
+      "epoch": 1.3156118143459916,
+      "grad_norm": 0.39038679003715515,
+      "learning_rate": 0.00018891786613668704,
+      "loss": 0.7262985706329346,
+      "step": 3118
+    },
+    {
+      "epoch": 1.3164556962025316,
+      "grad_norm": 0.4827214479446411,
+      "learning_rate": 0.0001888958796140871,
+      "loss": 0.8121856451034546,
+      "step": 3120
+    },
+    {
+      "epoch": 1.3172995780590717,
+      "grad_norm": 0.47739601135253906,
+      "learning_rate": 0.000188873872584915,
+      "loss": 0.7165566086769104,
+      "step": 3122
+    },
+    {
+      "epoch": 1.3181434599156119,
+      "grad_norm": 0.4274037778377533,
+      "learning_rate": 0.0001888518450542473,
+      "loss": 0.6698412299156189,
+      "step": 3124
+    },
+    {
+      "epoch": 1.3189873417721518,
+      "grad_norm": 0.44825276732444763,
+      "learning_rate": 0.00018882979702716544,
+      "loss": 0.7271128296852112,
+      "step": 3126
+    },
+    {
+      "epoch": 1.319831223628692,
+      "grad_norm": 0.42956918478012085,
+      "learning_rate": 0.00018880772850875537,
+      "loss": 0.7924301624298096,
+      "step": 3128
+    },
+    {
+      "epoch": 1.3206751054852321,
+      "grad_norm": 0.4848136305809021,
+      "learning_rate": 0.00018878563950410792,
+      "loss": 0.7846497893333435,
+      "step": 3130
+    },
+    {
+      "epoch": 1.321518987341772,
+      "grad_norm": 0.46651726961135864,
+      "learning_rate": 0.00018876353001831864,
+      "loss": 0.7262214422225952,
+      "step": 3132
+    },
+    {
+      "epoch": 1.3223628691983123,
+      "grad_norm": 0.44598284363746643,
+      "learning_rate": 0.0001887414000564877,
+      "loss": 0.7640091180801392,
+      "step": 3134
+    },
+    {
+      "epoch": 1.3232067510548524,
+      "grad_norm": 0.43744903802871704,
+      "learning_rate": 0.00018871924962372006,
+      "loss": 0.6960080862045288,
+      "step": 3136
+    },
+    {
+      "epoch": 1.3240506329113924,
+      "grad_norm": 0.4737822413444519,
+      "learning_rate": 0.00018869707872512544,
+      "loss": 0.7121244072914124,
+      "step": 3138
+    },
+    {
+      "epoch": 1.3248945147679325,
+      "grad_norm": 0.44452038407325745,
+      "learning_rate": 0.00018867488736581817,
+      "loss": 0.713148295879364,
+      "step": 3140
+    },
+    {
+      "epoch": 1.3257383966244727,
+      "grad_norm": 0.47142693400382996,
+      "learning_rate": 0.0001886526755509175,
+      "loss": 0.6904974579811096,
+      "step": 3142
+    },
+    {
+      "epoch": 1.3265822784810126,
+      "grad_norm": 0.40572887659072876,
+      "learning_rate": 0.00018863044328554712,
+      "loss": 0.7319386005401611,
+      "step": 3144
+    },
+    {
+      "epoch": 1.3274261603375528,
+      "grad_norm": 0.44173532724380493,
+      "learning_rate": 0.00018860819057483566,
+      "loss": 0.7268120646476746,
+      "step": 3146
+    },
+    {
+      "epoch": 1.328270042194093,
+      "grad_norm": 0.43025046586990356,
+      "learning_rate": 0.00018858591742391642,
+      "loss": 0.7129524946212769,
+      "step": 3148
+    },
+    {
+      "epoch": 1.3291139240506329,
+      "grad_norm": 0.41418036818504333,
+      "learning_rate": 0.00018856362383792732,
+      "loss": 0.6917402148246765,
+      "step": 3150
+    },
+    {
+      "epoch": 1.329957805907173,
+      "grad_norm": 0.46275854110717773,
+      "learning_rate": 0.00018854130982201111,
+      "loss": 0.7653918862342834,
+      "step": 3152
+    },
+    {
+      "epoch": 1.3308016877637132,
+      "grad_norm": 0.38927653431892395,
+      "learning_rate": 0.0001885189753813152,
+      "loss": 0.7311477065086365,
+      "step": 3154
+    },
+    {
+      "epoch": 1.3316455696202532,
+      "grad_norm": 0.44294974207878113,
+      "learning_rate": 0.0001884966205209917,
+      "loss": 0.7978816628456116,
+      "step": 3156
+    },
+    {
+      "epoch": 1.3324894514767933,
+      "grad_norm": 0.3772844970226288,
+      "learning_rate": 0.00018847424524619745,
+      "loss": 0.7497738599777222,
+      "step": 3158
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.4538787603378296,
+      "learning_rate": 0.000188451849562094,
+      "loss": 0.712806761264801,
+      "step": 3160
+    },
+    {
+      "epoch": 1.3341772151898734,
+      "grad_norm": 0.4194336533546448,
+      "learning_rate": 0.00018842943347384763,
+      "loss": 0.7282580137252808,
+      "step": 3162
+    },
+    {
+      "epoch": 1.3350210970464136,
+      "grad_norm": 0.4186438024044037,
+      "learning_rate": 0.00018840699698662925,
+      "loss": 0.797752320766449,
+      "step": 3164
+    },
+    {
+      "epoch": 1.3358649789029535,
+      "grad_norm": 0.47711506485939026,
+      "learning_rate": 0.00018838454010561457,
+      "loss": 0.8175379633903503,
+      "step": 3166
+    },
+    {
+      "epoch": 1.3367088607594937,
+      "grad_norm": 0.4380176067352295,
+      "learning_rate": 0.00018836206283598394,
+      "loss": 0.659176766872406,
+      "step": 3168
+    },
+    {
+      "epoch": 1.3375527426160336,
+      "grad_norm": 0.4505249559879303,
+      "learning_rate": 0.00018833956518292245,
+      "loss": 0.732314944267273,
+      "step": 3170
+    },
+    {
+      "epoch": 1.3383966244725738,
+      "grad_norm": 0.3832913935184479,
+      "learning_rate": 0.00018831704715161984,
+      "loss": 0.6791377067565918,
+      "step": 3172
+    },
+    {
+      "epoch": 1.339240506329114,
+      "grad_norm": 0.42996498942375183,
+      "learning_rate": 0.00018829450874727064,
+      "loss": 0.7395796775817871,
+      "step": 3174
+    },
+    {
+      "epoch": 1.340084388185654,
+      "grad_norm": 0.3926190435886383,
+      "learning_rate": 0.000188271949975074,
+      "loss": 0.6893631219863892,
+      "step": 3176
+    },
+    {
+      "epoch": 1.340928270042194,
+      "grad_norm": 0.4353938102722168,
+      "learning_rate": 0.0001882493708402338,
+      "loss": 0.6708382964134216,
+      "step": 3178
+    },
+    {
+      "epoch": 1.3417721518987342,
+      "grad_norm": 0.43043214082717896,
+      "learning_rate": 0.0001882267713479586,
+      "loss": 0.8447695374488831,
+      "step": 3180
+    },
+    {
+      "epoch": 1.3426160337552742,
+      "grad_norm": 0.40693536400794983,
+      "learning_rate": 0.00018820415150346163,
+      "loss": 0.7081091403961182,
+      "step": 3182
+    },
+    {
+      "epoch": 1.3434599156118143,
+      "grad_norm": 0.45037057995796204,
+      "learning_rate": 0.00018818151131196097,
+      "loss": 0.6822336316108704,
+      "step": 3184
+    },
+    {
+      "epoch": 1.3443037974683545,
+      "grad_norm": 0.3731948137283325,
+      "learning_rate": 0.00018815885077867915,
+      "loss": 0.6316909790039062,
+      "step": 3186
+    },
+    {
+      "epoch": 1.3451476793248944,
+      "grad_norm": 0.4248095750808716,
+      "learning_rate": 0.00018813616990884363,
+      "loss": 0.7452375292778015,
+      "step": 3188
+    },
+    {
+      "epoch": 1.3459915611814346,
+      "grad_norm": 0.39229142665863037,
+      "learning_rate": 0.00018811346870768638,
+      "loss": 0.6493782997131348,
+      "step": 3190
+    },
+    {
+      "epoch": 1.3468354430379748,
+      "grad_norm": 0.45015549659729004,
+      "learning_rate": 0.00018809074718044414,
+      "loss": 0.735163688659668,
+      "step": 3192
+    },
+    {
+      "epoch": 1.3476793248945147,
+      "grad_norm": 0.39756354689598083,
+      "learning_rate": 0.00018806800533235836,
+      "loss": 0.6656006574630737,
+      "step": 3194
+    },
+    {
+      "epoch": 1.3485232067510549,
+      "grad_norm": 0.4410010576248169,
+      "learning_rate": 0.0001880452431686751,
+      "loss": 0.7040987014770508,
+      "step": 3196
+    },
+    {
+      "epoch": 1.349367088607595,
+      "grad_norm": 0.48144352436065674,
+      "learning_rate": 0.0001880224606946452,
+      "loss": 0.7722331285476685,
+      "step": 3198
+    },
+    {
+      "epoch": 1.350210970464135,
+      "grad_norm": 0.38644957542419434,
+      "learning_rate": 0.00018799965791552414,
+      "loss": 0.6681689023971558,
+      "step": 3200
+    },
+    {
+      "epoch": 1.350210970464135,
+      "eval_loss": 0.797160804271698,
+      "eval_runtime": 680.976,
+      "eval_samples_per_second": 3.094,
+      "eval_steps_per_second": 3.094,
+      "step": 3200
+    },
+    {
+      "epoch": 1.3510548523206751,
+      "grad_norm": 0.4471014440059662,
+      "learning_rate": 0.00018797683483657201,
+      "loss": 0.6995527744293213,
+      "step": 3202
+    },
+    {
+      "epoch": 1.3518987341772153,
+      "grad_norm": 0.4065002202987671,
+      "learning_rate": 0.00018795399146305378,
+      "loss": 0.659622848033905,
+      "step": 3204
+    },
+    {
+      "epoch": 1.3527426160337552,
+      "grad_norm": 0.4787910580635071,
+      "learning_rate": 0.00018793112780023888,
+      "loss": 0.7927116751670837,
+      "step": 3206
+    },
+    {
+      "epoch": 1.3535864978902954,
+      "grad_norm": 0.43351876735687256,
+      "learning_rate": 0.0001879082438534015,
+      "loss": 0.6738835573196411,
+      "step": 3208
+    },
+    {
+      "epoch": 1.3544303797468356,
+      "grad_norm": 0.4275745749473572,
+      "learning_rate": 0.00018788533962782063,
+      "loss": 0.7785354852676392,
+      "step": 3210
+    },
+    {
+      "epoch": 1.3552742616033755,
+      "grad_norm": 0.4506145119667053,
+      "learning_rate": 0.00018786241512877975,
+      "loss": 0.7325424551963806,
+      "step": 3212
+    },
+    {
+      "epoch": 1.3561181434599157,
+      "grad_norm": 0.42040735483169556,
+      "learning_rate": 0.00018783947036156713,
+      "loss": 0.6898744702339172,
+      "step": 3214
+    },
+    {
+      "epoch": 1.3569620253164558,
+      "grad_norm": 0.4650438129901886,
+      "learning_rate": 0.0001878165053314757,
+      "loss": 0.7498815655708313,
+      "step": 3216
+    },
+    {
+      "epoch": 1.3578059071729958,
+      "grad_norm": 0.4194130599498749,
+      "learning_rate": 0.000187793520043803,
+      "loss": 0.6891883611679077,
+      "step": 3218
+    },
+    {
+      "epoch": 1.358649789029536,
+      "grad_norm": 0.4447539746761322,
+      "learning_rate": 0.0001877705145038514,
+      "loss": 0.753805935382843,
+      "step": 3220
+    },
+    {
+      "epoch": 1.3594936708860759,
+      "grad_norm": 0.4047110974788666,
+      "learning_rate": 0.00018774748871692773,
+      "loss": 0.6476293802261353,
+      "step": 3222
+    },
+    {
+      "epoch": 1.360337552742616,
+      "grad_norm": 0.39818957448005676,
+      "learning_rate": 0.00018772444268834364,
+      "loss": 0.6765931844711304,
+      "step": 3224
+    },
+    {
+      "epoch": 1.3611814345991562,
+      "grad_norm": 0.41843733191490173,
+      "learning_rate": 0.0001877013764234154,
+      "loss": 0.7360382676124573,
+      "step": 3226
+    },
+    {
+      "epoch": 1.3620253164556961,
+      "grad_norm": 0.41397789120674133,
+      "learning_rate": 0.00018767828992746394,
+      "loss": 0.7149524688720703,
+      "step": 3228
+    },
+    {
+      "epoch": 1.3628691983122363,
+      "grad_norm": 0.43585047125816345,
+      "learning_rate": 0.00018765518320581492,
+      "loss": 0.6932925581932068,
+      "step": 3230
+    },
+    {
+      "epoch": 1.3637130801687762,
+      "grad_norm": 0.3651057481765747,
+      "learning_rate": 0.00018763205626379857,
+      "loss": 0.6561182141304016,
+      "step": 3232
+    },
+    {
+      "epoch": 1.3645569620253164,
+      "grad_norm": 0.44937077164649963,
+      "learning_rate": 0.00018760890910674984,
+      "loss": 0.7436831593513489,
+      "step": 3234
+    },
+    {
+      "epoch": 1.3654008438818566,
+      "grad_norm": 0.5045920014381409,
+      "learning_rate": 0.00018758574174000832,
+      "loss": 0.7849522233009338,
+      "step": 3236
+    },
+    {
+      "epoch": 1.3662447257383965,
+      "grad_norm": 0.4625580906867981,
+      "learning_rate": 0.00018756255416891833,
+      "loss": 0.7501618266105652,
+      "step": 3238
+    },
+    {
+      "epoch": 1.3670886075949367,
+      "grad_norm": 0.3721269369125366,
+      "learning_rate": 0.00018753934639882875,
+      "loss": 0.7656083106994629,
+      "step": 3240
+    },
+    {
+      "epoch": 1.3679324894514768,
+      "grad_norm": 0.41721680760383606,
+      "learning_rate": 0.00018751611843509318,
+      "loss": 0.7816168069839478,
+      "step": 3242
+    },
+    {
+      "epoch": 1.3687763713080168,
+      "grad_norm": 0.8452199101448059,
+      "learning_rate": 0.00018749287028306988,
+      "loss": 0.7002755403518677,
+      "step": 3244
+    },
+    {
+      "epoch": 1.369620253164557,
+      "grad_norm": 0.4320507347583771,
+      "learning_rate": 0.00018746960194812172,
+      "loss": 0.7358270883560181,
+      "step": 3246
+    },
+    {
+      "epoch": 1.370464135021097,
+      "grad_norm": 0.43323323130607605,
+      "learning_rate": 0.00018744631343561627,
+      "loss": 0.6528644561767578,
+      "step": 3248
+    },
+    {
+      "epoch": 1.371308016877637,
+      "grad_norm": 0.41487041115760803,
+      "learning_rate": 0.0001874230047509258,
+      "loss": 0.7317517995834351,
+      "step": 3250
+    },
+    {
+      "epoch": 1.3721518987341772,
+      "grad_norm": 0.5085384249687195,
+      "learning_rate": 0.00018739967589942707,
+      "loss": 0.8236643671989441,
+      "step": 3252
+    },
+    {
+      "epoch": 1.3729957805907174,
+      "grad_norm": 0.4126952290534973,
+      "learning_rate": 0.00018737632688650168,
+      "loss": 0.730254054069519,
+      "step": 3254
+    },
+    {
+      "epoch": 1.3738396624472573,
+      "grad_norm": 0.5272142291069031,
+      "learning_rate": 0.00018735295771753573,
+      "loss": 0.7637890577316284,
+      "step": 3256
+    },
+    {
+      "epoch": 1.3746835443037975,
+      "grad_norm": 0.4319530725479126,
+      "learning_rate": 0.00018732956839792012,
+      "loss": 0.7906988263130188,
+      "step": 3258
+    },
+    {
+      "epoch": 1.3755274261603376,
+      "grad_norm": 0.4393143355846405,
+      "learning_rate": 0.00018730615893305025,
+      "loss": 0.7261547446250916,
+      "step": 3260
+    },
+    {
+      "epoch": 1.3763713080168776,
+      "grad_norm": 0.4161636233329773,
+      "learning_rate": 0.00018728272932832632,
+      "loss": 0.6857497692108154,
+      "step": 3262
+    },
+    {
+      "epoch": 1.3772151898734177,
+      "grad_norm": 0.48862120509147644,
+      "learning_rate": 0.00018725927958915297,
+      "loss": 0.7488857507705688,
+      "step": 3264
+    },
+    {
+      "epoch": 1.378059071729958,
+      "grad_norm": 0.4422304928302765,
+      "learning_rate": 0.0001872358097209397,
+      "loss": 0.7257721424102783,
+      "step": 3266
+    },
+    {
+      "epoch": 1.3789029535864978,
+      "grad_norm": 0.4577067494392395,
+      "learning_rate": 0.00018721231972910048,
+      "loss": 0.746697187423706,
+      "step": 3268
+    },
+    {
+      "epoch": 1.379746835443038,
+      "grad_norm": 0.42382001876831055,
+      "learning_rate": 0.00018718880961905406,
+      "loss": 0.768519401550293,
+      "step": 3270
+    },
+    {
+      "epoch": 1.3805907172995782,
+      "grad_norm": 0.43368014693260193,
+      "learning_rate": 0.00018716527939622378,
+      "loss": 0.7426102757453918,
+      "step": 3272
+    },
+    {
+      "epoch": 1.381434599156118,
+      "grad_norm": 0.3777220547199249,
+      "learning_rate": 0.00018714172906603757,
+      "loss": 0.7104197144508362,
+      "step": 3274
+    },
+    {
+      "epoch": 1.3822784810126583,
+      "grad_norm": 0.46071696281433105,
+      "learning_rate": 0.00018711815863392802,
+      "loss": 0.7502968907356262,
+      "step": 3276
+    },
+    {
+      "epoch": 1.3831223628691984,
+      "grad_norm": 0.42510855197906494,
+      "learning_rate": 0.00018709456810533245,
+      "loss": 0.6400540471076965,
+      "step": 3278
+    },
+    {
+      "epoch": 1.3839662447257384,
+      "grad_norm": 0.4217228293418884,
+      "learning_rate": 0.00018707095748569268,
+      "loss": 0.6648687720298767,
+      "step": 3280
+    },
+    {
+      "epoch": 1.3848101265822785,
+      "grad_norm": 0.48267799615859985,
+      "learning_rate": 0.00018704732678045526,
+      "loss": 0.7176171541213989,
+      "step": 3282
+    },
+    {
+      "epoch": 1.3856540084388187,
+      "grad_norm": 0.47114965319633484,
+      "learning_rate": 0.00018702367599507134,
+      "loss": 0.8402457237243652,
+      "step": 3284
+    },
+    {
+      "epoch": 1.3864978902953586,
+      "grad_norm": 0.5054492354393005,
+      "learning_rate": 0.00018700000513499667,
+      "loss": 0.8130836486816406,
+      "step": 3286
+    },
+    {
+      "epoch": 1.3873417721518988,
+      "grad_norm": 0.39435645937919617,
+      "learning_rate": 0.00018697631420569168,
+      "loss": 0.6432579755783081,
+      "step": 3288
+    },
+    {
+      "epoch": 1.3881856540084387,
+      "grad_norm": 0.3997228443622589,
+      "learning_rate": 0.00018695260321262142,
+      "loss": 0.6921643614768982,
+      "step": 3290
+    },
+    {
+      "epoch": 1.389029535864979,
+      "grad_norm": 0.37304237484931946,
+      "learning_rate": 0.00018692887216125557,
+      "loss": 0.7162153124809265,
+      "step": 3292
+    },
+    {
+      "epoch": 1.389873417721519,
+      "grad_norm": 0.7997981309890747,
+      "learning_rate": 0.00018690512105706838,
+      "loss": 0.7194803953170776,
+      "step": 3294
+    },
+    {
+      "epoch": 1.390717299578059,
+      "grad_norm": 0.41268959641456604,
+      "learning_rate": 0.00018688134990553883,
+      "loss": 0.6888322830200195,
+      "step": 3296
+    },
+    {
+      "epoch": 1.3915611814345992,
+      "grad_norm": 0.4485826790332794,
+      "learning_rate": 0.00018685755871215043,
+      "loss": 0.7387221455574036,
+      "step": 3298
+    },
+    {
+      "epoch": 1.3924050632911391,
+      "grad_norm": 0.47835296392440796,
+      "learning_rate": 0.0001868337474823914,
+      "loss": 0.8061952590942383,
+      "step": 3300
+    },
+    {
+      "epoch": 1.3924050632911391,
+      "eval_loss": 0.795619547367096,
+      "eval_runtime": 692.7157,
+      "eval_samples_per_second": 3.042,
+      "eval_steps_per_second": 3.042,
+      "step": 3300
+    },
+    {
+      "epoch": 1.3932489451476793,
+      "grad_norm": 0.4024396240711212,
+      "learning_rate": 0.00018680991622175446,
+      "loss": 0.7317537069320679,
+      "step": 3302
+    },
+    {
+      "epoch": 1.3940928270042194,
+      "grad_norm": 0.4526490271091461,
+      "learning_rate": 0.00018678606493573708,
+      "loss": 0.7470992207527161,
+      "step": 3304
+    },
+    {
+      "epoch": 1.3949367088607594,
+      "grad_norm": 0.5055129528045654,
+      "learning_rate": 0.00018676219362984126,
+      "loss": 0.7627943754196167,
+      "step": 3306
+    },
+    {
+      "epoch": 1.3957805907172995,
+      "grad_norm": 0.4030216634273529,
+      "learning_rate": 0.0001867383023095737,
+      "loss": 0.684033989906311,
+      "step": 3308
+    },
+    {
+      "epoch": 1.3966244725738397,
+      "grad_norm": 0.4353308379650116,
+      "learning_rate": 0.0001867143909804456,
+      "loss": 0.7366722822189331,
+      "step": 3310
+    },
+    {
+      "epoch": 1.3974683544303796,
+      "grad_norm": 0.5360553860664368,
+      "learning_rate": 0.0001866904596479729,
+      "loss": 0.8050951957702637,
+      "step": 3312
+    },
+    {
+      "epoch": 1.3983122362869198,
+      "grad_norm": 0.46081677079200745,
+      "learning_rate": 0.00018666650831767608,
+      "loss": 0.7227963805198669,
+      "step": 3314
+    },
+    {
+      "epoch": 1.39915611814346,
+      "grad_norm": 0.4325845539569855,
+      "learning_rate": 0.00018664253699508029,
+      "loss": 0.714216411113739,
+      "step": 3316
+    },
+    {
+      "epoch": 1.4,
+      "grad_norm": 0.4373100996017456,
+      "learning_rate": 0.0001866185456857152,
+      "loss": 0.7378593683242798,
+      "step": 3318
+    },
+    {
+      "epoch": 1.40084388185654,
+      "grad_norm": 0.43221279978752136,
+      "learning_rate": 0.00018659453439511512,
+      "loss": 0.7565981149673462,
+      "step": 3320
+    },
+    {
+      "epoch": 1.4016877637130802,
+      "grad_norm": 0.3938370943069458,
+      "learning_rate": 0.00018657050312881904,
+      "loss": 0.6409488916397095,
+      "step": 3322
+    },
+    {
+      "epoch": 1.4025316455696202,
+      "grad_norm": 0.43609440326690674,
+      "learning_rate": 0.00018654645189237055,
+      "loss": 0.6911652088165283,
+      "step": 3324
+    },
+    {
+      "epoch": 1.4033755274261603,
+      "grad_norm": 0.4526350200176239,
+      "learning_rate": 0.00018652238069131774,
+      "loss": 0.7275710701942444,
+      "step": 3326
+    },
+    {
+      "epoch": 1.4042194092827005,
+      "grad_norm": 0.3990516662597656,
+      "learning_rate": 0.0001864982895312134,
+      "loss": 0.6466895341873169,
+      "step": 3328
+    },
+    {
+      "epoch": 1.4050632911392404,
+      "grad_norm": 0.44321122765541077,
+      "learning_rate": 0.0001864741784176149,
+      "loss": 0.6520540714263916,
+      "step": 3330
+    },
+    {
+      "epoch": 1.4059071729957806,
+      "grad_norm": 0.44733941555023193,
+      "learning_rate": 0.0001864500473560842,
+      "loss": 0.7289671301841736,
+      "step": 3332
+    },
+    {
+      "epoch": 1.4067510548523208,
+      "grad_norm": 0.4569414556026459,
+      "learning_rate": 0.00018642589635218789,
+      "loss": 0.7664391398429871,
+      "step": 3334
+    },
+    {
+      "epoch": 1.4075949367088607,
+      "grad_norm": 0.4172186851501465,
+      "learning_rate": 0.00018640172541149709,
+      "loss": 0.7747462391853333,
+      "step": 3336
+    },
+    {
+      "epoch": 1.4084388185654009,
+      "grad_norm": 0.4540993869304657,
+      "learning_rate": 0.0001863775345395877,
+      "loss": 0.8337607383728027,
+      "step": 3338
+    },
+    {
+      "epoch": 1.409282700421941,
+      "grad_norm": 0.4306154251098633,
+      "learning_rate": 0.00018635332374203993,
+      "loss": 0.6851577758789062,
+      "step": 3340
+    },
+    {
+      "epoch": 1.410126582278481,
+      "grad_norm": 0.45615455508232117,
+      "learning_rate": 0.00018632909302443884,
+      "loss": 0.6841398477554321,
+      "step": 3342
+    },
+    {
+      "epoch": 1.4109704641350211,
+      "grad_norm": 0.35296666622161865,
+      "learning_rate": 0.00018630484239237397,
+      "loss": 0.6672312021255493,
+      "step": 3344
+    },
+    {
+      "epoch": 1.4118143459915613,
+      "grad_norm": 0.4424770176410675,
+      "learning_rate": 0.00018628057185143945,
+      "loss": 0.7027004361152649,
+      "step": 3346
+    },
+    {
+      "epoch": 1.4126582278481012,
+      "grad_norm": 0.429817259311676,
+      "learning_rate": 0.0001862562814072341,
+      "loss": 0.7397928237915039,
+      "step": 3348
+    },
+    {
+      "epoch": 1.4135021097046414,
+      "grad_norm": 0.4720051884651184,
+      "learning_rate": 0.00018623197106536118,
+      "loss": 0.8009215593338013,
+      "step": 3350
+    },
+    {
+      "epoch": 1.4143459915611816,
+      "grad_norm": 0.42197802662849426,
+      "learning_rate": 0.0001862076408314287,
+      "loss": 0.6875046491622925,
+      "step": 3352
+    },
+    {
+      "epoch": 1.4151898734177215,
+      "grad_norm": 0.3674047589302063,
+      "learning_rate": 0.00018618329071104907,
+      "loss": 0.7202261090278625,
+      "step": 3354
+    },
+    {
+      "epoch": 1.4160337552742617,
+      "grad_norm": 0.42327889800071716,
+      "learning_rate": 0.0001861589207098395,
+      "loss": 0.6981248259544373,
+      "step": 3356
+    },
+    {
+      "epoch": 1.4168776371308016,
+      "grad_norm": 0.40644150972366333,
+      "learning_rate": 0.00018613453083342162,
+      "loss": 0.7081356048583984,
+      "step": 3358
+    },
+    {
+      "epoch": 1.4177215189873418,
+      "grad_norm": 0.3948482275009155,
+      "learning_rate": 0.00018611012108742168,
+      "loss": 0.6420091986656189,
+      "step": 3360
+    },
+    {
+      "epoch": 1.4185654008438817,
+      "grad_norm": 0.44622674584388733,
+      "learning_rate": 0.00018608569147747064,
+      "loss": 0.7662023305892944,
+      "step": 3362
+    },
+    {
+      "epoch": 1.4194092827004219,
+      "grad_norm": 0.39772045612335205,
+      "learning_rate": 0.00018606124200920386,
+      "loss": 0.7243877649307251,
+      "step": 3364
+    },
+    {
+      "epoch": 1.420253164556962,
+      "grad_norm": 0.4284326136112213,
+      "learning_rate": 0.00018603677268826138,
+      "loss": 0.7399296164512634,
+      "step": 3366
+    },
+    {
+      "epoch": 1.421097046413502,
+      "grad_norm": 0.3749450445175171,
+      "learning_rate": 0.0001860122835202878,
+      "loss": 0.6978716850280762,
+      "step": 3368
+    },
+    {
+      "epoch": 1.4219409282700421,
+      "grad_norm": 0.46721795201301575,
+      "learning_rate": 0.00018598777451093234,
+      "loss": 0.7733059525489807,
+      "step": 3370
+    },
+    {
+      "epoch": 1.4227848101265823,
+      "grad_norm": 0.39606085419654846,
+      "learning_rate": 0.0001859632456658487,
+      "loss": 0.6721867918968201,
+      "step": 3372
+    },
+    {
+      "epoch": 1.4236286919831223,
+      "grad_norm": 0.3879879117012024,
+      "learning_rate": 0.00018593869699069527,
+      "loss": 0.715825080871582,
+      "step": 3374
+    },
+    {
+      "epoch": 1.4244725738396624,
+      "grad_norm": 0.3641676902770996,
+      "learning_rate": 0.0001859141284911349,
+      "loss": 0.6226941347122192,
+      "step": 3376
+    },
+    {
+      "epoch": 1.4253164556962026,
+      "grad_norm": 0.4612879753112793,
+      "learning_rate": 0.0001858895401728351,
+      "loss": 0.7972121238708496,
+      "step": 3378
+    },
+    {
+      "epoch": 1.4261603375527425,
+      "grad_norm": 0.4406103193759918,
+      "learning_rate": 0.00018586493204146795,
+      "loss": 0.6663658022880554,
+      "step": 3380
+    },
+    {
+      "epoch": 1.4270042194092827,
+      "grad_norm": 0.420205295085907,
+      "learning_rate": 0.00018584030410271,
+      "loss": 0.6867321133613586,
+      "step": 3382
+    },
+    {
+      "epoch": 1.4278481012658228,
+      "grad_norm": 0.4499560594558716,
+      "learning_rate": 0.00018581565636224253,
+      "loss": 0.66688072681427,
+      "step": 3384
+    },
+    {
+      "epoch": 1.4286919831223628,
+      "grad_norm": 0.42507919669151306,
+      "learning_rate": 0.00018579098882575123,
+      "loss": 0.6714702844619751,
+      "step": 3386
+    },
+    {
+      "epoch": 1.429535864978903,
+      "grad_norm": 0.43856513500213623,
+      "learning_rate": 0.0001857663014989265,
+      "loss": 0.7219351530075073,
+      "step": 3388
+    },
+    {
+      "epoch": 1.4303797468354431,
+      "grad_norm": 0.5167419910430908,
+      "learning_rate": 0.00018574159438746314,
+      "loss": 0.7157991528511047,
+      "step": 3390
+    },
+    {
+      "epoch": 1.431223628691983,
+      "grad_norm": 0.4443167746067047,
+      "learning_rate": 0.00018571686749706067,
+      "loss": 0.7509840726852417,
+      "step": 3392
+    },
+    {
+      "epoch": 1.4320675105485232,
+      "grad_norm": 0.41321757435798645,
+      "learning_rate": 0.0001856921208334231,
+      "loss": 0.756000816822052,
+      "step": 3394
+    },
+    {
+      "epoch": 1.4329113924050634,
+      "grad_norm": 0.4094153046607971,
+      "learning_rate": 0.000185667354402259,
+      "loss": 0.7047854065895081,
+      "step": 3396
+    },
+    {
+      "epoch": 1.4337552742616033,
+      "grad_norm": 0.4248482286930084,
+      "learning_rate": 0.0001856425682092815,
+      "loss": 0.7620865106582642,
+      "step": 3398
+    },
+    {
+      "epoch": 1.4345991561181435,
+      "grad_norm": 0.4190550148487091,
+      "learning_rate": 0.0001856177622602083,
+      "loss": 0.7441845536231995,
+      "step": 3400
+    },
+    {
+      "epoch": 1.4345991561181435,
+      "eval_loss": 0.7917885780334473,
+      "eval_runtime": 686.1689,
+      "eval_samples_per_second": 3.071,
+      "eval_steps_per_second": 3.071,
+      "step": 3400
+    },
+    {
+      "epoch": 1.4354430379746836,
+      "grad_norm": 0.45602327585220337,
+      "learning_rate": 0.00018559293656076166,
+      "loss": 0.7123498916625977,
+      "step": 3402
+    },
+    {
+      "epoch": 1.4362869198312236,
+      "grad_norm": 0.5209269523620605,
+      "learning_rate": 0.00018556809111666843,
+      "loss": 0.7096527814865112,
+      "step": 3404
+    },
+    {
+      "epoch": 1.4371308016877637,
+      "grad_norm": 0.44521182775497437,
+      "learning_rate": 0.00018554322593365995,
+      "loss": 0.7116838693618774,
+      "step": 3406
+    },
+    {
+      "epoch": 1.437974683544304,
+      "grad_norm": 0.44571003317832947,
+      "learning_rate": 0.00018551834101747212,
+      "loss": 0.7756352424621582,
+      "step": 3408
+    },
+    {
+      "epoch": 1.4388185654008439,
+      "grad_norm": 0.4030776023864746,
+      "learning_rate": 0.00018549343637384543,
+      "loss": 0.689633309841156,
+      "step": 3410
+    },
+    {
+      "epoch": 1.439662447257384,
+      "grad_norm": 0.4025530517101288,
+      "learning_rate": 0.0001854685120085249,
+      "loss": 0.7173827290534973,
+      "step": 3412
+    },
+    {
+      "epoch": 1.4405063291139242,
+      "grad_norm": 0.4376870095729828,
+      "learning_rate": 0.0001854435679272601,
+      "loss": 0.7212072610855103,
+      "step": 3414
+    },
+    {
+      "epoch": 1.4413502109704641,
+      "grad_norm": 0.4469084143638611,
+      "learning_rate": 0.00018541860413580515,
+      "loss": 0.8260660171508789,
+      "step": 3416
+    },
+    {
+      "epoch": 1.4421940928270043,
+      "grad_norm": 0.48033684492111206,
+      "learning_rate": 0.00018539362063991872,
+      "loss": 0.764930248260498,
+      "step": 3418
+    },
+    {
+      "epoch": 1.4430379746835442,
+      "grad_norm": 0.4428006708621979,
+      "learning_rate": 0.00018536861744536403,
+      "loss": 0.7436494827270508,
+      "step": 3420
+    },
+    {
+      "epoch": 1.4438818565400844,
+      "grad_norm": 0.41812989115715027,
+      "learning_rate": 0.00018534359455790885,
+      "loss": 0.7695552706718445,
+      "step": 3422
+    },
+    {
+      "epoch": 1.4447257383966245,
+      "grad_norm": 0.4243115186691284,
+      "learning_rate": 0.00018531855198332544,
+      "loss": 0.7430856823921204,
+      "step": 3424
+    },
+    {
+      "epoch": 1.4455696202531645,
+      "grad_norm": 0.44387027621269226,
+      "learning_rate": 0.00018529348972739067,
+      "loss": 0.7015851140022278,
+      "step": 3426
+    },
+    {
+      "epoch": 1.4464135021097047,
+      "grad_norm": 0.5105122923851013,
+      "learning_rate": 0.0001852684077958859,
+      "loss": 0.7073075771331787,
+      "step": 3428
+    },
+    {
+      "epoch": 1.4472573839662446,
+      "grad_norm": 0.40193185210227966,
+      "learning_rate": 0.00018524330619459708,
+      "loss": 0.7478930950164795,
+      "step": 3430
+    },
+    {
+      "epoch": 1.4481012658227848,
+      "grad_norm": 0.47935256361961365,
+      "learning_rate": 0.00018521818492931463,
+      "loss": 0.6987152099609375,
+      "step": 3432
+    },
+    {
+      "epoch": 1.448945147679325,
+      "grad_norm": 0.42298686504364014,
+      "learning_rate": 0.00018519304400583356,
+      "loss": 0.6677015423774719,
+      "step": 3434
+    },
+    {
+      "epoch": 1.4497890295358649,
+      "grad_norm": 0.37220221757888794,
+      "learning_rate": 0.0001851678834299534,
+      "loss": 0.5727065205574036,
+      "step": 3436
+    },
+    {
+      "epoch": 1.450632911392405,
+      "grad_norm": 0.4990854561328888,
+      "learning_rate": 0.00018514270320747823,
+      "loss": 0.7245276570320129,
+      "step": 3438
+    },
+    {
+      "epoch": 1.4514767932489452,
+      "grad_norm": 0.4366965889930725,
+      "learning_rate": 0.00018511750334421665,
+      "loss": 0.5907649993896484,
+      "step": 3440
+    },
+    {
+      "epoch": 1.4523206751054851,
+      "grad_norm": 0.4355267286300659,
+      "learning_rate": 0.00018509228384598172,
+      "loss": 0.6898093223571777,
+      "step": 3442
+    },
+    {
+      "epoch": 1.4531645569620253,
+      "grad_norm": 0.4055832028388977,
+      "learning_rate": 0.00018506704471859116,
+      "loss": 0.6242932081222534,
+      "step": 3444
+    },
+    {
+      "epoch": 1.4540084388185655,
+      "grad_norm": 0.42599600553512573,
+      "learning_rate": 0.00018504178596786712,
+      "loss": 0.7214970588684082,
+      "step": 3446
+    },
+    {
+      "epoch": 1.4548523206751054,
+      "grad_norm": 0.41228237748146057,
+      "learning_rate": 0.0001850165075996363,
+      "loss": 0.7120789885520935,
+      "step": 3448
+    },
+    {
+      "epoch": 1.4556962025316456,
+      "grad_norm": 0.4529595375061035,
+      "learning_rate": 0.00018499120961972996,
+      "loss": 0.7376183867454529,
+      "step": 3450
+    },
+    {
+      "epoch": 1.4565400843881857,
+      "grad_norm": 0.4260895550251007,
+      "learning_rate": 0.00018496589203398386,
+      "loss": 0.7321018576622009,
+      "step": 3452
+    },
+    {
+      "epoch": 1.4573839662447257,
+      "grad_norm": 0.46325981616973877,
+      "learning_rate": 0.00018494055484823824,
+      "loss": 0.7099342346191406,
+      "step": 3454
+    },
+    {
+      "epoch": 1.4582278481012658,
+      "grad_norm": 0.43322354555130005,
+      "learning_rate": 0.00018491519806833795,
+      "loss": 0.6663834452629089,
+      "step": 3456
+    },
+    {
+      "epoch": 1.459071729957806,
+      "grad_norm": 0.421683132648468,
+      "learning_rate": 0.0001848898217001323,
+      "loss": 0.6634764671325684,
+      "step": 3458
+    },
+    {
+      "epoch": 1.459915611814346,
+      "grad_norm": 0.44365957379341125,
+      "learning_rate": 0.00018486442574947511,
+      "loss": 0.697095513343811,
+      "step": 3460
+    },
+    {
+      "epoch": 1.460759493670886,
+      "grad_norm": 0.4892813265323639,
+      "learning_rate": 0.00018483901022222474,
+      "loss": 0.8009715676307678,
+      "step": 3462
+    },
+    {
+      "epoch": 1.4616033755274263,
+      "grad_norm": 0.4230288863182068,
+      "learning_rate": 0.00018481357512424408,
+      "loss": 0.631499707698822,
+      "step": 3464
+    },
+    {
+      "epoch": 1.4624472573839662,
+      "grad_norm": 0.49490758776664734,
+      "learning_rate": 0.00018478812046140056,
+      "loss": 0.8045525550842285,
+      "step": 3466
+    },
+    {
+      "epoch": 1.4632911392405064,
+      "grad_norm": 0.5035375952720642,
+      "learning_rate": 0.00018476264623956597,
+      "loss": 0.7299133539199829,
+      "step": 3468
+    },
+    {
+      "epoch": 1.4641350210970465,
+      "grad_norm": 0.45048409700393677,
+      "learning_rate": 0.00018473715246461684,
+      "loss": 0.7313228845596313,
+      "step": 3470
+    },
+    {
+      "epoch": 1.4649789029535865,
+      "grad_norm": 0.41414541006088257,
+      "learning_rate": 0.00018471163914243405,
+      "loss": 0.7515901923179626,
+      "step": 3472
+    },
+    {
+      "epoch": 1.4658227848101266,
+      "grad_norm": 0.4440646767616272,
+      "learning_rate": 0.00018468610627890298,
+      "loss": 0.7425532341003418,
+      "step": 3474
+    },
+    {
+      "epoch": 1.4666666666666668,
+      "grad_norm": 0.4749740660190582,
+      "learning_rate": 0.00018466055387991362,
+      "loss": 0.7509122490882874,
+      "step": 3476
+    },
+    {
+      "epoch": 1.4675105485232067,
+      "grad_norm": 0.498951256275177,
+      "learning_rate": 0.00018463498195136046,
+      "loss": 0.7395774126052856,
+      "step": 3478
+    },
+    {
+      "epoch": 1.4683544303797469,
+      "grad_norm": 0.45996859669685364,
+      "learning_rate": 0.00018460939049914237,
+      "loss": 0.7860153317451477,
+      "step": 3480
+    },
+    {
+      "epoch": 1.469198312236287,
+      "grad_norm": 0.5989604592323303,
+      "learning_rate": 0.0001845837795291629,
+      "loss": 0.8193565607070923,
+      "step": 3482
+    },
+    {
+      "epoch": 1.470042194092827,
+      "grad_norm": 0.3903721272945404,
+      "learning_rate": 0.0001845581490473299,
+      "loss": 0.6569210290908813,
+      "step": 3484
+    },
+    {
+      "epoch": 1.4708860759493672,
+      "grad_norm": 0.4316142201423645,
+      "learning_rate": 0.00018453249905955592,
+      "loss": 0.7226642370223999,
+      "step": 3486
+    },
+    {
+      "epoch": 1.471729957805907,
+      "grad_norm": 0.4447585344314575,
+      "learning_rate": 0.00018450682957175786,
+      "loss": 0.8134398460388184,
+      "step": 3488
+    },
+    {
+      "epoch": 1.4725738396624473,
+      "grad_norm": 0.42515450716018677,
+      "learning_rate": 0.0001844811405898572,
+      "loss": 0.7668985724449158,
+      "step": 3490
+    },
+    {
+      "epoch": 1.4734177215189874,
+      "grad_norm": 0.39883363246917725,
+      "learning_rate": 0.0001844554321197799,
+      "loss": 0.7289790511131287,
+      "step": 3492
+    },
+    {
+      "epoch": 1.4742616033755274,
+      "grad_norm": 0.4419080317020416,
+      "learning_rate": 0.00018442970416745645,
+      "loss": 0.7611824870109558,
+      "step": 3494
+    },
+    {
+      "epoch": 1.4751054852320675,
+      "grad_norm": 0.37460315227508545,
+      "learning_rate": 0.0001844039567388217,
+      "loss": 0.6926230788230896,
+      "step": 3496
+    },
+    {
+      "epoch": 1.4759493670886075,
+      "grad_norm": 0.47706854343414307,
+      "learning_rate": 0.00018437818983981513,
+      "loss": 0.7889325618743896,
+      "step": 3498
+    },
+    {
+      "epoch": 1.4767932489451476,
+      "grad_norm": 0.449028879404068,
+      "learning_rate": 0.00018435240347638074,
+      "loss": 0.7231656312942505,
+      "step": 3500
+    },
+    {
+      "epoch": 1.4767932489451476,
+      "eval_loss": 0.7902651429176331,
+      "eval_runtime": 672.312,
+      "eval_samples_per_second": 3.134,
+      "eval_steps_per_second": 3.134,
+      "step": 3500
+    },
+    {
+      "epoch": 1.4776371308016878,
+      "grad_norm": 0.44085755944252014,
+      "learning_rate": 0.00018432659765446686,
+      "loss": 0.7464927434921265,
+      "step": 3502
+    },
+    {
+      "epoch": 1.4784810126582277,
+      "grad_norm": 0.4243546724319458,
+      "learning_rate": 0.0001843007723800264,
+      "loss": 0.7312393188476562,
+      "step": 3504
+    },
+    {
+      "epoch": 1.479324894514768,
+      "grad_norm": 0.4437529444694519,
+      "learning_rate": 0.00018427492765901677,
+      "loss": 0.7169944643974304,
+      "step": 3506
+    },
+    {
+      "epoch": 1.480168776371308,
+      "grad_norm": 0.3639313280582428,
+      "learning_rate": 0.0001842490634973999,
+      "loss": 0.6169725060462952,
+      "step": 3508
+    },
+    {
+      "epoch": 1.481012658227848,
+      "grad_norm": 0.4452592730522156,
+      "learning_rate": 0.0001842231799011421,
+      "loss": 0.7034649848937988,
+      "step": 3510
+    },
+    {
+      "epoch": 1.4818565400843882,
+      "grad_norm": 0.5229591131210327,
+      "learning_rate": 0.00018419727687621421,
+      "loss": 0.7535615563392639,
+      "step": 3512
+    },
+    {
+      "epoch": 1.4827004219409283,
+      "grad_norm": 0.40748512744903564,
+      "learning_rate": 0.00018417135442859163,
+      "loss": 0.6858918070793152,
+      "step": 3514
+    },
+    {
+      "epoch": 1.4835443037974683,
+      "grad_norm": 0.4532773792743683,
+      "learning_rate": 0.0001841454125642541,
+      "loss": 0.7120109796524048,
+      "step": 3516
+    },
+    {
+      "epoch": 1.4843881856540084,
+      "grad_norm": 0.4281020760536194,
+      "learning_rate": 0.00018411945128918593,
+      "loss": 0.6837015748023987,
+      "step": 3518
+    },
+    {
+      "epoch": 1.4852320675105486,
+      "grad_norm": 0.4891088306903839,
+      "learning_rate": 0.0001840934706093759,
+      "loss": 0.6639618873596191,
+      "step": 3520
+    },
+    {
+      "epoch": 1.4860759493670885,
+      "grad_norm": 0.44358906149864197,
+      "learning_rate": 0.00018406747053081724,
+      "loss": 0.693483829498291,
+      "step": 3522
+    },
+    {
+      "epoch": 1.4869198312236287,
+      "grad_norm": 0.40134626626968384,
+      "learning_rate": 0.00018404145105950767,
+      "loss": 0.7046290636062622,
+      "step": 3524
+    },
+    {
+      "epoch": 1.4877637130801689,
+      "grad_norm": 0.4601934552192688,
+      "learning_rate": 0.0001840154122014494,
+      "loss": 0.6991957426071167,
+      "step": 3526
+    },
+    {
+      "epoch": 1.4886075949367088,
+      "grad_norm": 0.39107242226600647,
+      "learning_rate": 0.00018398935396264906,
+      "loss": 0.667023241519928,
+      "step": 3528
+    },
+    {
+      "epoch": 1.489451476793249,
+      "grad_norm": 0.4498196542263031,
+      "learning_rate": 0.00018396327634911784,
+      "loss": 0.7187332510948181,
+      "step": 3530
+    },
+    {
+      "epoch": 1.4902953586497891,
+      "grad_norm": 0.41769179701805115,
+      "learning_rate": 0.0001839371793668713,
+      "loss": 0.6854014992713928,
+      "step": 3532
+    },
+    {
+      "epoch": 1.491139240506329,
+      "grad_norm": 0.40170612931251526,
+      "learning_rate": 0.0001839110630219295,
+      "loss": 0.6643120050430298,
+      "step": 3534
+    },
+    {
+      "epoch": 1.4919831223628692,
+      "grad_norm": 0.43697425723075867,
+      "learning_rate": 0.00018388492732031701,
+      "loss": 0.6822295188903809,
+      "step": 3536
+    },
+    {
+      "epoch": 1.4928270042194094,
+      "grad_norm": 0.4400080442428589,
+      "learning_rate": 0.00018385877226806288,
+      "loss": 0.7500558495521545,
+      "step": 3538
+    },
+    {
+      "epoch": 1.4936708860759493,
+      "grad_norm": 0.5162575244903564,
+      "learning_rate": 0.00018383259787120048,
+      "loss": 0.7742161750793457,
+      "step": 3540
+    },
+    {
+      "epoch": 1.4945147679324895,
+      "grad_norm": 0.4234789311885834,
+      "learning_rate": 0.0001838064041357678,
+      "loss": 0.7334147095680237,
+      "step": 3542
+    },
+    {
+      "epoch": 1.4953586497890297,
+      "grad_norm": 0.40562525391578674,
+      "learning_rate": 0.0001837801910678072,
+      "loss": 0.691303014755249,
+      "step": 3544
+    },
+    {
+      "epoch": 1.4962025316455696,
+      "grad_norm": 0.3651523292064667,
+      "learning_rate": 0.00018375395867336556,
+      "loss": 0.6478174328804016,
+      "step": 3546
+    },
+    {
+      "epoch": 1.4970464135021098,
+      "grad_norm": 0.45428016781806946,
+      "learning_rate": 0.00018372770695849417,
+      "loss": 0.6800187826156616,
+      "step": 3548
+    },
+    {
+      "epoch": 1.49789029535865,
+      "grad_norm": 0.4258574843406677,
+      "learning_rate": 0.00018370143592924882,
+      "loss": 0.7386951446533203,
+      "step": 3550
+    },
+    {
+      "epoch": 1.4987341772151899,
+      "grad_norm": 0.4260908365249634,
+      "learning_rate": 0.00018367514559168972,
+      "loss": 0.757290780544281,
+      "step": 3552
+    },
+    {
+      "epoch": 1.49957805907173,
+      "grad_norm": 0.4037017226219177,
+      "learning_rate": 0.00018364883595188152,
+      "loss": 0.7254101037979126,
+      "step": 3554
+    },
+    {
+      "epoch": 1.5004219409282702,
+      "grad_norm": 0.4212391674518585,
+      "learning_rate": 0.0001836225070158934,
+      "loss": 0.7342239618301392,
+      "step": 3556
+    },
+    {
+      "epoch": 1.5012658227848101,
+      "grad_norm": 0.4902292490005493,
+      "learning_rate": 0.0001835961587897989,
+      "loss": 0.7147436738014221,
+      "step": 3558
+    },
+    {
+      "epoch": 1.50210970464135,
+      "grad_norm": 0.39230459928512573,
+      "learning_rate": 0.00018356979127967605,
+      "loss": 0.7473883628845215,
+      "step": 3560
+    },
+    {
+      "epoch": 1.5029535864978905,
+      "grad_norm": 0.4203200340270996,
+      "learning_rate": 0.00018354340449160735,
+      "loss": 0.7026845812797546,
+      "step": 3562
+    },
+    {
+      "epoch": 1.5037974683544304,
+      "grad_norm": 0.44363775849342346,
+      "learning_rate": 0.00018351699843167972,
+      "loss": 0.7426526546478271,
+      "step": 3564
+    },
+    {
+      "epoch": 1.5046413502109703,
+      "grad_norm": 0.4479754865169525,
+      "learning_rate": 0.0001834905731059845,
+      "loss": 0.6800932884216309,
+      "step": 3566
+    },
+    {
+      "epoch": 1.5054852320675105,
+      "grad_norm": 0.43898341059684753,
+      "learning_rate": 0.0001834641285206176,
+      "loss": 0.6681315898895264,
+      "step": 3568
+    },
+    {
+      "epoch": 1.5063291139240507,
+      "grad_norm": 0.44789746403694153,
+      "learning_rate": 0.00018343766468167916,
+      "loss": 0.7319603562355042,
+      "step": 3570
+    },
+    {
+      "epoch": 1.5071729957805906,
+      "grad_norm": 0.44450023770332336,
+      "learning_rate": 0.00018341118159527392,
+      "loss": 0.7205765843391418,
+      "step": 3572
+    },
+    {
+      "epoch": 1.5080168776371308,
+      "grad_norm": 0.4461601674556732,
+      "learning_rate": 0.00018338467926751103,
+      "loss": 0.8053180575370789,
+      "step": 3574
+    },
+    {
+      "epoch": 1.508860759493671,
+      "grad_norm": 0.415216863155365,
+      "learning_rate": 0.00018335815770450408,
+      "loss": 0.6944835186004639,
+      "step": 3576
+    },
+    {
+      "epoch": 1.5097046413502109,
+      "grad_norm": 0.48354068398475647,
+      "learning_rate": 0.00018333161691237105,
+      "loss": 0.7333522439002991,
+      "step": 3578
+    },
+    {
+      "epoch": 1.510548523206751,
+      "grad_norm": 0.44450995326042175,
+      "learning_rate": 0.00018330505689723446,
+      "loss": 0.7036639451980591,
+      "step": 3580
+    },
+    {
+      "epoch": 1.5113924050632912,
+      "grad_norm": 0.355945348739624,
+      "learning_rate": 0.00018327847766522111,
+      "loss": 0.6589286923408508,
+      "step": 3582
+    },
+    {
+      "epoch": 1.5122362869198311,
+      "grad_norm": 0.4532116651535034,
+      "learning_rate": 0.00018325187922246236,
+      "loss": 0.7284913063049316,
+      "step": 3584
+    },
+    {
+      "epoch": 1.5130801687763713,
+      "grad_norm": 0.45910802483558655,
+      "learning_rate": 0.00018322526157509396,
+      "loss": 0.7253853678703308,
+      "step": 3586
+    },
+    {
+      "epoch": 1.5139240506329115,
+      "grad_norm": 0.4729263186454773,
+      "learning_rate": 0.0001831986247292561,
+      "loss": 0.7626010179519653,
+      "step": 3588
+    },
+    {
+      "epoch": 1.5147679324894514,
+      "grad_norm": 0.42071840167045593,
+      "learning_rate": 0.00018317196869109336,
+      "loss": 0.705466091632843,
+      "step": 3590
+    },
+    {
+      "epoch": 1.5156118143459916,
+      "grad_norm": 0.40275856852531433,
+      "learning_rate": 0.00018314529346675479,
+      "loss": 0.6900550723075867,
+      "step": 3592
+    },
+    {
+      "epoch": 1.5164556962025317,
+      "grad_norm": 0.43557414412498474,
+      "learning_rate": 0.00018311859906239387,
+      "loss": 0.6951512694358826,
+      "step": 3594
+    },
+    {
+      "epoch": 1.5172995780590717,
+      "grad_norm": 0.4007355570793152,
+      "learning_rate": 0.00018309188548416844,
+      "loss": 0.7453750967979431,
+      "step": 3596
+    },
+    {
+      "epoch": 1.5181434599156118,
+      "grad_norm": 0.4686451256275177,
+      "learning_rate": 0.00018306515273824086,
+      "loss": 0.7549434900283813,
+      "step": 3598
+    },
+    {
+      "epoch": 1.518987341772152,
+      "grad_norm": 0.47308704257011414,
+      "learning_rate": 0.0001830384008307778,
+      "loss": 0.7708864212036133,
+      "step": 3600
+    },
+    {
+      "epoch": 1.518987341772152,
+      "eval_loss": 0.785450279712677,
+      "eval_runtime": 675.2312,
+      "eval_samples_per_second": 3.12,
+      "eval_steps_per_second": 3.12,
+      "step": 3600
+    },
+    {
+      "epoch": 1.519831223628692,
+      "grad_norm": 0.4343389570713043,
+      "learning_rate": 0.0001830116297679505,
+      "loss": 0.7501338124275208,
+      "step": 3602
+    },
+    {
+      "epoch": 1.520675105485232,
+      "grad_norm": 0.4971345067024231,
+      "learning_rate": 0.00018298483955593445,
+      "loss": 0.7358015179634094,
+      "step": 3604
+    },
+    {
+      "epoch": 1.5215189873417723,
+      "grad_norm": 0.42013493180274963,
+      "learning_rate": 0.0001829580302009097,
+      "loss": 0.7972728610038757,
+      "step": 3606
+    },
+    {
+      "epoch": 1.5223628691983122,
+      "grad_norm": 0.44829806685447693,
+      "learning_rate": 0.00018293120170906062,
+      "loss": 0.5908644199371338,
+      "step": 3608
+    },
+    {
+      "epoch": 1.5232067510548524,
+      "grad_norm": 0.47412917017936707,
+      "learning_rate": 0.000182904354086576,
+      "loss": 0.7872797846794128,
+      "step": 3610
+    },
+    {
+      "epoch": 1.5240506329113925,
+      "grad_norm": 0.4224211275577545,
+      "learning_rate": 0.0001828774873396491,
+      "loss": 0.7327616214752197,
+      "step": 3612
+    },
+    {
+      "epoch": 1.5248945147679325,
+      "grad_norm": 0.4055147171020508,
+      "learning_rate": 0.00018285060147447756,
+      "loss": 0.6649892330169678,
+      "step": 3614
+    },
+    {
+      "epoch": 1.5257383966244724,
+      "grad_norm": 0.4058035910129547,
+      "learning_rate": 0.00018282369649726346,
+      "loss": 0.6939427852630615,
+      "step": 3616
+    },
+    {
+      "epoch": 1.5265822784810128,
+      "grad_norm": 0.4192068874835968,
+      "learning_rate": 0.00018279677241421325,
+      "loss": 0.6351069211959839,
+      "step": 3618
+    },
+    {
+      "epoch": 1.5274261603375527,
+      "grad_norm": 0.4659916162490845,
+      "learning_rate": 0.00018276982923153776,
+      "loss": 0.6963213682174683,
+      "step": 3620
+    },
+    {
+      "epoch": 1.5282700421940927,
+      "grad_norm": 0.4993051290512085,
+      "learning_rate": 0.00018274286695545229,
+      "loss": 0.7463255524635315,
+      "step": 3622
+    },
+    {
+      "epoch": 1.529113924050633,
+      "grad_norm": 0.39232251048088074,
+      "learning_rate": 0.00018271588559217654,
+      "loss": 0.6566499471664429,
+      "step": 3624
+    },
+    {
+      "epoch": 1.529957805907173,
+      "grad_norm": 0.4482063353061676,
+      "learning_rate": 0.00018268888514793457,
+      "loss": 0.7695494890213013,
+      "step": 3626
+    },
+    {
+      "epoch": 1.530801687763713,
+      "grad_norm": 0.4000336825847626,
+      "learning_rate": 0.00018266186562895483,
+      "loss": 0.7357239723205566,
+      "step": 3628
+    },
+    {
+      "epoch": 1.5316455696202531,
+      "grad_norm": 0.4220462441444397,
+      "learning_rate": 0.0001826348270414703,
+      "loss": 0.7234170436859131,
+      "step": 3630
+    },
+    {
+      "epoch": 1.5324894514767933,
+      "grad_norm": 0.4787490665912628,
+      "learning_rate": 0.00018260776939171814,
+      "loss": 0.7864061594009399,
+      "step": 3632
+    },
+    {
+      "epoch": 1.5333333333333332,
+      "grad_norm": 0.3670683205127716,
+      "learning_rate": 0.00018258069268594013,
+      "loss": 0.670537531375885,
+      "step": 3634
+    },
+    {
+      "epoch": 1.5341772151898734,
+      "grad_norm": 0.41599538922309875,
+      "learning_rate": 0.0001825535969303823,
+      "loss": 0.7535691261291504,
+      "step": 3636
+    },
+    {
+      "epoch": 1.5350210970464135,
+      "grad_norm": 0.5033245086669922,
+      "learning_rate": 0.00018252648213129514,
+      "loss": 0.7553712725639343,
+      "step": 3638
+    },
+    {
+      "epoch": 1.5358649789029535,
+      "grad_norm": 0.4235823452472687,
+      "learning_rate": 0.0001824993482949335,
+      "loss": 0.7149441242218018,
+      "step": 3640
+    },
+    {
+      "epoch": 1.5367088607594936,
+      "grad_norm": 0.37993329763412476,
+      "learning_rate": 0.00018247219542755664,
+      "loss": 0.735568642616272,
+      "step": 3642
+    },
+    {
+      "epoch": 1.5375527426160338,
+      "grad_norm": 0.39534425735473633,
+      "learning_rate": 0.00018244502353542818,
+      "loss": 0.693510115146637,
+      "step": 3644
+    },
+    {
+      "epoch": 1.5383966244725737,
+      "grad_norm": 0.45447441935539246,
+      "learning_rate": 0.00018241783262481622,
+      "loss": 0.7261984348297119,
+      "step": 3646
+    },
+    {
+      "epoch": 1.539240506329114,
+      "grad_norm": 0.4131409823894501,
+      "learning_rate": 0.0001823906227019931,
+      "loss": 0.7518824338912964,
+      "step": 3648
+    },
+    {
+      "epoch": 1.540084388185654,
+      "grad_norm": 0.46603885293006897,
+      "learning_rate": 0.00018236339377323567,
+      "loss": 0.7650659680366516,
+      "step": 3650
+    },
+    {
+      "epoch": 1.540928270042194,
+      "grad_norm": 0.3998747766017914,
+      "learning_rate": 0.00018233614584482513,
+      "loss": 0.6436064839363098,
+      "step": 3652
+    },
+    {
+      "epoch": 1.5417721518987342,
+      "grad_norm": 0.4314839839935303,
+      "learning_rate": 0.00018230887892304703,
+      "loss": 0.7068451642990112,
+      "step": 3654
+    },
+    {
+      "epoch": 1.5426160337552743,
+      "grad_norm": 0.4324610233306885,
+      "learning_rate": 0.0001822815930141913,
+      "loss": 0.7342977523803711,
+      "step": 3656
+    },
+    {
+      "epoch": 1.5434599156118143,
+      "grad_norm": 0.4385458827018738,
+      "learning_rate": 0.0001822542881245524,
+      "loss": 0.6482785940170288,
+      "step": 3658
+    },
+    {
+      "epoch": 1.5443037974683544,
+      "grad_norm": 0.3736587166786194,
+      "learning_rate": 0.0001822269642604289,
+      "loss": 0.6319621205329895,
+      "step": 3660
+    },
+    {
+      "epoch": 1.5451476793248946,
+      "grad_norm": 0.42113035917282104,
+      "learning_rate": 0.00018219962142812393,
+      "loss": 0.6603307723999023,
+      "step": 3662
+    },
+    {
+      "epoch": 1.5459915611814345,
+      "grad_norm": 0.422158420085907,
+      "learning_rate": 0.00018217225963394496,
+      "loss": 0.6874772906303406,
+      "step": 3664
+    },
+    {
+      "epoch": 1.5468354430379747,
+      "grad_norm": 0.42724940180778503,
+      "learning_rate": 0.00018214487888420388,
+      "loss": 0.7231854796409607,
+      "step": 3666
+    },
+    {
+      "epoch": 1.5476793248945149,
+      "grad_norm": 0.3747844398021698,
+      "learning_rate": 0.00018211747918521685,
+      "loss": 0.6418231129646301,
+      "step": 3668
+    },
+    {
+      "epoch": 1.5485232067510548,
+      "grad_norm": 0.37007269263267517,
+      "learning_rate": 0.00018209006054330446,
+      "loss": 0.7019823789596558,
+      "step": 3670
+    },
+    {
+      "epoch": 1.549367088607595,
+      "grad_norm": 0.409701943397522,
+      "learning_rate": 0.00018206262296479168,
+      "loss": 0.6632764339447021,
+      "step": 3672
+    },
+    {
+      "epoch": 1.5502109704641351,
+      "grad_norm": 0.49566221237182617,
+      "learning_rate": 0.00018203516645600782,
+      "loss": 0.7327075600624084,
+      "step": 3674
+    },
+    {
+      "epoch": 1.551054852320675,
+      "grad_norm": 0.4549940526485443,
+      "learning_rate": 0.0001820076910232866,
+      "loss": 0.6535149812698364,
+      "step": 3676
+    },
+    {
+      "epoch": 1.5518987341772152,
+      "grad_norm": 0.42619478702545166,
+      "learning_rate": 0.000181980196672966,
+      "loss": 0.7761320471763611,
+      "step": 3678
+    },
+    {
+      "epoch": 1.5527426160337554,
+      "grad_norm": 0.4332887828350067,
+      "learning_rate": 0.00018195268341138852,
+      "loss": 0.6728626489639282,
+      "step": 3680
+    },
+    {
+      "epoch": 1.5535864978902953,
+      "grad_norm": 0.4972975552082062,
+      "learning_rate": 0.0001819251512449009,
+      "loss": 0.7369232177734375,
+      "step": 3682
+    },
+    {
+      "epoch": 1.5544303797468353,
+      "grad_norm": 0.3619556725025177,
+      "learning_rate": 0.00018189760017985428,
+      "loss": 0.6703663468360901,
+      "step": 3684
+    },
+    {
+      "epoch": 1.5552742616033757,
+      "grad_norm": 0.46435829997062683,
+      "learning_rate": 0.00018187003022260417,
+      "loss": 0.7183756828308105,
+      "step": 3686
+    },
+    {
+      "epoch": 1.5561181434599156,
+      "grad_norm": 0.4605063498020172,
+      "learning_rate": 0.00018184244137951045,
+      "loss": 0.680042564868927,
+      "step": 3688
+    },
+    {
+      "epoch": 1.5569620253164556,
+      "grad_norm": 0.4023124873638153,
+      "learning_rate": 0.00018181483365693732,
+      "loss": 0.679821789264679,
+      "step": 3690
+    },
+    {
+      "epoch": 1.557805907172996,
+      "grad_norm": 0.459499716758728,
+      "learning_rate": 0.00018178720706125333,
+      "loss": 0.7408921718597412,
+      "step": 3692
+    },
+    {
+      "epoch": 1.5586497890295359,
+      "grad_norm": 0.4397135078907013,
+      "learning_rate": 0.00018175956159883146,
+      "loss": 0.6926741003990173,
+      "step": 3694
+    },
+    {
+      "epoch": 1.5594936708860758,
+      "grad_norm": 0.4476170837879181,
+      "learning_rate": 0.0001817318972760489,
+      "loss": 0.752724826335907,
+      "step": 3696
+    },
+    {
+      "epoch": 1.560337552742616,
+      "grad_norm": 0.43684154748916626,
+      "learning_rate": 0.00018170421409928735,
+      "loss": 0.7217722535133362,
+      "step": 3698
+    },
+    {
+      "epoch": 1.5611814345991561,
+      "grad_norm": 0.4118131995201111,
+      "learning_rate": 0.00018167651207493278,
+      "loss": 0.7264936566352844,
+      "step": 3700
+    },
+    {
+      "epoch": 1.5611814345991561,
+      "eval_loss": 0.7854447960853577,
+      "eval_runtime": 687.7907,
+      "eval_samples_per_second": 3.063,
+      "eval_steps_per_second": 3.063,
+      "step": 3700
+    },
+    {
+      "epoch": 1.562025316455696,
+      "grad_norm": 0.40639030933380127,
+      "learning_rate": 0.0001816487912093755,
+      "loss": 0.701478123664856,
+      "step": 3702
+    },
+    {
+      "epoch": 1.5628691983122363,
+      "grad_norm": 0.4205699563026428,
+      "learning_rate": 0.00018162105150901017,
+      "loss": 0.696711003780365,
+      "step": 3704
+    },
+    {
+      "epoch": 1.5637130801687764,
+      "grad_norm": 0.46437183022499084,
+      "learning_rate": 0.0001815932929802358,
+      "loss": 0.7130293250083923,
+      "step": 3706
+    },
+    {
+      "epoch": 1.5645569620253164,
+      "grad_norm": 0.4413023591041565,
+      "learning_rate": 0.0001815655156294558,
+      "loss": 0.7180310487747192,
+      "step": 3708
+    },
+    {
+      "epoch": 1.5654008438818565,
+      "grad_norm": 0.3964478075504303,
+      "learning_rate": 0.00018153771946307783,
+      "loss": 0.6883325576782227,
+      "step": 3710
+    },
+    {
+      "epoch": 1.5662447257383967,
+      "grad_norm": 0.34486109018325806,
+      "learning_rate": 0.00018150990448751394,
+      "loss": 0.6683162450790405,
+      "step": 3712
+    },
+    {
+      "epoch": 1.5670886075949366,
+      "grad_norm": 0.4209035336971283,
+      "learning_rate": 0.00018148207070918052,
+      "loss": 0.6969789862632751,
+      "step": 3714
+    },
+    {
+      "epoch": 1.5679324894514768,
+      "grad_norm": 0.41674816608428955,
+      "learning_rate": 0.00018145421813449829,
+      "loss": 0.714320182800293,
+      "step": 3716
+    },
+    {
+      "epoch": 1.568776371308017,
+      "grad_norm": 0.4365934729576111,
+      "learning_rate": 0.00018142634676989228,
+      "loss": 0.7125279307365417,
+      "step": 3718
+    },
+    {
+      "epoch": 1.5696202531645569,
+      "grad_norm": 0.43599721789360046,
+      "learning_rate": 0.00018139845662179188,
+      "loss": 0.6787398457527161,
+      "step": 3720
+    },
+    {
+      "epoch": 1.570464135021097,
+      "grad_norm": 0.36320003867149353,
+      "learning_rate": 0.0001813705476966308,
+      "loss": 0.6401328444480896,
+      "step": 3722
+    },
+    {
+      "epoch": 1.5713080168776372,
+      "grad_norm": 0.43504419922828674,
+      "learning_rate": 0.00018134262000084718,
+      "loss": 0.6875129342079163,
+      "step": 3724
+    },
+    {
+      "epoch": 1.5721518987341772,
+      "grad_norm": 0.476723313331604,
+      "learning_rate": 0.00018131467354088332,
+      "loss": 0.6903918385505676,
+      "step": 3726
+    },
+    {
+      "epoch": 1.5729957805907173,
+      "grad_norm": 0.4465550184249878,
+      "learning_rate": 0.00018128670832318592,
+      "loss": 0.690018892288208,
+      "step": 3728
+    },
+    {
+      "epoch": 1.5738396624472575,
+      "grad_norm": 0.43816477060317993,
+      "learning_rate": 0.0001812587243542061,
+      "loss": 0.6912158727645874,
+      "step": 3730
+    },
+    {
+      "epoch": 1.5746835443037974,
+      "grad_norm": 0.40546250343322754,
+      "learning_rate": 0.00018123072164039913,
+      "loss": 0.7415492534637451,
+      "step": 3732
+    },
+    {
+      "epoch": 1.5755274261603376,
+      "grad_norm": 0.4253178536891937,
+      "learning_rate": 0.00018120270018822477,
+      "loss": 0.7664727568626404,
+      "step": 3734
+    },
+    {
+      "epoch": 1.5763713080168777,
+      "grad_norm": 0.503524661064148,
+      "learning_rate": 0.000181174660004147,
+      "loss": 0.782779335975647,
+      "step": 3736
+    },
+    {
+      "epoch": 1.5772151898734177,
+      "grad_norm": 0.5260395407676697,
+      "learning_rate": 0.00018114660109463414,
+      "loss": 0.806460976600647,
+      "step": 3738
+    },
+    {
+      "epoch": 1.5780590717299579,
+      "grad_norm": 0.3641359806060791,
+      "learning_rate": 0.0001811185234661589,
+      "loss": 0.6859704852104187,
+      "step": 3740
+    },
+    {
+      "epoch": 1.578902953586498,
+      "grad_norm": 0.4468916058540344,
+      "learning_rate": 0.00018109042712519818,
+      "loss": 0.7798553109169006,
+      "step": 3742
+    },
+    {
+      "epoch": 1.579746835443038,
+      "grad_norm": 0.4078207015991211,
+      "learning_rate": 0.00018106231207823327,
+      "loss": 0.7086471915245056,
+      "step": 3744
+    },
+    {
+      "epoch": 1.580590717299578,
+      "grad_norm": 0.39144742488861084,
+      "learning_rate": 0.00018103417833174981,
+      "loss": 0.6669760346412659,
+      "step": 3746
+    },
+    {
+      "epoch": 1.5814345991561183,
+      "grad_norm": 0.4366845190525055,
+      "learning_rate": 0.0001810060258922377,
+      "loss": 0.7296954989433289,
+      "step": 3748
+    },
+    {
+      "epoch": 1.5822784810126582,
+      "grad_norm": 0.46795082092285156,
+      "learning_rate": 0.00018097785476619118,
+      "loss": 0.7663856148719788,
+      "step": 3750
+    },
+    {
+      "epoch": 1.5831223628691982,
+      "grad_norm": 0.40331873297691345,
+      "learning_rate": 0.00018094966496010878,
+      "loss": 0.7105234861373901,
+      "step": 3752
+    },
+    {
+      "epoch": 1.5839662447257385,
+      "grad_norm": 0.38210469484329224,
+      "learning_rate": 0.00018092145648049334,
+      "loss": 0.666635274887085,
+      "step": 3754
+    },
+    {
+      "epoch": 1.5848101265822785,
+      "grad_norm": 0.45946040749549866,
+      "learning_rate": 0.000180893229333852,
+      "loss": 0.7287175059318542,
+      "step": 3756
+    },
+    {
+      "epoch": 1.5856540084388184,
+      "grad_norm": 0.4035777747631073,
+      "learning_rate": 0.00018086498352669624,
+      "loss": 0.7145618796348572,
+      "step": 3758
+    },
+    {
+      "epoch": 1.5864978902953588,
+      "grad_norm": 0.42843320965766907,
+      "learning_rate": 0.0001808367190655418,
+      "loss": 0.6550979018211365,
+      "step": 3760
+    },
+    {
+      "epoch": 1.5873417721518988,
+      "grad_norm": 0.4016468822956085,
+      "learning_rate": 0.0001808084359569088,
+      "loss": 0.7040430307388306,
+      "step": 3762
+    },
+    {
+      "epoch": 1.5881856540084387,
+      "grad_norm": 0.4126848578453064,
+      "learning_rate": 0.00018078013420732157,
+      "loss": 0.7523179054260254,
+      "step": 3764
+    },
+    {
+      "epoch": 1.5890295358649789,
+      "grad_norm": 0.45878592133522034,
+      "learning_rate": 0.00018075181382330878,
+      "loss": 0.7497967481613159,
+      "step": 3766
+    },
+    {
+      "epoch": 1.589873417721519,
+      "grad_norm": 0.44179272651672363,
+      "learning_rate": 0.00018072347481140345,
+      "loss": 0.7694459557533264,
+      "step": 3768
+    },
+    {
+      "epoch": 1.590717299578059,
+      "grad_norm": 0.3706412613391876,
+      "learning_rate": 0.00018069511717814276,
+      "loss": 0.6766634583473206,
+      "step": 3770
+    },
+    {
+      "epoch": 1.5915611814345991,
+      "grad_norm": 0.40555453300476074,
+      "learning_rate": 0.0001806667409300683,
+      "loss": 0.6394028067588806,
+      "step": 3772
+    },
+    {
+      "epoch": 1.5924050632911393,
+      "grad_norm": 0.3752219080924988,
+      "learning_rate": 0.000180638346073726,
+      "loss": 0.6775637865066528,
+      "step": 3774
+    },
+    {
+      "epoch": 1.5932489451476792,
+      "grad_norm": 0.40960127115249634,
+      "learning_rate": 0.00018060993261566593,
+      "loss": 0.7200632095336914,
+      "step": 3776
+    },
+    {
+      "epoch": 1.5940928270042194,
+      "grad_norm": 0.4215535521507263,
+      "learning_rate": 0.00018058150056244254,
+      "loss": 0.7314276695251465,
+      "step": 3778
+    },
+    {
+      "epoch": 1.5949367088607596,
+      "grad_norm": 0.42734605073928833,
+      "learning_rate": 0.00018055304992061456,
+      "loss": 0.7954897284507751,
+      "step": 3780
+    },
+    {
+      "epoch": 1.5957805907172995,
+      "grad_norm": 0.3980397582054138,
+      "learning_rate": 0.00018052458069674505,
+      "loss": 0.6552245020866394,
+      "step": 3782
+    },
+    {
+      "epoch": 1.5966244725738397,
+      "grad_norm": 0.44602933526039124,
+      "learning_rate": 0.00018049609289740124,
+      "loss": 0.7638092041015625,
+      "step": 3784
+    },
+    {
+      "epoch": 1.5974683544303798,
+      "grad_norm": 0.43835216760635376,
+      "learning_rate": 0.00018046758652915474,
+      "loss": 0.6886481046676636,
+      "step": 3786
+    },
+    {
+      "epoch": 1.5983122362869198,
+      "grad_norm": 0.38229402899742126,
+      "learning_rate": 0.00018043906159858147,
+      "loss": 0.6726123690605164,
+      "step": 3788
+    },
+    {
+      "epoch": 1.59915611814346,
+      "grad_norm": 0.43049392104148865,
+      "learning_rate": 0.00018041051811226155,
+      "loss": 0.7416896224021912,
+      "step": 3790
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 0.4027816951274872,
+      "learning_rate": 0.00018038195607677943,
+      "loss": 0.682691752910614,
+      "step": 3792
+    },
+    {
+      "epoch": 1.60084388185654,
+      "grad_norm": 0.4143599569797516,
+      "learning_rate": 0.00018035337549872376,
+      "loss": 0.7199558615684509,
+      "step": 3794
+    },
+    {
+      "epoch": 1.6016877637130802,
+      "grad_norm": 0.41221415996551514,
+      "learning_rate": 0.00018032477638468762,
+      "loss": 0.6586267948150635,
+      "step": 3796
+    },
+    {
+      "epoch": 1.6025316455696204,
+      "grad_norm": 0.40655994415283203,
+      "learning_rate": 0.0001802961587412682,
+      "loss": 0.7579189538955688,
+      "step": 3798
+    },
+    {
+      "epoch": 1.6033755274261603,
+      "grad_norm": 0.4163990914821625,
+      "learning_rate": 0.00018026752257506708,
+      "loss": 0.7246071100234985,
+      "step": 3800
+    },
+    {
+      "epoch": 1.6033755274261603,
+      "eval_loss": 0.778353214263916,
+      "eval_runtime": 692.5522,
+      "eval_samples_per_second": 3.042,
+      "eval_steps_per_second": 3.042,
+      "step": 3800
+    },
+    {
+      "epoch": 1.6042194092827005,
+      "grad_norm": 0.378628671169281,
+      "learning_rate": 0.0001802388678926901,
+      "loss": 0.7114223837852478,
+      "step": 3802
+    },
+    {
+      "epoch": 1.6050632911392406,
+      "grad_norm": 0.5050750374794006,
+      "learning_rate": 0.0001802101947007473,
+      "loss": 0.7519527077674866,
+      "step": 3804
+    },
+    {
+      "epoch": 1.6059071729957806,
+      "grad_norm": 0.5003950595855713,
+      "learning_rate": 0.00018018150300585305,
+      "loss": 0.7553771138191223,
+      "step": 3806
+    },
+    {
+      "epoch": 1.6067510548523207,
+      "grad_norm": 0.3851728141307831,
+      "learning_rate": 0.00018015279281462595,
+      "loss": 0.6584326028823853,
+      "step": 3808
+    },
+    {
+      "epoch": 1.6075949367088609,
+      "grad_norm": 0.41256240010261536,
+      "learning_rate": 0.00018012406413368893,
+      "loss": 0.7215287089347839,
+      "step": 3810
+    },
+    {
+      "epoch": 1.6084388185654008,
+      "grad_norm": 0.41479718685150146,
+      "learning_rate": 0.00018009531696966912,
+      "loss": 0.7753872275352478,
+      "step": 3812
+    },
+    {
+      "epoch": 1.6092827004219408,
+      "grad_norm": 0.398509681224823,
+      "learning_rate": 0.00018006655132919797,
+      "loss": 0.6984832882881165,
+      "step": 3814
+    },
+    {
+      "epoch": 1.6101265822784812,
+      "grad_norm": 0.42210307717323303,
+      "learning_rate": 0.0001800377672189111,
+      "loss": 0.8117644786834717,
+      "step": 3816
+    },
+    {
+      "epoch": 1.610970464135021,
+      "grad_norm": 0.3714556396007538,
+      "learning_rate": 0.0001800089646454485,
+      "loss": 0.6302977204322815,
+      "step": 3818
+    },
+    {
+      "epoch": 1.611814345991561,
+      "grad_norm": 0.4478890299797058,
+      "learning_rate": 0.00017998014361545438,
+      "loss": 0.7327325940132141,
+      "step": 3820
+    },
+    {
+      "epoch": 1.6126582278481014,
+      "grad_norm": 0.43516477942466736,
+      "learning_rate": 0.00017995130413557713,
+      "loss": 0.6877713203430176,
+      "step": 3822
+    },
+    {
+      "epoch": 1.6135021097046414,
+      "grad_norm": 0.42912542819976807,
+      "learning_rate": 0.0001799224462124695,
+      "loss": 0.6797093152999878,
+      "step": 3824
+    },
+    {
+      "epoch": 1.6143459915611813,
+      "grad_norm": 0.40646228194236755,
+      "learning_rate": 0.00017989356985278853,
+      "loss": 0.7402470111846924,
+      "step": 3826
+    },
+    {
+      "epoch": 1.6151898734177215,
+      "grad_norm": 0.42576655745506287,
+      "learning_rate": 0.00017986467506319535,
+      "loss": 0.6698583364486694,
+      "step": 3828
+    },
+    {
+      "epoch": 1.6160337552742616,
+      "grad_norm": 0.4298654794692993,
+      "learning_rate": 0.0001798357618503555,
+      "loss": 0.6736086010932922,
+      "step": 3830
+    },
+    {
+      "epoch": 1.6168776371308016,
+      "grad_norm": 0.4205976128578186,
+      "learning_rate": 0.0001798068302209386,
+      "loss": 0.7724249958992004,
+      "step": 3832
+    },
+    {
+      "epoch": 1.6177215189873417,
+      "grad_norm": 0.39669108390808105,
+      "learning_rate": 0.00017977788018161872,
+      "loss": 0.6468521952629089,
+      "step": 3834
+    },
+    {
+      "epoch": 1.618565400843882,
+      "grad_norm": 0.428551584482193,
+      "learning_rate": 0.00017974891173907405,
+      "loss": 0.7086918950080872,
+      "step": 3836
+    },
+    {
+      "epoch": 1.6194092827004218,
+      "grad_norm": 0.43773409724235535,
+      "learning_rate": 0.00017971992489998703,
+      "loss": 0.6483711004257202,
+      "step": 3838
+    },
+    {
+      "epoch": 1.620253164556962,
+      "grad_norm": 0.40540891885757446,
+      "learning_rate": 0.00017969091967104438,
+      "loss": 0.6947301030158997,
+      "step": 3840
+    },
+    {
+      "epoch": 1.6210970464135022,
+      "grad_norm": 0.3685019016265869,
+      "learning_rate": 0.00017966189605893708,
+      "loss": 0.667517900466919,
+      "step": 3842
+    },
+    {
+      "epoch": 1.621940928270042,
+      "grad_norm": 0.44813284277915955,
+      "learning_rate": 0.0001796328540703603,
+      "loss": 0.6957448124885559,
+      "step": 3844
+    },
+    {
+      "epoch": 1.6227848101265823,
+      "grad_norm": 0.4524349868297577,
+      "learning_rate": 0.00017960379371201345,
+      "loss": 0.7735851407051086,
+      "step": 3846
+    },
+    {
+      "epoch": 1.6236286919831224,
+      "grad_norm": 0.42349138855934143,
+      "learning_rate": 0.0001795747149906002,
+      "loss": 0.7052056193351746,
+      "step": 3848
+    },
+    {
+      "epoch": 1.6244725738396624,
+      "grad_norm": 0.47317537665367126,
+      "learning_rate": 0.0001795456179128285,
+      "loss": 0.7402311563491821,
+      "step": 3850
+    },
+    {
+      "epoch": 1.6253164556962025,
+      "grad_norm": 0.38915711641311646,
+      "learning_rate": 0.0001795165024854104,
+      "loss": 0.6743243932723999,
+      "step": 3852
+    },
+    {
+      "epoch": 1.6261603375527427,
+      "grad_norm": 0.4430672526359558,
+      "learning_rate": 0.00017948736871506235,
+      "loss": 0.6983860731124878,
+      "step": 3854
+    },
+    {
+      "epoch": 1.6270042194092826,
+      "grad_norm": 0.44286689162254333,
+      "learning_rate": 0.00017945821660850494,
+      "loss": 0.8012514710426331,
+      "step": 3856
+    },
+    {
+      "epoch": 1.6278481012658228,
+      "grad_norm": 0.43364086747169495,
+      "learning_rate": 0.00017942904617246295,
+      "loss": 0.8064966201782227,
+      "step": 3858
+    },
+    {
+      "epoch": 1.628691983122363,
+      "grad_norm": 0.45848122239112854,
+      "learning_rate": 0.00017939985741366547,
+      "loss": 0.7069969177246094,
+      "step": 3860
+    },
+    {
+      "epoch": 1.629535864978903,
+      "grad_norm": 0.4199717044830322,
+      "learning_rate": 0.0001793706503388458,
+      "loss": 0.6722052097320557,
+      "step": 3862
+    },
+    {
+      "epoch": 1.630379746835443,
+      "grad_norm": 0.3820365071296692,
+      "learning_rate": 0.00017934142495474142,
+      "loss": 0.663442075252533,
+      "step": 3864
+    },
+    {
+      "epoch": 1.6312236286919832,
+      "grad_norm": 0.4470944106578827,
+      "learning_rate": 0.00017931218126809412,
+      "loss": 0.70677250623703,
+      "step": 3866
+    },
+    {
+      "epoch": 1.6320675105485232,
+      "grad_norm": 0.3941600024700165,
+      "learning_rate": 0.0001792829192856498,
+      "loss": 0.7125341892242432,
+      "step": 3868
+    },
+    {
+      "epoch": 1.6329113924050633,
+      "grad_norm": 0.44924396276474,
+      "learning_rate": 0.00017925363901415863,
+      "loss": 0.7245699763298035,
+      "step": 3870
+    },
+    {
+      "epoch": 1.6337552742616035,
+      "grad_norm": 0.4070027470588684,
+      "learning_rate": 0.00017922434046037508,
+      "loss": 0.6977120637893677,
+      "step": 3872
+    },
+    {
+      "epoch": 1.6345991561181434,
+      "grad_norm": 0.45040908455848694,
+      "learning_rate": 0.00017919502363105772,
+      "loss": 0.7631482481956482,
+      "step": 3874
+    },
+    {
+      "epoch": 1.6354430379746834,
+      "grad_norm": 0.4083724617958069,
+      "learning_rate": 0.00017916568853296934,
+      "loss": 0.6656452417373657,
+      "step": 3876
+    },
+    {
+      "epoch": 1.6362869198312238,
+      "grad_norm": 0.3762308359146118,
+      "learning_rate": 0.00017913633517287704,
+      "loss": 0.6302213072776794,
+      "step": 3878
+    },
+    {
+      "epoch": 1.6371308016877637,
+      "grad_norm": 0.36366912722587585,
+      "learning_rate": 0.00017910696355755205,
+      "loss": 0.7119566202163696,
+      "step": 3880
+    },
+    {
+      "epoch": 1.6379746835443036,
+      "grad_norm": 0.43168532848358154,
+      "learning_rate": 0.00017907757369376985,
+      "loss": 0.7271678447723389,
+      "step": 3882
+    },
+    {
+      "epoch": 1.638818565400844,
+      "grad_norm": 0.37720009684562683,
+      "learning_rate": 0.00017904816558831014,
+      "loss": 0.6340095400810242,
+      "step": 3884
+    },
+    {
+      "epoch": 1.639662447257384,
+      "grad_norm": 0.36539995670318604,
+      "learning_rate": 0.00017901873924795677,
+      "loss": 0.6910572052001953,
+      "step": 3886
+    },
+    {
+      "epoch": 1.640506329113924,
+      "grad_norm": 0.42437905073165894,
+      "learning_rate": 0.00017898929467949782,
+      "loss": 0.7570282220840454,
+      "step": 3888
+    },
+    {
+      "epoch": 1.6413502109704643,
+      "grad_norm": 0.43482181429862976,
+      "learning_rate": 0.00017895983188972558,
+      "loss": 0.7835264205932617,
+      "step": 3890
+    },
+    {
+      "epoch": 1.6421940928270042,
+      "grad_norm": 0.40460237860679626,
+      "learning_rate": 0.00017893035088543662,
+      "loss": 0.7058968544006348,
+      "step": 3892
+    },
+    {
+      "epoch": 1.6430379746835442,
+      "grad_norm": 0.39779701828956604,
+      "learning_rate": 0.00017890085167343158,
+      "loss": 0.723969042301178,
+      "step": 3894
+    },
+    {
+      "epoch": 1.6438818565400843,
+      "grad_norm": 0.46131494641304016,
+      "learning_rate": 0.00017887133426051536,
+      "loss": 0.7773922681808472,
+      "step": 3896
+    },
+    {
+      "epoch": 1.6447257383966245,
+      "grad_norm": 0.48412877321243286,
+      "learning_rate": 0.0001788417986534971,
+      "loss": 0.7742162942886353,
+      "step": 3898
+    },
+    {
+      "epoch": 1.6455696202531644,
+      "grad_norm": 0.40083935856819153,
+      "learning_rate": 0.00017881224485919006,
+      "loss": 0.6674869656562805,
+      "step": 3900
+    },
+    {
+      "epoch": 1.6455696202531644,
+      "eval_loss": 0.7763351202011108,
+      "eval_runtime": 682.0824,
+      "eval_samples_per_second": 3.089,
+      "eval_steps_per_second": 3.089,
+      "step": 3900
+    },
+    {
+      "epoch": 1.6464135021097046,
+      "grad_norm": 0.3944476842880249,
+      "learning_rate": 0.00017878267288441176,
+      "loss": 0.6777821183204651,
+      "step": 3902
+    },
+    {
+      "epoch": 1.6472573839662448,
+      "grad_norm": 0.44863224029541016,
+      "learning_rate": 0.0001787530827359839,
+      "loss": 0.7398322224617004,
+      "step": 3904
+    },
+    {
+      "epoch": 1.6481012658227847,
+      "grad_norm": 0.4665529131889343,
+      "learning_rate": 0.00017872347442073232,
+      "loss": 0.7203068733215332,
+      "step": 3906
+    },
+    {
+      "epoch": 1.6489451476793249,
+      "grad_norm": 0.3603050410747528,
+      "learning_rate": 0.00017869384794548708,
+      "loss": 0.6232115030288696,
+      "step": 3908
+    },
+    {
+      "epoch": 1.649789029535865,
+      "grad_norm": 0.3545325696468353,
+      "learning_rate": 0.0001786642033170825,
+      "loss": 0.6120040416717529,
+      "step": 3910
+    },
+    {
+      "epoch": 1.650632911392405,
+      "grad_norm": 0.3947487473487854,
+      "learning_rate": 0.00017863454054235697,
+      "loss": 0.7285997867584229,
+      "step": 3912
+    },
+    {
+      "epoch": 1.6514767932489451,
+      "grad_norm": 0.33153054118156433,
+      "learning_rate": 0.00017860485962815312,
+      "loss": 0.6486109495162964,
+      "step": 3914
+    },
+    {
+      "epoch": 1.6523206751054853,
+      "grad_norm": 0.3886185884475708,
+      "learning_rate": 0.00017857516058131782,
+      "loss": 0.6870376467704773,
+      "step": 3916
+    },
+    {
+      "epoch": 1.6531645569620252,
+      "grad_norm": 0.46399059891700745,
+      "learning_rate": 0.00017854544340870201,
+      "loss": 0.7351298928260803,
+      "step": 3918
+    },
+    {
+      "epoch": 1.6540084388185654,
+      "grad_norm": 0.49280229210853577,
+      "learning_rate": 0.00017851570811716093,
+      "loss": 0.7283915281295776,
+      "step": 3920
+    },
+    {
+      "epoch": 1.6548523206751056,
+      "grad_norm": 0.41570642590522766,
+      "learning_rate": 0.00017848595471355388,
+      "loss": 0.7436906099319458,
+      "step": 3922
+    },
+    {
+      "epoch": 1.6556962025316455,
+      "grad_norm": 0.44591766595840454,
+      "learning_rate": 0.00017845618320474444,
+      "loss": 0.7427505254745483,
+      "step": 3924
+    },
+    {
+      "epoch": 1.6565400843881857,
+      "grad_norm": 0.4804318845272064,
+      "learning_rate": 0.00017842639359760032,
+      "loss": 0.7642725706100464,
+      "step": 3926
+    },
+    {
+      "epoch": 1.6573839662447258,
+      "grad_norm": 0.4164349138736725,
+      "learning_rate": 0.00017839658589899343,
+      "loss": 0.7097790837287903,
+      "step": 3928
+    },
+    {
+      "epoch": 1.6582278481012658,
+      "grad_norm": 0.5307883620262146,
+      "learning_rate": 0.00017836676011579976,
+      "loss": 0.8346033096313477,
+      "step": 3930
+    },
+    {
+      "epoch": 1.659071729957806,
+      "grad_norm": 0.37231504917144775,
+      "learning_rate": 0.0001783369162548996,
+      "loss": 0.6327160596847534,
+      "step": 3932
+    },
+    {
+      "epoch": 1.659915611814346,
+      "grad_norm": 0.4178757965564728,
+      "learning_rate": 0.00017830705432317738,
+      "loss": 0.6441237330436707,
+      "step": 3934
+    },
+    {
+      "epoch": 1.660759493670886,
+      "grad_norm": 0.42698073387145996,
+      "learning_rate": 0.00017827717432752162,
+      "loss": 0.7267951369285583,
+      "step": 3936
+    },
+    {
+      "epoch": 1.6616033755274262,
+      "grad_norm": 0.48763108253479004,
+      "learning_rate": 0.0001782472762748251,
+      "loss": 0.7248380780220032,
+      "step": 3938
+    },
+    {
+      "epoch": 1.6624472573839664,
+      "grad_norm": 0.38882505893707275,
+      "learning_rate": 0.0001782173601719847,
+      "loss": 0.6665275692939758,
+      "step": 3940
+    },
+    {
+      "epoch": 1.6632911392405063,
+      "grad_norm": 0.4684160649776459,
+      "learning_rate": 0.0001781874260259015,
+      "loss": 0.8017846941947937,
+      "step": 3942
+    },
+    {
+      "epoch": 1.6641350210970463,
+      "grad_norm": 0.4200204908847809,
+      "learning_rate": 0.00017815747384348075,
+      "loss": 0.6810902953147888,
+      "step": 3944
+    },
+    {
+      "epoch": 1.6649789029535866,
+      "grad_norm": 0.45843279361724854,
+      "learning_rate": 0.0001781275036316318,
+      "loss": 0.714082658290863,
+      "step": 3946
+    },
+    {
+      "epoch": 1.6658227848101266,
+      "grad_norm": 0.522654116153717,
+      "learning_rate": 0.00017809751539726824,
+      "loss": 0.842085599899292,
+      "step": 3948
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 0.43468809127807617,
+      "learning_rate": 0.00017806750914730777,
+      "loss": 0.6917972564697266,
+      "step": 3950
+    },
+    {
+      "epoch": 1.667510548523207,
+      "grad_norm": 0.4071558117866516,
+      "learning_rate": 0.00017803748488867224,
+      "loss": 0.6676424741744995,
+      "step": 3952
+    },
+    {
+      "epoch": 1.6683544303797468,
+      "grad_norm": 0.4478702247142792,
+      "learning_rate": 0.00017800744262828772,
+      "loss": 0.7563748359680176,
+      "step": 3954
+    },
+    {
+      "epoch": 1.6691983122362868,
+      "grad_norm": 0.42986053228378296,
+      "learning_rate": 0.00017797738237308432,
+      "loss": 0.685714066028595,
+      "step": 3956
+    },
+    {
+      "epoch": 1.6700421940928272,
+      "grad_norm": 0.4410569667816162,
+      "learning_rate": 0.0001779473041299964,
+      "loss": 0.7753605842590332,
+      "step": 3958
+    },
+    {
+      "epoch": 1.6708860759493671,
+      "grad_norm": 0.44055965542793274,
+      "learning_rate": 0.00017791720790596242,
+      "loss": 0.7826018333435059,
+      "step": 3960
+    },
+    {
+      "epoch": 1.671729957805907,
+      "grad_norm": 0.4660222828388214,
+      "learning_rate": 0.00017788709370792502,
+      "loss": 0.7012629508972168,
+      "step": 3962
+    },
+    {
+      "epoch": 1.6725738396624472,
+      "grad_norm": 0.3655732274055481,
+      "learning_rate": 0.00017785696154283093,
+      "loss": 0.6647007465362549,
+      "step": 3964
+    },
+    {
+      "epoch": 1.6734177215189874,
+      "grad_norm": 0.45735979080200195,
+      "learning_rate": 0.0001778268114176311,
+      "loss": 0.7238346338272095,
+      "step": 3966
+    },
+    {
+      "epoch": 1.6742616033755273,
+      "grad_norm": 0.4894043207168579,
+      "learning_rate": 0.00017779664333928053,
+      "loss": 0.7595300078392029,
+      "step": 3968
+    },
+    {
+      "epoch": 1.6751054852320675,
+      "grad_norm": 0.3613647222518921,
+      "learning_rate": 0.00017776645731473848,
+      "loss": 0.6534504294395447,
+      "step": 3970
+    },
+    {
+      "epoch": 1.6759493670886076,
+      "grad_norm": 0.4402320683002472,
+      "learning_rate": 0.00017773625335096828,
+      "loss": 0.7699148058891296,
+      "step": 3972
+    },
+    {
+      "epoch": 1.6767932489451476,
+      "grad_norm": 0.4391469955444336,
+      "learning_rate": 0.0001777060314549374,
+      "loss": 0.771800696849823,
+      "step": 3974
+    },
+    {
+      "epoch": 1.6776371308016877,
+      "grad_norm": 0.41348180174827576,
+      "learning_rate": 0.00017767579163361735,
+      "loss": 0.6957393884658813,
+      "step": 3976
+    },
+    {
+      "epoch": 1.678481012658228,
+      "grad_norm": 0.4115046560764313,
+      "learning_rate": 0.00017764553389398407,
+      "loss": 0.6666664481163025,
+      "step": 3978
+    },
+    {
+      "epoch": 1.6793248945147679,
+      "grad_norm": 0.44647306203842163,
+      "learning_rate": 0.00017761525824301725,
+      "loss": 0.7364352345466614,
+      "step": 3980
+    },
+    {
+      "epoch": 1.680168776371308,
+      "grad_norm": 0.424234002828598,
+      "learning_rate": 0.00017758496468770103,
+      "loss": 0.7229774594306946,
+      "step": 3982
+    },
+    {
+      "epoch": 1.6810126582278482,
+      "grad_norm": 0.4850127696990967,
+      "learning_rate": 0.00017755465323502346,
+      "loss": 0.6992678642272949,
+      "step": 3984
+    },
+    {
+      "epoch": 1.6818565400843881,
+      "grad_norm": 0.44637173414230347,
+      "learning_rate": 0.00017752432389197688,
+      "loss": 0.6957634091377258,
+      "step": 3986
+    },
+    {
+      "epoch": 1.6827004219409283,
+      "grad_norm": 0.477393239736557,
+      "learning_rate": 0.00017749397666555768,
+      "loss": 0.6842859983444214,
+      "step": 3988
+    },
+    {
+      "epoch": 1.6835443037974684,
+      "grad_norm": 0.3634146451950073,
+      "learning_rate": 0.00017746361156276632,
+      "loss": 0.6478199362754822,
+      "step": 3990
+    },
+    {
+      "epoch": 1.6843881856540084,
+      "grad_norm": 0.4806389808654785,
+      "learning_rate": 0.00017743322859060752,
+      "loss": 0.7166973352432251,
+      "step": 3992
+    },
+    {
+      "epoch": 1.6852320675105485,
+      "grad_norm": 0.4273380935192108,
+      "learning_rate": 0.00017740282775609,
+      "loss": 0.6624643206596375,
+      "step": 3994
+    },
+    {
+      "epoch": 1.6860759493670887,
+      "grad_norm": 0.4340716600418091,
+      "learning_rate": 0.0001773724090662267,
+      "loss": 0.6408489346504211,
+      "step": 3996
+    },
+    {
+      "epoch": 1.6869198312236287,
+      "grad_norm": 0.50480717420578,
+      "learning_rate": 0.00017734197252803458,
+      "loss": 0.8096797466278076,
+      "step": 3998
+    },
+    {
+      "epoch": 1.6877637130801688,
+      "grad_norm": 0.4549327790737152,
+      "learning_rate": 0.00017731151814853475,
+      "loss": 0.6902434825897217,
+      "step": 4000
+    },
+    {
+      "epoch": 1.6877637130801688,
+      "eval_loss": 0.7721371650695801,
+      "eval_runtime": 668.395,
+      "eval_samples_per_second": 3.152,
+      "eval_steps_per_second": 3.152,
+      "step": 4000
+    },
+    {
+      "epoch": 1.688607594936709,
+      "grad_norm": 0.38909077644348145,
+      "learning_rate": 0.00017728104593475247,
+      "loss": 0.6510542035102844,
+      "step": 4002
+    },
+    {
+      "epoch": 1.689451476793249,
+      "grad_norm": 0.444608598947525,
+      "learning_rate": 0.00017725055589371715,
+      "loss": 0.7331066727638245,
+      "step": 4004
+    },
+    {
+      "epoch": 1.690295358649789,
+      "grad_norm": 0.3364178538322449,
+      "learning_rate": 0.0001772200480324622,
+      "loss": 0.5950609445571899,
+      "step": 4006
+    },
+    {
+      "epoch": 1.6911392405063292,
+      "grad_norm": 0.37264227867126465,
+      "learning_rate": 0.0001771895223580252,
+      "loss": 0.7024590969085693,
+      "step": 4008
+    },
+    {
+      "epoch": 1.6919831223628692,
+      "grad_norm": 0.3576899766921997,
+      "learning_rate": 0.00017715897887744784,
+      "loss": 0.6620895266532898,
+      "step": 4010
+    },
+    {
+      "epoch": 1.6928270042194091,
+      "grad_norm": 0.45855066180229187,
+      "learning_rate": 0.00017712841759777591,
+      "loss": 0.6726759076118469,
+      "step": 4012
+    },
+    {
+      "epoch": 1.6936708860759495,
+      "grad_norm": 0.44610634446144104,
+      "learning_rate": 0.00017709783852605933,
+      "loss": 0.6550953984260559,
+      "step": 4014
+    },
+    {
+      "epoch": 1.6945147679324895,
+      "grad_norm": 0.4262036681175232,
+      "learning_rate": 0.00017706724166935208,
+      "loss": 0.7338476777076721,
+      "step": 4016
+    },
+    {
+      "epoch": 1.6953586497890294,
+      "grad_norm": 0.44584009051322937,
+      "learning_rate": 0.0001770366270347123,
+      "loss": 0.6785868406295776,
+      "step": 4018
+    },
+    {
+      "epoch": 1.6962025316455698,
+      "grad_norm": 0.44828134775161743,
+      "learning_rate": 0.0001770059946292022,
+      "loss": 0.7113003134727478,
+      "step": 4020
+    },
+    {
+      "epoch": 1.6970464135021097,
+      "grad_norm": 0.39716270565986633,
+      "learning_rate": 0.00017697534445988803,
+      "loss": 0.6420866847038269,
+      "step": 4022
+    },
+    {
+      "epoch": 1.6978902953586497,
+      "grad_norm": 0.43546608090400696,
+      "learning_rate": 0.00017694467653384024,
+      "loss": 0.8087385892868042,
+      "step": 4024
+    },
+    {
+      "epoch": 1.6987341772151898,
+      "grad_norm": 0.3698647618293762,
+      "learning_rate": 0.00017691399085813334,
+      "loss": 0.6378461718559265,
+      "step": 4026
+    },
+    {
+      "epoch": 1.69957805907173,
+      "grad_norm": 0.4256688058376312,
+      "learning_rate": 0.0001768832874398459,
+      "loss": 0.6504215002059937,
+      "step": 4028
+    },
+    {
+      "epoch": 1.70042194092827,
+      "grad_norm": 0.4014805555343628,
+      "learning_rate": 0.00017685256628606063,
+      "loss": 0.6648795008659363,
+      "step": 4030
+    },
+    {
+      "epoch": 1.70126582278481,
+      "grad_norm": 0.43924233317375183,
+      "learning_rate": 0.00017682182740386427,
+      "loss": 0.6905859112739563,
+      "step": 4032
+    },
+    {
+      "epoch": 1.7021097046413503,
+      "grad_norm": 0.43565094470977783,
+      "learning_rate": 0.00017679107080034774,
+      "loss": 0.7800544500350952,
+      "step": 4034
+    },
+    {
+      "epoch": 1.7029535864978902,
+      "grad_norm": 0.4681089520454407,
+      "learning_rate": 0.00017676029648260597,
+      "loss": 0.7487648725509644,
+      "step": 4036
+    },
+    {
+      "epoch": 1.7037974683544304,
+      "grad_norm": 0.41400185227394104,
+      "learning_rate": 0.00017672950445773804,
+      "loss": 0.7374032735824585,
+      "step": 4038
+    },
+    {
+      "epoch": 1.7046413502109705,
+      "grad_norm": 0.5120159983634949,
+      "learning_rate": 0.00017669869473284702,
+      "loss": 0.7539847493171692,
+      "step": 4040
+    },
+    {
+      "epoch": 1.7054852320675105,
+      "grad_norm": 0.4043315052986145,
+      "learning_rate": 0.00017666786731504013,
+      "loss": 0.7029932737350464,
+      "step": 4042
+    },
+    {
+      "epoch": 1.7063291139240506,
+      "grad_norm": 0.462801992893219,
+      "learning_rate": 0.00017663702221142874,
+      "loss": 0.6464276313781738,
+      "step": 4044
+    },
+    {
+      "epoch": 1.7071729957805908,
+      "grad_norm": 0.4751932621002197,
+      "learning_rate": 0.00017660615942912813,
+      "loss": 0.7646113634109497,
+      "step": 4046
+    },
+    {
+      "epoch": 1.7080168776371307,
+      "grad_norm": 0.41720855236053467,
+      "learning_rate": 0.0001765752789752578,
+      "loss": 0.7191748023033142,
+      "step": 4048
+    },
+    {
+      "epoch": 1.7088607594936709,
+      "grad_norm": 0.40235280990600586,
+      "learning_rate": 0.00017654438085694128,
+      "loss": 0.6946488618850708,
+      "step": 4050
+    },
+    {
+      "epoch": 1.709704641350211,
+      "grad_norm": 0.4385916292667389,
+      "learning_rate": 0.00017651346508130612,
+      "loss": 0.7655195593833923,
+      "step": 4052
+    },
+    {
+      "epoch": 1.710548523206751,
+      "grad_norm": 0.4192262887954712,
+      "learning_rate": 0.00017648253165548406,
+      "loss": 0.7192418575286865,
+      "step": 4054
+    },
+    {
+      "epoch": 1.7113924050632912,
+      "grad_norm": 0.4348951280117035,
+      "learning_rate": 0.00017645158058661082,
+      "loss": 0.6501081585884094,
+      "step": 4056
+    },
+    {
+      "epoch": 1.7122362869198313,
+      "grad_norm": 0.4178772270679474,
+      "learning_rate": 0.0001764206118818262,
+      "loss": 0.741439163684845,
+      "step": 4058
+    },
+    {
+      "epoch": 1.7130801687763713,
+      "grad_norm": 0.46996670961380005,
+      "learning_rate": 0.00017638962554827414,
+      "loss": 0.7087798714637756,
+      "step": 4060
+    },
+    {
+      "epoch": 1.7139240506329114,
+      "grad_norm": 0.4775761365890503,
+      "learning_rate": 0.00017635862159310253,
+      "loss": 0.6716644167900085,
+      "step": 4062
+    },
+    {
+      "epoch": 1.7147679324894516,
+      "grad_norm": 0.4094069004058838,
+      "learning_rate": 0.00017632760002346345,
+      "loss": 0.6630500555038452,
+      "step": 4064
+    },
+    {
+      "epoch": 1.7156118143459915,
+      "grad_norm": 0.3968847393989563,
+      "learning_rate": 0.0001762965608465129,
+      "loss": 0.7678145170211792,
+      "step": 4066
+    },
+    {
+      "epoch": 1.7164556962025317,
+      "grad_norm": 0.43751946091651917,
+      "learning_rate": 0.0001762655040694111,
+      "loss": 0.6981875896453857,
+      "step": 4068
+    },
+    {
+      "epoch": 1.7172995780590719,
+      "grad_norm": 0.44504284858703613,
+      "learning_rate": 0.00017623442969932218,
+      "loss": 0.6919799447059631,
+      "step": 4070
+    },
+    {
+      "epoch": 1.7181434599156118,
+      "grad_norm": 0.36895114183425903,
+      "learning_rate": 0.00017620333774341447,
+      "loss": 0.6310772895812988,
+      "step": 4072
+    },
+    {
+      "epoch": 1.7189873417721517,
+      "grad_norm": 0.4312762916088104,
+      "learning_rate": 0.00017617222820886022,
+      "loss": 0.6954092383384705,
+      "step": 4074
+    },
+    {
+      "epoch": 1.7198312236286921,
+      "grad_norm": 0.40909886360168457,
+      "learning_rate": 0.00017614110110283587,
+      "loss": 0.6973698139190674,
+      "step": 4076
+    },
+    {
+      "epoch": 1.720675105485232,
+      "grad_norm": 0.43111470341682434,
+      "learning_rate": 0.0001761099564325218,
+      "loss": 0.6775513291358948,
+      "step": 4078
+    },
+    {
+      "epoch": 1.721518987341772,
+      "grad_norm": 0.45573189854621887,
+      "learning_rate": 0.0001760787942051025,
+      "loss": 0.6889033913612366,
+      "step": 4080
+    },
+    {
+      "epoch": 1.7223628691983124,
+      "grad_norm": 0.44934743642807007,
+      "learning_rate": 0.0001760476144277665,
+      "loss": 0.6894892454147339,
+      "step": 4082
+    },
+    {
+      "epoch": 1.7232067510548523,
+      "grad_norm": 0.39716506004333496,
+      "learning_rate": 0.0001760164171077064,
+      "loss": 0.6841691732406616,
+      "step": 4084
+    },
+    {
+      "epoch": 1.7240506329113923,
+      "grad_norm": 0.4486255645751953,
+      "learning_rate": 0.00017598520225211876,
+      "loss": 0.7197495102882385,
+      "step": 4086
+    },
+    {
+      "epoch": 1.7248945147679327,
+      "grad_norm": 0.40130025148391724,
+      "learning_rate": 0.00017595396986820432,
+      "loss": 0.7388272881507874,
+      "step": 4088
+    },
+    {
+      "epoch": 1.7257383966244726,
+      "grad_norm": 0.40419334173202515,
+      "learning_rate": 0.0001759227199631677,
+      "loss": 0.7447546720504761,
+      "step": 4090
+    },
+    {
+      "epoch": 1.7265822784810125,
+      "grad_norm": 0.39778804779052734,
+      "learning_rate": 0.00017589145254421774,
+      "loss": 0.6348387598991394,
+      "step": 4092
+    },
+    {
+      "epoch": 1.7274261603375527,
+      "grad_norm": 0.3934648036956787,
+      "learning_rate": 0.0001758601676185672,
+      "loss": 0.6813296675682068,
+      "step": 4094
+    },
+    {
+      "epoch": 1.7282700421940929,
+      "grad_norm": 0.4080725908279419,
+      "learning_rate": 0.0001758288651934329,
+      "loss": 0.6369423270225525,
+      "step": 4096
+    },
+    {
+      "epoch": 1.7291139240506328,
+      "grad_norm": 0.5186997652053833,
+      "learning_rate": 0.00017579754527603575,
+      "loss": 0.6664180159568787,
+      "step": 4098
+    },
+    {
+      "epoch": 1.729957805907173,
+      "grad_norm": 0.45290765166282654,
+      "learning_rate": 0.0001757662078736006,
+      "loss": 0.7181394696235657,
+      "step": 4100
+    },
+    {
+      "epoch": 1.729957805907173,
+      "eval_loss": 0.7690847516059875,
+      "eval_runtime": 673.6323,
+      "eval_samples_per_second": 3.128,
+      "eval_steps_per_second": 3.128,
+      "step": 4100
+    },
+    {
+      "epoch": 1.7308016877637131,
+      "grad_norm": 0.45677125453948975,
+      "learning_rate": 0.00017573485299335643,
+      "loss": 0.6850123405456543,
+      "step": 4102
+    },
+    {
+      "epoch": 1.731645569620253,
+      "grad_norm": 0.4854084253311157,
+      "learning_rate": 0.0001757034806425362,
+      "loss": 0.7225809693336487,
+      "step": 4104
+    },
+    {
+      "epoch": 1.7324894514767932,
+      "grad_norm": 0.486583411693573,
+      "learning_rate": 0.0001756720908283769,
+      "loss": 0.7303882837295532,
+      "step": 4106
+    },
+    {
+      "epoch": 1.7333333333333334,
+      "grad_norm": 0.4609545171260834,
+      "learning_rate": 0.00017564068355811952,
+      "loss": 0.6789878010749817,
+      "step": 4108
+    },
+    {
+      "epoch": 1.7341772151898733,
+      "grad_norm": 0.3853047788143158,
+      "learning_rate": 0.00017560925883900914,
+      "loss": 0.7311062812805176,
+      "step": 4110
+    },
+    {
+      "epoch": 1.7350210970464135,
+      "grad_norm": 0.41957035660743713,
+      "learning_rate": 0.00017557781667829487,
+      "loss": 0.7106070518493652,
+      "step": 4112
+    },
+    {
+      "epoch": 1.7358649789029537,
+      "grad_norm": 0.37419262528419495,
+      "learning_rate": 0.0001755463570832298,
+      "loss": 0.6345783472061157,
+      "step": 4114
+    },
+    {
+      "epoch": 1.7367088607594936,
+      "grad_norm": 0.5050614476203918,
+      "learning_rate": 0.00017551488006107104,
+      "loss": 0.7923876047134399,
+      "step": 4116
+    },
+    {
+      "epoch": 1.7375527426160338,
+      "grad_norm": 0.49726125597953796,
+      "learning_rate": 0.00017548338561907976,
+      "loss": 0.7502455115318298,
+      "step": 4118
+    },
+    {
+      "epoch": 1.738396624472574,
+      "grad_norm": 0.38956591486930847,
+      "learning_rate": 0.00017545187376452105,
+      "loss": 0.6333624720573425,
+      "step": 4120
+    },
+    {
+      "epoch": 1.7392405063291139,
+      "grad_norm": 0.430318146944046,
+      "learning_rate": 0.0001754203445046642,
+      "loss": 0.6884907484054565,
+      "step": 4122
+    },
+    {
+      "epoch": 1.740084388185654,
+      "grad_norm": 0.39826440811157227,
+      "learning_rate": 0.0001753887978467823,
+      "loss": 0.7345985174179077,
+      "step": 4124
+    },
+    {
+      "epoch": 1.7409282700421942,
+      "grad_norm": 0.4263492822647095,
+      "learning_rate": 0.00017535723379815267,
+      "loss": 0.6537752151489258,
+      "step": 4126
+    },
+    {
+      "epoch": 1.7417721518987341,
+      "grad_norm": 0.4012296199798584,
+      "learning_rate": 0.0001753256523660564,
+      "loss": 0.6368000507354736,
+      "step": 4128
+    },
+    {
+      "epoch": 1.7426160337552743,
+      "grad_norm": 0.3881204426288605,
+      "learning_rate": 0.0001752940535577788,
+      "loss": 0.6038334369659424,
+      "step": 4130
+    },
+    {
+      "epoch": 1.7434599156118145,
+      "grad_norm": 0.38410526514053345,
+      "learning_rate": 0.0001752624373806091,
+      "loss": 0.6657052636146545,
+      "step": 4132
+    },
+    {
+      "epoch": 1.7443037974683544,
+      "grad_norm": 0.3970552980899811,
+      "learning_rate": 0.00017523080384184057,
+      "loss": 0.6518114805221558,
+      "step": 4134
+    },
+    {
+      "epoch": 1.7451476793248946,
+      "grad_norm": 0.37542012333869934,
+      "learning_rate": 0.0001751991529487704,
+      "loss": 0.609853208065033,
+      "step": 4136
+    },
+    {
+      "epoch": 1.7459915611814347,
+      "grad_norm": 0.49812132120132446,
+      "learning_rate": 0.00017516748470869988,
+      "loss": 0.7356346845626831,
+      "step": 4138
+    },
+    {
+      "epoch": 1.7468354430379747,
+      "grad_norm": 0.48197710514068604,
+      "learning_rate": 0.00017513579912893426,
+      "loss": 0.7198674082756042,
+      "step": 4140
+    },
+    {
+      "epoch": 1.7476793248945146,
+      "grad_norm": 0.38266444206237793,
+      "learning_rate": 0.0001751040962167828,
+      "loss": 0.6841102242469788,
+      "step": 4142
+    },
+    {
+      "epoch": 1.748523206751055,
+      "grad_norm": 0.4165695309638977,
+      "learning_rate": 0.0001750723759795587,
+      "loss": 0.7348533272743225,
+      "step": 4144
+    },
+    {
+      "epoch": 1.749367088607595,
+      "grad_norm": 0.4414960443973541,
+      "learning_rate": 0.0001750406384245793,
+      "loss": 0.7344526648521423,
+      "step": 4146
+    },
+    {
+      "epoch": 1.7502109704641349,
+      "grad_norm": 0.4324318468570709,
+      "learning_rate": 0.00017500888355916576,
+      "loss": 0.7047654986381531,
+      "step": 4148
+    },
+    {
+      "epoch": 1.7510548523206753,
+      "grad_norm": 0.44670364260673523,
+      "learning_rate": 0.00017497711139064336,
+      "loss": 0.7317671179771423,
+      "step": 4150
+    },
+    {
+      "epoch": 1.7518987341772152,
+      "grad_norm": 0.43074679374694824,
+      "learning_rate": 0.00017494532192634138,
+      "loss": 0.7447407245635986,
+      "step": 4152
+    },
+    {
+      "epoch": 1.7527426160337551,
+      "grad_norm": 0.43057894706726074,
+      "learning_rate": 0.00017491351517359292,
+      "loss": 0.7352449893951416,
+      "step": 4154
+    },
+    {
+      "epoch": 1.7535864978902953,
+      "grad_norm": 0.49488556385040283,
+      "learning_rate": 0.00017488169113973528,
+      "loss": 0.7244091033935547,
+      "step": 4156
+    },
+    {
+      "epoch": 1.7544303797468355,
+      "grad_norm": 0.4565380811691284,
+      "learning_rate": 0.0001748498498321096,
+      "loss": 0.7515795230865479,
+      "step": 4158
+    },
+    {
+      "epoch": 1.7552742616033754,
+      "grad_norm": 0.45722341537475586,
+      "learning_rate": 0.0001748179912580611,
+      "loss": 0.7451042532920837,
+      "step": 4160
+    },
+    {
+      "epoch": 1.7561181434599156,
+      "grad_norm": 0.47247326374053955,
+      "learning_rate": 0.0001747861154249389,
+      "loss": 0.7133217453956604,
+      "step": 4162
+    },
+    {
+      "epoch": 1.7569620253164557,
+      "grad_norm": 0.48307132720947266,
+      "learning_rate": 0.00017475422234009623,
+      "loss": 0.643742561340332,
+      "step": 4164
+    },
+    {
+      "epoch": 1.7578059071729957,
+      "grad_norm": 0.49163761734962463,
+      "learning_rate": 0.00017472231201089012,
+      "loss": 0.7587857842445374,
+      "step": 4166
+    },
+    {
+      "epoch": 1.7586497890295358,
+      "grad_norm": 0.4726962149143219,
+      "learning_rate": 0.00017469038444468165,
+      "loss": 0.6980416774749756,
+      "step": 4168
+    },
+    {
+      "epoch": 1.759493670886076,
+      "grad_norm": 0.4192392826080322,
+      "learning_rate": 0.00017465843964883598,
+      "loss": 0.6929903626441956,
+      "step": 4170
+    },
+    {
+      "epoch": 1.760337552742616,
+      "grad_norm": 0.43000563979148865,
+      "learning_rate": 0.00017462647763072216,
+      "loss": 0.7076222896575928,
+      "step": 4172
+    },
+    {
+      "epoch": 1.761181434599156,
+      "grad_norm": 0.45402681827545166,
+      "learning_rate": 0.00017459449839771313,
+      "loss": 0.7260748147964478,
+      "step": 4174
+    },
+    {
+      "epoch": 1.7620253164556963,
+      "grad_norm": 0.4077165126800537,
+      "learning_rate": 0.00017456250195718595,
+      "loss": 0.6901930570602417,
+      "step": 4176
+    },
+    {
+      "epoch": 1.7628691983122362,
+      "grad_norm": 0.4167994558811188,
+      "learning_rate": 0.00017453048831652157,
+      "loss": 0.7085300087928772,
+      "step": 4178
+    },
+    {
+      "epoch": 1.7637130801687764,
+      "grad_norm": 0.45262131094932556,
+      "learning_rate": 0.00017449845748310493,
+      "loss": 0.7646799087524414,
+      "step": 4180
+    },
+    {
+      "epoch": 1.7645569620253165,
+      "grad_norm": 0.4639943540096283,
+      "learning_rate": 0.0001744664094643249,
+      "loss": 0.7847923040390015,
+      "step": 4182
+    },
+    {
+      "epoch": 1.7654008438818565,
+      "grad_norm": 0.33605465292930603,
+      "learning_rate": 0.00017443434426757439,
+      "loss": 0.7192286252975464,
+      "step": 4184
+    },
+    {
+      "epoch": 1.7662447257383966,
+      "grad_norm": 0.4036388099193573,
+      "learning_rate": 0.00017440226190025015,
+      "loss": 0.6554746627807617,
+      "step": 4186
+    },
+    {
+      "epoch": 1.7670886075949368,
+      "grad_norm": 0.45648789405822754,
+      "learning_rate": 0.00017437016236975303,
+      "loss": 0.6433794498443604,
+      "step": 4188
+    },
+    {
+      "epoch": 1.7679324894514767,
+      "grad_norm": 0.38929006457328796,
+      "learning_rate": 0.00017433804568348776,
+      "loss": 0.6608775854110718,
+      "step": 4190
+    },
+    {
+      "epoch": 1.768776371308017,
+      "grad_norm": 0.37158843874931335,
+      "learning_rate": 0.00017430591184886308,
+      "loss": 0.6470816135406494,
+      "step": 4192
+    },
+    {
+      "epoch": 1.769620253164557,
+      "grad_norm": 0.39842307567596436,
+      "learning_rate": 0.00017427376087329158,
+      "loss": 0.6825124025344849,
+      "step": 4194
+    },
+    {
+      "epoch": 1.770464135021097,
+      "grad_norm": 0.44220465421676636,
+      "learning_rate": 0.00017424159276418987,
+      "loss": 0.7641696333885193,
+      "step": 4196
+    },
+    {
+      "epoch": 1.7713080168776372,
+      "grad_norm": 0.44530072808265686,
+      "learning_rate": 0.0001742094075289786,
+      "loss": 0.7783681154251099,
+      "step": 4198
+    },
+    {
+      "epoch": 1.7721518987341773,
+      "grad_norm": 0.4497349262237549,
+      "learning_rate": 0.00017417720517508216,
+      "loss": 0.6718224883079529,
+      "step": 4200
+    },
+    {
+      "epoch": 1.7721518987341773,
+      "eval_loss": 0.7676366567611694,
+      "eval_runtime": 687.9619,
+      "eval_samples_per_second": 3.063,
+      "eval_steps_per_second": 3.063,
+      "step": 4200
+    },
+    {
+      "epoch": 1.7729957805907173,
+      "grad_norm": 0.3857066333293915,
+      "learning_rate": 0.00017414498570992914,
+      "loss": 0.7050814032554626,
+      "step": 4202
+    },
+    {
+      "epoch": 1.7738396624472574,
+      "grad_norm": 0.4533625543117523,
+      "learning_rate": 0.00017411274914095188,
+      "loss": 0.7222742438316345,
+      "step": 4204
+    },
+    {
+      "epoch": 1.7746835443037976,
+      "grad_norm": 0.40805840492248535,
+      "learning_rate": 0.00017408049547558676,
+      "loss": 0.6434906125068665,
+      "step": 4206
+    },
+    {
+      "epoch": 1.7755274261603375,
+      "grad_norm": 0.5041740536689758,
+      "learning_rate": 0.00017404822472127406,
+      "loss": 0.686283528804779,
+      "step": 4208
+    },
+    {
+      "epoch": 1.7763713080168775,
+      "grad_norm": 0.41846802830696106,
+      "learning_rate": 0.00017401593688545807,
+      "loss": 0.6785043478012085,
+      "step": 4210
+    },
+    {
+      "epoch": 1.7772151898734179,
+      "grad_norm": 0.4696303606033325,
+      "learning_rate": 0.00017398363197558694,
+      "loss": 0.7566665410995483,
+      "step": 4212
+    },
+    {
+      "epoch": 1.7780590717299578,
+      "grad_norm": 0.3731461763381958,
+      "learning_rate": 0.0001739513099991128,
+      "loss": 0.6178455948829651,
+      "step": 4214
+    },
+    {
+      "epoch": 1.7789029535864977,
+      "grad_norm": 0.48098450899124146,
+      "learning_rate": 0.00017391897096349167,
+      "loss": 0.771265983581543,
+      "step": 4216
+    },
+    {
+      "epoch": 1.7797468354430381,
+      "grad_norm": 0.4164000451564789,
+      "learning_rate": 0.0001738866148761836,
+      "loss": 0.6372597813606262,
+      "step": 4218
+    },
+    {
+      "epoch": 1.780590717299578,
+      "grad_norm": 0.4490415155887604,
+      "learning_rate": 0.00017385424174465251,
+      "loss": 0.7213301658630371,
+      "step": 4220
+    },
+    {
+      "epoch": 1.781434599156118,
+      "grad_norm": 0.44518229365348816,
+      "learning_rate": 0.00017382185157636623,
+      "loss": 0.7575726509094238,
+      "step": 4222
+    },
+    {
+      "epoch": 1.7822784810126582,
+      "grad_norm": 0.34899434447288513,
+      "learning_rate": 0.00017378944437879658,
+      "loss": 0.6473780274391174,
+      "step": 4224
+    },
+    {
+      "epoch": 1.7831223628691983,
+      "grad_norm": 0.4068683087825775,
+      "learning_rate": 0.00017375702015941924,
+      "loss": 0.7238336205482483,
+      "step": 4226
+    },
+    {
+      "epoch": 1.7839662447257383,
+      "grad_norm": 0.4025551974773407,
+      "learning_rate": 0.0001737245789257139,
+      "loss": 0.7797979116439819,
+      "step": 4228
+    },
+    {
+      "epoch": 1.7848101265822784,
+      "grad_norm": 0.35764896869659424,
+      "learning_rate": 0.00017369212068516412,
+      "loss": 0.6304885149002075,
+      "step": 4230
+    },
+    {
+      "epoch": 1.7856540084388186,
+      "grad_norm": 0.43337100744247437,
+      "learning_rate": 0.00017365964544525738,
+      "loss": 0.6606637835502625,
+      "step": 4232
+    },
+    {
+      "epoch": 1.7864978902953585,
+      "grad_norm": 0.4630814492702484,
+      "learning_rate": 0.0001736271532134851,
+      "loss": 0.712163507938385,
+      "step": 4234
+    },
+    {
+      "epoch": 1.7873417721518987,
+      "grad_norm": 0.38714799284935,
+      "learning_rate": 0.00017359464399734255,
+      "loss": 0.6649466753005981,
+      "step": 4236
+    },
+    {
+      "epoch": 1.7881856540084389,
+      "grad_norm": 0.38999977707862854,
+      "learning_rate": 0.0001735621178043291,
+      "loss": 0.6223326921463013,
+      "step": 4238
+    },
+    {
+      "epoch": 1.7890295358649788,
+      "grad_norm": 0.4736211597919464,
+      "learning_rate": 0.00017352957464194786,
+      "loss": 0.7207517027854919,
+      "step": 4240
+    },
+    {
+      "epoch": 1.789873417721519,
+      "grad_norm": 0.423476904630661,
+      "learning_rate": 0.00017349701451770588,
+      "loss": 0.7203661203384399,
+      "step": 4242
+    },
+    {
+      "epoch": 1.7907172995780591,
+      "grad_norm": 0.4487479627132416,
+      "learning_rate": 0.0001734644374391142,
+      "loss": 0.709998607635498,
+      "step": 4244
+    },
+    {
+      "epoch": 1.791561181434599,
+      "grad_norm": 0.43721267580986023,
+      "learning_rate": 0.00017343184341368773,
+      "loss": 0.7079582214355469,
+      "step": 4246
+    },
+    {
+      "epoch": 1.7924050632911392,
+      "grad_norm": 0.426864892244339,
+      "learning_rate": 0.00017339923244894524,
+      "loss": 0.6772971749305725,
+      "step": 4248
+    },
+    {
+      "epoch": 1.7932489451476794,
+      "grad_norm": 0.3580208420753479,
+      "learning_rate": 0.0001733666045524095,
+      "loss": 0.6882370710372925,
+      "step": 4250
+    },
+    {
+      "epoch": 1.7940928270042193,
+      "grad_norm": 0.487248957157135,
+      "learning_rate": 0.00017333395973160714,
+      "loss": 0.7534610629081726,
+      "step": 4252
+    },
+    {
+      "epoch": 1.7949367088607595,
+      "grad_norm": 0.44908860325813293,
+      "learning_rate": 0.00017330129799406866,
+      "loss": 0.6897269487380981,
+      "step": 4254
+    },
+    {
+      "epoch": 1.7957805907172997,
+      "grad_norm": 0.43617817759513855,
+      "learning_rate": 0.00017326861934732852,
+      "loss": 0.7081538438796997,
+      "step": 4256
+    },
+    {
+      "epoch": 1.7966244725738396,
+      "grad_norm": 0.428666353225708,
+      "learning_rate": 0.00017323592379892504,
+      "loss": 0.6852753758430481,
+      "step": 4258
+    },
+    {
+      "epoch": 1.7974683544303798,
+      "grad_norm": 0.4522199034690857,
+      "learning_rate": 0.00017320321135640045,
+      "loss": 0.7463182210922241,
+      "step": 4260
+    },
+    {
+      "epoch": 1.79831223628692,
+      "grad_norm": 0.38632410764694214,
+      "learning_rate": 0.00017317048202730088,
+      "loss": 0.6643774509429932,
+      "step": 4262
+    },
+    {
+      "epoch": 1.7991561181434599,
+      "grad_norm": 0.41171637177467346,
+      "learning_rate": 0.00017313773581917642,
+      "loss": 0.6726668477058411,
+      "step": 4264
+    },
+    {
+      "epoch": 1.8,
+      "grad_norm": 0.398520290851593,
+      "learning_rate": 0.0001731049727395809,
+      "loss": 0.6278629302978516,
+      "step": 4266
+    },
+    {
+      "epoch": 1.8008438818565402,
+      "grad_norm": 0.4000356197357178,
+      "learning_rate": 0.0001730721927960722,
+      "loss": 0.5823352336883545,
+      "step": 4268
+    },
+    {
+      "epoch": 1.8016877637130801,
+      "grad_norm": 0.47376710176467896,
+      "learning_rate": 0.000173039395996212,
+      "loss": 0.7492824196815491,
+      "step": 4270
+    },
+    {
+      "epoch": 1.80253164556962,
+      "grad_norm": 0.4624378979206085,
+      "learning_rate": 0.00017300658234756588,
+      "loss": 0.7630324363708496,
+      "step": 4272
+    },
+    {
+      "epoch": 1.8033755274261605,
+      "grad_norm": 0.4178096354007721,
+      "learning_rate": 0.00017297375185770336,
+      "loss": 0.6484803557395935,
+      "step": 4274
+    },
+    {
+      "epoch": 1.8042194092827004,
+      "grad_norm": 0.511215329170227,
+      "learning_rate": 0.00017294090453419774,
+      "loss": 0.776732325553894,
+      "step": 4276
+    },
+    {
+      "epoch": 1.8050632911392404,
+      "grad_norm": 0.44643548130989075,
+      "learning_rate": 0.00017290804038462633,
+      "loss": 0.6801077723503113,
+      "step": 4278
+    },
+    {
+      "epoch": 1.8059071729957807,
+      "grad_norm": 0.5090142488479614,
+      "learning_rate": 0.0001728751594165702,
+      "loss": 0.8024200797080994,
+      "step": 4280
+    },
+    {
+      "epoch": 1.8067510548523207,
+      "grad_norm": 0.45888981223106384,
+      "learning_rate": 0.00017284226163761436,
+      "loss": 0.7810231447219849,
+      "step": 4282
+    },
+    {
+      "epoch": 1.8075949367088606,
+      "grad_norm": 0.3985820710659027,
+      "learning_rate": 0.0001728093470553478,
+      "loss": 0.6768189072608948,
+      "step": 4284
+    },
+    {
+      "epoch": 1.808438818565401,
+      "grad_norm": 0.3988003730773926,
+      "learning_rate": 0.00017277641567736316,
+      "loss": 0.6093943119049072,
+      "step": 4286
+    },
+    {
+      "epoch": 1.809282700421941,
+      "grad_norm": 0.46131962537765503,
+      "learning_rate": 0.0001727434675112571,
+      "loss": 0.6836619973182678,
+      "step": 4288
+    },
+    {
+      "epoch": 1.810126582278481,
+      "grad_norm": 0.42959487438201904,
+      "learning_rate": 0.0001727105025646302,
+      "loss": 0.7144147157669067,
+      "step": 4290
+    },
+    {
+      "epoch": 1.810970464135021,
+      "grad_norm": 0.4110259413719177,
+      "learning_rate": 0.00017267752084508675,
+      "loss": 0.7190625667572021,
+      "step": 4292
+    },
+    {
+      "epoch": 1.8118143459915612,
+      "grad_norm": 0.4501377046108246,
+      "learning_rate": 0.00017264452236023505,
+      "loss": 0.7272495031356812,
+      "step": 4294
+    },
+    {
+      "epoch": 1.8126582278481012,
+      "grad_norm": 0.467817097902298,
+      "learning_rate": 0.0001726115071176872,
+      "loss": 0.758166491985321,
+      "step": 4296
+    },
+    {
+      "epoch": 1.8135021097046413,
+      "grad_norm": 0.42394939064979553,
+      "learning_rate": 0.0001725784751250592,
+      "loss": 0.6712231040000916,
+      "step": 4298
+    },
+    {
+      "epoch": 1.8143459915611815,
+      "grad_norm": 0.37668925523757935,
+      "learning_rate": 0.00017254542638997088,
+      "loss": 0.5999719500541687,
+      "step": 4300
+    },
+    {
+      "epoch": 1.8143459915611815,
+      "eval_loss": 0.7672964930534363,
+      "eval_runtime": 688.4249,
+      "eval_samples_per_second": 3.061,
+      "eval_steps_per_second": 3.061,
+      "step": 4300
+    },
+    {
+      "epoch": 1.8151898734177214,
+      "grad_norm": 0.44964465498924255,
+      "learning_rate": 0.00017251236092004596,
+      "loss": 0.7188607454299927,
+      "step": 4302
+    },
+    {
+      "epoch": 1.8160337552742616,
+      "grad_norm": 0.4108492434024811,
+      "learning_rate": 0.000172479278722912,
+      "loss": 0.707834005355835,
+      "step": 4304
+    },
+    {
+      "epoch": 1.8168776371308017,
+      "grad_norm": 0.4455305337905884,
+      "learning_rate": 0.00017244617980620043,
+      "loss": 0.6915075182914734,
+      "step": 4306
+    },
+    {
+      "epoch": 1.8177215189873417,
+      "grad_norm": 0.3977123200893402,
+      "learning_rate": 0.00017241306417754655,
+      "loss": 0.6591874361038208,
+      "step": 4308
+    },
+    {
+      "epoch": 1.8185654008438819,
+      "grad_norm": 0.4028434157371521,
+      "learning_rate": 0.00017237993184458946,
+      "loss": 0.6254785060882568,
+      "step": 4310
+    },
+    {
+      "epoch": 1.819409282700422,
+      "grad_norm": 0.41024482250213623,
+      "learning_rate": 0.00017234678281497217,
+      "loss": 0.6968662142753601,
+      "step": 4312
+    },
+    {
+      "epoch": 1.820253164556962,
+      "grad_norm": 0.45941361784935,
+      "learning_rate": 0.00017231361709634153,
+      "loss": 0.6535647511482239,
+      "step": 4314
+    },
+    {
+      "epoch": 1.8210970464135021,
+      "grad_norm": 0.39477071166038513,
+      "learning_rate": 0.00017228043469634825,
+      "loss": 0.6378563642501831,
+      "step": 4316
+    },
+    {
+      "epoch": 1.8219409282700423,
+      "grad_norm": 0.403826504945755,
+      "learning_rate": 0.00017224723562264687,
+      "loss": 0.7021255493164062,
+      "step": 4318
+    },
+    {
+      "epoch": 1.8227848101265822,
+      "grad_norm": 0.4179201126098633,
+      "learning_rate": 0.00017221401988289575,
+      "loss": 0.7411118745803833,
+      "step": 4320
+    },
+    {
+      "epoch": 1.8236286919831224,
+      "grad_norm": 0.4719425141811371,
+      "learning_rate": 0.0001721807874847571,
+      "loss": 0.7435210943222046,
+      "step": 4322
+    },
+    {
+      "epoch": 1.8244725738396625,
+      "grad_norm": 0.39946448802948,
+      "learning_rate": 0.0001721475384358971,
+      "loss": 0.6813794374465942,
+      "step": 4324
+    },
+    {
+      "epoch": 1.8253164556962025,
+      "grad_norm": 0.4367005527019501,
+      "learning_rate": 0.00017211427274398559,
+      "loss": 0.6468828320503235,
+      "step": 4326
+    },
+    {
+      "epoch": 1.8261603375527427,
+      "grad_norm": 0.4078138470649719,
+      "learning_rate": 0.00017208099041669632,
+      "loss": 0.6628729701042175,
+      "step": 4328
+    },
+    {
+      "epoch": 1.8270042194092828,
+      "grad_norm": 0.4141235649585724,
+      "learning_rate": 0.0001720476914617069,
+      "loss": 0.7028043866157532,
+      "step": 4330
+    },
+    {
+      "epoch": 1.8278481012658228,
+      "grad_norm": 0.448733925819397,
+      "learning_rate": 0.00017201437588669878,
+      "loss": 0.6710682511329651,
+      "step": 4332
+    },
+    {
+      "epoch": 1.828691983122363,
+      "grad_norm": 0.5084340572357178,
+      "learning_rate": 0.00017198104369935718,
+      "loss": 0.792453408241272,
+      "step": 4334
+    },
+    {
+      "epoch": 1.829535864978903,
+      "grad_norm": 0.43463611602783203,
+      "learning_rate": 0.00017194769490737124,
+      "loss": 0.7536583542823792,
+      "step": 4336
+    },
+    {
+      "epoch": 1.830379746835443,
+      "grad_norm": 0.46039044857025146,
+      "learning_rate": 0.00017191432951843387,
+      "loss": 0.7445555925369263,
+      "step": 4338
+    },
+    {
+      "epoch": 1.831223628691983,
+      "grad_norm": 0.3894030451774597,
+      "learning_rate": 0.0001718809475402418,
+      "loss": 0.7003044486045837,
+      "step": 4340
+    },
+    {
+      "epoch": 1.8320675105485233,
+      "grad_norm": 0.5851747393608093,
+      "learning_rate": 0.00017184754898049568,
+      "loss": 0.6948950886726379,
+      "step": 4342
+    },
+    {
+      "epoch": 1.8329113924050633,
+      "grad_norm": 0.40906789898872375,
+      "learning_rate": 0.00017181413384689982,
+      "loss": 0.675811231136322,
+      "step": 4344
+    },
+    {
+      "epoch": 1.8337552742616032,
+      "grad_norm": 0.4235765039920807,
+      "learning_rate": 0.0001717807021471625,
+      "loss": 0.7422580122947693,
+      "step": 4346
+    },
+    {
+      "epoch": 1.8345991561181436,
+      "grad_norm": 0.3898368775844574,
+      "learning_rate": 0.0001717472538889958,
+      "loss": 0.7014731168746948,
+      "step": 4348
+    },
+    {
+      "epoch": 1.8354430379746836,
+      "grad_norm": 0.4122314751148224,
+      "learning_rate": 0.00017171378908011553,
+      "loss": 0.6952776312828064,
+      "step": 4350
+    },
+    {
+      "epoch": 1.8362869198312235,
+      "grad_norm": 0.38085612654685974,
+      "learning_rate": 0.00017168030772824144,
+      "loss": 0.6758598685264587,
+      "step": 4352
+    },
+    {
+      "epoch": 1.8371308016877637,
+      "grad_norm": 0.43519729375839233,
+      "learning_rate": 0.00017164680984109695,
+      "loss": 0.7139424085617065,
+      "step": 4354
+    },
+    {
+      "epoch": 1.8379746835443038,
+      "grad_norm": 0.4404325783252716,
+      "learning_rate": 0.0001716132954264095,
+      "loss": 0.7125198841094971,
+      "step": 4356
+    },
+    {
+      "epoch": 1.8388185654008438,
+      "grad_norm": 0.3965512812137604,
+      "learning_rate": 0.00017157976449191012,
+      "loss": 0.7011314630508423,
+      "step": 4358
+    },
+    {
+      "epoch": 1.839662447257384,
+      "grad_norm": 0.4273630976676941,
+      "learning_rate": 0.0001715462170453338,
+      "loss": 0.6485645174980164,
+      "step": 4360
+    },
+    {
+      "epoch": 1.840506329113924,
+      "grad_norm": 0.4639208912849426,
+      "learning_rate": 0.00017151265309441925,
+      "loss": 0.6669548153877258,
+      "step": 4362
+    },
+    {
+      "epoch": 1.841350210970464,
+      "grad_norm": 0.4190906286239624,
+      "learning_rate": 0.00017147907264690907,
+      "loss": 0.6276374459266663,
+      "step": 4364
+    },
+    {
+      "epoch": 1.8421940928270042,
+      "grad_norm": 0.4146515130996704,
+      "learning_rate": 0.00017144547571054962,
+      "loss": 0.5975915193557739,
+      "step": 4366
+    },
+    {
+      "epoch": 1.8430379746835444,
+      "grad_norm": 0.38717755675315857,
+      "learning_rate": 0.00017141186229309105,
+      "loss": 0.6536507606506348,
+      "step": 4368
+    },
+    {
+      "epoch": 1.8438818565400843,
+      "grad_norm": 0.3988541066646576,
+      "learning_rate": 0.00017137823240228735,
+      "loss": 0.7078962326049805,
+      "step": 4370
+    },
+    {
+      "epoch": 1.8447257383966245,
+      "grad_norm": 0.43783870339393616,
+      "learning_rate": 0.00017134458604589628,
+      "loss": 0.6795231699943542,
+      "step": 4372
+    },
+    {
+      "epoch": 1.8455696202531646,
+      "grad_norm": 0.439636766910553,
+      "learning_rate": 0.00017131092323167938,
+      "loss": 0.7030490040779114,
+      "step": 4374
+    },
+    {
+      "epoch": 1.8464135021097046,
+      "grad_norm": 0.4426078498363495,
+      "learning_rate": 0.00017127724396740206,
+      "loss": 0.7611671686172485,
+      "step": 4376
+    },
+    {
+      "epoch": 1.8472573839662447,
+      "grad_norm": 0.40058812499046326,
+      "learning_rate": 0.00017124354826083348,
+      "loss": 0.737910270690918,
+      "step": 4378
+    },
+    {
+      "epoch": 1.8481012658227849,
+      "grad_norm": 0.43037283420562744,
+      "learning_rate": 0.00017120983611974654,
+      "loss": 0.683937132358551,
+      "step": 4380
+    },
+    {
+      "epoch": 1.8489451476793248,
+      "grad_norm": 0.4479113221168518,
+      "learning_rate": 0.00017117610755191807,
+      "loss": 0.6987335681915283,
+      "step": 4382
+    },
+    {
+      "epoch": 1.849789029535865,
+      "grad_norm": 0.4218229651451111,
+      "learning_rate": 0.0001711423625651285,
+      "loss": 0.6964257955551147,
+      "step": 4384
+    },
+    {
+      "epoch": 1.8506329113924052,
+      "grad_norm": 0.397733211517334,
+      "learning_rate": 0.00017110860116716222,
+      "loss": 0.6983811855316162,
+      "step": 4386
+    },
+    {
+      "epoch": 1.851476793248945,
+      "grad_norm": 0.4844493865966797,
+      "learning_rate": 0.00017107482336580735,
+      "loss": 0.7579324245452881,
+      "step": 4388
+    },
+    {
+      "epoch": 1.8523206751054853,
+      "grad_norm": 0.4494120478630066,
+      "learning_rate": 0.0001710410291688557,
+      "loss": 0.703558623790741,
+      "step": 4390
+    },
+    {
+      "epoch": 1.8531645569620254,
+      "grad_norm": 0.4721727967262268,
+      "learning_rate": 0.000171007218584103,
+      "loss": 0.7386981248855591,
+      "step": 4392
+    },
+    {
+      "epoch": 1.8540084388185654,
+      "grad_norm": 0.4816451370716095,
+      "learning_rate": 0.0001709733916193487,
+      "loss": 0.7769017219543457,
+      "step": 4394
+    },
+    {
+      "epoch": 1.8548523206751055,
+      "grad_norm": 0.43638876080513,
+      "learning_rate": 0.00017093954828239602,
+      "loss": 0.7315034866333008,
+      "step": 4396
+    },
+    {
+      "epoch": 1.8556962025316457,
+      "grad_norm": 0.4035055935382843,
+      "learning_rate": 0.000170905688581052,
+      "loss": 0.7289890050888062,
+      "step": 4398
+    },
+    {
+      "epoch": 1.8565400843881856,
+      "grad_norm": 0.44265708327293396,
+      "learning_rate": 0.00017087181252312735,
+      "loss": 0.7190408110618591,
+      "step": 4400
+    },
+    {
+      "epoch": 1.8565400843881856,
+      "eval_loss": 0.7635221481323242,
+      "eval_runtime": 678.243,
+      "eval_samples_per_second": 3.107,
+      "eval_steps_per_second": 3.107,
+      "step": 4400
+    },
+    {
+      "epoch": 1.8573839662447258,
+      "grad_norm": 0.39691466093063354,
+      "learning_rate": 0.0001708379201164367,
+      "loss": 0.6318185925483704,
+      "step": 4402
+    },
+    {
+      "epoch": 1.858227848101266,
+      "grad_norm": 0.47801393270492554,
+      "learning_rate": 0.0001708040113687983,
+      "loss": 0.7040913105010986,
+      "step": 4404
+    },
+    {
+      "epoch": 1.859071729957806,
+      "grad_norm": 0.47457239031791687,
+      "learning_rate": 0.00017077008628803436,
+      "loss": 0.7834659218788147,
+      "step": 4406
+    },
+    {
+      "epoch": 1.8599156118143458,
+      "grad_norm": 0.41362807154655457,
+      "learning_rate": 0.00017073614488197065,
+      "loss": 0.7027983665466309,
+      "step": 4408
+    },
+    {
+      "epoch": 1.8607594936708862,
+      "grad_norm": 0.457927942276001,
+      "learning_rate": 0.0001707021871584368,
+      "loss": 0.7143591642379761,
+      "step": 4410
+    },
+    {
+      "epoch": 1.8616033755274262,
+      "grad_norm": 0.42931967973709106,
+      "learning_rate": 0.00017066821312526625,
+      "loss": 0.7328225374221802,
+      "step": 4412
+    },
+    {
+      "epoch": 1.862447257383966,
+      "grad_norm": 0.3821638524532318,
+      "learning_rate": 0.00017063422279029616,
+      "loss": 0.7199252843856812,
+      "step": 4414
+    },
+    {
+      "epoch": 1.8632911392405065,
+      "grad_norm": 0.3954671621322632,
+      "learning_rate": 0.00017060021616136742,
+      "loss": 0.7345157265663147,
+      "step": 4416
+    },
+    {
+      "epoch": 1.8641350210970464,
+      "grad_norm": 0.38824960589408875,
+      "learning_rate": 0.0001705661932463247,
+      "loss": 0.587895393371582,
+      "step": 4418
+    },
+    {
+      "epoch": 1.8649789029535864,
+      "grad_norm": 0.510923445224762,
+      "learning_rate": 0.00017053215405301648,
+      "loss": 0.7227945327758789,
+      "step": 4420
+    },
+    {
+      "epoch": 1.8658227848101265,
+      "grad_norm": 0.38688936829566956,
+      "learning_rate": 0.0001704980985892949,
+      "loss": 0.6605637073516846,
+      "step": 4422
+    },
+    {
+      "epoch": 1.8666666666666667,
+      "grad_norm": 0.4886745512485504,
+      "learning_rate": 0.0001704640268630159,
+      "loss": 0.7581667304039001,
+      "step": 4424
+    },
+    {
+      "epoch": 1.8675105485232066,
+      "grad_norm": 0.4214446246623993,
+      "learning_rate": 0.0001704299388820392,
+      "loss": 0.7051677107810974,
+      "step": 4426
+    },
+    {
+      "epoch": 1.8683544303797468,
+      "grad_norm": 0.4189293384552002,
+      "learning_rate": 0.00017039583465422824,
+      "loss": 0.6489346623420715,
+      "step": 4428
+    },
+    {
+      "epoch": 1.869198312236287,
+      "grad_norm": 0.4264903962612152,
+      "learning_rate": 0.0001703617141874502,
+      "loss": 0.6896629929542542,
+      "step": 4430
+    },
+    {
+      "epoch": 1.870042194092827,
+      "grad_norm": 0.5351413488388062,
+      "learning_rate": 0.00017032757748957603,
+      "loss": 0.7468283772468567,
+      "step": 4432
+    },
+    {
+      "epoch": 1.870886075949367,
+      "grad_norm": 0.40163540840148926,
+      "learning_rate": 0.00017029342456848035,
+      "loss": 0.6029976010322571,
+      "step": 4434
+    },
+    {
+      "epoch": 1.8717299578059072,
+      "grad_norm": 0.3786795139312744,
+      "learning_rate": 0.0001702592554320417,
+      "loss": 0.6891162991523743,
+      "step": 4436
+    },
+    {
+      "epoch": 1.8725738396624472,
+      "grad_norm": 0.4266849458217621,
+      "learning_rate": 0.00017022507008814214,
+      "loss": 0.6189853549003601,
+      "step": 4438
+    },
+    {
+      "epoch": 1.8734177215189873,
+      "grad_norm": 0.4076935052871704,
+      "learning_rate": 0.0001701908685446676,
+      "loss": 0.6858769655227661,
+      "step": 4440
+    },
+    {
+      "epoch": 1.8742616033755275,
+      "grad_norm": 0.4209558665752411,
+      "learning_rate": 0.00017015665080950774,
+      "loss": 0.6650485396385193,
+      "step": 4442
+    },
+    {
+      "epoch": 1.8751054852320674,
+      "grad_norm": 0.4114036560058594,
+      "learning_rate": 0.0001701224168905559,
+      "loss": 0.6887878179550171,
+      "step": 4444
+    },
+    {
+      "epoch": 1.8759493670886076,
+      "grad_norm": 0.34596455097198486,
+      "learning_rate": 0.00017008816679570926,
+      "loss": 0.6950712203979492,
+      "step": 4446
+    },
+    {
+      "epoch": 1.8767932489451478,
+      "grad_norm": 0.45132747292518616,
+      "learning_rate": 0.0001700539005328686,
+      "loss": 0.7360581159591675,
+      "step": 4448
+    },
+    {
+      "epoch": 1.8776371308016877,
+      "grad_norm": 0.4258243441581726,
+      "learning_rate": 0.00017001961810993846,
+      "loss": 0.6672278046607971,
+      "step": 4450
+    },
+    {
+      "epoch": 1.8784810126582279,
+      "grad_norm": 0.39144474267959595,
+      "learning_rate": 0.0001699853195348272,
+      "loss": 0.6740089654922485,
+      "step": 4452
+    },
+    {
+      "epoch": 1.879324894514768,
+      "grad_norm": 0.42183035612106323,
+      "learning_rate": 0.00016995100481544681,
+      "loss": 0.7747141122817993,
+      "step": 4454
+    },
+    {
+      "epoch": 1.880168776371308,
+      "grad_norm": 0.4110005795955658,
+      "learning_rate": 0.00016991667395971303,
+      "loss": 0.7189701199531555,
+      "step": 4456
+    },
+    {
+      "epoch": 1.8810126582278481,
+      "grad_norm": 0.4073067307472229,
+      "learning_rate": 0.00016988232697554538,
+      "loss": 0.6880105137825012,
+      "step": 4458
+    },
+    {
+      "epoch": 1.8818565400843883,
+      "grad_norm": 0.467443585395813,
+      "learning_rate": 0.00016984796387086698,
+      "loss": 0.6994736790657043,
+      "step": 4460
+    },
+    {
+      "epoch": 1.8827004219409282,
+      "grad_norm": 0.4769018292427063,
+      "learning_rate": 0.00016981358465360482,
+      "loss": 0.7389267683029175,
+      "step": 4462
+    },
+    {
+      "epoch": 1.8835443037974684,
+      "grad_norm": 0.3834097981452942,
+      "learning_rate": 0.00016977918933168938,
+      "loss": 0.6375740170478821,
+      "step": 4464
+    },
+    {
+      "epoch": 1.8843881856540086,
+      "grad_norm": 0.44548940658569336,
+      "learning_rate": 0.00016974477791305517,
+      "loss": 0.6845629811286926,
+      "step": 4466
+    },
+    {
+      "epoch": 1.8852320675105485,
+      "grad_norm": 0.39344266057014465,
+      "learning_rate": 0.00016971035040564016,
+      "loss": 0.7307721376419067,
+      "step": 4468
+    },
+    {
+      "epoch": 1.8860759493670884,
+      "grad_norm": 0.37901216745376587,
+      "learning_rate": 0.0001696759068173861,
+      "loss": 0.6733962297439575,
+      "step": 4470
+    },
+    {
+      "epoch": 1.8869198312236288,
+      "grad_norm": 0.44465333223342896,
+      "learning_rate": 0.00016964144715623852,
+      "loss": 0.7294619679450989,
+      "step": 4472
+    },
+    {
+      "epoch": 1.8877637130801688,
+      "grad_norm": 0.39566636085510254,
+      "learning_rate": 0.00016960697143014657,
+      "loss": 0.6537288427352905,
+      "step": 4474
+    },
+    {
+      "epoch": 1.8886075949367087,
+      "grad_norm": 0.4047028124332428,
+      "learning_rate": 0.0001695724796470631,
+      "loss": 0.623903751373291,
+      "step": 4476
+    },
+    {
+      "epoch": 1.889451476793249,
+      "grad_norm": 0.4559679329395294,
+      "learning_rate": 0.00016953797181494475,
+      "loss": 0.661283552646637,
+      "step": 4478
+    },
+    {
+      "epoch": 1.890295358649789,
+      "grad_norm": 0.3980371952056885,
+      "learning_rate": 0.00016950344794175183,
+      "loss": 0.6587702035903931,
+      "step": 4480
+    },
+    {
+      "epoch": 1.891139240506329,
+      "grad_norm": 0.41418227553367615,
+      "learning_rate": 0.0001694689080354483,
+      "loss": 0.7898983955383301,
+      "step": 4482
+    },
+    {
+      "epoch": 1.8919831223628694,
+      "grad_norm": 0.41184720396995544,
+      "learning_rate": 0.00016943435210400184,
+      "loss": 0.6213424205780029,
+      "step": 4484
+    },
+    {
+      "epoch": 1.8928270042194093,
+      "grad_norm": 0.39218711853027344,
+      "learning_rate": 0.00016939978015538388,
+      "loss": 0.6624190807342529,
+      "step": 4486
+    },
+    {
+      "epoch": 1.8936708860759492,
+      "grad_norm": 0.4875730872154236,
+      "learning_rate": 0.0001693651921975695,
+      "loss": 0.7063044309616089,
+      "step": 4488
+    },
+    {
+      "epoch": 1.8945147679324894,
+      "grad_norm": 0.38058432936668396,
+      "learning_rate": 0.00016933058823853748,
+      "loss": 0.6471660137176514,
+      "step": 4490
+    },
+    {
+      "epoch": 1.8953586497890296,
+      "grad_norm": 0.5368654131889343,
+      "learning_rate": 0.0001692959682862702,
+      "loss": 0.7694545388221741,
+      "step": 4492
+    },
+    {
+      "epoch": 1.8962025316455695,
+      "grad_norm": 0.41736599802970886,
+      "learning_rate": 0.00016926133234875396,
+      "loss": 0.700016438961029,
+      "step": 4494
+    },
+    {
+      "epoch": 1.8970464135021097,
+      "grad_norm": 0.4086865484714508,
+      "learning_rate": 0.0001692266804339785,
+      "loss": 0.6926955580711365,
+      "step": 4496
+    },
+    {
+      "epoch": 1.8978902953586498,
+      "grad_norm": 0.447516530752182,
+      "learning_rate": 0.00016919201254993738,
+      "loss": 0.6776167154312134,
+      "step": 4498
+    },
+    {
+      "epoch": 1.8987341772151898,
+      "grad_norm": 0.43951377272605896,
+      "learning_rate": 0.00016915732870462783,
+      "loss": 0.6860454678535461,
+      "step": 4500
+    },
+    {
+      "epoch": 1.8987341772151898,
+      "eval_loss": 0.7600579857826233,
+      "eval_runtime": 674.2593,
+      "eval_samples_per_second": 3.125,
+      "eval_steps_per_second": 3.125,
+      "step": 4500
+    },
+    {
+      "epoch": 1.89957805907173,
+      "grad_norm": 0.380329966545105,
+      "learning_rate": 0.00016912262890605076,
+      "loss": 0.6525024175643921,
+      "step": 4502
+    },
+    {
+      "epoch": 1.90042194092827,
+      "grad_norm": 0.3461032807826996,
+      "learning_rate": 0.00016908791316221073,
+      "loss": 0.6473777294158936,
+      "step": 4504
+    },
+    {
+      "epoch": 1.90126582278481,
+      "grad_norm": 0.47325965762138367,
+      "learning_rate": 0.00016905318148111595,
+      "loss": 0.760007381439209,
+      "step": 4506
+    },
+    {
+      "epoch": 1.9021097046413502,
+      "grad_norm": 0.39158931374549866,
+      "learning_rate": 0.00016901843387077844,
+      "loss": 0.659562349319458,
+      "step": 4508
+    },
+    {
+      "epoch": 1.9029535864978904,
+      "grad_norm": 0.36785584688186646,
+      "learning_rate": 0.0001689836703392137,
+      "loss": 0.5873214602470398,
+      "step": 4510
+    },
+    {
+      "epoch": 1.9037974683544303,
+      "grad_norm": 0.3843834102153778,
+      "learning_rate": 0.00016894889089444115,
+      "loss": 0.6236514449119568,
+      "step": 4512
+    },
+    {
+      "epoch": 1.9046413502109705,
+      "grad_norm": 0.41675102710723877,
+      "learning_rate": 0.0001689140955444836,
+      "loss": 0.7129080891609192,
+      "step": 4514
+    },
+    {
+      "epoch": 1.9054852320675106,
+      "grad_norm": 0.46025362610816956,
+      "learning_rate": 0.00016887928429736775,
+      "loss": 0.7348461747169495,
+      "step": 4516
+    },
+    {
+      "epoch": 1.9063291139240506,
+      "grad_norm": 0.43589887022972107,
+      "learning_rate": 0.00016884445716112387,
+      "loss": 0.6797979474067688,
+      "step": 4518
+    },
+    {
+      "epoch": 1.9071729957805907,
+      "grad_norm": 0.4301862120628357,
+      "learning_rate": 0.0001688096141437859,
+      "loss": 0.6754241585731506,
+      "step": 4520
+    },
+    {
+      "epoch": 1.908016877637131,
+      "grad_norm": 0.41446763277053833,
+      "learning_rate": 0.00016877475525339146,
+      "loss": 0.681866466999054,
+      "step": 4522
+    },
+    {
+      "epoch": 1.9088607594936708,
+      "grad_norm": 0.4051108658313751,
+      "learning_rate": 0.0001687398804979818,
+      "loss": 0.6626620888710022,
+      "step": 4524
+    },
+    {
+      "epoch": 1.909704641350211,
+      "grad_norm": 0.3896785080432892,
+      "learning_rate": 0.00016870498988560192,
+      "loss": 0.6753385066986084,
+      "step": 4526
+    },
+    {
+      "epoch": 1.9105485232067512,
+      "grad_norm": 0.4307839274406433,
+      "learning_rate": 0.00016867008342430037,
+      "loss": 0.6971913576126099,
+      "step": 4528
+    },
+    {
+      "epoch": 1.9113924050632911,
+      "grad_norm": 0.4120950698852539,
+      "learning_rate": 0.0001686351611221294,
+      "loss": 0.7674229145050049,
+      "step": 4530
+    },
+    {
+      "epoch": 1.9122362869198313,
+      "grad_norm": 0.4321785569190979,
+      "learning_rate": 0.00016860022298714492,
+      "loss": 0.7074664235115051,
+      "step": 4532
+    },
+    {
+      "epoch": 1.9130801687763714,
+      "grad_norm": 0.38313060998916626,
+      "learning_rate": 0.00016856526902740652,
+      "loss": 0.6178744435310364,
+      "step": 4534
+    },
+    {
+      "epoch": 1.9139240506329114,
+      "grad_norm": 0.3940589129924774,
+      "learning_rate": 0.0001685302992509773,
+      "loss": 0.7168449759483337,
+      "step": 4536
+    },
+    {
+      "epoch": 1.9147679324894513,
+      "grad_norm": 0.3719513416290283,
+      "learning_rate": 0.0001684953136659243,
+      "loss": 0.6892635226249695,
+      "step": 4538
+    },
+    {
+      "epoch": 1.9156118143459917,
+      "grad_norm": 0.40150535106658936,
+      "learning_rate": 0.00016846031228031784,
+      "loss": 0.7352524399757385,
+      "step": 4540
+    },
+    {
+      "epoch": 1.9164556962025316,
+      "grad_norm": 0.4212523102760315,
+      "learning_rate": 0.00016842529510223213,
+      "loss": 0.6469122767448425,
+      "step": 4542
+    },
+    {
+      "epoch": 1.9172995780590716,
+      "grad_norm": 0.45365604758262634,
+      "learning_rate": 0.00016839026213974503,
+      "loss": 0.6883940696716309,
+      "step": 4544
+    },
+    {
+      "epoch": 1.918143459915612,
+      "grad_norm": 0.436541885137558,
+      "learning_rate": 0.00016835521340093787,
+      "loss": 0.7378416657447815,
+      "step": 4546
+    },
+    {
+      "epoch": 1.918987341772152,
+      "grad_norm": 0.41461536288261414,
+      "learning_rate": 0.0001683201488938958,
+      "loss": 0.7432116270065308,
+      "step": 4548
+    },
+    {
+      "epoch": 1.9198312236286919,
+      "grad_norm": 0.40218585729599,
+      "learning_rate": 0.00016828506862670746,
+      "loss": 0.6801432967185974,
+      "step": 4550
+    },
+    {
+      "epoch": 1.920675105485232,
+      "grad_norm": 0.3994768559932709,
+      "learning_rate": 0.00016824997260746525,
+      "loss": 0.6835547685623169,
+      "step": 4552
+    },
+    {
+      "epoch": 1.9215189873417722,
+      "grad_norm": 0.460328072309494,
+      "learning_rate": 0.00016821486084426512,
+      "loss": 0.7723629474639893,
+      "step": 4554
+    },
+    {
+      "epoch": 1.9223628691983121,
+      "grad_norm": 0.4739786386489868,
+      "learning_rate": 0.00016817973334520668,
+      "loss": 0.7513449192047119,
+      "step": 4556
+    },
+    {
+      "epoch": 1.9232067510548523,
+      "grad_norm": 0.46618419885635376,
+      "learning_rate": 0.00016814459011839315,
+      "loss": 0.7000700235366821,
+      "step": 4558
+    },
+    {
+      "epoch": 1.9240506329113924,
+      "grad_norm": 0.4187009632587433,
+      "learning_rate": 0.00016810943117193146,
+      "loss": 0.6381427049636841,
+      "step": 4560
+    },
+    {
+      "epoch": 1.9248945147679324,
+      "grad_norm": 0.5057410597801208,
+      "learning_rate": 0.00016807425651393207,
+      "loss": 0.7823911309242249,
+      "step": 4562
+    },
+    {
+      "epoch": 1.9257383966244725,
+      "grad_norm": 0.4235171973705292,
+      "learning_rate": 0.00016803906615250907,
+      "loss": 0.7210549712181091,
+      "step": 4564
+    },
+    {
+      "epoch": 1.9265822784810127,
+      "grad_norm": 0.45755431056022644,
+      "learning_rate": 0.00016800386009578025,
+      "loss": 0.794435441493988,
+      "step": 4566
+    },
+    {
+      "epoch": 1.9274261603375527,
+      "grad_norm": 0.4225042760372162,
+      "learning_rate": 0.0001679686383518669,
+      "loss": 0.6490750312805176,
+      "step": 4568
+    },
+    {
+      "epoch": 1.9282700421940928,
+      "grad_norm": 0.41081616282463074,
+      "learning_rate": 0.00016793340092889408,
+      "loss": 0.6976916790008545,
+      "step": 4570
+    },
+    {
+      "epoch": 1.929113924050633,
+      "grad_norm": 0.42354175448417664,
+      "learning_rate": 0.00016789814783499033,
+      "loss": 0.6559295058250427,
+      "step": 4572
+    },
+    {
+      "epoch": 1.929957805907173,
+      "grad_norm": 0.39702117443084717,
+      "learning_rate": 0.0001678628790782879,
+      "loss": 0.7900391817092896,
+      "step": 4574
+    },
+    {
+      "epoch": 1.930801687763713,
+      "grad_norm": 0.4400629997253418,
+      "learning_rate": 0.00016782759466692256,
+      "loss": 0.7020596265792847,
+      "step": 4576
+    },
+    {
+      "epoch": 1.9316455696202532,
+      "grad_norm": 0.42563396692276,
+      "learning_rate": 0.00016779229460903383,
+      "loss": 0.7107458710670471,
+      "step": 4578
+    },
+    {
+      "epoch": 1.9324894514767932,
+      "grad_norm": 0.40650051832199097,
+      "learning_rate": 0.0001677569789127647,
+      "loss": 0.58088618516922,
+      "step": 4580
+    },
+    {
+      "epoch": 1.9333333333333333,
+      "grad_norm": 0.42461681365966797,
+      "learning_rate": 0.00016772164758626184,
+      "loss": 0.6815688610076904,
+      "step": 4582
+    },
+    {
+      "epoch": 1.9341772151898735,
+      "grad_norm": 0.4248030483722687,
+      "learning_rate": 0.0001676863006376755,
+      "loss": 0.6617820262908936,
+      "step": 4584
+    },
+    {
+      "epoch": 1.9350210970464135,
+      "grad_norm": 0.4336024522781372,
+      "learning_rate": 0.0001676509380751595,
+      "loss": 0.7462296485900879,
+      "step": 4586
+    },
+    {
+      "epoch": 1.9358649789029536,
+      "grad_norm": 0.42542141675949097,
+      "learning_rate": 0.00016761555990687136,
+      "loss": 0.7801896333694458,
+      "step": 4588
+    },
+    {
+      "epoch": 1.9367088607594938,
+      "grad_norm": 0.41240575909614563,
+      "learning_rate": 0.00016758016614097217,
+      "loss": 0.7308001518249512,
+      "step": 4590
+    },
+    {
+      "epoch": 1.9375527426160337,
+      "grad_norm": 0.3811012804508209,
+      "learning_rate": 0.00016754475678562654,
+      "loss": 0.7126680016517639,
+      "step": 4592
+    },
+    {
+      "epoch": 1.9383966244725739,
+      "grad_norm": 0.4859951436519623,
+      "learning_rate": 0.00016750933184900272,
+      "loss": 0.679752767086029,
+      "step": 4594
+    },
+    {
+      "epoch": 1.939240506329114,
+      "grad_norm": 0.36403799057006836,
+      "learning_rate": 0.00016747389133927261,
+      "loss": 0.6077587604522705,
+      "step": 4596
+    },
+    {
+      "epoch": 1.940084388185654,
+      "grad_norm": 0.3960701525211334,
+      "learning_rate": 0.00016743843526461158,
+      "loss": 0.6919949650764465,
+      "step": 4598
+    },
+    {
+      "epoch": 1.9409282700421941,
+      "grad_norm": 0.4058634042739868,
+      "learning_rate": 0.00016740296363319877,
+      "loss": 0.728607177734375,
+      "step": 4600
+    },
+    {
+      "epoch": 1.9409282700421941,
+      "eval_loss": 0.7585541009902954,
+      "eval_runtime": 679.0866,
+      "eval_samples_per_second": 3.103,
+      "eval_steps_per_second": 3.103,
+      "step": 4600
+    },
+    {
+      "epoch": 1.9417721518987343,
+      "grad_norm": 0.45107272267341614,
+      "learning_rate": 0.00016736747645321672,
+      "loss": 0.669603168964386,
+      "step": 4602
+    },
+    {
+      "epoch": 1.9426160337552743,
+      "grad_norm": 0.4244096577167511,
+      "learning_rate": 0.00016733197373285165,
+      "loss": 0.6975913643836975,
+      "step": 4604
+    },
+    {
+      "epoch": 1.9434599156118142,
+      "grad_norm": 0.37567681074142456,
+      "learning_rate": 0.00016729645548029342,
+      "loss": 0.5990989804267883,
+      "step": 4606
+    },
+    {
+      "epoch": 1.9443037974683546,
+      "grad_norm": 0.40256237983703613,
+      "learning_rate": 0.00016726092170373533,
+      "loss": 0.723430871963501,
+      "step": 4608
+    },
+    {
+      "epoch": 1.9451476793248945,
+      "grad_norm": 0.4884983003139496,
+      "learning_rate": 0.0001672253724113744,
+      "loss": 0.7460100650787354,
+      "step": 4610
+    },
+    {
+      "epoch": 1.9459915611814345,
+      "grad_norm": 0.3629630208015442,
+      "learning_rate": 0.0001671898076114111,
+      "loss": 0.5577017664909363,
+      "step": 4612
+    },
+    {
+      "epoch": 1.9468354430379748,
+      "grad_norm": 0.42751777172088623,
+      "learning_rate": 0.00016715422731204967,
+      "loss": 0.7394112944602966,
+      "step": 4614
+    },
+    {
+      "epoch": 1.9476793248945148,
+      "grad_norm": 0.35809609293937683,
+      "learning_rate": 0.00016711863152149765,
+      "loss": 0.5905179381370544,
+      "step": 4616
+    },
+    {
+      "epoch": 1.9485232067510547,
+      "grad_norm": 0.4111083745956421,
+      "learning_rate": 0.0001670830202479664,
+      "loss": 0.7035338878631592,
+      "step": 4618
+    },
+    {
+      "epoch": 1.9493670886075949,
+      "grad_norm": 0.4530862867832184,
+      "learning_rate": 0.0001670473934996707,
+      "loss": 0.7048936486244202,
+      "step": 4620
+    },
+    {
+      "epoch": 1.950210970464135,
+      "grad_norm": 0.41778260469436646,
+      "learning_rate": 0.00016701175128482903,
+      "loss": 0.6594944596290588,
+      "step": 4622
+    },
+    {
+      "epoch": 1.951054852320675,
+      "grad_norm": 0.40512707829475403,
+      "learning_rate": 0.0001669760936116633,
+      "loss": 0.6902912259101868,
+      "step": 4624
+    },
+    {
+      "epoch": 1.9518987341772152,
+      "grad_norm": 0.4226403832435608,
+      "learning_rate": 0.0001669404204883991,
+      "loss": 0.7086893320083618,
+      "step": 4626
+    },
+    {
+      "epoch": 1.9527426160337553,
+      "grad_norm": 0.397930383682251,
+      "learning_rate": 0.00016690473192326549,
+      "loss": 0.6647822856903076,
+      "step": 4628
+    },
+    {
+      "epoch": 1.9535864978902953,
+      "grad_norm": 0.3985617458820343,
+      "learning_rate": 0.00016686902792449515,
+      "loss": 0.747227668762207,
+      "step": 4630
+    },
+    {
+      "epoch": 1.9544303797468354,
+      "grad_norm": 0.4174083173274994,
+      "learning_rate": 0.00016683330850032432,
+      "loss": 0.7526347637176514,
+      "step": 4632
+    },
+    {
+      "epoch": 1.9552742616033756,
+      "grad_norm": 0.425070583820343,
+      "learning_rate": 0.00016679757365899275,
+      "loss": 0.6674112677574158,
+      "step": 4634
+    },
+    {
+      "epoch": 1.9561181434599155,
+      "grad_norm": 0.4049135148525238,
+      "learning_rate": 0.00016676182340874386,
+      "loss": 0.6515055298805237,
+      "step": 4636
+    },
+    {
+      "epoch": 1.9569620253164557,
+      "grad_norm": 0.4312315285205841,
+      "learning_rate": 0.00016672605775782447,
+      "loss": 0.645137369632721,
+      "step": 4638
+    },
+    {
+      "epoch": 1.9578059071729959,
+      "grad_norm": 0.39540281891822815,
+      "learning_rate": 0.00016669027671448503,
+      "loss": 0.6519104838371277,
+      "step": 4640
+    },
+    {
+      "epoch": 1.9586497890295358,
+      "grad_norm": 0.5028185844421387,
+      "learning_rate": 0.00016665448028697961,
+      "loss": 0.6815812587738037,
+      "step": 4642
+    },
+    {
+      "epoch": 1.959493670886076,
+      "grad_norm": 0.4395635724067688,
+      "learning_rate": 0.00016661866848356567,
+      "loss": 0.6713913083076477,
+      "step": 4644
+    },
+    {
+      "epoch": 1.9603375527426161,
+      "grad_norm": 0.42733073234558105,
+      "learning_rate": 0.00016658284131250436,
+      "loss": 0.6989195942878723,
+      "step": 4646
+    },
+    {
+      "epoch": 1.961181434599156,
+      "grad_norm": 0.41415879130363464,
+      "learning_rate": 0.00016654699878206031,
+      "loss": 0.6501366496086121,
+      "step": 4648
+    },
+    {
+      "epoch": 1.9620253164556962,
+      "grad_norm": 0.39072170853614807,
+      "learning_rate": 0.0001665111409005017,
+      "loss": 0.6437955498695374,
+      "step": 4650
+    },
+    {
+      "epoch": 1.9628691983122364,
+      "grad_norm": 0.41802850365638733,
+      "learning_rate": 0.00016647526767610025,
+      "loss": 0.6521314382553101,
+      "step": 4652
+    },
+    {
+      "epoch": 1.9637130801687763,
+      "grad_norm": 0.41475021839141846,
+      "learning_rate": 0.00016643937911713124,
+      "loss": 0.7300742864608765,
+      "step": 4654
+    },
+    {
+      "epoch": 1.9645569620253165,
+      "grad_norm": 0.41763070225715637,
+      "learning_rate": 0.00016640347523187344,
+      "loss": 0.6217485070228577,
+      "step": 4656
+    },
+    {
+      "epoch": 1.9654008438818567,
+      "grad_norm": 0.4488297402858734,
+      "learning_rate": 0.00016636755602860922,
+      "loss": 0.7974472641944885,
+      "step": 4658
+    },
+    {
+      "epoch": 1.9662447257383966,
+      "grad_norm": 0.37688007950782776,
+      "learning_rate": 0.00016633162151562445,
+      "loss": 0.728827714920044,
+      "step": 4660
+    },
+    {
+      "epoch": 1.9670886075949368,
+      "grad_norm": 0.4636184275150299,
+      "learning_rate": 0.0001662956717012085,
+      "loss": 0.7563744783401489,
+      "step": 4662
+    },
+    {
+      "epoch": 1.967932489451477,
+      "grad_norm": 0.432020902633667,
+      "learning_rate": 0.00016625970659365434,
+      "loss": 0.7408145666122437,
+      "step": 4664
+    },
+    {
+      "epoch": 1.9687763713080169,
+      "grad_norm": 0.4098186790943146,
+      "learning_rate": 0.0001662237262012584,
+      "loss": 0.6961510181427002,
+      "step": 4666
+    },
+    {
+      "epoch": 1.9696202531645568,
+      "grad_norm": 0.44124510884284973,
+      "learning_rate": 0.00016618773053232068,
+      "loss": 0.6959825754165649,
+      "step": 4668
+    },
+    {
+      "epoch": 1.9704641350210972,
+      "grad_norm": 0.3729988634586334,
+      "learning_rate": 0.00016615171959514466,
+      "loss": 0.6693721413612366,
+      "step": 4670
+    },
+    {
+      "epoch": 1.9713080168776371,
+      "grad_norm": 0.4506371021270752,
+      "learning_rate": 0.00016611569339803744,
+      "loss": 0.6943060159683228,
+      "step": 4672
+    },
+    {
+      "epoch": 1.972151898734177,
+      "grad_norm": 0.43713298439979553,
+      "learning_rate": 0.00016607965194930947,
+      "loss": 0.6915813684463501,
+      "step": 4674
+    },
+    {
+      "epoch": 1.9729957805907175,
+      "grad_norm": 0.4375544786453247,
+      "learning_rate": 0.00016604359525727492,
+      "loss": 0.7430137991905212,
+      "step": 4676
+    },
+    {
+      "epoch": 1.9738396624472574,
+      "grad_norm": 0.4393104314804077,
+      "learning_rate": 0.00016600752333025134,
+      "loss": 0.7260952591896057,
+      "step": 4678
+    },
+    {
+      "epoch": 1.9746835443037973,
+      "grad_norm": 0.4471568167209625,
+      "learning_rate": 0.00016597143617655982,
+      "loss": 0.7350760102272034,
+      "step": 4680
+    },
+    {
+      "epoch": 1.9755274261603377,
+      "grad_norm": 0.39821380376815796,
+      "learning_rate": 0.00016593533380452499,
+      "loss": 0.7295693159103394,
+      "step": 4682
+    },
+    {
+      "epoch": 1.9763713080168777,
+      "grad_norm": 0.40367990732192993,
+      "learning_rate": 0.00016589921622247493,
+      "loss": 0.6461828947067261,
+      "step": 4684
+    },
+    {
+      "epoch": 1.9772151898734176,
+      "grad_norm": 0.4723254442214966,
+      "learning_rate": 0.00016586308343874136,
+      "loss": 0.699313223361969,
+      "step": 4686
+    },
+    {
+      "epoch": 1.9780590717299578,
+      "grad_norm": 0.45741328597068787,
+      "learning_rate": 0.0001658269354616593,
+      "loss": 0.7299131751060486,
+      "step": 4688
+    },
+    {
+      "epoch": 1.978902953586498,
+      "grad_norm": 0.4011725187301636,
+      "learning_rate": 0.0001657907722995675,
+      "loss": 0.7643240690231323,
+      "step": 4690
+    },
+    {
+      "epoch": 1.9797468354430379,
+      "grad_norm": 0.3904820382595062,
+      "learning_rate": 0.00016575459396080805,
+      "loss": 0.6945428848266602,
+      "step": 4692
+    },
+    {
+      "epoch": 1.980590717299578,
+      "grad_norm": 0.3643582761287689,
+      "learning_rate": 0.00016571840045372663,
+      "loss": 0.5986754298210144,
+      "step": 4694
+    },
+    {
+      "epoch": 1.9814345991561182,
+      "grad_norm": 0.4371087849140167,
+      "learning_rate": 0.00016568219178667235,
+      "loss": 0.7462178468704224,
+      "step": 4696
+    },
+    {
+      "epoch": 1.9822784810126581,
+      "grad_norm": 0.4440254867076874,
+      "learning_rate": 0.0001656459679679979,
+      "loss": 0.7553940415382385,
+      "step": 4698
+    },
+    {
+      "epoch": 1.9831223628691983,
+      "grad_norm": 0.41370099782943726,
+      "learning_rate": 0.00016560972900605937,
+      "loss": 0.7119964361190796,
+      "step": 4700
+    },
+    {
+      "epoch": 1.9831223628691983,
+      "eval_loss": 0.7582268714904785,
+      "eval_runtime": 690.081,
+      "eval_samples_per_second": 3.053,
+      "eval_steps_per_second": 3.053,
+      "step": 4700
+    },
+    {
+      "epoch": 1.9839662447257385,
+      "grad_norm": 0.4408273696899414,
+      "learning_rate": 0.0001655734749092164,
+      "loss": 0.6260318160057068,
+      "step": 4702
+    },
+    {
+      "epoch": 1.9848101265822784,
+      "grad_norm": 0.4697865843772888,
+      "learning_rate": 0.0001655372056858322,
+      "loss": 0.7382199168205261,
+      "step": 4704
+    },
+    {
+      "epoch": 1.9856540084388186,
+      "grad_norm": 0.423520028591156,
+      "learning_rate": 0.00016550092134427327,
+      "loss": 0.7037063241004944,
+      "step": 4706
+    },
+    {
+      "epoch": 1.9864978902953587,
+      "grad_norm": 0.38633009791374207,
+      "learning_rate": 0.00016546462189290974,
+      "loss": 0.7173316478729248,
+      "step": 4708
+    },
+    {
+      "epoch": 1.9873417721518987,
+      "grad_norm": 0.410230815410614,
+      "learning_rate": 0.00016542830734011523,
+      "loss": 0.6884669065475464,
+      "step": 4710
+    },
+    {
+      "epoch": 1.9881856540084388,
+      "grad_norm": 0.4486774802207947,
+      "learning_rate": 0.00016539197769426678,
+      "loss": 0.7181811332702637,
+      "step": 4712
+    },
+    {
+      "epoch": 1.989029535864979,
+      "grad_norm": 0.47757431864738464,
+      "learning_rate": 0.00016535563296374495,
+      "loss": 0.7737504243850708,
+      "step": 4714
+    },
+    {
+      "epoch": 1.989873417721519,
+      "grad_norm": 0.40281420946121216,
+      "learning_rate": 0.00016531927315693382,
+      "loss": 0.683630645275116,
+      "step": 4716
+    },
+    {
+      "epoch": 1.990717299578059,
+      "grad_norm": 0.42349720001220703,
+      "learning_rate": 0.00016528289828222082,
+      "loss": 0.7447960376739502,
+      "step": 4718
+    },
+    {
+      "epoch": 1.9915611814345993,
+      "grad_norm": 0.40754765272140503,
+      "learning_rate": 0.00016524650834799694,
+      "loss": 0.6800103783607483,
+      "step": 4720
+    },
+    {
+      "epoch": 1.9924050632911392,
+      "grad_norm": 0.3766736090183258,
+      "learning_rate": 0.0001652101033626567,
+      "loss": 0.6609475612640381,
+      "step": 4722
+    },
+    {
+      "epoch": 1.9932489451476794,
+      "grad_norm": 0.3600929081439972,
+      "learning_rate": 0.00016517368333459798,
+      "loss": 0.6496920585632324,
+      "step": 4724
+    },
+    {
+      "epoch": 1.9940928270042195,
+      "grad_norm": 0.39590612053871155,
+      "learning_rate": 0.00016513724827222227,
+      "loss": 0.6305265426635742,
+      "step": 4726
+    },
+    {
+      "epoch": 1.9949367088607595,
+      "grad_norm": 0.41002732515335083,
+      "learning_rate": 0.00016510079818393426,
+      "loss": 0.712282657623291,
+      "step": 4728
+    },
+    {
+      "epoch": 1.9957805907172996,
+      "grad_norm": 0.41948413848876953,
+      "learning_rate": 0.00016506433307814246,
+      "loss": 0.6974884271621704,
+      "step": 4730
+    },
+    {
+      "epoch": 1.9966244725738398,
+      "grad_norm": 0.4283764958381653,
+      "learning_rate": 0.00016502785296325858,
+      "loss": 0.6844019889831543,
+      "step": 4732
+    },
+    {
+      "epoch": 1.9974683544303797,
+      "grad_norm": 0.39830246567726135,
+      "learning_rate": 0.0001649913578476979,
+      "loss": 0.7453671097755432,
+      "step": 4734
+    },
+    {
+      "epoch": 1.9983122362869197,
+      "grad_norm": 0.3875660300254822,
+      "learning_rate": 0.00016495484773987917,
+      "loss": 0.6571418046951294,
+      "step": 4736
+    },
+    {
+      "epoch": 1.99915611814346,
+      "grad_norm": 0.379986971616745,
+      "learning_rate": 0.00016491832264822453,
+      "loss": 0.6844568848609924,
+      "step": 4738
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.38605886697769165,
+      "learning_rate": 0.00016488178258115965,
+      "loss": 0.6115277409553528,
+      "step": 4740
+    },
+    {
+      "epoch": 2.00084388185654,
+      "grad_norm": 0.3991186022758484,
+      "learning_rate": 0.0001648452275471136,
+      "loss": 0.6192594170570374,
+      "step": 4742
+    },
+    {
+      "epoch": 2.0016877637130803,
+      "grad_norm": 0.41378065943717957,
+      "learning_rate": 0.00016480865755451893,
+      "loss": 0.6478689312934875,
+      "step": 4744
+    },
+    {
+      "epoch": 2.0025316455696203,
+      "grad_norm": 0.41938909888267517,
+      "learning_rate": 0.00016477207261181162,
+      "loss": 0.6162620186805725,
+      "step": 4746
+    },
+    {
+      "epoch": 2.00337552742616,
+      "grad_norm": 0.40436387062072754,
+      "learning_rate": 0.00016473547272743115,
+      "loss": 0.5328736901283264,
+      "step": 4748
+    },
+    {
+      "epoch": 2.0042194092827006,
+      "grad_norm": 0.4148593544960022,
+      "learning_rate": 0.00016469885790982037,
+      "loss": 0.547049880027771,
+      "step": 4750
+    },
+    {
+      "epoch": 2.0050632911392405,
+      "grad_norm": 0.42139601707458496,
+      "learning_rate": 0.00016466222816742564,
+      "loss": 0.5509368181228638,
+      "step": 4752
+    },
+    {
+      "epoch": 2.0059071729957805,
+      "grad_norm": 0.4509783089160919,
+      "learning_rate": 0.00016462558350869677,
+      "loss": 0.5994526147842407,
+      "step": 4754
+    },
+    {
+      "epoch": 2.006751054852321,
+      "grad_norm": 0.4232040345668793,
+      "learning_rate": 0.0001645889239420869,
+      "loss": 0.6348381638526917,
+      "step": 4756
+    },
+    {
+      "epoch": 2.007594936708861,
+      "grad_norm": 0.4259628355503082,
+      "learning_rate": 0.00016455224947605274,
+      "loss": 0.5487880110740662,
+      "step": 4758
+    },
+    {
+      "epoch": 2.0084388185654007,
+      "grad_norm": 0.5485594868659973,
+      "learning_rate": 0.0001645155601190544,
+      "loss": 0.5468270778656006,
+      "step": 4760
+    },
+    {
+      "epoch": 2.009282700421941,
+      "grad_norm": 0.4243668019771576,
+      "learning_rate": 0.00016447885587955537,
+      "loss": 0.6165785789489746,
+      "step": 4762
+    },
+    {
+      "epoch": 2.010126582278481,
+      "grad_norm": 0.46513810753822327,
+      "learning_rate": 0.00016444213676602263,
+      "loss": 0.5868126749992371,
+      "step": 4764
+    },
+    {
+      "epoch": 2.010970464135021,
+      "grad_norm": 0.4555842876434326,
+      "learning_rate": 0.00016440540278692654,
+      "loss": 0.548373818397522,
+      "step": 4766
+    },
+    {
+      "epoch": 2.0118143459915614,
+      "grad_norm": 0.4402005672454834,
+      "learning_rate": 0.000164368653950741,
+      "loss": 0.6254384517669678,
+      "step": 4768
+    },
+    {
+      "epoch": 2.0126582278481013,
+      "grad_norm": 0.5100982785224915,
+      "learning_rate": 0.0001643318902659432,
+      "loss": 0.6335576772689819,
+      "step": 4770
+    },
+    {
+      "epoch": 2.0135021097046413,
+      "grad_norm": 0.508076548576355,
+      "learning_rate": 0.0001642951117410138,
+      "loss": 0.6601388454437256,
+      "step": 4772
+    },
+    {
+      "epoch": 2.014345991561181,
+      "grad_norm": 0.47249099612236023,
+      "learning_rate": 0.000164258318384437,
+      "loss": 0.6183612942695618,
+      "step": 4774
+    },
+    {
+      "epoch": 2.0151898734177216,
+      "grad_norm": 0.459249883890152,
+      "learning_rate": 0.00016422151020470022,
+      "loss": 0.6172632575035095,
+      "step": 4776
+    },
+    {
+      "epoch": 2.0160337552742615,
+      "grad_norm": 0.44374513626098633,
+      "learning_rate": 0.00016418468721029442,
+      "loss": 0.6392593383789062,
+      "step": 4778
+    },
+    {
+      "epoch": 2.0168776371308015,
+      "grad_norm": 0.4633971154689789,
+      "learning_rate": 0.000164147849409714,
+      "loss": 0.6331546902656555,
+      "step": 4780
+    },
+    {
+      "epoch": 2.017721518987342,
+      "grad_norm": 0.4080882668495178,
+      "learning_rate": 0.00016411099681145667,
+      "loss": 0.5368412137031555,
+      "step": 4782
+    },
+    {
+      "epoch": 2.018565400843882,
+      "grad_norm": 0.3852706253528595,
+      "learning_rate": 0.00016407412942402366,
+      "loss": 0.4763215184211731,
+      "step": 4784
+    },
+    {
+      "epoch": 2.0194092827004217,
+      "grad_norm": 0.45222651958465576,
+      "learning_rate": 0.00016403724725591958,
+      "loss": 0.6510329842567444,
+      "step": 4786
+    },
+    {
+      "epoch": 2.020253164556962,
+      "grad_norm": 0.4027794897556305,
+      "learning_rate": 0.00016400035031565238,
+      "loss": 0.5856698751449585,
+      "step": 4788
+    },
+    {
+      "epoch": 2.021097046413502,
+      "grad_norm": 0.3780474066734314,
+      "learning_rate": 0.00016396343861173356,
+      "loss": 0.5960753560066223,
+      "step": 4790
+    },
+    {
+      "epoch": 2.021940928270042,
+      "grad_norm": 0.4297144412994385,
+      "learning_rate": 0.00016392651215267785,
+      "loss": 0.5696688890457153,
+      "step": 4792
+    },
+    {
+      "epoch": 2.0227848101265824,
+      "grad_norm": 0.4521540701389313,
+      "learning_rate": 0.00016388957094700356,
+      "loss": 0.6610676050186157,
+      "step": 4794
+    },
+    {
+      "epoch": 2.0236286919831223,
+      "grad_norm": 0.41184982657432556,
+      "learning_rate": 0.00016385261500323224,
+      "loss": 0.5387522578239441,
+      "step": 4796
+    },
+    {
+      "epoch": 2.0244725738396623,
+      "grad_norm": 0.489391565322876,
+      "learning_rate": 0.00016381564432988897,
+      "loss": 0.6614878177642822,
+      "step": 4798
+    },
+    {
+      "epoch": 2.0253164556962027,
+      "grad_norm": 0.3741590678691864,
+      "learning_rate": 0.00016377865893550214,
+      "loss": 0.5513694286346436,
+      "step": 4800
+    },
+    {
+      "epoch": 2.0253164556962027,
+      "eval_loss": 0.7633068561553955,
+      "eval_runtime": 688.8684,
+      "eval_samples_per_second": 3.059,
+      "eval_steps_per_second": 3.059,
+      "step": 4800
+    },
+    {
+      "epoch": 2.0261603375527426,
+      "grad_norm": 0.45757874846458435,
+      "learning_rate": 0.0001637416588286036,
+      "loss": 0.5732831358909607,
+      "step": 4802
+    },
+    {
+      "epoch": 2.0270042194092825,
+      "grad_norm": 0.4009305238723755,
+      "learning_rate": 0.00016370464401772853,
+      "loss": 0.5994226932525635,
+      "step": 4804
+    },
+    {
+      "epoch": 2.027848101265823,
+      "grad_norm": 0.44907137751579285,
+      "learning_rate": 0.0001636676145114156,
+      "loss": 0.6229469776153564,
+      "step": 4806
+    },
+    {
+      "epoch": 2.028691983122363,
+      "grad_norm": 0.4153282344341278,
+      "learning_rate": 0.00016363057031820673,
+      "loss": 0.5126086473464966,
+      "step": 4808
+    },
+    {
+      "epoch": 2.029535864978903,
+      "grad_norm": 0.41859957575798035,
+      "learning_rate": 0.0001635935114466473,
+      "loss": 0.5779650211334229,
+      "step": 4810
+    },
+    {
+      "epoch": 2.030379746835443,
+      "grad_norm": 0.44624394178390503,
+      "learning_rate": 0.00016355643790528617,
+      "loss": 0.5561378598213196,
+      "step": 4812
+    },
+    {
+      "epoch": 2.031223628691983,
+      "grad_norm": 0.44554266333580017,
+      "learning_rate": 0.00016351934970267542,
+      "loss": 0.5381612181663513,
+      "step": 4814
+    },
+    {
+      "epoch": 2.032067510548523,
+      "grad_norm": 0.4529551565647125,
+      "learning_rate": 0.00016348224684737065,
+      "loss": 0.6434502601623535,
+      "step": 4816
+    },
+    {
+      "epoch": 2.0329113924050635,
+      "grad_norm": 0.891188383102417,
+      "learning_rate": 0.00016344512934793066,
+      "loss": 0.6316392421722412,
+      "step": 4818
+    },
+    {
+      "epoch": 2.0337552742616034,
+      "grad_norm": 0.4621472656726837,
+      "learning_rate": 0.00016340799721291783,
+      "loss": 0.5569570064544678,
+      "step": 4820
+    },
+    {
+      "epoch": 2.0345991561181433,
+      "grad_norm": 0.38671162724494934,
+      "learning_rate": 0.00016337085045089786,
+      "loss": 0.5403776168823242,
+      "step": 4822
+    },
+    {
+      "epoch": 2.0354430379746837,
+      "grad_norm": 0.3896256685256958,
+      "learning_rate": 0.00016333368907043972,
+      "loss": 0.5629472732543945,
+      "step": 4824
+    },
+    {
+      "epoch": 2.0362869198312237,
+      "grad_norm": 0.4890784025192261,
+      "learning_rate": 0.00016329651308011583,
+      "loss": 0.5740808844566345,
+      "step": 4826
+    },
+    {
+      "epoch": 2.0371308016877636,
+      "grad_norm": 0.493118554353714,
+      "learning_rate": 0.00016325932248850207,
+      "loss": 0.6362084746360779,
+      "step": 4828
+    },
+    {
+      "epoch": 2.037974683544304,
+      "grad_norm": 0.4112105965614319,
+      "learning_rate": 0.0001632221173041775,
+      "loss": 0.5512531399726868,
+      "step": 4830
+    },
+    {
+      "epoch": 2.038818565400844,
+      "grad_norm": 0.48197677731513977,
+      "learning_rate": 0.00016318489753572464,
+      "loss": 0.5507554411888123,
+      "step": 4832
+    },
+    {
+      "epoch": 2.039662447257384,
+      "grad_norm": 0.5130576491355896,
+      "learning_rate": 0.00016314766319172947,
+      "loss": 0.7607873678207397,
+      "step": 4834
+    },
+    {
+      "epoch": 2.0405063291139243,
+      "grad_norm": 0.4454704523086548,
+      "learning_rate": 0.00016311041428078112,
+      "loss": 0.6044831275939941,
+      "step": 4836
+    },
+    {
+      "epoch": 2.041350210970464,
+      "grad_norm": 0.4688933789730072,
+      "learning_rate": 0.00016307315081147227,
+      "loss": 0.587941586971283,
+      "step": 4838
+    },
+    {
+      "epoch": 2.042194092827004,
+      "grad_norm": 0.45465248823165894,
+      "learning_rate": 0.0001630358727923989,
+      "loss": 0.5822482109069824,
+      "step": 4840
+    },
+    {
+      "epoch": 2.043037974683544,
+      "grad_norm": 0.4670163691043854,
+      "learning_rate": 0.00016299858023216028,
+      "loss": 0.584092378616333,
+      "step": 4842
+    },
+    {
+      "epoch": 2.0438818565400845,
+      "grad_norm": 0.43894922733306885,
+      "learning_rate": 0.0001629612731393591,
+      "loss": 0.5876650214195251,
+      "step": 4844
+    },
+    {
+      "epoch": 2.0447257383966244,
+      "grad_norm": 0.4144253730773926,
+      "learning_rate": 0.0001629239515226014,
+      "loss": 0.5628184676170349,
+      "step": 4846
+    },
+    {
+      "epoch": 2.0455696202531644,
+      "grad_norm": 0.47422146797180176,
+      "learning_rate": 0.0001628866153904966,
+      "loss": 0.6156699657440186,
+      "step": 4848
+    },
+    {
+      "epoch": 2.0464135021097047,
+      "grad_norm": 0.46102508902549744,
+      "learning_rate": 0.00016284926475165735,
+      "loss": 0.5761739015579224,
+      "step": 4850
+    },
+    {
+      "epoch": 2.0472573839662447,
+      "grad_norm": 0.4987565577030182,
+      "learning_rate": 0.00016281189961469977,
+      "loss": 0.6326378583908081,
+      "step": 4852
+    },
+    {
+      "epoch": 2.0481012658227846,
+      "grad_norm": 0.46640047430992126,
+      "learning_rate": 0.0001627745199882433,
+      "loss": 0.6072084903717041,
+      "step": 4854
+    },
+    {
+      "epoch": 2.048945147679325,
+      "grad_norm": 0.5362635254859924,
+      "learning_rate": 0.00016273712588091066,
+      "loss": 0.6507929563522339,
+      "step": 4856
+    },
+    {
+      "epoch": 2.049789029535865,
+      "grad_norm": 0.459285706281662,
+      "learning_rate": 0.00016269971730132796,
+      "loss": 0.6264472007751465,
+      "step": 4858
+    },
+    {
+      "epoch": 2.050632911392405,
+      "grad_norm": 0.4467318058013916,
+      "learning_rate": 0.00016266229425812467,
+      "loss": 0.6185184717178345,
+      "step": 4860
+    },
+    {
+      "epoch": 2.0514767932489453,
+      "grad_norm": 0.4965224266052246,
+      "learning_rate": 0.0001626248567599335,
+      "loss": 0.5723293423652649,
+      "step": 4862
+    },
+    {
+      "epoch": 2.052320675105485,
+      "grad_norm": 0.46917736530303955,
+      "learning_rate": 0.0001625874048153907,
+      "loss": 0.578831672668457,
+      "step": 4864
+    },
+    {
+      "epoch": 2.053164556962025,
+      "grad_norm": 0.498439222574234,
+      "learning_rate": 0.00016254993843313554,
+      "loss": 0.6448227763175964,
+      "step": 4866
+    },
+    {
+      "epoch": 2.0540084388185655,
+      "grad_norm": 0.4801621735095978,
+      "learning_rate": 0.00016251245762181088,
+      "loss": 0.6240531802177429,
+      "step": 4868
+    },
+    {
+      "epoch": 2.0548523206751055,
+      "grad_norm": 0.3652440309524536,
+      "learning_rate": 0.00016247496239006287,
+      "loss": 0.4955635070800781,
+      "step": 4870
+    },
+    {
+      "epoch": 2.0556962025316454,
+      "grad_norm": 0.504094123840332,
+      "learning_rate": 0.00016243745274654084,
+      "loss": 0.6184053421020508,
+      "step": 4872
+    },
+    {
+      "epoch": 2.056540084388186,
+      "grad_norm": 0.4311911463737488,
+      "learning_rate": 0.0001623999286998976,
+      "loss": 0.5592952370643616,
+      "step": 4874
+    },
+    {
+      "epoch": 2.0573839662447257,
+      "grad_norm": 0.5173681974411011,
+      "learning_rate": 0.00016236239025878928,
+      "loss": 0.6024063229560852,
+      "step": 4876
+    },
+    {
+      "epoch": 2.0582278481012657,
+      "grad_norm": 0.41021528840065,
+      "learning_rate": 0.0001623248374318752,
+      "loss": 0.5788836479187012,
+      "step": 4878
+    },
+    {
+      "epoch": 2.059071729957806,
+      "grad_norm": 0.5167528986930847,
+      "learning_rate": 0.00016228727022781807,
+      "loss": 0.6215962767601013,
+      "step": 4880
+    },
+    {
+      "epoch": 2.059915611814346,
+      "grad_norm": 0.5254554152488708,
+      "learning_rate": 0.00016224968865528395,
+      "loss": 0.5767302513122559,
+      "step": 4882
+    },
+    {
+      "epoch": 2.060759493670886,
+      "grad_norm": 0.3826330602169037,
+      "learning_rate": 0.00016221209272294218,
+      "loss": 0.5494593381881714,
+      "step": 4884
+    },
+    {
+      "epoch": 2.0616033755274263,
+      "grad_norm": 0.48100414872169495,
+      "learning_rate": 0.00016217448243946542,
+      "loss": 0.5395565032958984,
+      "step": 4886
+    },
+    {
+      "epoch": 2.0624472573839663,
+      "grad_norm": 0.4237285852432251,
+      "learning_rate": 0.00016213685781352966,
+      "loss": 0.5737000107765198,
+      "step": 4888
+    },
+    {
+      "epoch": 2.0632911392405062,
+      "grad_norm": 0.46409112215042114,
+      "learning_rate": 0.0001620992188538142,
+      "loss": 0.6150708794593811,
+      "step": 4890
+    },
+    {
+      "epoch": 2.0641350210970466,
+      "grad_norm": 0.5053852200508118,
+      "learning_rate": 0.0001620615655690015,
+      "loss": 0.5868361592292786,
+      "step": 4892
+    },
+    {
+      "epoch": 2.0649789029535865,
+      "grad_norm": 0.4552644193172455,
+      "learning_rate": 0.00016202389796777763,
+      "loss": 0.543509304523468,
+      "step": 4894
+    },
+    {
+      "epoch": 2.0658227848101265,
+      "grad_norm": 0.49399474263191223,
+      "learning_rate": 0.00016198621605883164,
+      "loss": 0.6006189584732056,
+      "step": 4896
+    },
+    {
+      "epoch": 2.066666666666667,
+      "grad_norm": 0.4846300184726715,
+      "learning_rate": 0.00016194851985085608,
+      "loss": 0.6238788962364197,
+      "step": 4898
+    },
+    {
+      "epoch": 2.067510548523207,
+      "grad_norm": 0.4207721948623657,
+      "learning_rate": 0.00016191080935254673,
+      "loss": 0.5164834856987,
+      "step": 4900
+    },
+    {
+      "epoch": 2.067510548523207,
+      "eval_loss": 0.7676681280136108,
+      "eval_runtime": 676.0104,
+      "eval_samples_per_second": 3.117,
+      "eval_steps_per_second": 3.117,
+      "step": 4900
+    },
+    {
+      "epoch": 2.0683544303797468,
+      "grad_norm": 0.49213138222694397,
+      "learning_rate": 0.00016187308457260268,
+      "loss": 0.6363024115562439,
+      "step": 4902
+    },
+    {
+      "epoch": 2.0691983122362867,
+      "grad_norm": 0.4619614779949188,
+      "learning_rate": 0.00016183534551972634,
+      "loss": 0.5472462177276611,
+      "step": 4904
+    },
+    {
+      "epoch": 2.070042194092827,
+      "grad_norm": 0.45133984088897705,
+      "learning_rate": 0.00016179759220262335,
+      "loss": 0.5588397979736328,
+      "step": 4906
+    },
+    {
+      "epoch": 2.070886075949367,
+      "grad_norm": 0.43284907937049866,
+      "learning_rate": 0.00016175982463000266,
+      "loss": 0.6439877152442932,
+      "step": 4908
+    },
+    {
+      "epoch": 2.071729957805907,
+      "grad_norm": 0.4439920485019684,
+      "learning_rate": 0.00016172204281057656,
+      "loss": 0.5987089276313782,
+      "step": 4910
+    },
+    {
+      "epoch": 2.0725738396624473,
+      "grad_norm": 0.376004695892334,
+      "learning_rate": 0.0001616842467530606,
+      "loss": 0.53819739818573,
+      "step": 4912
+    },
+    {
+      "epoch": 2.0734177215189873,
+      "grad_norm": 0.42577606439590454,
+      "learning_rate": 0.00016164643646617358,
+      "loss": 0.5739918947219849,
+      "step": 4914
+    },
+    {
+      "epoch": 2.0742616033755272,
+      "grad_norm": 0.4607972502708435,
+      "learning_rate": 0.00016160861195863756,
+      "loss": 0.580957293510437,
+      "step": 4916
+    },
+    {
+      "epoch": 2.0751054852320676,
+      "grad_norm": 0.4114130735397339,
+      "learning_rate": 0.00016157077323917802,
+      "loss": 0.6063762903213501,
+      "step": 4918
+    },
+    {
+      "epoch": 2.0759493670886076,
+      "grad_norm": 0.4415706694126129,
+      "learning_rate": 0.00016153292031652358,
+      "loss": 0.5996072292327881,
+      "step": 4920
+    },
+    {
+      "epoch": 2.0767932489451475,
+      "grad_norm": 0.5539509654045105,
+      "learning_rate": 0.00016149505319940617,
+      "loss": 0.6442107558250427,
+      "step": 4922
+    },
+    {
+      "epoch": 2.077637130801688,
+      "grad_norm": 0.510313868522644,
+      "learning_rate": 0.00016145717189656105,
+      "loss": 0.6175990104675293,
+      "step": 4924
+    },
+    {
+      "epoch": 2.078481012658228,
+      "grad_norm": 0.442271888256073,
+      "learning_rate": 0.00016141927641672666,
+      "loss": 0.6108829975128174,
+      "step": 4926
+    },
+    {
+      "epoch": 2.0793248945147678,
+      "grad_norm": 0.4499345123767853,
+      "learning_rate": 0.00016138136676864477,
+      "loss": 0.6193580031394958,
+      "step": 4928
+    },
+    {
+      "epoch": 2.080168776371308,
+      "grad_norm": 0.36362606287002563,
+      "learning_rate": 0.00016134344296106043,
+      "loss": 0.5283842086791992,
+      "step": 4930
+    },
+    {
+      "epoch": 2.081012658227848,
+      "grad_norm": 0.460282564163208,
+      "learning_rate": 0.00016130550500272195,
+      "loss": 0.5338225960731506,
+      "step": 4932
+    },
+    {
+      "epoch": 2.081856540084388,
+      "grad_norm": 0.4316645860671997,
+      "learning_rate": 0.0001612675529023808,
+      "loss": 0.6192039251327515,
+      "step": 4934
+    },
+    {
+      "epoch": 2.0827004219409284,
+      "grad_norm": 0.4333813488483429,
+      "learning_rate": 0.0001612295866687919,
+      "loss": 0.5953296422958374,
+      "step": 4936
+    },
+    {
+      "epoch": 2.0835443037974684,
+      "grad_norm": 0.560342013835907,
+      "learning_rate": 0.0001611916063107133,
+      "loss": 0.6765958070755005,
+      "step": 4938
+    },
+    {
+      "epoch": 2.0843881856540083,
+      "grad_norm": 0.4263869822025299,
+      "learning_rate": 0.00016115361183690631,
+      "loss": 0.5475618243217468,
+      "step": 4940
+    },
+    {
+      "epoch": 2.0852320675105487,
+      "grad_norm": 0.41631078720092773,
+      "learning_rate": 0.00016111560325613553,
+      "loss": 0.570278525352478,
+      "step": 4942
+    },
+    {
+      "epoch": 2.0860759493670886,
+      "grad_norm": 0.3922317326068878,
+      "learning_rate": 0.00016107758057716884,
+      "loss": 0.5691710710525513,
+      "step": 4944
+    },
+    {
+      "epoch": 2.0869198312236286,
+      "grad_norm": 0.518889307975769,
+      "learning_rate": 0.00016103954380877736,
+      "loss": 0.6014654040336609,
+      "step": 4946
+    },
+    {
+      "epoch": 2.087763713080169,
+      "grad_norm": 0.560624897480011,
+      "learning_rate": 0.00016100149295973537,
+      "loss": 0.6084938049316406,
+      "step": 4948
+    },
+    {
+      "epoch": 2.088607594936709,
+      "grad_norm": 0.46847012639045715,
+      "learning_rate": 0.00016096342803882053,
+      "loss": 0.6078231334686279,
+      "step": 4950
+    },
+    {
+      "epoch": 2.089451476793249,
+      "grad_norm": 0.43910443782806396,
+      "learning_rate": 0.00016092534905481367,
+      "loss": 0.5664147734642029,
+      "step": 4952
+    },
+    {
+      "epoch": 2.090295358649789,
+      "grad_norm": 0.40413206815719604,
+      "learning_rate": 0.00016088725601649887,
+      "loss": 0.5793747901916504,
+      "step": 4954
+    },
+    {
+      "epoch": 2.091139240506329,
+      "grad_norm": 0.47664782404899597,
+      "learning_rate": 0.00016084914893266347,
+      "loss": 0.6121629476547241,
+      "step": 4956
+    },
+    {
+      "epoch": 2.091983122362869,
+      "grad_norm": 0.4575684070587158,
+      "learning_rate": 0.00016081102781209805,
+      "loss": 0.6086400747299194,
+      "step": 4958
+    },
+    {
+      "epoch": 2.0928270042194095,
+      "grad_norm": 0.4286920726299286,
+      "learning_rate": 0.0001607728926635964,
+      "loss": 0.5971894860267639,
+      "step": 4960
+    },
+    {
+      "epoch": 2.0936708860759494,
+      "grad_norm": 0.5314067006111145,
+      "learning_rate": 0.00016073474349595555,
+      "loss": 0.6195604801177979,
+      "step": 4962
+    },
+    {
+      "epoch": 2.0945147679324894,
+      "grad_norm": 0.5221946239471436,
+      "learning_rate": 0.00016069658031797585,
+      "loss": 0.6141531467437744,
+      "step": 4964
+    },
+    {
+      "epoch": 2.0953586497890297,
+      "grad_norm": 0.46874696016311646,
+      "learning_rate": 0.00016065840313846076,
+      "loss": 0.6354720592498779,
+      "step": 4966
+    },
+    {
+      "epoch": 2.0962025316455697,
+      "grad_norm": 0.44518616795539856,
+      "learning_rate": 0.00016062021196621703,
+      "loss": 0.5759334564208984,
+      "step": 4968
+    },
+    {
+      "epoch": 2.0970464135021096,
+      "grad_norm": 0.44092273712158203,
+      "learning_rate": 0.00016058200681005462,
+      "loss": 0.5387811064720154,
+      "step": 4970
+    },
+    {
+      "epoch": 2.09789029535865,
+      "grad_norm": 0.4952426552772522,
+      "learning_rate": 0.00016054378767878678,
+      "loss": 0.5712288022041321,
+      "step": 4972
+    },
+    {
+      "epoch": 2.09873417721519,
+      "grad_norm": 0.4890255331993103,
+      "learning_rate": 0.00016050555458122984,
+      "loss": 0.6320934891700745,
+      "step": 4974
+    },
+    {
+      "epoch": 2.09957805907173,
+      "grad_norm": 0.5182738304138184,
+      "learning_rate": 0.00016046730752620352,
+      "loss": 0.6141278147697449,
+      "step": 4976
+    },
+    {
+      "epoch": 2.10042194092827,
+      "grad_norm": 0.5344594717025757,
+      "learning_rate": 0.00016042904652253064,
+      "loss": 0.6079437732696533,
+      "step": 4978
+    },
+    {
+      "epoch": 2.1012658227848102,
+      "grad_norm": 0.436641126871109,
+      "learning_rate": 0.0001603907715790373,
+      "loss": 0.5634552836418152,
+      "step": 4980
+    },
+    {
+      "epoch": 2.10210970464135,
+      "grad_norm": 0.44045236706733704,
+      "learning_rate": 0.00016035248270455273,
+      "loss": 0.5707821846008301,
+      "step": 4982
+    },
+    {
+      "epoch": 2.10295358649789,
+      "grad_norm": 0.43434852361679077,
+      "learning_rate": 0.00016031417990790953,
+      "loss": 0.5635452270507812,
+      "step": 4984
+    },
+    {
+      "epoch": 2.1037974683544305,
+      "grad_norm": 0.5538138747215271,
+      "learning_rate": 0.00016027586319794341,
+      "loss": 0.632486879825592,
+      "step": 4986
+    },
+    {
+      "epoch": 2.1046413502109704,
+      "grad_norm": 0.5425865054130554,
+      "learning_rate": 0.00016023753258349323,
+      "loss": 0.6509168744087219,
+      "step": 4988
+    },
+    {
+      "epoch": 2.1054852320675104,
+      "grad_norm": 0.4603687524795532,
+      "learning_rate": 0.0001601991880734012,
+      "loss": 0.615444540977478,
+      "step": 4990
+    },
+    {
+      "epoch": 2.1063291139240508,
+      "grad_norm": 0.4866391718387604,
+      "learning_rate": 0.0001601608296765126,
+      "loss": 0.684465765953064,
+      "step": 4992
+    },
+    {
+      "epoch": 2.1071729957805907,
+      "grad_norm": 0.47539305686950684,
+      "learning_rate": 0.000160122457401676,
+      "loss": 0.5981772541999817,
+      "step": 4994
+    },
+    {
+      "epoch": 2.1080168776371306,
+      "grad_norm": 0.4135447144508362,
+      "learning_rate": 0.00016008407125774315,
+      "loss": 0.6041272878646851,
+      "step": 4996
+    },
+    {
+      "epoch": 2.108860759493671,
+      "grad_norm": 0.5324689745903015,
+      "learning_rate": 0.00016004567125356905,
+      "loss": 0.6491074562072754,
+      "step": 4998
+    },
+    {
+      "epoch": 2.109704641350211,
+      "grad_norm": 0.4004698395729065,
+      "learning_rate": 0.00016000725739801172,
+      "loss": 0.6178454756736755,
+      "step": 5000
+    },
+    {
+      "epoch": 2.109704641350211,
+      "eval_loss": 0.7635271549224854,
+      "eval_runtime": 669.5049,
+      "eval_samples_per_second": 3.147,
+      "eval_steps_per_second": 3.147,
+      "step": 5000
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 14220,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 6,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 5,
+        "early_stopping_threshold": 0.001
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 4
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.9899897855995863e+18,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/sft_qwen_14B_v2/checkpoints/checkpoint-5000/training_args.bin b/sft_qwen_14B_v2/checkpoints/checkpoint-5000/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..0bd5f45757bdc6c7fc2d67718b18675b51983e0e
--- /dev/null
+++ b/sft_qwen_14B_v2/checkpoints/checkpoint-5000/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:afae4a8ce1391f149eb00b14eed8f891c715e892ea546bf754d22db2c2bc5969
+size 4792
diff --git a/sft_qwen_14B_v2/config_resolved.yaml b/sft_qwen_14B_v2/config_resolved.yaml
new file mode 100644
index 0000000000000000000000000000000000000000..7864fd23399b7ddbde0319c26c473806dea1a4b3
--- /dev/null
+++ b/sft_qwen_14B_v2/config_resolved.yaml
@@ -0,0 +1,102 @@
+run:
+  run_dir: ./task2file/sft_qwen_14B_v2
+  seed: 42
+wandb:
+  enabled: true
+  project: sft-training
+  entity: null
+  name: null
+  tags:
+  - sft-lora
+  - instruction-tuning
+  notes: null
+model:
+  repo_id: ./Models/Qwen2.5-Coder-14B-CPT
+  revision: null
+  base_local_dir: base_model
+  trust_remote_code: true
+  tokenizer_use_fast: true
+  device_map: auto
+  torch_dtype: bfloat16
+  use_4bit: false
+  bnb_4bit_quant_type: nf4
+  bnb_4bit_use_double_quant: false
+  bnb_4bit_compute_dtype: bfloat16
+  attn_implementation: null
+data:
+  train_jsonl: sft_dataset.jsonl
+  eval_jsonl: null
+  eval_split_ratio: 0.1
+  instruction_field: instruction
+  input_field: input
+  output_field: output
+  format_type: custom
+  system_prompt: "You are a Hyperswitch Rust code analyzer. Identify functions/structs\
+    \ that need modification for a given task.\n\n## Output Format\n\n##OUTPUT\nExplain\
+    \ the data flow and why each component must change:\n- Flow: [Input \u2192 Processing\
+    \ \u2192 Output with arrows]\n- For each component: \"The [ComponentName] ([path])\
+    \ must [action] because [reason]\u2014without this, [consequence]\"\n- Explain\
+    \ coupling between components\n\n##SELECT\nmodify::crates/path/to/file.rs::impl::ComponentName\n\
+    add::crates/another/file.rs::function::AnotherComponent\n<EOS>\n\n## Rules\n\n\
+    1. Use full paths: `remove::crates/folder/file.rs::Type::Name`\n2. Use `::` for\
+    \ nested items: `status::StructName::Type::Name`\n3. Always explain \"must change\
+    \ because\" and \"without this\"\n3. Types of components: function, struct, enum,\
+    \ impl, trait\n4. If there is extra information (e.g., enum variants), include\
+    \ that too.\n5. Start with ##OUTPUT, end with ##SELECT, terminate with <EOS>\n\
+    \n## Example\n\n##TASK\nAdd webhook subscription support\n\n##OUTPUT\nThe webhook\
+    \ system routes events via EventClass enum. Flow: webhook \u2192 EventClass \u2192\
+    \ handler \u2192 processing. The EventClass enum (crates/common_enums/src/enums.rs::EventClass)\
+    \ must add Subscriptions variant because it defines event routing\u2014without\
+    \ this, subscription events cannot be processed. The SubscriptionStatus impl (crates/common_enums/src/transformers.rs::SubscriptionStatus)\
+    \ must map to EventType because it converts status to events\u2014without this,\
+    \ status changes don't trigger webhooks. These are coupled: EventClass routes\
+    \ to handlers that use SubscriptionStatus mappings.\n\n##SELECT\ncrates/common_enums/src/enums.rs::EventClass\n\
+    crates/common_enums/src/transformers.rs::SubscriptionStatus\n<EOS>\n"
+  custom_template: '##INSTRUCTION
+
+    {instruction}<|im_end|>
+
+    {input}<|im_end|>
+
+    {output}<|im_end|>'
+  max_length: 2048
+  shuffle: true
+  num_proc: 4
+peft:
+  enabled: true
+  r: 32
+  lora_alpha: 64
+  lora_dropout: 0.05
+  bias: none
+  target_modules: auto
+train:
+  num_train_epochs: 6
+  per_device_train_batch_size: 1
+  per_device_eval_batch_size: 1
+  gradient_accumulation_steps: 8
+  learning_rate: 2e-4
+  weight_decay: 0.0
+  warmup_ratio: 0.08
+  lr_scheduler_type: cosine
+  optim: adamw_torch
+  max_grad_norm: 1.0
+  gradient_checkpointing: true
+  logging_steps: 2
+  save_strategy: steps
+  save_steps: 500
+  save_total_limit: 20
+  evaluation_strategy: steps
+  eval_steps: 100
+  load_best_model_at_end: true
+  early_stopping:
+    enabled: true
+    patience: 5
+    min_delta: 0.001
+    metric: eval_loss
+    mode: min
+  resume_from_checkpoint: auto
+merge:
+  enabled: true
+  merged_dtype: float16
+  max_shard_size: 2GB
+  output_dir: ./Models/Qwen2.5-Coder-14B-CPT-SFT_v2
diff --git a/sft_qwen_14B_v2/eval_final.json b/sft_qwen_14B_v2/eval_final.json
new file mode 100644
index 0000000000000000000000000000000000000000..00ac3e1711f59e93b9a4dc42a894fd3d24fc0ef1
--- /dev/null
+++ b/sft_qwen_14B_v2/eval_final.json
@@ -0,0 +1,8 @@
+{
+  "eval_loss": 0.7600579857826233,
+  "eval_runtime": 674.048,
+  "eval_samples_per_second": 3.126,
+  "eval_steps_per_second": 3.126,
+  "epoch": 2.151898734177215,
+  "perplexity": 2.138400213711816
+}
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/logs/eval.jsonl b/sft_qwen_14B_v2/logs/eval.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..17e5041402f1d4a7ca4d5d2612d79c5e3f7afd2b
--- /dev/null
+++ b/sft_qwen_14B_v2/logs/eval.jsonl
@@ -0,0 +1,52 @@
+{"ts": "2025-12-26T18:41:10", "event": "eval", "step": 100, "epoch": 0.04219409282700422, "eval_loss": 1.2979938983917236, "eval_runtime": 682.1979, "eval_samples_per_second": 3.089, "eval_steps_per_second": 3.089, "perplexity": 3.661943064177116}
+{"ts": "2025-12-26T19:05:13", "event": "eval", "step": 200, "epoch": 0.08438818565400844, "eval_loss": 1.142486810684204, "eval_runtime": 668.2356, "eval_samples_per_second": 3.153, "eval_steps_per_second": 3.153, "perplexity": 3.134553722506413}
+{"ts": "2025-12-26T19:29:29", "event": "eval", "step": 300, "epoch": 0.12658227848101267, "eval_loss": 1.0952109098434448, "eval_runtime": 677.0652, "eval_samples_per_second": 3.112, "eval_steps_per_second": 3.112, "perplexity": 2.98981319793367}
+{"ts": "2025-12-26T19:53:55", "event": "eval", "step": 400, "epoch": 0.16877637130801687, "eval_loss": 1.0625108480453491, "eval_runtime": 691.0068, "eval_samples_per_second": 3.049, "eval_steps_per_second": 3.049, "perplexity": 2.893627334202045}
+{"ts": "2025-12-26T20:18:00", "event": "eval", "step": 500, "epoch": 0.2109704641350211, "eval_loss": 1.042096495628357, "eval_runtime": 692.4361, "eval_samples_per_second": 3.043, "eval_steps_per_second": 3.043, "perplexity": 2.8351546774213405}
+{"ts": "2025-12-26T20:42:00", "event": "eval", "step": 600, "epoch": 0.25316455696202533, "eval_loss": 1.0193854570388794, "eval_runtime": 677.9523, "eval_samples_per_second": 3.108, "eval_steps_per_second": 3.108, "perplexity": 2.7714910402016297}
+{"ts": "2025-12-26T21:06:13", "event": "eval", "step": 700, "epoch": 0.29535864978902954, "eval_loss": 0.996929407119751, "eval_runtime": 668.6398, "eval_samples_per_second": 3.151, "eval_steps_per_second": 3.151, "perplexity": 2.7099478932392134}
+{"ts": "2025-12-26T21:30:25", "event": "eval", "step": 800, "epoch": 0.33755274261603374, "eval_loss": 0.9800403714179993, "eval_runtime": 678.8306, "eval_samples_per_second": 3.104, "eval_steps_per_second": 3.104, "perplexity": 2.6645638119774637}
+{"ts": "2025-12-26T21:54:42", "event": "eval", "step": 900, "epoch": 0.379746835443038, "eval_loss": 0.9643027186393738, "eval_runtime": 691.7929, "eval_samples_per_second": 3.046, "eval_steps_per_second": 3.046, "perplexity": 2.6229580789054108}
+{"ts": "2025-12-26T22:18:39", "event": "eval", "step": 1000, "epoch": 0.4219409282700422, "eval_loss": 0.9487298727035522, "eval_runtime": 689.4288, "eval_samples_per_second": 3.056, "eval_steps_per_second": 3.056, "perplexity": 2.5824275636777196}
+{"ts": "2025-12-26T22:42:41", "event": "eval", "step": 1100, "epoch": 0.4641350210970464, "eval_loss": 0.9357889294624329, "eval_runtime": 676.9573, "eval_samples_per_second": 3.112, "eval_steps_per_second": 3.112, "perplexity": 2.549223822396605}
+{"ts": "2025-12-26T23:06:55", "event": "eval", "step": 1200, "epoch": 0.5063291139240507, "eval_loss": 0.9224098324775696, "eval_runtime": 669.7542, "eval_samples_per_second": 3.146, "eval_steps_per_second": 3.146, "perplexity": 2.515344651361619}
+{"ts": "2025-12-26T23:31:25", "event": "eval", "step": 1300, "epoch": 0.5485232067510548, "eval_loss": 0.9068717360496521, "eval_runtime": 680.7718, "eval_samples_per_second": 3.095, "eval_steps_per_second": 3.095, "perplexity": 2.476563059931004}
+{"ts": "2025-12-26T23:55:39", "event": "eval", "step": 1400, "epoch": 0.5907172995780591, "eval_loss": 0.8971880674362183, "eval_runtime": 692.8046, "eval_samples_per_second": 3.041, "eval_steps_per_second": 3.041, "perplexity": 2.452696587964245}
+{"ts": "2025-12-27T00:19:35", "event": "eval", "step": 1500, "epoch": 0.6329113924050633, "eval_loss": 0.887488842010498, "eval_runtime": 686.2804, "eval_samples_per_second": 3.07, "eval_steps_per_second": 3.07, "perplexity": 2.4290223274474503}
+{"ts": "2025-12-27T00:43:47", "event": "eval", "step": 1600, "epoch": 0.6751054852320675, "eval_loss": 0.8769772052764893, "eval_runtime": 677.9338, "eval_samples_per_second": 3.108, "eval_steps_per_second": 3.108, "perplexity": 2.403623054958293}
+{"ts": "2025-12-27T01:08:03", "event": "eval", "step": 1700, "epoch": 0.7172995780590717, "eval_loss": 0.8708170056343079, "eval_runtime": 670.3019, "eval_samples_per_second": 3.143, "eval_steps_per_second": 3.143, "perplexity": 2.388861769986548}
+{"ts": "2025-12-27T01:32:23", "event": "eval", "step": 1800, "epoch": 0.759493670886076, "eval_loss": 0.8625519275665283, "eval_runtime": 686.4271, "eval_samples_per_second": 3.07, "eval_steps_per_second": 3.07, "perplexity": 2.369199010020167}
+{"ts": "2025-12-27T01:56:20", "event": "eval", "step": 1900, "epoch": 0.8016877637130801, "eval_loss": 0.8546335697174072, "eval_runtime": 688.5301, "eval_samples_per_second": 3.06, "eval_steps_per_second": 3.06, "perplexity": 2.3505129236377402}
+{"ts": "2025-12-27T02:20:26", "event": "eval", "step": 2000, "epoch": 0.8438818565400844, "eval_loss": 0.8460908532142639, "eval_runtime": 685.2518, "eval_samples_per_second": 3.075, "eval_steps_per_second": 3.075, "perplexity": 2.330518682256874}
+{"ts": "2025-12-27T02:44:39", "event": "eval", "step": 2100, "epoch": 0.8860759493670886, "eval_loss": 0.8401098847389221, "eval_runtime": 669.1149, "eval_samples_per_second": 3.149, "eval_steps_per_second": 3.149, "perplexity": 2.3166215241467625}
+{"ts": "2025-12-27T03:09:05", "event": "eval", "step": 2200, "epoch": 0.9282700421940928, "eval_loss": 0.8336610198020935, "eval_runtime": 674.5134, "eval_samples_per_second": 3.124, "eval_steps_per_second": 3.124, "perplexity": 2.3017300131082887}
+{"ts": "2025-12-27T03:33:21", "event": "eval", "step": 2300, "epoch": 0.9704641350210971, "eval_loss": 0.8281980156898499, "eval_runtime": 688.6136, "eval_samples_per_second": 3.06, "eval_steps_per_second": 3.06, "perplexity": 2.289189937012629}
+{"ts": "2025-12-27T03:57:18", "event": "eval", "step": 2400, "epoch": 1.0126582278481013, "eval_loss": 0.8250564932823181, "eval_runtime": 691.5833, "eval_samples_per_second": 3.047, "eval_steps_per_second": 3.047, "perplexity": 2.282009679904965}
+{"ts": "2025-12-27T04:21:22", "event": "eval", "step": 2500, "epoch": 1.0548523206751055, "eval_loss": 0.8249453902244568, "eval_runtime": 679.4446, "eval_samples_per_second": 3.101, "eval_steps_per_second": 3.101, "perplexity": 2.2817561557353745}
+{"ts": "2025-12-27T04:45:36", "event": "eval", "step": 2600, "epoch": 1.0970464135021096, "eval_loss": 0.8211485743522644, "eval_runtime": 670.2276, "eval_samples_per_second": 3.144, "eval_steps_per_second": 3.144, "perplexity": 2.2731091736340194}
+{"ts": "2025-12-27T05:10:00", "event": "eval", "step": 2700, "epoch": 1.139240506329114, "eval_loss": 0.8155058026313782, "eval_runtime": 678.284, "eval_samples_per_second": 3.106, "eval_steps_per_second": 3.106, "perplexity": 2.2603186583878263}
+{"ts": "2025-12-27T05:34:19", "event": "eval", "step": 2800, "epoch": 1.1814345991561181, "eval_loss": 0.8124309182167053, "eval_runtime": 688.4759, "eval_samples_per_second": 3.06, "eval_steps_per_second": 3.06, "perplexity": 2.2533791143885313}
+{"ts": "2025-12-27T05:58:11", "event": "eval", "step": 2900, "epoch": 1.2236286919831223, "eval_loss": 0.8077136278152466, "eval_runtime": 685.6042, "eval_samples_per_second": 3.073, "eval_steps_per_second": 3.073, "perplexity": 2.2427743033735634}
+{"ts": "2025-12-27T06:22:11", "event": "eval", "step": 3000, "epoch": 1.2658227848101267, "eval_loss": 0.8033165335655212, "eval_runtime": 678.7554, "eval_samples_per_second": 3.104, "eval_steps_per_second": 3.104, "perplexity": 2.232934263027478}
+{"ts": "2025-12-27T06:46:29", "event": "eval", "step": 3100, "epoch": 1.3080168776371308, "eval_loss": 0.8010181784629822, "eval_runtime": 668.1688, "eval_samples_per_second": 3.153, "eval_steps_per_second": 3.153, "perplexity": 2.2278080803210654}
+{"ts": "2025-12-27T07:10:54", "event": "eval", "step": 3200, "epoch": 1.350210970464135, "eval_loss": 0.797160804271698, "eval_runtime": 680.976, "eval_samples_per_second": 3.094, "eval_steps_per_second": 3.094, "perplexity": 2.2192311437906307}
+{"ts": "2025-12-27T07:35:02", "event": "eval", "step": 3300, "epoch": 1.3924050632911391, "eval_loss": 0.795619547367096, "eval_runtime": 692.7157, "eval_samples_per_second": 3.042, "eval_steps_per_second": 3.042, "perplexity": 2.215813372975358}
+{"ts": "2025-12-27T07:59:01", "event": "eval", "step": 3400, "epoch": 1.4345991561181435, "eval_loss": 0.7917885780334473, "eval_runtime": 686.1689, "eval_samples_per_second": 3.071, "eval_steps_per_second": 3.071, "perplexity": 2.2073408991501657}
+{"ts": "2025-12-27T08:22:55", "event": "eval", "step": 3500, "epoch": 1.4767932489451476, "eval_loss": 0.7902651429176331, "eval_runtime": 672.312, "eval_samples_per_second": 3.134, "eval_steps_per_second": 3.134, "perplexity": 2.203980718670171}
+{"ts": "2025-12-27T08:47:16", "event": "eval", "step": 3600, "epoch": 1.518987341772152, "eval_loss": 0.785450279712677, "eval_runtime": 675.2312, "eval_samples_per_second": 3.12, "eval_steps_per_second": 3.12, "perplexity": 2.1933943593911716}
+{"ts": "2025-12-27T09:11:38", "event": "eval", "step": 3700, "epoch": 1.5611814345991561, "eval_loss": 0.7854447960853577, "eval_runtime": 687.7907, "eval_samples_per_second": 3.063, "eval_steps_per_second": 3.063, "perplexity": 2.193382331666918}
+{"ts": "2025-12-27T09:35:44", "event": "eval", "step": 3800, "epoch": 1.6033755274261603, "eval_loss": 0.778353214263916, "eval_runtime": 692.5522, "eval_samples_per_second": 3.042, "eval_steps_per_second": 3.042, "perplexity": 2.1778828044355443}
+{"ts": "2025-12-27T09:59:49", "event": "eval", "step": 3900, "epoch": 1.6455696202531644, "eval_loss": 0.7763351202011108, "eval_runtime": 682.0824, "eval_samples_per_second": 3.089, "eval_steps_per_second": 3.089, "perplexity": 2.173492064032179}
+{"ts": "2025-12-27T10:23:55", "event": "eval", "step": 4000, "epoch": 1.6877637130801688, "eval_loss": 0.7721371650695801, "eval_runtime": 668.395, "eval_samples_per_second": 3.152, "eval_steps_per_second": 3.152, "perplexity": 2.1643869666352633}
+{"ts": "2025-12-27T10:48:15", "event": "eval", "step": 4100, "epoch": 1.729957805907173, "eval_loss": 0.7690847516059875, "eval_runtime": 673.6323, "eval_samples_per_second": 3.128, "eval_steps_per_second": 3.128, "perplexity": 2.157790435509873}
+{"ts": "2025-12-27T11:12:31", "event": "eval", "step": 4200, "epoch": 1.7721518987341773, "eval_loss": 0.7676366567611694, "eval_runtime": 687.9619, "eval_samples_per_second": 3.063, "eval_steps_per_second": 3.063, "perplexity": 2.1546680116326113}
+{"ts": "2025-12-27T11:36:30", "event": "eval", "step": 4300, "epoch": 1.8143459915611815, "eval_loss": 0.7672964930534363, "eval_runtime": 688.4249, "eval_samples_per_second": 3.061, "eval_steps_per_second": 3.061, "perplexity": 2.1539351964184767}
+{"ts": "2025-12-27T12:00:29", "event": "eval", "step": 4400, "epoch": 1.8565400843881856, "eval_loss": 0.7635221481323242, "eval_runtime": 678.243, "eval_samples_per_second": 3.107, "eval_steps_per_second": 3.107, "perplexity": 2.1458208249008255}
+{"ts": "2025-12-27T12:24:47", "event": "eval", "step": 4500, "epoch": 1.8987341772151898, "eval_loss": 0.7600579857826233, "eval_runtime": 674.2593, "eval_samples_per_second": 3.125, "eval_steps_per_second": 3.125, "perplexity": 2.138400213711816}
+{"ts": "2025-12-27T12:49:10", "event": "eval", "step": 4600, "epoch": 1.9409282700421941, "eval_loss": 0.7585541009902954, "eval_runtime": 679.0866, "eval_samples_per_second": 3.103, "eval_steps_per_second": 3.103, "perplexity": 2.1351867231159773}
+{"ts": "2025-12-27T13:13:19", "event": "eval", "step": 4700, "epoch": 1.9831223628691983, "eval_loss": 0.7582268714904785, "eval_runtime": 690.081, "eval_samples_per_second": 3.053, "eval_steps_per_second": 3.053, "perplexity": 2.134488141337073}
+{"ts": "2025-12-27T13:37:21", "event": "eval", "step": 4800, "epoch": 2.0253164556962027, "eval_loss": 0.7633068561553955, "eval_runtime": 688.8684, "eval_samples_per_second": 3.059, "eval_steps_per_second": 3.059, "perplexity": 2.145358896619808}
+{"ts": "2025-12-27T14:01:30", "event": "eval", "step": 4900, "epoch": 2.067510548523207, "eval_loss": 0.7676681280136108, "eval_runtime": 676.0104, "eval_samples_per_second": 3.117, "eval_steps_per_second": 3.117, "perplexity": 2.1547358228005784}
+{"ts": "2025-12-27T14:25:57", "event": "eval", "step": 5000, "epoch": 2.109704641350211, "eval_loss": 0.7635271549224854, "eval_runtime": 669.5049, "eval_samples_per_second": 3.147, "eval_steps_per_second": 3.147, "perplexity": 2.145831568602315}
+{"ts": "2025-12-27T14:50:12", "event": "eval", "step": 5100, "epoch": 2.151898734177215, "eval_loss": 0.7654595971107483, "eval_runtime": 681.4966, "eval_samples_per_second": 3.092, "eval_steps_per_second": 3.092, "perplexity": 2.149982273261109}
+{"ts": "2025-12-27T15:01:27", "event": "eval", "step": 5100, "epoch": 2.151898734177215, "eval_loss": 0.7600579857826233, "eval_runtime": 674.048, "eval_samples_per_second": 3.126, "eval_steps_per_second": 3.126, "perplexity": 2.138400213711816}
diff --git a/sft_qwen_14B_v2/logs/train.jsonl b/sft_qwen_14B_v2/logs/train.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..3ebf38ac0b63423a8ef6e40f4363f73a76bc492a
--- /dev/null
+++ b/sft_qwen_14B_v2/logs/train.jsonl
@@ -0,0 +1,2603 @@
+{"ts": "2025-12-26T18:17:26", "event": "train_log", "step": 2, "epoch": 0.0008438818565400844, "progress_pct": 0.01, "epoch_pct": 0.01, "eta": "31:09:53", "max_grad_norm": 1.0, "loss": 1.6941628456115723, "grad_norm": 0.5386583805084229, "learning_rate": 1.7574692442882248e-07}
+{"ts": "2025-12-26T18:17:42", "event": "train_log", "step": 4, "epoch": 0.0016877637130801688, "progress_pct": 0.03, "epoch_pct": 0.03, "eta": "31:14:54", "max_grad_norm": 1.0, "loss": 1.7132279872894287, "grad_norm": 0.5477277636528015, "learning_rate": 5.272407732864675e-07}
+{"ts": "2025-12-26T18:17:57", "event": "train_log", "step": 6, "epoch": 0.002531645569620253, "progress_pct": 0.04, "epoch_pct": 0.04, "eta": "31:02:32", "max_grad_norm": 1.0, "loss": 1.641180396080017, "grad_norm": 0.5390765070915222, "learning_rate": 8.787346221441126e-07}
+{"ts": "2025-12-26T18:18:13", "event": "train_log", "step": 8, "epoch": 0.0033755274261603376, "progress_pct": 0.06, "epoch_pct": 0.06, "eta": "31:07:59", "max_grad_norm": 1.0, "loss": 1.5616240501403809, "grad_norm": 0.5023683905601501, "learning_rate": 1.2302284710017575e-06}
+{"ts": "2025-12-26T18:18:31", "event": "train_log", "step": 10, "epoch": 0.004219409282700422, "progress_pct": 0.07, "epoch_pct": 0.07, "eta": "31:44:10", "max_grad_norm": 1.0, "loss": 1.572033405303955, "grad_norm": 0.4899154603481293, "learning_rate": 1.5817223198594026e-06}
+{"ts": "2025-12-26T18:18:47", "event": "train_log", "step": 12, "epoch": 0.005063291139240506, "progress_pct": 0.08, "epoch_pct": 0.08, "eta": "31:44:10", "max_grad_norm": 1.0, "loss": 1.6242921352386475, "grad_norm": 0.5239788293838501, "learning_rate": 1.9332161687170474e-06}
+{"ts": "2025-12-26T18:19:03", "event": "train_log", "step": 14, "epoch": 0.00590717299578059, "progress_pct": 0.1, "epoch_pct": 0.1, "eta": "31:40:24", "max_grad_norm": 1.0, "loss": 1.6800041198730469, "grad_norm": 0.5172926783561707, "learning_rate": 2.2847100175746925e-06}
+{"ts": "2025-12-26T18:19:20", "event": "train_log", "step": 16, "epoch": 0.006751054852320675, "progress_pct": 0.11, "epoch_pct": 0.11, "eta": "31:54:47", "max_grad_norm": 1.0, "loss": 1.6450834274291992, "grad_norm": 0.5539224743843079, "learning_rate": 2.6362038664323376e-06}
+{"ts": "2025-12-26T18:19:35", "event": "train_log", "step": 18, "epoch": 0.007594936708860759, "progress_pct": 0.13, "epoch_pct": 0.13, "eta": "31:50:46", "max_grad_norm": 1.0, "loss": 1.6673263311386108, "grad_norm": 0.5255337953567505, "learning_rate": 2.9876977152899827e-06}
+{"ts": "2025-12-26T18:19:51", "event": "train_log", "step": 20, "epoch": 0.008438818565400843, "progress_pct": 0.14, "epoch_pct": 0.14, "eta": "31:40:51", "max_grad_norm": 1.0, "loss": 1.531802773475647, "grad_norm": 0.5074548721313477, "learning_rate": 3.3391915641476277e-06}
+{"ts": "2025-12-26T18:20:06", "event": "train_log", "step": 22, "epoch": 0.009282700421940928, "progress_pct": 0.15, "epoch_pct": 0.15, "eta": "31:30:04", "max_grad_norm": 1.0, "loss": 1.599354863166809, "grad_norm": 0.4160279333591461, "learning_rate": 3.6906854130052724e-06}
+{"ts": "2025-12-26T18:20:22", "event": "train_log", "step": 24, "epoch": 0.010126582278481013, "progress_pct": 0.17, "epoch_pct": 0.17, "eta": "31:26:56", "max_grad_norm": 1.0, "loss": 1.6700962781906128, "grad_norm": 0.5716474652290344, "learning_rate": 4.0421792618629174e-06}
+{"ts": "2025-12-26T18:20:38", "event": "train_log", "step": 26, "epoch": 0.010970464135021098, "progress_pct": 0.18, "epoch_pct": 0.18, "eta": "31:27:18", "max_grad_norm": 1.0, "loss": 1.66217839717865, "grad_norm": 0.5148899555206299, "learning_rate": 4.3936731107205625e-06}
+{"ts": "2025-12-26T18:20:53", "event": "train_log", "step": 28, "epoch": 0.01181434599156118, "progress_pct": 0.2, "epoch_pct": 0.2, "eta": "31:18:25", "max_grad_norm": 1.0, "loss": 1.6692266464233398, "grad_norm": 0.575722336769104, "learning_rate": 4.7451669595782076e-06}
+{"ts": "2025-12-26T18:21:07", "event": "train_log", "step": 30, "epoch": 0.012658227848101266, "progress_pct": 0.21, "epoch_pct": 0.21, "eta": "31:10:37", "max_grad_norm": 1.0, "loss": 1.5518689155578613, "grad_norm": 0.5345953106880188, "learning_rate": 5.096660808435853e-06}
+{"ts": "2025-12-26T18:21:22", "event": "train_log", "step": 32, "epoch": 0.01350210970464135, "progress_pct": 0.23, "epoch_pct": 0.23, "eta": "31:03:45", "max_grad_norm": 1.0, "loss": 1.5930007696151733, "grad_norm": 0.4462043344974518, "learning_rate": 5.448154657293498e-06}
+{"ts": "2025-12-26T18:21:37", "event": "train_log", "step": 34, "epoch": 0.014345991561181435, "progress_pct": 0.24, "epoch_pct": 0.24, "eta": "30:55:24", "max_grad_norm": 1.0, "loss": 1.6069684028625488, "grad_norm": 0.5119605660438538, "learning_rate": 5.799648506151143e-06}
+{"ts": "2025-12-26T18:21:52", "event": "train_log", "step": 36, "epoch": 0.015189873417721518, "progress_pct": 0.25, "epoch_pct": 0.25, "eta": "30:50:57", "max_grad_norm": 1.0, "loss": 1.5838109254837036, "grad_norm": 0.5328608751296997, "learning_rate": 6.151142355008788e-06}
+{"ts": "2025-12-26T18:22:07", "event": "train_log", "step": 38, "epoch": 0.016033755274261603, "progress_pct": 0.27, "epoch_pct": 0.27, "eta": "30:48:40", "max_grad_norm": 1.0, "loss": 1.608130931854248, "grad_norm": 0.5065920352935791, "learning_rate": 6.502636203866433e-06}
+{"ts": "2025-12-26T18:22:23", "event": "train_log", "step": 40, "epoch": 0.016877637130801686, "progress_pct": 0.28, "epoch_pct": 0.28, "eta": "30:48:19", "max_grad_norm": 1.0, "loss": 1.5942182540893555, "grad_norm": 0.4479359984397888, "learning_rate": 6.854130052724078e-06}
+{"ts": "2025-12-26T18:22:37", "event": "train_log", "step": 42, "epoch": 0.017721518987341773, "progress_pct": 0.3, "epoch_pct": 0.3, "eta": "30:41:16", "max_grad_norm": 1.0, "loss": 1.6441553831100464, "grad_norm": 0.42844903469085693, "learning_rate": 7.205623901581722e-06}
+{"ts": "2025-12-26T18:22:52", "event": "train_log", "step": 44, "epoch": 0.018565400843881856, "progress_pct": 0.31, "epoch_pct": 0.31, "eta": "30:36:38", "max_grad_norm": 1.0, "loss": 1.6068111658096313, "grad_norm": 0.476630836725235, "learning_rate": 7.557117750439367e-06}
+{"ts": "2025-12-26T18:23:07", "event": "train_log", "step": 46, "epoch": 0.019409282700421943, "progress_pct": 0.32, "epoch_pct": 0.32, "eta": "30:29:59", "max_grad_norm": 1.0, "loss": 1.6618021726608276, "grad_norm": 0.4532654881477356, "learning_rate": 7.908611599297012e-06}
+{"ts": "2025-12-26T18:23:22", "event": "train_log", "step": 48, "epoch": 0.020253164556962026, "progress_pct": 0.34, "epoch_pct": 0.34, "eta": "30:27:44", "max_grad_norm": 1.0, "loss": 1.4730033874511719, "grad_norm": 0.3701118230819702, "learning_rate": 8.260105448154657e-06}
+{"ts": "2025-12-26T18:23:37", "event": "train_log", "step": 50, "epoch": 0.02109704641350211, "progress_pct": 0.35, "epoch_pct": 0.35, "eta": "30:26:49", "max_grad_norm": 1.0, "loss": 1.4828267097473145, "grad_norm": 0.38471561670303345, "learning_rate": 8.611599297012302e-06}
+{"ts": "2025-12-26T18:23:52", "event": "train_log", "step": 52, "epoch": 0.021940928270042195, "progress_pct": 0.37, "epoch_pct": 0.37, "eta": "30:23:44", "max_grad_norm": 1.0, "loss": 1.3877452611923218, "grad_norm": 0.3602336347103119, "learning_rate": 8.963093145869948e-06}
+{"ts": "2025-12-26T18:24:06", "event": "train_log", "step": 54, "epoch": 0.02278481012658228, "progress_pct": 0.38, "epoch_pct": 0.38, "eta": "30:16:46", "max_grad_norm": 1.0, "loss": 1.49052894115448, "grad_norm": 0.40318572521209717, "learning_rate": 9.314586994727593e-06}
+{"ts": "2025-12-26T18:24:21", "event": "train_log", "step": 56, "epoch": 0.02362869198312236, "progress_pct": 0.39, "epoch_pct": 0.39, "eta": "30:15:39", "max_grad_norm": 1.0, "loss": 1.4912524223327637, "grad_norm": 0.3223826587200165, "learning_rate": 9.666080843585238e-06}
+{"ts": "2025-12-26T18:24:35", "event": "train_log", "step": 58, "epoch": 0.024472573839662448, "progress_pct": 0.41, "epoch_pct": 0.41, "eta": "30:12:10", "max_grad_norm": 1.0, "loss": 1.526674509048462, "grad_norm": 0.3873065114021301, "learning_rate": 1.0017574692442883e-05}
+{"ts": "2025-12-26T18:24:50", "event": "train_log", "step": 60, "epoch": 0.02531645569620253, "progress_pct": 0.42, "epoch_pct": 0.42, "eta": "30:08:52", "max_grad_norm": 1.0, "loss": 1.4480271339416504, "grad_norm": 0.410159707069397, "learning_rate": 1.0369068541300528e-05}
+{"ts": "2025-12-26T18:25:05", "event": "train_log", "step": 62, "epoch": 0.026160337552742614, "progress_pct": 0.44, "epoch_pct": 0.44, "eta": "30:07:18", "max_grad_norm": 1.0, "loss": 1.4222990274429321, "grad_norm": 0.3632003962993622, "learning_rate": 1.0720562390158173e-05}
+{"ts": "2025-12-26T18:25:20", "event": "train_log", "step": 64, "epoch": 0.0270042194092827, "progress_pct": 0.45, "epoch_pct": 0.45, "eta": "30:05:16", "max_grad_norm": 1.0, "loss": 1.387171745300293, "grad_norm": 0.33118435740470886, "learning_rate": 1.1072056239015818e-05}
+{"ts": "2025-12-26T18:25:34", "event": "train_log", "step": 66, "epoch": 0.027848101265822784, "progress_pct": 0.46, "epoch_pct": 0.46, "eta": "30:02:30", "max_grad_norm": 1.0, "loss": 1.3523777723312378, "grad_norm": 0.3301764726638794, "learning_rate": 1.1423550087873463e-05}
+{"ts": "2025-12-26T18:25:50", "event": "train_log", "step": 68, "epoch": 0.02869198312236287, "progress_pct": 0.48, "epoch_pct": 0.48, "eta": "30:02:26", "max_grad_norm": 1.0, "loss": 1.4515162706375122, "grad_norm": 0.34342435002326965, "learning_rate": 1.1775043936731108e-05}
+{"ts": "2025-12-26T18:26:05", "event": "train_log", "step": 70, "epoch": 0.029535864978902954, "progress_pct": 0.49, "epoch_pct": 0.49, "eta": "30:02:54", "max_grad_norm": 1.0, "loss": 1.3509243726730347, "grad_norm": 0.3243122100830078, "learning_rate": 1.2126537785588753e-05}
+{"ts": "2025-12-26T18:26:19", "event": "train_log", "step": 72, "epoch": 0.030379746835443037, "progress_pct": 0.51, "epoch_pct": 0.51, "eta": "29:57:27", "max_grad_norm": 1.0, "loss": 1.4936245679855347, "grad_norm": 0.3450150787830353, "learning_rate": 1.2478031634446398e-05}
+{"ts": "2025-12-26T18:26:34", "event": "train_log", "step": 74, "epoch": 0.031223628691983123, "progress_pct": 0.52, "epoch_pct": 0.52, "eta": "29:57:49", "max_grad_norm": 1.0, "loss": 1.3419109582901, "grad_norm": 0.38912028074264526, "learning_rate": 1.2829525483304042e-05}
+{"ts": "2025-12-26T18:26:49", "event": "train_log", "step": 76, "epoch": 0.032067510548523206, "progress_pct": 0.53, "epoch_pct": 0.53, "eta": "29:56:09", "max_grad_norm": 1.0, "loss": 1.4284154176712036, "grad_norm": 0.3019310235977173, "learning_rate": 1.3181019332161687e-05}
+{"ts": "2025-12-26T18:27:04", "event": "train_log", "step": 78, "epoch": 0.03291139240506329, "progress_pct": 0.55, "epoch_pct": 0.55, "eta": "29:53:57", "max_grad_norm": 1.0, "loss": 1.4256561994552612, "grad_norm": 0.37803682684898376, "learning_rate": 1.3532513181019332e-05}
+{"ts": "2025-12-26T18:27:19", "event": "train_log", "step": 80, "epoch": 0.03375527426160337, "progress_pct": 0.56, "epoch_pct": 0.56, "eta": "29:52:20", "max_grad_norm": 1.0, "loss": 1.3256909847259521, "grad_norm": 0.34191736578941345, "learning_rate": 1.3884007029876977e-05}
+{"ts": "2025-12-26T18:27:34", "event": "train_log", "step": 82, "epoch": 0.03459915611814346, "progress_pct": 0.58, "epoch_pct": 0.58, "eta": "29:52:58", "max_grad_norm": 1.0, "loss": 1.2710685729980469, "grad_norm": 0.35242700576782227, "learning_rate": 1.4235500878734624e-05}
+{"ts": "2025-12-26T18:27:50", "event": "train_log", "step": 84, "epoch": 0.035443037974683546, "progress_pct": 0.59, "epoch_pct": 0.59, "eta": "29:53:41", "max_grad_norm": 1.0, "loss": 1.253411889076233, "grad_norm": 0.38094228506088257, "learning_rate": 1.4586994727592269e-05}
+{"ts": "2025-12-26T18:28:05", "event": "train_log", "step": 86, "epoch": 0.036286919831223625, "progress_pct": 0.6, "epoch_pct": 0.6, "eta": "29:54:14", "max_grad_norm": 1.0, "loss": 1.3064342737197876, "grad_norm": 0.36837366223335266, "learning_rate": 1.4938488576449914e-05}
+{"ts": "2025-12-26T18:28:20", "event": "train_log", "step": 88, "epoch": 0.03713080168776371, "progress_pct": 0.62, "epoch_pct": 0.62, "eta": "29:52:02", "max_grad_norm": 1.0, "loss": 1.293562412261963, "grad_norm": 0.3443569242954254, "learning_rate": 1.5289982425307557e-05}
+{"ts": "2025-12-26T18:28:35", "event": "train_log", "step": 90, "epoch": 0.0379746835443038, "progress_pct": 0.63, "epoch_pct": 0.63, "eta": "29:50:57", "max_grad_norm": 1.0, "loss": 1.3382648229599, "grad_norm": 0.3799338936805725, "learning_rate": 1.5641476274165202e-05}
+{"ts": "2025-12-26T18:28:49", "event": "train_log", "step": 92, "epoch": 0.038818565400843885, "progress_pct": 0.65, "epoch_pct": 0.65, "eta": "29:49:38", "max_grad_norm": 1.0, "loss": 1.3925724029541016, "grad_norm": 0.40501922369003296, "learning_rate": 1.599297012302285e-05}
+{"ts": "2025-12-26T18:29:04", "event": "train_log", "step": 94, "epoch": 0.039662447257383965, "progress_pct": 0.66, "epoch_pct": 0.66, "eta": "29:46:49", "max_grad_norm": 1.0, "loss": 1.357171893119812, "grad_norm": 0.4419630467891693, "learning_rate": 1.6344463971880492e-05}
+{"ts": "2025-12-26T18:29:18", "event": "train_log", "step": 96, "epoch": 0.04050632911392405, "progress_pct": 0.68, "epoch_pct": 0.68, "eta": "29:45:24", "max_grad_norm": 1.0, "loss": 1.3029985427856445, "grad_norm": 0.3619817793369293, "learning_rate": 1.6695957820738137e-05}
+{"ts": "2025-12-26T18:29:32", "event": "train_log", "step": 98, "epoch": 0.04135021097046414, "progress_pct": 0.69, "epoch_pct": 0.69, "eta": "29:42:16", "max_grad_norm": 1.0, "loss": 1.3498191833496094, "grad_norm": 0.4851357340812683, "learning_rate": 1.7047451669595782e-05}
+{"ts": "2025-12-26T18:29:48", "event": "train_log", "step": 100, "epoch": 0.04219409282700422, "progress_pct": 0.7, "epoch_pct": 0.7, "eta": "29:42:35", "max_grad_norm": 1.0, "loss": 1.185287356376648, "grad_norm": 0.418658584356308, "learning_rate": 1.7398945518453427e-05}
+{"ts": "2025-12-26T18:41:10", "event": "train_log", "step": 100, "epoch": 0.04219409282700422, "progress_pct": 0.7, "epoch_pct": 0.7, "eta": "56:28:01", "max_grad_norm": 1.0, "eval_loss": 1.2979938983917236, "eval_runtime": 682.1979, "eval_samples_per_second": 3.089, "eval_steps_per_second": 3.089}
+{"ts": "2025-12-26T18:41:25", "event": "train_log", "step": 102, "epoch": 0.043037974683544304, "progress_pct": 0.72, "epoch_pct": 0.72, "eta": "55:56:35", "max_grad_norm": 1.0, "loss": 1.2217272520065308, "grad_norm": 0.4464418888092041, "learning_rate": 1.7750439367311073e-05}
+{"ts": "2025-12-26T18:41:41", "event": "train_log", "step": 104, "epoch": 0.04388185654008439, "progress_pct": 0.73, "epoch_pct": 0.73, "eta": "55:27:29", "max_grad_norm": 1.0, "loss": 1.2052050828933716, "grad_norm": 0.4706237316131592, "learning_rate": 1.8101933216168718e-05}
+{"ts": "2025-12-26T18:41:55", "event": "train_log", "step": 106, "epoch": 0.04472573839662447, "progress_pct": 0.75, "epoch_pct": 0.75, "eta": "54:55:49", "max_grad_norm": 1.0, "loss": 1.221343994140625, "grad_norm": 0.46394404768943787, "learning_rate": 1.8453427065026363e-05}
+{"ts": "2025-12-26T18:42:11", "event": "train_log", "step": 108, "epoch": 0.04556962025316456, "progress_pct": 0.76, "epoch_pct": 0.76, "eta": "54:27:45", "max_grad_norm": 1.0, "loss": 1.2387475967407227, "grad_norm": 0.4726889431476593, "learning_rate": 1.8804920913884008e-05}
+{"ts": "2025-12-26T18:42:26", "event": "train_log", "step": 110, "epoch": 0.046413502109704644, "progress_pct": 0.77, "epoch_pct": 0.77, "eta": "54:01:37", "max_grad_norm": 1.0, "loss": 1.2851309776306152, "grad_norm": 0.42130985856056213, "learning_rate": 1.9156414762741653e-05}
+{"ts": "2025-12-26T18:42:43", "event": "train_log", "step": 112, "epoch": 0.04725738396624472, "progress_pct": 0.79, "epoch_pct": 0.79, "eta": "53:37:09", "max_grad_norm": 1.0, "loss": 1.2753145694732666, "grad_norm": 0.4504576623439789, "learning_rate": 1.9507908611599298e-05}
+{"ts": "2025-12-26T18:42:59", "event": "train_log", "step": 114, "epoch": 0.04810126582278481, "progress_pct": 0.8, "epoch_pct": 0.8, "eta": "53:13:19", "max_grad_norm": 1.0, "loss": 1.2427717447280884, "grad_norm": 0.396085262298584, "learning_rate": 1.9859402460456943e-05}
+{"ts": "2025-12-26T18:43:14", "event": "train_log", "step": 116, "epoch": 0.048945147679324896, "progress_pct": 0.82, "epoch_pct": 0.82, "eta": "52:49:08", "max_grad_norm": 1.0, "loss": 1.2943825721740723, "grad_norm": 0.5106491446495056, "learning_rate": 2.0210896309314588e-05}
+{"ts": "2025-12-26T18:43:29", "event": "train_log", "step": 118, "epoch": 0.049789029535864976, "progress_pct": 0.83, "epoch_pct": 0.83, "eta": "52:25:12", "max_grad_norm": 1.0, "loss": 1.263301134109497, "grad_norm": 0.42351317405700684, "learning_rate": 2.0562390158172233e-05}
+{"ts": "2025-12-26T18:43:44", "event": "train_log", "step": 120, "epoch": 0.05063291139240506, "progress_pct": 0.84, "epoch_pct": 0.84, "eta": "52:01:59", "max_grad_norm": 1.0, "loss": 1.2647849321365356, "grad_norm": 0.4403539299964905, "learning_rate": 2.0913884007029878e-05}
+{"ts": "2025-12-26T18:43:59", "event": "train_log", "step": 122, "epoch": 0.05147679324894515, "progress_pct": 0.86, "epoch_pct": 0.86, "eta": "51:38:39", "max_grad_norm": 1.0, "loss": 1.2351393699645996, "grad_norm": 0.5260752439498901, "learning_rate": 2.1265377855887523e-05}
+{"ts": "2025-12-26T18:44:16", "event": "train_log", "step": 124, "epoch": 0.05232067510548523, "progress_pct": 0.87, "epoch_pct": 0.87, "eta": "51:20:15", "max_grad_norm": 1.0, "loss": 1.0384471416473389, "grad_norm": 0.44978851079940796, "learning_rate": 2.1616871704745168e-05}
+{"ts": "2025-12-26T18:44:33", "event": "train_log", "step": 126, "epoch": 0.053164556962025315, "progress_pct": 0.89, "epoch_pct": 0.89, "eta": "51:02:43", "max_grad_norm": 1.0, "loss": 1.1518068313598633, "grad_norm": 0.47732362151145935, "learning_rate": 2.1968365553602813e-05}
+{"ts": "2025-12-26T18:44:48", "event": "train_log", "step": 128, "epoch": 0.0540084388185654, "progress_pct": 0.9, "epoch_pct": 0.9, "eta": "50:41:43", "max_grad_norm": 1.0, "loss": 1.2264912128448486, "grad_norm": 0.5473551750183105, "learning_rate": 2.231985940246046e-05}
+{"ts": "2025-12-26T18:45:03", "event": "train_log", "step": 130, "epoch": 0.05485232067510549, "progress_pct": 0.91, "epoch_pct": 0.91, "eta": "50:21:17", "max_grad_norm": 1.0, "loss": 1.1615246534347534, "grad_norm": 0.4473855197429657, "learning_rate": 2.2671353251318103e-05}
+{"ts": "2025-12-26T18:45:19", "event": "train_log", "step": 132, "epoch": 0.05569620253164557, "progress_pct": 0.93, "epoch_pct": 0.93, "eta": "50:03:15", "max_grad_norm": 1.0, "loss": 1.1334880590438843, "grad_norm": 0.5980377197265625, "learning_rate": 2.302284710017575e-05}
+{"ts": "2025-12-26T18:45:35", "event": "train_log", "step": 134, "epoch": 0.056540084388185655, "progress_pct": 0.94, "epoch_pct": 0.94, "eta": "49:46:01", "max_grad_norm": 1.0, "loss": 1.1546804904937744, "grad_norm": 0.5987792015075684, "learning_rate": 2.3374340949033394e-05}
+{"ts": "2025-12-26T18:45:50", "event": "train_log", "step": 136, "epoch": 0.05738396624472574, "progress_pct": 0.96, "epoch_pct": 0.96, "eta": "49:29:14", "max_grad_norm": 1.0, "loss": 1.194953441619873, "grad_norm": 0.45355498790740967, "learning_rate": 2.372583479789104e-05}
+{"ts": "2025-12-26T18:46:07", "event": "train_log", "step": 138, "epoch": 0.05822784810126582, "progress_pct": 0.97, "epoch_pct": 0.97, "eta": "49:13:04", "max_grad_norm": 1.0, "loss": 1.1067466735839844, "grad_norm": 0.5373698472976685, "learning_rate": 2.4077328646748684e-05}
+{"ts": "2025-12-26T18:46:22", "event": "train_log", "step": 140, "epoch": 0.05907172995780591, "progress_pct": 0.98, "epoch_pct": 0.98, "eta": "48:55:54", "max_grad_norm": 1.0, "loss": 1.188468098640442, "grad_norm": 0.48734328150749207, "learning_rate": 2.442882249560633e-05}
+{"ts": "2025-12-26T18:46:38", "event": "train_log", "step": 142, "epoch": 0.059915611814345994, "progress_pct": 1.0, "epoch_pct": 1.0, "eta": "48:40:36", "max_grad_norm": 1.0, "loss": 1.1624362468719482, "grad_norm": 0.4692173898220062, "learning_rate": 2.478031634446397e-05}
+{"ts": "2025-12-26T18:46:54", "event": "train_log", "step": 144, "epoch": 0.060759493670886074, "progress_pct": 1.01, "epoch_pct": 1.01, "eta": "48:26:30", "max_grad_norm": 1.0, "loss": 1.0978907346725464, "grad_norm": 0.532554030418396, "learning_rate": 2.513181019332162e-05}
+{"ts": "2025-12-26T18:47:09", "event": "train_log", "step": 146, "epoch": 0.06160337552742616, "progress_pct": 1.03, "epoch_pct": 1.03, "eta": "48:09:34", "max_grad_norm": 1.0, "loss": 1.2030781507492065, "grad_norm": 0.5853802561759949, "learning_rate": 2.5483304042179264e-05}
+{"ts": "2025-12-26T18:47:25", "event": "train_log", "step": 148, "epoch": 0.06244725738396625, "progress_pct": 1.04, "epoch_pct": 1.04, "eta": "47:55:44", "max_grad_norm": 1.0, "loss": 1.082366943359375, "grad_norm": 0.5061611533164978, "learning_rate": 2.583479789103691e-05}
+{"ts": "2025-12-26T18:47:41", "event": "train_log", "step": 150, "epoch": 0.06329113924050633, "progress_pct": 1.05, "epoch_pct": 1.05, "eta": "47:42:57", "max_grad_norm": 1.0, "loss": 1.10564386844635, "grad_norm": 0.49426141381263733, "learning_rate": 2.6186291739894554e-05}
+{"ts": "2025-12-26T18:47:57", "event": "train_log", "step": 152, "epoch": 0.06413502109704641, "progress_pct": 1.07, "epoch_pct": 1.07, "eta": "47:28:55", "max_grad_norm": 1.0, "loss": 1.1992807388305664, "grad_norm": 0.5846618413925171, "learning_rate": 2.6537785588752196e-05}
+{"ts": "2025-12-26T18:48:12", "event": "train_log", "step": 154, "epoch": 0.06497890295358649, "progress_pct": 1.08, "epoch_pct": 1.08, "eta": "47:14:42", "max_grad_norm": 1.0, "loss": 1.1757566928863525, "grad_norm": 0.5517552495002747, "learning_rate": 2.6889279437609844e-05}
+{"ts": "2025-12-26T18:48:28", "event": "train_log", "step": 156, "epoch": 0.06582278481012659, "progress_pct": 1.1, "epoch_pct": 1.1, "eta": "47:01:26", "max_grad_norm": 1.0, "loss": 1.0548783540725708, "grad_norm": 0.5667305588722229, "learning_rate": 2.7240773286467486e-05}
+{"ts": "2025-12-26T18:48:44", "event": "train_log", "step": 158, "epoch": 0.06666666666666667, "progress_pct": 1.11, "epoch_pct": 1.11, "eta": "46:49:16", "max_grad_norm": 1.0, "loss": 1.184364914894104, "grad_norm": 0.6760414242744446, "learning_rate": 2.7592267135325134e-05}
+{"ts": "2025-12-26T18:48:59", "event": "train_log", "step": 160, "epoch": 0.06751054852320675, "progress_pct": 1.13, "epoch_pct": 1.13, "eta": "46:36:11", "max_grad_norm": 1.0, "loss": 1.1945042610168457, "grad_norm": 0.5261430740356445, "learning_rate": 2.7943760984182776e-05}
+{"ts": "2025-12-26T18:49:14", "event": "train_log", "step": 162, "epoch": 0.06835443037974684, "progress_pct": 1.14, "epoch_pct": 1.14, "eta": "46:22:52", "max_grad_norm": 1.0, "loss": 1.2021973133087158, "grad_norm": 0.6155015230178833, "learning_rate": 2.8295254833040425e-05}
+{"ts": "2025-12-26T18:49:30", "event": "train_log", "step": 164, "epoch": 0.06919831223628692, "progress_pct": 1.15, "epoch_pct": 1.15, "eta": "46:10:34", "max_grad_norm": 1.0, "loss": 1.144123911857605, "grad_norm": 0.6131619215011597, "learning_rate": 2.8646748681898066e-05}
+{"ts": "2025-12-26T18:49:46", "event": "train_log", "step": 166, "epoch": 0.070042194092827, "progress_pct": 1.17, "epoch_pct": 1.17, "eta": "46:00:05", "max_grad_norm": 1.0, "loss": 1.1329256296157837, "grad_norm": 0.5749185681343079, "learning_rate": 2.8998242530755715e-05}
+{"ts": "2025-12-26T18:50:03", "event": "train_log", "step": 168, "epoch": 0.07088607594936709, "progress_pct": 1.18, "epoch_pct": 1.18, "eta": "45:49:46", "max_grad_norm": 1.0, "loss": 1.0892387628555298, "grad_norm": 0.5243118405342102, "learning_rate": 2.9349736379613356e-05}
+{"ts": "2025-12-26T18:50:17", "event": "train_log", "step": 170, "epoch": 0.07172995780590717, "progress_pct": 1.2, "epoch_pct": 1.2, "eta": "45:37:02", "max_grad_norm": 1.0, "loss": 1.163260817527771, "grad_norm": 0.7190104722976685, "learning_rate": 2.9701230228471005e-05}
+{"ts": "2025-12-26T18:50:33", "event": "train_log", "step": 172, "epoch": 0.07257383966244725, "progress_pct": 1.21, "epoch_pct": 1.21, "eta": "45:25:46", "max_grad_norm": 1.0, "loss": 1.0880777835845947, "grad_norm": 0.5486982464790344, "learning_rate": 3.0052724077328647e-05}
+{"ts": "2025-12-26T18:50:49", "event": "train_log", "step": 174, "epoch": 0.07341772151898734, "progress_pct": 1.22, "epoch_pct": 1.22, "eta": "45:16:37", "max_grad_norm": 1.0, "loss": 1.0433368682861328, "grad_norm": 0.5020889043807983, "learning_rate": 3.0404217926186295e-05}
+{"ts": "2025-12-26T18:51:05", "event": "train_log", "step": 176, "epoch": 0.07426160337552742, "progress_pct": 1.24, "epoch_pct": 1.24, "eta": "45:05:47", "max_grad_norm": 1.0, "loss": 1.0528991222381592, "grad_norm": 0.47329774498939514, "learning_rate": 3.075571177504394e-05}
+{"ts": "2025-12-26T18:51:20", "event": "train_log", "step": 178, "epoch": 0.0751054852320675, "progress_pct": 1.25, "epoch_pct": 1.25, "eta": "44:55:26", "max_grad_norm": 1.0, "loss": 1.1627811193466187, "grad_norm": 0.6635547876358032, "learning_rate": 3.110720562390158e-05}
+{"ts": "2025-12-26T18:51:36", "event": "train_log", "step": 180, "epoch": 0.0759493670886076, "progress_pct": 1.27, "epoch_pct": 1.27, "eta": "44:45:10", "max_grad_norm": 1.0, "loss": 1.084869384765625, "grad_norm": 0.5624618530273438, "learning_rate": 3.145869947275923e-05}
+{"ts": "2025-12-26T18:51:51", "event": "train_log", "step": 182, "epoch": 0.07679324894514768, "progress_pct": 1.28, "epoch_pct": 1.28, "eta": "44:34:28", "max_grad_norm": 1.0, "loss": 1.2227671146392822, "grad_norm": 0.6029536724090576, "learning_rate": 3.181019332161687e-05}
+{"ts": "2025-12-26T18:52:05", "event": "train_log", "step": 184, "epoch": 0.07763713080168777, "progress_pct": 1.29, "epoch_pct": 1.29, "eta": "44:23:43", "max_grad_norm": 1.0, "loss": 1.0955452919006348, "grad_norm": 0.930959939956665, "learning_rate": 3.216168717047452e-05}
+{"ts": "2025-12-26T18:52:21", "event": "train_log", "step": 186, "epoch": 0.07848101265822785, "progress_pct": 1.31, "epoch_pct": 1.31, "eta": "44:14:24", "max_grad_norm": 1.0, "loss": 1.0640798807144165, "grad_norm": 0.5326952338218689, "learning_rate": 3.251318101933216e-05}
+{"ts": "2025-12-26T18:52:36", "event": "train_log", "step": 188, "epoch": 0.07932489451476793, "progress_pct": 1.32, "epoch_pct": 1.32, "eta": "44:04:53", "max_grad_norm": 1.0, "loss": 1.0700589418411255, "grad_norm": 0.5484727621078491, "learning_rate": 3.286467486818981e-05}
+{"ts": "2025-12-26T18:52:51", "event": "train_log", "step": 190, "epoch": 0.08016877637130802, "progress_pct": 1.34, "epoch_pct": 1.34, "eta": "43:54:36", "max_grad_norm": 1.0, "loss": 1.1593081951141357, "grad_norm": 0.605273425579071, "learning_rate": 3.3216168717047456e-05}
+{"ts": "2025-12-26T18:53:05", "event": "train_log", "step": 192, "epoch": 0.0810126582278481, "progress_pct": 1.35, "epoch_pct": 1.35, "eta": "43:44:25", "max_grad_norm": 1.0, "loss": 1.1617076396942139, "grad_norm": 0.5704394578933716, "learning_rate": 3.35676625659051e-05}
+{"ts": "2025-12-26T18:53:21", "event": "train_log", "step": 194, "epoch": 0.08185654008438818, "progress_pct": 1.36, "epoch_pct": 1.36, "eta": "43:35:18", "max_grad_norm": 1.0, "loss": 1.1346839666366577, "grad_norm": 0.5929452180862427, "learning_rate": 3.3919156414762746e-05}
+{"ts": "2025-12-26T18:53:35", "event": "train_log", "step": 196, "epoch": 0.08270042194092828, "progress_pct": 1.38, "epoch_pct": 1.38, "eta": "43:25:58", "max_grad_norm": 1.0, "loss": 1.0934710502624512, "grad_norm": 0.5624077916145325, "learning_rate": 3.427065026362039e-05}
+{"ts": "2025-12-26T18:53:50", "event": "train_log", "step": 198, "epoch": 0.08354430379746836, "progress_pct": 1.39, "epoch_pct": 1.39, "eta": "43:16:01", "max_grad_norm": 1.0, "loss": 1.1810534000396729, "grad_norm": 0.6717425584793091, "learning_rate": 3.4622144112478036e-05}
+{"ts": "2025-12-26T18:54:05", "event": "train_log", "step": 200, "epoch": 0.08438818565400844, "progress_pct": 1.41, "epoch_pct": 1.41, "eta": "43:07:42", "max_grad_norm": 1.0, "loss": 1.1525514125823975, "grad_norm": 0.5120199918746948, "learning_rate": 3.4973637961335674e-05}
+{"ts": "2025-12-26T19:05:13", "event": "train_log", "step": 200, "epoch": 0.08438818565400844, "progress_pct": 1.41, "epoch_pct": 1.41, "eta": "56:08:25", "max_grad_norm": 1.0, "eval_loss": 1.142486810684204, "eval_runtime": 668.2356, "eval_samples_per_second": 3.153, "eval_steps_per_second": 3.153}
+{"ts": "2025-12-26T19:05:29", "event": "train_log", "step": 202, "epoch": 0.08523206751054853, "progress_pct": 1.42, "epoch_pct": 1.42, "eta": "55:52:50", "max_grad_norm": 1.0, "loss": 1.0243735313415527, "grad_norm": 0.5144487023353577, "learning_rate": 3.5325131810193326e-05}
+{"ts": "2025-12-26T19:05:44", "event": "train_log", "step": 204, "epoch": 0.08607594936708861, "progress_pct": 1.43, "epoch_pct": 1.43, "eta": "55:36:55", "max_grad_norm": 1.0, "loss": 1.118743896484375, "grad_norm": 0.6325069069862366, "learning_rate": 3.5676625659050964e-05}
+{"ts": "2025-12-26T19:06:00", "event": "train_log", "step": 206, "epoch": 0.08691983122362869, "progress_pct": 1.45, "epoch_pct": 1.45, "eta": "55:21:52", "max_grad_norm": 1.0, "loss": 1.0380504131317139, "grad_norm": 0.5501633882522583, "learning_rate": 3.6028119507908616e-05}
+{"ts": "2025-12-26T19:06:16", "event": "train_log", "step": 208, "epoch": 0.08776371308016878, "progress_pct": 1.46, "epoch_pct": 1.46, "eta": "55:07:35", "max_grad_norm": 1.0, "loss": 0.9837555885314941, "grad_norm": 0.6133899688720703, "learning_rate": 3.6379613356766254e-05}
+{"ts": "2025-12-26T19:06:31", "event": "train_log", "step": 210, "epoch": 0.08860759493670886, "progress_pct": 1.48, "epoch_pct": 1.48, "eta": "54:52:41", "max_grad_norm": 1.0, "loss": 1.090720295906067, "grad_norm": 0.5799810886383057, "learning_rate": 3.6731107205623906e-05}
+{"ts": "2025-12-26T19:06:47", "event": "train_log", "step": 212, "epoch": 0.08945147679324894, "progress_pct": 1.49, "epoch_pct": 1.49, "eta": "54:37:49", "max_grad_norm": 1.0, "loss": 1.120232343673706, "grad_norm": 0.6039511561393738, "learning_rate": 3.7082601054481544e-05}
+{"ts": "2025-12-26T19:07:01", "event": "train_log", "step": 214, "epoch": 0.09029535864978903, "progress_pct": 1.5, "epoch_pct": 1.5, "eta": "54:22:58", "max_grad_norm": 1.0, "loss": 1.096949815750122, "grad_norm": 0.5983024835586548, "learning_rate": 3.7434094903339196e-05}
+{"ts": "2025-12-26T19:07:17", "event": "train_log", "step": 216, "epoch": 0.09113924050632911, "progress_pct": 1.52, "epoch_pct": 1.52, "eta": "54:09:30", "max_grad_norm": 1.0, "loss": 1.1226298809051514, "grad_norm": 0.5641079545021057, "learning_rate": 3.7785588752196835e-05}
+{"ts": "2025-12-26T19:07:33", "event": "train_log", "step": 218, "epoch": 0.0919831223628692, "progress_pct": 1.53, "epoch_pct": 1.53, "eta": "53:55:26", "max_grad_norm": 1.0, "loss": 1.1260643005371094, "grad_norm": 0.655717134475708, "learning_rate": 3.8137082601054486e-05}
+{"ts": "2025-12-26T19:07:48", "event": "train_log", "step": 220, "epoch": 0.09282700421940929, "progress_pct": 1.55, "epoch_pct": 1.55, "eta": "53:42:00", "max_grad_norm": 1.0, "loss": 1.0777709484100342, "grad_norm": 0.6111898422241211, "learning_rate": 3.848857644991213e-05}
+{"ts": "2025-12-26T19:08:04", "event": "train_log", "step": 222, "epoch": 0.09367088607594937, "progress_pct": 1.56, "epoch_pct": 1.56, "eta": "53:29:16", "max_grad_norm": 1.0, "loss": 1.10588800907135, "grad_norm": 0.6821302771568298, "learning_rate": 3.884007029876977e-05}
+{"ts": "2025-12-26T19:08:19", "event": "train_log", "step": 224, "epoch": 0.09451476793248945, "progress_pct": 1.58, "epoch_pct": 1.58, "eta": "53:16:02", "max_grad_norm": 1.0, "loss": 1.1498671770095825, "grad_norm": 0.693175733089447, "learning_rate": 3.919156414762742e-05}
+{"ts": "2025-12-26T19:08:35", "event": "train_log", "step": 226, "epoch": 0.09535864978902954, "progress_pct": 1.59, "epoch_pct": 1.59, "eta": "53:03:34", "max_grad_norm": 1.0, "loss": 1.0587562322616577, "grad_norm": 0.5288166403770447, "learning_rate": 3.954305799648506e-05}
+{"ts": "2025-12-26T19:08:50", "event": "train_log", "step": 228, "epoch": 0.09620253164556962, "progress_pct": 1.6, "epoch_pct": 1.6, "eta": "52:50:35", "max_grad_norm": 1.0, "loss": 1.1107512712478638, "grad_norm": 0.6882867813110352, "learning_rate": 3.989455184534271e-05}
+{"ts": "2025-12-26T19:09:07", "event": "train_log", "step": 230, "epoch": 0.0970464135021097, "progress_pct": 1.62, "epoch_pct": 1.62, "eta": "52:40:01", "max_grad_norm": 1.0, "loss": 1.020510196685791, "grad_norm": 0.5834154486656189, "learning_rate": 4.024604569420035e-05}
+{"ts": "2025-12-26T19:09:22", "event": "train_log", "step": 232, "epoch": 0.09789029535864979, "progress_pct": 1.63, "epoch_pct": 1.63, "eta": "52:27:27", "max_grad_norm": 1.0, "loss": 1.0642449855804443, "grad_norm": 0.7157064080238342, "learning_rate": 4.0597539543058e-05}
+{"ts": "2025-12-26T19:09:39", "event": "train_log", "step": 234, "epoch": 0.09873417721518987, "progress_pct": 1.65, "epoch_pct": 1.65, "eta": "52:16:19", "max_grad_norm": 1.0, "loss": 1.0359872579574585, "grad_norm": 0.6530708074569702, "learning_rate": 4.094903339191564e-05}
+{"ts": "2025-12-26T19:09:53", "event": "train_log", "step": 236, "epoch": 0.09957805907172995, "progress_pct": 1.66, "epoch_pct": 1.66, "eta": "52:03:41", "max_grad_norm": 1.0, "loss": 1.050504446029663, "grad_norm": 0.6329686045646667, "learning_rate": 4.130052724077329e-05}
+{"ts": "2025-12-26T19:10:08", "event": "train_log", "step": 238, "epoch": 0.10042194092827005, "progress_pct": 1.67, "epoch_pct": 1.67, "eta": "51:51:35", "max_grad_norm": 1.0, "loss": 1.2621175050735474, "grad_norm": 0.6597026586532593, "learning_rate": 4.165202108963093e-05}
+{"ts": "2025-12-26T19:10:24", "event": "train_log", "step": 240, "epoch": 0.10126582278481013, "progress_pct": 1.69, "epoch_pct": 1.69, "eta": "51:41:04", "max_grad_norm": 1.0, "loss": 1.1218310594558716, "grad_norm": 0.6195225119590759, "learning_rate": 4.200351493848858e-05}
+{"ts": "2025-12-26T19:10:40", "event": "train_log", "step": 242, "epoch": 0.1021097046413502, "progress_pct": 1.7, "epoch_pct": 1.7, "eta": "51:29:35", "max_grad_norm": 1.0, "loss": 1.1250728368759155, "grad_norm": 0.6764137744903564, "learning_rate": 4.235500878734622e-05}
+{"ts": "2025-12-26T19:10:56", "event": "train_log", "step": 244, "epoch": 0.1029535864978903, "progress_pct": 1.72, "epoch_pct": 1.72, "eta": "51:19:19", "max_grad_norm": 1.0, "loss": 1.028212308883667, "grad_norm": 0.552363395690918, "learning_rate": 4.270650263620387e-05}
+{"ts": "2025-12-26T19:11:12", "event": "train_log", "step": 246, "epoch": 0.10379746835443038, "progress_pct": 1.73, "epoch_pct": 1.73, "eta": "51:09:20", "max_grad_norm": 1.0, "loss": 1.0425450801849365, "grad_norm": 0.5620495676994324, "learning_rate": 4.305799648506151e-05}
+{"ts": "2025-12-26T19:11:27", "event": "train_log", "step": 248, "epoch": 0.10464135021097046, "progress_pct": 1.74, "epoch_pct": 1.74, "eta": "50:57:59", "max_grad_norm": 1.0, "loss": 1.144278883934021, "grad_norm": 0.6860032081604004, "learning_rate": 4.3409490333919156e-05}
+{"ts": "2025-12-26T19:11:41", "event": "train_log", "step": 250, "epoch": 0.10548523206751055, "progress_pct": 1.76, "epoch_pct": 1.76, "eta": "50:46:30", "max_grad_norm": 1.0, "loss": 1.1223982572555542, "grad_norm": 0.6033259034156799, "learning_rate": 4.37609841827768e-05}
+{"ts": "2025-12-26T19:11:57", "event": "train_log", "step": 252, "epoch": 0.10632911392405063, "progress_pct": 1.77, "epoch_pct": 1.77, "eta": "50:36:22", "max_grad_norm": 1.0, "loss": 1.1609960794448853, "grad_norm": 0.6292146444320679, "learning_rate": 4.4112478031634446e-05}
+{"ts": "2025-12-26T19:12:12", "event": "train_log", "step": 254, "epoch": 0.10717299578059072, "progress_pct": 1.79, "epoch_pct": 1.79, "eta": "50:25:56", "max_grad_norm": 1.0, "loss": 1.063547968864441, "grad_norm": 0.7982883453369141, "learning_rate": 4.44639718804921e-05}
+{"ts": "2025-12-26T19:12:27", "event": "train_log", "step": 256, "epoch": 0.1080168776371308, "progress_pct": 1.8, "epoch_pct": 1.8, "eta": "50:14:56", "max_grad_norm": 1.0, "loss": 1.0719804763793945, "grad_norm": 0.7719110250473022, "learning_rate": 4.4815465729349736e-05}
+{"ts": "2025-12-26T19:12:43", "event": "train_log", "step": 258, "epoch": 0.10886075949367088, "progress_pct": 1.81, "epoch_pct": 1.81, "eta": "50:05:41", "max_grad_norm": 1.0, "loss": 1.0778400897979736, "grad_norm": 0.6101011633872986, "learning_rate": 4.516695957820739e-05}
+{"ts": "2025-12-26T19:12:58", "event": "train_log", "step": 260, "epoch": 0.10970464135021098, "progress_pct": 1.83, "epoch_pct": 1.83, "eta": "49:55:27", "max_grad_norm": 1.0, "loss": 1.2129558324813843, "grad_norm": 0.7300994396209717, "learning_rate": 4.5518453427065026e-05}
+{"ts": "2025-12-26T19:13:13", "event": "train_log", "step": 262, "epoch": 0.11054852320675106, "progress_pct": 1.84, "epoch_pct": 1.84, "eta": "49:45:54", "max_grad_norm": 1.0, "loss": 1.221714735031128, "grad_norm": 0.8348747491836548, "learning_rate": 4.586994727592268e-05}
+{"ts": "2025-12-26T19:13:28", "event": "train_log", "step": 264, "epoch": 0.11139240506329114, "progress_pct": 1.86, "epoch_pct": 1.86, "eta": "49:36:28", "max_grad_norm": 1.0, "loss": 1.0187978744506836, "grad_norm": 0.5445612072944641, "learning_rate": 4.6221441124780316e-05}
+{"ts": "2025-12-26T19:13:45", "event": "train_log", "step": 266, "epoch": 0.11223628691983123, "progress_pct": 1.87, "epoch_pct": 1.87, "eta": "49:28:20", "max_grad_norm": 1.0, "loss": 1.096561312675476, "grad_norm": 0.6230319738388062, "learning_rate": 4.657293497363797e-05}
+{"ts": "2025-12-26T19:14:02", "event": "train_log", "step": 268, "epoch": 0.11308016877637131, "progress_pct": 1.88, "epoch_pct": 1.88, "eta": "49:20:11", "max_grad_norm": 1.0, "loss": 1.089842438697815, "grad_norm": 0.6231237649917603, "learning_rate": 4.6924428822495606e-05}
+{"ts": "2025-12-26T19:14:16", "event": "train_log", "step": 270, "epoch": 0.11392405063291139, "progress_pct": 1.9, "epoch_pct": 1.9, "eta": "49:10:11", "max_grad_norm": 1.0, "loss": 1.0696645975112915, "grad_norm": 0.7178627252578735, "learning_rate": 4.727592267135325e-05}
+{"ts": "2025-12-26T19:14:32", "event": "train_log", "step": 272, "epoch": 0.11476793248945148, "progress_pct": 1.91, "epoch_pct": 1.91, "eta": "49:01:48", "max_grad_norm": 1.0, "loss": 1.0511361360549927, "grad_norm": 0.6895854473114014, "learning_rate": 4.7627416520210896e-05}
+{"ts": "2025-12-26T19:14:48", "event": "train_log", "step": 274, "epoch": 0.11561181434599156, "progress_pct": 1.93, "epoch_pct": 1.93, "eta": "48:53:38", "max_grad_norm": 1.0, "loss": 1.1373958587646484, "grad_norm": 0.6046878695487976, "learning_rate": 4.797891036906854e-05}
+{"ts": "2025-12-26T19:15:04", "event": "train_log", "step": 276, "epoch": 0.11645569620253164, "progress_pct": 1.94, "epoch_pct": 1.94, "eta": "48:44:50", "max_grad_norm": 1.0, "loss": 1.0734186172485352, "grad_norm": 0.6524552702903748, "learning_rate": 4.833040421792619e-05}
+{"ts": "2025-12-26T19:15:19", "event": "train_log", "step": 278, "epoch": 0.11729957805907174, "progress_pct": 1.95, "epoch_pct": 1.95, "eta": "48:36:17", "max_grad_norm": 1.0, "loss": 1.123913049697876, "grad_norm": 0.6331019997596741, "learning_rate": 4.868189806678383e-05}
+{"ts": "2025-12-26T19:15:34", "event": "train_log", "step": 280, "epoch": 0.11814345991561181, "progress_pct": 1.97, "epoch_pct": 1.97, "eta": "48:27:43", "max_grad_norm": 1.0, "loss": 1.0635710954666138, "grad_norm": 0.5919018983840942, "learning_rate": 4.903339191564148e-05}
+{"ts": "2025-12-26T19:15:52", "event": "train_log", "step": 282, "epoch": 0.1189873417721519, "progress_pct": 1.98, "epoch_pct": 1.98, "eta": "48:20:49", "max_grad_norm": 1.0, "loss": 1.0429247617721558, "grad_norm": 0.6067633032798767, "learning_rate": 4.938488576449912e-05}
+{"ts": "2025-12-26T19:16:07", "event": "train_log", "step": 284, "epoch": 0.11983122362869199, "progress_pct": 2.0, "epoch_pct": 2.0, "eta": "48:12:20", "max_grad_norm": 1.0, "loss": 1.1397464275360107, "grad_norm": 0.6583750247955322, "learning_rate": 4.9736379613356774e-05}
+{"ts": "2025-12-26T19:16:23", "event": "train_log", "step": 286, "epoch": 0.12067510548523207, "progress_pct": 2.01, "epoch_pct": 2.01, "eta": "48:04:38", "max_grad_norm": 1.0, "loss": 1.0590803623199463, "grad_norm": 0.6200069785118103, "learning_rate": 5.008787346221442e-05}
+{"ts": "2025-12-26T19:16:38", "event": "train_log", "step": 288, "epoch": 0.12151898734177215, "progress_pct": 2.03, "epoch_pct": 2.03, "eta": "47:56:39", "max_grad_norm": 1.0, "loss": 1.1318789720535278, "grad_norm": 0.6798665523529053, "learning_rate": 5.0439367311072064e-05}
+{"ts": "2025-12-26T19:16:55", "event": "train_log", "step": 290, "epoch": 0.12236286919831224, "progress_pct": 2.04, "epoch_pct": 2.04, "eta": "47:49:56", "max_grad_norm": 1.0, "loss": 1.0934956073760986, "grad_norm": 0.7508794069290161, "learning_rate": 5.07908611599297e-05}
+{"ts": "2025-12-26T19:17:10", "event": "train_log", "step": 292, "epoch": 0.12320675105485232, "progress_pct": 2.05, "epoch_pct": 2.05, "eta": "47:42:06", "max_grad_norm": 1.0, "loss": 1.163407802581787, "grad_norm": 0.6901452541351318, "learning_rate": 5.114235500878735e-05}
+{"ts": "2025-12-26T19:17:25", "event": "train_log", "step": 294, "epoch": 0.1240506329113924, "progress_pct": 2.07, "epoch_pct": 2.07, "eta": "47:34:06", "max_grad_norm": 1.0, "loss": 1.09059476852417, "grad_norm": 0.6423285603523254, "learning_rate": 5.1493848857644985e-05}
+{"ts": "2025-12-26T19:17:42", "event": "train_log", "step": 296, "epoch": 0.1248945147679325, "progress_pct": 2.08, "epoch_pct": 2.08, "eta": "47:27:04", "max_grad_norm": 1.0, "loss": 1.0690211057662964, "grad_norm": 0.6839275360107422, "learning_rate": 5.1845342706502644e-05}
+{"ts": "2025-12-26T19:17:58", "event": "train_log", "step": 298, "epoch": 0.1257383966244726, "progress_pct": 2.1, "epoch_pct": 2.1, "eta": "47:20:03", "max_grad_norm": 1.0, "loss": 0.982322096824646, "grad_norm": 0.6350128054618835, "learning_rate": 5.219683655536028e-05}
+{"ts": "2025-12-26T19:18:12", "event": "train_log", "step": 300, "epoch": 0.12658227848101267, "progress_pct": 2.11, "epoch_pct": 2.11, "eta": "47:12:00", "max_grad_norm": 1.0, "loss": 1.1132930517196655, "grad_norm": 0.7136530876159668, "learning_rate": 5.254833040421793e-05}
+{"ts": "2025-12-26T19:29:29", "event": "train_log", "step": 300, "epoch": 0.12658227848101267, "progress_pct": 2.11, "epoch_pct": 2.11, "eta": "55:55:36", "max_grad_norm": 1.0, "eval_loss": 1.0952109098434448, "eval_runtime": 677.0652, "eval_samples_per_second": 3.112, "eval_steps_per_second": 3.112}
+{"ts": "2025-12-26T19:29:45", "event": "train_log", "step": 302, "epoch": 0.12742616033755275, "progress_pct": 2.12, "epoch_pct": 2.12, "eta": "55:44:49", "max_grad_norm": 1.0, "loss": 0.973595917224884, "grad_norm": 0.7339721322059631, "learning_rate": 5.289982425307557e-05}
+{"ts": "2025-12-26T19:30:00", "event": "train_log", "step": 304, "epoch": 0.12827004219409283, "progress_pct": 2.14, "epoch_pct": 2.14, "eta": "55:33:54", "max_grad_norm": 1.0, "loss": 0.9819849729537964, "grad_norm": 0.5941481590270996, "learning_rate": 5.3251318101933224e-05}
+{"ts": "2025-12-26T19:30:15", "event": "train_log", "step": 306, "epoch": 0.1291139240506329, "progress_pct": 2.15, "epoch_pct": 2.15, "eta": "55:23:03", "max_grad_norm": 1.0, "loss": 1.0315470695495605, "grad_norm": 0.7153938412666321, "learning_rate": 5.360281195079086e-05}
+{"ts": "2025-12-26T19:30:31", "event": "train_log", "step": 308, "epoch": 0.12995780590717299, "progress_pct": 2.17, "epoch_pct": 2.17, "eta": "55:13:01", "max_grad_norm": 1.0, "loss": 0.9492001533508301, "grad_norm": 0.5167180299758911, "learning_rate": 5.395430579964851e-05}
+{"ts": "2025-12-26T19:30:46", "event": "train_log", "step": 310, "epoch": 0.1308016877637131, "progress_pct": 2.18, "epoch_pct": 2.18, "eta": "55:02:21", "max_grad_norm": 1.0, "loss": 1.0156209468841553, "grad_norm": 0.6055944561958313, "learning_rate": 5.430579964850615e-05}
+{"ts": "2025-12-26T19:31:00", "event": "train_log", "step": 312, "epoch": 0.13164556962025317, "progress_pct": 2.19, "epoch_pct": 2.19, "eta": "54:51:12", "max_grad_norm": 1.0, "loss": 1.1791651248931885, "grad_norm": 0.7662386298179626, "learning_rate": 5.4657293497363805e-05}
+{"ts": "2025-12-26T19:31:16", "event": "train_log", "step": 314, "epoch": 0.13248945147679325, "progress_pct": 2.21, "epoch_pct": 2.21, "eta": "54:41:26", "max_grad_norm": 1.0, "loss": 1.0009297132492065, "grad_norm": 0.6065546274185181, "learning_rate": 5.500878734622145e-05}
+{"ts": "2025-12-26T19:31:31", "event": "train_log", "step": 316, "epoch": 0.13333333333333333, "progress_pct": 2.22, "epoch_pct": 2.22, "eta": "54:31:32", "max_grad_norm": 1.0, "loss": 1.0208244323730469, "grad_norm": 0.604225754737854, "learning_rate": 5.536028119507909e-05}
+{"ts": "2025-12-26T19:31:47", "event": "train_log", "step": 318, "epoch": 0.1341772151898734, "progress_pct": 2.24, "epoch_pct": 2.24, "eta": "54:21:35", "max_grad_norm": 1.0, "loss": 0.9968416690826416, "grad_norm": 0.6186763048171997, "learning_rate": 5.571177504393673e-05}
+{"ts": "2025-12-26T19:32:02", "event": "train_log", "step": 320, "epoch": 0.1350210970464135, "progress_pct": 2.25, "epoch_pct": 2.25, "eta": "54:11:46", "max_grad_norm": 1.0, "loss": 0.9540256857872009, "grad_norm": 0.7100363969802856, "learning_rate": 5.606326889279437e-05}
+{"ts": "2025-12-26T19:32:16", "event": "train_log", "step": 322, "epoch": 0.1358649789029536, "progress_pct": 2.26, "epoch_pct": 2.26, "eta": "54:01:33", "max_grad_norm": 1.0, "loss": 1.0631953477859497, "grad_norm": 0.6979711055755615, "learning_rate": 5.641476274165203e-05}
+{"ts": "2025-12-26T19:32:32", "event": "train_log", "step": 324, "epoch": 0.13670886075949368, "progress_pct": 2.28, "epoch_pct": 2.28, "eta": "53:52:08", "max_grad_norm": 1.0, "loss": 1.0170501470565796, "grad_norm": 0.6237109303474426, "learning_rate": 5.676625659050967e-05}
+{"ts": "2025-12-26T19:32:48", "event": "train_log", "step": 326, "epoch": 0.13755274261603376, "progress_pct": 2.29, "epoch_pct": 2.29, "eta": "53:43:14", "max_grad_norm": 1.0, "loss": 1.0715603828430176, "grad_norm": 0.6525548696517944, "learning_rate": 5.711775043936731e-05}
+{"ts": "2025-12-26T19:33:03", "event": "train_log", "step": 328, "epoch": 0.13839662447257384, "progress_pct": 2.31, "epoch_pct": 2.31, "eta": "53:33:47", "max_grad_norm": 1.0, "loss": 1.0111541748046875, "grad_norm": 0.6869221329689026, "learning_rate": 5.746924428822495e-05}
+{"ts": "2025-12-26T19:33:18", "event": "train_log", "step": 330, "epoch": 0.13924050632911392, "progress_pct": 2.32, "epoch_pct": 2.32, "eta": "53:24:42", "max_grad_norm": 1.0, "loss": 1.0311682224273682, "grad_norm": 0.553188145160675, "learning_rate": 5.782073813708261e-05}
+{"ts": "2025-12-26T19:33:33", "event": "train_log", "step": 332, "epoch": 0.140084388185654, "progress_pct": 2.33, "epoch_pct": 2.33, "eta": "53:15:19", "max_grad_norm": 1.0, "loss": 1.0213634967803955, "grad_norm": 0.6760852932929993, "learning_rate": 5.817223198594025e-05}
+{"ts": "2025-12-26T19:33:49", "event": "train_log", "step": 334, "epoch": 0.1409282700421941, "progress_pct": 2.35, "epoch_pct": 2.35, "eta": "53:06:28", "max_grad_norm": 1.0, "loss": 0.9748594164848328, "grad_norm": 0.5907419919967651, "learning_rate": 5.8523725834797894e-05}
+{"ts": "2025-12-26T19:34:03", "event": "train_log", "step": 336, "epoch": 0.14177215189873418, "progress_pct": 2.36, "epoch_pct": 2.36, "eta": "52:56:30", "max_grad_norm": 1.0, "loss": 1.05863356590271, "grad_norm": 0.7044920921325684, "learning_rate": 5.887521968365554e-05}
+{"ts": "2025-12-26T19:34:17", "event": "train_log", "step": 338, "epoch": 0.14261603375527426, "progress_pct": 2.38, "epoch_pct": 2.38, "eta": "52:47:06", "max_grad_norm": 1.0, "loss": 1.1341127157211304, "grad_norm": 0.679073691368103, "learning_rate": 5.922671353251318e-05}
+{"ts": "2025-12-26T19:34:32", "event": "train_log", "step": 340, "epoch": 0.14345991561181434, "progress_pct": 2.39, "epoch_pct": 2.39, "eta": "52:38:12", "max_grad_norm": 1.0, "loss": 0.9540836215019226, "grad_norm": 0.7676237225532532, "learning_rate": 5.957820738137083e-05}
+{"ts": "2025-12-26T19:34:47", "event": "train_log", "step": 342, "epoch": 0.14430379746835442, "progress_pct": 2.41, "epoch_pct": 2.41, "eta": "52:29:26", "max_grad_norm": 1.0, "loss": 1.0585911273956299, "grad_norm": 0.6313899755477905, "learning_rate": 5.9929701230228474e-05}
+{"ts": "2025-12-26T19:35:02", "event": "train_log", "step": 344, "epoch": 0.1451476793248945, "progress_pct": 2.42, "epoch_pct": 2.42, "eta": "52:20:40", "max_grad_norm": 1.0, "loss": 1.0760118961334229, "grad_norm": 0.7123099565505981, "learning_rate": 6.028119507908612e-05}
+{"ts": "2025-12-26T19:35:17", "event": "train_log", "step": 346, "epoch": 0.1459915611814346, "progress_pct": 2.43, "epoch_pct": 2.43, "eta": "52:12:28", "max_grad_norm": 1.0, "loss": 1.036866307258606, "grad_norm": 0.585935652256012, "learning_rate": 6.063268892794376e-05}
+{"ts": "2025-12-26T19:35:32", "event": "train_log", "step": 348, "epoch": 0.1468354430379747, "progress_pct": 2.45, "epoch_pct": 2.45, "eta": "52:03:40", "max_grad_norm": 1.0, "loss": 1.0242938995361328, "grad_norm": 0.5643263459205627, "learning_rate": 6.0984182776801416e-05}
+{"ts": "2025-12-26T19:35:47", "event": "train_log", "step": 350, "epoch": 0.14767932489451477, "progress_pct": 2.46, "epoch_pct": 2.46, "eta": "51:55:02", "max_grad_norm": 1.0, "loss": 1.0497376918792725, "grad_norm": 0.626761794090271, "learning_rate": 6.133567662565906e-05}
+{"ts": "2025-12-26T19:36:03", "event": "train_log", "step": 352, "epoch": 0.14852320675105485, "progress_pct": 2.48, "epoch_pct": 2.48, "eta": "51:47:28", "max_grad_norm": 1.0, "loss": 0.9811885356903076, "grad_norm": 0.5106956958770752, "learning_rate": 6.16871704745167e-05}
+{"ts": "2025-12-26T19:36:19", "event": "train_log", "step": 354, "epoch": 0.14936708860759493, "progress_pct": 2.49, "epoch_pct": 2.49, "eta": "51:40:08", "max_grad_norm": 1.0, "loss": 1.0715330839157104, "grad_norm": 0.6948089003562927, "learning_rate": 6.203866432337434e-05}
+{"ts": "2025-12-26T19:36:35", "event": "train_log", "step": 356, "epoch": 0.150210970464135, "progress_pct": 2.5, "epoch_pct": 2.5, "eta": "51:32:27", "max_grad_norm": 1.0, "loss": 1.0405226945877075, "grad_norm": 0.699713945388794, "learning_rate": 6.239015817223199e-05}
+{"ts": "2025-12-26T19:36:51", "event": "train_log", "step": 358, "epoch": 0.15105485232067511, "progress_pct": 2.52, "epoch_pct": 2.52, "eta": "51:25:30", "max_grad_norm": 1.0, "loss": 1.0490930080413818, "grad_norm": 0.6437667012214661, "learning_rate": 6.274165202108964e-05}
+{"ts": "2025-12-26T19:37:07", "event": "train_log", "step": 360, "epoch": 0.1518987341772152, "progress_pct": 2.53, "epoch_pct": 2.53, "eta": "51:17:47", "max_grad_norm": 1.0, "loss": 0.9267548322677612, "grad_norm": 0.6952699422836304, "learning_rate": 6.309314586994728e-05}
+{"ts": "2025-12-26T19:37:22", "event": "train_log", "step": 362, "epoch": 0.15274261603375527, "progress_pct": 2.55, "epoch_pct": 2.55, "eta": "51:10:08", "max_grad_norm": 1.0, "loss": 1.0427420139312744, "grad_norm": 0.6713186502456665, "learning_rate": 6.344463971880492e-05}
+{"ts": "2025-12-26T19:37:38", "event": "train_log", "step": 364, "epoch": 0.15358649789029535, "progress_pct": 2.56, "epoch_pct": 2.56, "eta": "51:02:50", "max_grad_norm": 1.0, "loss": 1.048950433731079, "grad_norm": 0.6750379800796509, "learning_rate": 6.379613356766257e-05}
+{"ts": "2025-12-26T19:37:54", "event": "train_log", "step": 366, "epoch": 0.15443037974683543, "progress_pct": 2.57, "epoch_pct": 2.57, "eta": "50:55:57", "max_grad_norm": 1.0, "loss": 1.0156004428863525, "grad_norm": 0.6053379774093628, "learning_rate": 6.414762741652022e-05}
+{"ts": "2025-12-26T19:38:10", "event": "train_log", "step": 368, "epoch": 0.15527426160337554, "progress_pct": 2.59, "epoch_pct": 2.59, "eta": "50:48:36", "max_grad_norm": 1.0, "loss": 1.0020819902420044, "grad_norm": 0.8063633441925049, "learning_rate": 6.449912126537786e-05}
+{"ts": "2025-12-26T19:38:26", "event": "train_log", "step": 370, "epoch": 0.15611814345991562, "progress_pct": 2.6, "epoch_pct": 2.6, "eta": "50:41:51", "max_grad_norm": 1.0, "loss": 1.055633783340454, "grad_norm": 0.8027494549751282, "learning_rate": 6.48506151142355e-05}
+{"ts": "2025-12-26T19:38:42", "event": "train_log", "step": 372, "epoch": 0.1569620253164557, "progress_pct": 2.62, "epoch_pct": 2.62, "eta": "50:35:05", "max_grad_norm": 1.0, "loss": 1.0149940252304077, "grad_norm": 0.6580121517181396, "learning_rate": 6.520210896309315e-05}
+{"ts": "2025-12-26T19:38:58", "event": "train_log", "step": 374, "epoch": 0.15780590717299578, "progress_pct": 2.63, "epoch_pct": 2.63, "eta": "50:28:07", "max_grad_norm": 1.0, "loss": 0.9769611954689026, "grad_norm": 0.6561233997344971, "learning_rate": 6.55536028119508e-05}
+{"ts": "2025-12-26T19:39:16", "event": "train_log", "step": 376, "epoch": 0.15864978902953586, "progress_pct": 2.64, "epoch_pct": 2.64, "eta": "50:22:50", "max_grad_norm": 1.0, "loss": 0.9099349975585938, "grad_norm": 0.6444346308708191, "learning_rate": 6.590509666080844e-05}
+{"ts": "2025-12-26T19:39:32", "event": "train_log", "step": 378, "epoch": 0.15949367088607594, "progress_pct": 2.66, "epoch_pct": 2.66, "eta": "50:16:06", "max_grad_norm": 1.0, "loss": 1.0797548294067383, "grad_norm": 0.5879359245300293, "learning_rate": 6.625659050966608e-05}
+{"ts": "2025-12-26T19:39:47", "event": "train_log", "step": 380, "epoch": 0.16033755274261605, "progress_pct": 2.67, "epoch_pct": 2.67, "eta": "50:08:50", "max_grad_norm": 1.0, "loss": 1.0336791276931763, "grad_norm": 0.6994144916534424, "learning_rate": 6.660808435852373e-05}
+{"ts": "2025-12-26T19:40:03", "event": "train_log", "step": 382, "epoch": 0.16118143459915613, "progress_pct": 2.69, "epoch_pct": 2.69, "eta": "50:02:10", "max_grad_norm": 1.0, "loss": 1.018118143081665, "grad_norm": 0.6128669381141663, "learning_rate": 6.695957820738138e-05}
+{"ts": "2025-12-26T19:40:18", "event": "train_log", "step": 384, "epoch": 0.1620253164556962, "progress_pct": 2.7, "epoch_pct": 2.7, "eta": "49:55:31", "max_grad_norm": 1.0, "loss": 1.1405497789382935, "grad_norm": 1.0237540006637573, "learning_rate": 6.731107205623902e-05}
+{"ts": "2025-12-26T19:40:35", "event": "train_log", "step": 386, "epoch": 0.16286919831223629, "progress_pct": 2.71, "epoch_pct": 2.71, "eta": "49:49:24", "max_grad_norm": 1.0, "loss": 1.0314189195632935, "grad_norm": 0.6091578006744385, "learning_rate": 6.766256590509666e-05}
+{"ts": "2025-12-26T19:40:52", "event": "train_log", "step": 388, "epoch": 0.16371308016877636, "progress_pct": 2.73, "epoch_pct": 2.73, "eta": "49:43:57", "max_grad_norm": 1.0, "loss": 0.9564052820205688, "grad_norm": 0.5916037559509277, "learning_rate": 6.801405975395431e-05}
+{"ts": "2025-12-26T19:41:08", "event": "train_log", "step": 390, "epoch": 0.16455696202531644, "progress_pct": 2.74, "epoch_pct": 2.74, "eta": "49:37:22", "max_grad_norm": 1.0, "loss": 1.0023859739303589, "grad_norm": 0.771653950214386, "learning_rate": 6.836555360281195e-05}
+{"ts": "2025-12-26T19:41:23", "event": "train_log", "step": 392, "epoch": 0.16540084388185655, "progress_pct": 2.76, "epoch_pct": 2.76, "eta": "49:30:32", "max_grad_norm": 1.0, "loss": 1.07024085521698, "grad_norm": 0.654658317565918, "learning_rate": 6.87170474516696e-05}
+{"ts": "2025-12-26T19:41:38", "event": "train_log", "step": 394, "epoch": 0.16624472573839663, "progress_pct": 2.77, "epoch_pct": 2.77, "eta": "49:23:51", "max_grad_norm": 1.0, "loss": 1.0552500486373901, "grad_norm": 0.6611968874931335, "learning_rate": 6.906854130052724e-05}
+{"ts": "2025-12-26T19:41:53", "event": "train_log", "step": 396, "epoch": 0.1670886075949367, "progress_pct": 2.78, "epoch_pct": 2.78, "eta": "49:17:24", "max_grad_norm": 1.0, "loss": 1.0562875270843506, "grad_norm": 0.6955893039703369, "learning_rate": 6.942003514938489e-05}
+{"ts": "2025-12-26T19:42:08", "event": "train_log", "step": 398, "epoch": 0.1679324894514768, "progress_pct": 2.8, "epoch_pct": 2.8, "eta": "49:10:51", "max_grad_norm": 1.0, "loss": 0.9850592017173767, "grad_norm": 0.6666058301925659, "learning_rate": 6.977152899824253e-05}
+{"ts": "2025-12-26T19:42:24", "event": "train_log", "step": 400, "epoch": 0.16877637130801687, "progress_pct": 2.81, "epoch_pct": 2.81, "eta": "49:04:38", "max_grad_norm": 1.0, "loss": 1.0077755451202393, "grad_norm": 0.6131711006164551, "learning_rate": 7.012302284710018e-05}
+{"ts": "2025-12-26T19:53:55", "event": "train_log", "step": 400, "epoch": 0.16877637130801687, "progress_pct": 2.81, "epoch_pct": 2.81, "eta": "55:42:32", "max_grad_norm": 1.0, "eval_loss": 1.0625108480453491, "eval_runtime": 691.0068, "eval_samples_per_second": 3.049, "eval_steps_per_second": 3.049}
+{"ts": "2025-12-26T19:54:10", "event": "train_log", "step": 402, "epoch": 0.16962025316455695, "progress_pct": 2.83, "epoch_pct": 2.83, "eta": "55:33:53", "max_grad_norm": 1.0, "loss": 1.1012427806854248, "grad_norm": 0.6286499500274658, "learning_rate": 7.047451669595783e-05}
+{"ts": "2025-12-26T19:54:25", "event": "train_log", "step": 404, "epoch": 0.17046413502109706, "progress_pct": 2.84, "epoch_pct": 2.84, "eta": "55:25:32", "max_grad_norm": 1.0, "loss": 1.0379719734191895, "grad_norm": 0.6639351844787598, "learning_rate": 7.082601054481547e-05}
+{"ts": "2025-12-26T19:54:40", "event": "train_log", "step": 406, "epoch": 0.17130801687763714, "progress_pct": 2.86, "epoch_pct": 2.86, "eta": "55:17:01", "max_grad_norm": 1.0, "loss": 1.031856656074524, "grad_norm": 0.750401496887207, "learning_rate": 7.117750439367311e-05}
+{"ts": "2025-12-26T19:54:53", "event": "train_log", "step": 408, "epoch": 0.17215189873417722, "progress_pct": 2.87, "epoch_pct": 2.87, "eta": "55:08:10", "max_grad_norm": 1.0, "loss": 1.0493193864822388, "grad_norm": 0.8084847331047058, "learning_rate": 7.152899824253075e-05}
+{"ts": "2025-12-26T19:55:09", "event": "train_log", "step": 410, "epoch": 0.1729957805907173, "progress_pct": 2.88, "epoch_pct": 2.88, "eta": "55:00:00", "max_grad_norm": 1.0, "loss": 1.1012418270111084, "grad_norm": 0.7448462247848511, "learning_rate": 7.188049209138841e-05}
+{"ts": "2025-12-26T19:55:24", "event": "train_log", "step": 412, "epoch": 0.17383966244725738, "progress_pct": 2.9, "epoch_pct": 2.9, "eta": "54:52:18", "max_grad_norm": 1.0, "loss": 0.9926692247390747, "grad_norm": 0.5841867923736572, "learning_rate": 7.223198594024605e-05}
+{"ts": "2025-12-26T19:55:39", "event": "train_log", "step": 414, "epoch": 0.17468354430379746, "progress_pct": 2.91, "epoch_pct": 2.91, "eta": "54:44:09", "max_grad_norm": 1.0, "loss": 1.0588877201080322, "grad_norm": 0.7125606536865234, "learning_rate": 7.258347978910369e-05}
+{"ts": "2025-12-26T19:55:54", "event": "train_log", "step": 416, "epoch": 0.17552742616033756, "progress_pct": 2.93, "epoch_pct": 2.93, "eta": "54:36:25", "max_grad_norm": 1.0, "loss": 1.038270354270935, "grad_norm": 0.5750942230224609, "learning_rate": 7.293497363796134e-05}
+{"ts": "2025-12-26T19:56:10", "event": "train_log", "step": 418, "epoch": 0.17637130801687764, "progress_pct": 2.94, "epoch_pct": 2.94, "eta": "54:28:45", "max_grad_norm": 1.0, "loss": 0.9843021035194397, "grad_norm": 0.565444827079773, "learning_rate": 7.328646748681899e-05}
+{"ts": "2025-12-26T19:56:25", "event": "train_log", "step": 420, "epoch": 0.17721518987341772, "progress_pct": 2.95, "epoch_pct": 2.95, "eta": "54:21:02", "max_grad_norm": 1.0, "loss": 1.0731632709503174, "grad_norm": 0.5825693011283875, "learning_rate": 7.363796133567663e-05}
+{"ts": "2025-12-26T19:56:40", "event": "train_log", "step": 422, "epoch": 0.1780590717299578, "progress_pct": 2.97, "epoch_pct": 2.97, "eta": "54:13:08", "max_grad_norm": 1.0, "loss": 1.0061273574829102, "grad_norm": 0.6267391443252563, "learning_rate": 7.398945518453427e-05}
+{"ts": "2025-12-26T19:56:54", "event": "train_log", "step": 424, "epoch": 0.17890295358649788, "progress_pct": 2.98, "epoch_pct": 2.98, "eta": "54:05:00", "max_grad_norm": 1.0, "loss": 1.0461612939834595, "grad_norm": 0.6621372103691101, "learning_rate": 7.434094903339192e-05}
+{"ts": "2025-12-26T19:57:09", "event": "train_log", "step": 426, "epoch": 0.17974683544303796, "progress_pct": 3.0, "epoch_pct": 3.0, "eta": "53:57:21", "max_grad_norm": 1.0, "loss": 0.9789207577705383, "grad_norm": 0.6635435223579407, "learning_rate": 7.469244288224957e-05}
+{"ts": "2025-12-26T19:57:23", "event": "train_log", "step": 428, "epoch": 0.18059071729957807, "progress_pct": 3.01, "epoch_pct": 3.01, "eta": "53:49:30", "max_grad_norm": 1.0, "loss": 1.0527069568634033, "grad_norm": 0.6342346668243408, "learning_rate": 7.504393673110721e-05}
+{"ts": "2025-12-26T19:57:39", "event": "train_log", "step": 430, "epoch": 0.18143459915611815, "progress_pct": 3.02, "epoch_pct": 3.02, "eta": "53:42:34", "max_grad_norm": 1.0, "loss": 0.9708702564239502, "grad_norm": 0.6762149930000305, "learning_rate": 7.539543057996485e-05}
+{"ts": "2025-12-26T19:57:54", "event": "train_log", "step": 432, "epoch": 0.18227848101265823, "progress_pct": 3.04, "epoch_pct": 3.04, "eta": "53:35:00", "max_grad_norm": 1.0, "loss": 1.0509834289550781, "grad_norm": 0.7073282599449158, "learning_rate": 7.57469244288225e-05}
+{"ts": "2025-12-26T19:58:09", "event": "train_log", "step": 434, "epoch": 0.1831223628691983, "progress_pct": 3.05, "epoch_pct": 3.05, "eta": "53:27:34", "max_grad_norm": 1.0, "loss": 1.0128819942474365, "grad_norm": 0.6917856931686401, "learning_rate": 7.609841827768014e-05}
+{"ts": "2025-12-26T19:58:25", "event": "train_log", "step": 436, "epoch": 0.1839662447257384, "progress_pct": 3.07, "epoch_pct": 3.07, "eta": "53:21:04", "max_grad_norm": 1.0, "loss": 0.989395797252655, "grad_norm": 0.5574942231178284, "learning_rate": 7.644991212653779e-05}
+{"ts": "2025-12-26T19:58:41", "event": "train_log", "step": 438, "epoch": 0.1848101265822785, "progress_pct": 3.08, "epoch_pct": 3.08, "eta": "53:14:12", "max_grad_norm": 1.0, "loss": 0.9846042990684509, "grad_norm": 0.640765905380249, "learning_rate": 7.680140597539543e-05}
+{"ts": "2025-12-26T19:58:56", "event": "train_log", "step": 440, "epoch": 0.18565400843881857, "progress_pct": 3.09, "epoch_pct": 3.09, "eta": "53:06:51", "max_grad_norm": 1.0, "loss": 1.0344442129135132, "grad_norm": 0.6699127554893494, "learning_rate": 7.715289982425308e-05}
+{"ts": "2025-12-26T19:59:11", "event": "train_log", "step": 442, "epoch": 0.18649789029535865, "progress_pct": 3.11, "epoch_pct": 3.11, "eta": "53:00:08", "max_grad_norm": 1.0, "loss": 1.0179373025894165, "grad_norm": 0.6164930462837219, "learning_rate": 7.750439367311072e-05}
+{"ts": "2025-12-26T19:59:26", "event": "train_log", "step": 444, "epoch": 0.18734177215189873, "progress_pct": 3.12, "epoch_pct": 3.12, "eta": "52:53:05", "max_grad_norm": 1.0, "loss": 1.0518895387649536, "grad_norm": 0.6880720853805542, "learning_rate": 7.785588752196837e-05}
+{"ts": "2025-12-26T19:59:41", "event": "train_log", "step": 446, "epoch": 0.1881856540084388, "progress_pct": 3.14, "epoch_pct": 3.14, "eta": "52:45:49", "max_grad_norm": 1.0, "loss": 1.0442606210708618, "grad_norm": 0.6501413583755493, "learning_rate": 7.820738137082601e-05}
+{"ts": "2025-12-26T19:59:55", "event": "train_log", "step": 448, "epoch": 0.1890295358649789, "progress_pct": 3.15, "epoch_pct": 3.15, "eta": "52:38:48", "max_grad_norm": 1.0, "loss": 0.9828442335128784, "grad_norm": 0.6076085567474365, "learning_rate": 7.855887521968366e-05}
+{"ts": "2025-12-26T20:00:10", "event": "train_log", "step": 450, "epoch": 0.189873417721519, "progress_pct": 3.16, "epoch_pct": 3.16, "eta": "52:31:43", "max_grad_norm": 1.0, "loss": 1.0573710203170776, "grad_norm": 0.6418202519416809, "learning_rate": 7.89103690685413e-05}
+{"ts": "2025-12-26T20:00:25", "event": "train_log", "step": 452, "epoch": 0.19071729957805908, "progress_pct": 3.18, "epoch_pct": 3.18, "eta": "52:24:44", "max_grad_norm": 1.0, "loss": 1.0216103792190552, "grad_norm": 0.7055076360702515, "learning_rate": 7.926186291739895e-05}
+{"ts": "2025-12-26T20:00:39", "event": "train_log", "step": 454, "epoch": 0.19156118143459916, "progress_pct": 3.19, "epoch_pct": 3.19, "eta": "52:17:54", "max_grad_norm": 1.0, "loss": 0.9837722778320312, "grad_norm": 0.5668330192565918, "learning_rate": 7.961335676625659e-05}
+{"ts": "2025-12-26T20:00:54", "event": "train_log", "step": 456, "epoch": 0.19240506329113924, "progress_pct": 3.21, "epoch_pct": 3.21, "eta": "52:10:53", "max_grad_norm": 1.0, "loss": 1.0003894567489624, "grad_norm": 0.6419380307197571, "learning_rate": 7.996485061511424e-05}
+{"ts": "2025-12-26T20:01:09", "event": "train_log", "step": 458, "epoch": 0.19324894514767932, "progress_pct": 3.22, "epoch_pct": 3.22, "eta": "52:04:16", "max_grad_norm": 1.0, "loss": 1.0609031915664673, "grad_norm": 0.5949198007583618, "learning_rate": 8.031634446397188e-05}
+{"ts": "2025-12-26T20:01:24", "event": "train_log", "step": 460, "epoch": 0.1940928270042194, "progress_pct": 3.23, "epoch_pct": 3.23, "eta": "51:57:39", "max_grad_norm": 1.0, "loss": 1.0543403625488281, "grad_norm": 0.7032039761543274, "learning_rate": 8.066783831282952e-05}
+{"ts": "2025-12-26T20:01:39", "event": "train_log", "step": 462, "epoch": 0.1949367088607595, "progress_pct": 3.25, "epoch_pct": 3.25, "eta": "51:51:20", "max_grad_norm": 1.0, "loss": 0.9819303154945374, "grad_norm": 0.5775868892669678, "learning_rate": 8.101933216168718e-05}
+{"ts": "2025-12-26T20:01:55", "event": "train_log", "step": 464, "epoch": 0.19578059071729959, "progress_pct": 3.26, "epoch_pct": 3.26, "eta": "51:45:12", "max_grad_norm": 1.0, "loss": 1.0542067289352417, "grad_norm": 0.9301062226295471, "learning_rate": 8.137082601054482e-05}
+{"ts": "2025-12-26T20:02:10", "event": "train_log", "step": 466, "epoch": 0.19662447257383966, "progress_pct": 3.28, "epoch_pct": 3.28, "eta": "51:38:58", "max_grad_norm": 1.0, "loss": 0.9966341257095337, "grad_norm": 0.6193217039108276, "learning_rate": 8.172231985940246e-05}
+{"ts": "2025-12-26T20:02:25", "event": "train_log", "step": 468, "epoch": 0.19746835443037974, "progress_pct": 3.29, "epoch_pct": 3.29, "eta": "51:32:35", "max_grad_norm": 1.0, "loss": 1.0474121570587158, "grad_norm": 0.6286146640777588, "learning_rate": 8.20738137082601e-05}
+{"ts": "2025-12-26T20:02:40", "event": "train_log", "step": 470, "epoch": 0.19831223628691982, "progress_pct": 3.31, "epoch_pct": 3.31, "eta": "51:26:23", "max_grad_norm": 1.0, "loss": 0.9549239277839661, "grad_norm": 0.7418972253799438, "learning_rate": 8.242530755711776e-05}
+{"ts": "2025-12-26T20:02:56", "event": "train_log", "step": 472, "epoch": 0.1991561181434599, "progress_pct": 3.32, "epoch_pct": 3.32, "eta": "51:20:28", "max_grad_norm": 1.0, "loss": 1.0191338062286377, "grad_norm": 0.6122808456420898, "learning_rate": 8.27768014059754e-05}
+{"ts": "2025-12-26T20:03:11", "event": "train_log", "step": 474, "epoch": 0.2, "progress_pct": 3.33, "epoch_pct": 3.33, "eta": "51:14:22", "max_grad_norm": 1.0, "loss": 1.0987539291381836, "grad_norm": 0.6375362277030945, "learning_rate": 8.312829525483304e-05}
+{"ts": "2025-12-26T20:03:26", "event": "train_log", "step": 476, "epoch": 0.2008438818565401, "progress_pct": 3.35, "epoch_pct": 3.35, "eta": "51:08:04", "max_grad_norm": 1.0, "loss": 1.0369136333465576, "grad_norm": 0.6459513306617737, "learning_rate": 8.347978910369068e-05}
+{"ts": "2025-12-26T20:03:41", "event": "train_log", "step": 478, "epoch": 0.20168776371308017, "progress_pct": 3.36, "epoch_pct": 3.36, "eta": "51:02:13", "max_grad_norm": 1.0, "loss": 1.0582096576690674, "grad_norm": 0.7029640674591064, "learning_rate": 8.383128295254833e-05}
+{"ts": "2025-12-26T20:03:56", "event": "train_log", "step": 480, "epoch": 0.20253164556962025, "progress_pct": 3.38, "epoch_pct": 3.38, "eta": "50:56:07", "max_grad_norm": 1.0, "loss": 1.022916316986084, "grad_norm": 0.6345387697219849, "learning_rate": 8.418277680140598e-05}
+{"ts": "2025-12-26T20:04:11", "event": "train_log", "step": 482, "epoch": 0.20337552742616033, "progress_pct": 3.39, "epoch_pct": 3.39, "eta": "50:50:12", "max_grad_norm": 1.0, "loss": 0.973024308681488, "grad_norm": 0.5764590501785278, "learning_rate": 8.453427065026362e-05}
+{"ts": "2025-12-26T20:04:26", "event": "train_log", "step": 484, "epoch": 0.2042194092827004, "progress_pct": 3.4, "epoch_pct": 3.4, "eta": "50:44:12", "max_grad_norm": 1.0, "loss": 1.0292812585830688, "grad_norm": 0.5884482860565186, "learning_rate": 8.488576449912127e-05}
+{"ts": "2025-12-26T20:04:41", "event": "train_log", "step": 486, "epoch": 0.20506329113924052, "progress_pct": 3.42, "epoch_pct": 3.42, "eta": "50:38:27", "max_grad_norm": 1.0, "loss": 1.0083447694778442, "grad_norm": 0.616357147693634, "learning_rate": 8.523725834797891e-05}
+{"ts": "2025-12-26T20:04:56", "event": "train_log", "step": 488, "epoch": 0.2059071729957806, "progress_pct": 3.43, "epoch_pct": 3.43, "eta": "50:32:38", "max_grad_norm": 1.0, "loss": 0.9936985373497009, "grad_norm": 0.7671196460723877, "learning_rate": 8.558875219683656e-05}
+{"ts": "2025-12-26T20:05:11", "event": "train_log", "step": 490, "epoch": 0.20675105485232068, "progress_pct": 3.45, "epoch_pct": 3.45, "eta": "50:26:41", "max_grad_norm": 1.0, "loss": 1.051513910293579, "grad_norm": 0.6197299957275391, "learning_rate": 8.59402460456942e-05}
+{"ts": "2025-12-26T20:05:26", "event": "train_log", "step": 492, "epoch": 0.20759493670886076, "progress_pct": 3.46, "epoch_pct": 3.46, "eta": "50:20:53", "max_grad_norm": 1.0, "loss": 0.9474978446960449, "grad_norm": 0.6912890672683716, "learning_rate": 8.629173989455185e-05}
+{"ts": "2025-12-26T20:05:41", "event": "train_log", "step": 494, "epoch": 0.20843881856540084, "progress_pct": 3.47, "epoch_pct": 3.47, "eta": "50:15:02", "max_grad_norm": 1.0, "loss": 1.0671660900115967, "grad_norm": 0.6941592693328857, "learning_rate": 8.664323374340949e-05}
+{"ts": "2025-12-26T20:05:56", "event": "train_log", "step": 496, "epoch": 0.20928270042194091, "progress_pct": 3.49, "epoch_pct": 3.49, "eta": "50:09:32", "max_grad_norm": 1.0, "loss": 1.0020159482955933, "grad_norm": 0.5889528393745422, "learning_rate": 8.699472759226714e-05}
+{"ts": "2025-12-26T20:06:12", "event": "train_log", "step": 498, "epoch": 0.21012658227848102, "progress_pct": 3.5, "epoch_pct": 3.5, "eta": "50:04:05", "max_grad_norm": 1.0, "loss": 1.0165860652923584, "grad_norm": 0.6478549838066101, "learning_rate": 8.734622144112478e-05}
+{"ts": "2025-12-26T20:06:28", "event": "train_log", "step": 500, "epoch": 0.2109704641350211, "progress_pct": 3.52, "epoch_pct": 3.52, "eta": "49:58:53", "max_grad_norm": 1.0, "loss": 0.8798263072967529, "grad_norm": 0.6018255949020386, "learning_rate": 8.769771528998243e-05}
+{"ts": "2025-12-26T20:18:00", "event": "train_log", "step": 500, "epoch": 0.2109704641350211, "progress_pct": 3.52, "epoch_pct": 3.52, "eta": "55:15:34", "max_grad_norm": 1.0, "eval_loss": 1.042096495628357, "eval_runtime": 692.4361, "eval_samples_per_second": 3.043, "eval_steps_per_second": 3.043}
+{"ts": "2025-12-26T20:18:16", "event": "train_log", "step": 502, "epoch": 0.21181434599156118, "progress_pct": 3.53, "epoch_pct": 3.53, "eta": "55:09:15", "max_grad_norm": 1.0, "loss": 1.092096209526062, "grad_norm": 0.578990638256073, "learning_rate": 8.804920913884007e-05}
+{"ts": "2025-12-26T20:18:32", "event": "train_log", "step": 504, "epoch": 0.21265822784810126, "progress_pct": 3.54, "epoch_pct": 3.54, "eta": "55:02:40", "max_grad_norm": 1.0, "loss": 1.0413451194763184, "grad_norm": 0.6597883701324463, "learning_rate": 8.840070298769771e-05}
+{"ts": "2025-12-26T20:18:48", "event": "train_log", "step": 506, "epoch": 0.21350210970464134, "progress_pct": 3.56, "epoch_pct": 3.56, "eta": "54:56:20", "max_grad_norm": 1.0, "loss": 1.0073142051696777, "grad_norm": 0.6660305261611938, "learning_rate": 8.875219683655536e-05}
+{"ts": "2025-12-26T20:19:03", "event": "train_log", "step": 508, "epoch": 0.21434599156118145, "progress_pct": 3.57, "epoch_pct": 3.57, "eta": "54:49:52", "max_grad_norm": 1.0, "loss": 1.0319768190383911, "grad_norm": 0.6283115148544312, "learning_rate": 8.910369068541301e-05}
+{"ts": "2025-12-26T20:19:20", "event": "train_log", "step": 510, "epoch": 0.21518987341772153, "progress_pct": 3.59, "epoch_pct": 3.59, "eta": "54:44:05", "max_grad_norm": 1.0, "loss": 1.0046353340148926, "grad_norm": 0.6257343292236328, "learning_rate": 8.945518453427065e-05}
+{"ts": "2025-12-26T20:19:36", "event": "train_log", "step": 512, "epoch": 0.2160337552742616, "progress_pct": 3.6, "epoch_pct": 3.6, "eta": "54:37:47", "max_grad_norm": 1.0, "loss": 0.9169099926948547, "grad_norm": 0.5530875325202942, "learning_rate": 8.980667838312829e-05}
+{"ts": "2025-12-26T20:19:52", "event": "train_log", "step": 514, "epoch": 0.2168776371308017, "progress_pct": 3.61, "epoch_pct": 3.61, "eta": "54:31:36", "max_grad_norm": 1.0, "loss": 0.9081505537033081, "grad_norm": 0.5369633436203003, "learning_rate": 9.015817223198594e-05}
+{"ts": "2025-12-26T20:20:07", "event": "train_log", "step": 516, "epoch": 0.21772151898734177, "progress_pct": 3.63, "epoch_pct": 3.63, "eta": "54:25:10", "max_grad_norm": 1.0, "loss": 0.9165045022964478, "grad_norm": 0.6618232131004333, "learning_rate": 9.05096660808436e-05}
+{"ts": "2025-12-26T20:20:22", "event": "train_log", "step": 518, "epoch": 0.21856540084388185, "progress_pct": 3.64, "epoch_pct": 3.64, "eta": "54:18:59", "max_grad_norm": 1.0, "loss": 0.91348797082901, "grad_norm": 0.600666344165802, "learning_rate": 9.086115992970123e-05}
+{"ts": "2025-12-26T20:20:38", "event": "train_log", "step": 520, "epoch": 0.21940928270042195, "progress_pct": 3.66, "epoch_pct": 3.66, "eta": "54:12:36", "max_grad_norm": 1.0, "loss": 1.006508469581604, "grad_norm": 0.5919831991195679, "learning_rate": 9.121265377855887e-05}
+{"ts": "2025-12-26T20:20:53", "event": "train_log", "step": 522, "epoch": 0.22025316455696203, "progress_pct": 3.67, "epoch_pct": 3.67, "eta": "54:06:25", "max_grad_norm": 1.0, "loss": 1.0013236999511719, "grad_norm": 0.688058614730835, "learning_rate": 9.156414762741654e-05}
+{"ts": "2025-12-26T20:21:09", "event": "train_log", "step": 524, "epoch": 0.2210970464135021, "progress_pct": 3.68, "epoch_pct": 3.68, "eta": "54:00:23", "max_grad_norm": 1.0, "loss": 1.0909923315048218, "grad_norm": 0.6721227765083313, "learning_rate": 9.191564147627418e-05}
+{"ts": "2025-12-26T20:21:25", "event": "train_log", "step": 526, "epoch": 0.2219409282700422, "progress_pct": 3.7, "epoch_pct": 3.7, "eta": "53:54:36", "max_grad_norm": 1.0, "loss": 1.0117096900939941, "grad_norm": 0.5987313389778137, "learning_rate": 9.226713532513181e-05}
+{"ts": "2025-12-26T20:21:40", "event": "train_log", "step": 528, "epoch": 0.22278481012658227, "progress_pct": 3.71, "epoch_pct": 3.71, "eta": "53:48:39", "max_grad_norm": 1.0, "loss": 1.0153647661209106, "grad_norm": 0.6191489696502686, "learning_rate": 9.261862917398945e-05}
+{"ts": "2025-12-26T20:21:56", "event": "train_log", "step": 530, "epoch": 0.22362869198312235, "progress_pct": 3.73, "epoch_pct": 3.73, "eta": "53:42:46", "max_grad_norm": 1.0, "loss": 0.9649755954742432, "grad_norm": 0.6821563243865967, "learning_rate": 9.29701230228471e-05}
+{"ts": "2025-12-26T20:22:12", "event": "train_log", "step": 532, "epoch": 0.22447257383966246, "progress_pct": 3.74, "epoch_pct": 3.74, "eta": "53:36:44", "max_grad_norm": 1.0, "loss": 0.8673232197761536, "grad_norm": 1.760398268699646, "learning_rate": 9.332161687170476e-05}
+{"ts": "2025-12-26T20:22:27", "event": "train_log", "step": 534, "epoch": 0.22531645569620254, "progress_pct": 3.76, "epoch_pct": 3.76, "eta": "53:30:57", "max_grad_norm": 1.0, "loss": 0.9942440986633301, "grad_norm": 0.6670058369636536, "learning_rate": 9.36731107205624e-05}
+{"ts": "2025-12-26T20:22:42", "event": "train_log", "step": 536, "epoch": 0.22616033755274262, "progress_pct": 3.77, "epoch_pct": 3.77, "eta": "53:24:51", "max_grad_norm": 1.0, "loss": 1.0364389419555664, "grad_norm": 0.7345916032791138, "learning_rate": 9.402460456942003e-05}
+{"ts": "2025-12-26T20:22:57", "event": "train_log", "step": 538, "epoch": 0.2270042194092827, "progress_pct": 3.78, "epoch_pct": 3.78, "eta": "53:18:50", "max_grad_norm": 1.0, "loss": 0.9314924478530884, "grad_norm": 0.5946128964424133, "learning_rate": 9.437609841827768e-05}
+{"ts": "2025-12-26T20:23:12", "event": "train_log", "step": 540, "epoch": 0.22784810126582278, "progress_pct": 3.8, "epoch_pct": 3.8, "eta": "53:12:58", "max_grad_norm": 1.0, "loss": 1.0694862604141235, "grad_norm": 0.5800848603248596, "learning_rate": 9.472759226713534e-05}
+{"ts": "2025-12-26T20:23:27", "event": "train_log", "step": 542, "epoch": 0.22869198312236286, "progress_pct": 3.81, "epoch_pct": 3.81, "eta": "53:06:38", "max_grad_norm": 1.0, "loss": 1.03531014919281, "grad_norm": 0.6712192893028259, "learning_rate": 9.507908611599297e-05}
+{"ts": "2025-12-26T20:23:42", "event": "train_log", "step": 544, "epoch": 0.22953586497890296, "progress_pct": 3.83, "epoch_pct": 3.83, "eta": "53:01:02", "max_grad_norm": 1.0, "loss": 0.9795235991477966, "grad_norm": 0.5641416311264038, "learning_rate": 9.543057996485063e-05}
+{"ts": "2025-12-26T20:23:58", "event": "train_log", "step": 546, "epoch": 0.23037974683544304, "progress_pct": 3.84, "epoch_pct": 3.84, "eta": "52:55:25", "max_grad_norm": 1.0, "loss": 0.9641494750976562, "grad_norm": 0.50412517786026, "learning_rate": 9.578207381370826e-05}
+{"ts": "2025-12-26T20:24:13", "event": "train_log", "step": 548, "epoch": 0.23122362869198312, "progress_pct": 3.85, "epoch_pct": 3.85, "eta": "52:49:35", "max_grad_norm": 1.0, "loss": 0.9375281929969788, "grad_norm": 0.579118549823761, "learning_rate": 9.61335676625659e-05}
+{"ts": "2025-12-26T20:24:28", "event": "train_log", "step": 550, "epoch": 0.2320675105485232, "progress_pct": 3.87, "epoch_pct": 3.87, "eta": "52:43:58", "max_grad_norm": 1.0, "loss": 0.9414046406745911, "grad_norm": 0.5888341665267944, "learning_rate": 9.648506151142355e-05}
+{"ts": "2025-12-26T20:24:43", "event": "train_log", "step": 552, "epoch": 0.23291139240506328, "progress_pct": 3.88, "epoch_pct": 3.88, "eta": "52:38:11", "max_grad_norm": 1.0, "loss": 0.9005617499351501, "grad_norm": 0.5595056414604187, "learning_rate": 9.68365553602812e-05}
+{"ts": "2025-12-26T20:24:58", "event": "train_log", "step": 554, "epoch": 0.23375527426160336, "progress_pct": 3.9, "epoch_pct": 3.9, "eta": "52:32:28", "max_grad_norm": 1.0, "loss": 1.0283968448638916, "grad_norm": 0.6605326533317566, "learning_rate": 9.718804920913884e-05}
+{"ts": "2025-12-26T20:25:14", "event": "train_log", "step": 556, "epoch": 0.23459915611814347, "progress_pct": 3.91, "epoch_pct": 3.91, "eta": "52:27:16", "max_grad_norm": 1.0, "loss": 1.0058249235153198, "grad_norm": 0.5657313466072083, "learning_rate": 9.753954305799648e-05}
+{"ts": "2025-12-26T20:25:30", "event": "train_log", "step": 558, "epoch": 0.23544303797468355, "progress_pct": 3.92, "epoch_pct": 3.92, "eta": "52:21:57", "max_grad_norm": 1.0, "loss": 0.9835494756698608, "grad_norm": 0.5433364510536194, "learning_rate": 9.789103690685413e-05}
+{"ts": "2025-12-26T20:25:44", "event": "train_log", "step": 560, "epoch": 0.23628691983122363, "progress_pct": 3.94, "epoch_pct": 3.94, "eta": "52:16:10", "max_grad_norm": 1.0, "loss": 1.054532527923584, "grad_norm": 0.6129802465438843, "learning_rate": 9.824253075571179e-05}
+{"ts": "2025-12-26T20:25:59", "event": "train_log", "step": 562, "epoch": 0.2371308016877637, "progress_pct": 3.95, "epoch_pct": 3.95, "eta": "52:10:39", "max_grad_norm": 1.0, "loss": 1.0240973234176636, "grad_norm": 0.6496239304542542, "learning_rate": 9.859402460456942e-05}
+{"ts": "2025-12-26T20:26:14", "event": "train_log", "step": 564, "epoch": 0.2379746835443038, "progress_pct": 3.97, "epoch_pct": 3.97, "eta": "52:04:58", "max_grad_norm": 1.0, "loss": 1.0229179859161377, "grad_norm": 0.6380873918533325, "learning_rate": 9.894551845342706e-05}
+{"ts": "2025-12-26T20:26:29", "event": "train_log", "step": 566, "epoch": 0.23881856540084387, "progress_pct": 3.98, "epoch_pct": 3.98, "eta": "51:59:24", "max_grad_norm": 1.0, "loss": 1.0111570358276367, "grad_norm": 0.6151993870735168, "learning_rate": 9.929701230228471e-05}
+{"ts": "2025-12-26T20:26:43", "event": "train_log", "step": 568, "epoch": 0.23966244725738398, "progress_pct": 3.99, "epoch_pct": 3.99, "eta": "51:53:51", "max_grad_norm": 1.0, "loss": 0.9450829029083252, "grad_norm": 0.5727584958076477, "learning_rate": 9.964850615114237e-05}
+{"ts": "2025-12-26T20:26:58", "event": "train_log", "step": 570, "epoch": 0.24050632911392406, "progress_pct": 4.01, "epoch_pct": 4.01, "eta": "51:48:12", "max_grad_norm": 1.0, "loss": 0.9800319075584412, "grad_norm": 0.6620725989341736, "learning_rate": 0.0001}
+{"ts": "2025-12-26T20:27:12", "event": "train_log", "step": 572, "epoch": 0.24135021097046414, "progress_pct": 4.02, "epoch_pct": 4.02, "eta": "51:42:42", "max_grad_norm": 1.0, "loss": 0.9757438898086548, "grad_norm": 0.6151163578033447, "learning_rate": 0.00010035149384885764}
+{"ts": "2025-12-26T20:27:27", "event": "train_log", "step": 574, "epoch": 0.24219409282700421, "progress_pct": 4.04, "epoch_pct": 4.04, "eta": "51:37:25", "max_grad_norm": 1.0, "loss": 0.9104921817779541, "grad_norm": 0.5672140717506409, "learning_rate": 0.0001007029876977153}
+{"ts": "2025-12-26T20:27:43", "event": "train_log", "step": 576, "epoch": 0.2430379746835443, "progress_pct": 4.05, "epoch_pct": 4.05, "eta": "51:32:22", "max_grad_norm": 1.0, "loss": 1.1027376651763916, "grad_norm": 0.5697256326675415, "learning_rate": 0.00010105448154657293}
+{"ts": "2025-12-26T20:27:58", "event": "train_log", "step": 578, "epoch": 0.2438818565400844, "progress_pct": 4.06, "epoch_pct": 4.06, "eta": "51:27:07", "max_grad_norm": 1.0, "loss": 1.0055404901504517, "grad_norm": 0.5590381622314453, "learning_rate": 0.00010140597539543057}
+{"ts": "2025-12-26T20:28:13", "event": "train_log", "step": 580, "epoch": 0.24472573839662448, "progress_pct": 4.08, "epoch_pct": 4.08, "eta": "51:21:54", "max_grad_norm": 1.0, "loss": 1.020835518836975, "grad_norm": 0.5518567562103271, "learning_rate": 0.00010175746924428824}
+{"ts": "2025-12-26T20:28:28", "event": "train_log", "step": 582, "epoch": 0.24556962025316456, "progress_pct": 4.09, "epoch_pct": 4.09, "eta": "51:16:45", "max_grad_norm": 1.0, "loss": 0.9528344869613647, "grad_norm": 0.6338496208190918, "learning_rate": 0.00010210896309314588}
+{"ts": "2025-12-26T20:28:42", "event": "train_log", "step": 584, "epoch": 0.24641350210970464, "progress_pct": 4.11, "epoch_pct": 4.11, "eta": "51:11:14", "max_grad_norm": 1.0, "loss": 1.0088670253753662, "grad_norm": 0.6497329473495483, "learning_rate": 0.00010246045694200353}
+{"ts": "2025-12-26T20:28:58", "event": "train_log", "step": 586, "epoch": 0.24725738396624472, "progress_pct": 4.12, "epoch_pct": 4.12, "eta": "51:06:18", "max_grad_norm": 1.0, "loss": 0.9961200952529907, "grad_norm": 0.49888095259666443, "learning_rate": 0.00010281195079086117}
+{"ts": "2025-12-26T20:29:12", "event": "train_log", "step": 588, "epoch": 0.2481012658227848, "progress_pct": 4.14, "epoch_pct": 4.14, "eta": "51:01:08", "max_grad_norm": 1.0, "loss": 0.9635610580444336, "grad_norm": 0.5680158734321594, "learning_rate": 0.0001031634446397188}
+{"ts": "2025-12-26T20:29:27", "event": "train_log", "step": 590, "epoch": 0.2489451476793249, "progress_pct": 4.15, "epoch_pct": 4.15, "eta": "50:55:56", "max_grad_norm": 1.0, "loss": 0.9392287135124207, "grad_norm": 0.658168375492096, "learning_rate": 0.00010351493848857646}
+{"ts": "2025-12-26T20:29:42", "event": "train_log", "step": 592, "epoch": 0.249789029535865, "progress_pct": 4.16, "epoch_pct": 4.16, "eta": "50:50:55", "max_grad_norm": 1.0, "loss": 0.9600516557693481, "grad_norm": 0.618262767791748, "learning_rate": 0.0001038664323374341}
+{"ts": "2025-12-26T20:29:56", "event": "train_log", "step": 594, "epoch": 0.25063291139240507, "progress_pct": 4.18, "epoch_pct": 4.18, "eta": "50:45:40", "max_grad_norm": 1.0, "loss": 1.005476713180542, "grad_norm": 0.6003909111022949, "learning_rate": 0.00010421792618629173}
+{"ts": "2025-12-26T20:30:11", "event": "train_log", "step": 596, "epoch": 0.2514767932489452, "progress_pct": 4.19, "epoch_pct": 4.19, "eta": "50:40:42", "max_grad_norm": 1.0, "loss": 0.9523017406463623, "grad_norm": 0.5437078475952148, "learning_rate": 0.00010456942003514938}
+{"ts": "2025-12-26T20:30:27", "event": "train_log", "step": 598, "epoch": 0.2523206751054852, "progress_pct": 4.21, "epoch_pct": 4.21, "eta": "50:35:58", "max_grad_norm": 1.0, "loss": 0.9526668787002563, "grad_norm": 0.5524541735649109, "learning_rate": 0.00010492091388400705}
+{"ts": "2025-12-26T20:30:42", "event": "train_log", "step": 600, "epoch": 0.25316455696202533, "progress_pct": 4.22, "epoch_pct": 4.22, "eta": "50:31:16", "max_grad_norm": 1.0, "loss": 1.019660472869873, "grad_norm": 0.679504930973053, "learning_rate": 0.00010527240773286469}
+{"ts": "2025-12-26T20:42:00", "event": "train_log", "step": 600, "epoch": 0.25316455696202533, "progress_pct": 4.22, "epoch_pct": 4.22, "eta": "54:47:45", "max_grad_norm": 1.0, "eval_loss": 1.0193854570388794, "eval_runtime": 677.9523, "eval_samples_per_second": 3.108, "eval_steps_per_second": 3.108}
+{"ts": "2025-12-26T20:42:16", "event": "train_log", "step": 602, "epoch": 0.2540084388185654, "progress_pct": 4.23, "epoch_pct": 4.23, "eta": "54:42:19", "max_grad_norm": 1.0, "loss": 0.9910882711410522, "grad_norm": 0.5646136999130249, "learning_rate": 0.00010562390158172233}
+{"ts": "2025-12-26T20:42:33", "event": "train_log", "step": 604, "epoch": 0.2548523206751055, "progress_pct": 4.25, "epoch_pct": 4.25, "eta": "54:37:11", "max_grad_norm": 1.0, "loss": 0.9616432785987854, "grad_norm": 0.5238093137741089, "learning_rate": 0.00010597539543057998}
+{"ts": "2025-12-26T20:42:47", "event": "train_log", "step": 606, "epoch": 0.25569620253164554, "progress_pct": 4.26, "epoch_pct": 4.26, "eta": "54:31:19", "max_grad_norm": 1.0, "loss": 1.0078275203704834, "grad_norm": 0.7483857274055481, "learning_rate": 0.00010632688927943762}
+{"ts": "2025-12-26T20:43:03", "event": "train_log", "step": 608, "epoch": 0.25654008438818565, "progress_pct": 4.28, "epoch_pct": 4.28, "eta": "54:26:10", "max_grad_norm": 1.0, "loss": 0.9827103018760681, "grad_norm": 0.578948974609375, "learning_rate": 0.00010667838312829525}
+{"ts": "2025-12-26T20:43:20", "event": "train_log", "step": 610, "epoch": 0.25738396624472576, "progress_pct": 4.29, "epoch_pct": 4.29, "eta": "54:21:03", "max_grad_norm": 1.0, "loss": 1.0423277616500854, "grad_norm": 0.5525906085968018, "learning_rate": 0.00010702987697715289}
+{"ts": "2025-12-26T20:43:36", "event": "train_log", "step": 612, "epoch": 0.2582278481012658, "progress_pct": 4.3, "epoch_pct": 4.3, "eta": "54:15:54", "max_grad_norm": 1.0, "loss": 0.9561693072319031, "grad_norm": 0.6721326112747192, "learning_rate": 0.00010738137082601054}
+{"ts": "2025-12-26T20:43:51", "event": "train_log", "step": 614, "epoch": 0.2590717299578059, "progress_pct": 4.32, "epoch_pct": 4.32, "eta": "54:10:30", "max_grad_norm": 1.0, "loss": 0.9602992534637451, "grad_norm": 0.5701051354408264, "learning_rate": 0.00010773286467486821}
+{"ts": "2025-12-26T20:44:06", "event": "train_log", "step": 616, "epoch": 0.25991561181434597, "progress_pct": 4.33, "epoch_pct": 4.33, "eta": "54:04:48", "max_grad_norm": 1.0, "loss": 1.1422650814056396, "grad_norm": 0.6349860429763794, "learning_rate": 0.00010808435852372585}
+{"ts": "2025-12-26T20:44:21", "event": "train_log", "step": 618, "epoch": 0.2607594936708861, "progress_pct": 4.35, "epoch_pct": 4.35, "eta": "53:59:30", "max_grad_norm": 1.0, "loss": 0.9762773513793945, "grad_norm": 0.5496085286140442, "learning_rate": 0.00010843585237258349}
+{"ts": "2025-12-26T20:44:37", "event": "train_log", "step": 620, "epoch": 0.2616033755274262, "progress_pct": 4.36, "epoch_pct": 4.36, "eta": "53:54:10", "max_grad_norm": 1.0, "loss": 1.0133616924285889, "grad_norm": 0.6080722808837891, "learning_rate": 0.00010878734622144114}
+{"ts": "2025-12-26T20:44:53", "event": "train_log", "step": 622, "epoch": 0.26244725738396624, "progress_pct": 4.37, "epoch_pct": 4.37, "eta": "53:49:12", "max_grad_norm": 1.0, "loss": 0.9385587573051453, "grad_norm": 0.5450218915939331, "learning_rate": 0.00010913884007029878}
+{"ts": "2025-12-26T20:45:10", "event": "train_log", "step": 624, "epoch": 0.26329113924050634, "progress_pct": 4.39, "epoch_pct": 4.39, "eta": "53:44:42", "max_grad_norm": 1.0, "loss": 0.9359989762306213, "grad_norm": 0.592106819152832, "learning_rate": 0.00010949033391915641}
+{"ts": "2025-12-26T20:45:25", "event": "train_log", "step": 626, "epoch": 0.2641350210970464, "progress_pct": 4.4, "epoch_pct": 4.4, "eta": "53:39:18", "max_grad_norm": 1.0, "loss": 1.0266027450561523, "grad_norm": 0.6449427604675293, "learning_rate": 0.00010984182776801407}
+{"ts": "2025-12-26T20:45:41", "event": "train_log", "step": 628, "epoch": 0.2649789029535865, "progress_pct": 4.42, "epoch_pct": 4.42, "eta": "53:34:30", "max_grad_norm": 1.0, "loss": 0.9303187131881714, "grad_norm": 0.538299560546875, "learning_rate": 0.0001101933216168717}
+{"ts": "2025-12-26T20:45:57", "event": "train_log", "step": 630, "epoch": 0.26582278481012656, "progress_pct": 4.43, "epoch_pct": 4.43, "eta": "53:29:35", "max_grad_norm": 1.0, "loss": 0.9368857145309448, "grad_norm": 0.546316921710968, "learning_rate": 0.00011054481546572934}
+{"ts": "2025-12-26T20:46:13", "event": "train_log", "step": 632, "epoch": 0.26666666666666666, "progress_pct": 4.44, "epoch_pct": 4.44, "eta": "53:24:21", "max_grad_norm": 1.0, "loss": 0.9573145508766174, "grad_norm": 0.5818730592727661, "learning_rate": 0.00011089630931458701}
+{"ts": "2025-12-26T20:46:28", "event": "train_log", "step": 634, "epoch": 0.26751054852320677, "progress_pct": 4.46, "epoch_pct": 4.46, "eta": "53:19:26", "max_grad_norm": 1.0, "loss": 0.9345449805259705, "grad_norm": 0.5958262085914612, "learning_rate": 0.00011124780316344465}
+{"ts": "2025-12-26T20:46:43", "event": "train_log", "step": 636, "epoch": 0.2683544303797468, "progress_pct": 4.47, "epoch_pct": 4.47, "eta": "53:14:04", "max_grad_norm": 1.0, "loss": 1.0906590223312378, "grad_norm": 0.6259077787399292, "learning_rate": 0.0001115992970123023}
+{"ts": "2025-12-26T20:46:58", "event": "train_log", "step": 638, "epoch": 0.26919831223628693, "progress_pct": 4.49, "epoch_pct": 4.49, "eta": "53:09:05", "max_grad_norm": 1.0, "loss": 1.0757447481155396, "grad_norm": 0.589672863483429, "learning_rate": 0.00011195079086115994}
+{"ts": "2025-12-26T20:47:14", "event": "train_log", "step": 640, "epoch": 0.270042194092827, "progress_pct": 4.5, "epoch_pct": 4.5, "eta": "53:04:07", "max_grad_norm": 1.0, "loss": 0.9310855269432068, "grad_norm": 0.5714080333709717, "learning_rate": 0.00011230228471001758}
+{"ts": "2025-12-26T20:47:29", "event": "train_log", "step": 642, "epoch": 0.2708860759493671, "progress_pct": 4.51, "epoch_pct": 4.51, "eta": "52:59:05", "max_grad_norm": 1.0, "loss": 0.9276360273361206, "grad_norm": 0.45342639088630676, "learning_rate": 0.00011265377855887523}
+{"ts": "2025-12-26T20:47:44", "event": "train_log", "step": 644, "epoch": 0.2717299578059072, "progress_pct": 4.53, "epoch_pct": 4.53, "eta": "52:54:02", "max_grad_norm": 1.0, "loss": 1.084719181060791, "grad_norm": 0.6386750340461731, "learning_rate": 0.00011300527240773287}
+{"ts": "2025-12-26T20:48:01", "event": "train_log", "step": 646, "epoch": 0.27257383966244725, "progress_pct": 4.54, "epoch_pct": 4.54, "eta": "52:49:30", "max_grad_norm": 1.0, "loss": 0.9763918519020081, "grad_norm": 0.6446163654327393, "learning_rate": 0.0001133567662565905}
+{"ts": "2025-12-26T20:48:15", "event": "train_log", "step": 648, "epoch": 0.27341772151898736, "progress_pct": 4.56, "epoch_pct": 4.56, "eta": "52:44:16", "max_grad_norm": 1.0, "loss": 0.9517921805381775, "grad_norm": 0.5925686359405518, "learning_rate": 0.00011370826010544816}
+{"ts": "2025-12-26T20:48:31", "event": "train_log", "step": 650, "epoch": 0.2742616033755274, "progress_pct": 4.57, "epoch_pct": 4.57, "eta": "52:39:42", "max_grad_norm": 1.0, "loss": 1.0587927103042603, "grad_norm": 0.5399773716926575, "learning_rate": 0.00011405975395430582}
+{"ts": "2025-12-26T20:48:46", "event": "train_log", "step": 652, "epoch": 0.2751054852320675, "progress_pct": 4.59, "epoch_pct": 4.59, "eta": "52:34:44", "max_grad_norm": 1.0, "loss": 0.883341908454895, "grad_norm": 0.5872456431388855, "learning_rate": 0.00011441124780316346}
+{"ts": "2025-12-26T20:49:04", "event": "train_log", "step": 654, "epoch": 0.2759493670886076, "progress_pct": 4.6, "epoch_pct": 4.6, "eta": "52:30:39", "max_grad_norm": 1.0, "loss": 1.0306891202926636, "grad_norm": 0.5574564337730408, "learning_rate": 0.0001147627416520211}
+{"ts": "2025-12-26T20:49:21", "event": "train_log", "step": 656, "epoch": 0.2767932489451477, "progress_pct": 4.61, "epoch_pct": 4.61, "eta": "52:26:30", "max_grad_norm": 1.0, "loss": 0.8814032077789307, "grad_norm": 0.47789013385772705, "learning_rate": 0.00011511423550087874}
+{"ts": "2025-12-26T20:49:36", "event": "train_log", "step": 658, "epoch": 0.2776371308016878, "progress_pct": 4.63, "epoch_pct": 4.63, "eta": "52:21:51", "max_grad_norm": 1.0, "loss": 0.9460552334785461, "grad_norm": 0.5565530061721802, "learning_rate": 0.00011546572934973639}
+{"ts": "2025-12-26T20:49:51", "event": "train_log", "step": 660, "epoch": 0.27848101265822783, "progress_pct": 4.64, "epoch_pct": 4.64, "eta": "52:17:00", "max_grad_norm": 1.0, "loss": 0.9475110769271851, "grad_norm": 0.5299761295318604, "learning_rate": 0.00011581722319859403}
+{"ts": "2025-12-26T20:50:07", "event": "train_log", "step": 662, "epoch": 0.27932489451476794, "progress_pct": 4.66, "epoch_pct": 4.66, "eta": "52:12:15", "max_grad_norm": 1.0, "loss": 1.0630913972854614, "grad_norm": 0.6503344178199768, "learning_rate": 0.00011616871704745166}
+{"ts": "2025-12-26T20:50:23", "event": "train_log", "step": 664, "epoch": 0.280168776371308, "progress_pct": 4.67, "epoch_pct": 4.67, "eta": "52:07:53", "max_grad_norm": 1.0, "loss": 0.9389138221740723, "grad_norm": 0.5794585943222046, "learning_rate": 0.00011652021089630932}
+{"ts": "2025-12-26T20:50:39", "event": "train_log", "step": 666, "epoch": 0.2810126582278481, "progress_pct": 4.68, "epoch_pct": 4.68, "eta": "52:03:31", "max_grad_norm": 1.0, "loss": 0.8934136033058167, "grad_norm": 0.5762867331504822, "learning_rate": 0.00011687170474516695}
+{"ts": "2025-12-26T20:50:55", "event": "train_log", "step": 668, "epoch": 0.2818565400843882, "progress_pct": 4.7, "epoch_pct": 4.7, "eta": "51:59:04", "max_grad_norm": 1.0, "loss": 1.1072614192962646, "grad_norm": 0.6565435528755188, "learning_rate": 0.00011722319859402462}
+{"ts": "2025-12-26T20:51:10", "event": "train_log", "step": 670, "epoch": 0.28270042194092826, "progress_pct": 4.71, "epoch_pct": 4.71, "eta": "51:54:16", "max_grad_norm": 1.0, "loss": 1.0501434803009033, "grad_norm": 0.5819830298423767, "learning_rate": 0.00011757469244288226}
+{"ts": "2025-12-26T20:51:25", "event": "train_log", "step": 672, "epoch": 0.28354430379746837, "progress_pct": 4.73, "epoch_pct": 4.73, "eta": "51:49:35", "max_grad_norm": 1.0, "loss": 0.9880793690681458, "grad_norm": 0.6071487069129944, "learning_rate": 0.00011792618629173991}
+{"ts": "2025-12-26T20:51:40", "event": "train_log", "step": 674, "epoch": 0.2843881856540084, "progress_pct": 4.74, "epoch_pct": 4.74, "eta": "51:44:55", "max_grad_norm": 1.0, "loss": 0.9670693874359131, "grad_norm": 0.5765058398246765, "learning_rate": 0.00011827768014059755}
+{"ts": "2025-12-26T20:51:55", "event": "train_log", "step": 676, "epoch": 0.2852320675105485, "progress_pct": 4.75, "epoch_pct": 4.75, "eta": "51:40:28", "max_grad_norm": 1.0, "loss": 0.9602360725402832, "grad_norm": 0.5245351791381836, "learning_rate": 0.00011862917398945519}
+{"ts": "2025-12-26T20:52:10", "event": "train_log", "step": 678, "epoch": 0.28607594936708863, "progress_pct": 4.77, "epoch_pct": 4.77, "eta": "51:35:46", "max_grad_norm": 1.0, "loss": 0.9684560894966125, "grad_norm": 0.6189922094345093, "learning_rate": 0.00011898066783831282}
+{"ts": "2025-12-26T20:52:27", "event": "train_log", "step": 680, "epoch": 0.2869198312236287, "progress_pct": 4.78, "epoch_pct": 4.78, "eta": "51:31:50", "max_grad_norm": 1.0, "loss": 0.9465792775154114, "grad_norm": 0.6138690710067749, "learning_rate": 0.00011933216168717048}
+{"ts": "2025-12-26T20:52:45", "event": "train_log", "step": 682, "epoch": 0.2877637130801688, "progress_pct": 4.8, "epoch_pct": 4.8, "eta": "51:28:15", "max_grad_norm": 1.0, "loss": 0.8495944738388062, "grad_norm": 0.5371595621109009, "learning_rate": 0.00011968365553602812}
+{"ts": "2025-12-26T20:53:02", "event": "train_log", "step": 684, "epoch": 0.28860759493670884, "progress_pct": 4.81, "epoch_pct": 4.81, "eta": "51:24:18", "max_grad_norm": 1.0, "loss": 0.9663267135620117, "grad_norm": 0.5549944639205933, "learning_rate": 0.00012003514938488578}
+{"ts": "2025-12-26T20:53:18", "event": "train_log", "step": 686, "epoch": 0.28945147679324895, "progress_pct": 4.82, "epoch_pct": 4.82, "eta": "51:20:19", "max_grad_norm": 1.0, "loss": 0.9736058712005615, "grad_norm": 0.6484189033508301, "learning_rate": 0.00012038664323374342}
+{"ts": "2025-12-26T20:53:33", "event": "train_log", "step": 688, "epoch": 0.290295358649789, "progress_pct": 4.84, "epoch_pct": 4.84, "eta": "51:15:56", "max_grad_norm": 1.0, "loss": 1.0591845512390137, "grad_norm": 0.540351390838623, "learning_rate": 0.00012073813708260107}
+{"ts": "2025-12-26T20:53:49", "event": "train_log", "step": 690, "epoch": 0.2911392405063291, "progress_pct": 4.85, "epoch_pct": 4.85, "eta": "51:11:39", "max_grad_norm": 1.0, "loss": 0.944908618927002, "grad_norm": 0.5657922029495239, "learning_rate": 0.00012108963093145871}
+{"ts": "2025-12-26T20:54:03", "event": "train_log", "step": 692, "epoch": 0.2919831223628692, "progress_pct": 4.87, "epoch_pct": 4.87, "eta": "51:07:00", "max_grad_norm": 1.0, "loss": 1.0018219947814941, "grad_norm": 0.6040505170822144, "learning_rate": 0.00012144112478031635}
+{"ts": "2025-12-26T20:54:19", "event": "train_log", "step": 694, "epoch": 0.29282700421940927, "progress_pct": 4.88, "epoch_pct": 4.88, "eta": "51:02:55", "max_grad_norm": 1.0, "loss": 1.0351502895355225, "grad_norm": 0.5435477495193481, "learning_rate": 0.000121792618629174}
+{"ts": "2025-12-26T20:54:35", "event": "train_log", "step": 696, "epoch": 0.2936708860759494, "progress_pct": 4.89, "epoch_pct": 4.89, "eta": "50:58:37", "max_grad_norm": 1.0, "loss": 0.9935672283172607, "grad_norm": 0.5712518692016602, "learning_rate": 0.00012214411247803164}
+{"ts": "2025-12-26T20:54:49", "event": "train_log", "step": 698, "epoch": 0.29451476793248943, "progress_pct": 4.91, "epoch_pct": 4.91, "eta": "50:54:05", "max_grad_norm": 1.0, "loss": 1.0165108442306519, "grad_norm": 0.6138222813606262, "learning_rate": 0.00012249560632688928}
+{"ts": "2025-12-26T20:55:05", "event": "train_log", "step": 700, "epoch": 0.29535864978902954, "progress_pct": 4.92, "epoch_pct": 4.92, "eta": "50:49:50", "max_grad_norm": 1.0, "loss": 0.9334425926208496, "grad_norm": 0.4495212435722351, "learning_rate": 0.00012284710017574691}
+{"ts": "2025-12-26T21:06:13", "event": "train_log", "step": 700, "epoch": 0.29535864978902954, "progress_pct": 4.92, "epoch_pct": 4.92, "eta": "54:25:04", "max_grad_norm": 1.0, "eval_loss": 0.996929407119751, "eval_runtime": 668.6398, "eval_samples_per_second": 3.151, "eval_steps_per_second": 3.151}
+{"ts": "2025-12-26T21:06:28", "event": "train_log", "step": 702, "epoch": 0.29620253164556964, "progress_pct": 4.94, "epoch_pct": 4.94, "eta": "54:20:05", "max_grad_norm": 1.0, "loss": 1.0516537427902222, "grad_norm": 0.5321539044380188, "learning_rate": 0.00012319859402460458}
+{"ts": "2025-12-26T21:06:43", "event": "train_log", "step": 704, "epoch": 0.2970464135021097, "progress_pct": 4.95, "epoch_pct": 4.95, "eta": "54:15:07", "max_grad_norm": 1.0, "loss": 0.9387198686599731, "grad_norm": 0.5716516971588135, "learning_rate": 0.00012355008787346222}
+{"ts": "2025-12-26T21:06:57", "event": "train_log", "step": 706, "epoch": 0.2978902953586498, "progress_pct": 4.96, "epoch_pct": 4.96, "eta": "54:09:58", "max_grad_norm": 1.0, "loss": 0.99737948179245, "grad_norm": 0.5617920160293579, "learning_rate": 0.00012390158172231988}
+{"ts": "2025-12-26T21:07:12", "event": "train_log", "step": 708, "epoch": 0.29873417721518986, "progress_pct": 4.98, "epoch_pct": 4.98, "eta": "54:04:59", "max_grad_norm": 1.0, "loss": 0.9955025911331177, "grad_norm": 0.4922899007797241, "learning_rate": 0.00012425307557117752}
+{"ts": "2025-12-26T21:07:27", "event": "train_log", "step": 710, "epoch": 0.29957805907172996, "progress_pct": 4.99, "epoch_pct": 4.99, "eta": "54:00:09", "max_grad_norm": 1.0, "loss": 0.9124280214309692, "grad_norm": 0.543501615524292, "learning_rate": 0.00012460456942003516}
+{"ts": "2025-12-26T21:07:43", "event": "train_log", "step": 712, "epoch": 0.30042194092827, "progress_pct": 5.01, "epoch_pct": 5.01, "eta": "53:55:30", "max_grad_norm": 1.0, "loss": 0.9820216298103333, "grad_norm": 0.49590054154396057, "learning_rate": 0.0001249560632688928}
+{"ts": "2025-12-26T21:07:57", "event": "train_log", "step": 714, "epoch": 0.3012658227848101, "progress_pct": 5.02, "epoch_pct": 5.02, "eta": "53:50:31", "max_grad_norm": 1.0, "loss": 1.0152074098587036, "grad_norm": 0.5984305739402771, "learning_rate": 0.00012530755711775044}
+{"ts": "2025-12-26T21:08:13", "event": "train_log", "step": 716, "epoch": 0.30210970464135023, "progress_pct": 5.04, "epoch_pct": 5.04, "eta": "53:46:08", "max_grad_norm": 1.0, "loss": 1.0577725172042847, "grad_norm": 0.9343504905700684, "learning_rate": 0.00012565905096660807}
+{"ts": "2025-12-26T21:08:29", "event": "train_log", "step": 718, "epoch": 0.3029535864978903, "progress_pct": 5.05, "epoch_pct": 5.05, "eta": "53:41:40", "max_grad_norm": 1.0, "loss": 0.9830358028411865, "grad_norm": 0.5118702054023743, "learning_rate": 0.0001260105448154657}
+{"ts": "2025-12-26T21:08:46", "event": "train_log", "step": 720, "epoch": 0.3037974683544304, "progress_pct": 5.06, "epoch_pct": 5.06, "eta": "53:37:36", "max_grad_norm": 1.0, "loss": 0.9466043710708618, "grad_norm": 0.4940392076969147, "learning_rate": 0.00012636203866432338}
+{"ts": "2025-12-26T21:09:01", "event": "train_log", "step": 722, "epoch": 0.30464135021097044, "progress_pct": 5.08, "epoch_pct": 5.08, "eta": "53:32:47", "max_grad_norm": 1.0, "loss": 1.015270709991455, "grad_norm": 0.5965693593025208, "learning_rate": 0.00012671353251318104}
+{"ts": "2025-12-26T21:09:18", "event": "train_log", "step": 724, "epoch": 0.30548523206751055, "progress_pct": 5.09, "epoch_pct": 5.09, "eta": "53:28:39", "max_grad_norm": 1.0, "loss": 0.9703927636146545, "grad_norm": 0.5020529627799988, "learning_rate": 0.00012706502636203868}
+{"ts": "2025-12-26T21:09:33", "event": "train_log", "step": 726, "epoch": 0.30632911392405066, "progress_pct": 5.11, "epoch_pct": 5.11, "eta": "53:23:58", "max_grad_norm": 1.0, "loss": 1.0255526304244995, "grad_norm": 0.6067010164260864, "learning_rate": 0.00012741652021089632}
+{"ts": "2025-12-26T21:09:48", "event": "train_log", "step": 728, "epoch": 0.3071729957805907, "progress_pct": 5.12, "epoch_pct": 5.12, "eta": "53:19:18", "max_grad_norm": 1.0, "loss": 0.9335633516311646, "grad_norm": 0.5931884050369263, "learning_rate": 0.00012776801405975396}
+{"ts": "2025-12-26T21:10:03", "event": "train_log", "step": 730, "epoch": 0.3080168776371308, "progress_pct": 5.13, "epoch_pct": 5.13, "eta": "53:14:39", "max_grad_norm": 1.0, "loss": 1.0921578407287598, "grad_norm": 0.5938752293586731, "learning_rate": 0.0001281195079086116}
+{"ts": "2025-12-26T21:10:19", "event": "train_log", "step": 732, "epoch": 0.30886075949367087, "progress_pct": 5.15, "epoch_pct": 5.15, "eta": "53:10:26", "max_grad_norm": 1.0, "loss": 0.963066041469574, "grad_norm": 0.49728086590766907, "learning_rate": 0.00012847100175746923}
+{"ts": "2025-12-26T21:10:34", "event": "train_log", "step": 734, "epoch": 0.309704641350211, "progress_pct": 5.16, "epoch_pct": 5.16, "eta": "53:05:50", "max_grad_norm": 1.0, "loss": 0.9513075351715088, "grad_norm": 0.5452080965042114, "learning_rate": 0.0001288224956063269}
+{"ts": "2025-12-26T21:10:50", "event": "train_log", "step": 736, "epoch": 0.3105485232067511, "progress_pct": 5.18, "epoch_pct": 5.18, "eta": "53:01:38", "max_grad_norm": 1.0, "loss": 0.8576077222824097, "grad_norm": 0.5497731566429138, "learning_rate": 0.00012917398945518454}
+{"ts": "2025-12-26T21:11:04", "event": "train_log", "step": 738, "epoch": 0.31139240506329113, "progress_pct": 5.19, "epoch_pct": 5.19, "eta": "52:56:46", "max_grad_norm": 1.0, "loss": 0.9542577862739563, "grad_norm": 0.5580397248268127, "learning_rate": 0.0001295254833040422}
+{"ts": "2025-12-26T21:11:18", "event": "train_log", "step": 740, "epoch": 0.31223628691983124, "progress_pct": 5.2, "epoch_pct": 5.2, "eta": "52:52:06", "max_grad_norm": 1.0, "loss": 0.8992732167243958, "grad_norm": 0.5890427827835083, "learning_rate": 0.00012987697715289984}
+{"ts": "2025-12-26T21:11:33", "event": "train_log", "step": 742, "epoch": 0.3130801687763713, "progress_pct": 5.22, "epoch_pct": 5.22, "eta": "52:47:38", "max_grad_norm": 1.0, "loss": 1.0322896242141724, "grad_norm": 0.5942965745925903, "learning_rate": 0.00013022847100175748}
+{"ts": "2025-12-26T21:11:49", "event": "train_log", "step": 744, "epoch": 0.3139240506329114, "progress_pct": 5.23, "epoch_pct": 5.23, "eta": "52:43:20", "max_grad_norm": 1.0, "loss": 0.9217103719711304, "grad_norm": 0.6341713070869446, "learning_rate": 0.00013057996485061512}
+{"ts": "2025-12-26T21:12:04", "event": "train_log", "step": 746, "epoch": 0.31476793248945145, "progress_pct": 5.25, "epoch_pct": 5.25, "eta": "52:38:57", "max_grad_norm": 1.0, "loss": 0.951789915561676, "grad_norm": 0.5294105410575867, "learning_rate": 0.00013093145869947276}
+{"ts": "2025-12-26T21:12:19", "event": "train_log", "step": 748, "epoch": 0.31561181434599156, "progress_pct": 5.26, "epoch_pct": 5.26, "eta": "52:34:32", "max_grad_norm": 1.0, "loss": 0.9459875226020813, "grad_norm": 0.6372058391571045, "learning_rate": 0.0001312829525483304}
+{"ts": "2025-12-26T21:12:34", "event": "train_log", "step": 750, "epoch": 0.31645569620253167, "progress_pct": 5.27, "epoch_pct": 5.27, "eta": "52:30:09", "max_grad_norm": 1.0, "loss": 0.9626097679138184, "grad_norm": 0.5979796648025513, "learning_rate": 0.00013163444639718806}
+{"ts": "2025-12-26T21:12:50", "event": "train_log", "step": 752, "epoch": 0.3172995780590717, "progress_pct": 5.29, "epoch_pct": 5.29, "eta": "52:26:08", "max_grad_norm": 1.0, "loss": 1.0261781215667725, "grad_norm": 0.5682399868965149, "learning_rate": 0.0001319859402460457}
+{"ts": "2025-12-26T21:13:06", "event": "train_log", "step": 754, "epoch": 0.3181434599156118, "progress_pct": 5.3, "epoch_pct": 5.3, "eta": "52:22:05", "max_grad_norm": 1.0, "loss": 0.9319828152656555, "grad_norm": 0.5349125266075134, "learning_rate": 0.00013233743409490336}
+{"ts": "2025-12-26T21:13:22", "event": "train_log", "step": 756, "epoch": 0.3189873417721519, "progress_pct": 5.32, "epoch_pct": 5.32, "eta": "52:17:54", "max_grad_norm": 1.0, "loss": 0.9216550588607788, "grad_norm": 0.6093934178352356, "learning_rate": 0.000132688927943761}
+{"ts": "2025-12-26T21:13:37", "event": "train_log", "step": 758, "epoch": 0.319831223628692, "progress_pct": 5.33, "epoch_pct": 5.33, "eta": "52:13:46", "max_grad_norm": 1.0, "loss": 0.901739776134491, "grad_norm": 0.5188612341880798, "learning_rate": 0.00013304042179261864}
+{"ts": "2025-12-26T21:13:52", "event": "train_log", "step": 760, "epoch": 0.3206751054852321, "progress_pct": 5.34, "epoch_pct": 5.34, "eta": "52:09:30", "max_grad_norm": 1.0, "loss": 1.0362589359283447, "grad_norm": 0.5877130627632141, "learning_rate": 0.00013339191564147628}
+{"ts": "2025-12-26T21:14:08", "event": "train_log", "step": 762, "epoch": 0.32151898734177214, "progress_pct": 5.36, "epoch_pct": 5.36, "eta": "52:05:22", "max_grad_norm": 1.0, "loss": 0.8787116408348083, "grad_norm": 0.5542771816253662, "learning_rate": 0.00013374340949033392}
+{"ts": "2025-12-26T21:14:24", "event": "train_log", "step": 764, "epoch": 0.32236286919831225, "progress_pct": 5.37, "epoch_pct": 5.37, "eta": "52:01:24", "max_grad_norm": 1.0, "loss": 0.9237037301063538, "grad_norm": 0.5084902048110962, "learning_rate": 0.00013409490333919156}
+{"ts": "2025-12-26T21:14:40", "event": "train_log", "step": 766, "epoch": 0.3232067510548523, "progress_pct": 5.39, "epoch_pct": 5.39, "eta": "51:57:27", "max_grad_norm": 1.0, "loss": 1.0150731801986694, "grad_norm": 0.5461528301239014, "learning_rate": 0.00013444639718804922}
+{"ts": "2025-12-26T21:14:56", "event": "train_log", "step": 768, "epoch": 0.3240506329113924, "progress_pct": 5.4, "epoch_pct": 5.4, "eta": "51:53:36", "max_grad_norm": 1.0, "loss": 0.8985214829444885, "grad_norm": 0.53483647108078, "learning_rate": 0.00013479789103690686}
+{"ts": "2025-12-26T21:15:11", "event": "train_log", "step": 770, "epoch": 0.32489451476793246, "progress_pct": 5.41, "epoch_pct": 5.41, "eta": "51:49:35", "max_grad_norm": 1.0, "loss": 1.0225775241851807, "grad_norm": 0.5580531358718872, "learning_rate": 0.0001351493848857645}
+{"ts": "2025-12-26T21:15:27", "event": "train_log", "step": 772, "epoch": 0.32573839662447257, "progress_pct": 5.43, "epoch_pct": 5.43, "eta": "51:45:42", "max_grad_norm": 1.0, "loss": 0.9571293592453003, "grad_norm": 0.5203377604484558, "learning_rate": 0.00013550087873462216}
+{"ts": "2025-12-26T21:15:43", "event": "train_log", "step": 774, "epoch": 0.3265822784810127, "progress_pct": 5.44, "epoch_pct": 5.44, "eta": "51:41:44", "max_grad_norm": 1.0, "loss": 1.0468909740447998, "grad_norm": 0.5049671530723572, "learning_rate": 0.0001358523725834798}
+{"ts": "2025-12-26T21:16:00", "event": "train_log", "step": 776, "epoch": 0.32742616033755273, "progress_pct": 5.46, "epoch_pct": 5.46, "eta": "51:38:13", "max_grad_norm": 1.0, "loss": 0.7743215560913086, "grad_norm": 0.4723063111305237, "learning_rate": 0.00013620386643233744}
+{"ts": "2025-12-26T21:16:16", "event": "train_log", "step": 778, "epoch": 0.32827004219409284, "progress_pct": 5.47, "epoch_pct": 5.47, "eta": "51:34:17", "max_grad_norm": 1.0, "loss": 1.021510362625122, "grad_norm": 0.6310980916023254, "learning_rate": 0.00013655536028119508}
+{"ts": "2025-12-26T21:16:31", "event": "train_log", "step": 780, "epoch": 0.3291139240506329, "progress_pct": 5.49, "epoch_pct": 5.49, "eta": "51:30:25", "max_grad_norm": 1.0, "loss": 0.9134382605552673, "grad_norm": 0.47066664695739746, "learning_rate": 0.00013690685413005274}
+{"ts": "2025-12-26T21:16:47", "event": "train_log", "step": 782, "epoch": 0.329957805907173, "progress_pct": 5.5, "epoch_pct": 5.5, "eta": "51:26:32", "max_grad_norm": 1.0, "loss": 0.9797834753990173, "grad_norm": 0.5725092887878418, "learning_rate": 0.00013725834797891038}
+{"ts": "2025-12-26T21:17:04", "event": "train_log", "step": 784, "epoch": 0.3308016877637131, "progress_pct": 5.51, "epoch_pct": 5.51, "eta": "51:23:03", "max_grad_norm": 1.0, "loss": 0.9372621178627014, "grad_norm": 0.5139563083648682, "learning_rate": 0.00013760984182776802}
+{"ts": "2025-12-26T21:17:19", "event": "train_log", "step": 786, "epoch": 0.33164556962025316, "progress_pct": 5.53, "epoch_pct": 5.53, "eta": "51:18:56", "max_grad_norm": 1.0, "loss": 0.9528245329856873, "grad_norm": 0.5275821685791016, "learning_rate": 0.00013796133567662566}
+{"ts": "2025-12-26T21:17:34", "event": "train_log", "step": 788, "epoch": 0.33248945147679326, "progress_pct": 5.54, "epoch_pct": 5.54, "eta": "51:14:59", "max_grad_norm": 1.0, "loss": 0.9750176072120667, "grad_norm": 0.5702582001686096, "learning_rate": 0.0001383128295254833}
+{"ts": "2025-12-26T21:17:49", "event": "train_log", "step": 790, "epoch": 0.3333333333333333, "progress_pct": 5.56, "epoch_pct": 5.56, "eta": "51:11:03", "max_grad_norm": 1.0, "loss": 0.9412306547164917, "grad_norm": 0.5281293392181396, "learning_rate": 0.00013866432337434096}
+{"ts": "2025-12-26T21:18:05", "event": "train_log", "step": 792, "epoch": 0.3341772151898734, "progress_pct": 5.57, "epoch_pct": 5.57, "eta": "51:07:18", "max_grad_norm": 1.0, "loss": 0.997580885887146, "grad_norm": 0.5578986406326294, "learning_rate": 0.0001390158172231986}
+{"ts": "2025-12-26T21:18:21", "event": "train_log", "step": 794, "epoch": 0.33502109704641353, "progress_pct": 5.58, "epoch_pct": 5.58, "eta": "51:03:36", "max_grad_norm": 1.0, "loss": 0.9082320928573608, "grad_norm": 0.50461345911026, "learning_rate": 0.00013936731107205624}
+{"ts": "2025-12-26T21:18:36", "event": "train_log", "step": 796, "epoch": 0.3358649789029536, "progress_pct": 5.6, "epoch_pct": 5.6, "eta": "50:59:41", "max_grad_norm": 1.0, "loss": 1.0082844495773315, "grad_norm": 0.5258530378341675, "learning_rate": 0.0001397188049209139}
+{"ts": "2025-12-26T21:18:50", "event": "train_log", "step": 798, "epoch": 0.3367088607594937, "progress_pct": 5.61, "epoch_pct": 5.61, "eta": "50:55:38", "max_grad_norm": 1.0, "loss": 0.9729003310203552, "grad_norm": 0.5548169016838074, "learning_rate": 0.00014007029876977154}
+{"ts": "2025-12-26T21:19:06", "event": "train_log", "step": 800, "epoch": 0.33755274261603374, "progress_pct": 5.63, "epoch_pct": 5.63, "eta": "50:51:55", "max_grad_norm": 1.0, "loss": 0.9099526405334473, "grad_norm": 0.48601076006889343, "learning_rate": 0.00014042179261862918}
+{"ts": "2025-12-26T21:30:25", "event": "train_log", "step": 800, "epoch": 0.33755274261603374, "progress_pct": 5.63, "epoch_pct": 5.63, "eta": "54:01:42", "max_grad_norm": 1.0, "eval_loss": 0.9800403714179993, "eval_runtime": 678.8306, "eval_samples_per_second": 3.104, "eval_steps_per_second": 3.104}
+{"ts": "2025-12-26T21:30:40", "event": "train_log", "step": 802, "epoch": 0.33839662447257385, "progress_pct": 5.64, "epoch_pct": 5.64, "eta": "53:57:18", "max_grad_norm": 1.0, "loss": 0.8610644936561584, "grad_norm": 0.5413158535957336, "learning_rate": 0.00014077328646748682}
+{"ts": "2025-12-26T21:30:55", "event": "train_log", "step": 804, "epoch": 0.3392405063291139, "progress_pct": 5.65, "epoch_pct": 5.65, "eta": "53:53:04", "max_grad_norm": 1.0, "loss": 0.9584825038909912, "grad_norm": 0.5147035717964172, "learning_rate": 0.00014112478031634446}
+{"ts": "2025-12-26T21:31:10", "event": "train_log", "step": 806, "epoch": 0.340084388185654, "progress_pct": 5.67, "epoch_pct": 5.67, "eta": "53:48:43", "max_grad_norm": 1.0, "loss": 1.0142558813095093, "grad_norm": 0.5931771397590637, "learning_rate": 0.0001414762741652021}
+{"ts": "2025-12-26T21:31:25", "event": "train_log", "step": 808, "epoch": 0.3409282700421941, "progress_pct": 5.68, "epoch_pct": 5.68, "eta": "53:44:25", "max_grad_norm": 1.0, "loss": 1.0078763961791992, "grad_norm": 0.5178377032279968, "learning_rate": 0.00014182776801405976}
+{"ts": "2025-12-26T21:31:41", "event": "train_log", "step": 810, "epoch": 0.34177215189873417, "progress_pct": 5.7, "epoch_pct": 5.7, "eta": "53:40:22", "max_grad_norm": 1.0, "loss": 0.9107215404510498, "grad_norm": 0.5453237295150757, "learning_rate": 0.0001421792618629174}
+{"ts": "2025-12-26T21:31:56", "event": "train_log", "step": 812, "epoch": 0.3426160337552743, "progress_pct": 5.71, "epoch_pct": 5.71, "eta": "53:35:55", "max_grad_norm": 1.0, "loss": 0.9981362819671631, "grad_norm": 0.5886152982711792, "learning_rate": 0.00014253075571177506}
+{"ts": "2025-12-26T21:32:11", "event": "train_log", "step": 814, "epoch": 0.3434599156118143, "progress_pct": 5.72, "epoch_pct": 5.72, "eta": "53:31:45", "max_grad_norm": 1.0, "loss": 0.9636131525039673, "grad_norm": 0.48040178418159485, "learning_rate": 0.0001428822495606327}
+{"ts": "2025-12-26T21:32:26", "event": "train_log", "step": 816, "epoch": 0.34430379746835443, "progress_pct": 5.74, "epoch_pct": 5.74, "eta": "53:27:21", "max_grad_norm": 1.0, "loss": 0.9590586423873901, "grad_norm": 0.5011753439903259, "learning_rate": 0.00014323374340949034}
+{"ts": "2025-12-26T21:32:40", "event": "train_log", "step": 818, "epoch": 0.34514767932489454, "progress_pct": 5.75, "epoch_pct": 5.75, "eta": "53:23:00", "max_grad_norm": 1.0, "loss": 0.978246808052063, "grad_norm": 0.57858806848526, "learning_rate": 0.00014358523725834798}
+{"ts": "2025-12-26T21:32:56", "event": "train_log", "step": 820, "epoch": 0.3459915611814346, "progress_pct": 5.77, "epoch_pct": 5.77, "eta": "53:19:00", "max_grad_norm": 1.0, "loss": 0.8549934029579163, "grad_norm": 0.46092939376831055, "learning_rate": 0.00014393673110720562}
+{"ts": "2025-12-26T21:33:11", "event": "train_log", "step": 822, "epoch": 0.3468354430379747, "progress_pct": 5.78, "epoch_pct": 5.78, "eta": "53:14:47", "max_grad_norm": 1.0, "loss": 0.9771265387535095, "grad_norm": 0.5756489038467407, "learning_rate": 0.00014428822495606326}
+{"ts": "2025-12-26T21:33:26", "event": "train_log", "step": 824, "epoch": 0.34767932489451475, "progress_pct": 5.79, "epoch_pct": 5.79, "eta": "53:10:41", "max_grad_norm": 1.0, "loss": 0.8739748001098633, "grad_norm": 0.5501731634140015, "learning_rate": 0.00014463971880492092}
+{"ts": "2025-12-26T21:33:40", "event": "train_log", "step": 826, "epoch": 0.34852320675105486, "progress_pct": 5.81, "epoch_pct": 5.81, "eta": "53:06:23", "max_grad_norm": 1.0, "loss": 0.9129468202590942, "grad_norm": 0.5451868176460266, "learning_rate": 0.0001449912126537786}
+{"ts": "2025-12-26T21:33:56", "event": "train_log", "step": 828, "epoch": 0.3493670886075949, "progress_pct": 5.82, "epoch_pct": 5.82, "eta": "53:02:18", "max_grad_norm": 1.0, "loss": 0.9196704030036926, "grad_norm": 0.4624619781970978, "learning_rate": 0.00014534270650263623}
+{"ts": "2025-12-26T21:34:10", "event": "train_log", "step": 830, "epoch": 0.350210970464135, "progress_pct": 5.84, "epoch_pct": 5.84, "eta": "52:58:07", "max_grad_norm": 1.0, "loss": 0.9976527690887451, "grad_norm": 0.520878791809082, "learning_rate": 0.00014569420035149386}
+{"ts": "2025-12-26T21:34:26", "event": "train_log", "step": 832, "epoch": 0.3510548523206751, "progress_pct": 5.85, "epoch_pct": 5.85, "eta": "52:54:12", "max_grad_norm": 1.0, "loss": 0.7753443717956543, "grad_norm": 0.4469171464443207, "learning_rate": 0.0001460456942003515}
+{"ts": "2025-12-26T21:34:40", "event": "train_log", "step": 834, "epoch": 0.3518987341772152, "progress_pct": 5.86, "epoch_pct": 5.86, "eta": "52:49:58", "max_grad_norm": 1.0, "loss": 0.9584846496582031, "grad_norm": 0.5105249881744385, "learning_rate": 0.00014639718804920914}
+{"ts": "2025-12-26T21:34:57", "event": "train_log", "step": 836, "epoch": 0.3527426160337553, "progress_pct": 5.88, "epoch_pct": 5.88, "eta": "52:46:15", "max_grad_norm": 1.0, "loss": 0.9013500213623047, "grad_norm": 0.5043913125991821, "learning_rate": 0.00014674868189806678}
+{"ts": "2025-12-26T21:35:11", "event": "train_log", "step": 838, "epoch": 0.35358649789029534, "progress_pct": 5.89, "epoch_pct": 5.89, "eta": "52:41:58", "max_grad_norm": 1.0, "loss": 0.9775562286376953, "grad_norm": 0.575850784778595, "learning_rate": 0.00014710017574692442}
+{"ts": "2025-12-26T21:35:25", "event": "train_log", "step": 840, "epoch": 0.35443037974683544, "progress_pct": 5.91, "epoch_pct": 5.91, "eta": "52:37:44", "max_grad_norm": 1.0, "loss": 0.9278940558433533, "grad_norm": 0.5128876566886902, "learning_rate": 0.00014745166959578208}
+{"ts": "2025-12-26T21:35:40", "event": "train_log", "step": 842, "epoch": 0.35527426160337555, "progress_pct": 5.92, "epoch_pct": 5.92, "eta": "52:33:47", "max_grad_norm": 1.0, "loss": 1.0091488361358643, "grad_norm": 0.5757885575294495, "learning_rate": 0.00014780316344463975}
+{"ts": "2025-12-26T21:35:55", "event": "train_log", "step": 844, "epoch": 0.3561181434599156, "progress_pct": 5.94, "epoch_pct": 5.94, "eta": "52:29:53", "max_grad_norm": 1.0, "loss": 0.9286836981773376, "grad_norm": 0.500934898853302, "learning_rate": 0.00014815465729349739}
+{"ts": "2025-12-26T21:36:10", "event": "train_log", "step": 846, "epoch": 0.3569620253164557, "progress_pct": 5.95, "epoch_pct": 5.95, "eta": "52:25:58", "max_grad_norm": 1.0, "loss": 0.9484171867370605, "grad_norm": 0.5220686197280884, "learning_rate": 0.00014850615114235502}
+{"ts": "2025-12-26T21:36:26", "event": "train_log", "step": 848, "epoch": 0.35780590717299576, "progress_pct": 5.96, "epoch_pct": 5.96, "eta": "52:22:05", "max_grad_norm": 1.0, "loss": 1.0556397438049316, "grad_norm": 0.5494697690010071, "learning_rate": 0.00014885764499121266}
+{"ts": "2025-12-26T21:36:41", "event": "train_log", "step": 850, "epoch": 0.35864978902953587, "progress_pct": 5.98, "epoch_pct": 5.98, "eta": "52:18:16", "max_grad_norm": 1.0, "loss": 0.933089017868042, "grad_norm": 0.46633943915367126, "learning_rate": 0.0001492091388400703}
+{"ts": "2025-12-26T21:36:56", "event": "train_log", "step": 852, "epoch": 0.3594936708860759, "progress_pct": 5.99, "epoch_pct": 5.99, "eta": "52:14:19", "max_grad_norm": 1.0, "loss": 1.0016963481903076, "grad_norm": 0.4864962697029114, "learning_rate": 0.00014956063268892794}
+{"ts": "2025-12-26T21:37:10", "event": "train_log", "step": 854, "epoch": 0.36033755274261603, "progress_pct": 6.01, "epoch_pct": 6.01, "eta": "52:10:17", "max_grad_norm": 1.0, "loss": 0.9085348844528198, "grad_norm": 0.5032764673233032, "learning_rate": 0.0001499121265377856}
+{"ts": "2025-12-26T21:37:25", "event": "train_log", "step": 856, "epoch": 0.36118143459915614, "progress_pct": 6.02, "epoch_pct": 6.02, "eta": "52:06:22", "max_grad_norm": 1.0, "loss": 0.923502504825592, "grad_norm": 0.5424998998641968, "learning_rate": 0.00015026362038664324}
+{"ts": "2025-12-26T21:37:40", "event": "train_log", "step": 858, "epoch": 0.3620253164556962, "progress_pct": 6.03, "epoch_pct": 6.03, "eta": "52:02:28", "max_grad_norm": 1.0, "loss": 0.919174313545227, "grad_norm": 0.5181655287742615, "learning_rate": 0.00015061511423550088}
+{"ts": "2025-12-26T21:37:56", "event": "train_log", "step": 860, "epoch": 0.3628691983122363, "progress_pct": 6.05, "epoch_pct": 6.05, "eta": "51:58:46", "max_grad_norm": 1.0, "loss": 0.851981520652771, "grad_norm": 0.5129443407058716, "learning_rate": 0.00015096660808435855}
+{"ts": "2025-12-26T21:38:11", "event": "train_log", "step": 862, "epoch": 0.36371308016877635, "progress_pct": 6.06, "epoch_pct": 6.06, "eta": "51:55:02", "max_grad_norm": 1.0, "loss": 0.8633858561515808, "grad_norm": 0.49540698528289795, "learning_rate": 0.00015131810193321618}
+{"ts": "2025-12-26T21:38:26", "event": "train_log", "step": 864, "epoch": 0.36455696202531646, "progress_pct": 6.08, "epoch_pct": 6.08, "eta": "51:51:17", "max_grad_norm": 1.0, "loss": 0.9473677277565002, "grad_norm": 0.4706701934337616, "learning_rate": 0.00015166959578207382}
+{"ts": "2025-12-26T21:38:41", "event": "train_log", "step": 866, "epoch": 0.36540084388185656, "progress_pct": 6.09, "epoch_pct": 6.09, "eta": "51:47:31", "max_grad_norm": 1.0, "loss": 0.942383885383606, "grad_norm": 0.587704598903656, "learning_rate": 0.00015202108963093146}
+{"ts": "2025-12-26T21:38:58", "event": "train_log", "step": 868, "epoch": 0.3662447257383966, "progress_pct": 6.1, "epoch_pct": 6.1, "eta": "51:44:02", "max_grad_norm": 1.0, "loss": 0.9811251163482666, "grad_norm": 0.5851273536682129, "learning_rate": 0.0001523725834797891}
+{"ts": "2025-12-26T21:39:13", "event": "train_log", "step": 870, "epoch": 0.3670886075949367, "progress_pct": 6.12, "epoch_pct": 6.12, "eta": "51:40:24", "max_grad_norm": 1.0, "loss": 0.8151084184646606, "grad_norm": 0.46357613801956177, "learning_rate": 0.00015272407732864676}
+{"ts": "2025-12-26T21:39:30", "event": "train_log", "step": 872, "epoch": 0.3679324894514768, "progress_pct": 6.13, "epoch_pct": 6.13, "eta": "51:37:04", "max_grad_norm": 1.0, "loss": 0.9023333191871643, "grad_norm": 0.47250670194625854, "learning_rate": 0.0001530755711775044}
+{"ts": "2025-12-26T21:39:45", "event": "train_log", "step": 874, "epoch": 0.3687763713080169, "progress_pct": 6.15, "epoch_pct": 6.15, "eta": "51:33:29", "max_grad_norm": 1.0, "loss": 0.8758499622344971, "grad_norm": 0.47510042786598206, "learning_rate": 0.00015342706502636204}
+{"ts": "2025-12-26T21:40:00", "event": "train_log", "step": 876, "epoch": 0.369620253164557, "progress_pct": 6.16, "epoch_pct": 6.16, "eta": "51:29:37", "max_grad_norm": 1.0, "loss": 0.9837421774864197, "grad_norm": 0.5687124729156494, "learning_rate": 0.00015377855887521968}
+{"ts": "2025-12-26T21:40:15", "event": "train_log", "step": 878, "epoch": 0.37046413502109704, "progress_pct": 6.17, "epoch_pct": 6.17, "eta": "51:25:57", "max_grad_norm": 1.0, "loss": 0.9084216356277466, "grad_norm": 0.49064236879348755, "learning_rate": 0.00015413005272407735}
+{"ts": "2025-12-26T21:40:31", "event": "train_log", "step": 880, "epoch": 0.37130801687763715, "progress_pct": 6.19, "epoch_pct": 6.19, "eta": "51:22:31", "max_grad_norm": 1.0, "loss": 0.8425542712211609, "grad_norm": 0.5288164615631104, "learning_rate": 0.00015448154657293498}
+{"ts": "2025-12-26T21:40:46", "event": "train_log", "step": 882, "epoch": 0.3721518987341772, "progress_pct": 6.2, "epoch_pct": 6.2, "eta": "51:18:48", "max_grad_norm": 1.0, "loss": 1.006197452545166, "grad_norm": 0.5446951985359192, "learning_rate": 0.00015483304042179262}
+{"ts": "2025-12-26T21:41:02", "event": "train_log", "step": 884, "epoch": 0.3729957805907173, "progress_pct": 6.22, "epoch_pct": 6.22, "eta": "51:15:32", "max_grad_norm": 1.0, "loss": 0.9779444932937622, "grad_norm": 0.46872642636299133, "learning_rate": 0.00015518453427065026}
+{"ts": "2025-12-26T21:41:19", "event": "train_log", "step": 886, "epoch": 0.37383966244725736, "progress_pct": 6.23, "epoch_pct": 6.23, "eta": "51:12:13", "max_grad_norm": 1.0, "loss": 0.9099963307380676, "grad_norm": 0.4852714240550995, "learning_rate": 0.00015553602811950793}
+{"ts": "2025-12-26T21:41:35", "event": "train_log", "step": 888, "epoch": 0.37468354430379747, "progress_pct": 6.24, "epoch_pct": 6.24, "eta": "51:08:57", "max_grad_norm": 1.0, "loss": 0.9730570316314697, "grad_norm": 0.5219841003417969, "learning_rate": 0.00015588752196836556}
+{"ts": "2025-12-26T21:41:50", "event": "train_log", "step": 890, "epoch": 0.3755274261603376, "progress_pct": 6.26, "epoch_pct": 6.26, "eta": "51:05:18", "max_grad_norm": 1.0, "loss": 0.9304586052894592, "grad_norm": 0.5258626341819763, "learning_rate": 0.0001562390158172232}
+{"ts": "2025-12-26T21:42:07", "event": "train_log", "step": 892, "epoch": 0.3763713080168776, "progress_pct": 6.27, "epoch_pct": 6.27, "eta": "51:02:19", "max_grad_norm": 1.0, "loss": 0.8680716156959534, "grad_norm": 0.4266716241836548, "learning_rate": 0.00015659050966608084}
+{"ts": "2025-12-26T21:42:23", "event": "train_log", "step": 894, "epoch": 0.37721518987341773, "progress_pct": 6.29, "epoch_pct": 6.29, "eta": "50:59:00", "max_grad_norm": 1.0, "loss": 0.8428018093109131, "grad_norm": 0.46361327171325684, "learning_rate": 0.0001569420035149385}
+{"ts": "2025-12-26T21:42:39", "event": "train_log", "step": 896, "epoch": 0.3780590717299578, "progress_pct": 6.3, "epoch_pct": 6.3, "eta": "50:55:40", "max_grad_norm": 1.0, "loss": 0.8465750217437744, "grad_norm": 0.5313687920570374, "learning_rate": 0.00015729349736379614}
+{"ts": "2025-12-26T21:42:54", "event": "train_log", "step": 898, "epoch": 0.3789029535864979, "progress_pct": 6.32, "epoch_pct": 6.32, "eta": "50:52:08", "max_grad_norm": 1.0, "loss": 0.9120327234268188, "grad_norm": 0.47166210412979126, "learning_rate": 0.00015764499121265378}
+{"ts": "2025-12-26T21:43:10", "event": "train_log", "step": 900, "epoch": 0.379746835443038, "progress_pct": 6.33, "epoch_pct": 6.33, "eta": "50:48:52", "max_grad_norm": 1.0, "loss": 0.8950425982475281, "grad_norm": 0.4647318720817566, "learning_rate": 0.00015799648506151145}
+{"ts": "2025-12-26T21:54:42", "event": "train_log", "step": 900, "epoch": 0.379746835443038, "progress_pct": 6.33, "epoch_pct": 6.33, "eta": "53:39:30", "max_grad_norm": 1.0, "eval_loss": 0.9643027186393738, "eval_runtime": 691.7929, "eval_samples_per_second": 3.046, "eval_steps_per_second": 3.046}
+{"ts": "2025-12-26T21:54:57", "event": "train_log", "step": 902, "epoch": 0.38059071729957805, "progress_pct": 6.34, "epoch_pct": 6.34, "eta": "53:35:36", "max_grad_norm": 1.0, "loss": 0.908163845539093, "grad_norm": 0.5445119738578796, "learning_rate": 0.00015834797891036909}
+{"ts": "2025-12-26T21:55:13", "event": "train_log", "step": 904, "epoch": 0.38143459915611816, "progress_pct": 6.36, "epoch_pct": 6.36, "eta": "53:31:48", "max_grad_norm": 1.0, "loss": 0.8945821523666382, "grad_norm": 0.4311858117580414, "learning_rate": 0.00015869947275922672}
+{"ts": "2025-12-26T21:55:27", "event": "train_log", "step": 906, "epoch": 0.3822784810126582, "progress_pct": 6.37, "epoch_pct": 6.37, "eta": "53:27:41", "max_grad_norm": 1.0, "loss": 0.9478458762168884, "grad_norm": 0.5590984225273132, "learning_rate": 0.00015905096660808436}
+{"ts": "2025-12-26T21:55:41", "event": "train_log", "step": 908, "epoch": 0.3831223628691983, "progress_pct": 6.39, "epoch_pct": 6.39, "eta": "53:23:40", "max_grad_norm": 1.0, "loss": 0.9259957671165466, "grad_norm": 0.5470241904258728, "learning_rate": 0.000159402460456942}
+{"ts": "2025-12-26T21:55:56", "event": "train_log", "step": 910, "epoch": 0.38396624472573837, "progress_pct": 6.4, "epoch_pct": 6.4, "eta": "53:19:46", "max_grad_norm": 1.0, "loss": 0.8824930787086487, "grad_norm": 0.5498791337013245, "learning_rate": 0.00015975395430579964}
+{"ts": "2025-12-26T21:56:12", "event": "train_log", "step": 912, "epoch": 0.3848101265822785, "progress_pct": 6.41, "epoch_pct": 6.41, "eta": "53:16:04", "max_grad_norm": 1.0, "loss": 0.8842340707778931, "grad_norm": 0.4779198467731476, "learning_rate": 0.0001601054481546573}
+{"ts": "2025-12-26T21:56:27", "event": "train_log", "step": 914, "epoch": 0.3856540084388186, "progress_pct": 6.43, "epoch_pct": 6.43, "eta": "53:12:23", "max_grad_norm": 1.0, "loss": 0.92950040102005, "grad_norm": 0.5390620827674866, "learning_rate": 0.00016045694200351494}
+{"ts": "2025-12-26T21:56:42", "event": "train_log", "step": 916, "epoch": 0.38649789029535864, "progress_pct": 6.44, "epoch_pct": 6.44, "eta": "53:08:24", "max_grad_norm": 1.0, "loss": 0.8420897126197815, "grad_norm": 0.505519688129425, "learning_rate": 0.0001608084358523726}
+{"ts": "2025-12-26T21:56:57", "event": "train_log", "step": 918, "epoch": 0.38734177215189874, "progress_pct": 6.46, "epoch_pct": 6.46, "eta": "53:04:38", "max_grad_norm": 1.0, "loss": 0.8256624341011047, "grad_norm": 0.4463907778263092, "learning_rate": 0.00016115992970123025}
+{"ts": "2025-12-26T21:57:12", "event": "train_log", "step": 920, "epoch": 0.3881856540084388, "progress_pct": 6.47, "epoch_pct": 6.47, "eta": "53:00:53", "max_grad_norm": 1.0, "loss": 0.8701168298721313, "grad_norm": 0.5321422219276428, "learning_rate": 0.00016151142355008788}
+{"ts": "2025-12-26T21:57:27", "event": "train_log", "step": 922, "epoch": 0.3890295358649789, "progress_pct": 6.48, "epoch_pct": 6.48, "eta": "52:57:05", "max_grad_norm": 1.0, "loss": 0.9019309878349304, "grad_norm": 0.4343073070049286, "learning_rate": 0.00016186291739894552}
+{"ts": "2025-12-26T21:57:42", "event": "train_log", "step": 924, "epoch": 0.389873417721519, "progress_pct": 6.5, "epoch_pct": 6.5, "eta": "52:53:23", "max_grad_norm": 1.0, "loss": 0.8560551404953003, "grad_norm": 0.5311984419822693, "learning_rate": 0.00016221441124780316}
+{"ts": "2025-12-26T21:57:56", "event": "train_log", "step": 926, "epoch": 0.39071729957805906, "progress_pct": 6.51, "epoch_pct": 6.51, "eta": "52:49:26", "max_grad_norm": 1.0, "loss": 0.9700050354003906, "grad_norm": 0.588691771030426, "learning_rate": 0.0001625659050966608}
+{"ts": "2025-12-26T21:58:10", "event": "train_log", "step": 928, "epoch": 0.39156118143459917, "progress_pct": 6.53, "epoch_pct": 6.53, "eta": "52:45:28", "max_grad_norm": 1.0, "loss": 1.043768048286438, "grad_norm": 0.5425586104393005, "learning_rate": 0.00016291739894551844}
+{"ts": "2025-12-26T21:58:25", "event": "train_log", "step": 930, "epoch": 0.3924050632911392, "progress_pct": 6.54, "epoch_pct": 6.54, "eta": "52:41:36", "max_grad_norm": 1.0, "loss": 0.9501712918281555, "grad_norm": 0.5228736996650696, "learning_rate": 0.0001632688927943761}
+{"ts": "2025-12-26T21:58:40", "event": "train_log", "step": 932, "epoch": 0.39324894514767933, "progress_pct": 6.55, "epoch_pct": 6.55, "eta": "52:37:56", "max_grad_norm": 1.0, "loss": 0.9223058223724365, "grad_norm": 0.48960360884666443, "learning_rate": 0.00016362038664323377}
+{"ts": "2025-12-26T21:58:55", "event": "train_log", "step": 934, "epoch": 0.39409282700421944, "progress_pct": 6.57, "epoch_pct": 6.57, "eta": "52:34:22", "max_grad_norm": 1.0, "loss": 0.9692960977554321, "grad_norm": 0.45204755663871765, "learning_rate": 0.0001639718804920914}
+{"ts": "2025-12-26T21:59:10", "event": "train_log", "step": 936, "epoch": 0.3949367088607595, "progress_pct": 6.58, "epoch_pct": 6.58, "eta": "52:30:37", "max_grad_norm": 1.0, "loss": 0.9467466473579407, "grad_norm": 0.5299274921417236, "learning_rate": 0.00016432337434094905}
+{"ts": "2025-12-26T21:59:25", "event": "train_log", "step": 938, "epoch": 0.3957805907172996, "progress_pct": 6.6, "epoch_pct": 6.6, "eta": "52:26:58", "max_grad_norm": 1.0, "loss": 0.9118053317070007, "grad_norm": 0.5607715249061584, "learning_rate": 0.00016467486818980668}
+{"ts": "2025-12-26T21:59:39", "event": "train_log", "step": 940, "epoch": 0.39662447257383965, "progress_pct": 6.61, "epoch_pct": 6.61, "eta": "52:23:10", "max_grad_norm": 1.0, "loss": 0.9131460189819336, "grad_norm": 0.5271831154823303, "learning_rate": 0.00016502636203866432}
+{"ts": "2025-12-26T21:59:55", "event": "train_log", "step": 942, "epoch": 0.39746835443037976, "progress_pct": 6.62, "epoch_pct": 6.62, "eta": "52:19:40", "max_grad_norm": 1.0, "loss": 0.9358300566673279, "grad_norm": 0.5075286030769348, "learning_rate": 0.00016537785588752196}
+{"ts": "2025-12-26T22:00:10", "event": "train_log", "step": 944, "epoch": 0.3983122362869198, "progress_pct": 6.64, "epoch_pct": 6.64, "eta": "52:16:02", "max_grad_norm": 1.0, "loss": 0.8908210396766663, "grad_norm": 0.515731155872345, "learning_rate": 0.00016572934973637963}
+{"ts": "2025-12-26T22:00:24", "event": "train_log", "step": 946, "epoch": 0.3991561181434599, "progress_pct": 6.65, "epoch_pct": 6.65, "eta": "52:12:24", "max_grad_norm": 1.0, "loss": 0.9775290489196777, "grad_norm": 0.4856977164745331, "learning_rate": 0.00016608084358523726}
+{"ts": "2025-12-26T22:00:39", "event": "train_log", "step": 948, "epoch": 0.4, "progress_pct": 6.67, "epoch_pct": 6.67, "eta": "52:08:42", "max_grad_norm": 1.0, "loss": 0.8957490921020508, "grad_norm": 0.48846355080604553, "learning_rate": 0.00016643233743409493}
+{"ts": "2025-12-26T22:00:54", "event": "train_log", "step": 950, "epoch": 0.4008438818565401, "progress_pct": 6.68, "epoch_pct": 6.68, "eta": "52:05:14", "max_grad_norm": 1.0, "loss": 0.9036174416542053, "grad_norm": 0.42990800738334656, "learning_rate": 0.00016678383128295257}
+{"ts": "2025-12-26T22:01:09", "event": "train_log", "step": 952, "epoch": 0.4016877637130802, "progress_pct": 6.69, "epoch_pct": 6.69, "eta": "52:01:39", "max_grad_norm": 1.0, "loss": 0.991032600402832, "grad_norm": 0.49552062153816223, "learning_rate": 0.0001671353251318102}
+{"ts": "2025-12-26T22:01:24", "event": "train_log", "step": 954, "epoch": 0.40253164556962023, "progress_pct": 6.71, "epoch_pct": 6.71, "eta": "51:58:06", "max_grad_norm": 1.0, "loss": 0.823063313961029, "grad_norm": 0.4565040171146393, "learning_rate": 0.00016748681898066784}
+{"ts": "2025-12-26T22:01:39", "event": "train_log", "step": 956, "epoch": 0.40337552742616034, "progress_pct": 6.72, "epoch_pct": 6.72, "eta": "51:54:35", "max_grad_norm": 1.0, "loss": 0.8785063624382019, "grad_norm": 0.4290153682231903, "learning_rate": 0.00016783831282952548}
+{"ts": "2025-12-26T22:01:54", "event": "train_log", "step": 958, "epoch": 0.40421940928270045, "progress_pct": 6.74, "epoch_pct": 6.74, "eta": "51:51:01", "max_grad_norm": 1.0, "loss": 0.8763971924781799, "grad_norm": 0.5419702529907227, "learning_rate": 0.00016818980667838312}
+{"ts": "2025-12-26T22:02:09", "event": "train_log", "step": 960, "epoch": 0.4050632911392405, "progress_pct": 6.75, "epoch_pct": 6.75, "eta": "51:47:34", "max_grad_norm": 1.0, "loss": 0.9470553398132324, "grad_norm": 0.5177501440048218, "learning_rate": 0.00016854130052724079}
+{"ts": "2025-12-26T22:02:24", "event": "train_log", "step": 962, "epoch": 0.4059071729957806, "progress_pct": 6.77, "epoch_pct": 6.77, "eta": "51:44:07", "max_grad_norm": 1.0, "loss": 0.9235025644302368, "grad_norm": 0.539725661277771, "learning_rate": 0.00016889279437609842}
+{"ts": "2025-12-26T22:02:38", "event": "train_log", "step": 964, "epoch": 0.40675105485232066, "progress_pct": 6.78, "epoch_pct": 6.78, "eta": "51:40:25", "max_grad_norm": 1.0, "loss": 1.0248996019363403, "grad_norm": 0.5324983596801758, "learning_rate": 0.0001692442882249561}
+{"ts": "2025-12-26T22:02:54", "event": "train_log", "step": 966, "epoch": 0.40759493670886077, "progress_pct": 6.79, "epoch_pct": 6.79, "eta": "51:37:07", "max_grad_norm": 1.0, "loss": 0.9076873660087585, "grad_norm": 0.4936407506465912, "learning_rate": 0.00016959578207381373}
+{"ts": "2025-12-26T22:03:09", "event": "train_log", "step": 968, "epoch": 0.4084388185654008, "progress_pct": 6.81, "epoch_pct": 6.81, "eta": "51:33:42", "max_grad_norm": 1.0, "loss": 0.9162673950195312, "grad_norm": 0.4960501194000244, "learning_rate": 0.00016994727592267137}
+{"ts": "2025-12-26T22:03:24", "event": "train_log", "step": 970, "epoch": 0.4092827004219409, "progress_pct": 6.82, "epoch_pct": 6.82, "eta": "51:30:15", "max_grad_norm": 1.0, "loss": 0.904100775718689, "grad_norm": 0.45093682408332825, "learning_rate": 0.000170298769771529}
+{"ts": "2025-12-26T22:03:39", "event": "train_log", "step": 972, "epoch": 0.41012658227848103, "progress_pct": 6.84, "epoch_pct": 6.84, "eta": "51:26:56", "max_grad_norm": 1.0, "loss": 0.8633337020874023, "grad_norm": 0.4560275077819824, "learning_rate": 0.00017065026362038664}
+{"ts": "2025-12-26T22:03:54", "event": "train_log", "step": 974, "epoch": 0.4109704641350211, "progress_pct": 6.85, "epoch_pct": 6.85, "eta": "51:23:30", "max_grad_norm": 1.0, "loss": 0.8454209566116333, "grad_norm": 0.44885000586509705, "learning_rate": 0.00017100175746924428}
+{"ts": "2025-12-26T22:04:10", "event": "train_log", "step": 976, "epoch": 0.4118143459915612, "progress_pct": 6.86, "epoch_pct": 6.86, "eta": "51:20:11", "max_grad_norm": 1.0, "loss": 0.824730396270752, "grad_norm": 0.47251659631729126, "learning_rate": 0.00017135325131810195}
+{"ts": "2025-12-26T22:04:25", "event": "train_log", "step": 978, "epoch": 0.41265822784810124, "progress_pct": 6.88, "epoch_pct": 6.88, "eta": "51:16:48", "max_grad_norm": 1.0, "loss": 0.9496501684188843, "grad_norm": 0.6597666144371033, "learning_rate": 0.00017170474516695959}
+{"ts": "2025-12-26T22:04:40", "event": "train_log", "step": 980, "epoch": 0.41350210970464135, "progress_pct": 6.89, "epoch_pct": 6.89, "eta": "51:13:28", "max_grad_norm": 1.0, "loss": 0.9419087171554565, "grad_norm": 0.39806297421455383, "learning_rate": 0.00017205623901581722}
+{"ts": "2025-12-26T22:04:55", "event": "train_log", "step": 982, "epoch": 0.41434599156118146, "progress_pct": 6.91, "epoch_pct": 6.91, "eta": "51:10:04", "max_grad_norm": 1.0, "loss": 0.9182976484298706, "grad_norm": 0.48231109976768494, "learning_rate": 0.0001724077328646749}
+{"ts": "2025-12-26T22:05:09", "event": "train_log", "step": 984, "epoch": 0.4151898734177215, "progress_pct": 6.92, "epoch_pct": 6.92, "eta": "51:06:38", "max_grad_norm": 1.0, "loss": 0.9386967420578003, "grad_norm": 0.5438776612281799, "learning_rate": 0.00017275922671353253}
+{"ts": "2025-12-26T22:05:24", "event": "train_log", "step": 986, "epoch": 0.4160337552742616, "progress_pct": 6.93, "epoch_pct": 6.93, "eta": "51:03:21", "max_grad_norm": 1.0, "loss": 0.897849440574646, "grad_norm": 0.4959667921066284, "learning_rate": 0.00017311072056239017}
+{"ts": "2025-12-26T22:05:39", "event": "train_log", "step": 988, "epoch": 0.41687763713080167, "progress_pct": 6.95, "epoch_pct": 6.95, "eta": "51:00:03", "max_grad_norm": 1.0, "loss": 0.8776953816413879, "grad_norm": 0.43533357977867126, "learning_rate": 0.0001734622144112478}
+{"ts": "2025-12-26T22:05:55", "event": "train_log", "step": 990, "epoch": 0.4177215189873418, "progress_pct": 6.96, "epoch_pct": 6.96, "eta": "50:56:52", "max_grad_norm": 1.0, "loss": 0.9162989854812622, "grad_norm": 0.47513946890830994, "learning_rate": 0.00017381370826010547}
+{"ts": "2025-12-26T22:06:10", "event": "train_log", "step": 992, "epoch": 0.41856540084388183, "progress_pct": 6.98, "epoch_pct": 6.98, "eta": "50:53:35", "max_grad_norm": 1.0, "loss": 0.9482660889625549, "grad_norm": 0.4907188415527344, "learning_rate": 0.0001741652021089631}
+{"ts": "2025-12-26T22:06:25", "event": "train_log", "step": 994, "epoch": 0.41940928270042194, "progress_pct": 6.99, "epoch_pct": 6.99, "eta": "50:50:13", "max_grad_norm": 1.0, "loss": 0.8812930583953857, "grad_norm": 0.44499966502189636, "learning_rate": 0.00017451669595782075}
+{"ts": "2025-12-26T22:06:39", "event": "train_log", "step": 996, "epoch": 0.42025316455696204, "progress_pct": 7.0, "epoch_pct": 7.0, "eta": "50:46:55", "max_grad_norm": 1.0, "loss": 0.9439874887466431, "grad_norm": 0.4535730481147766, "learning_rate": 0.00017486818980667838}
+{"ts": "2025-12-26T22:06:54", "event": "train_log", "step": 998, "epoch": 0.4210970464135021, "progress_pct": 7.02, "epoch_pct": 7.02, "eta": "50:43:37", "max_grad_norm": 1.0, "loss": 0.8818395137786865, "grad_norm": 0.5240745544433594, "learning_rate": 0.00017521968365553602}
+{"ts": "2025-12-26T22:07:10", "event": "train_log", "step": 1000, "epoch": 0.4219409282700422, "progress_pct": 7.03, "epoch_pct": 7.03, "eta": "50:40:27", "max_grad_norm": 1.0, "loss": 0.886186957359314, "grad_norm": 0.5301211476325989, "learning_rate": 0.0001755711775043937}
+{"ts": "2025-12-26T22:18:39", "event": "train_log", "step": 1000, "epoch": 0.4219409282700422, "progress_pct": 7.03, "epoch_pct": 7.03, "eta": "53:12:22", "max_grad_norm": 1.0, "eval_loss": 0.9487298727035522, "eval_runtime": 689.4288, "eval_samples_per_second": 3.056, "eval_steps_per_second": 3.056}
+{"ts": "2025-12-26T22:18:56", "event": "train_log", "step": 1002, "epoch": 0.42278481012658226, "progress_pct": 7.05, "epoch_pct": 7.05, "eta": "53:09:18", "max_grad_norm": 1.0, "loss": 0.8814021348953247, "grad_norm": 0.47876957058906555, "learning_rate": 0.00017592267135325133}
+{"ts": "2025-12-26T22:19:13", "event": "train_log", "step": 1004, "epoch": 0.42362869198312236, "progress_pct": 7.06, "epoch_pct": 7.06, "eta": "53:06:07", "max_grad_norm": 1.0, "loss": 0.8295068740844727, "grad_norm": 0.4929780960083008, "learning_rate": 0.00017627416520210896}
+{"ts": "2025-12-26T22:19:30", "event": "train_log", "step": 1006, "epoch": 0.42447257383966247, "progress_pct": 7.07, "epoch_pct": 7.07, "eta": "53:03:03", "max_grad_norm": 1.0, "loss": 0.8364827036857605, "grad_norm": 0.41888436675071716, "learning_rate": 0.00017662565905096663}
+{"ts": "2025-12-26T22:19:46", "event": "train_log", "step": 1008, "epoch": 0.4253164556962025, "progress_pct": 7.09, "epoch_pct": 7.09, "eta": "52:59:39", "max_grad_norm": 1.0, "loss": 1.0167189836502075, "grad_norm": 0.5175151824951172, "learning_rate": 0.00017697715289982427}
+{"ts": "2025-12-26T22:20:01", "event": "train_log", "step": 1010, "epoch": 0.42616033755274263, "progress_pct": 7.1, "epoch_pct": 7.1, "eta": "52:56:17", "max_grad_norm": 1.0, "loss": 0.8460752964019775, "grad_norm": 0.4815356135368347, "learning_rate": 0.0001773286467486819}
+{"ts": "2025-12-26T22:20:16", "event": "train_log", "step": 1012, "epoch": 0.4270042194092827, "progress_pct": 7.12, "epoch_pct": 7.12, "eta": "52:52:48", "max_grad_norm": 1.0, "loss": 0.8595574498176575, "grad_norm": 0.5210875272750854, "learning_rate": 0.00017768014059753954}
+{"ts": "2025-12-26T22:20:32", "event": "train_log", "step": 1014, "epoch": 0.4278481012658228, "progress_pct": 7.13, "epoch_pct": 7.13, "eta": "52:49:35", "max_grad_norm": 1.0, "loss": 0.8764723539352417, "grad_norm": 0.4400486946105957, "learning_rate": 0.00017803163444639718}
+{"ts": "2025-12-26T22:20:48", "event": "train_log", "step": 1016, "epoch": 0.4286919831223629, "progress_pct": 7.14, "epoch_pct": 7.14, "eta": "52:46:12", "max_grad_norm": 1.0, "loss": 0.9706798791885376, "grad_norm": 0.5282127857208252, "learning_rate": 0.00017838312829525482}
+{"ts": "2025-12-26T22:21:04", "event": "train_log", "step": 1018, "epoch": 0.42953586497890295, "progress_pct": 7.16, "epoch_pct": 7.16, "eta": "52:42:56", "max_grad_norm": 1.0, "loss": 0.8912002444267273, "grad_norm": 0.47523441910743713, "learning_rate": 0.00017873462214411249}
+{"ts": "2025-12-26T22:21:19", "event": "train_log", "step": 1020, "epoch": 0.43037974683544306, "progress_pct": 7.17, "epoch_pct": 7.17, "eta": "52:39:37", "max_grad_norm": 1.0, "loss": 0.8835636973381042, "grad_norm": 0.49640706181526184, "learning_rate": 0.00017908611599297012}
+{"ts": "2025-12-26T22:21:35", "event": "train_log", "step": 1022, "epoch": 0.4312236286919831, "progress_pct": 7.19, "epoch_pct": 7.19, "eta": "52:36:18", "max_grad_norm": 1.0, "loss": 0.8711735606193542, "grad_norm": 0.5253039002418518, "learning_rate": 0.0001794376098418278}
+{"ts": "2025-12-26T22:21:50", "event": "train_log", "step": 1024, "epoch": 0.4320675105485232, "progress_pct": 7.2, "epoch_pct": 7.2, "eta": "52:32:59", "max_grad_norm": 1.0, "loss": 0.9064869284629822, "grad_norm": 0.49285009503364563, "learning_rate": 0.00017978910369068543}
+{"ts": "2025-12-26T22:22:05", "event": "train_log", "step": 1026, "epoch": 0.43291139240506327, "progress_pct": 7.22, "epoch_pct": 7.22, "eta": "52:29:36", "max_grad_norm": 1.0, "loss": 0.8993359208106995, "grad_norm": 0.48758041858673096, "learning_rate": 0.00018014059753954307}
+{"ts": "2025-12-26T22:22:20", "event": "train_log", "step": 1028, "epoch": 0.4337552742616034, "progress_pct": 7.23, "epoch_pct": 7.23, "eta": "52:26:09", "max_grad_norm": 1.0, "loss": 0.8747937679290771, "grad_norm": 0.4815461039543152, "learning_rate": 0.0001804920913884007}
+{"ts": "2025-12-26T22:22:35", "event": "train_log", "step": 1030, "epoch": 0.4345991561181435, "progress_pct": 7.24, "epoch_pct": 7.24, "eta": "52:22:49", "max_grad_norm": 1.0, "loss": 0.8879084587097168, "grad_norm": 0.4930349886417389, "learning_rate": 0.00018084358523725834}
+{"ts": "2025-12-26T22:22:51", "event": "train_log", "step": 1032, "epoch": 0.43544303797468353, "progress_pct": 7.26, "epoch_pct": 7.26, "eta": "52:19:29", "max_grad_norm": 1.0, "loss": 1.0023083686828613, "grad_norm": 0.49303027987480164, "learning_rate": 0.00018119507908611598}
+{"ts": "2025-12-26T22:23:06", "event": "train_log", "step": 1034, "epoch": 0.43628691983122364, "progress_pct": 7.27, "epoch_pct": 7.27, "eta": "52:16:17", "max_grad_norm": 1.0, "loss": 0.938680112361908, "grad_norm": 0.5312249064445496, "learning_rate": 0.00018154657293497365}
+{"ts": "2025-12-26T22:23:23", "event": "train_log", "step": 1036, "epoch": 0.4371308016877637, "progress_pct": 7.29, "epoch_pct": 7.29, "eta": "52:13:13", "max_grad_norm": 1.0, "loss": 0.854195773601532, "grad_norm": 0.5174582004547119, "learning_rate": 0.0001818980667838313}
+{"ts": "2025-12-26T22:23:38", "event": "train_log", "step": 1038, "epoch": 0.4379746835443038, "progress_pct": 7.3, "epoch_pct": 7.3, "eta": "52:09:54", "max_grad_norm": 1.0, "loss": 0.9059375524520874, "grad_norm": 0.5452545881271362, "learning_rate": 0.00018224956063268895}
+{"ts": "2025-12-26T22:23:54", "event": "train_log", "step": 1040, "epoch": 0.4388185654008439, "progress_pct": 7.31, "epoch_pct": 7.31, "eta": "52:06:46", "max_grad_norm": 1.0, "loss": 0.9708920121192932, "grad_norm": 0.5480839014053345, "learning_rate": 0.0001826010544815466}
+{"ts": "2025-12-26T22:24:10", "event": "train_log", "step": 1042, "epoch": 0.43966244725738396, "progress_pct": 7.33, "epoch_pct": 7.33, "eta": "52:03:46", "max_grad_norm": 1.0, "loss": 0.8561046123504639, "grad_norm": 0.44693151116371155, "learning_rate": 0.00018295254833040423}
+{"ts": "2025-12-26T22:24:27", "event": "train_log", "step": 1044, "epoch": 0.44050632911392407, "progress_pct": 7.34, "epoch_pct": 7.34, "eta": "52:00:46", "max_grad_norm": 1.0, "loss": 0.8811968564987183, "grad_norm": 0.4024234116077423, "learning_rate": 0.00018330404217926187}
+{"ts": "2025-12-26T22:24:42", "event": "train_log", "step": 1046, "epoch": 0.4413502109704641, "progress_pct": 7.36, "epoch_pct": 7.36, "eta": "51:57:36", "max_grad_norm": 1.0, "loss": 0.978068470954895, "grad_norm": 0.5586408376693726, "learning_rate": 0.0001836555360281195}
+{"ts": "2025-12-26T22:24:58", "event": "train_log", "step": 1048, "epoch": 0.4421940928270042, "progress_pct": 7.37, "epoch_pct": 7.37, "eta": "51:54:26", "max_grad_norm": 1.0, "loss": 0.8640981316566467, "grad_norm": 0.4918624758720398, "learning_rate": 0.00018400702987697714}
+{"ts": "2025-12-26T22:25:14", "event": "train_log", "step": 1050, "epoch": 0.4430379746835443, "progress_pct": 7.38, "epoch_pct": 7.38, "eta": "51:51:27", "max_grad_norm": 1.0, "loss": 0.9077964425086975, "grad_norm": 0.4335230588912964, "learning_rate": 0.0001843585237258348}
+{"ts": "2025-12-26T22:25:30", "event": "train_log", "step": 1052, "epoch": 0.4438818565400844, "progress_pct": 7.4, "epoch_pct": 7.4, "eta": "51:48:21", "max_grad_norm": 1.0, "loss": 0.960682213306427, "grad_norm": 0.5275123715400696, "learning_rate": 0.00018471001757469247}
+{"ts": "2025-12-26T22:25:45", "event": "train_log", "step": 1054, "epoch": 0.4447257383966245, "progress_pct": 7.41, "epoch_pct": 7.41, "eta": "51:45:04", "max_grad_norm": 1.0, "loss": 0.8549577593803406, "grad_norm": 0.5144415497779846, "learning_rate": 0.0001850615114235501}
+{"ts": "2025-12-26T22:25:59", "event": "train_log", "step": 1056, "epoch": 0.44556962025316454, "progress_pct": 7.43, "epoch_pct": 7.43, "eta": "51:41:39", "max_grad_norm": 1.0, "loss": 0.9320827126502991, "grad_norm": 0.5166662931442261, "learning_rate": 0.00018541300527240775}
+{"ts": "2025-12-26T22:26:13", "event": "train_log", "step": 1058, "epoch": 0.44641350210970465, "progress_pct": 7.44, "epoch_pct": 7.44, "eta": "51:38:16", "max_grad_norm": 1.0, "loss": 0.9632431864738464, "grad_norm": 0.5011980533599854, "learning_rate": 0.0001857644991212654}
+{"ts": "2025-12-26T22:26:28", "event": "train_log", "step": 1060, "epoch": 0.4472573839662447, "progress_pct": 7.45, "epoch_pct": 7.45, "eta": "51:34:58", "max_grad_norm": 1.0, "loss": 0.9442946910858154, "grad_norm": 0.49312469363212585, "learning_rate": 0.00018611599297012303}
+{"ts": "2025-12-26T22:26:42", "event": "train_log", "step": 1062, "epoch": 0.4481012658227848, "progress_pct": 7.47, "epoch_pct": 7.47, "eta": "51:31:39", "max_grad_norm": 1.0, "loss": 0.952802300453186, "grad_norm": 0.49958568811416626, "learning_rate": 0.00018646748681898066}
+{"ts": "2025-12-26T22:26:57", "event": "train_log", "step": 1064, "epoch": 0.4489451476793249, "progress_pct": 7.48, "epoch_pct": 7.48, "eta": "51:28:21", "max_grad_norm": 1.0, "loss": 0.9630650877952576, "grad_norm": 0.49827462434768677, "learning_rate": 0.00018681898066783833}
+{"ts": "2025-12-26T22:27:11", "event": "train_log", "step": 1066, "epoch": 0.44978902953586497, "progress_pct": 7.5, "epoch_pct": 7.5, "eta": "51:24:58", "max_grad_norm": 1.0, "loss": 0.9081395268440247, "grad_norm": 0.523980438709259, "learning_rate": 0.00018717047451669597}
+{"ts": "2025-12-26T22:27:26", "event": "train_log", "step": 1068, "epoch": 0.4506329113924051, "progress_pct": 7.51, "epoch_pct": 7.51, "eta": "51:21:57", "max_grad_norm": 1.0, "loss": 0.9398958683013916, "grad_norm": 0.5108568668365479, "learning_rate": 0.0001875219683655536}
+{"ts": "2025-12-26T22:27:41", "event": "train_log", "step": 1070, "epoch": 0.45147679324894513, "progress_pct": 7.52, "epoch_pct": 7.52, "eta": "51:18:43", "max_grad_norm": 1.0, "loss": 0.93592768907547, "grad_norm": 0.4453965127468109, "learning_rate": 0.00018787346221441127}
+{"ts": "2025-12-26T22:27:55", "event": "train_log", "step": 1072, "epoch": 0.45232067510548524, "progress_pct": 7.54, "epoch_pct": 7.54, "eta": "51:15:29", "max_grad_norm": 1.0, "loss": 0.8879633545875549, "grad_norm": 0.4675683081150055, "learning_rate": 0.0001882249560632689}
+{"ts": "2025-12-26T22:28:11", "event": "train_log", "step": 1074, "epoch": 0.4531645569620253, "progress_pct": 7.55, "epoch_pct": 7.55, "eta": "51:12:22", "max_grad_norm": 1.0, "loss": 0.9966004490852356, "grad_norm": 0.47398847341537476, "learning_rate": 0.00018857644991212655}
+{"ts": "2025-12-26T22:28:25", "event": "train_log", "step": 1076, "epoch": 0.4540084388185654, "progress_pct": 7.57, "epoch_pct": 7.57, "eta": "51:09:13", "max_grad_norm": 1.0, "loss": 0.9184179902076721, "grad_norm": 0.46053192019462585, "learning_rate": 0.0001889279437609842}
+{"ts": "2025-12-26T22:28:40", "event": "train_log", "step": 1078, "epoch": 0.4548523206751055, "progress_pct": 7.58, "epoch_pct": 7.58, "eta": "51:06:06", "max_grad_norm": 1.0, "loss": 0.9434974193572998, "grad_norm": 0.5601398348808289, "learning_rate": 0.00018927943760984182}
+{"ts": "2025-12-26T22:28:55", "event": "train_log", "step": 1080, "epoch": 0.45569620253164556, "progress_pct": 7.59, "epoch_pct": 7.59, "eta": "51:02:54", "max_grad_norm": 1.0, "loss": 0.9522465467453003, "grad_norm": 0.48422637581825256, "learning_rate": 0.0001896309314586995}
+{"ts": "2025-12-26T22:29:09", "event": "train_log", "step": 1082, "epoch": 0.45654008438818566, "progress_pct": 7.61, "epoch_pct": 7.61, "eta": "50:59:42", "max_grad_norm": 1.0, "loss": 0.9315434694290161, "grad_norm": 0.46280911564826965, "learning_rate": 0.00018998242530755713}
+{"ts": "2025-12-26T22:29:25", "event": "train_log", "step": 1084, "epoch": 0.4573839662447257, "progress_pct": 7.62, "epoch_pct": 7.62, "eta": "50:56:41", "max_grad_norm": 1.0, "loss": 0.8165783286094666, "grad_norm": 0.4658683240413666, "learning_rate": 0.00019033391915641477}
+{"ts": "2025-12-26T22:29:40", "event": "train_log", "step": 1086, "epoch": 0.4582278481012658, "progress_pct": 7.64, "epoch_pct": 7.64, "eta": "50:53:37", "max_grad_norm": 1.0, "loss": 0.8024274110794067, "grad_norm": 0.4192182123661041, "learning_rate": 0.0001906854130052724}
+{"ts": "2025-12-26T22:29:56", "event": "train_log", "step": 1088, "epoch": 0.45907172995780593, "progress_pct": 7.65, "epoch_pct": 7.65, "eta": "50:50:52", "max_grad_norm": 1.0, "loss": 0.8620653748512268, "grad_norm": 0.4178735613822937, "learning_rate": 0.00019103690685413007}
+{"ts": "2025-12-26T22:30:11", "event": "train_log", "step": 1090, "epoch": 0.459915611814346, "progress_pct": 7.67, "epoch_pct": 7.67, "eta": "50:47:46", "max_grad_norm": 1.0, "loss": 0.9121530055999756, "grad_norm": 0.4488574266433716, "learning_rate": 0.0001913884007029877}
+{"ts": "2025-12-26T22:30:25", "event": "train_log", "step": 1092, "epoch": 0.4607594936708861, "progress_pct": 7.68, "epoch_pct": 7.68, "eta": "50:44:34", "max_grad_norm": 1.0, "loss": 0.9496700763702393, "grad_norm": 0.5164965987205505, "learning_rate": 0.00019173989455184535}
+{"ts": "2025-12-26T22:30:40", "event": "train_log", "step": 1094, "epoch": 0.46160337552742614, "progress_pct": 7.69, "epoch_pct": 7.69, "eta": "50:41:34", "max_grad_norm": 1.0, "loss": 0.8757708668708801, "grad_norm": 0.41563132405281067, "learning_rate": 0.00019209138840070299}
+{"ts": "2025-12-26T22:30:56", "event": "train_log", "step": 1096, "epoch": 0.46244725738396625, "progress_pct": 7.71, "epoch_pct": 7.71, "eta": "50:38:35", "max_grad_norm": 1.0, "loss": 0.8406533002853394, "grad_norm": 2.3222429752349854, "learning_rate": 0.00019244288224956065}
+{"ts": "2025-12-26T22:31:10", "event": "train_log", "step": 1098, "epoch": 0.46329113924050636, "progress_pct": 7.72, "epoch_pct": 7.72, "eta": "50:35:33", "max_grad_norm": 1.0, "loss": 0.9459465742111206, "grad_norm": 0.4813845157623291, "learning_rate": 0.0001927943760984183}
+{"ts": "2025-12-26T22:31:24", "event": "train_log", "step": 1100, "epoch": 0.4641350210970464, "progress_pct": 7.74, "epoch_pct": 7.74, "eta": "50:32:13", "max_grad_norm": 1.0, "loss": 0.9376904368400574, "grad_norm": 0.6233882308006287, "learning_rate": 0.00019314586994727593}
+{"ts": "2025-12-26T22:42:41", "event": "train_log", "step": 1100, "epoch": 0.4641350210970464, "progress_pct": 7.74, "epoch_pct": 7.74, "eta": "52:46:47", "max_grad_norm": 1.0, "eval_loss": 0.9357889294624329, "eval_runtime": 676.9573, "eval_samples_per_second": 3.112, "eval_steps_per_second": 3.112}
+{"ts": "2025-12-26T22:42:57", "event": "train_log", "step": 1102, "epoch": 0.4649789029535865, "progress_pct": 7.75, "epoch_pct": 7.75, "eta": "52:43:45", "max_grad_norm": 1.0, "loss": 0.8998825550079346, "grad_norm": 0.5125579237937927, "learning_rate": 0.00019349736379613357}
+{"ts": "2025-12-26T22:43:13", "event": "train_log", "step": 1104, "epoch": 0.46582278481012657, "progress_pct": 7.76, "epoch_pct": 7.76, "eta": "52:40:41", "max_grad_norm": 1.0, "loss": 0.9213768839836121, "grad_norm": 0.4534320831298828, "learning_rate": 0.00019384885764499123}
+{"ts": "2025-12-26T22:43:29", "event": "train_log", "step": 1106, "epoch": 0.4666666666666667, "progress_pct": 7.78, "epoch_pct": 7.78, "eta": "52:37:39", "max_grad_norm": 1.0, "loss": 0.9739661812782288, "grad_norm": 0.4715143144130707, "learning_rate": 0.00019420035149384887}
+{"ts": "2025-12-26T22:43:43", "event": "train_log", "step": 1108, "epoch": 0.4675105485232067, "progress_pct": 7.79, "epoch_pct": 7.79, "eta": "52:34:20", "max_grad_norm": 1.0, "loss": 0.9113216400146484, "grad_norm": 0.45529672503471375, "learning_rate": 0.0001945518453427065}
+{"ts": "2025-12-26T22:43:58", "event": "train_log", "step": 1110, "epoch": 0.46835443037974683, "progress_pct": 7.81, "epoch_pct": 7.81, "eta": "52:31:08", "max_grad_norm": 1.0, "loss": 0.92528235912323, "grad_norm": 0.48199015855789185, "learning_rate": 0.00019490333919156417}
+{"ts": "2025-12-26T22:44:14", "event": "train_log", "step": 1112, "epoch": 0.46919831223628694, "progress_pct": 7.82, "epoch_pct": 7.82, "eta": "52:28:02", "max_grad_norm": 1.0, "loss": 0.8612716197967529, "grad_norm": 0.4425188899040222, "learning_rate": 0.0001952548330404218}
+{"ts": "2025-12-26T22:44:33", "event": "train_log", "step": 1114, "epoch": 0.470042194092827, "progress_pct": 7.83, "epoch_pct": 7.83, "eta": "52:25:38", "max_grad_norm": 1.0, "loss": 0.8901699185371399, "grad_norm": 0.4311593770980835, "learning_rate": 0.00019560632688927945}
+{"ts": "2025-12-26T22:44:48", "event": "train_log", "step": 1116, "epoch": 0.4708860759493671, "progress_pct": 7.85, "epoch_pct": 7.85, "eta": "52:22:27", "max_grad_norm": 1.0, "loss": 0.9271994829177856, "grad_norm": 0.497806578874588, "learning_rate": 0.0001959578207381371}
+{"ts": "2025-12-26T22:45:04", "event": "train_log", "step": 1118, "epoch": 0.47172995780590715, "progress_pct": 7.86, "epoch_pct": 7.86, "eta": "52:19:34", "max_grad_norm": 1.0, "loss": 0.8740925788879395, "grad_norm": 0.47149473428726196, "learning_rate": 0.00019630931458699473}
+{"ts": "2025-12-26T22:45:20", "event": "train_log", "step": 1120, "epoch": 0.47257383966244726, "progress_pct": 7.88, "epoch_pct": 7.88, "eta": "52:16:38", "max_grad_norm": 1.0, "loss": 0.9190115928649902, "grad_norm": 0.5029966235160828, "learning_rate": 0.00019666080843585236}
+{"ts": "2025-12-26T22:45:36", "event": "train_log", "step": 1122, "epoch": 0.47341772151898737, "progress_pct": 7.89, "epoch_pct": 7.89, "eta": "52:13:34", "max_grad_norm": 1.0, "loss": 1.0043057203292847, "grad_norm": 0.47972601652145386, "learning_rate": 0.00019701230228471003}
+{"ts": "2025-12-26T22:45:52", "event": "train_log", "step": 1124, "epoch": 0.4742616033755274, "progress_pct": 7.9, "epoch_pct": 7.9, "eta": "52:10:36", "max_grad_norm": 1.0, "loss": 0.9472925066947937, "grad_norm": 0.45096471905708313, "learning_rate": 0.00019736379613356767}
+{"ts": "2025-12-26T22:46:07", "event": "train_log", "step": 1126, "epoch": 0.4751054852320675, "progress_pct": 7.92, "epoch_pct": 7.92, "eta": "52:07:33", "max_grad_norm": 1.0, "loss": 0.8043124079704285, "grad_norm": 0.459852933883667, "learning_rate": 0.00019771528998242533}
+{"ts": "2025-12-26T22:46:23", "event": "train_log", "step": 1128, "epoch": 0.4759493670886076, "progress_pct": 7.93, "epoch_pct": 7.93, "eta": "52:04:34", "max_grad_norm": 1.0, "loss": 0.9662142395973206, "grad_norm": 0.5313422679901123, "learning_rate": 0.00019806678383128297}
+{"ts": "2025-12-26T22:46:39", "event": "train_log", "step": 1130, "epoch": 0.4767932489451477, "progress_pct": 7.95, "epoch_pct": 7.95, "eta": "52:01:39", "max_grad_norm": 1.0, "loss": 0.8691151738166809, "grad_norm": 0.43474531173706055, "learning_rate": 0.0001984182776801406}
+{"ts": "2025-12-26T22:46:55", "event": "train_log", "step": 1132, "epoch": 0.47763713080168774, "progress_pct": 7.96, "epoch_pct": 7.96, "eta": "51:58:43", "max_grad_norm": 1.0, "loss": 0.9024254083633423, "grad_norm": 0.5232312083244324, "learning_rate": 0.00019876977152899825}
+{"ts": "2025-12-26T22:47:12", "event": "train_log", "step": 1134, "epoch": 0.47848101265822784, "progress_pct": 7.97, "epoch_pct": 7.97, "eta": "51:56:07", "max_grad_norm": 1.0, "loss": 0.8996873497962952, "grad_norm": 0.47671905159950256, "learning_rate": 0.0001991212653778559}
+{"ts": "2025-12-26T22:47:28", "event": "train_log", "step": 1136, "epoch": 0.47932489451476795, "progress_pct": 7.99, "epoch_pct": 7.99, "eta": "51:53:07", "max_grad_norm": 1.0, "loss": 0.8336917757987976, "grad_norm": 0.42289480566978455, "learning_rate": 0.00019947275922671353}
+{"ts": "2025-12-26T22:47:43", "event": "train_log", "step": 1138, "epoch": 0.480168776371308, "progress_pct": 8.0, "epoch_pct": 8.0, "eta": "51:50:03", "max_grad_norm": 1.0, "loss": 0.8943206071853638, "grad_norm": 0.4700844883918762, "learning_rate": 0.0001998242530755712}
+{"ts": "2025-12-26T22:47:59", "event": "train_log", "step": 1140, "epoch": 0.4810126582278481, "progress_pct": 8.02, "epoch_pct": 8.02, "eta": "51:47:11", "max_grad_norm": 1.0, "loss": 0.9340365529060364, "grad_norm": 0.5341399312019348, "learning_rate": 0.00019999999711649004}
+{"ts": "2025-12-26T22:48:14", "event": "train_log", "step": 1142, "epoch": 0.48185654008438816, "progress_pct": 8.03, "epoch_pct": 8.03, "eta": "51:44:14", "max_grad_norm": 1.0, "loss": 0.8241778016090393, "grad_norm": 0.46169522404670715, "learning_rate": 0.00019999997404841123}
+{"ts": "2025-12-26T22:48:28", "event": "train_log", "step": 1144, "epoch": 0.48270042194092827, "progress_pct": 8.05, "epoch_pct": 8.05, "eta": "51:40:52", "max_grad_norm": 1.0, "loss": 1.0096158981323242, "grad_norm": 0.5475223064422607, "learning_rate": 0.00019999992791225896}
+{"ts": "2025-12-26T22:48:44", "event": "train_log", "step": 1146, "epoch": 0.4835443037974684, "progress_pct": 8.06, "epoch_pct": 8.06, "eta": "51:38:04", "max_grad_norm": 1.0, "loss": 0.9650378227233887, "grad_norm": 0.524641215801239, "learning_rate": 0.00019999985870804385}
+{"ts": "2025-12-26T22:49:00", "event": "train_log", "step": 1148, "epoch": 0.48438818565400843, "progress_pct": 8.07, "epoch_pct": 8.07, "eta": "51:35:11", "max_grad_norm": 1.0, "loss": 0.9003009796142578, "grad_norm": 0.4326174855232239, "learning_rate": 0.00019999976643578186}
+{"ts": "2025-12-26T22:49:14", "event": "train_log", "step": 1150, "epoch": 0.48523206751054854, "progress_pct": 8.09, "epoch_pct": 8.09, "eta": "51:32:06", "max_grad_norm": 1.0, "loss": 0.9793432950973511, "grad_norm": 0.49034059047698975, "learning_rate": 0.0001999996510954943}
+{"ts": "2025-12-26T22:49:29", "event": "train_log", "step": 1152, "epoch": 0.4860759493670886, "progress_pct": 8.1, "epoch_pct": 8.1, "eta": "51:29:06", "max_grad_norm": 1.0, "loss": 0.8793007731437683, "grad_norm": 0.4760092496871948, "learning_rate": 0.00019999951268720776}
+{"ts": "2025-12-26T22:49:44", "event": "train_log", "step": 1154, "epoch": 0.4869198312236287, "progress_pct": 8.12, "epoch_pct": 8.12, "eta": "51:26:05", "max_grad_norm": 1.0, "loss": 0.888630747795105, "grad_norm": 0.44606879353523254, "learning_rate": 0.00019999935121095417}
+{"ts": "2025-12-26T22:49:59", "event": "train_log", "step": 1156, "epoch": 0.4877637130801688, "progress_pct": 8.13, "epoch_pct": 8.13, "eta": "51:23:09", "max_grad_norm": 1.0, "loss": 0.9660000801086426, "grad_norm": 0.5030332207679749, "learning_rate": 0.0001999991666667708}
+{"ts": "2025-12-26T22:50:14", "event": "train_log", "step": 1158, "epoch": 0.48860759493670886, "progress_pct": 8.14, "epoch_pct": 8.14, "eta": "51:20:04", "max_grad_norm": 1.0, "loss": 0.824654757976532, "grad_norm": 0.4295555055141449, "learning_rate": 0.00019999895905470014}
+{"ts": "2025-12-26T22:50:31", "event": "train_log", "step": 1160, "epoch": 0.48945147679324896, "progress_pct": 8.16, "epoch_pct": 8.16, "eta": "51:17:31", "max_grad_norm": 1.0, "loss": 0.8159562349319458, "grad_norm": 0.392167866230011, "learning_rate": 0.0001999987283747902}
+{"ts": "2025-12-26T22:50:46", "event": "train_log", "step": 1162, "epoch": 0.490295358649789, "progress_pct": 8.17, "epoch_pct": 8.17, "eta": "51:14:33", "max_grad_norm": 1.0, "loss": 0.9630686044692993, "grad_norm": 0.484611839056015, "learning_rate": 0.00019999847462709412}
+{"ts": "2025-12-26T22:51:03", "event": "train_log", "step": 1164, "epoch": 0.4911392405063291, "progress_pct": 8.19, "epoch_pct": 8.19, "eta": "51:11:53", "max_grad_norm": 1.0, "loss": 0.8396129608154297, "grad_norm": 0.45891445875167847, "learning_rate": 0.00019999819781167042}
+{"ts": "2025-12-26T22:51:19", "event": "train_log", "step": 1166, "epoch": 0.4919831223628692, "progress_pct": 8.2, "epoch_pct": 8.2, "eta": "51:09:17", "max_grad_norm": 1.0, "loss": 0.8194513916969299, "grad_norm": 0.5553452968597412, "learning_rate": 0.00019999789792858304}
+{"ts": "2025-12-26T22:51:35", "event": "train_log", "step": 1168, "epoch": 0.4928270042194093, "progress_pct": 8.21, "epoch_pct": 8.21, "eta": "51:06:31", "max_grad_norm": 1.0, "loss": 0.9268721342086792, "grad_norm": 0.38998672366142273, "learning_rate": 0.00019999757497790106}
+{"ts": "2025-12-26T22:51:50", "event": "train_log", "step": 1170, "epoch": 0.4936708860759494, "progress_pct": 8.23, "epoch_pct": 8.23, "eta": "51:03:38", "max_grad_norm": 1.0, "loss": 0.9204684495925903, "grad_norm": 0.45445355772972107, "learning_rate": 0.00019999722895969904}
+{"ts": "2025-12-26T22:52:05", "event": "train_log", "step": 1172, "epoch": 0.49451476793248944, "progress_pct": 8.24, "epoch_pct": 8.24, "eta": "51:00:41", "max_grad_norm": 1.0, "loss": 0.9270301461219788, "grad_norm": 0.484225332736969, "learning_rate": 0.00019999685987405678}
+{"ts": "2025-12-26T22:52:21", "event": "train_log", "step": 1174, "epoch": 0.49535864978902955, "progress_pct": 8.26, "epoch_pct": 8.26, "eta": "50:57:56", "max_grad_norm": 1.0, "loss": 0.8782645463943481, "grad_norm": 0.45215415954589844, "learning_rate": 0.00019999646772105942}
+{"ts": "2025-12-26T22:52:38", "event": "train_log", "step": 1176, "epoch": 0.4962025316455696, "progress_pct": 8.27, "epoch_pct": 8.27, "eta": "50:55:21", "max_grad_norm": 1.0, "loss": 0.873112678527832, "grad_norm": 0.41578832268714905, "learning_rate": 0.00019999605250079744}
+{"ts": "2025-12-26T22:52:55", "event": "train_log", "step": 1178, "epoch": 0.4970464135021097, "progress_pct": 8.28, "epoch_pct": 8.28, "eta": "50:52:47", "max_grad_norm": 1.0, "loss": 0.9593189358711243, "grad_norm": 0.4632788300514221, "learning_rate": 0.0001999956142133666}
+{"ts": "2025-12-26T22:53:10", "event": "train_log", "step": 1180, "epoch": 0.4978902953586498, "progress_pct": 8.3, "epoch_pct": 8.3, "eta": "50:49:58", "max_grad_norm": 1.0, "loss": 0.9215621948242188, "grad_norm": 0.5007622838020325, "learning_rate": 0.000199995152858868}
+{"ts": "2025-12-26T22:53:24", "event": "train_log", "step": 1182, "epoch": 0.49873417721518987, "progress_pct": 8.31, "epoch_pct": 8.31, "eta": "50:46:55", "max_grad_norm": 1.0, "loss": 0.8964219689369202, "grad_norm": 0.5201935768127441, "learning_rate": 0.0001999946684374081}
+{"ts": "2025-12-26T22:53:38", "event": "train_log", "step": 1184, "epoch": 0.49957805907173, "progress_pct": 8.33, "epoch_pct": 8.33, "eta": "50:43:54", "max_grad_norm": 1.0, "loss": 0.8528663516044617, "grad_norm": 0.5053967237472534, "learning_rate": 0.0001999941609490986}
+{"ts": "2025-12-26T22:53:54", "event": "train_log", "step": 1186, "epoch": 0.5004219409282701, "progress_pct": 8.34, "epoch_pct": 8.34, "eta": "50:41:18", "max_grad_norm": 1.0, "loss": 0.8214734792709351, "grad_norm": 0.44397974014282227, "learning_rate": 0.0001999936303940566}
+{"ts": "2025-12-26T22:54:10", "event": "train_log", "step": 1188, "epoch": 0.5012658227848101, "progress_pct": 8.35, "epoch_pct": 8.35, "eta": "50:38:29", "max_grad_norm": 1.0, "loss": 0.8882166743278503, "grad_norm": 0.4563101530075073, "learning_rate": 0.0001999930767724045}
+{"ts": "2025-12-26T22:54:25", "event": "train_log", "step": 1190, "epoch": 0.5021097046413502, "progress_pct": 8.37, "epoch_pct": 8.37, "eta": "50:35:43", "max_grad_norm": 1.0, "loss": 0.8882588148117065, "grad_norm": 0.4666728675365448, "learning_rate": 0.00019999250008426997}
+{"ts": "2025-12-26T22:54:41", "event": "train_log", "step": 1192, "epoch": 0.5029535864978903, "progress_pct": 8.38, "epoch_pct": 8.38, "eta": "50:33:07", "max_grad_norm": 1.0, "loss": 0.8353691697120667, "grad_norm": 0.4423629939556122, "learning_rate": 0.00019999190032978607}
+{"ts": "2025-12-26T22:54:58", "event": "train_log", "step": 1194, "epoch": 0.5037974683544304, "progress_pct": 8.4, "epoch_pct": 8.4, "eta": "50:30:39", "max_grad_norm": 1.0, "loss": 0.8721219301223755, "grad_norm": 0.4514595866203308, "learning_rate": 0.00019999127750909118}
+{"ts": "2025-12-26T22:55:13", "event": "train_log", "step": 1196, "epoch": 0.5046413502109705, "progress_pct": 8.41, "epoch_pct": 8.41, "eta": "50:27:54", "max_grad_norm": 1.0, "loss": 0.902795135974884, "grad_norm": 0.48469996452331543, "learning_rate": 0.00019999063162232886}
+{"ts": "2025-12-26T22:55:29", "event": "train_log", "step": 1198, "epoch": 0.5054852320675105, "progress_pct": 8.42, "epoch_pct": 8.42, "eta": "50:25:15", "max_grad_norm": 1.0, "loss": 0.8843175768852234, "grad_norm": 0.4931983947753906, "learning_rate": 0.00019998996266964823}
+{"ts": "2025-12-26T22:55:45", "event": "train_log", "step": 1200, "epoch": 0.5063291139240507, "progress_pct": 8.44, "epoch_pct": 8.44, "eta": "50:22:33", "max_grad_norm": 1.0, "loss": 0.9179208278656006, "grad_norm": 0.507408082485199, "learning_rate": 0.00019998927065120357}
+{"ts": "2025-12-26T23:06:55", "event": "train_log", "step": 1200, "epoch": 0.5063291139240507, "progress_pct": 8.44, "epoch_pct": 8.44, "eta": "52:23:40", "max_grad_norm": 1.0, "eval_loss": 0.9224098324775696, "eval_runtime": 669.7542, "eval_samples_per_second": 3.146, "eval_steps_per_second": 3.146}
+{"ts": "2025-12-26T23:07:09", "event": "train_log", "step": 1202, "epoch": 0.5071729957805907, "progress_pct": 8.45, "epoch_pct": 8.45, "eta": "52:20:34", "max_grad_norm": 1.0, "loss": 0.9078981876373291, "grad_norm": 0.4024188220500946, "learning_rate": 0.00019998855556715447}
+{"ts": "2025-12-26T23:07:24", "event": "train_log", "step": 1204, "epoch": 0.5080168776371308, "progress_pct": 8.47, "epoch_pct": 8.47, "eta": "52:17:36", "max_grad_norm": 1.0, "loss": 0.8089252710342407, "grad_norm": 0.6558667421340942, "learning_rate": 0.00019998781741766594}
+{"ts": "2025-12-26T23:07:38", "event": "train_log", "step": 1206, "epoch": 0.5088607594936709, "progress_pct": 8.48, "epoch_pct": 8.48, "eta": "52:14:29", "max_grad_norm": 1.0, "loss": 0.9316248297691345, "grad_norm": 0.45128145813941956, "learning_rate": 0.00019998705620290823}
+{"ts": "2025-12-26T23:07:53", "event": "train_log", "step": 1208, "epoch": 0.509704641350211, "progress_pct": 8.5, "epoch_pct": 8.5, "eta": "52:11:31", "max_grad_norm": 1.0, "loss": 0.9047867655754089, "grad_norm": 0.4733511507511139, "learning_rate": 0.00019998627192305694}
+{"ts": "2025-12-26T23:08:09", "event": "train_log", "step": 1210, "epoch": 0.510548523206751, "progress_pct": 8.51, "epoch_pct": 8.51, "eta": "52:08:34", "max_grad_norm": 1.0, "loss": 0.8507166504859924, "grad_norm": 0.4026021659374237, "learning_rate": 0.00019998546457829298}
+{"ts": "2025-12-26T23:08:23", "event": "train_log", "step": 1212, "epoch": 0.5113924050632911, "progress_pct": 8.52, "epoch_pct": 8.52, "eta": "52:05:33", "max_grad_norm": 1.0, "loss": 0.8490090370178223, "grad_norm": 0.43416184186935425, "learning_rate": 0.00019998463416880262}
+{"ts": "2025-12-26T23:08:41", "event": "train_log", "step": 1214, "epoch": 0.5122362869198313, "progress_pct": 8.54, "epoch_pct": 8.54, "eta": "52:03:01", "max_grad_norm": 1.0, "loss": 0.7730492949485779, "grad_norm": 0.3678364157676697, "learning_rate": 0.0001999837806947774}
+{"ts": "2025-12-26T23:08:55", "event": "train_log", "step": 1216, "epoch": 0.5130801687763713, "progress_pct": 8.55, "epoch_pct": 8.55, "eta": "51:59:55", "max_grad_norm": 1.0, "loss": 0.9625269174575806, "grad_norm": 0.49042677879333496, "learning_rate": 0.00019998290415641415}
+{"ts": "2025-12-26T23:09:09", "event": "train_log", "step": 1218, "epoch": 0.5139240506329114, "progress_pct": 8.57, "epoch_pct": 8.57, "eta": "51:56:54", "max_grad_norm": 1.0, "loss": 0.8961732983589172, "grad_norm": 0.4576701819896698, "learning_rate": 0.00019998200455391516}
+{"ts": "2025-12-26T23:09:24", "event": "train_log", "step": 1220, "epoch": 0.5147679324894515, "progress_pct": 8.58, "epoch_pct": 8.58, "eta": "51:53:54", "max_grad_norm": 1.0, "loss": 0.8508996367454529, "grad_norm": 0.4465518593788147, "learning_rate": 0.00019998108188748793}
+{"ts": "2025-12-26T23:09:38", "event": "train_log", "step": 1222, "epoch": 0.5156118143459916, "progress_pct": 8.59, "epoch_pct": 8.59, "eta": "51:50:52", "max_grad_norm": 1.0, "loss": 0.8664930462837219, "grad_norm": 0.4267960488796234, "learning_rate": 0.00019998013615734524}
+{"ts": "2025-12-26T23:09:54", "event": "train_log", "step": 1224, "epoch": 0.5164556962025316, "progress_pct": 8.61, "epoch_pct": 8.61, "eta": "51:48:10", "max_grad_norm": 1.0, "loss": 0.8841317892074585, "grad_norm": 0.39936602115631104, "learning_rate": 0.0001999791673637053}
+{"ts": "2025-12-26T23:10:09", "event": "train_log", "step": 1226, "epoch": 0.5172995780590718, "progress_pct": 8.62, "epoch_pct": 8.62, "eta": "51:45:16", "max_grad_norm": 1.0, "loss": 0.8601276874542236, "grad_norm": 0.4442414343357086, "learning_rate": 0.0001999781755067916}
+{"ts": "2025-12-26T23:10:25", "event": "train_log", "step": 1228, "epoch": 0.5181434599156118, "progress_pct": 8.64, "epoch_pct": 8.64, "eta": "51:42:32", "max_grad_norm": 1.0, "loss": 0.8377046585083008, "grad_norm": 0.4397028088569641, "learning_rate": 0.00019997716058683292}
+{"ts": "2025-12-26T23:10:41", "event": "train_log", "step": 1230, "epoch": 0.5189873417721519, "progress_pct": 8.65, "epoch_pct": 8.65, "eta": "51:39:42", "max_grad_norm": 1.0, "loss": 0.9853615164756775, "grad_norm": 0.5327648520469666, "learning_rate": 0.0001999761226040634}
+{"ts": "2025-12-26T23:10:57", "event": "train_log", "step": 1232, "epoch": 0.5198312236286919, "progress_pct": 8.66, "epoch_pct": 8.66, "eta": "51:37:10", "max_grad_norm": 1.0, "loss": 0.9235715866088867, "grad_norm": 0.40931421518325806, "learning_rate": 0.00019997506155872244}
+{"ts": "2025-12-26T23:11:13", "event": "train_log", "step": 1234, "epoch": 0.5206751054852321, "progress_pct": 8.68, "epoch_pct": 8.68, "eta": "51:34:23", "max_grad_norm": 1.0, "loss": 0.868396520614624, "grad_norm": 0.4599161148071289, "learning_rate": 0.00019997397745105487}
+{"ts": "2025-12-26T23:11:28", "event": "train_log", "step": 1236, "epoch": 0.5215189873417722, "progress_pct": 8.69, "epoch_pct": 8.69, "eta": "51:31:36", "max_grad_norm": 1.0, "loss": 0.9946733117103577, "grad_norm": 0.4813833236694336, "learning_rate": 0.0001999728702813107}
+{"ts": "2025-12-26T23:11:44", "event": "train_log", "step": 1238, "epoch": 0.5223628691983122, "progress_pct": 8.71, "epoch_pct": 8.71, "eta": "51:28:52", "max_grad_norm": 1.0, "loss": 0.8608635067939758, "grad_norm": 0.43518269062042236, "learning_rate": 0.00019997174004974543}
+{"ts": "2025-12-26T23:12:00", "event": "train_log", "step": 1240, "epoch": 0.5232067510548524, "progress_pct": 8.72, "epoch_pct": 8.72, "eta": "51:26:15", "max_grad_norm": 1.0, "loss": 0.819354772567749, "grad_norm": 0.45646214485168457, "learning_rate": 0.0001999705867566197}
+{"ts": "2025-12-26T23:12:16", "event": "train_log", "step": 1242, "epoch": 0.5240506329113924, "progress_pct": 8.73, "epoch_pct": 8.73, "eta": "51:23:35", "max_grad_norm": 1.0, "loss": 0.8934658765792847, "grad_norm": 0.46834197640419006, "learning_rate": 0.00019996941040219954}
+{"ts": "2025-12-26T23:12:33", "event": "train_log", "step": 1244, "epoch": 0.5248945147679325, "progress_pct": 8.75, "epoch_pct": 8.75, "eta": "51:20:59", "max_grad_norm": 1.0, "loss": 0.8624778985977173, "grad_norm": 0.46065032482147217, "learning_rate": 0.0001999682109867564}
+{"ts": "2025-12-26T23:12:48", "event": "train_log", "step": 1246, "epoch": 0.5257383966244725, "progress_pct": 8.76, "epoch_pct": 8.76, "eta": "51:18:12", "max_grad_norm": 1.0, "loss": 0.8540882468223572, "grad_norm": 0.4506741166114807, "learning_rate": 0.00019996698851056688}
+{"ts": "2025-12-26T23:13:03", "event": "train_log", "step": 1248, "epoch": 0.5265822784810127, "progress_pct": 8.78, "epoch_pct": 8.78, "eta": "51:15:30", "max_grad_norm": 1.0, "loss": 0.9003006219863892, "grad_norm": 0.4550519287586212, "learning_rate": 0.00019996574297391302}
+{"ts": "2025-12-26T23:13:19", "event": "train_log", "step": 1250, "epoch": 0.5274261603375527, "progress_pct": 8.79, "epoch_pct": 8.79, "eta": "51:12:47", "max_grad_norm": 1.0, "loss": 0.8490248918533325, "grad_norm": 0.4097813367843628, "learning_rate": 0.00019996447437708214}
+{"ts": "2025-12-26T23:13:35", "event": "train_log", "step": 1252, "epoch": 0.5282700421940928, "progress_pct": 8.8, "epoch_pct": 8.8, "eta": "51:10:10", "max_grad_norm": 1.0, "loss": 0.9204569458961487, "grad_norm": 0.4375711679458618, "learning_rate": 0.00019996318272036688}
+{"ts": "2025-12-26T23:13:51", "event": "train_log", "step": 1254, "epoch": 0.529113924050633, "progress_pct": 8.82, "epoch_pct": 8.82, "eta": "51:07:35", "max_grad_norm": 1.0, "loss": 0.8831789493560791, "grad_norm": 0.4335198998451233, "learning_rate": 0.00019996186800406518}
+{"ts": "2025-12-26T23:14:06", "event": "train_log", "step": 1256, "epoch": 0.529957805907173, "progress_pct": 8.83, "epoch_pct": 8.83, "eta": "51:04:53", "max_grad_norm": 1.0, "loss": 0.8578064441680908, "grad_norm": 0.4840933084487915, "learning_rate": 0.00019996053022848035}
+{"ts": "2025-12-26T23:14:21", "event": "train_log", "step": 1258, "epoch": 0.5308016877637131, "progress_pct": 8.85, "epoch_pct": 8.85, "eta": "51:02:04", "max_grad_norm": 1.0, "loss": 0.9235416650772095, "grad_norm": 0.48911216855049133, "learning_rate": 0.00019995916939392097}
+{"ts": "2025-12-26T23:14:38", "event": "train_log", "step": 1260, "epoch": 0.5316455696202531, "progress_pct": 8.86, "epoch_pct": 8.86, "eta": "50:59:42", "max_grad_norm": 1.0, "loss": 0.7897764444351196, "grad_norm": 0.41106730699539185, "learning_rate": 0.00019995778550070098}
+{"ts": "2025-12-26T23:14:56", "event": "train_log", "step": 1262, "epoch": 0.5324894514767933, "progress_pct": 8.87, "epoch_pct": 8.87, "eta": "50:57:18", "max_grad_norm": 1.0, "loss": 0.9392989873886108, "grad_norm": 0.42233356833457947, "learning_rate": 0.00019995637854913957}
+{"ts": "2025-12-26T23:15:12", "event": "train_log", "step": 1264, "epoch": 0.5333333333333333, "progress_pct": 8.89, "epoch_pct": 8.89, "eta": "50:54:49", "max_grad_norm": 1.0, "loss": 0.9130091667175293, "grad_norm": 0.39145180583000183, "learning_rate": 0.0001999549485395613}
+{"ts": "2025-12-26T23:15:28", "event": "train_log", "step": 1266, "epoch": 0.5341772151898734, "progress_pct": 8.9, "epoch_pct": 8.9, "eta": "50:52:17", "max_grad_norm": 1.0, "loss": 0.9690561890602112, "grad_norm": 0.42780739068984985, "learning_rate": 0.00019995349547229614}
+{"ts": "2025-12-26T23:15:44", "event": "train_log", "step": 1268, "epoch": 0.5350210970464135, "progress_pct": 8.92, "epoch_pct": 8.92, "eta": "50:49:37", "max_grad_norm": 1.0, "loss": 0.9990131258964539, "grad_norm": 0.4823327362537384, "learning_rate": 0.0001999520193476792}
+{"ts": "2025-12-26T23:15:58", "event": "train_log", "step": 1270, "epoch": 0.5358649789029536, "progress_pct": 8.93, "epoch_pct": 8.93, "eta": "50:46:50", "max_grad_norm": 1.0, "loss": 0.8343052864074707, "grad_norm": 0.4669715166091919, "learning_rate": 0.00019995052016605097}
+{"ts": "2025-12-26T23:16:14", "event": "train_log", "step": 1272, "epoch": 0.5367088607594936, "progress_pct": 8.95, "epoch_pct": 8.95, "eta": "50:44:11", "max_grad_norm": 1.0, "loss": 0.9605970978736877, "grad_norm": 0.4024743437767029, "learning_rate": 0.0001999489979277574}
+{"ts": "2025-12-26T23:16:30", "event": "train_log", "step": 1274, "epoch": 0.5375527426160338, "progress_pct": 8.96, "epoch_pct": 8.96, "eta": "50:41:44", "max_grad_norm": 1.0, "loss": 0.8164438009262085, "grad_norm": 0.42063432931900024, "learning_rate": 0.0001999474526331495}
+{"ts": "2025-12-26T23:16:46", "event": "train_log", "step": 1276, "epoch": 0.5383966244725739, "progress_pct": 8.97, "epoch_pct": 8.97, "eta": "50:39:10", "max_grad_norm": 1.0, "loss": 0.879767656326294, "grad_norm": 0.47953495383262634, "learning_rate": 0.00019994588428258383}
+{"ts": "2025-12-26T23:17:02", "event": "train_log", "step": 1278, "epoch": 0.5392405063291139, "progress_pct": 8.99, "epoch_pct": 8.99, "eta": "50:36:40", "max_grad_norm": 1.0, "loss": 0.7658726572990417, "grad_norm": 0.39668557047843933, "learning_rate": 0.00019994429287642214}
+{"ts": "2025-12-26T23:17:18", "event": "train_log", "step": 1280, "epoch": 0.540084388185654, "progress_pct": 9.0, "epoch_pct": 9.0, "eta": "50:34:06", "max_grad_norm": 1.0, "loss": 0.9477384686470032, "grad_norm": 0.5136121511459351, "learning_rate": 0.00019994267841503155}
+{"ts": "2025-12-26T23:17:36", "event": "train_log", "step": 1282, "epoch": 0.5409282700421941, "progress_pct": 9.02, "epoch_pct": 9.02, "eta": "50:32:00", "max_grad_norm": 1.0, "loss": 0.7883896827697754, "grad_norm": 0.4039286673069, "learning_rate": 0.00019994104089878452}
+{"ts": "2025-12-26T23:17:52", "event": "train_log", "step": 1284, "epoch": 0.5417721518987342, "progress_pct": 9.03, "epoch_pct": 9.03, "eta": "50:29:27", "max_grad_norm": 1.0, "loss": 0.8526129126548767, "grad_norm": 0.4425487816333771, "learning_rate": 0.0001999393803280587}
+{"ts": "2025-12-26T23:18:09", "event": "train_log", "step": 1286, "epoch": 0.5426160337552742, "progress_pct": 9.04, "epoch_pct": 9.04, "eta": "50:27:06", "max_grad_norm": 1.0, "loss": 0.8694683909416199, "grad_norm": 0.4544559121131897, "learning_rate": 0.00019993769670323725}
+{"ts": "2025-12-26T23:18:28", "event": "train_log", "step": 1288, "epoch": 0.5434599156118144, "progress_pct": 9.06, "epoch_pct": 9.06, "eta": "50:25:06", "max_grad_norm": 1.0, "loss": 0.8211527466773987, "grad_norm": 0.42872917652130127, "learning_rate": 0.0001999359900247085}
+{"ts": "2025-12-26T23:18:44", "event": "train_log", "step": 1290, "epoch": 0.5443037974683544, "progress_pct": 9.07, "epoch_pct": 9.07, "eta": "50:22:40", "max_grad_norm": 1.0, "loss": 0.7967932224273682, "grad_norm": 0.41549673676490784, "learning_rate": 0.00019993426029286617}
+{"ts": "2025-12-26T23:19:00", "event": "train_log", "step": 1292, "epoch": 0.5451476793248945, "progress_pct": 9.09, "epoch_pct": 9.09, "eta": "50:20:13", "max_grad_norm": 1.0, "loss": 0.8291563987731934, "grad_norm": 0.42888355255126953, "learning_rate": 0.00019993250750810926}
+{"ts": "2025-12-26T23:19:16", "event": "train_log", "step": 1294, "epoch": 0.5459915611814345, "progress_pct": 9.1, "epoch_pct": 9.1, "eta": "50:17:39", "max_grad_norm": 1.0, "loss": 0.8787803649902344, "grad_norm": 0.44985485076904297, "learning_rate": 0.0001999307316708421}
+{"ts": "2025-12-26T23:19:31", "event": "train_log", "step": 1296, "epoch": 0.5468354430379747, "progress_pct": 9.11, "epoch_pct": 9.11, "eta": "50:15:05", "max_grad_norm": 1.0, "loss": 0.8375519514083862, "grad_norm": 0.43326008319854736, "learning_rate": 0.00019992893278147436}
+{"ts": "2025-12-26T23:19:48", "event": "train_log", "step": 1298, "epoch": 0.5476793248945148, "progress_pct": 9.13, "epoch_pct": 9.13, "eta": "50:12:47", "max_grad_norm": 1.0, "loss": 0.8538051247596741, "grad_norm": 0.406556099653244, "learning_rate": 0.00019992711084042096}
+{"ts": "2025-12-26T23:20:04", "event": "train_log", "step": 1300, "epoch": 0.5485232067510548, "progress_pct": 9.14, "epoch_pct": 9.14, "eta": "50:10:17", "max_grad_norm": 1.0, "loss": 0.9722132086753845, "grad_norm": 0.46090877056121826, "learning_rate": 0.00019992526584810226}
+{"ts": "2025-12-26T23:31:25", "event": "train_log", "step": 1300, "epoch": 0.5485232067510548, "progress_pct": 9.14, "epoch_pct": 9.14, "eta": "52:03:03", "max_grad_norm": 1.0, "eval_loss": 0.9068717360496521, "eval_runtime": 680.7718, "eval_samples_per_second": 3.095, "eval_steps_per_second": 3.095}
+{"ts": "2025-12-26T23:31:39", "event": "train_log", "step": 1302, "epoch": 0.549367088607595, "progress_pct": 9.16, "epoch_pct": 9.16, "eta": "52:00:15", "max_grad_norm": 1.0, "loss": 0.8474152684211731, "grad_norm": 0.44932207465171814, "learning_rate": 0.00019992339780494382}
+{"ts": "2025-12-26T23:31:54", "event": "train_log", "step": 1304, "epoch": 0.550210970464135, "progress_pct": 9.17, "epoch_pct": 9.17, "eta": "51:57:26", "max_grad_norm": 1.0, "loss": 0.8440775871276855, "grad_norm": 0.5185276865959167, "learning_rate": 0.0001999215067113766}
+{"ts": "2025-12-26T23:32:10", "event": "train_log", "step": 1306, "epoch": 0.5510548523206751, "progress_pct": 9.18, "epoch_pct": 9.18, "eta": "51:54:42", "max_grad_norm": 1.0, "loss": 0.8848010897636414, "grad_norm": 0.43811365962028503, "learning_rate": 0.0001999195925678368}
+{"ts": "2025-12-26T23:32:25", "event": "train_log", "step": 1308, "epoch": 0.5518987341772152, "progress_pct": 9.2, "epoch_pct": 9.2, "eta": "51:51:54", "max_grad_norm": 1.0, "loss": 0.8265100121498108, "grad_norm": 0.4861660301685333, "learning_rate": 0.00019991765537476596}
+{"ts": "2025-12-26T23:32:39", "event": "train_log", "step": 1310, "epoch": 0.5527426160337553, "progress_pct": 9.21, "epoch_pct": 9.21, "eta": "51:49:03", "max_grad_norm": 1.0, "loss": 0.9327963590621948, "grad_norm": 0.5068721771240234, "learning_rate": 0.000199915695132611}
+{"ts": "2025-12-26T23:32:54", "event": "train_log", "step": 1312, "epoch": 0.5535864978902953, "progress_pct": 9.23, "epoch_pct": 9.23, "eta": "51:46:20", "max_grad_norm": 1.0, "loss": 0.9315155744552612, "grad_norm": 0.4902805685997009, "learning_rate": 0.0001999137118418241}
+{"ts": "2025-12-26T23:33:09", "event": "train_log", "step": 1314, "epoch": 0.5544303797468354, "progress_pct": 9.24, "epoch_pct": 9.24, "eta": "51:43:33", "max_grad_norm": 1.0, "loss": 0.9344852566719055, "grad_norm": 0.4530964493751526, "learning_rate": 0.00019991170550286272}
+{"ts": "2025-12-26T23:33:24", "event": "train_log", "step": 1316, "epoch": 0.5552742616033756, "progress_pct": 9.25, "epoch_pct": 9.25, "eta": "51:40:48", "max_grad_norm": 1.0, "loss": 0.8326173424720764, "grad_norm": 0.4106673300266266, "learning_rate": 0.00019990967611618974}
+{"ts": "2025-12-26T23:33:40", "event": "train_log", "step": 1318, "epoch": 0.5561181434599156, "progress_pct": 9.27, "epoch_pct": 9.27, "eta": "51:38:09", "max_grad_norm": 1.0, "loss": 0.8934909105300903, "grad_norm": 0.4439711570739746, "learning_rate": 0.0001999076236822733}
+{"ts": "2025-12-26T23:33:55", "event": "train_log", "step": 1320, "epoch": 0.5569620253164557, "progress_pct": 9.28, "epoch_pct": 9.28, "eta": "51:35:33", "max_grad_norm": 1.0, "loss": 0.6691107153892517, "grad_norm": 0.40245553851127625, "learning_rate": 0.0001999055482015868}
+{"ts": "2025-12-26T23:34:10", "event": "train_log", "step": 1322, "epoch": 0.5578059071729958, "progress_pct": 9.3, "epoch_pct": 9.3, "eta": "51:32:47", "max_grad_norm": 1.0, "loss": 0.8560307621955872, "grad_norm": 0.43904435634613037, "learning_rate": 0.0001999034496746091}
+{"ts": "2025-12-26T23:34:24", "event": "train_log", "step": 1324, "epoch": 0.5586497890295359, "progress_pct": 9.31, "epoch_pct": 9.31, "eta": "51:29:54", "max_grad_norm": 1.0, "loss": 0.9024442434310913, "grad_norm": 0.5131705403327942, "learning_rate": 0.00019990132810182422}
+{"ts": "2025-12-26T23:34:39", "event": "train_log", "step": 1326, "epoch": 0.5594936708860759, "progress_pct": 9.32, "epoch_pct": 9.32, "eta": "51:27:10", "max_grad_norm": 1.0, "loss": 0.8675428628921509, "grad_norm": 0.4539531171321869, "learning_rate": 0.00019989918348372156}
+{"ts": "2025-12-26T23:34:54", "event": "train_log", "step": 1328, "epoch": 0.560337552742616, "progress_pct": 9.34, "epoch_pct": 9.34, "eta": "51:24:27", "max_grad_norm": 1.0, "loss": 0.8853429555892944, "grad_norm": 0.44469985365867615, "learning_rate": 0.00019989701582079591}
+{"ts": "2025-12-26T23:35:08", "event": "train_log", "step": 1330, "epoch": 0.5611814345991561, "progress_pct": 9.35, "epoch_pct": 9.35, "eta": "51:21:37", "max_grad_norm": 1.0, "loss": 0.9675378799438477, "grad_norm": 0.4485546946525574, "learning_rate": 0.00019989482511354725}
+{"ts": "2025-12-26T23:35:23", "event": "train_log", "step": 1332, "epoch": 0.5620253164556962, "progress_pct": 9.37, "epoch_pct": 9.37, "eta": "51:18:52", "max_grad_norm": 1.0, "loss": 0.8256655335426331, "grad_norm": 0.43316468596458435, "learning_rate": 0.00019989261136248097}
+{"ts": "2025-12-26T23:35:38", "event": "train_log", "step": 1334, "epoch": 0.5628691983122363, "progress_pct": 9.38, "epoch_pct": 9.38, "eta": "51:16:17", "max_grad_norm": 1.0, "loss": 0.7727690935134888, "grad_norm": 0.4315780699253082, "learning_rate": 0.00019989037456810772}
+{"ts": "2025-12-26T23:35:53", "event": "train_log", "step": 1336, "epoch": 0.5637130801687764, "progress_pct": 9.4, "epoch_pct": 9.4, "eta": "51:13:34", "max_grad_norm": 1.0, "loss": 0.9240618348121643, "grad_norm": 0.45147502422332764, "learning_rate": 0.00019988811473094348}
+{"ts": "2025-12-26T23:36:08", "event": "train_log", "step": 1338, "epoch": 0.5645569620253165, "progress_pct": 9.41, "epoch_pct": 9.41, "eta": "51:10:57", "max_grad_norm": 1.0, "loss": 0.8005949258804321, "grad_norm": 0.4606908857822418, "learning_rate": 0.00019988583185150957}
+{"ts": "2025-12-26T23:36:23", "event": "train_log", "step": 1340, "epoch": 0.5654008438818565, "progress_pct": 9.42, "epoch_pct": 9.42, "eta": "51:08:12", "max_grad_norm": 1.0, "loss": 0.9042718410491943, "grad_norm": 0.48932451009750366, "learning_rate": 0.00019988352593033255}
+{"ts": "2025-12-26T23:36:37", "event": "train_log", "step": 1342, "epoch": 0.5662447257383966, "progress_pct": 9.44, "epoch_pct": 9.44, "eta": "51:05:23", "max_grad_norm": 1.0, "loss": 0.8724613785743713, "grad_norm": 0.46022218465805054, "learning_rate": 0.00019988119696794443}
+{"ts": "2025-12-26T23:36:52", "event": "train_log", "step": 1344, "epoch": 0.5670886075949367, "progress_pct": 9.45, "epoch_pct": 9.45, "eta": "51:02:44", "max_grad_norm": 1.0, "loss": 0.8653066158294678, "grad_norm": 0.4414621889591217, "learning_rate": 0.00019987884496488242}
+{"ts": "2025-12-26T23:37:06", "event": "train_log", "step": 1346, "epoch": 0.5679324894514768, "progress_pct": 9.47, "epoch_pct": 9.47, "eta": "51:00:04", "max_grad_norm": 1.0, "loss": 0.8843849301338196, "grad_norm": 0.4490342140197754, "learning_rate": 0.0001998764699216891}
+{"ts": "2025-12-26T23:37:22", "event": "train_log", "step": 1348, "epoch": 0.5687763713080168, "progress_pct": 9.48, "epoch_pct": 9.48, "eta": "50:57:30", "max_grad_norm": 1.0, "loss": 0.8385393023490906, "grad_norm": 0.47690069675445557, "learning_rate": 0.00019987407183891232}
+{"ts": "2025-12-26T23:37:37", "event": "train_log", "step": 1350, "epoch": 0.569620253164557, "progress_pct": 9.49, "epoch_pct": 9.49, "eta": "50:54:52", "max_grad_norm": 1.0, "loss": 0.77492755651474, "grad_norm": 0.41409361362457275, "learning_rate": 0.00019987165071710527}
+{"ts": "2025-12-26T23:37:52", "event": "train_log", "step": 1352, "epoch": 0.570464135021097, "progress_pct": 9.51, "epoch_pct": 9.51, "eta": "50:52:19", "max_grad_norm": 1.0, "loss": 0.8368680477142334, "grad_norm": 0.4151647984981537, "learning_rate": 0.0001998692065568265}
+{"ts": "2025-12-26T23:38:08", "event": "train_log", "step": 1354, "epoch": 0.5713080168776371, "progress_pct": 9.52, "epoch_pct": 9.52, "eta": "50:49:53", "max_grad_norm": 1.0, "loss": 0.833220362663269, "grad_norm": 0.4105067849159241, "learning_rate": 0.0001998667393586398}
+{"ts": "2025-12-26T23:38:24", "event": "train_log", "step": 1356, "epoch": 0.5721518987341773, "progress_pct": 9.54, "epoch_pct": 9.54, "eta": "50:47:24", "max_grad_norm": 1.0, "loss": 0.8041552305221558, "grad_norm": 0.44701820611953735, "learning_rate": 0.0001998642491231143}
+{"ts": "2025-12-26T23:38:39", "event": "train_log", "step": 1358, "epoch": 0.5729957805907173, "progress_pct": 9.55, "epoch_pct": 9.55, "eta": "50:44:48", "max_grad_norm": 1.0, "loss": 0.9052709937095642, "grad_norm": 0.4621582329273224, "learning_rate": 0.00019986173585082444}
+{"ts": "2025-12-26T23:38:53", "event": "train_log", "step": 1360, "epoch": 0.5738396624472574, "progress_pct": 9.56, "epoch_pct": 9.56, "eta": "50:42:04", "max_grad_norm": 1.0, "loss": 0.8036054372787476, "grad_norm": 0.4934713840484619, "learning_rate": 0.00019985919954235005}
+{"ts": "2025-12-26T23:39:09", "event": "train_log", "step": 1362, "epoch": 0.5746835443037974, "progress_pct": 9.58, "epoch_pct": 9.58, "eta": "50:39:38", "max_grad_norm": 1.0, "loss": 0.8340095281600952, "grad_norm": 0.4557384252548218, "learning_rate": 0.0001998566401982761}
+{"ts": "2025-12-26T23:39:24", "event": "train_log", "step": 1364, "epoch": 0.5755274261603376, "progress_pct": 9.59, "epoch_pct": 9.59, "eta": "50:37:04", "max_grad_norm": 1.0, "loss": 0.8703633546829224, "grad_norm": 0.5048403739929199, "learning_rate": 0.00019985405781919305}
+{"ts": "2025-12-26T23:39:39", "event": "train_log", "step": 1366, "epoch": 0.5763713080168776, "progress_pct": 9.61, "epoch_pct": 9.61, "eta": "50:34:28", "max_grad_norm": 1.0, "loss": 0.9115830659866333, "grad_norm": 0.4707714319229126, "learning_rate": 0.0001998514524056966}
+{"ts": "2025-12-26T23:39:54", "event": "train_log", "step": 1368, "epoch": 0.5772151898734177, "progress_pct": 9.62, "epoch_pct": 9.62, "eta": "50:31:56", "max_grad_norm": 1.0, "loss": 0.8603149056434631, "grad_norm": 0.4502186179161072, "learning_rate": 0.00019984882395838777}
+{"ts": "2025-12-26T23:40:10", "event": "train_log", "step": 1370, "epoch": 0.5780590717299579, "progress_pct": 9.63, "epoch_pct": 9.63, "eta": "50:29:33", "max_grad_norm": 1.0, "loss": 0.7482197284698486, "grad_norm": 0.4229136109352112, "learning_rate": 0.00019984617247787288}
+{"ts": "2025-12-26T23:40:25", "event": "train_log", "step": 1372, "epoch": 0.5789029535864979, "progress_pct": 9.65, "epoch_pct": 9.65, "eta": "50:26:57", "max_grad_norm": 1.0, "loss": 0.9071239233016968, "grad_norm": 0.49208030104637146, "learning_rate": 0.0001998434979647636}
+{"ts": "2025-12-26T23:40:39", "event": "train_log", "step": 1374, "epoch": 0.579746835443038, "progress_pct": 9.66, "epoch_pct": 9.66, "eta": "50:24:20", "max_grad_norm": 1.0, "loss": 0.9370644092559814, "grad_norm": 0.47393161058425903, "learning_rate": 0.00019984080041967687}
+{"ts": "2025-12-26T23:40:54", "event": "train_log", "step": 1376, "epoch": 0.580590717299578, "progress_pct": 9.68, "epoch_pct": 9.68, "eta": "50:21:46", "max_grad_norm": 1.0, "loss": 0.8536180257797241, "grad_norm": 0.4483324885368347, "learning_rate": 0.00019983807984323492}
+{"ts": "2025-12-26T23:41:09", "event": "train_log", "step": 1378, "epoch": 0.5814345991561182, "progress_pct": 9.69, "epoch_pct": 9.69, "eta": "50:19:16", "max_grad_norm": 1.0, "loss": 0.7360405325889587, "grad_norm": 0.4163796007633209, "learning_rate": 0.00019983533623606543}
+{"ts": "2025-12-26T23:41:25", "event": "train_log", "step": 1380, "epoch": 0.5822784810126582, "progress_pct": 9.7, "epoch_pct": 9.7, "eta": "50:16:56", "max_grad_norm": 1.0, "loss": 0.8162824511528015, "grad_norm": 0.44147446751594543, "learning_rate": 0.00019983256959880122}
+{"ts": "2025-12-26T23:41:41", "event": "train_log", "step": 1382, "epoch": 0.5831223628691983, "progress_pct": 9.72, "epoch_pct": 9.72, "eta": "50:14:36", "max_grad_norm": 1.0, "loss": 0.9169327616691589, "grad_norm": 0.44443491101264954, "learning_rate": 0.0001998297799320805}
+{"ts": "2025-12-26T23:41:58", "event": "train_log", "step": 1384, "epoch": 0.5839662447257384, "progress_pct": 9.73, "epoch_pct": 9.73, "eta": "50:12:18", "max_grad_norm": 1.0, "loss": 0.8875693678855896, "grad_norm": 0.4641949236392975, "learning_rate": 0.00019982696723654686}
+{"ts": "2025-12-26T23:42:15", "event": "train_log", "step": 1386, "epoch": 0.5848101265822785, "progress_pct": 9.75, "epoch_pct": 9.75, "eta": "50:10:13", "max_grad_norm": 1.0, "loss": 0.7747344970703125, "grad_norm": 0.4327974319458008, "learning_rate": 0.00019982413151284906}
+{"ts": "2025-12-26T23:42:32", "event": "train_log", "step": 1388, "epoch": 0.5856540084388185, "progress_pct": 9.76, "epoch_pct": 9.76, "eta": "50:07:54", "max_grad_norm": 1.0, "loss": 0.8490481972694397, "grad_norm": 0.44200772047042847, "learning_rate": 0.0001998212727616413}
+{"ts": "2025-12-26T23:42:47", "event": "train_log", "step": 1390, "epoch": 0.5864978902953587, "progress_pct": 9.77, "epoch_pct": 9.77, "eta": "50:05:32", "max_grad_norm": 1.0, "loss": 0.8014808297157288, "grad_norm": 0.39084959030151367, "learning_rate": 0.000199818390983583}
+{"ts": "2025-12-26T23:43:03", "event": "train_log", "step": 1392, "epoch": 0.5873417721518988, "progress_pct": 9.79, "epoch_pct": 9.79, "eta": "50:03:11", "max_grad_norm": 1.0, "loss": 0.8650928139686584, "grad_norm": 0.3691277801990509, "learning_rate": 0.00019981548617933897}
+{"ts": "2025-12-26T23:43:20", "event": "train_log", "step": 1394, "epoch": 0.5881856540084388, "progress_pct": 9.8, "epoch_pct": 9.8, "eta": "50:00:54", "max_grad_norm": 1.0, "loss": 0.7713267803192139, "grad_norm": 0.480002760887146, "learning_rate": 0.00019981255834957925}
+{"ts": "2025-12-26T23:43:35", "event": "train_log", "step": 1396, "epoch": 0.5890295358649789, "progress_pct": 9.82, "epoch_pct": 9.82, "eta": "49:58:27", "max_grad_norm": 1.0, "loss": 0.9053656458854675, "grad_norm": 0.4397226572036743, "learning_rate": 0.00019980960749497927}
+{"ts": "2025-12-26T23:43:51", "event": "train_log", "step": 1398, "epoch": 0.589873417721519, "progress_pct": 9.83, "epoch_pct": 9.83, "eta": "49:56:12", "max_grad_norm": 1.0, "loss": 0.8737669587135315, "grad_norm": 0.4263727366924286, "learning_rate": 0.00019980663361621973}
+{"ts": "2025-12-26T23:44:06", "event": "train_log", "step": 1400, "epoch": 0.5907172995780591, "progress_pct": 9.85, "epoch_pct": 9.85, "eta": "49:53:45", "max_grad_norm": 1.0, "loss": 0.9396650791168213, "grad_norm": 0.4468817710876465, "learning_rate": 0.0001998036367139866}
+{"ts": "2025-12-26T23:55:39", "event": "train_log", "step": 1400, "epoch": 0.5907172995780591, "progress_pct": 9.85, "epoch_pct": 9.85, "eta": "51:39:29", "max_grad_norm": 1.0, "eval_loss": 0.8971880674362183, "eval_runtime": 692.8046, "eval_samples_per_second": 3.041, "eval_steps_per_second": 3.041}
+{"ts": "2025-12-26T23:55:55", "event": "train_log", "step": 1402, "epoch": 0.5915611814345991, "progress_pct": 9.86, "epoch_pct": 9.86, "eta": "51:36:59", "max_grad_norm": 1.0, "loss": 0.926134467124939, "grad_norm": 0.4128144383430481, "learning_rate": 0.00019980061678897127}
+{"ts": "2025-12-26T23:56:11", "event": "train_log", "step": 1404, "epoch": 0.5924050632911393, "progress_pct": 9.87, "epoch_pct": 9.87, "eta": "51:34:34", "max_grad_norm": 1.0, "loss": 0.826360285282135, "grad_norm": 0.45524629950523376, "learning_rate": 0.00019979757384187035}
+{"ts": "2025-12-26T23:56:26", "event": "train_log", "step": 1406, "epoch": 0.5932489451476793, "progress_pct": 9.89, "epoch_pct": 9.89, "eta": "51:32:01", "max_grad_norm": 1.0, "loss": 0.8147702813148499, "grad_norm": 0.42521336674690247, "learning_rate": 0.0001997945078733858}
+{"ts": "2025-12-26T23:56:42", "event": "train_log", "step": 1408, "epoch": 0.5940928270042194, "progress_pct": 9.9, "epoch_pct": 9.9, "eta": "51:29:32", "max_grad_norm": 1.0, "loss": 0.7711596488952637, "grad_norm": 0.4172525107860565, "learning_rate": 0.00019979141888422489}
+{"ts": "2025-12-26T23:56:57", "event": "train_log", "step": 1410, "epoch": 0.5949367088607594, "progress_pct": 9.92, "epoch_pct": 9.92, "eta": "51:26:58", "max_grad_norm": 1.0, "loss": 0.7875980138778687, "grad_norm": 0.46193623542785645, "learning_rate": 0.00019978830687510013}
+{"ts": "2025-12-26T23:57:13", "event": "train_log", "step": 1412, "epoch": 0.5957805907172996, "progress_pct": 9.93, "epoch_pct": 9.93, "eta": "51:24:26", "max_grad_norm": 1.0, "loss": 0.8590195178985596, "grad_norm": 0.4455825686454773, "learning_rate": 0.00019978517184672946}
+{"ts": "2025-12-26T23:57:27", "event": "train_log", "step": 1414, "epoch": 0.5966244725738397, "progress_pct": 9.94, "epoch_pct": 9.94, "eta": "51:21:45", "max_grad_norm": 1.0, "loss": 0.9318227767944336, "grad_norm": 0.4266505539417267, "learning_rate": 0.00019978201379983608}
+{"ts": "2025-12-26T23:57:42", "event": "train_log", "step": 1416, "epoch": 0.5974683544303797, "progress_pct": 9.96, "epoch_pct": 9.96, "eta": "51:19:14", "max_grad_norm": 1.0, "loss": 0.8644474744796753, "grad_norm": 0.42627251148223877, "learning_rate": 0.00019977883273514843}
+{"ts": "2025-12-26T23:57:58", "event": "train_log", "step": 1418, "epoch": 0.5983122362869199, "progress_pct": 9.97, "epoch_pct": 9.97, "eta": "51:16:47", "max_grad_norm": 1.0, "loss": 0.7760446071624756, "grad_norm": 0.393873393535614, "learning_rate": 0.00019977562865340038}
+{"ts": "2025-12-26T23:58:14", "event": "train_log", "step": 1420, "epoch": 0.5991561181434599, "progress_pct": 9.99, "epoch_pct": 9.99, "eta": "51:14:18", "max_grad_norm": 1.0, "loss": 0.8071584701538086, "grad_norm": 0.44334208965301514, "learning_rate": 0.000199772401555331}
+{"ts": "2025-12-26T23:58:28", "event": "train_log", "step": 1422, "epoch": 0.6, "progress_pct": 10.0, "epoch_pct": 10.0, "eta": "51:11:43", "max_grad_norm": 1.0, "loss": 0.9076889753341675, "grad_norm": 0.39390063285827637, "learning_rate": 0.0001997691514416848}
+{"ts": "2025-12-26T23:58:43", "event": "train_log", "step": 1424, "epoch": 0.60084388185654, "progress_pct": 10.01, "epoch_pct": 10.01, "eta": "51:09:08", "max_grad_norm": 1.0, "loss": 0.8534318804740906, "grad_norm": 0.46753421425819397, "learning_rate": 0.00019976587831321144}
+{"ts": "2025-12-26T23:58:57", "event": "train_log", "step": 1426, "epoch": 0.6016877637130802, "progress_pct": 10.03, "epoch_pct": 10.03, "eta": "51:06:29", "max_grad_norm": 1.0, "loss": 0.9049596190452576, "grad_norm": 0.4840681254863739, "learning_rate": 0.00019976258217066598}
+{"ts": "2025-12-26T23:59:12", "event": "train_log", "step": 1428, "epoch": 0.6025316455696202, "progress_pct": 10.04, "epoch_pct": 10.04, "eta": "51:03:54", "max_grad_norm": 1.0, "loss": 0.8520918488502502, "grad_norm": 0.4740557372570038, "learning_rate": 0.00019975926301480878}
+{"ts": "2025-12-26T23:59:27", "event": "train_log", "step": 1430, "epoch": 0.6033755274261603, "progress_pct": 10.06, "epoch_pct": 10.06, "eta": "51:01:21", "max_grad_norm": 1.0, "loss": 0.8221305012702942, "grad_norm": 0.46498584747314453, "learning_rate": 0.00019975592084640553}
+{"ts": "2025-12-26T23:59:42", "event": "train_log", "step": 1432, "epoch": 0.6042194092827005, "progress_pct": 10.07, "epoch_pct": 10.07, "eta": "50:58:52", "max_grad_norm": 1.0, "loss": 0.8132198452949524, "grad_norm": 0.40885692834854126, "learning_rate": 0.0001997525556662272}
+{"ts": "2025-12-26T23:59:57", "event": "train_log", "step": 1434, "epoch": 0.6050632911392405, "progress_pct": 10.08, "epoch_pct": 10.08, "eta": "50:56:19", "max_grad_norm": 1.0, "loss": 0.870534360408783, "grad_norm": 0.4375821053981781, "learning_rate": 0.00019974916747505006}
+{"ts": "2025-12-27T00:00:12", "event": "train_log", "step": 1436, "epoch": 0.6059071729957806, "progress_pct": 10.1, "epoch_pct": 10.1, "eta": "50:53:46", "max_grad_norm": 1.0, "loss": 0.8825759291648865, "grad_norm": 0.4780935049057007, "learning_rate": 0.00019974575627365569}
+{"ts": "2025-12-27T00:00:27", "event": "train_log", "step": 1438, "epoch": 0.6067510548523207, "progress_pct": 10.11, "epoch_pct": 10.11, "eta": "50:51:15", "max_grad_norm": 1.0, "loss": 0.8187641501426697, "grad_norm": 0.4623970687389374, "learning_rate": 0.00019974232206283098}
+{"ts": "2025-12-27T00:00:42", "event": "train_log", "step": 1440, "epoch": 0.6075949367088608, "progress_pct": 10.13, "epoch_pct": 10.13, "eta": "50:48:47", "max_grad_norm": 1.0, "loss": 0.8569821119308472, "grad_norm": 0.42172595858573914, "learning_rate": 0.0001997388648433682}
+{"ts": "2025-12-27T00:00:56", "event": "train_log", "step": 1442, "epoch": 0.6084388185654008, "progress_pct": 10.14, "epoch_pct": 10.14, "eta": "50:46:13", "max_grad_norm": 1.0, "loss": 0.9173614382743835, "grad_norm": 0.49386659264564514, "learning_rate": 0.0001997353846160648}
+{"ts": "2025-12-27T00:01:11", "event": "train_log", "step": 1444, "epoch": 0.6092827004219409, "progress_pct": 10.15, "epoch_pct": 10.15, "eta": "50:43:46", "max_grad_norm": 1.0, "loss": 0.9444975256919861, "grad_norm": 0.47122615575790405, "learning_rate": 0.00019973188138172363}
+{"ts": "2025-12-27T00:01:26", "event": "train_log", "step": 1446, "epoch": 0.610126582278481, "progress_pct": 10.17, "epoch_pct": 10.17, "eta": "50:41:13", "max_grad_norm": 1.0, "loss": 0.9537245631217957, "grad_norm": 0.4831530451774597, "learning_rate": 0.0001997283551411528}
+{"ts": "2025-12-27T00:01:42", "event": "train_log", "step": 1448, "epoch": 0.6109704641350211, "progress_pct": 10.18, "epoch_pct": 10.18, "eta": "50:38:53", "max_grad_norm": 1.0, "loss": 0.8481367230415344, "grad_norm": 0.44460317492485046, "learning_rate": 0.0001997248058951658}
+{"ts": "2025-12-27T00:01:57", "event": "train_log", "step": 1450, "epoch": 0.6118143459915611, "progress_pct": 10.2, "epoch_pct": 10.2, "eta": "50:36:23", "max_grad_norm": 1.0, "loss": 0.8572371006011963, "grad_norm": 0.4315263628959656, "learning_rate": 0.00019972123364458128}
+{"ts": "2025-12-27T00:02:12", "event": "train_log", "step": 1452, "epoch": 0.6126582278481013, "progress_pct": 10.21, "epoch_pct": 10.21, "eta": "50:33:58", "max_grad_norm": 1.0, "loss": 0.762590765953064, "grad_norm": 0.38559582829475403, "learning_rate": 0.00019971763839022336}
+{"ts": "2025-12-27T00:02:27", "event": "train_log", "step": 1454, "epoch": 0.6135021097046414, "progress_pct": 10.23, "epoch_pct": 10.23, "eta": "50:31:35", "max_grad_norm": 1.0, "loss": 0.7879500389099121, "grad_norm": 0.3820290267467499, "learning_rate": 0.00019971402013292138}
+{"ts": "2025-12-27T00:02:42", "event": "train_log", "step": 1456, "epoch": 0.6143459915611814, "progress_pct": 10.24, "epoch_pct": 10.24, "eta": "50:29:07", "max_grad_norm": 1.0, "loss": 0.8696863055229187, "grad_norm": 0.42509570717811584, "learning_rate": 0.00019971037887351}
+{"ts": "2025-12-27T00:02:57", "event": "train_log", "step": 1458, "epoch": 0.6151898734177215, "progress_pct": 10.25, "epoch_pct": 10.25, "eta": "50:26:34", "max_grad_norm": 1.0, "loss": 0.8425862789154053, "grad_norm": 0.46750548481941223, "learning_rate": 0.00019970671461282916}
+{"ts": "2025-12-27T00:03:12", "event": "train_log", "step": 1460, "epoch": 0.6160337552742616, "progress_pct": 10.27, "epoch_pct": 10.27, "eta": "50:24:12", "max_grad_norm": 1.0, "loss": 0.8028931021690369, "grad_norm": 0.3937121033668518, "learning_rate": 0.0001997030273517242}
+{"ts": "2025-12-27T00:03:27", "event": "train_log", "step": 1462, "epoch": 0.6168776371308017, "progress_pct": 10.28, "epoch_pct": 10.28, "eta": "50:21:46", "max_grad_norm": 1.0, "loss": 0.8662509918212891, "grad_norm": 0.4361220896244049, "learning_rate": 0.00019969931709104565}
+{"ts": "2025-12-27T00:03:41", "event": "train_log", "step": 1464, "epoch": 0.6177215189873417, "progress_pct": 10.3, "epoch_pct": 10.3, "eta": "50:19:13", "max_grad_norm": 1.0, "loss": 0.8962596654891968, "grad_norm": 0.4737963378429413, "learning_rate": 0.00019969558383164943}
+{"ts": "2025-12-27T00:03:57", "event": "train_log", "step": 1466, "epoch": 0.6185654008438819, "progress_pct": 10.31, "epoch_pct": 10.31, "eta": "50:16:54", "max_grad_norm": 1.0, "loss": 0.8630867600440979, "grad_norm": 0.3952867388725281, "learning_rate": 0.00019969182757439668}
+{"ts": "2025-12-27T00:04:12", "event": "train_log", "step": 1468, "epoch": 0.619409282700422, "progress_pct": 10.32, "epoch_pct": 10.32, "eta": "50:14:32", "max_grad_norm": 1.0, "loss": 0.8759240508079529, "grad_norm": 0.41688743233680725, "learning_rate": 0.00019968804832015393}
+{"ts": "2025-12-27T00:04:26", "event": "train_log", "step": 1470, "epoch": 0.620253164556962, "progress_pct": 10.34, "epoch_pct": 10.34, "eta": "50:12:03", "max_grad_norm": 1.0, "loss": 0.8459385633468628, "grad_norm": 0.4599224627017975, "learning_rate": 0.00019968424606979298}
+{"ts": "2025-12-27T00:04:41", "event": "train_log", "step": 1472, "epoch": 0.6210970464135022, "progress_pct": 10.35, "epoch_pct": 10.35, "eta": "50:09:36", "max_grad_norm": 1.0, "loss": 0.8844659328460693, "grad_norm": 0.5476765632629395, "learning_rate": 0.00019968042082419094}
+{"ts": "2025-12-27T00:04:56", "event": "train_log", "step": 1474, "epoch": 0.6219409282700422, "progress_pct": 10.37, "epoch_pct": 10.37, "eta": "50:07:12", "max_grad_norm": 1.0, "loss": 0.860946536064148, "grad_norm": 0.46202269196510315, "learning_rate": 0.00019967657258423022}
+{"ts": "2025-12-27T00:05:12", "event": "train_log", "step": 1476, "epoch": 0.6227848101265823, "progress_pct": 10.38, "epoch_pct": 10.38, "eta": "50:04:53", "max_grad_norm": 1.0, "loss": 0.7587860822677612, "grad_norm": 0.39446666836738586, "learning_rate": 0.00019967270135079853}
+{"ts": "2025-12-27T00:05:27", "event": "train_log", "step": 1478, "epoch": 0.6236286919831223, "progress_pct": 10.39, "epoch_pct": 10.39, "eta": "50:02:33", "max_grad_norm": 1.0, "loss": 0.9143000245094299, "grad_norm": 0.4500375986099243, "learning_rate": 0.0001996688071247889}
+{"ts": "2025-12-27T00:05:42", "event": "train_log", "step": 1480, "epoch": 0.6244725738396625, "progress_pct": 10.41, "epoch_pct": 10.41, "eta": "50:00:13", "max_grad_norm": 1.0, "loss": 0.7865519523620605, "grad_norm": 0.4203545153141022, "learning_rate": 0.0001996648899070996}
+{"ts": "2025-12-27T00:05:56", "event": "train_log", "step": 1482, "epoch": 0.6253164556962025, "progress_pct": 10.42, "epoch_pct": 10.42, "eta": "49:57:42", "max_grad_norm": 1.0, "loss": 0.9192912578582764, "grad_norm": 0.45623889565467834, "learning_rate": 0.00019966094969863432}
+{"ts": "2025-12-27T00:06:11", "event": "train_log", "step": 1484, "epoch": 0.6261603375527426, "progress_pct": 10.44, "epoch_pct": 10.44, "eta": "49:55:22", "max_grad_norm": 1.0, "loss": 0.8459161520004272, "grad_norm": 0.4139637053012848, "learning_rate": 0.00019965698650030195}
+{"ts": "2025-12-27T00:06:26", "event": "train_log", "step": 1486, "epoch": 0.6270042194092827, "progress_pct": 10.45, "epoch_pct": 10.45, "eta": "49:52:56", "max_grad_norm": 1.0, "loss": 0.8159077763557434, "grad_norm": 0.4410824179649353, "learning_rate": 0.00019965300031301678}
+{"ts": "2025-12-27T00:06:41", "event": "train_log", "step": 1488, "epoch": 0.6278481012658228, "progress_pct": 10.46, "epoch_pct": 10.46, "eta": "49:50:34", "max_grad_norm": 1.0, "loss": 0.884467363357544, "grad_norm": 0.43821197748184204, "learning_rate": 0.00019964899113769828}
+{"ts": "2025-12-27T00:06:55", "event": "train_log", "step": 1490, "epoch": 0.6286919831223629, "progress_pct": 10.48, "epoch_pct": 10.48, "eta": "49:48:06", "max_grad_norm": 1.0, "loss": 0.8693601489067078, "grad_norm": 0.4692091643810272, "learning_rate": 0.00019964495897527133}
+{"ts": "2025-12-27T00:07:10", "event": "train_log", "step": 1492, "epoch": 0.6295358649789029, "progress_pct": 10.49, "epoch_pct": 10.49, "eta": "49:45:43", "max_grad_norm": 1.0, "loss": 0.8897743225097656, "grad_norm": 0.40715447068214417, "learning_rate": 0.00019964090382666608}
+{"ts": "2025-12-27T00:07:25", "event": "train_log", "step": 1494, "epoch": 0.6303797468354431, "progress_pct": 10.51, "epoch_pct": 10.51, "eta": "49:43:25", "max_grad_norm": 1.0, "loss": 0.8274733424186707, "grad_norm": 0.4123334586620331, "learning_rate": 0.0001996368256928179}
+{"ts": "2025-12-27T00:07:40", "event": "train_log", "step": 1496, "epoch": 0.6312236286919831, "progress_pct": 10.52, "epoch_pct": 10.52, "eta": "49:41:03", "max_grad_norm": 1.0, "loss": 0.8158749938011169, "grad_norm": 0.44347211718559265, "learning_rate": 0.00019963272457466767}
+{"ts": "2025-12-27T00:07:54", "event": "train_log", "step": 1498, "epoch": 0.6320675105485232, "progress_pct": 10.53, "epoch_pct": 10.53, "eta": "49:38:40", "max_grad_norm": 1.0, "loss": 0.8439569473266602, "grad_norm": 0.4700329601764679, "learning_rate": 0.00019962860047316135}
+{"ts": "2025-12-27T00:08:09", "event": "train_log", "step": 1500, "epoch": 0.6329113924050633, "progress_pct": 10.55, "epoch_pct": 10.55, "eta": "49:36:19", "max_grad_norm": 1.0, "loss": 0.8316822648048401, "grad_norm": 0.46415844559669495, "learning_rate": 0.00019962445338925027}
+{"ts": "2025-12-27T00:19:35", "event": "train_log", "step": 1500, "epoch": 0.6329113924050633, "progress_pct": 10.55, "epoch_pct": 10.55, "eta": "51:13:18", "max_grad_norm": 1.0, "eval_loss": 0.887488842010498, "eval_runtime": 686.2804, "eval_samples_per_second": 3.07, "eval_steps_per_second": 3.07}
+{"ts": "2025-12-27T00:19:52", "event": "train_log", "step": 1502, "epoch": 0.6337552742616034, "progress_pct": 10.56, "epoch_pct": 10.56, "eta": "51:11:03", "max_grad_norm": 1.0, "loss": 0.9121994972229004, "grad_norm": 0.44183167815208435, "learning_rate": 0.0001996202833238911}
+{"ts": "2025-12-27T00:20:08", "event": "train_log", "step": 1504, "epoch": 0.6345991561181434, "progress_pct": 10.58, "epoch_pct": 10.58, "eta": "51:08:43", "max_grad_norm": 1.0, "loss": 0.9160211086273193, "grad_norm": 0.4279208779335022, "learning_rate": 0.00019961609027804587}
+{"ts": "2025-12-27T00:20:22", "event": "train_log", "step": 1506, "epoch": 0.6354430379746835, "progress_pct": 10.59, "epoch_pct": 10.59, "eta": "51:06:13", "max_grad_norm": 1.0, "loss": 0.861677348613739, "grad_norm": 0.3967169523239136, "learning_rate": 0.00019961187425268176}
+{"ts": "2025-12-27T00:20:38", "event": "train_log", "step": 1508, "epoch": 0.6362869198312237, "progress_pct": 10.6, "epoch_pct": 10.6, "eta": "51:03:57", "max_grad_norm": 1.0, "loss": 0.8478423357009888, "grad_norm": 0.4796451926231384, "learning_rate": 0.0001996076352487713}
+{"ts": "2025-12-27T00:20:55", "event": "train_log", "step": 1510, "epoch": 0.6371308016877637, "progress_pct": 10.62, "epoch_pct": 10.62, "eta": "51:01:45", "max_grad_norm": 1.0, "loss": 0.8739159107208252, "grad_norm": 0.3861426115036011, "learning_rate": 0.00019960337326729245}
+{"ts": "2025-12-27T00:21:12", "event": "train_log", "step": 1512, "epoch": 0.6379746835443038, "progress_pct": 10.63, "epoch_pct": 10.63, "eta": "50:59:31", "max_grad_norm": 1.0, "loss": 0.7693920731544495, "grad_norm": 0.4567820429801941, "learning_rate": 0.00019959908830922824}
+{"ts": "2025-12-27T00:21:27", "event": "train_log", "step": 1514, "epoch": 0.6388185654008439, "progress_pct": 10.65, "epoch_pct": 10.65, "eta": "50:57:13", "max_grad_norm": 1.0, "loss": 0.8482301831245422, "grad_norm": 0.4457108676433563, "learning_rate": 0.00019959478037556724}
+{"ts": "2025-12-27T00:21:43", "event": "train_log", "step": 1516, "epoch": 0.639662447257384, "progress_pct": 10.66, "epoch_pct": 10.66, "eta": "50:54:57", "max_grad_norm": 1.0, "loss": 0.8626812696456909, "grad_norm": 0.43918928503990173, "learning_rate": 0.00019959044946730314}
+{"ts": "2025-12-27T00:22:01", "event": "train_log", "step": 1518, "epoch": 0.640506329113924, "progress_pct": 10.68, "epoch_pct": 10.68, "eta": "50:52:51", "max_grad_norm": 1.0, "loss": 0.8342230916023254, "grad_norm": 0.4095900058746338, "learning_rate": 0.00019958609558543504}
+{"ts": "2025-12-27T00:22:17", "event": "train_log", "step": 1520, "epoch": 0.6413502109704642, "progress_pct": 10.69, "epoch_pct": 10.69, "eta": "50:50:34", "max_grad_norm": 1.0, "loss": 0.8794118165969849, "grad_norm": 0.4181270897388458, "learning_rate": 0.00019958171873096724}
+{"ts": "2025-12-27T00:22:32", "event": "train_log", "step": 1522, "epoch": 0.6421940928270042, "progress_pct": 10.7, "epoch_pct": 10.7, "eta": "50:48:15", "max_grad_norm": 1.0, "loss": 0.7787677049636841, "grad_norm": 0.4750959277153015, "learning_rate": 0.00019957731890490947}
+{"ts": "2025-12-27T00:22:47", "event": "train_log", "step": 1524, "epoch": 0.6430379746835443, "progress_pct": 10.72, "epoch_pct": 10.72, "eta": "50:45:53", "max_grad_norm": 1.0, "loss": 0.9084368348121643, "grad_norm": 0.5129296779632568, "learning_rate": 0.00019957289610827663}
+{"ts": "2025-12-27T00:23:03", "event": "train_log", "step": 1526, "epoch": 0.6438818565400843, "progress_pct": 10.73, "epoch_pct": 10.73, "eta": "50:43:31", "max_grad_norm": 1.0, "loss": 0.9066952466964722, "grad_norm": 0.4829029142856598, "learning_rate": 0.000199568450342089}
+{"ts": "2025-12-27T00:23:18", "event": "train_log", "step": 1528, "epoch": 0.6447257383966245, "progress_pct": 10.75, "epoch_pct": 10.75, "eta": "50:41:13", "max_grad_norm": 1.0, "loss": 0.8127874732017517, "grad_norm": 0.4225057363510132, "learning_rate": 0.00019956398160737214}
+{"ts": "2025-12-27T00:23:33", "event": "train_log", "step": 1530, "epoch": 0.6455696202531646, "progress_pct": 10.76, "epoch_pct": 10.76, "eta": "50:38:51", "max_grad_norm": 1.0, "loss": 0.8781921863555908, "grad_norm": 0.4231826364994049, "learning_rate": 0.00019955948990515684}
+{"ts": "2025-12-27T00:23:48", "event": "train_log", "step": 1532, "epoch": 0.6464135021097046, "progress_pct": 10.77, "epoch_pct": 10.77, "eta": "50:36:29", "max_grad_norm": 1.0, "loss": 0.8136996626853943, "grad_norm": 0.4436216950416565, "learning_rate": 0.00019955497523647933}
+{"ts": "2025-12-27T00:24:05", "event": "train_log", "step": 1534, "epoch": 0.6472573839662448, "progress_pct": 10.79, "epoch_pct": 10.79, "eta": "50:34:17", "max_grad_norm": 1.0, "loss": 0.8003877401351929, "grad_norm": 0.4018244743347168, "learning_rate": 0.00019955043760238098}
+{"ts": "2025-12-27T00:24:22", "event": "train_log", "step": 1536, "epoch": 0.6481012658227848, "progress_pct": 10.8, "epoch_pct": 10.8, "eta": "50:32:17", "max_grad_norm": 1.0, "loss": 0.8996267914772034, "grad_norm": 0.40024611353874207, "learning_rate": 0.0001995458770039086}
+{"ts": "2025-12-27T00:24:39", "event": "train_log", "step": 1538, "epoch": 0.6489451476793249, "progress_pct": 10.82, "epoch_pct": 10.82, "eta": "50:30:05", "max_grad_norm": 1.0, "loss": 0.8610522150993347, "grad_norm": 0.42845702171325684, "learning_rate": 0.00019954129344211424}
+{"ts": "2025-12-27T00:24:55", "event": "train_log", "step": 1540, "epoch": 0.6497890295358649, "progress_pct": 10.83, "epoch_pct": 10.83, "eta": "50:27:56", "max_grad_norm": 1.0, "loss": 0.8991837501525879, "grad_norm": 0.4258182644844055, "learning_rate": 0.00019953668691805517}
+{"ts": "2025-12-27T00:25:10", "event": "train_log", "step": 1542, "epoch": 0.6506329113924051, "progress_pct": 10.84, "epoch_pct": 10.84, "eta": "50:25:34", "max_grad_norm": 1.0, "loss": 0.8150189518928528, "grad_norm": 0.4483257830142975, "learning_rate": 0.0001995320574327941}
+{"ts": "2025-12-27T00:25:25", "event": "train_log", "step": 1544, "epoch": 0.6514767932489451, "progress_pct": 10.86, "epoch_pct": 10.86, "eta": "50:23:15", "max_grad_norm": 1.0, "loss": 0.8572840690612793, "grad_norm": 0.4320627748966217, "learning_rate": 0.0001995274049873989}
+{"ts": "2025-12-27T00:25:42", "event": "train_log", "step": 1546, "epoch": 0.6523206751054852, "progress_pct": 10.87, "epoch_pct": 10.87, "eta": "50:21:12", "max_grad_norm": 1.0, "loss": 0.7766129374504089, "grad_norm": 0.39950594305992126, "learning_rate": 0.00019952272958294288}
+{"ts": "2025-12-27T00:25:58", "event": "train_log", "step": 1548, "epoch": 0.6531645569620254, "progress_pct": 10.89, "epoch_pct": 10.89, "eta": "50:18:58", "max_grad_norm": 1.0, "loss": 0.9032199382781982, "grad_norm": 0.4130117893218994, "learning_rate": 0.00019951803122050446}
+{"ts": "2025-12-27T00:26:13", "event": "train_log", "step": 1550, "epoch": 0.6540084388185654, "progress_pct": 10.9, "epoch_pct": 10.9, "eta": "50:16:35", "max_grad_norm": 1.0, "loss": 0.8361048102378845, "grad_norm": 0.4436919391155243, "learning_rate": 0.00019951330990116754}
+{"ts": "2025-12-27T00:26:29", "event": "train_log", "step": 1552, "epoch": 0.6548523206751055, "progress_pct": 10.91, "epoch_pct": 10.91, "eta": "50:14:26", "max_grad_norm": 1.0, "loss": 0.8661768436431885, "grad_norm": 0.4229227900505066, "learning_rate": 0.00019950856562602121}
+{"ts": "2025-12-27T00:26:44", "event": "train_log", "step": 1554, "epoch": 0.6556962025316456, "progress_pct": 10.93, "epoch_pct": 10.93, "eta": "50:12:13", "max_grad_norm": 1.0, "loss": 0.9737826585769653, "grad_norm": 0.4839727580547333, "learning_rate": 0.00019950379839615988}
+{"ts": "2025-12-27T00:27:01", "event": "train_log", "step": 1556, "epoch": 0.6565400843881857, "progress_pct": 10.94, "epoch_pct": 10.94, "eta": "50:10:04", "max_grad_norm": 1.0, "loss": 0.7542453408241272, "grad_norm": 0.3859386742115021, "learning_rate": 0.00019949900821268328}
+{"ts": "2025-12-27T00:27:18", "event": "train_log", "step": 1558, "epoch": 0.6573839662447257, "progress_pct": 10.96, "epoch_pct": 10.96, "eta": "50:08:02", "max_grad_norm": 1.0, "loss": 0.9226290583610535, "grad_norm": 0.39643988013267517, "learning_rate": 0.0001994941950766964}
+{"ts": "2025-12-27T00:27:32", "event": "train_log", "step": 1560, "epoch": 0.6582278481012658, "progress_pct": 10.97, "epoch_pct": 10.97, "eta": "50:05:42", "max_grad_norm": 1.0, "loss": 0.8375602960586548, "grad_norm": 0.4261355400085449, "learning_rate": 0.00019948935898930954}
+{"ts": "2025-12-27T00:27:49", "event": "train_log", "step": 1562, "epoch": 0.6590717299578059, "progress_pct": 10.98, "epoch_pct": 10.98, "eta": "50:03:35", "max_grad_norm": 1.0, "loss": 0.7847310304641724, "grad_norm": 0.445931077003479, "learning_rate": 0.00019948449995163826}
+{"ts": "2025-12-27T00:28:04", "event": "train_log", "step": 1564, "epoch": 0.659915611814346, "progress_pct": 11.0, "epoch_pct": 11.0, "eta": "50:01:16", "max_grad_norm": 1.0, "loss": 0.851150631904602, "grad_norm": 0.4680459797382355, "learning_rate": 0.00019947961796480353}
+{"ts": "2025-12-27T00:28:19", "event": "train_log", "step": 1566, "epoch": 0.660759493670886, "progress_pct": 11.01, "epoch_pct": 11.01, "eta": "49:59:00", "max_grad_norm": 1.0, "loss": 0.8259562253952026, "grad_norm": 0.6826074719429016, "learning_rate": 0.00019947471302993143}
+{"ts": "2025-12-27T00:28:33", "event": "train_log", "step": 1568, "epoch": 0.6616033755274262, "progress_pct": 11.03, "epoch_pct": 11.03, "eta": "49:56:37", "max_grad_norm": 1.0, "loss": 0.879021406173706, "grad_norm": 0.47369205951690674, "learning_rate": 0.00019946978514815352}
+{"ts": "2025-12-27T00:28:47", "event": "train_log", "step": 1570, "epoch": 0.6624472573839663, "progress_pct": 11.04, "epoch_pct": 11.04, "eta": "49:54:14", "max_grad_norm": 1.0, "loss": 0.9296056032180786, "grad_norm": 0.41111937165260315, "learning_rate": 0.0001994648343206065}
+{"ts": "2025-12-27T00:29:02", "event": "train_log", "step": 1572, "epoch": 0.6632911392405063, "progress_pct": 11.05, "epoch_pct": 11.05, "eta": "49:51:58", "max_grad_norm": 1.0, "loss": 0.873822808265686, "grad_norm": 0.4297783076763153, "learning_rate": 0.00019945986054843248}
+{"ts": "2025-12-27T00:29:17", "event": "train_log", "step": 1574, "epoch": 0.6641350210970464, "progress_pct": 11.07, "epoch_pct": 11.07, "eta": "49:49:39", "max_grad_norm": 1.0, "loss": 0.937151312828064, "grad_norm": 0.48124316334724426, "learning_rate": 0.00019945486383277878}
+{"ts": "2025-12-27T00:29:32", "event": "train_log", "step": 1576, "epoch": 0.6649789029535865, "progress_pct": 11.08, "epoch_pct": 11.08, "eta": "49:47:21", "max_grad_norm": 1.0, "loss": 0.8398929834365845, "grad_norm": 0.4048190116882324, "learning_rate": 0.00019944984417479805}
+{"ts": "2025-12-27T00:29:47", "event": "train_log", "step": 1578, "epoch": 0.6658227848101266, "progress_pct": 11.1, "epoch_pct": 11.1, "eta": "49:45:09", "max_grad_norm": 1.0, "loss": 0.8269520998001099, "grad_norm": 0.4023774266242981, "learning_rate": 0.00019944480157564826}
+{"ts": "2025-12-27T00:30:02", "event": "train_log", "step": 1580, "epoch": 0.6666666666666666, "progress_pct": 11.11, "epoch_pct": 11.11, "eta": "49:42:52", "max_grad_norm": 1.0, "loss": 0.8237608671188354, "grad_norm": 0.3784184157848358, "learning_rate": 0.00019943973603649262}
+{"ts": "2025-12-27T00:30:17", "event": "train_log", "step": 1582, "epoch": 0.6675105485232068, "progress_pct": 11.13, "epoch_pct": 11.13, "eta": "49:40:36", "max_grad_norm": 1.0, "loss": 0.8469099998474121, "grad_norm": 0.4448573887348175, "learning_rate": 0.00019943464755849965}
+{"ts": "2025-12-27T00:30:32", "event": "train_log", "step": 1584, "epoch": 0.6683544303797468, "progress_pct": 11.14, "epoch_pct": 11.14, "eta": "49:38:23", "max_grad_norm": 1.0, "loss": 0.8630822896957397, "grad_norm": 0.4690774083137512, "learning_rate": 0.00019942953614284317}
+{"ts": "2025-12-27T00:30:47", "event": "train_log", "step": 1586, "epoch": 0.6691983122362869, "progress_pct": 11.15, "epoch_pct": 11.15, "eta": "49:36:07", "max_grad_norm": 1.0, "loss": 0.8904082179069519, "grad_norm": 0.43670853972435, "learning_rate": 0.00019942440179070227}
+{"ts": "2025-12-27T00:31:02", "event": "train_log", "step": 1588, "epoch": 0.6700421940928271, "progress_pct": 11.17, "epoch_pct": 11.17, "eta": "49:33:56", "max_grad_norm": 1.0, "loss": 0.8355640172958374, "grad_norm": 0.4236849844455719, "learning_rate": 0.00019941924450326138}
+{"ts": "2025-12-27T00:31:16", "event": "train_log", "step": 1590, "epoch": 0.6708860759493671, "progress_pct": 11.18, "epoch_pct": 11.18, "eta": "49:31:36", "max_grad_norm": 1.0, "loss": 0.8898435831069946, "grad_norm": 0.48624539375305176, "learning_rate": 0.00019941406428171013}
+{"ts": "2025-12-27T00:31:30", "event": "train_log", "step": 1592, "epoch": 0.6717299578059072, "progress_pct": 11.2, "epoch_pct": 11.2, "eta": "49:29:16", "max_grad_norm": 1.0, "loss": 0.9000012278556824, "grad_norm": 0.4453780949115753, "learning_rate": 0.00019940886112724355}
+{"ts": "2025-12-27T00:31:45", "event": "train_log", "step": 1594, "epoch": 0.6725738396624472, "progress_pct": 11.21, "epoch_pct": 11.21, "eta": "49:27:00", "max_grad_norm": 1.0, "loss": 0.8171058893203735, "grad_norm": 0.4417199194431305, "learning_rate": 0.00019940363504106193}
+{"ts": "2025-12-27T00:31:59", "event": "train_log", "step": 1596, "epoch": 0.6734177215189874, "progress_pct": 11.22, "epoch_pct": 11.22, "eta": "49:24:43", "max_grad_norm": 1.0, "loss": 0.8703798055648804, "grad_norm": 0.45055362582206726, "learning_rate": 0.00019939838602437074}
+{"ts": "2025-12-27T00:32:14", "event": "train_log", "step": 1598, "epoch": 0.6742616033755274, "progress_pct": 11.24, "epoch_pct": 11.24, "eta": "49:22:27", "max_grad_norm": 1.0, "loss": 0.8677281737327576, "grad_norm": 0.42316532135009766, "learning_rate": 0.00019939311407838088}
+{"ts": "2025-12-27T00:32:29", "event": "train_log", "step": 1600, "epoch": 0.6751054852320675, "progress_pct": 11.25, "epoch_pct": 11.25, "eta": "49:20:19", "max_grad_norm": 1.0, "loss": 0.7314013242721558, "grad_norm": 0.3935781717300415, "learning_rate": 0.0001993878192043085}
+{"ts": "2025-12-27T00:43:47", "event": "train_log", "step": 1600, "epoch": 0.6751054852320675, "progress_pct": 11.25, "epoch_pct": 11.25, "eta": "50:49:27", "max_grad_norm": 1.0, "eval_loss": 0.8769772052764893, "eval_runtime": 677.9338, "eval_samples_per_second": 3.108, "eval_steps_per_second": 3.108}
+{"ts": "2025-12-27T00:44:03", "event": "train_log", "step": 1602, "epoch": 0.6759493670886076, "progress_pct": 11.27, "epoch_pct": 11.27, "eta": "50:47:17", "max_grad_norm": 1.0, "loss": 0.8075380325317383, "grad_norm": 0.4197012186050415, "learning_rate": 0.000199382501403375}
+{"ts": "2025-12-27T00:44:20", "event": "train_log", "step": 1604, "epoch": 0.6767932489451477, "progress_pct": 11.28, "epoch_pct": 11.28, "eta": "50:45:10", "max_grad_norm": 1.0, "loss": 0.7885794639587402, "grad_norm": 0.45227327942848206, "learning_rate": 0.00019937716067680713}
+{"ts": "2025-12-27T00:44:35", "event": "train_log", "step": 1606, "epoch": 0.6776371308016877, "progress_pct": 11.29, "epoch_pct": 11.29, "eta": "50:42:55", "max_grad_norm": 1.0, "loss": 0.868261992931366, "grad_norm": 0.45660385489463806, "learning_rate": 0.00019937179702583682}
+{"ts": "2025-12-27T00:44:51", "event": "train_log", "step": 1608, "epoch": 0.6784810126582278, "progress_pct": 11.31, "epoch_pct": 11.31, "eta": "50:40:44", "max_grad_norm": 1.0, "loss": 0.8306655287742615, "grad_norm": 0.3955288231372833, "learning_rate": 0.00019936641045170144}
+{"ts": "2025-12-27T00:45:07", "event": "train_log", "step": 1610, "epoch": 0.679324894514768, "progress_pct": 11.32, "epoch_pct": 11.32, "eta": "50:38:31", "max_grad_norm": 1.0, "loss": 0.7505315542221069, "grad_norm": 0.3651324510574341, "learning_rate": 0.00019936100095564353}
+{"ts": "2025-12-27T00:45:23", "event": "train_log", "step": 1612, "epoch": 0.680168776371308, "progress_pct": 11.34, "epoch_pct": 11.34, "eta": "50:36:19", "max_grad_norm": 1.0, "loss": 0.8792756795883179, "grad_norm": 0.4240955412387848, "learning_rate": 0.00019935556853891096}
+{"ts": "2025-12-27T00:45:40", "event": "train_log", "step": 1614, "epoch": 0.6810126582278481, "progress_pct": 11.35, "epoch_pct": 11.35, "eta": "50:34:20", "max_grad_norm": 1.0, "loss": 0.7328272461891174, "grad_norm": 0.4142507314682007, "learning_rate": 0.00019935011320275687}
+{"ts": "2025-12-27T00:45:54", "event": "train_log", "step": 1616, "epoch": 0.6818565400843882, "progress_pct": 11.36, "epoch_pct": 11.36, "eta": "50:31:57", "max_grad_norm": 1.0, "loss": 0.8414849042892456, "grad_norm": 0.48339322209358215, "learning_rate": 0.00019934463494843975}
+{"ts": "2025-12-27T00:46:09", "event": "train_log", "step": 1618, "epoch": 0.6827004219409283, "progress_pct": 11.38, "epoch_pct": 11.38, "eta": "50:29:41", "max_grad_norm": 1.0, "loss": 0.8588716983795166, "grad_norm": 0.4312443435192108, "learning_rate": 0.00019933913377722328}
+{"ts": "2025-12-27T00:46:24", "event": "train_log", "step": 1620, "epoch": 0.6835443037974683, "progress_pct": 11.39, "epoch_pct": 11.39, "eta": "50:27:24", "max_grad_norm": 1.0, "loss": 0.846511960029602, "grad_norm": 0.43849512934684753, "learning_rate": 0.0001993336096903765}
+{"ts": "2025-12-27T00:46:40", "event": "train_log", "step": 1622, "epoch": 0.6843881856540084, "progress_pct": 11.41, "epoch_pct": 11.41, "eta": "50:25:12", "max_grad_norm": 1.0, "loss": 0.9352323412895203, "grad_norm": 0.45767566561698914, "learning_rate": 0.0001993280626891737}
+{"ts": "2025-12-27T00:46:55", "event": "train_log", "step": 1624, "epoch": 0.6852320675105485, "progress_pct": 11.42, "epoch_pct": 11.42, "eta": "50:22:55", "max_grad_norm": 1.0, "loss": 0.8884757161140442, "grad_norm": 0.438412606716156, "learning_rate": 0.00019932249277489444}
+{"ts": "2025-12-27T00:47:10", "event": "train_log", "step": 1626, "epoch": 0.6860759493670886, "progress_pct": 11.43, "epoch_pct": 11.43, "eta": "50:20:41", "max_grad_norm": 1.0, "loss": 0.7739649415016174, "grad_norm": 0.45893922448158264, "learning_rate": 0.0001993168999488237}
+{"ts": "2025-12-27T00:47:26", "event": "train_log", "step": 1628, "epoch": 0.6869198312236287, "progress_pct": 11.45, "epoch_pct": 11.45, "eta": "50:18:32", "max_grad_norm": 1.0, "loss": 0.7948258519172668, "grad_norm": 0.38631772994995117, "learning_rate": 0.0001993112842122515}
+{"ts": "2025-12-27T00:47:42", "event": "train_log", "step": 1630, "epoch": 0.6877637130801688, "progress_pct": 11.46, "epoch_pct": 11.46, "eta": "50:16:27", "max_grad_norm": 1.0, "loss": 0.750511884689331, "grad_norm": 0.3824027180671692, "learning_rate": 0.00019930564556647334}
+{"ts": "2025-12-27T00:47:57", "event": "train_log", "step": 1632, "epoch": 0.6886075949367089, "progress_pct": 11.48, "epoch_pct": 11.48, "eta": "50:14:09", "max_grad_norm": 1.0, "loss": 0.9006738662719727, "grad_norm": 0.3980776071548462, "learning_rate": 0.00019929998401278996}
+{"ts": "2025-12-27T00:48:12", "event": "train_log", "step": 1634, "epoch": 0.6894514767932489, "progress_pct": 11.49, "epoch_pct": 11.49, "eta": "50:11:56", "max_grad_norm": 1.0, "loss": 0.6996869444847107, "grad_norm": 0.41419896483421326, "learning_rate": 0.00019929429955250734}
+{"ts": "2025-12-27T00:48:28", "event": "train_log", "step": 1636, "epoch": 0.6902953586497891, "progress_pct": 11.5, "epoch_pct": 11.5, "eta": "50:09:49", "max_grad_norm": 1.0, "loss": 0.8523393273353577, "grad_norm": 0.43783676624298096, "learning_rate": 0.00019928859218693682}
+{"ts": "2025-12-27T00:48:44", "event": "train_log", "step": 1638, "epoch": 0.6911392405063291, "progress_pct": 11.52, "epoch_pct": 11.52, "eta": "50:07:46", "max_grad_norm": 1.0, "loss": 0.8885987401008606, "grad_norm": 0.4378126561641693, "learning_rate": 0.00019928286191739497}
+{"ts": "2025-12-27T00:49:00", "event": "train_log", "step": 1640, "epoch": 0.6919831223628692, "progress_pct": 11.53, "epoch_pct": 11.53, "eta": "50:05:35", "max_grad_norm": 1.0, "loss": 0.8563809990882874, "grad_norm": 0.4276133179664612, "learning_rate": 0.00019927710874520363}
+{"ts": "2025-12-27T00:49:15", "event": "train_log", "step": 1642, "epoch": 0.6928270042194092, "progress_pct": 11.55, "epoch_pct": 11.55, "eta": "50:03:22", "max_grad_norm": 1.0, "loss": 0.854114294052124, "grad_norm": 0.4600282311439514, "learning_rate": 0.00019927133267168992}
+{"ts": "2025-12-27T00:49:29", "event": "train_log", "step": 1644, "epoch": 0.6936708860759494, "progress_pct": 11.56, "epoch_pct": 11.56, "eta": "50:01:05", "max_grad_norm": 1.0, "loss": 0.8557889461517334, "grad_norm": 0.5058461427688599, "learning_rate": 0.00019926553369818635}
+{"ts": "2025-12-27T00:49:44", "event": "train_log", "step": 1646, "epoch": 0.6945147679324895, "progress_pct": 11.58, "epoch_pct": 11.58, "eta": "49:58:54", "max_grad_norm": 1.0, "loss": 0.8622767925262451, "grad_norm": 0.4947412312030792, "learning_rate": 0.00019925971182603057}
+{"ts": "2025-12-27T00:50:01", "event": "train_log", "step": 1648, "epoch": 0.6953586497890295, "progress_pct": 11.59, "epoch_pct": 11.59, "eta": "49:56:50", "max_grad_norm": 1.0, "loss": 0.7836558818817139, "grad_norm": 0.42555496096611023, "learning_rate": 0.00019925386705656558}
+{"ts": "2025-12-27T00:50:16", "event": "train_log", "step": 1650, "epoch": 0.6962025316455697, "progress_pct": 11.6, "epoch_pct": 11.6, "eta": "49:54:44", "max_grad_norm": 1.0, "loss": 0.7877846956253052, "grad_norm": 0.47560831904411316, "learning_rate": 0.00019924799939113967}
+{"ts": "2025-12-27T00:50:33", "event": "train_log", "step": 1652, "epoch": 0.6970464135021097, "progress_pct": 11.62, "epoch_pct": 11.62, "eta": "49:52:41", "max_grad_norm": 1.0, "loss": 0.775681734085083, "grad_norm": 0.391702264547348, "learning_rate": 0.0001992421088311064}
+{"ts": "2025-12-27T00:50:49", "event": "train_log", "step": 1654, "epoch": 0.6978902953586498, "progress_pct": 11.63, "epoch_pct": 11.63, "eta": "49:50:42", "max_grad_norm": 1.0, "loss": 0.8809882402420044, "grad_norm": 0.463679701089859, "learning_rate": 0.0001992361953778246}
+{"ts": "2025-12-27T00:51:06", "event": "train_log", "step": 1656, "epoch": 0.6987341772151898, "progress_pct": 11.65, "epoch_pct": 11.65, "eta": "49:48:43", "max_grad_norm": 1.0, "loss": 0.8513299226760864, "grad_norm": 0.44096484780311584, "learning_rate": 0.00019923025903265836}
+{"ts": "2025-12-27T00:51:22", "event": "train_log", "step": 1658, "epoch": 0.69957805907173, "progress_pct": 11.66, "epoch_pct": 11.66, "eta": "49:46:39", "max_grad_norm": 1.0, "loss": 0.8309667706489563, "grad_norm": 0.47818854451179504, "learning_rate": 0.0001992242997969772}
+{"ts": "2025-12-27T00:51:37", "event": "train_log", "step": 1660, "epoch": 0.70042194092827, "progress_pct": 11.67, "epoch_pct": 11.67, "eta": "49:44:29", "max_grad_norm": 1.0, "loss": 0.8817558884620667, "grad_norm": 0.4192790985107422, "learning_rate": 0.00019921831767215565}
+{"ts": "2025-12-27T00:51:54", "event": "train_log", "step": 1662, "epoch": 0.7012658227848101, "progress_pct": 11.69, "epoch_pct": 11.69, "eta": "49:42:31", "max_grad_norm": 1.0, "loss": 0.8315755724906921, "grad_norm": 0.45285239815711975, "learning_rate": 0.00019921231265957373}
+{"ts": "2025-12-27T00:52:10", "event": "train_log", "step": 1664, "epoch": 0.7021097046413503, "progress_pct": 11.7, "epoch_pct": 11.7, "eta": "49:40:30", "max_grad_norm": 1.0, "loss": 0.8698058724403381, "grad_norm": 0.47118625044822693, "learning_rate": 0.00019920628476061673}
+{"ts": "2025-12-27T00:52:26", "event": "train_log", "step": 1666, "epoch": 0.7029535864978903, "progress_pct": 11.72, "epoch_pct": 11.72, "eta": "49:38:25", "max_grad_norm": 1.0, "loss": 0.953620970249176, "grad_norm": 0.4769238829612732, "learning_rate": 0.0001992002339766751}
+{"ts": "2025-12-27T00:52:41", "event": "train_log", "step": 1668, "epoch": 0.7037974683544304, "progress_pct": 11.73, "epoch_pct": 11.73, "eta": "49:36:21", "max_grad_norm": 1.0, "loss": 0.8536615371704102, "grad_norm": 0.39397355914115906, "learning_rate": 0.00019919416030914468}
+{"ts": "2025-12-27T00:52:59", "event": "train_log", "step": 1670, "epoch": 0.7046413502109705, "progress_pct": 11.74, "epoch_pct": 11.74, "eta": "49:34:31", "max_grad_norm": 1.0, "loss": 0.8159828782081604, "grad_norm": 0.3659648299217224, "learning_rate": 0.00019918806375942655}
+{"ts": "2025-12-27T00:53:14", "event": "train_log", "step": 1672, "epoch": 0.7054852320675106, "progress_pct": 11.76, "epoch_pct": 11.76, "eta": "49:32:25", "max_grad_norm": 1.0, "loss": 0.782463550567627, "grad_norm": 0.4381789565086365, "learning_rate": 0.00019918194432892703}
+{"ts": "2025-12-27T00:53:29", "event": "train_log", "step": 1674, "epoch": 0.7063291139240506, "progress_pct": 11.77, "epoch_pct": 11.77, "eta": "49:30:15", "max_grad_norm": 1.0, "loss": 0.9314272999763489, "grad_norm": 0.44745752215385437, "learning_rate": 0.00019917580201905778}
+{"ts": "2025-12-27T00:53:46", "event": "train_log", "step": 1676, "epoch": 0.7071729957805907, "progress_pct": 11.79, "epoch_pct": 11.79, "eta": "49:28:20", "max_grad_norm": 1.0, "loss": 0.860816240310669, "grad_norm": 0.4515164792537689, "learning_rate": 0.00019916963683123574}
+{"ts": "2025-12-27T00:54:02", "event": "train_log", "step": 1678, "epoch": 0.7080168776371308, "progress_pct": 11.8, "epoch_pct": 11.8, "eta": "49:26:16", "max_grad_norm": 1.0, "loss": 0.8008456826210022, "grad_norm": 0.42735886573791504, "learning_rate": 0.00019916344876688303}
+{"ts": "2025-12-27T00:54:15", "event": "train_log", "step": 1680, "epoch": 0.7088607594936709, "progress_pct": 11.81, "epoch_pct": 11.81, "eta": "49:23:57", "max_grad_norm": 1.0, "loss": 0.7796595096588135, "grad_norm": 0.418275386095047, "learning_rate": 0.00019915723782742718}
+{"ts": "2025-12-27T00:54:30", "event": "train_log", "step": 1682, "epoch": 0.7097046413502109, "progress_pct": 11.83, "epoch_pct": 11.83, "eta": "49:21:49", "max_grad_norm": 1.0, "loss": 0.9006506204605103, "grad_norm": 0.44347989559173584, "learning_rate": 0.00019915100401430089}
+{"ts": "2025-12-27T00:54:47", "event": "train_log", "step": 1684, "epoch": 0.7105485232067511, "progress_pct": 11.84, "epoch_pct": 11.84, "eta": "49:19:51", "max_grad_norm": 1.0, "loss": 0.7750146985054016, "grad_norm": 0.3932702839374542, "learning_rate": 0.00019914474732894225}
+{"ts": "2025-12-27T00:55:02", "event": "train_log", "step": 1686, "epoch": 0.7113924050632912, "progress_pct": 11.86, "epoch_pct": 11.86, "eta": "49:17:44", "max_grad_norm": 1.0, "loss": 0.7508097290992737, "grad_norm": 0.4810502231121063, "learning_rate": 0.00019913846777279447}
+{"ts": "2025-12-27T00:55:16", "event": "train_log", "step": 1688, "epoch": 0.7122362869198312, "progress_pct": 11.87, "epoch_pct": 11.87, "eta": "49:15:36", "max_grad_norm": 1.0, "loss": 0.7926796078681946, "grad_norm": 0.4166778028011322, "learning_rate": 0.00019913216534730622}
+{"ts": "2025-12-27T00:55:32", "event": "train_log", "step": 1690, "epoch": 0.7130801687763713, "progress_pct": 11.88, "epoch_pct": 11.88, "eta": "49:13:33", "max_grad_norm": 1.0, "loss": 0.8363928198814392, "grad_norm": 0.47900474071502686, "learning_rate": 0.00019912584005393123}
+{"ts": "2025-12-27T00:55:49", "event": "train_log", "step": 1692, "epoch": 0.7139240506329114, "progress_pct": 11.9, "epoch_pct": 11.9, "eta": "49:11:39", "max_grad_norm": 1.0, "loss": 0.8649623990058899, "grad_norm": 0.435143381357193, "learning_rate": 0.00019911949189412872}
+{"ts": "2025-12-27T00:56:05", "event": "train_log", "step": 1694, "epoch": 0.7147679324894515, "progress_pct": 11.91, "epoch_pct": 11.91, "eta": "49:09:44", "max_grad_norm": 1.0, "loss": 0.8196499943733215, "grad_norm": 0.4070208966732025, "learning_rate": 0.00019911312086936305}
+{"ts": "2025-12-27T00:56:21", "event": "train_log", "step": 1696, "epoch": 0.7156118143459915, "progress_pct": 11.93, "epoch_pct": 11.93, "eta": "49:07:45", "max_grad_norm": 1.0, "loss": 0.9157413840293884, "grad_norm": 0.4717724621295929, "learning_rate": 0.0001991067269811039}
+{"ts": "2025-12-27T00:56:36", "event": "train_log", "step": 1698, "epoch": 0.7164556962025317, "progress_pct": 11.94, "epoch_pct": 11.94, "eta": "49:05:37", "max_grad_norm": 1.0, "loss": 0.7880456447601318, "grad_norm": 0.3942880928516388, "learning_rate": 0.0001991003102308262}
+{"ts": "2025-12-27T00:56:52", "event": "train_log", "step": 1700, "epoch": 0.7172995780590717, "progress_pct": 11.95, "epoch_pct": 11.95, "eta": "49:03:40", "max_grad_norm": 1.0, "loss": 0.8088646531105042, "grad_norm": 0.4037776589393616, "learning_rate": 0.0001990938706200102}
+{"ts": "2025-12-27T01:08:03", "event": "train_log", "step": 1700, "epoch": 0.7172995780590717, "progress_pct": 11.95, "epoch_pct": 11.95, "eta": "50:25:57", "max_grad_norm": 1.0, "eval_loss": 0.8708170056343079, "eval_runtime": 670.3019, "eval_samples_per_second": 3.143, "eval_steps_per_second": 3.143}
+{"ts": "2025-12-27T01:08:17", "event": "train_log", "step": 1702, "epoch": 0.7181434599156118, "progress_pct": 11.97, "epoch_pct": 11.97, "eta": "50:23:38", "max_grad_norm": 1.0, "loss": 0.8740324378013611, "grad_norm": 0.4443536400794983, "learning_rate": 0.0001990874081501414}
+{"ts": "2025-12-27T01:08:32", "event": "train_log", "step": 1704, "epoch": 0.7189873417721518, "progress_pct": 11.98, "epoch_pct": 11.98, "eta": "50:21:26", "max_grad_norm": 1.0, "loss": 0.8305023908615112, "grad_norm": 0.45089733600616455, "learning_rate": 0.00019908092282271053}
+{"ts": "2025-12-27T01:08:47", "event": "train_log", "step": 1706, "epoch": 0.719831223628692, "progress_pct": 12.0, "epoch_pct": 12.0, "eta": "50:19:21", "max_grad_norm": 1.0, "loss": 0.7718394994735718, "grad_norm": 0.3903036415576935, "learning_rate": 0.00019907441463921366}
+{"ts": "2025-12-27T01:09:02", "event": "train_log", "step": 1708, "epoch": 0.7206751054852321, "progress_pct": 12.01, "epoch_pct": 12.01, "eta": "50:17:10", "max_grad_norm": 1.0, "loss": 0.843289315700531, "grad_norm": 0.42990729212760925, "learning_rate": 0.00019906788360115208}
+{"ts": "2025-12-27T01:09:17", "event": "train_log", "step": 1710, "epoch": 0.7215189873417721, "progress_pct": 12.03, "epoch_pct": 12.03, "eta": "50:14:56", "max_grad_norm": 1.0, "loss": 0.8332282900810242, "grad_norm": 0.4287833571434021, "learning_rate": 0.0001990613297100324}
+{"ts": "2025-12-27T01:09:32", "event": "train_log", "step": 1712, "epoch": 0.7223628691983123, "progress_pct": 12.04, "epoch_pct": 12.04, "eta": "50:12:43", "max_grad_norm": 1.0, "loss": 0.784180223941803, "grad_norm": 0.4557269513607025, "learning_rate": 0.00019905475296736646}
+{"ts": "2025-12-27T01:09:47", "event": "train_log", "step": 1714, "epoch": 0.7232067510548523, "progress_pct": 12.05, "epoch_pct": 12.05, "eta": "50:10:39", "max_grad_norm": 1.0, "loss": 0.7807716131210327, "grad_norm": 0.4001525640487671, "learning_rate": 0.0001990481533746714}
+{"ts": "2025-12-27T01:10:03", "event": "train_log", "step": 1716, "epoch": 0.7240506329113924, "progress_pct": 12.07, "epoch_pct": 12.07, "eta": "50:08:29", "max_grad_norm": 1.0, "loss": 0.7936414480209351, "grad_norm": 0.5208709836006165, "learning_rate": 0.0001990415309334696}
+{"ts": "2025-12-27T01:10:18", "event": "train_log", "step": 1718, "epoch": 0.7248945147679325, "progress_pct": 12.08, "epoch_pct": 12.08, "eta": "50:06:24", "max_grad_norm": 1.0, "loss": 0.7964845895767212, "grad_norm": 0.46469953656196594, "learning_rate": 0.00019903488564528875}
+{"ts": "2025-12-27T01:10:34", "event": "train_log", "step": 1720, "epoch": 0.7257383966244726, "progress_pct": 12.1, "epoch_pct": 12.1, "eta": "50:04:17", "max_grad_norm": 1.0, "loss": 0.8302215337753296, "grad_norm": 0.405773401260376, "learning_rate": 0.00019902821751166178}
+{"ts": "2025-12-27T01:10:49", "event": "train_log", "step": 1722, "epoch": 0.7265822784810126, "progress_pct": 12.11, "epoch_pct": 12.11, "eta": "50:02:10", "max_grad_norm": 1.0, "loss": 0.8486512899398804, "grad_norm": 0.3874973654747009, "learning_rate": 0.00019902152653412692}
+{"ts": "2025-12-27T01:11:03", "event": "train_log", "step": 1724, "epoch": 0.7274261603375527, "progress_pct": 12.12, "epoch_pct": 12.12, "eta": "49:59:57", "max_grad_norm": 1.0, "loss": 0.837841272354126, "grad_norm": 0.46865755319595337, "learning_rate": 0.00019901481271422758}
+{"ts": "2025-12-27T01:11:19", "event": "train_log", "step": 1726, "epoch": 0.7282700421940929, "progress_pct": 12.14, "epoch_pct": 12.14, "eta": "49:57:51", "max_grad_norm": 1.0, "loss": 0.7640942335128784, "grad_norm": 0.4139123558998108, "learning_rate": 0.0001990080760535126}
+{"ts": "2025-12-27T01:11:33", "event": "train_log", "step": 1728, "epoch": 0.7291139240506329, "progress_pct": 12.15, "epoch_pct": 12.15, "eta": "49:55:34", "max_grad_norm": 1.0, "loss": 0.8236244320869446, "grad_norm": 0.5003024935722351, "learning_rate": 0.000199001316553536}
+{"ts": "2025-12-27T01:11:47", "event": "train_log", "step": 1730, "epoch": 0.729957805907173, "progress_pct": 12.17, "epoch_pct": 12.17, "eta": "49:53:23", "max_grad_norm": 1.0, "loss": 0.7792612314224243, "grad_norm": 0.38495776057243347, "learning_rate": 0.00019899453421585696}
+{"ts": "2025-12-27T01:12:03", "event": "train_log", "step": 1732, "epoch": 0.7308016877637131, "progress_pct": 12.18, "epoch_pct": 12.18, "eta": "49:51:19", "max_grad_norm": 1.0, "loss": 0.7733872532844543, "grad_norm": 0.4462733268737793, "learning_rate": 0.00019898772904204014}
+{"ts": "2025-12-27T01:12:18", "event": "train_log", "step": 1734, "epoch": 0.7316455696202532, "progress_pct": 12.19, "epoch_pct": 12.19, "eta": "49:49:13", "max_grad_norm": 1.0, "loss": 0.8933147192001343, "grad_norm": 0.398248553276062, "learning_rate": 0.00019898090103365532}
+{"ts": "2025-12-27T01:12:33", "event": "train_log", "step": 1736, "epoch": 0.7324894514767932, "progress_pct": 12.21, "epoch_pct": 12.21, "eta": "49:47:06", "max_grad_norm": 1.0, "loss": 0.8198356628417969, "grad_norm": 0.4231175482273102, "learning_rate": 0.00019897405019227756}
+{"ts": "2025-12-27T01:12:48", "event": "train_log", "step": 1738, "epoch": 0.7333333333333333, "progress_pct": 12.22, "epoch_pct": 12.22, "eta": "49:45:00", "max_grad_norm": 1.0, "loss": 0.8266391158103943, "grad_norm": 0.43639084696769714, "learning_rate": 0.0001989671765194873}
+{"ts": "2025-12-27T01:13:05", "event": "train_log", "step": 1740, "epoch": 0.7341772151898734, "progress_pct": 12.24, "epoch_pct": 12.24, "eta": "49:43:03", "max_grad_norm": 1.0, "loss": 0.7839997410774231, "grad_norm": 0.4235835671424866, "learning_rate": 0.00019896028001687008}
+{"ts": "2025-12-27T01:13:21", "event": "train_log", "step": 1742, "epoch": 0.7350210970464135, "progress_pct": 12.25, "epoch_pct": 12.25, "eta": "49:41:05", "max_grad_norm": 1.0, "loss": 0.7153518199920654, "grad_norm": 0.4033111035823822, "learning_rate": 0.00019895336068601687}
+{"ts": "2025-12-27T01:13:36", "event": "train_log", "step": 1744, "epoch": 0.7358649789029535, "progress_pct": 12.26, "epoch_pct": 12.26, "eta": "49:38:57", "max_grad_norm": 1.0, "loss": 0.9079289436340332, "grad_norm": 0.44281265139579773, "learning_rate": 0.00019894641852852377}
+{"ts": "2025-12-27T01:13:51", "event": "train_log", "step": 1746, "epoch": 0.7367088607594937, "progress_pct": 12.28, "epoch_pct": 12.28, "eta": "49:36:52", "max_grad_norm": 1.0, "loss": 0.8645254373550415, "grad_norm": 0.4882747232913971, "learning_rate": 0.0001989394535459922}
+{"ts": "2025-12-27T01:14:08", "event": "train_log", "step": 1748, "epoch": 0.7375527426160338, "progress_pct": 12.29, "epoch_pct": 12.29, "eta": "49:35:01", "max_grad_norm": 1.0, "loss": 0.8134214878082275, "grad_norm": 0.34858641028404236, "learning_rate": 0.00019893246574002887}
+{"ts": "2025-12-27T01:14:24", "event": "train_log", "step": 1750, "epoch": 0.7383966244725738, "progress_pct": 12.31, "epoch_pct": 12.31, "eta": "49:33:01", "max_grad_norm": 1.0, "loss": 0.8142663240432739, "grad_norm": 0.4273303747177124, "learning_rate": 0.00019892545511224574}
+{"ts": "2025-12-27T01:14:38", "event": "train_log", "step": 1752, "epoch": 0.739240506329114, "progress_pct": 12.32, "epoch_pct": 12.32, "eta": "49:30:52", "max_grad_norm": 1.0, "loss": 0.7957643866539001, "grad_norm": 0.4289957284927368, "learning_rate": 0.00019891842166426004}
+{"ts": "2025-12-27T01:14:54", "event": "train_log", "step": 1754, "epoch": 0.740084388185654, "progress_pct": 12.33, "epoch_pct": 12.33, "eta": "49:28:50", "max_grad_norm": 1.0, "loss": 0.9000363945960999, "grad_norm": 0.4812224805355072, "learning_rate": 0.00019891136539769416}
+{"ts": "2025-12-27T01:15:09", "event": "train_log", "step": 1756, "epoch": 0.7409282700421941, "progress_pct": 12.35, "epoch_pct": 12.35, "eta": "49:26:51", "max_grad_norm": 1.0, "loss": 0.7719792723655701, "grad_norm": 0.4078022837638855, "learning_rate": 0.00019890428631417595}
+{"ts": "2025-12-27T01:15:25", "event": "train_log", "step": 1758, "epoch": 0.7417721518987341, "progress_pct": 12.36, "epoch_pct": 12.36, "eta": "49:24:49", "max_grad_norm": 1.0, "loss": 0.7681847214698792, "grad_norm": 0.40169858932495117, "learning_rate": 0.00019889718441533838}
+{"ts": "2025-12-27T01:15:42", "event": "train_log", "step": 1760, "epoch": 0.7426160337552743, "progress_pct": 12.38, "epoch_pct": 12.38, "eta": "49:22:58", "max_grad_norm": 1.0, "loss": 0.7536827325820923, "grad_norm": 0.3713594377040863, "learning_rate": 0.00019889005970281972}
+{"ts": "2025-12-27T01:15:56", "event": "train_log", "step": 1762, "epoch": 0.7434599156118143, "progress_pct": 12.39, "epoch_pct": 12.39, "eta": "49:20:52", "max_grad_norm": 1.0, "loss": 0.9000160694122314, "grad_norm": 0.4183627665042877, "learning_rate": 0.0001988829121782635}
+{"ts": "2025-12-27T01:16:12", "event": "train_log", "step": 1764, "epoch": 0.7443037974683544, "progress_pct": 12.41, "epoch_pct": 12.41, "eta": "49:18:51", "max_grad_norm": 1.0, "loss": 0.8047307133674622, "grad_norm": 0.4241594672203064, "learning_rate": 0.00019887574184331851}
+{"ts": "2025-12-27T01:16:27", "event": "train_log", "step": 1766, "epoch": 0.7451476793248946, "progress_pct": 12.42, "epoch_pct": 12.42, "eta": "49:16:51", "max_grad_norm": 1.0, "loss": 0.8932583928108215, "grad_norm": 0.4249233603477478, "learning_rate": 0.00019886854869963883}
+{"ts": "2025-12-27T01:16:45", "event": "train_log", "step": 1768, "epoch": 0.7459915611814346, "progress_pct": 12.43, "epoch_pct": 12.43, "eta": "49:15:07", "max_grad_norm": 1.0, "loss": 0.7684977054595947, "grad_norm": 0.35815340280532837, "learning_rate": 0.00019886133274888378}
+{"ts": "2025-12-27T01:17:00", "event": "train_log", "step": 1770, "epoch": 0.7468354430379747, "progress_pct": 12.45, "epoch_pct": 12.45, "eta": "49:13:05", "max_grad_norm": 1.0, "loss": 0.8644338250160217, "grad_norm": 0.41035401821136475, "learning_rate": 0.00019885409399271795}
+{"ts": "2025-12-27T01:17:16", "event": "train_log", "step": 1772, "epoch": 0.7476793248945147, "progress_pct": 12.46, "epoch_pct": 12.46, "eta": "49:11:07", "max_grad_norm": 1.0, "loss": 0.8035860657691956, "grad_norm": 0.5450286269187927, "learning_rate": 0.00019884683243281116}
+{"ts": "2025-12-27T01:17:31", "event": "train_log", "step": 1774, "epoch": 0.7485232067510549, "progress_pct": 12.48, "epoch_pct": 12.48, "eta": "49:09:03", "max_grad_norm": 1.0, "loss": 0.8861207962036133, "grad_norm": 0.5280999541282654, "learning_rate": 0.0001988395480708385}
+{"ts": "2025-12-27T01:17:47", "event": "train_log", "step": 1776, "epoch": 0.7493670886075949, "progress_pct": 12.49, "epoch_pct": 12.49, "eta": "49:07:05", "max_grad_norm": 1.0, "loss": 0.7506847977638245, "grad_norm": 0.42665231227874756, "learning_rate": 0.00019883224090848036}
+{"ts": "2025-12-27T01:18:04", "event": "train_log", "step": 1778, "epoch": 0.750210970464135, "progress_pct": 12.5, "epoch_pct": 12.5, "eta": "49:05:22", "max_grad_norm": 1.0, "loss": 0.7662046551704407, "grad_norm": 0.39029282331466675, "learning_rate": 0.00019882491094742232}
+{"ts": "2025-12-27T01:18:21", "event": "train_log", "step": 1780, "epoch": 0.7510548523206751, "progress_pct": 12.52, "epoch_pct": 12.52, "eta": "49:03:34", "max_grad_norm": 1.0, "loss": 0.7590143084526062, "grad_norm": 0.43870019912719727, "learning_rate": 0.00019881755818935534}
+{"ts": "2025-12-27T01:18:36", "event": "train_log", "step": 1782, "epoch": 0.7518987341772152, "progress_pct": 12.53, "epoch_pct": 12.53, "eta": "49:01:32", "max_grad_norm": 1.0, "loss": 0.9340365529060364, "grad_norm": 0.4195050001144409, "learning_rate": 0.0001988101826359755}
+{"ts": "2025-12-27T01:18:51", "event": "train_log", "step": 1784, "epoch": 0.7527426160337553, "progress_pct": 12.55, "epoch_pct": 12.55, "eta": "48:59:28", "max_grad_norm": 1.0, "loss": 0.7969209551811218, "grad_norm": 0.4236123263835907, "learning_rate": 0.0001988027842889842}
+{"ts": "2025-12-27T01:19:06", "event": "train_log", "step": 1786, "epoch": 0.7535864978902953, "progress_pct": 12.56, "epoch_pct": 12.56, "eta": "48:57:27", "max_grad_norm": 1.0, "loss": 0.7892382144927979, "grad_norm": 0.4274357259273529, "learning_rate": 0.00019879536315008808}
+{"ts": "2025-12-27T01:19:22", "event": "train_log", "step": 1788, "epoch": 0.7544303797468355, "progress_pct": 12.57, "epoch_pct": 12.57, "eta": "48:55:31", "max_grad_norm": 1.0, "loss": 0.8752562403678894, "grad_norm": 0.4622916579246521, "learning_rate": 0.00019878791922099912}
+{"ts": "2025-12-27T01:19:38", "event": "train_log", "step": 1790, "epoch": 0.7552742616033755, "progress_pct": 12.59, "epoch_pct": 12.59, "eta": "48:53:37", "max_grad_norm": 1.0, "loss": 0.813011884689331, "grad_norm": 0.3648734986782074, "learning_rate": 0.00019878045250343445}
+{"ts": "2025-12-27T01:19:53", "event": "train_log", "step": 1792, "epoch": 0.7561181434599156, "progress_pct": 12.6, "epoch_pct": 12.6, "eta": "48:51:39", "max_grad_norm": 1.0, "loss": 0.7310198545455933, "grad_norm": 0.41361239552497864, "learning_rate": 0.00019877296299911648}
+{"ts": "2025-12-27T01:20:09", "event": "train_log", "step": 1794, "epoch": 0.7569620253164557, "progress_pct": 12.62, "epoch_pct": 12.62, "eta": "48:49:44", "max_grad_norm": 1.0, "loss": 0.858386754989624, "grad_norm": 0.4113222062587738, "learning_rate": 0.00019876545070977294}
+{"ts": "2025-12-27T01:20:25", "event": "train_log", "step": 1796, "epoch": 0.7578059071729958, "progress_pct": 12.63, "epoch_pct": 12.63, "eta": "48:47:48", "max_grad_norm": 1.0, "loss": 0.8377325534820557, "grad_norm": 0.4295370876789093, "learning_rate": 0.00019875791563713676}
+{"ts": "2025-12-27T01:20:40", "event": "train_log", "step": 1798, "epoch": 0.7586497890295358, "progress_pct": 12.64, "epoch_pct": 12.64, "eta": "48:45:54", "max_grad_norm": 1.0, "loss": 0.9139418005943298, "grad_norm": 0.4227522015571594, "learning_rate": 0.0001987503577829461}
+{"ts": "2025-12-27T01:20:57", "event": "train_log", "step": 1800, "epoch": 0.759493670886076, "progress_pct": 12.66, "epoch_pct": 12.66, "eta": "48:44:02", "max_grad_norm": 1.0, "loss": 0.6994872689247131, "grad_norm": 0.3638151288032532, "learning_rate": 0.00019874277714894442}
+{"ts": "2025-12-27T01:32:23", "event": "train_log", "step": 1800, "epoch": 0.759493670886076, "progress_pct": 12.66, "epoch_pct": 12.66, "eta": "50:02:59", "max_grad_norm": 1.0, "eval_loss": 0.8625519275665283, "eval_runtime": 686.4271, "eval_samples_per_second": 3.07, "eval_steps_per_second": 3.07}
+{"ts": "2025-12-27T01:32:38", "event": "train_log", "step": 1802, "epoch": 0.760337552742616, "progress_pct": 12.67, "epoch_pct": 12.67, "eta": "50:00:51", "max_grad_norm": 1.0, "loss": 0.8681167960166931, "grad_norm": 0.46954411268234253, "learning_rate": 0.00019873517373688047}
+{"ts": "2025-12-27T01:32:52", "event": "train_log", "step": 1804, "epoch": 0.7611814345991561, "progress_pct": 12.69, "epoch_pct": 12.69, "eta": "49:58:42", "max_grad_norm": 1.0, "loss": 0.7420852184295654, "grad_norm": 0.41958674788475037, "learning_rate": 0.00019872754754850819}
+{"ts": "2025-12-27T01:33:07", "event": "train_log", "step": 1806, "epoch": 0.7620253164556962, "progress_pct": 12.7, "epoch_pct": 12.7, "eta": "49:56:37", "max_grad_norm": 1.0, "loss": 0.8018608093261719, "grad_norm": 0.431226521730423, "learning_rate": 0.00019871989858558678}
+{"ts": "2025-12-27T01:33:22", "event": "train_log", "step": 1808, "epoch": 0.7628691983122363, "progress_pct": 12.71, "epoch_pct": 12.71, "eta": "49:54:32", "max_grad_norm": 1.0, "loss": 0.8793904781341553, "grad_norm": 0.45129454135894775, "learning_rate": 0.0001987122268498807}
+{"ts": "2025-12-27T01:33:36", "event": "train_log", "step": 1810, "epoch": 0.7637130801687764, "progress_pct": 12.73, "epoch_pct": 12.73, "eta": "49:52:21", "max_grad_norm": 1.0, "loss": 0.8416730761528015, "grad_norm": 0.44278961420059204, "learning_rate": 0.00019870453234315972}
+{"ts": "2025-12-27T01:33:51", "event": "train_log", "step": 1812, "epoch": 0.7645569620253164, "progress_pct": 12.74, "epoch_pct": 12.74, "eta": "49:50:15", "max_grad_norm": 1.0, "loss": 0.8225743174552917, "grad_norm": 0.4689098000526428, "learning_rate": 0.00019869681506719876}
+{"ts": "2025-12-27T01:34:05", "event": "train_log", "step": 1814, "epoch": 0.7654008438818566, "progress_pct": 12.76, "epoch_pct": 12.76, "eta": "49:48:05", "max_grad_norm": 1.0, "loss": 0.7871913313865662, "grad_norm": 0.4413386881351471, "learning_rate": 0.00019868907502377806}
+{"ts": "2025-12-27T01:34:20", "event": "train_log", "step": 1816, "epoch": 0.7662447257383966, "progress_pct": 12.77, "epoch_pct": 12.77, "eta": "49:46:00", "max_grad_norm": 1.0, "loss": 0.8616237044334412, "grad_norm": 0.48395273089408875, "learning_rate": 0.00019868131221468315}
+{"ts": "2025-12-27T01:34:36", "event": "train_log", "step": 1818, "epoch": 0.7670886075949367, "progress_pct": 12.78, "epoch_pct": 12.78, "eta": "49:44:01", "max_grad_norm": 1.0, "loss": 0.7243452668190002, "grad_norm": 0.370263934135437, "learning_rate": 0.0001986735266417047}
+{"ts": "2025-12-27T01:34:51", "event": "train_log", "step": 1820, "epoch": 0.7679324894514767, "progress_pct": 12.8, "epoch_pct": 12.8, "eta": "49:41:57", "max_grad_norm": 1.0, "loss": 0.8623812198638916, "grad_norm": 0.4522445797920227, "learning_rate": 0.00019866571830663875}
+{"ts": "2025-12-27T01:35:06", "event": "train_log", "step": 1822, "epoch": 0.7687763713080169, "progress_pct": 12.81, "epoch_pct": 12.81, "eta": "49:39:56", "max_grad_norm": 1.0, "loss": 0.776618480682373, "grad_norm": 0.43821004033088684, "learning_rate": 0.00019865788721128643}
+{"ts": "2025-12-27T01:35:21", "event": "train_log", "step": 1824, "epoch": 0.769620253164557, "progress_pct": 12.83, "epoch_pct": 12.83, "eta": "49:37:55", "max_grad_norm": 1.0, "loss": 0.8722774982452393, "grad_norm": 0.3825650215148926, "learning_rate": 0.00019865003335745436}
+{"ts": "2025-12-27T01:35:36", "event": "train_log", "step": 1826, "epoch": 0.770464135021097, "progress_pct": 12.84, "epoch_pct": 12.84, "eta": "49:35:51", "max_grad_norm": 1.0, "loss": 0.907802939414978, "grad_norm": 0.43684205412864685, "learning_rate": 0.00019864215674695416}
+{"ts": "2025-12-27T01:35:51", "event": "train_log", "step": 1828, "epoch": 0.7713080168776372, "progress_pct": 12.86, "epoch_pct": 12.86, "eta": "49:33:49", "max_grad_norm": 1.0, "loss": 0.8505539298057556, "grad_norm": 0.4128975570201874, "learning_rate": 0.00019863425738160287}
+{"ts": "2025-12-27T01:36:06", "event": "train_log", "step": 1830, "epoch": 0.7721518987341772, "progress_pct": 12.87, "epoch_pct": 12.87, "eta": "49:31:46", "max_grad_norm": 1.0, "loss": 0.83274906873703, "grad_norm": 0.4254065454006195, "learning_rate": 0.0001986263352632227}
+{"ts": "2025-12-27T01:36:21", "event": "train_log", "step": 1832, "epoch": 0.7729957805907173, "progress_pct": 12.88, "epoch_pct": 12.88, "eta": "49:29:42", "max_grad_norm": 1.0, "loss": 0.8172078728675842, "grad_norm": 0.4137375056743622, "learning_rate": 0.00019861839039364118}
+{"ts": "2025-12-27T01:36:35", "event": "train_log", "step": 1834, "epoch": 0.7738396624472574, "progress_pct": 12.9, "epoch_pct": 12.9, "eta": "49:27:34", "max_grad_norm": 1.0, "loss": 0.8610842227935791, "grad_norm": 0.4387703239917755, "learning_rate": 0.00019861042277469096}
+{"ts": "2025-12-27T01:36:50", "event": "train_log", "step": 1836, "epoch": 0.7746835443037975, "progress_pct": 12.91, "epoch_pct": 12.91, "eta": "49:25:32", "max_grad_norm": 1.0, "loss": 0.7961188554763794, "grad_norm": 0.398952454328537, "learning_rate": 0.00019860243240821005}
+{"ts": "2025-12-27T01:37:05", "event": "train_log", "step": 1838, "epoch": 0.7755274261603375, "progress_pct": 12.93, "epoch_pct": 12.93, "eta": "49:23:32", "max_grad_norm": 1.0, "loss": 0.8078321218490601, "grad_norm": 0.4079921543598175, "learning_rate": 0.0001985944192960417}
+{"ts": "2025-12-27T01:37:19", "event": "train_log", "step": 1840, "epoch": 0.7763713080168776, "progress_pct": 12.94, "epoch_pct": 12.94, "eta": "49:21:27", "max_grad_norm": 1.0, "loss": 0.8743909001350403, "grad_norm": 0.4701489508152008, "learning_rate": 0.00019858638344003433}
+{"ts": "2025-12-27T01:37:34", "event": "train_log", "step": 1842, "epoch": 0.7772151898734178, "progress_pct": 12.95, "epoch_pct": 12.95, "eta": "49:19:25", "max_grad_norm": 1.0, "loss": 0.7681707143783569, "grad_norm": 0.5391269326210022, "learning_rate": 0.00019857832484204173}
+{"ts": "2025-12-27T01:37:50", "event": "train_log", "step": 1844, "epoch": 0.7780590717299578, "progress_pct": 12.97, "epoch_pct": 12.97, "eta": "49:17:30", "max_grad_norm": 1.0, "loss": 0.7817525863647461, "grad_norm": 0.36250734329223633, "learning_rate": 0.00019857024350392277}
+{"ts": "2025-12-27T01:38:05", "event": "train_log", "step": 1846, "epoch": 0.7789029535864979, "progress_pct": 12.98, "epoch_pct": 12.98, "eta": "49:15:29", "max_grad_norm": 1.0, "loss": 0.7614551782608032, "grad_norm": 0.41237136721611023, "learning_rate": 0.00019856213942754175}
+{"ts": "2025-12-27T01:38:20", "event": "train_log", "step": 1848, "epoch": 0.779746835443038, "progress_pct": 13.0, "epoch_pct": 13.0, "eta": "49:13:31", "max_grad_norm": 1.0, "loss": 0.7492313981056213, "grad_norm": 0.3968869745731354, "learning_rate": 0.00019855401261476807}
+{"ts": "2025-12-27T01:38:35", "event": "train_log", "step": 1850, "epoch": 0.7805907172995781, "progress_pct": 13.01, "epoch_pct": 13.01, "eta": "49:11:29", "max_grad_norm": 1.0, "loss": 0.8615695238113403, "grad_norm": 0.441476434469223, "learning_rate": 0.00019854586306747646}
+{"ts": "2025-12-27T01:38:49", "event": "train_log", "step": 1852, "epoch": 0.7814345991561181, "progress_pct": 13.02, "epoch_pct": 13.02, "eta": "49:09:26", "max_grad_norm": 1.0, "loss": 0.8416724801063538, "grad_norm": 0.4486154317855835, "learning_rate": 0.00019853769078754686}
+{"ts": "2025-12-27T01:39:04", "event": "train_log", "step": 1854, "epoch": 0.7822784810126582, "progress_pct": 13.04, "epoch_pct": 13.04, "eta": "49:07:24", "max_grad_norm": 1.0, "loss": 0.8987806439399719, "grad_norm": 0.45148879289627075, "learning_rate": 0.0001985294957768644}
+{"ts": "2025-12-27T01:39:20", "event": "train_log", "step": 1856, "epoch": 0.7831223628691983, "progress_pct": 13.05, "epoch_pct": 13.05, "eta": "49:05:32", "max_grad_norm": 1.0, "loss": 0.7597590088844299, "grad_norm": 0.39330095052719116, "learning_rate": 0.0001985212780373196}
+{"ts": "2025-12-27T01:39:35", "event": "train_log", "step": 1858, "epoch": 0.7839662447257384, "progress_pct": 13.07, "epoch_pct": 13.07, "eta": "49:03:30", "max_grad_norm": 1.0, "loss": 0.8429927229881287, "grad_norm": 0.4260408580303192, "learning_rate": 0.00019851303757080808}
+{"ts": "2025-12-27T01:39:50", "event": "train_log", "step": 1860, "epoch": 0.7848101265822784, "progress_pct": 13.08, "epoch_pct": 13.08, "eta": "49:01:34", "max_grad_norm": 1.0, "loss": 0.7443564534187317, "grad_norm": 0.37519171833992004, "learning_rate": 0.00019850477437923075}
+{"ts": "2025-12-27T01:40:04", "event": "train_log", "step": 1862, "epoch": 0.7856540084388186, "progress_pct": 13.09, "epoch_pct": 13.09, "eta": "48:59:28", "max_grad_norm": 1.0, "loss": 0.934238612651825, "grad_norm": 0.4991084337234497, "learning_rate": 0.00019849648846449383}
+{"ts": "2025-12-27T01:40:19", "event": "train_log", "step": 1864, "epoch": 0.7864978902953587, "progress_pct": 13.11, "epoch_pct": 13.11, "eta": "48:57:27", "max_grad_norm": 1.0, "loss": 0.8938905000686646, "grad_norm": 0.4711335003376007, "learning_rate": 0.00019848817982850867}
+{"ts": "2025-12-27T01:40:34", "event": "train_log", "step": 1866, "epoch": 0.7873417721518987, "progress_pct": 13.12, "epoch_pct": 13.12, "eta": "48:55:31", "max_grad_norm": 1.0, "loss": 0.8745643496513367, "grad_norm": 0.4145370423793793, "learning_rate": 0.00019847984847319193}
+{"ts": "2025-12-27T01:40:49", "event": "train_log", "step": 1868, "epoch": 0.7881856540084389, "progress_pct": 13.14, "epoch_pct": 13.14, "eta": "48:53:36", "max_grad_norm": 1.0, "loss": 0.7178786396980286, "grad_norm": 0.3444044888019562, "learning_rate": 0.00019847149440046548}
+{"ts": "2025-12-27T01:41:04", "event": "train_log", "step": 1870, "epoch": 0.7890295358649789, "progress_pct": 13.15, "epoch_pct": 13.15, "eta": "48:51:38", "max_grad_norm": 1.0, "loss": 0.9114303588867188, "grad_norm": 0.34632962942123413, "learning_rate": 0.00019846311761225646}
+{"ts": "2025-12-27T01:41:20", "event": "train_log", "step": 1872, "epoch": 0.789873417721519, "progress_pct": 13.16, "epoch_pct": 13.16, "eta": "48:49:43", "max_grad_norm": 1.0, "loss": 0.836281955242157, "grad_norm": 0.38630741834640503, "learning_rate": 0.0001984547181104972}
+{"ts": "2025-12-27T01:41:35", "event": "train_log", "step": 1874, "epoch": 0.790717299578059, "progress_pct": 13.18, "epoch_pct": 13.18, "eta": "48:47:46", "max_grad_norm": 1.0, "loss": 0.8247858881950378, "grad_norm": 0.40947434306144714, "learning_rate": 0.00019844629589712534}
+{"ts": "2025-12-27T01:41:49", "event": "train_log", "step": 1876, "epoch": 0.7915611814345992, "progress_pct": 13.19, "epoch_pct": 13.19, "eta": "48:45:48", "max_grad_norm": 1.0, "loss": 0.8446351289749146, "grad_norm": 0.45878684520721436, "learning_rate": 0.00019843785097408372}
+{"ts": "2025-12-27T01:42:05", "event": "train_log", "step": 1878, "epoch": 0.7924050632911392, "progress_pct": 13.21, "epoch_pct": 13.21, "eta": "48:43:54", "max_grad_norm": 1.0, "loss": 0.8567686676979065, "grad_norm": 0.37202852964401245, "learning_rate": 0.0001984293833433204}
+{"ts": "2025-12-27T01:42:21", "event": "train_log", "step": 1880, "epoch": 0.7932489451476793, "progress_pct": 13.22, "epoch_pct": 13.22, "eta": "48:42:03", "max_grad_norm": 1.0, "loss": 0.8134047985076904, "grad_norm": 0.3839069902896881, "learning_rate": 0.00019842089300678876}
+{"ts": "2025-12-27T01:42:35", "event": "train_log", "step": 1882, "epoch": 0.7940928270042195, "progress_pct": 13.23, "epoch_pct": 13.23, "eta": "48:40:05", "max_grad_norm": 1.0, "loss": 0.8670110106468201, "grad_norm": 0.4018687605857849, "learning_rate": 0.00019841237996644727}
+{"ts": "2025-12-27T01:42:51", "event": "train_log", "step": 1884, "epoch": 0.7949367088607595, "progress_pct": 13.25, "epoch_pct": 13.25, "eta": "48:38:11", "max_grad_norm": 1.0, "loss": 0.7975355386734009, "grad_norm": 0.372089684009552, "learning_rate": 0.0001984038442242598}
+{"ts": "2025-12-27T01:43:05", "event": "train_log", "step": 1886, "epoch": 0.7957805907172996, "progress_pct": 13.26, "epoch_pct": 13.26, "eta": "48:36:12", "max_grad_norm": 1.0, "loss": 0.8755611777305603, "grad_norm": 0.4492705166339874, "learning_rate": 0.0001983952857821953}
+{"ts": "2025-12-27T01:43:20", "event": "train_log", "step": 1888, "epoch": 0.7966244725738396, "progress_pct": 13.28, "epoch_pct": 13.28, "eta": "48:34:17", "max_grad_norm": 1.0, "loss": 0.8680241107940674, "grad_norm": 0.451054185628891, "learning_rate": 0.00019838670464222813}
+{"ts": "2025-12-27T01:43:35", "event": "train_log", "step": 1890, "epoch": 0.7974683544303798, "progress_pct": 13.29, "epoch_pct": 13.29, "eta": "48:32:20", "max_grad_norm": 1.0, "loss": 0.7990990877151489, "grad_norm": 0.4036900997161865, "learning_rate": 0.00019837810080633773}
+{"ts": "2025-12-27T01:43:50", "event": "train_log", "step": 1892, "epoch": 0.7983122362869198, "progress_pct": 13.31, "epoch_pct": 13.31, "eta": "48:30:22", "max_grad_norm": 1.0, "loss": 0.8598953485488892, "grad_norm": 0.46125802397727966, "learning_rate": 0.0001983694742765089}
+{"ts": "2025-12-27T01:44:06", "event": "train_log", "step": 1894, "epoch": 0.7991561181434599, "progress_pct": 13.32, "epoch_pct": 13.32, "eta": "48:28:36", "max_grad_norm": 1.0, "loss": 0.7707474827766418, "grad_norm": 0.4458785951137543, "learning_rate": 0.00019836082505473153}
+{"ts": "2025-12-27T01:44:22", "event": "train_log", "step": 1896, "epoch": 0.8, "progress_pct": 13.33, "epoch_pct": 13.33, "eta": "48:26:44", "max_grad_norm": 1.0, "loss": 0.8069534301757812, "grad_norm": 0.44501590728759766, "learning_rate": 0.0001983521531430009}
+{"ts": "2025-12-27T01:44:38", "event": "train_log", "step": 1898, "epoch": 0.8008438818565401, "progress_pct": 13.35, "epoch_pct": 13.35, "eta": "48:24:56", "max_grad_norm": 1.0, "loss": 0.862334668636322, "grad_norm": 0.400771826505661, "learning_rate": 0.0001983434585433174}
+{"ts": "2025-12-27T01:44:52", "event": "train_log", "step": 1900, "epoch": 0.8016877637130801, "progress_pct": 13.36, "epoch_pct": 13.36, "eta": "48:22:56", "max_grad_norm": 1.0, "loss": 0.8125433325767517, "grad_norm": 0.39193347096443176, "learning_rate": 0.00019833474125768676}
+{"ts": "2025-12-27T01:56:20", "event": "train_log", "step": 1900, "epoch": 0.8016877637130801, "progress_pct": 13.36, "epoch_pct": 13.36, "eta": "49:37:21", "max_grad_norm": 1.0, "eval_loss": 0.8546335697174072, "eval_runtime": 688.5301, "eval_samples_per_second": 3.06, "eval_steps_per_second": 3.06}
+{"ts": "2025-12-27T01:56:36", "event": "train_log", "step": 1902, "epoch": 0.8025316455696202, "progress_pct": 13.38, "epoch_pct": 13.38, "eta": "49:35:25", "max_grad_norm": 1.0, "loss": 0.8246012330055237, "grad_norm": 0.441129207611084, "learning_rate": 0.00019832600128811986}
+{"ts": "2025-12-27T01:56:51", "event": "train_log", "step": 1904, "epoch": 0.8033755274261604, "progress_pct": 13.39, "epoch_pct": 13.39, "eta": "49:33:24", "max_grad_norm": 1.0, "loss": 0.8387641310691833, "grad_norm": 0.4089467227458954, "learning_rate": 0.00019831723863663285}
+{"ts": "2025-12-27T01:57:06", "event": "train_log", "step": 1906, "epoch": 0.8042194092827004, "progress_pct": 13.4, "epoch_pct": 13.4, "eta": "49:31:29", "max_grad_norm": 1.0, "loss": 0.7891429662704468, "grad_norm": 0.4871654510498047, "learning_rate": 0.0001983084533052471}
+{"ts": "2025-12-27T01:57:22", "event": "train_log", "step": 1908, "epoch": 0.8050632911392405, "progress_pct": 13.42, "epoch_pct": 13.42, "eta": "49:29:35", "max_grad_norm": 1.0, "loss": 0.8008774518966675, "grad_norm": 0.40085604786872864, "learning_rate": 0.00019829964529598923}
+{"ts": "2025-12-27T01:57:38", "event": "train_log", "step": 1910, "epoch": 0.8059071729957806, "progress_pct": 13.43, "epoch_pct": 13.43, "eta": "49:27:39", "max_grad_norm": 1.0, "loss": 0.7921834588050842, "grad_norm": 0.38775014877319336, "learning_rate": 0.00019829081461089109}
+{"ts": "2025-12-27T01:57:55", "event": "train_log", "step": 1912, "epoch": 0.8067510548523207, "progress_pct": 13.45, "epoch_pct": 13.45, "eta": "49:25:57", "max_grad_norm": 1.0, "loss": 0.7752519845962524, "grad_norm": 0.37830594182014465, "learning_rate": 0.0001982819612519897}
+{"ts": "2025-12-27T01:58:11", "event": "train_log", "step": 1914, "epoch": 0.8075949367088607, "progress_pct": 13.46, "epoch_pct": 13.46, "eta": "49:24:03", "max_grad_norm": 1.0, "loss": 0.7958255410194397, "grad_norm": 0.41137194633483887, "learning_rate": 0.0001982730852213274}
+{"ts": "2025-12-27T01:58:26", "event": "train_log", "step": 1916, "epoch": 0.8084388185654009, "progress_pct": 13.47, "epoch_pct": 13.47, "eta": "49:22:03", "max_grad_norm": 1.0, "loss": 0.8177281618118286, "grad_norm": 0.4871830940246582, "learning_rate": 0.0001982641865209517}
+{"ts": "2025-12-27T01:58:43", "event": "train_log", "step": 1918, "epoch": 0.809282700421941, "progress_pct": 13.49, "epoch_pct": 13.49, "eta": "49:20:18", "max_grad_norm": 1.0, "loss": 0.7726616859436035, "grad_norm": 0.4157388210296631, "learning_rate": 0.0001982552651529154}
+{"ts": "2025-12-27T01:59:00", "event": "train_log", "step": 1920, "epoch": 0.810126582278481, "progress_pct": 13.5, "epoch_pct": 13.5, "eta": "49:18:35", "max_grad_norm": 1.0, "loss": 0.6601400375366211, "grad_norm": 0.32338014245033264, "learning_rate": 0.00019824632111927645}
+{"ts": "2025-12-27T01:59:15", "event": "train_log", "step": 1922, "epoch": 0.810970464135021, "progress_pct": 13.52, "epoch_pct": 13.52, "eta": "49:16:36", "max_grad_norm": 1.0, "loss": 0.7866622805595398, "grad_norm": 0.4642949104309082, "learning_rate": 0.00019823735442209804}
+{"ts": "2025-12-27T01:59:32", "event": "train_log", "step": 1924, "epoch": 0.8118143459915612, "progress_pct": 13.53, "epoch_pct": 13.53, "eta": "49:14:52", "max_grad_norm": 1.0, "loss": 0.7857986092567444, "grad_norm": 0.43558797240257263, "learning_rate": 0.00019822836506344865}
+{"ts": "2025-12-27T01:59:47", "event": "train_log", "step": 1926, "epoch": 0.8126582278481013, "progress_pct": 13.54, "epoch_pct": 13.54, "eta": "49:12:59", "max_grad_norm": 1.0, "loss": 0.8407763838768005, "grad_norm": 0.3996953070163727, "learning_rate": 0.000198219353045402}
+{"ts": "2025-12-27T02:00:02", "event": "train_log", "step": 1928, "epoch": 0.8135021097046413, "progress_pct": 13.56, "epoch_pct": 13.56, "eta": "49:11:00", "max_grad_norm": 1.0, "loss": 0.8433752655982971, "grad_norm": 0.4415414333343506, "learning_rate": 0.00019821031837003686}
+{"ts": "2025-12-27T02:00:19", "event": "train_log", "step": 1930, "epoch": 0.8143459915611815, "progress_pct": 13.57, "epoch_pct": 13.57, "eta": "49:09:13", "max_grad_norm": 1.0, "loss": 0.6941158175468445, "grad_norm": 0.37329310178756714, "learning_rate": 0.00019820126103943752}
+{"ts": "2025-12-27T02:00:33", "event": "train_log", "step": 1932, "epoch": 0.8151898734177215, "progress_pct": 13.59, "epoch_pct": 13.59, "eta": "49:07:15", "max_grad_norm": 1.0, "loss": 0.775604248046875, "grad_norm": 0.4482601284980774, "learning_rate": 0.00019819218105569323}
+{"ts": "2025-12-27T02:00:49", "event": "train_log", "step": 1934, "epoch": 0.8160337552742616, "progress_pct": 13.6, "epoch_pct": 13.6, "eta": "49:05:22", "max_grad_norm": 1.0, "loss": 0.8496418595314026, "grad_norm": 0.426194429397583, "learning_rate": 0.00019818307842089854}
+{"ts": "2025-12-27T02:01:04", "event": "train_log", "step": 1936, "epoch": 0.8168776371308016, "progress_pct": 13.61, "epoch_pct": 13.61, "eta": "49:03:26", "max_grad_norm": 1.0, "loss": 0.7767958045005798, "grad_norm": 0.42379093170166016, "learning_rate": 0.00019817395313715335}
+{"ts": "2025-12-27T02:01:19", "event": "train_log", "step": 1938, "epoch": 0.8177215189873418, "progress_pct": 13.63, "epoch_pct": 13.63, "eta": "49:01:32", "max_grad_norm": 1.0, "loss": 0.7347666025161743, "grad_norm": 0.4149782061576843, "learning_rate": 0.0001981648052065626}
+{"ts": "2025-12-27T02:01:34", "event": "train_log", "step": 1940, "epoch": 0.8185654008438819, "progress_pct": 13.64, "epoch_pct": 13.64, "eta": "48:59:35", "max_grad_norm": 1.0, "loss": 0.7839647531509399, "grad_norm": 0.3927431106567383, "learning_rate": 0.0001981556346312366}
+{"ts": "2025-12-27T02:01:49", "event": "train_log", "step": 1942, "epoch": 0.8194092827004219, "progress_pct": 13.66, "epoch_pct": 13.66, "eta": "48:57:38", "max_grad_norm": 1.0, "loss": 0.8870531916618347, "grad_norm": 0.4483136236667633, "learning_rate": 0.00019814644141329083}
+{"ts": "2025-12-27T02:02:03", "event": "train_log", "step": 1944, "epoch": 0.8202531645569621, "progress_pct": 13.67, "epoch_pct": 13.67, "eta": "48:55:36", "max_grad_norm": 1.0, "loss": 0.8646620512008667, "grad_norm": 0.4400147497653961, "learning_rate": 0.00019813722555484594}
+{"ts": "2025-12-27T02:02:17", "event": "train_log", "step": 1946, "epoch": 0.8210970464135021, "progress_pct": 13.68, "epoch_pct": 13.68, "eta": "48:53:38", "max_grad_norm": 1.0, "loss": 0.9576541781425476, "grad_norm": 0.469971626996994, "learning_rate": 0.00019812798705802785}
+{"ts": "2025-12-27T02:02:32", "event": "train_log", "step": 1948, "epoch": 0.8219409282700422, "progress_pct": 13.7, "epoch_pct": 13.7, "eta": "48:51:44", "max_grad_norm": 1.0, "loss": 0.7956477403640747, "grad_norm": 0.388113409280777, "learning_rate": 0.00019811872592496776}
+{"ts": "2025-12-27T02:02:48", "event": "train_log", "step": 1950, "epoch": 0.8227848101265823, "progress_pct": 13.71, "epoch_pct": 13.71, "eta": "48:49:55", "max_grad_norm": 1.0, "loss": 0.8440352082252502, "grad_norm": 0.4271424114704132, "learning_rate": 0.00019810944215780201}
+{"ts": "2025-12-27T02:03:03", "event": "train_log", "step": 1952, "epoch": 0.8236286919831224, "progress_pct": 13.73, "epoch_pct": 13.73, "eta": "48:47:58", "max_grad_norm": 1.0, "loss": 0.8589065074920654, "grad_norm": 0.4097966253757477, "learning_rate": 0.00019810013575867217}
+{"ts": "2025-12-27T02:03:17", "event": "train_log", "step": 1954, "epoch": 0.8244725738396624, "progress_pct": 13.74, "epoch_pct": 13.74, "eta": "48:46:00", "max_grad_norm": 1.0, "loss": 0.8606626987457275, "grad_norm": 0.41550710797309875, "learning_rate": 0.00019809080672972504}
+{"ts": "2025-12-27T02:03:33", "event": "train_log", "step": 1956, "epoch": 0.8253164556962025, "progress_pct": 13.76, "epoch_pct": 13.76, "eta": "48:44:08", "max_grad_norm": 1.0, "loss": 0.7582436800003052, "grad_norm": 0.3680916130542755, "learning_rate": 0.0001980814550731127}
+{"ts": "2025-12-27T02:03:48", "event": "train_log", "step": 1958, "epoch": 0.8261603375527427, "progress_pct": 13.77, "epoch_pct": 13.77, "eta": "48:42:14", "max_grad_norm": 1.0, "loss": 0.8345186710357666, "grad_norm": 0.573297917842865, "learning_rate": 0.00019807208079099234}
+{"ts": "2025-12-27T02:04:02", "event": "train_log", "step": 1960, "epoch": 0.8270042194092827, "progress_pct": 13.78, "epoch_pct": 13.78, "eta": "48:40:16", "max_grad_norm": 1.0, "loss": 0.889683723449707, "grad_norm": 0.43631303310394287, "learning_rate": 0.00019806268388552646}
+{"ts": "2025-12-27T02:04:17", "event": "train_log", "step": 1962, "epoch": 0.8278481012658228, "progress_pct": 13.8, "epoch_pct": 13.8, "eta": "48:38:21", "max_grad_norm": 1.0, "loss": 0.8092973232269287, "grad_norm": 0.442531019449234, "learning_rate": 0.00019805326435888275}
+{"ts": "2025-12-27T02:04:31", "event": "train_log", "step": 1964, "epoch": 0.8286919831223629, "progress_pct": 13.81, "epoch_pct": 13.81, "eta": "48:36:24", "max_grad_norm": 1.0, "loss": 0.8919535279273987, "grad_norm": 0.45020824670791626, "learning_rate": 0.0001980438222132341}
+{"ts": "2025-12-27T02:04:46", "event": "train_log", "step": 1966, "epoch": 0.829535864978903, "progress_pct": 13.83, "epoch_pct": 13.83, "eta": "48:34:29", "max_grad_norm": 1.0, "loss": 0.7652381658554077, "grad_norm": 0.43131786584854126, "learning_rate": 0.0001980343574507586}
+{"ts": "2025-12-27T02:05:01", "event": "train_log", "step": 1968, "epoch": 0.830379746835443, "progress_pct": 13.84, "epoch_pct": 13.84, "eta": "48:32:37", "max_grad_norm": 1.0, "loss": 0.8512389659881592, "grad_norm": 0.4461326599121094, "learning_rate": 0.00019802487007363967}
+{"ts": "2025-12-27T02:05:16", "event": "train_log", "step": 1970, "epoch": 0.8312236286919831, "progress_pct": 13.85, "epoch_pct": 13.85, "eta": "48:30:45", "max_grad_norm": 1.0, "loss": 0.8681327700614929, "grad_norm": 0.4676894247531891, "learning_rate": 0.0001980153600840658}
+{"ts": "2025-12-27T02:05:32", "event": "train_log", "step": 1972, "epoch": 0.8320675105485232, "progress_pct": 13.87, "epoch_pct": 13.87, "eta": "48:28:56", "max_grad_norm": 1.0, "loss": 0.7476955652236938, "grad_norm": 0.38376888632774353, "learning_rate": 0.00019800582748423075}
+{"ts": "2025-12-27T02:05:47", "event": "train_log", "step": 1974, "epoch": 0.8329113924050633, "progress_pct": 13.88, "epoch_pct": 13.88, "eta": "48:27:05", "max_grad_norm": 1.0, "loss": 0.814192533493042, "grad_norm": 0.37173229455947876, "learning_rate": 0.00019799627227633354}
+{"ts": "2025-12-27T02:06:02", "event": "train_log", "step": 1976, "epoch": 0.8337552742616033, "progress_pct": 13.9, "epoch_pct": 13.9, "eta": "48:25:13", "max_grad_norm": 1.0, "loss": 0.7668994069099426, "grad_norm": 0.4345237910747528, "learning_rate": 0.00019798669446257844}
+{"ts": "2025-12-27T02:06:17", "event": "train_log", "step": 1978, "epoch": 0.8345991561181435, "progress_pct": 13.91, "epoch_pct": 13.91, "eta": "48:23:21", "max_grad_norm": 1.0, "loss": 0.8321775197982788, "grad_norm": 0.45872583985328674, "learning_rate": 0.0001979770940451747}
+{"ts": "2025-12-27T02:06:31", "event": "train_log", "step": 1980, "epoch": 0.8354430379746836, "progress_pct": 13.92, "epoch_pct": 13.92, "eta": "48:21:27", "max_grad_norm": 1.0, "loss": 0.7824342250823975, "grad_norm": 0.40618205070495605, "learning_rate": 0.0001979674710263371}
+{"ts": "2025-12-27T02:06:47", "event": "train_log", "step": 1982, "epoch": 0.8362869198312236, "progress_pct": 13.94, "epoch_pct": 13.94, "eta": "48:19:37", "max_grad_norm": 1.0, "loss": 0.8752480745315552, "grad_norm": 0.38377392292022705, "learning_rate": 0.00019795782540828544}
+{"ts": "2025-12-27T02:07:02", "event": "train_log", "step": 1984, "epoch": 0.8371308016877637, "progress_pct": 13.95, "epoch_pct": 13.95, "eta": "48:17:50", "max_grad_norm": 1.0, "loss": 0.66700679063797, "grad_norm": 0.36340072751045227, "learning_rate": 0.00019794815719324476}
+{"ts": "2025-12-27T02:07:17", "event": "train_log", "step": 1986, "epoch": 0.8379746835443038, "progress_pct": 13.97, "epoch_pct": 13.97, "eta": "48:15:55", "max_grad_norm": 1.0, "loss": 0.8582209944725037, "grad_norm": 0.4638384282588959, "learning_rate": 0.00019793846638344536}
+{"ts": "2025-12-27T02:07:32", "event": "train_log", "step": 1988, "epoch": 0.8388185654008439, "progress_pct": 13.98, "epoch_pct": 13.98, "eta": "48:14:04", "max_grad_norm": 1.0, "loss": 0.7724968791007996, "grad_norm": 0.38364699482917786, "learning_rate": 0.00019792875298112268}
+{"ts": "2025-12-27T02:07:47", "event": "train_log", "step": 1990, "epoch": 0.8396624472573839, "progress_pct": 13.99, "epoch_pct": 13.99, "eta": "48:12:13", "max_grad_norm": 1.0, "loss": 0.8542404174804688, "grad_norm": 0.439730703830719, "learning_rate": 0.00019791901698851749}
+{"ts": "2025-12-27T02:08:01", "event": "train_log", "step": 1992, "epoch": 0.8405063291139241, "progress_pct": 14.01, "epoch_pct": 14.01, "eta": "48:10:21", "max_grad_norm": 1.0, "loss": 0.8424296975135803, "grad_norm": 0.4622722566127777, "learning_rate": 0.00019790925840787563}
+{"ts": "2025-12-27T02:08:16", "event": "train_log", "step": 1994, "epoch": 0.8413502109704641, "progress_pct": 14.02, "epoch_pct": 14.02, "eta": "48:08:29", "max_grad_norm": 1.0, "loss": 0.8400710225105286, "grad_norm": 0.4491989314556122, "learning_rate": 0.0001978994772414482}
+{"ts": "2025-12-27T02:08:31", "event": "train_log", "step": 1996, "epoch": 0.8421940928270042, "progress_pct": 14.04, "epoch_pct": 14.04, "eta": "48:06:41", "max_grad_norm": 1.0, "loss": 0.7265452146530151, "grad_norm": 0.372295081615448, "learning_rate": 0.0001978896734914916}
+{"ts": "2025-12-27T02:08:46", "event": "train_log", "step": 1998, "epoch": 0.8430379746835444, "progress_pct": 14.05, "epoch_pct": 14.05, "eta": "48:04:49", "max_grad_norm": 1.0, "loss": 0.8458228707313538, "grad_norm": 0.4328666925430298, "learning_rate": 0.00019787984716026732}
+{"ts": "2025-12-27T02:09:01", "event": "train_log", "step": 2000, "epoch": 0.8438818565400844, "progress_pct": 14.06, "epoch_pct": 14.06, "eta": "48:02:57", "max_grad_norm": 1.0, "loss": 0.8517491817474365, "grad_norm": 0.42246317863464355, "learning_rate": 0.00019786999825004216}
+{"ts": "2025-12-27T02:20:26", "event": "train_log", "step": 2000, "epoch": 0.8438818565400844, "progress_pct": 14.06, "epoch_pct": 14.06, "eta": "49:12:44", "max_grad_norm": 1.0, "eval_loss": 0.8460908532142639, "eval_runtime": 685.2518, "eval_samples_per_second": 3.075, "eval_steps_per_second": 3.075}
+{"ts": "2025-12-27T02:20:44", "event": "train_log", "step": 2002, "epoch": 0.8447257383966245, "progress_pct": 14.08, "epoch_pct": 14.08, "eta": "49:11:05", "max_grad_norm": 1.0, "loss": 0.8045481443405151, "grad_norm": 0.3927323818206787, "learning_rate": 0.00019786012676308802}
+{"ts": "2025-12-27T02:20:59", "event": "train_log", "step": 2004, "epoch": 0.8455696202531645, "progress_pct": 14.09, "epoch_pct": 14.09, "eta": "49:09:17", "max_grad_norm": 1.0, "loss": 0.7509838938713074, "grad_norm": 0.38189369440078735, "learning_rate": 0.00019785023270168202}
+{"ts": "2025-12-27T02:21:15", "event": "train_log", "step": 2006, "epoch": 0.8464135021097047, "progress_pct": 14.11, "epoch_pct": 14.11, "eta": "49:07:27", "max_grad_norm": 1.0, "loss": 0.7730163335800171, "grad_norm": 0.40383535623550415, "learning_rate": 0.00019784031606810663}
+{"ts": "2025-12-27T02:21:31", "event": "train_log", "step": 2008, "epoch": 0.8472573839662447, "progress_pct": 14.12, "epoch_pct": 14.12, "eta": "49:05:37", "max_grad_norm": 1.0, "loss": 0.7096828818321228, "grad_norm": 0.3665419816970825, "learning_rate": 0.00019783037686464935}
+{"ts": "2025-12-27T02:21:47", "event": "train_log", "step": 2010, "epoch": 0.8481012658227848, "progress_pct": 14.14, "epoch_pct": 14.14, "eta": "49:03:50", "max_grad_norm": 1.0, "loss": 0.8186401128768921, "grad_norm": 0.36623379588127136, "learning_rate": 0.00019782041509360302}
+{"ts": "2025-12-27T02:22:03", "event": "train_log", "step": 2012, "epoch": 0.8489451476793249, "progress_pct": 14.15, "epoch_pct": 14.15, "eta": "49:02:01", "max_grad_norm": 1.0, "loss": 0.8502879738807678, "grad_norm": 0.477255642414093, "learning_rate": 0.0001978104307572656}
+{"ts": "2025-12-27T02:22:19", "event": "train_log", "step": 2014, "epoch": 0.849789029535865, "progress_pct": 14.16, "epoch_pct": 14.16, "eta": "49:00:14", "max_grad_norm": 1.0, "loss": 0.9010884761810303, "grad_norm": 0.44456636905670166, "learning_rate": 0.00019780042385794032}
+{"ts": "2025-12-27T02:22:34", "event": "train_log", "step": 2016, "epoch": 0.850632911392405, "progress_pct": 14.18, "epoch_pct": 14.18, "eta": "48:58:23", "max_grad_norm": 1.0, "loss": 0.774183452129364, "grad_norm": 0.42378175258636475, "learning_rate": 0.00019779039439793552}
+{"ts": "2025-12-27T02:22:49", "event": "train_log", "step": 2018, "epoch": 0.8514767932489451, "progress_pct": 14.19, "epoch_pct": 14.19, "eta": "48:56:32", "max_grad_norm": 1.0, "loss": 0.8487694263458252, "grad_norm": 0.43021997809410095, "learning_rate": 0.00019778034237956482}
+{"ts": "2025-12-27T02:23:06", "event": "train_log", "step": 2020, "epoch": 0.8523206751054853, "progress_pct": 14.21, "epoch_pct": 14.21, "eta": "48:54:47", "max_grad_norm": 1.0, "loss": 0.8004868030548096, "grad_norm": 0.38188812136650085, "learning_rate": 0.00019777026780514704}
+{"ts": "2025-12-27T02:23:21", "event": "train_log", "step": 2022, "epoch": 0.8531645569620253, "progress_pct": 14.22, "epoch_pct": 14.22, "eta": "48:52:57", "max_grad_norm": 1.0, "loss": 0.7680953145027161, "grad_norm": 0.3684130609035492, "learning_rate": 0.0001977601706770062}
+{"ts": "2025-12-27T02:23:37", "event": "train_log", "step": 2024, "epoch": 0.8540084388185654, "progress_pct": 14.23, "epoch_pct": 14.23, "eta": "48:51:11", "max_grad_norm": 1.0, "loss": 0.8990264534950256, "grad_norm": 0.444844514131546, "learning_rate": 0.00019775005099747153}
+{"ts": "2025-12-27T02:23:54", "event": "train_log", "step": 2026, "epoch": 0.8548523206751055, "progress_pct": 14.25, "epoch_pct": 14.25, "eta": "48:49:31", "max_grad_norm": 1.0, "loss": 0.7676048874855042, "grad_norm": 0.3432365655899048, "learning_rate": 0.00019773990876887737}
+{"ts": "2025-12-27T02:24:10", "event": "train_log", "step": 2028, "epoch": 0.8556962025316456, "progress_pct": 14.26, "epoch_pct": 14.26, "eta": "48:47:41", "max_grad_norm": 1.0, "loss": 0.8546769618988037, "grad_norm": 0.3728618025779724, "learning_rate": 0.00019772974399356343}
+{"ts": "2025-12-27T02:24:26", "event": "train_log", "step": 2030, "epoch": 0.8565400843881856, "progress_pct": 14.28, "epoch_pct": 14.28, "eta": "48:46:01", "max_grad_norm": 1.0, "loss": 0.8009977340698242, "grad_norm": 0.42767980694770813, "learning_rate": 0.00019771955667387449}
+{"ts": "2025-12-27T02:24:42", "event": "train_log", "step": 2032, "epoch": 0.8573839662447258, "progress_pct": 14.29, "epoch_pct": 14.29, "eta": "48:44:15", "max_grad_norm": 1.0, "loss": 0.8424299955368042, "grad_norm": 0.4125710129737854, "learning_rate": 0.00019770934681216055}
+{"ts": "2025-12-27T02:24:58", "event": "train_log", "step": 2034, "epoch": 0.8582278481012658, "progress_pct": 14.3, "epoch_pct": 14.3, "eta": "48:42:29", "max_grad_norm": 1.0, "loss": 0.7940502762794495, "grad_norm": 0.4399266839027405, "learning_rate": 0.00019769911441077685}
+{"ts": "2025-12-27T02:25:15", "event": "train_log", "step": 2036, "epoch": 0.8590717299578059, "progress_pct": 14.32, "epoch_pct": 14.32, "eta": "48:40:49", "max_grad_norm": 1.0, "loss": 0.9424314498901367, "grad_norm": 0.41797709465026855, "learning_rate": 0.0001976888594720838}
+{"ts": "2025-12-27T02:25:31", "event": "train_log", "step": 2038, "epoch": 0.859915611814346, "progress_pct": 14.33, "epoch_pct": 14.33, "eta": "48:39:02", "max_grad_norm": 1.0, "loss": 0.9102896451950073, "grad_norm": 0.5245009660720825, "learning_rate": 0.00019767858199844698}
+{"ts": "2025-12-27T02:25:48", "event": "train_log", "step": 2040, "epoch": 0.8607594936708861, "progress_pct": 14.35, "epoch_pct": 14.35, "eta": "48:37:22", "max_grad_norm": 1.0, "loss": 0.7434720396995544, "grad_norm": 0.4092593193054199, "learning_rate": 0.00019766828199223728}
+{"ts": "2025-12-27T02:26:03", "event": "train_log", "step": 2042, "epoch": 0.8616033755274262, "progress_pct": 14.36, "epoch_pct": 14.36, "eta": "48:35:34", "max_grad_norm": 1.0, "loss": 0.7441987991333008, "grad_norm": 0.40787473320961, "learning_rate": 0.00019765795945583068}
+{"ts": "2025-12-27T02:26:18", "event": "train_log", "step": 2044, "epoch": 0.8624472573839662, "progress_pct": 14.37, "epoch_pct": 14.37, "eta": "48:33:46", "max_grad_norm": 1.0, "loss": 0.8317872881889343, "grad_norm": 0.42438021302223206, "learning_rate": 0.0001976476143916084}
+{"ts": "2025-12-27T02:26:34", "event": "train_log", "step": 2046, "epoch": 0.8632911392405064, "progress_pct": 14.39, "epoch_pct": 14.39, "eta": "48:32:01", "max_grad_norm": 1.0, "loss": 0.764509379863739, "grad_norm": 0.38261914253234863, "learning_rate": 0.00019763724680195682}
+{"ts": "2025-12-27T02:26:50", "event": "train_log", "step": 2048, "epoch": 0.8641350210970464, "progress_pct": 14.4, "epoch_pct": 14.4, "eta": "48:30:16", "max_grad_norm": 1.0, "loss": 0.8666167855262756, "grad_norm": 0.43295741081237793, "learning_rate": 0.00019762685668926758}
+{"ts": "2025-12-27T02:27:06", "event": "train_log", "step": 2050, "epoch": 0.8649789029535865, "progress_pct": 14.42, "epoch_pct": 14.42, "eta": "48:28:32", "max_grad_norm": 1.0, "loss": 0.7606754302978516, "grad_norm": 0.36384403705596924, "learning_rate": 0.00019761644405593743}
+{"ts": "2025-12-27T02:27:21", "event": "train_log", "step": 2052, "epoch": 0.8658227848101265, "progress_pct": 14.43, "epoch_pct": 14.43, "eta": "48:26:41", "max_grad_norm": 1.0, "loss": 0.827495813369751, "grad_norm": 0.4296814501285553, "learning_rate": 0.0001976060089043684}
+{"ts": "2025-12-27T02:27:37", "event": "train_log", "step": 2054, "epoch": 0.8666666666666667, "progress_pct": 14.44, "epoch_pct": 14.44, "eta": "48:24:55", "max_grad_norm": 1.0, "loss": 0.7418577075004578, "grad_norm": 0.3831455111503601, "learning_rate": 0.0001975955512369677}
+{"ts": "2025-12-27T02:27:52", "event": "train_log", "step": 2056, "epoch": 0.8675105485232067, "progress_pct": 14.46, "epoch_pct": 14.46, "eta": "48:23:10", "max_grad_norm": 1.0, "loss": 0.8603045344352722, "grad_norm": 0.44710344076156616, "learning_rate": 0.0001975850710561477}
+{"ts": "2025-12-27T02:28:08", "event": "train_log", "step": 2058, "epoch": 0.8683544303797468, "progress_pct": 14.47, "epoch_pct": 14.47, "eta": "48:21:25", "max_grad_norm": 1.0, "loss": 0.900658905506134, "grad_norm": 0.44012558460235596, "learning_rate": 0.00019757456836432593}
+{"ts": "2025-12-27T02:28:24", "event": "train_log", "step": 2060, "epoch": 0.869198312236287, "progress_pct": 14.49, "epoch_pct": 14.49, "eta": "48:19:39", "max_grad_norm": 1.0, "loss": 0.8416497707366943, "grad_norm": 0.42166298627853394, "learning_rate": 0.0001975640431639252}
+{"ts": "2025-12-27T02:28:39", "event": "train_log", "step": 2062, "epoch": 0.870042194092827, "progress_pct": 14.5, "epoch_pct": 14.5, "eta": "48:17:50", "max_grad_norm": 1.0, "loss": 0.8281066417694092, "grad_norm": 0.40824490785598755, "learning_rate": 0.00019755349545737345}
+{"ts": "2025-12-27T02:28:54", "event": "train_log", "step": 2064, "epoch": 0.8708860759493671, "progress_pct": 14.51, "epoch_pct": 14.51, "eta": "48:16:01", "max_grad_norm": 1.0, "loss": 0.83869469165802, "grad_norm": 0.4272223711013794, "learning_rate": 0.00019754292524710388}
+{"ts": "2025-12-27T02:29:09", "event": "train_log", "step": 2066, "epoch": 0.8717299578059071, "progress_pct": 14.53, "epoch_pct": 14.53, "eta": "48:14:16", "max_grad_norm": 1.0, "loss": 0.8337644934654236, "grad_norm": 0.3732798397541046, "learning_rate": 0.00019753233253555478}
+{"ts": "2025-12-27T02:29:25", "event": "train_log", "step": 2068, "epoch": 0.8725738396624473, "progress_pct": 14.54, "epoch_pct": 14.54, "eta": "48:12:32", "max_grad_norm": 1.0, "loss": 0.7677452564239502, "grad_norm": 0.4048069417476654, "learning_rate": 0.00019752171732516973}
+{"ts": "2025-12-27T02:29:41", "event": "train_log", "step": 2070, "epoch": 0.8734177215189873, "progress_pct": 14.56, "epoch_pct": 14.56, "eta": "48:10:48", "max_grad_norm": 1.0, "loss": 0.8320774435997009, "grad_norm": 0.43628430366516113, "learning_rate": 0.00019751107961839742}
+{"ts": "2025-12-27T02:29:57", "event": "train_log", "step": 2072, "epoch": 0.8742616033755274, "progress_pct": 14.57, "epoch_pct": 14.57, "eta": "48:09:06", "max_grad_norm": 1.0, "loss": 0.8480262160301208, "grad_norm": 0.43674105405807495, "learning_rate": 0.00019750041941769174}
+{"ts": "2025-12-27T02:30:13", "event": "train_log", "step": 2074, "epoch": 0.8751054852320675, "progress_pct": 14.59, "epoch_pct": 14.59, "eta": "48:07:23", "max_grad_norm": 1.0, "loss": 0.7598503828048706, "grad_norm": 0.3678584098815918, "learning_rate": 0.0001974897367255119}
+{"ts": "2025-12-27T02:30:29", "event": "train_log", "step": 2076, "epoch": 0.8759493670886076, "progress_pct": 14.6, "epoch_pct": 14.6, "eta": "48:05:47", "max_grad_norm": 1.0, "loss": 0.7907084822654724, "grad_norm": 0.4316411316394806, "learning_rate": 0.0001974790315443221}
+{"ts": "2025-12-27T02:30:44", "event": "train_log", "step": 2078, "epoch": 0.8767932489451477, "progress_pct": 14.61, "epoch_pct": 14.61, "eta": "48:03:58", "max_grad_norm": 1.0, "loss": 0.8287727236747742, "grad_norm": 0.426465779542923, "learning_rate": 0.00019746830387659182}
+{"ts": "2025-12-27T02:30:59", "event": "train_log", "step": 2080, "epoch": 0.8776371308016878, "progress_pct": 14.63, "epoch_pct": 14.63, "eta": "48:02:08", "max_grad_norm": 1.0, "loss": 0.8464577794075012, "grad_norm": 0.447337806224823, "learning_rate": 0.00019745755372479578}
+{"ts": "2025-12-27T02:31:14", "event": "train_log", "step": 2082, "epoch": 0.8784810126582279, "progress_pct": 14.64, "epoch_pct": 14.64, "eta": "48:00:21", "max_grad_norm": 1.0, "loss": 0.7383086681365967, "grad_norm": 0.37877580523490906, "learning_rate": 0.0001974467810914138}
+{"ts": "2025-12-27T02:31:28", "event": "train_log", "step": 2084, "epoch": 0.8793248945147679, "progress_pct": 14.66, "epoch_pct": 14.66, "eta": "47:58:32", "max_grad_norm": 1.0, "loss": 0.729221522808075, "grad_norm": 0.3914664089679718, "learning_rate": 0.0001974359859789309}
+{"ts": "2025-12-27T02:31:44", "event": "train_log", "step": 2086, "epoch": 0.880168776371308, "progress_pct": 14.67, "epoch_pct": 14.67, "eta": "47:56:50", "max_grad_norm": 1.0, "loss": 0.692332923412323, "grad_norm": 0.39767366647720337, "learning_rate": 0.00019742516838983733}
+{"ts": "2025-12-27T02:31:59", "event": "train_log", "step": 2088, "epoch": 0.8810126582278481, "progress_pct": 14.68, "epoch_pct": 14.68, "eta": "47:55:00", "max_grad_norm": 1.0, "loss": 0.8540068864822388, "grad_norm": 0.4396754205226898, "learning_rate": 0.0001974143283266285}
+{"ts": "2025-12-27T02:32:13", "event": "train_log", "step": 2090, "epoch": 0.8818565400843882, "progress_pct": 14.7, "epoch_pct": 14.7, "eta": "47:53:10", "max_grad_norm": 1.0, "loss": 0.8570998907089233, "grad_norm": 0.4968172609806061, "learning_rate": 0.00019740346579180502}
+{"ts": "2025-12-27T02:32:29", "event": "train_log", "step": 2092, "epoch": 0.8827004219409282, "progress_pct": 14.71, "epoch_pct": 14.71, "eta": "47:51:29", "max_grad_norm": 1.0, "loss": 0.784139096736908, "grad_norm": 0.41686809062957764, "learning_rate": 0.00019739258078787264}
+{"ts": "2025-12-27T02:32:44", "event": "train_log", "step": 2094, "epoch": 0.8835443037974684, "progress_pct": 14.73, "epoch_pct": 14.73, "eta": "47:49:44", "max_grad_norm": 1.0, "loss": 0.8521397113800049, "grad_norm": 0.38788047432899475, "learning_rate": 0.0001973816733173423}
+{"ts": "2025-12-27T02:32:59", "event": "train_log", "step": 2096, "epoch": 0.8843881856540085, "progress_pct": 14.74, "epoch_pct": 14.74, "eta": "47:47:56", "max_grad_norm": 1.0, "loss": 0.8457935452461243, "grad_norm": 0.38369178771972656, "learning_rate": 0.00019737074338273023}
+{"ts": "2025-12-27T02:33:15", "event": "train_log", "step": 2098, "epoch": 0.8852320675105485, "progress_pct": 14.75, "epoch_pct": 14.75, "eta": "47:46:15", "max_grad_norm": 1.0, "loss": 0.7425959706306458, "grad_norm": 0.38869398832321167, "learning_rate": 0.0001973597909865577}
+{"ts": "2025-12-27T02:33:30", "event": "train_log", "step": 2100, "epoch": 0.8860759493670886, "progress_pct": 14.77, "epoch_pct": 14.77, "eta": "47:44:34", "max_grad_norm": 1.0, "loss": 0.7742393016815186, "grad_norm": 0.4067917466163635, "learning_rate": 0.00019734881613135122}
+{"ts": "2025-12-27T02:44:39", "event": "train_log", "step": 2100, "epoch": 0.8860759493670886, "progress_pct": 14.77, "epoch_pct": 14.77, "eta": "48:48:55", "max_grad_norm": 1.0, "eval_loss": 0.8401098847389221, "eval_runtime": 669.1149, "eval_samples_per_second": 3.149, "eval_steps_per_second": 3.149}
+{"ts": "2025-12-27T02:44:54", "event": "train_log", "step": 2102, "epoch": 0.8869198312236287, "progress_pct": 14.78, "epoch_pct": 14.78, "eta": "48:47:04", "max_grad_norm": 1.0, "loss": 0.9099814295768738, "grad_norm": 0.4282098412513733, "learning_rate": 0.00019733781881964242}
+{"ts": "2025-12-27T02:45:09", "event": "train_log", "step": 2104, "epoch": 0.8877637130801688, "progress_pct": 14.8, "epoch_pct": 14.8, "eta": "48:45:15", "max_grad_norm": 1.0, "loss": 0.7458856105804443, "grad_norm": 0.33882173895835876, "learning_rate": 0.00019732679905396826}
+{"ts": "2025-12-27T02:45:26", "event": "train_log", "step": 2106, "epoch": 0.8886075949367088, "progress_pct": 14.81, "epoch_pct": 14.81, "eta": "48:43:37", "max_grad_norm": 1.0, "loss": 0.7829693555831909, "grad_norm": 0.43959516286849976, "learning_rate": 0.00019731575683687078}
+{"ts": "2025-12-27T02:45:41", "event": "train_log", "step": 2108, "epoch": 0.889451476793249, "progress_pct": 14.82, "epoch_pct": 14.82, "eta": "48:41:47", "max_grad_norm": 1.0, "loss": 0.7898997068405151, "grad_norm": 0.40046483278274536, "learning_rate": 0.00019730469217089716}
+{"ts": "2025-12-27T02:45:57", "event": "train_log", "step": 2110, "epoch": 0.890295358649789, "progress_pct": 14.84, "epoch_pct": 14.84, "eta": "48:40:05", "max_grad_norm": 1.0, "loss": 0.7874186635017395, "grad_norm": 0.40572553873062134, "learning_rate": 0.00019729360505859984}
+{"ts": "2025-12-27T02:46:13", "event": "train_log", "step": 2112, "epoch": 0.8911392405063291, "progress_pct": 14.85, "epoch_pct": 14.85, "eta": "48:38:23", "max_grad_norm": 1.0, "loss": 0.8499898910522461, "grad_norm": 0.42086347937583923, "learning_rate": 0.0001972824955025364}
+{"ts": "2025-12-27T02:46:29", "event": "train_log", "step": 2114, "epoch": 0.8919831223628693, "progress_pct": 14.87, "epoch_pct": 14.87, "eta": "48:36:38", "max_grad_norm": 1.0, "loss": 0.7940409779548645, "grad_norm": 0.38845229148864746, "learning_rate": 0.00019727136350526956}
+{"ts": "2025-12-27T02:46:46", "event": "train_log", "step": 2116, "epoch": 0.8928270042194093, "progress_pct": 14.88, "epoch_pct": 14.88, "eta": "48:35:01", "max_grad_norm": 1.0, "loss": 0.8279082775115967, "grad_norm": 0.41575267910957336, "learning_rate": 0.00019726020906936728}
+{"ts": "2025-12-27T02:47:02", "event": "train_log", "step": 2118, "epoch": 0.8936708860759494, "progress_pct": 14.89, "epoch_pct": 14.89, "eta": "48:33:15", "max_grad_norm": 1.0, "loss": 0.8217757940292358, "grad_norm": 0.37615057826042175, "learning_rate": 0.0001972490321974027}
+{"ts": "2025-12-27T02:47:16", "event": "train_log", "step": 2120, "epoch": 0.8945147679324894, "progress_pct": 14.91, "epoch_pct": 14.91, "eta": "48:31:26", "max_grad_norm": 1.0, "loss": 0.7637750506401062, "grad_norm": 0.9318385124206543, "learning_rate": 0.00019723783289195415}
+{"ts": "2025-12-27T02:47:33", "event": "train_log", "step": 2122, "epoch": 0.8953586497890296, "progress_pct": 14.92, "epoch_pct": 14.92, "eta": "48:29:46", "max_grad_norm": 1.0, "loss": 0.8702366948127747, "grad_norm": 0.41948962211608887, "learning_rate": 0.000197226611155605}
+{"ts": "2025-12-27T02:47:49", "event": "train_log", "step": 2124, "epoch": 0.8962025316455696, "progress_pct": 14.94, "epoch_pct": 14.94, "eta": "48:28:06", "max_grad_norm": 1.0, "loss": 0.8146984577178955, "grad_norm": 0.4718417525291443, "learning_rate": 0.00019721536699094392}
+{"ts": "2025-12-27T02:48:04", "event": "train_log", "step": 2126, "epoch": 0.8970464135021097, "progress_pct": 14.95, "epoch_pct": 14.95, "eta": "48:26:20", "max_grad_norm": 1.0, "loss": 0.8646075129508972, "grad_norm": 0.4470657408237457, "learning_rate": 0.00019720410040056474}
+{"ts": "2025-12-27T02:48:20", "event": "train_log", "step": 2128, "epoch": 0.8978902953586498, "progress_pct": 14.96, "epoch_pct": 14.96, "eta": "48:24:34", "max_grad_norm": 1.0, "loss": 0.7925992012023926, "grad_norm": 0.4176540970802307, "learning_rate": 0.00019719281138706643}
+{"ts": "2025-12-27T02:48:35", "event": "train_log", "step": 2130, "epoch": 0.8987341772151899, "progress_pct": 14.98, "epoch_pct": 14.98, "eta": "48:22:51", "max_grad_norm": 1.0, "loss": 0.8142143487930298, "grad_norm": 0.4854719042778015, "learning_rate": 0.00019718149995305315}
+{"ts": "2025-12-27T02:48:53", "event": "train_log", "step": 2132, "epoch": 0.8995780590717299, "progress_pct": 14.99, "epoch_pct": 14.99, "eta": "48:21:18", "max_grad_norm": 1.0, "loss": 0.8050890564918518, "grad_norm": 0.4270440340042114, "learning_rate": 0.00019717016610113427}
+{"ts": "2025-12-27T02:49:09", "event": "train_log", "step": 2134, "epoch": 0.90042194092827, "progress_pct": 15.01, "epoch_pct": 15.01, "eta": "48:19:38", "max_grad_norm": 1.0, "loss": 0.7888354659080505, "grad_norm": 0.37710970640182495, "learning_rate": 0.00019715880983392422}
+{"ts": "2025-12-27T02:49:24", "event": "train_log", "step": 2136, "epoch": 0.9012658227848102, "progress_pct": 15.02, "epoch_pct": 15.02, "eta": "48:17:47", "max_grad_norm": 1.0, "loss": 0.7348225712776184, "grad_norm": 0.4181252717971802, "learning_rate": 0.00019714743115404273}
+{"ts": "2025-12-27T02:49:39", "event": "train_log", "step": 2138, "epoch": 0.9021097046413502, "progress_pct": 15.04, "epoch_pct": 15.04, "eta": "48:16:02", "max_grad_norm": 1.0, "loss": 0.7729623913764954, "grad_norm": 0.4475359618663788, "learning_rate": 0.0001971360300641146}
+{"ts": "2025-12-27T02:49:55", "event": "train_log", "step": 2140, "epoch": 0.9029535864978903, "progress_pct": 15.05, "epoch_pct": 15.05, "eta": "48:14:21", "max_grad_norm": 1.0, "loss": 0.8066384792327881, "grad_norm": 0.4052332937717438, "learning_rate": 0.00019712460656676992}
+{"ts": "2025-12-27T02:50:10", "event": "train_log", "step": 2142, "epoch": 0.9037974683544304, "progress_pct": 15.06, "epoch_pct": 15.06, "eta": "48:12:35", "max_grad_norm": 1.0, "loss": 0.8053154349327087, "grad_norm": 0.4567393362522125, "learning_rate": 0.0001971131606646438}
+{"ts": "2025-12-27T02:50:27", "event": "train_log", "step": 2144, "epoch": 0.9046413502109705, "progress_pct": 15.08, "epoch_pct": 15.08, "eta": "48:11:00", "max_grad_norm": 1.0, "loss": 0.7242560386657715, "grad_norm": 0.4304409325122833, "learning_rate": 0.00019710169236037656}
+{"ts": "2025-12-27T02:50:42", "event": "train_log", "step": 2146, "epoch": 0.9054852320675105, "progress_pct": 15.09, "epoch_pct": 15.09, "eta": "48:09:17", "max_grad_norm": 1.0, "loss": 0.8475301265716553, "grad_norm": 0.4785008132457733, "learning_rate": 0.0001970902016566138}
+{"ts": "2025-12-27T02:50:58", "event": "train_log", "step": 2148, "epoch": 0.9063291139240506, "progress_pct": 15.11, "epoch_pct": 15.11, "eta": "48:07:33", "max_grad_norm": 1.0, "loss": 0.928544282913208, "grad_norm": 0.45226508378982544, "learning_rate": 0.00019707868855600618}
+{"ts": "2025-12-27T02:51:14", "event": "train_log", "step": 2150, "epoch": 0.9071729957805907, "progress_pct": 15.12, "epoch_pct": 15.12, "eta": "48:05:53", "max_grad_norm": 1.0, "loss": 0.8135722875595093, "grad_norm": 0.3794468641281128, "learning_rate": 0.00019706715306120956}
+{"ts": "2025-12-27T02:51:29", "event": "train_log", "step": 2152, "epoch": 0.9080168776371308, "progress_pct": 15.13, "epoch_pct": 15.13, "eta": "48:04:12", "max_grad_norm": 1.0, "loss": 0.7808290719985962, "grad_norm": 0.4005141258239746, "learning_rate": 0.0001970555951748849}
+{"ts": "2025-12-27T02:51:45", "event": "train_log", "step": 2154, "epoch": 0.9088607594936708, "progress_pct": 15.15, "epoch_pct": 15.15, "eta": "48:02:29", "max_grad_norm": 1.0, "loss": 0.8292917013168335, "grad_norm": 0.37929919362068176, "learning_rate": 0.00019704401489969842}
+{"ts": "2025-12-27T02:52:00", "event": "train_log", "step": 2156, "epoch": 0.909704641350211, "progress_pct": 15.16, "epoch_pct": 15.16, "eta": "48:00:45", "max_grad_norm": 1.0, "loss": 0.772884726524353, "grad_norm": 0.4458678066730499, "learning_rate": 0.0001970324122383215}
+{"ts": "2025-12-27T02:52:15", "event": "train_log", "step": 2158, "epoch": 0.9105485232067511, "progress_pct": 15.18, "epoch_pct": 15.18, "eta": "47:59:01", "max_grad_norm": 1.0, "loss": 0.8204470872879028, "grad_norm": 0.41907885670661926, "learning_rate": 0.0001970207871934306}
+{"ts": "2025-12-27T02:52:31", "event": "train_log", "step": 2160, "epoch": 0.9113924050632911, "progress_pct": 15.19, "epoch_pct": 15.19, "eta": "47:57:20", "max_grad_norm": 1.0, "loss": 0.8361451625823975, "grad_norm": 0.40610212087631226, "learning_rate": 0.00019700913976770741}
+{"ts": "2025-12-27T02:52:46", "event": "train_log", "step": 2162, "epoch": 0.9122362869198313, "progress_pct": 15.2, "epoch_pct": 15.2, "eta": "47:55:39", "max_grad_norm": 1.0, "loss": 0.8456075191497803, "grad_norm": 0.4213273227214813, "learning_rate": 0.00019699746996383878}
+{"ts": "2025-12-27T02:53:02", "event": "train_log", "step": 2164, "epoch": 0.9130801687763713, "progress_pct": 15.22, "epoch_pct": 15.22, "eta": "47:53:56", "max_grad_norm": 1.0, "loss": 0.8731603622436523, "grad_norm": 0.49264612793922424, "learning_rate": 0.0001969857777845167}
+{"ts": "2025-12-27T02:53:18", "event": "train_log", "step": 2166, "epoch": 0.9139240506329114, "progress_pct": 15.23, "epoch_pct": 15.23, "eta": "47:52:21", "max_grad_norm": 1.0, "loss": 0.8149666786193848, "grad_norm": 0.42002132534980774, "learning_rate": 0.00019697406323243834}
+{"ts": "2025-12-27T02:53:34", "event": "train_log", "step": 2168, "epoch": 0.9147679324894514, "progress_pct": 15.25, "epoch_pct": 15.25, "eta": "47:50:41", "max_grad_norm": 1.0, "loss": 0.7508546710014343, "grad_norm": 0.37501421570777893, "learning_rate": 0.000196962326310306}
+{"ts": "2025-12-27T02:53:50", "event": "train_log", "step": 2170, "epoch": 0.9156118143459916, "progress_pct": 15.26, "epoch_pct": 15.26, "eta": "47:49:00", "max_grad_norm": 1.0, "loss": 0.830280601978302, "grad_norm": 0.3706159293651581, "learning_rate": 0.00019695056702082717}
+{"ts": "2025-12-27T02:54:07", "event": "train_log", "step": 2172, "epoch": 0.9164556962025316, "progress_pct": 15.27, "epoch_pct": 15.27, "eta": "47:47:27", "max_grad_norm": 1.0, "loss": 0.7297167778015137, "grad_norm": 0.3873467445373535, "learning_rate": 0.00019693878536671452}
+{"ts": "2025-12-27T02:54:22", "event": "train_log", "step": 2174, "epoch": 0.9172995780590717, "progress_pct": 15.29, "epoch_pct": 15.29, "eta": "47:45:47", "max_grad_norm": 1.0, "loss": 0.7924292087554932, "grad_norm": 0.42876651883125305, "learning_rate": 0.00019692698135068582}
+{"ts": "2025-12-27T02:54:38", "event": "train_log", "step": 2176, "epoch": 0.9181434599156119, "progress_pct": 15.3, "epoch_pct": 15.3, "eta": "47:44:05", "max_grad_norm": 1.0, "loss": 0.8098247051239014, "grad_norm": 0.4259159564971924, "learning_rate": 0.00019691515497546406}
+{"ts": "2025-12-27T02:54:54", "event": "train_log", "step": 2178, "epoch": 0.9189873417721519, "progress_pct": 15.32, "epoch_pct": 15.32, "eta": "47:42:29", "max_grad_norm": 1.0, "loss": 0.6789063215255737, "grad_norm": 0.38152721524238586, "learning_rate": 0.00019690330624377728}
+{"ts": "2025-12-27T02:55:11", "event": "train_log", "step": 2180, "epoch": 0.919831223628692, "progress_pct": 15.33, "epoch_pct": 15.33, "eta": "47:40:54", "max_grad_norm": 1.0, "loss": 0.8582620620727539, "grad_norm": 0.4135437607765198, "learning_rate": 0.00019689143515835887}
+{"ts": "2025-12-27T02:55:28", "event": "train_log", "step": 2182, "epoch": 0.920675105485232, "progress_pct": 15.34, "epoch_pct": 15.34, "eta": "47:39:23", "max_grad_norm": 1.0, "loss": 0.8059743642807007, "grad_norm": 0.3818798065185547, "learning_rate": 0.0001968795417219472}
+{"ts": "2025-12-27T02:55:46", "event": "train_log", "step": 2184, "epoch": 0.9215189873417722, "progress_pct": 15.36, "epoch_pct": 15.36, "eta": "47:37:56", "max_grad_norm": 1.0, "loss": 0.7432259917259216, "grad_norm": 0.39021286368370056, "learning_rate": 0.00019686762593728582}
+{"ts": "2025-12-27T02:56:02", "event": "train_log", "step": 2186, "epoch": 0.9223628691983122, "progress_pct": 15.37, "epoch_pct": 15.37, "eta": "47:36:22", "max_grad_norm": 1.0, "loss": 0.7561086416244507, "grad_norm": 0.37864068150520325, "learning_rate": 0.00019685568780712355}
+{"ts": "2025-12-27T02:56:18", "event": "train_log", "step": 2188, "epoch": 0.9232067510548523, "progress_pct": 15.39, "epoch_pct": 15.39, "eta": "47:34:44", "max_grad_norm": 1.0, "loss": 0.7552722692489624, "grad_norm": 0.450847864151001, "learning_rate": 0.00019684372733421425}
+{"ts": "2025-12-27T02:56:34", "event": "train_log", "step": 2190, "epoch": 0.9240506329113924, "progress_pct": 15.4, "epoch_pct": 15.4, "eta": "47:33:07", "max_grad_norm": 1.0, "loss": 0.6905336380004883, "grad_norm": 0.37909507751464844, "learning_rate": 0.00019683174452131695}
+{"ts": "2025-12-27T02:56:49", "event": "train_log", "step": 2192, "epoch": 0.9248945147679325, "progress_pct": 15.41, "epoch_pct": 15.41, "eta": "47:31:28", "max_grad_norm": 1.0, "loss": 0.7710593938827515, "grad_norm": 0.4439173638820648, "learning_rate": 0.0001968197393711959}
+{"ts": "2025-12-27T02:57:05", "event": "train_log", "step": 2194, "epoch": 0.9257383966244725, "progress_pct": 15.43, "epoch_pct": 15.43, "eta": "47:29:48", "max_grad_norm": 1.0, "loss": 0.8212814331054688, "grad_norm": 0.3908376395702362, "learning_rate": 0.00019680771188662044}
+{"ts": "2025-12-27T02:57:21", "event": "train_log", "step": 2196, "epoch": 0.9265822784810127, "progress_pct": 15.44, "epoch_pct": 15.44, "eta": "47:28:11", "max_grad_norm": 1.0, "loss": 0.751355767250061, "grad_norm": 0.3528299033641815, "learning_rate": 0.00019679566207036507}
+{"ts": "2025-12-27T02:57:35", "event": "train_log", "step": 2198, "epoch": 0.9274261603375528, "progress_pct": 15.46, "epoch_pct": 15.46, "eta": "47:26:26", "max_grad_norm": 1.0, "loss": 0.8204057812690735, "grad_norm": 0.4351397156715393, "learning_rate": 0.00019678358992520945}
+{"ts": "2025-12-27T02:57:51", "event": "train_log", "step": 2200, "epoch": 0.9282700421940928, "progress_pct": 15.47, "epoch_pct": 15.47, "eta": "47:24:48", "max_grad_norm": 1.0, "loss": 0.763841450214386, "grad_norm": 0.45566943287849426, "learning_rate": 0.0001967714954539384}
+{"ts": "2025-12-27T03:09:05", "event": "train_log", "step": 2200, "epoch": 0.9282700421940928, "progress_pct": 15.47, "epoch_pct": 15.47, "eta": "48:26:13", "max_grad_norm": 1.0, "eval_loss": 0.8336610198020935, "eval_runtime": 674.5134, "eval_samples_per_second": 3.124, "eval_steps_per_second": 3.124}
+{"ts": "2025-12-27T03:09:20", "event": "train_log", "step": 2202, "epoch": 0.9291139240506329, "progress_pct": 15.49, "epoch_pct": 15.49, "eta": "48:24:27", "max_grad_norm": 1.0, "loss": 0.861711323261261, "grad_norm": 0.4516221284866333, "learning_rate": 0.00019675937865934188}
+{"ts": "2025-12-27T03:09:36", "event": "train_log", "step": 2204, "epoch": 0.929957805907173, "progress_pct": 15.5, "epoch_pct": 15.5, "eta": "48:22:43", "max_grad_norm": 1.0, "loss": 0.8062718510627747, "grad_norm": 0.43421855568885803, "learning_rate": 0.00019674723954421502}
+{"ts": "2025-12-27T03:09:50", "event": "train_log", "step": 2206, "epoch": 0.9308016877637131, "progress_pct": 15.51, "epoch_pct": 15.51, "eta": "48:20:57", "max_grad_norm": 1.0, "loss": 0.8069695234298706, "grad_norm": 0.42442476749420166, "learning_rate": 0.00019673507811135803}
+{"ts": "2025-12-27T03:10:06", "event": "train_log", "step": 2208, "epoch": 0.9316455696202531, "progress_pct": 15.53, "epoch_pct": 15.53, "eta": "48:19:13", "max_grad_norm": 1.0, "loss": 0.7336108684539795, "grad_norm": 0.3976045250892639, "learning_rate": 0.00019672289436357637}
+{"ts": "2025-12-27T03:10:20", "event": "train_log", "step": 2210, "epoch": 0.9324894514767933, "progress_pct": 15.54, "epoch_pct": 15.54, "eta": "48:17:25", "max_grad_norm": 1.0, "loss": 0.7559890151023865, "grad_norm": 0.4131270945072174, "learning_rate": 0.00019671068830368058}
+{"ts": "2025-12-27T03:10:35", "event": "train_log", "step": 2212, "epoch": 0.9333333333333333, "progress_pct": 15.56, "epoch_pct": 15.56, "eta": "48:15:41", "max_grad_norm": 1.0, "loss": 0.8447075486183167, "grad_norm": 0.4599294364452362, "learning_rate": 0.00019669845993448634}
+{"ts": "2025-12-27T03:10:50", "event": "train_log", "step": 2214, "epoch": 0.9341772151898734, "progress_pct": 15.57, "epoch_pct": 15.57, "eta": "48:13:58", "max_grad_norm": 1.0, "loss": 0.8329233527183533, "grad_norm": 0.4291204810142517, "learning_rate": 0.00019668620925881455}
+{"ts": "2025-12-27T03:11:07", "event": "train_log", "step": 2216, "epoch": 0.9350210970464135, "progress_pct": 15.58, "epoch_pct": 15.58, "eta": "48:12:20", "max_grad_norm": 1.0, "loss": 0.749232292175293, "grad_norm": 0.4226539731025696, "learning_rate": 0.00019667393627949116}
+{"ts": "2025-12-27T03:11:22", "event": "train_log", "step": 2218, "epoch": 0.9358649789029536, "progress_pct": 15.6, "epoch_pct": 15.6, "eta": "48:10:38", "max_grad_norm": 1.0, "loss": 0.7470011115074158, "grad_norm": 0.4159580171108246, "learning_rate": 0.00019666164099934732}
+{"ts": "2025-12-27T03:11:37", "event": "train_log", "step": 2220, "epoch": 0.9367088607594937, "progress_pct": 15.61, "epoch_pct": 15.61, "eta": "48:08:51", "max_grad_norm": 1.0, "loss": 0.8094202280044556, "grad_norm": 0.4160766303539276, "learning_rate": 0.00019664932342121935}
+{"ts": "2025-12-27T03:11:52", "event": "train_log", "step": 2222, "epoch": 0.9375527426160337, "progress_pct": 15.63, "epoch_pct": 15.63, "eta": "48:07:10", "max_grad_norm": 1.0, "loss": 0.8398169875144958, "grad_norm": 0.45284298062324524, "learning_rate": 0.00019663698354794863}
+{"ts": "2025-12-27T03:12:07", "event": "train_log", "step": 2224, "epoch": 0.9383966244725739, "progress_pct": 15.64, "epoch_pct": 15.64, "eta": "48:05:24", "max_grad_norm": 1.0, "loss": 0.7652443051338196, "grad_norm": 0.3960540294647217, "learning_rate": 0.00019662462138238178}
+{"ts": "2025-12-27T03:12:21", "event": "train_log", "step": 2226, "epoch": 0.9392405063291139, "progress_pct": 15.65, "epoch_pct": 15.65, "eta": "48:03:39", "max_grad_norm": 1.0, "loss": 0.7909489870071411, "grad_norm": 0.4269275367259979, "learning_rate": 0.00019661223692737047}
+{"ts": "2025-12-27T03:12:37", "event": "train_log", "step": 2228, "epoch": 0.940084388185654, "progress_pct": 15.67, "epoch_pct": 15.67, "eta": "48:02:00", "max_grad_norm": 1.0, "loss": 0.7821935415267944, "grad_norm": 0.38187873363494873, "learning_rate": 0.00019659983018577157}
+{"ts": "2025-12-27T03:12:52", "event": "train_log", "step": 2230, "epoch": 0.9409282700421941, "progress_pct": 15.68, "epoch_pct": 15.68, "eta": "48:00:17", "max_grad_norm": 1.0, "loss": 0.7792721390724182, "grad_norm": 0.45898866653442383, "learning_rate": 0.0001965874011604471}
+{"ts": "2025-12-27T03:13:06", "event": "train_log", "step": 2232, "epoch": 0.9417721518987342, "progress_pct": 15.7, "epoch_pct": 15.7, "eta": "47:58:30", "max_grad_norm": 1.0, "loss": 0.885725200176239, "grad_norm": 0.4740699827671051, "learning_rate": 0.00019657494985426418}
+{"ts": "2025-12-27T03:13:21", "event": "train_log", "step": 2234, "epoch": 0.9426160337552743, "progress_pct": 15.71, "epoch_pct": 15.71, "eta": "47:56:46", "max_grad_norm": 1.0, "loss": 0.7898101806640625, "grad_norm": 0.3912813365459442, "learning_rate": 0.00019656247627009506}
+{"ts": "2025-12-27T03:13:36", "event": "train_log", "step": 2236, "epoch": 0.9434599156118143, "progress_pct": 15.72, "epoch_pct": 15.72, "eta": "47:55:03", "max_grad_norm": 1.0, "loss": 0.8646330833435059, "grad_norm": 0.4552418291568756, "learning_rate": 0.0001965499804108172}
+{"ts": "2025-12-27T03:13:51", "event": "train_log", "step": 2238, "epoch": 0.9443037974683545, "progress_pct": 15.74, "epoch_pct": 15.74, "eta": "47:53:17", "max_grad_norm": 1.0, "loss": 0.8068296313285828, "grad_norm": 0.4178548753261566, "learning_rate": 0.0001965374622793131}
+{"ts": "2025-12-27T03:14:06", "event": "train_log", "step": 2240, "epoch": 0.9451476793248945, "progress_pct": 15.75, "epoch_pct": 15.75, "eta": "47:51:36", "max_grad_norm": 1.0, "loss": 0.7953271865844727, "grad_norm": 0.38649076223373413, "learning_rate": 0.00019652492187847057}
+{"ts": "2025-12-27T03:14:20", "event": "train_log", "step": 2242, "epoch": 0.9459915611814346, "progress_pct": 15.77, "epoch_pct": 15.77, "eta": "47:49:49", "max_grad_norm": 1.0, "loss": 0.8423402309417725, "grad_norm": 0.39785075187683105, "learning_rate": 0.0001965123592111823}
+{"ts": "2025-12-27T03:14:35", "event": "train_log", "step": 2244, "epoch": 0.9468354430379747, "progress_pct": 15.78, "epoch_pct": 15.78, "eta": "47:48:06", "max_grad_norm": 1.0, "loss": 0.8883589506149292, "grad_norm": 0.47455263137817383, "learning_rate": 0.00019649977428034632}
+{"ts": "2025-12-27T03:14:50", "event": "train_log", "step": 2246, "epoch": 0.9476793248945148, "progress_pct": 15.79, "epoch_pct": 15.79, "eta": "47:46:26", "max_grad_norm": 1.0, "loss": 0.7232863903045654, "grad_norm": 0.3866841197013855, "learning_rate": 0.0001964871670888657}
+{"ts": "2025-12-27T03:15:06", "event": "train_log", "step": 2248, "epoch": 0.9485232067510548, "progress_pct": 15.81, "epoch_pct": 15.81, "eta": "47:44:47", "max_grad_norm": 1.0, "loss": 0.7265347838401794, "grad_norm": 0.3950609564781189, "learning_rate": 0.00019647453763964876}
+{"ts": "2025-12-27T03:15:21", "event": "train_log", "step": 2250, "epoch": 0.9493670886075949, "progress_pct": 15.82, "epoch_pct": 15.82, "eta": "47:43:04", "max_grad_norm": 1.0, "loss": 0.8000131249427795, "grad_norm": 0.4315829575061798, "learning_rate": 0.0001964618859356088}
+{"ts": "2025-12-27T03:15:37", "event": "train_log", "step": 2252, "epoch": 0.950210970464135, "progress_pct": 15.84, "epoch_pct": 15.84, "eta": "47:41:30", "max_grad_norm": 1.0, "loss": 0.823086678981781, "grad_norm": 0.403984934091568, "learning_rate": 0.00019644921197966433}
+{"ts": "2025-12-27T03:15:53", "event": "train_log", "step": 2254, "epoch": 0.9510548523206751, "progress_pct": 15.85, "epoch_pct": 15.85, "eta": "47:39:52", "max_grad_norm": 1.0, "loss": 0.8457553386688232, "grad_norm": 0.4311080574989319, "learning_rate": 0.00019643651577473897}
+{"ts": "2025-12-27T03:16:09", "event": "train_log", "step": 2256, "epoch": 0.9518987341772152, "progress_pct": 15.86, "epoch_pct": 15.86, "eta": "47:38:20", "max_grad_norm": 1.0, "loss": 0.7866189479827881, "grad_norm": 0.3884856104850769, "learning_rate": 0.00019642379732376158}
+{"ts": "2025-12-27T03:16:25", "event": "train_log", "step": 2258, "epoch": 0.9527426160337553, "progress_pct": 15.88, "epoch_pct": 15.88, "eta": "47:36:43", "max_grad_norm": 1.0, "loss": 0.7095813751220703, "grad_norm": 0.3384067714214325, "learning_rate": 0.00019641105662966595}
+{"ts": "2025-12-27T03:16:41", "event": "train_log", "step": 2260, "epoch": 0.9535864978902954, "progress_pct": 15.89, "epoch_pct": 15.89, "eta": "47:35:06", "max_grad_norm": 1.0, "loss": 0.8271632194519043, "grad_norm": 0.40426746010780334, "learning_rate": 0.00019639829369539117}
+{"ts": "2025-12-27T03:16:57", "event": "train_log", "step": 2262, "epoch": 0.9544303797468354, "progress_pct": 15.91, "epoch_pct": 15.91, "eta": "47:33:30", "max_grad_norm": 1.0, "loss": 0.8204861283302307, "grad_norm": 0.45495131611824036, "learning_rate": 0.00019638550852388144}
+{"ts": "2025-12-27T03:17:12", "event": "train_log", "step": 2264, "epoch": 0.9552742616033755, "progress_pct": 15.92, "epoch_pct": 15.92, "eta": "47:31:50", "max_grad_norm": 1.0, "loss": 0.884763777256012, "grad_norm": 0.47201067209243774, "learning_rate": 0.00019637270111808598}
+{"ts": "2025-12-27T03:17:27", "event": "train_log", "step": 2266, "epoch": 0.9561181434599156, "progress_pct": 15.94, "epoch_pct": 15.94, "eta": "47:30:13", "max_grad_norm": 1.0, "loss": 0.7630926370620728, "grad_norm": 0.3959396779537201, "learning_rate": 0.0001963598714809592}
+{"ts": "2025-12-27T03:17:43", "event": "train_log", "step": 2268, "epoch": 0.9569620253164557, "progress_pct": 15.95, "epoch_pct": 15.95, "eta": "47:28:36", "max_grad_norm": 1.0, "loss": 0.797870397567749, "grad_norm": 0.4197136461734772, "learning_rate": 0.00019634701961546077}
+{"ts": "2025-12-27T03:17:58", "event": "train_log", "step": 2270, "epoch": 0.9578059071729957, "progress_pct": 15.96, "epoch_pct": 15.96, "eta": "47:26:55", "max_grad_norm": 1.0, "loss": 0.7755846977233887, "grad_norm": 0.381290078163147, "learning_rate": 0.00019633414552455526}
+{"ts": "2025-12-27T03:18:13", "event": "train_log", "step": 2272, "epoch": 0.9586497890295359, "progress_pct": 15.98, "epoch_pct": 15.98, "eta": "47:25:15", "max_grad_norm": 1.0, "loss": 0.7553264498710632, "grad_norm": 0.3868124783039093, "learning_rate": 0.00019632124921121255}
+{"ts": "2025-12-27T03:18:29", "event": "train_log", "step": 2274, "epoch": 0.959493670886076, "progress_pct": 15.99, "epoch_pct": 15.99, "eta": "47:23:39", "max_grad_norm": 1.0, "loss": 0.7792160511016846, "grad_norm": 0.4350631833076477, "learning_rate": 0.00019630833067840747}
+{"ts": "2025-12-27T03:18:44", "event": "train_log", "step": 2276, "epoch": 0.960337552742616, "progress_pct": 16.01, "epoch_pct": 16.01, "eta": "47:22:01", "max_grad_norm": 1.0, "loss": 0.878220796585083, "grad_norm": 0.4717157781124115, "learning_rate": 0.00019629538992912016}
+{"ts": "2025-12-27T03:18:59", "event": "train_log", "step": 2278, "epoch": 0.9611814345991562, "progress_pct": 16.02, "epoch_pct": 16.02, "eta": "47:20:21", "max_grad_norm": 1.0, "loss": 0.8891014456748962, "grad_norm": 0.45545122027397156, "learning_rate": 0.00019628242696633584}
+{"ts": "2025-12-27T03:19:15", "event": "train_log", "step": 2280, "epoch": 0.9620253164556962, "progress_pct": 16.03, "epoch_pct": 16.03, "eta": "47:18:45", "max_grad_norm": 1.0, "loss": 0.8024723529815674, "grad_norm": 0.4421025514602661, "learning_rate": 0.00019626944179304467}
+{"ts": "2025-12-27T03:19:30", "event": "train_log", "step": 2282, "epoch": 0.9628691983122363, "progress_pct": 16.05, "epoch_pct": 16.05, "eta": "47:17:08", "max_grad_norm": 1.0, "loss": 0.8036280274391174, "grad_norm": 0.40385812520980835, "learning_rate": 0.00019625643441224222}
+{"ts": "2025-12-27T03:19:47", "event": "train_log", "step": 2284, "epoch": 0.9637130801687763, "progress_pct": 16.06, "epoch_pct": 16.06, "eta": "47:15:40", "max_grad_norm": 1.0, "loss": 0.8072395324707031, "grad_norm": 0.3866552412509918, "learning_rate": 0.00019624340482692898}
+{"ts": "2025-12-27T03:20:03", "event": "train_log", "step": 2286, "epoch": 0.9645569620253165, "progress_pct": 16.08, "epoch_pct": 16.08, "eta": "47:14:06", "max_grad_norm": 1.0, "loss": 0.8046114444732666, "grad_norm": 0.4113233685493469, "learning_rate": 0.0001962303530401106}
+{"ts": "2025-12-27T03:20:18", "event": "train_log", "step": 2288, "epoch": 0.9654008438818565, "progress_pct": 16.09, "epoch_pct": 16.09, "eta": "47:12:28", "max_grad_norm": 1.0, "loss": 0.8434037566184998, "grad_norm": 0.40720227360725403, "learning_rate": 0.00019621727905479798}
+{"ts": "2025-12-27T03:20:35", "event": "train_log", "step": 2290, "epoch": 0.9662447257383966, "progress_pct": 16.1, "epoch_pct": 16.1, "eta": "47:10:57", "max_grad_norm": 1.0, "loss": 0.7832419276237488, "grad_norm": 0.35759273171424866, "learning_rate": 0.00019620418287400693}
+{"ts": "2025-12-27T03:20:50", "event": "train_log", "step": 2292, "epoch": 0.9670886075949368, "progress_pct": 16.12, "epoch_pct": 16.12, "eta": "47:09:20", "max_grad_norm": 1.0, "loss": 0.7975693345069885, "grad_norm": 0.4170442223548889, "learning_rate": 0.0001961910645007585}
+{"ts": "2025-12-27T03:21:06", "event": "train_log", "step": 2294, "epoch": 0.9679324894514768, "progress_pct": 16.13, "epoch_pct": 16.13, "eta": "47:07:45", "max_grad_norm": 1.0, "loss": 0.8554494976997375, "grad_norm": 0.4126700162887573, "learning_rate": 0.00019617792393807895}
+{"ts": "2025-12-27T03:21:21", "event": "train_log", "step": 2296, "epoch": 0.9687763713080169, "progress_pct": 16.15, "epoch_pct": 16.15, "eta": "47:06:09", "max_grad_norm": 1.0, "loss": 0.824539065361023, "grad_norm": 0.4231695830821991, "learning_rate": 0.00019616476118899942}
+{"ts": "2025-12-27T03:21:37", "event": "train_log", "step": 2298, "epoch": 0.9696202531645569, "progress_pct": 16.16, "epoch_pct": 16.16, "eta": "47:04:35", "max_grad_norm": 1.0, "loss": 0.7008040547370911, "grad_norm": 0.3556209206581116, "learning_rate": 0.0001961515762565564}
+{"ts": "2025-12-27T03:21:53", "event": "train_log", "step": 2300, "epoch": 0.9704641350210971, "progress_pct": 16.17, "epoch_pct": 16.17, "eta": "47:02:59", "max_grad_norm": 1.0, "loss": 0.8192558288574219, "grad_norm": 0.4370400309562683, "learning_rate": 0.00019613836914379132}
+{"ts": "2025-12-27T03:33:21", "event": "train_log", "step": 2300, "epoch": 0.9704641350210971, "progress_pct": 16.17, "epoch_pct": 16.17, "eta": "48:02:28", "max_grad_norm": 1.0, "eval_loss": 0.8281980156898499, "eval_runtime": 688.6136, "eval_samples_per_second": 3.06, "eval_steps_per_second": 3.06}
+{"ts": "2025-12-27T03:33:36", "event": "train_log", "step": 2302, "epoch": 0.9713080168776371, "progress_pct": 16.19, "epoch_pct": 16.19, "eta": "48:00:45", "max_grad_norm": 1.0, "loss": 0.878928005695343, "grad_norm": 0.4207614064216614, "learning_rate": 0.00019612513985375088}
+{"ts": "2025-12-27T03:33:51", "event": "train_log", "step": 2304, "epoch": 0.9721518987341772, "progress_pct": 16.2, "epoch_pct": 16.2, "eta": "47:59:07", "max_grad_norm": 1.0, "loss": 0.736182689666748, "grad_norm": 0.37607815861701965, "learning_rate": 0.0001961118883894868}
+{"ts": "2025-12-27T03:34:06", "event": "train_log", "step": 2306, "epoch": 0.9729957805907173, "progress_pct": 16.22, "epoch_pct": 16.22, "eta": "47:57:24", "max_grad_norm": 1.0, "loss": 0.8403618335723877, "grad_norm": 0.43311476707458496, "learning_rate": 0.00019609861475405592}
+{"ts": "2025-12-27T03:34:21", "event": "train_log", "step": 2308, "epoch": 0.9738396624472574, "progress_pct": 16.23, "epoch_pct": 16.23, "eta": "47:55:44", "max_grad_norm": 1.0, "loss": 0.8414564728736877, "grad_norm": 0.4533176124095917, "learning_rate": 0.00019608531895052021}
+{"ts": "2025-12-27T03:34:37", "event": "train_log", "step": 2310, "epoch": 0.9746835443037974, "progress_pct": 16.24, "epoch_pct": 16.24, "eta": "47:54:04", "max_grad_norm": 1.0, "loss": 0.8105775117874146, "grad_norm": 0.4025697410106659, "learning_rate": 0.00019607200098194677}
+{"ts": "2025-12-27T03:34:51", "event": "train_log", "step": 2312, "epoch": 0.9755274261603376, "progress_pct": 16.26, "epoch_pct": 16.26, "eta": "47:52:22", "max_grad_norm": 1.0, "loss": 0.8053821325302124, "grad_norm": 0.37894198298454285, "learning_rate": 0.00019605866085140784}
+{"ts": "2025-12-27T03:35:06", "event": "train_log", "step": 2314, "epoch": 0.9763713080168777, "progress_pct": 16.27, "epoch_pct": 16.27, "eta": "47:50:41", "max_grad_norm": 1.0, "loss": 0.8092884421348572, "grad_norm": 0.46265891194343567, "learning_rate": 0.0001960452985619807}
+{"ts": "2025-12-27T03:35:21", "event": "train_log", "step": 2316, "epoch": 0.9772151898734177, "progress_pct": 16.29, "epoch_pct": 16.29, "eta": "47:48:59", "max_grad_norm": 1.0, "loss": 0.7456095218658447, "grad_norm": 0.3770054280757904, "learning_rate": 0.0001960319141167477}
+{"ts": "2025-12-27T03:35:35", "event": "train_log", "step": 2318, "epoch": 0.9780590717299578, "progress_pct": 16.3, "epoch_pct": 16.3, "eta": "47:47:15", "max_grad_norm": 1.0, "loss": 0.7739883661270142, "grad_norm": 0.44084450602531433, "learning_rate": 0.0001960185075187965}
+{"ts": "2025-12-27T03:35:51", "event": "train_log", "step": 2320, "epoch": 0.9789029535864979, "progress_pct": 16.32, "epoch_pct": 16.32, "eta": "47:45:37", "max_grad_norm": 1.0, "loss": 0.6961302757263184, "grad_norm": 0.3880341351032257, "learning_rate": 0.00019600507877121965}
+{"ts": "2025-12-27T03:36:05", "event": "train_log", "step": 2322, "epoch": 0.979746835443038, "progress_pct": 16.33, "epoch_pct": 16.33, "eta": "47:43:54", "max_grad_norm": 1.0, "loss": 0.8447984457015991, "grad_norm": 0.468037486076355, "learning_rate": 0.00019599162787711495}
+{"ts": "2025-12-27T03:36:20", "event": "train_log", "step": 2324, "epoch": 0.980590717299578, "progress_pct": 16.34, "epoch_pct": 16.34, "eta": "47:42:14", "max_grad_norm": 1.0, "loss": 0.8155615329742432, "grad_norm": 0.4123631417751312, "learning_rate": 0.00019597815483958527}
+{"ts": "2025-12-27T03:36:35", "event": "train_log", "step": 2326, "epoch": 0.9814345991561182, "progress_pct": 16.36, "epoch_pct": 16.36, "eta": "47:40:34", "max_grad_norm": 1.0, "loss": 0.7990952730178833, "grad_norm": 0.3827875554561615, "learning_rate": 0.00019596465966173853}
+{"ts": "2025-12-27T03:36:50", "event": "train_log", "step": 2328, "epoch": 0.9822784810126582, "progress_pct": 16.37, "epoch_pct": 16.37, "eta": "47:38:52", "max_grad_norm": 1.0, "loss": 0.8899750709533691, "grad_norm": 0.4644222557544708, "learning_rate": 0.00019595114234668788}
+{"ts": "2025-12-27T03:37:04", "event": "train_log", "step": 2330, "epoch": 0.9831223628691983, "progress_pct": 16.39, "epoch_pct": 16.39, "eta": "47:37:10", "max_grad_norm": 1.0, "loss": 0.7863160371780396, "grad_norm": 0.43699249625205994, "learning_rate": 0.00019593760289755144}
+{"ts": "2025-12-27T03:37:19", "event": "train_log", "step": 2332, "epoch": 0.9839662447257383, "progress_pct": 16.4, "epoch_pct": 16.4, "eta": "47:35:27", "max_grad_norm": 1.0, "loss": 0.7960637807846069, "grad_norm": 0.40713244676589966, "learning_rate": 0.00019592404131745254}
+{"ts": "2025-12-27T03:37:34", "event": "train_log", "step": 2334, "epoch": 0.9848101265822785, "progress_pct": 16.41, "epoch_pct": 16.41, "eta": "47:33:52", "max_grad_norm": 1.0, "loss": 0.714144229888916, "grad_norm": 0.37364068627357483, "learning_rate": 0.00019591045760951957}
+{"ts": "2025-12-27T03:37:50", "event": "train_log", "step": 2336, "epoch": 0.9856540084388186, "progress_pct": 16.43, "epoch_pct": 16.43, "eta": "47:32:14", "max_grad_norm": 1.0, "loss": 0.7311791777610779, "grad_norm": 0.4094736576080322, "learning_rate": 0.00019589685177688599}
+{"ts": "2025-12-27T03:38:05", "event": "train_log", "step": 2338, "epoch": 0.9864978902953586, "progress_pct": 16.44, "epoch_pct": 16.44, "eta": "47:30:36", "max_grad_norm": 1.0, "loss": 0.8191342353820801, "grad_norm": 0.4275180697441101, "learning_rate": 0.0001958832238226905}
+{"ts": "2025-12-27T03:38:20", "event": "train_log", "step": 2340, "epoch": 0.9873417721518988, "progress_pct": 16.46, "epoch_pct": 16.46, "eta": "47:28:56", "max_grad_norm": 1.0, "loss": 0.7476106286048889, "grad_norm": 0.4627651274204254, "learning_rate": 0.00019586957375007668}
+{"ts": "2025-12-27T03:38:34", "event": "train_log", "step": 2342, "epoch": 0.9881856540084388, "progress_pct": 16.47, "epoch_pct": 16.47, "eta": "47:27:13", "max_grad_norm": 1.0, "loss": 0.8257967233657837, "grad_norm": 0.44676414132118225, "learning_rate": 0.00019585590156219346}
+{"ts": "2025-12-27T03:38:49", "event": "train_log", "step": 2344, "epoch": 0.9890295358649789, "progress_pct": 16.48, "epoch_pct": 16.48, "eta": "47:25:38", "max_grad_norm": 1.0, "loss": 0.7922079563140869, "grad_norm": 0.40667134523391724, "learning_rate": 0.00019584220726219466}
+{"ts": "2025-12-27T03:39:04", "event": "train_log", "step": 2346, "epoch": 0.9898734177215189, "progress_pct": 16.5, "epoch_pct": 16.5, "eta": "47:23:59", "max_grad_norm": 1.0, "loss": 0.7892434597015381, "grad_norm": 0.40865346789360046, "learning_rate": 0.00019582849085323934}
+{"ts": "2025-12-27T03:39:18", "event": "train_log", "step": 2348, "epoch": 0.9907172995780591, "progress_pct": 16.51, "epoch_pct": 16.51, "eta": "47:22:16", "max_grad_norm": 1.0, "loss": 0.8766644597053528, "grad_norm": 0.43309709429740906, "learning_rate": 0.00019581475233849163}
+{"ts": "2025-12-27T03:39:34", "event": "train_log", "step": 2350, "epoch": 0.9915611814345991, "progress_pct": 16.53, "epoch_pct": 16.53, "eta": "47:20:40", "max_grad_norm": 1.0, "loss": 0.9183312058448792, "grad_norm": 0.41558781266212463, "learning_rate": 0.00019580099172112069}
+{"ts": "2025-12-27T03:39:48", "event": "train_log", "step": 2352, "epoch": 0.9924050632911392, "progress_pct": 16.54, "epoch_pct": 16.54, "eta": "47:18:59", "max_grad_norm": 1.0, "loss": 0.7905668020248413, "grad_norm": 0.432748019695282, "learning_rate": 0.00019578720900430082}
+{"ts": "2025-12-27T03:40:03", "event": "train_log", "step": 2354, "epoch": 0.9932489451476794, "progress_pct": 16.55, "epoch_pct": 16.55, "eta": "47:17:19", "max_grad_norm": 1.0, "loss": 0.8279712200164795, "grad_norm": 0.4116985499858856, "learning_rate": 0.0001957734041912115}
+{"ts": "2025-12-27T03:40:17", "event": "train_log", "step": 2356, "epoch": 0.9940928270042194, "progress_pct": 16.57, "epoch_pct": 16.57, "eta": "47:15:41", "max_grad_norm": 1.0, "loss": 0.7451362609863281, "grad_norm": 0.3927275240421295, "learning_rate": 0.00019575957728503718}
+{"ts": "2025-12-27T03:40:33", "event": "train_log", "step": 2358, "epoch": 0.9949367088607595, "progress_pct": 16.58, "epoch_pct": 16.58, "eta": "47:14:07", "max_grad_norm": 1.0, "loss": 0.6769390106201172, "grad_norm": 0.3389509320259094, "learning_rate": 0.0001957457282889675}
+{"ts": "2025-12-27T03:40:49", "event": "train_log", "step": 2360, "epoch": 0.9957805907172996, "progress_pct": 16.6, "epoch_pct": 16.6, "eta": "47:12:31", "max_grad_norm": 1.0, "loss": 0.7590897083282471, "grad_norm": 0.4220774471759796, "learning_rate": 0.00019573185720619712}
+{"ts": "2025-12-27T03:41:03", "event": "train_log", "step": 2362, "epoch": 0.9966244725738397, "progress_pct": 16.61, "epoch_pct": 16.61, "eta": "47:10:53", "max_grad_norm": 1.0, "loss": 0.8301284313201904, "grad_norm": 0.4299897849559784, "learning_rate": 0.00019571796403992583}
+{"ts": "2025-12-27T03:41:19", "event": "train_log", "step": 2364, "epoch": 0.9974683544303797, "progress_pct": 16.62, "epoch_pct": 16.62, "eta": "47:09:19", "max_grad_norm": 1.0, "loss": 0.8019471764564514, "grad_norm": 0.4031789302825928, "learning_rate": 0.00019570404879335856}
+{"ts": "2025-12-27T03:41:34", "event": "train_log", "step": 2366, "epoch": 0.9983122362869198, "progress_pct": 16.64, "epoch_pct": 16.64, "eta": "47:07:42", "max_grad_norm": 1.0, "loss": 0.7427027821540833, "grad_norm": 0.3957567811012268, "learning_rate": 0.00019569011146970526}
+{"ts": "2025-12-27T03:41:49", "event": "train_log", "step": 2368, "epoch": 0.99915611814346, "progress_pct": 16.65, "epoch_pct": 16.65, "eta": "47:06:04", "max_grad_norm": 1.0, "loss": 0.8221617341041565, "grad_norm": 0.41741347312927246, "learning_rate": 0.000195676152072181}
+{"ts": "2025-12-27T03:42:03", "event": "train_log", "step": 2370, "epoch": 1.0, "progress_pct": 16.67, "epoch_pct": 16.67, "eta": "47:04:25", "max_grad_norm": 1.0, "loss": 0.8143455982208252, "grad_norm": 0.47121551632881165, "learning_rate": 0.00019566217060400595}
+{"ts": "2025-12-27T03:42:18", "event": "train_log", "step": 2372, "epoch": 1.0008438818565402, "progress_pct": 16.68, "epoch_pct": 16.68, "eta": "47:02:48", "max_grad_norm": 1.0, "loss": 0.6971694231033325, "grad_norm": 0.3854716122150421, "learning_rate": 0.0001956481670684054}
+{"ts": "2025-12-27T03:42:33", "event": "train_log", "step": 2374, "epoch": 1.00168776371308, "progress_pct": 16.69, "epoch_pct": 16.69, "eta": "47:01:13", "max_grad_norm": 1.0, "loss": 0.6719155311584473, "grad_norm": 0.4029138386249542, "learning_rate": 0.00019563414146860964}
+{"ts": "2025-12-27T03:42:48", "event": "train_log", "step": 2376, "epoch": 1.0025316455696203, "progress_pct": 16.71, "epoch_pct": 16.71, "eta": "46:59:37", "max_grad_norm": 1.0, "loss": 0.7161872386932373, "grad_norm": 0.40136656165122986, "learning_rate": 0.00019562009380785412}
+{"ts": "2025-12-27T03:43:04", "event": "train_log", "step": 2378, "epoch": 1.0033755274261604, "progress_pct": 16.72, "epoch_pct": 16.72, "eta": "46:58:02", "max_grad_norm": 1.0, "loss": 0.71699059009552, "grad_norm": 0.40685850381851196, "learning_rate": 0.0001956060240893794}
+{"ts": "2025-12-27T03:43:18", "event": "train_log", "step": 2380, "epoch": 1.0042194092827004, "progress_pct": 16.74, "epoch_pct": 16.74, "eta": "46:56:25", "max_grad_norm": 1.0, "loss": 0.8030059337615967, "grad_norm": 0.46153366565704346, "learning_rate": 0.00019559193231643104}
+{"ts": "2025-12-27T03:43:33", "event": "train_log", "step": 2382, "epoch": 1.0050632911392405, "progress_pct": 16.75, "epoch_pct": 16.75, "eta": "46:54:46", "max_grad_norm": 1.0, "loss": 0.7666479349136353, "grad_norm": 0.45048823952674866, "learning_rate": 0.0001955778184922598}
+{"ts": "2025-12-27T03:43:47", "event": "train_log", "step": 2384, "epoch": 1.0059071729957807, "progress_pct": 16.77, "epoch_pct": 16.77, "eta": "46:53:07", "max_grad_norm": 1.0, "loss": 0.7481234073638916, "grad_norm": 0.4260413348674774, "learning_rate": 0.00019556368262012142}
+{"ts": "2025-12-27T03:44:02", "event": "train_log", "step": 2386, "epoch": 1.0067510548523206, "progress_pct": 16.78, "epoch_pct": 16.78, "eta": "46:51:31", "max_grad_norm": 1.0, "loss": 0.7943820953369141, "grad_norm": 0.4819575548171997, "learning_rate": 0.0001955495247032768}
+{"ts": "2025-12-27T03:44:18", "event": "train_log", "step": 2388, "epoch": 1.0075949367088608, "progress_pct": 16.79, "epoch_pct": 16.79, "eta": "46:50:01", "max_grad_norm": 1.0, "loss": 0.6401634812355042, "grad_norm": 0.45789605379104614, "learning_rate": 0.0001955353447449919}
+{"ts": "2025-12-27T03:44:33", "event": "train_log", "step": 2390, "epoch": 1.0084388185654007, "progress_pct": 16.81, "epoch_pct": 16.81, "eta": "46:48:26", "max_grad_norm": 1.0, "loss": 0.7058886885643005, "grad_norm": 0.37019097805023193, "learning_rate": 0.00019552114274853778}
+{"ts": "2025-12-27T03:44:48", "event": "train_log", "step": 2392, "epoch": 1.009282700421941, "progress_pct": 16.82, "epoch_pct": 16.82, "eta": "46:46:48", "max_grad_norm": 1.0, "loss": 0.7631242871284485, "grad_norm": 0.44416263699531555, "learning_rate": 0.0001955069187171905}
+{"ts": "2025-12-27T03:45:02", "event": "train_log", "step": 2394, "epoch": 1.010126582278481, "progress_pct": 16.84, "epoch_pct": 16.84, "eta": "46:45:10", "max_grad_norm": 1.0, "loss": 0.7873892784118652, "grad_norm": 0.4478379786014557, "learning_rate": 0.00019549267265423138}
+{"ts": "2025-12-27T03:45:17", "event": "train_log", "step": 2396, "epoch": 1.010970464135021, "progress_pct": 16.85, "epoch_pct": 16.85, "eta": "46:43:35", "max_grad_norm": 1.0, "loss": 0.7882059216499329, "grad_norm": 0.4596130847930908, "learning_rate": 0.00019547840456294658}
+{"ts": "2025-12-27T03:45:33", "event": "train_log", "step": 2398, "epoch": 1.0118143459915612, "progress_pct": 16.86, "epoch_pct": 16.86, "eta": "46:42:03", "max_grad_norm": 1.0, "loss": 0.6358574628829956, "grad_norm": 0.4062999188899994, "learning_rate": 0.0001954641144466276}
+{"ts": "2025-12-27T03:45:47", "event": "train_log", "step": 2400, "epoch": 1.0126582278481013, "progress_pct": 16.88, "epoch_pct": 16.88, "eta": "46:40:24", "max_grad_norm": 1.0, "loss": 0.7803886532783508, "grad_norm": 0.47627049684524536, "learning_rate": 0.0001954498023085708}
+{"ts": "2025-12-27T03:57:18", "event": "train_log", "step": 2400, "epoch": 1.0126582278481013, "progress_pct": 16.88, "epoch_pct": 16.88, "eta": "47:37:10", "max_grad_norm": 1.0, "eval_loss": 0.8250564932823181, "eval_runtime": 691.5833, "eval_samples_per_second": 3.047, "eval_steps_per_second": 3.047}
+{"ts": "2025-12-27T03:57:35", "event": "train_log", "step": 2402, "epoch": 1.0135021097046413, "progress_pct": 16.89, "epoch_pct": 16.89, "eta": "47:35:41", "max_grad_norm": 1.0, "loss": 0.7394700050354004, "grad_norm": 0.40690696239471436, "learning_rate": 0.00019543546815207777}
+{"ts": "2025-12-27T03:57:50", "event": "train_log", "step": 2404, "epoch": 1.0143459915611814, "progress_pct": 16.91, "epoch_pct": 16.91, "eta": "47:34:04", "max_grad_norm": 1.0, "loss": 0.7096555829048157, "grad_norm": 0.45543980598449707, "learning_rate": 0.0001954211119804551}
+{"ts": "2025-12-27T03:58:06", "event": "train_log", "step": 2406, "epoch": 1.0151898734177216, "progress_pct": 16.92, "epoch_pct": 16.92, "eta": "47:32:29", "max_grad_norm": 1.0, "loss": 0.7961843013763428, "grad_norm": 0.4367314279079437, "learning_rate": 0.00019540673379701454}
+{"ts": "2025-12-27T03:58:21", "event": "train_log", "step": 2408, "epoch": 1.0160337552742615, "progress_pct": 16.93, "epoch_pct": 16.93, "eta": "47:30:51", "max_grad_norm": 1.0, "loss": 0.743843138217926, "grad_norm": 0.4458340108394623, "learning_rate": 0.00019539233360507278}
+{"ts": "2025-12-27T03:58:36", "event": "train_log", "step": 2410, "epoch": 1.0168776371308017, "progress_pct": 16.95, "epoch_pct": 16.95, "eta": "47:29:16", "max_grad_norm": 1.0, "loss": 0.7273756265640259, "grad_norm": 0.3741893470287323, "learning_rate": 0.0001953779114079517}
+{"ts": "2025-12-27T03:58:52", "event": "train_log", "step": 2412, "epoch": 1.0177215189873419, "progress_pct": 16.96, "epoch_pct": 16.96, "eta": "47:27:43", "max_grad_norm": 1.0, "loss": 0.7922506928443909, "grad_norm": 0.4340079426765442, "learning_rate": 0.00019536346720897823}
+{"ts": "2025-12-27T03:59:08", "event": "train_log", "step": 2414, "epoch": 1.0185654008438818, "progress_pct": 16.98, "epoch_pct": 16.98, "eta": "47:26:12", "max_grad_norm": 1.0, "loss": 0.6492037773132324, "grad_norm": 0.4518018662929535, "learning_rate": 0.00019534900101148437}
+{"ts": "2025-12-27T03:59:24", "event": "train_log", "step": 2416, "epoch": 1.019409282700422, "progress_pct": 16.99, "epoch_pct": 16.99, "eta": "47:24:37", "max_grad_norm": 1.0, "loss": 0.7422950863838196, "grad_norm": 0.4160727560520172, "learning_rate": 0.00019533451281880716}
+{"ts": "2025-12-27T03:59:41", "event": "train_log", "step": 2418, "epoch": 1.0202531645569621, "progress_pct": 17.0, "epoch_pct": 17.0, "eta": "47:23:11", "max_grad_norm": 1.0, "loss": 0.6489207148551941, "grad_norm": 0.35953211784362793, "learning_rate": 0.00019532000263428883}
+{"ts": "2025-12-27T03:59:58", "event": "train_log", "step": 2420, "epoch": 1.021097046413502, "progress_pct": 17.02, "epoch_pct": 17.02, "eta": "47:21:43", "max_grad_norm": 1.0, "loss": 0.7449417114257812, "grad_norm": 0.38774433732032776, "learning_rate": 0.0001953054704612765}
+{"ts": "2025-12-27T04:00:13", "event": "train_log", "step": 2422, "epoch": 1.0219409282700422, "progress_pct": 17.03, "epoch_pct": 17.03, "eta": "47:20:09", "max_grad_norm": 1.0, "loss": 0.8481494188308716, "grad_norm": 0.48185858130455017, "learning_rate": 0.0001952909163031225}
+{"ts": "2025-12-27T04:00:29", "event": "train_log", "step": 2424, "epoch": 1.0227848101265822, "progress_pct": 17.05, "epoch_pct": 17.05, "eta": "47:18:34", "max_grad_norm": 1.0, "loss": 0.7442291378974915, "grad_norm": 0.4098491072654724, "learning_rate": 0.00019527634016318425}
+{"ts": "2025-12-27T04:00:45", "event": "train_log", "step": 2426, "epoch": 1.0236286919831223, "progress_pct": 17.06, "epoch_pct": 17.06, "eta": "47:17:02", "max_grad_norm": 1.0, "loss": 0.7564480900764465, "grad_norm": 0.41284480690956116, "learning_rate": 0.00019526174204482412}
+{"ts": "2025-12-27T04:01:00", "event": "train_log", "step": 2428, "epoch": 1.0244725738396625, "progress_pct": 17.07, "epoch_pct": 17.07, "eta": "47:15:28", "max_grad_norm": 1.0, "loss": 0.7234854102134705, "grad_norm": 0.44948479533195496, "learning_rate": 0.00019524712195140965}
+{"ts": "2025-12-27T04:01:15", "event": "train_log", "step": 2430, "epoch": 1.0253164556962024, "progress_pct": 17.09, "epoch_pct": 17.09, "eta": "47:13:52", "max_grad_norm": 1.0, "loss": 0.7638729810714722, "grad_norm": 0.39939233660697937, "learning_rate": 0.0001952324798863134}
+{"ts": "2025-12-27T04:01:31", "event": "train_log", "step": 2432, "epoch": 1.0261603375527426, "progress_pct": 17.1, "epoch_pct": 17.1, "eta": "47:12:22", "max_grad_norm": 1.0, "loss": 0.6922630071640015, "grad_norm": 0.40807482600212097, "learning_rate": 0.00019521781585291298}
+{"ts": "2025-12-27T04:01:46", "event": "train_log", "step": 2434, "epoch": 1.0270042194092828, "progress_pct": 17.12, "epoch_pct": 17.12, "eta": "47:10:47", "max_grad_norm": 1.0, "loss": 0.762996256351471, "grad_norm": 0.45735791325569153, "learning_rate": 0.00019520312985459116}
+{"ts": "2025-12-27T04:02:02", "event": "train_log", "step": 2436, "epoch": 1.0278481012658227, "progress_pct": 17.13, "epoch_pct": 17.13, "eta": "47:09:15", "max_grad_norm": 1.0, "loss": 0.7811596989631653, "grad_norm": 0.4280409812927246, "learning_rate": 0.0001951884218947357}
+{"ts": "2025-12-27T04:02:18", "event": "train_log", "step": 2438, "epoch": 1.0286919831223629, "progress_pct": 17.14, "epoch_pct": 17.14, "eta": "47:07:46", "max_grad_norm": 1.0, "loss": 0.7278305292129517, "grad_norm": 0.41957560181617737, "learning_rate": 0.00019517369197673943}
+{"ts": "2025-12-27T04:02:33", "event": "train_log", "step": 2440, "epoch": 1.029535864978903, "progress_pct": 17.16, "epoch_pct": 17.16, "eta": "47:06:09", "max_grad_norm": 1.0, "loss": 0.807456910610199, "grad_norm": 0.4832586646080017, "learning_rate": 0.00019515894010400028}
+{"ts": "2025-12-27T04:02:48", "event": "train_log", "step": 2442, "epoch": 1.030379746835443, "progress_pct": 17.17, "epoch_pct": 17.17, "eta": "47:04:30", "max_grad_norm": 1.0, "loss": 0.7421302795410156, "grad_norm": 0.46448788046836853, "learning_rate": 0.00019514416627992122}
+{"ts": "2025-12-27T04:03:02", "event": "train_log", "step": 2444, "epoch": 1.0312236286919831, "progress_pct": 17.19, "epoch_pct": 17.19, "eta": "47:02:54", "max_grad_norm": 1.0, "loss": 0.6892213225364685, "grad_norm": 0.41362619400024414, "learning_rate": 0.00019512937050791024}
+{"ts": "2025-12-27T04:03:17", "event": "train_log", "step": 2446, "epoch": 1.0320675105485233, "progress_pct": 17.2, "epoch_pct": 17.2, "eta": "47:01:18", "max_grad_norm": 1.0, "loss": 0.7305331826210022, "grad_norm": 0.4158109128475189, "learning_rate": 0.0001951145527913805}
+{"ts": "2025-12-27T04:03:31", "event": "train_log", "step": 2448, "epoch": 1.0329113924050632, "progress_pct": 17.22, "epoch_pct": 17.22, "eta": "46:59:40", "max_grad_norm": 1.0, "loss": 0.7682753205299377, "grad_norm": 0.4725087881088257, "learning_rate": 0.00019509971313375011}
+{"ts": "2025-12-27T04:03:46", "event": "train_log", "step": 2450, "epoch": 1.0337552742616034, "progress_pct": 17.23, "epoch_pct": 17.23, "eta": "46:58:04", "max_grad_norm": 1.0, "loss": 0.747041642665863, "grad_norm": 0.4341438114643097, "learning_rate": 0.00019508485153844237}
+{"ts": "2025-12-27T04:04:01", "event": "train_log", "step": 2452, "epoch": 1.0345991561181433, "progress_pct": 17.24, "epoch_pct": 17.24, "eta": "46:56:29", "max_grad_norm": 1.0, "loss": 0.7312062382698059, "grad_norm": 0.44515377283096313, "learning_rate": 0.00019506996800888548}
+{"ts": "2025-12-27T04:04:16", "event": "train_log", "step": 2454, "epoch": 1.0354430379746835, "progress_pct": 17.26, "epoch_pct": 17.26, "eta": "46:54:52", "max_grad_norm": 1.0, "loss": 0.7861107587814331, "grad_norm": 0.5313915014266968, "learning_rate": 0.00019505506254851283}
+{"ts": "2025-12-27T04:04:30", "event": "train_log", "step": 2456, "epoch": 1.0362869198312237, "progress_pct": 17.27, "epoch_pct": 17.27, "eta": "46:53:14", "max_grad_norm": 1.0, "loss": 0.8458115458488464, "grad_norm": 0.6094246506690979, "learning_rate": 0.00019504013516076285}
+{"ts": "2025-12-27T04:04:46", "event": "train_log", "step": 2458, "epoch": 1.0371308016877636, "progress_pct": 17.29, "epoch_pct": 17.29, "eta": "46:51:45", "max_grad_norm": 1.0, "loss": 0.6091256141662598, "grad_norm": 0.36419153213500977, "learning_rate": 0.00019502518584907895}
+{"ts": "2025-12-27T04:05:02", "event": "train_log", "step": 2460, "epoch": 1.0379746835443038, "progress_pct": 17.3, "epoch_pct": 17.3, "eta": "46:50:14", "max_grad_norm": 1.0, "loss": 0.6703370809555054, "grad_norm": 0.399271696805954, "learning_rate": 0.00019501021461690966}
+{"ts": "2025-12-27T04:05:17", "event": "train_log", "step": 2462, "epoch": 1.038818565400844, "progress_pct": 17.31, "epoch_pct": 17.31, "eta": "46:48:42", "max_grad_norm": 1.0, "loss": 0.7220943570137024, "grad_norm": 0.38560932874679565, "learning_rate": 0.00019499522146770856}
+{"ts": "2025-12-27T04:05:32", "event": "train_log", "step": 2464, "epoch": 1.0396624472573839, "progress_pct": 17.33, "epoch_pct": 17.33, "eta": "46:47:08", "max_grad_norm": 1.0, "loss": 0.7501406073570251, "grad_norm": 0.44199803471565247, "learning_rate": 0.00019498020640493428}
+{"ts": "2025-12-27T04:05:46", "event": "train_log", "step": 2466, "epoch": 1.040506329113924, "progress_pct": 17.34, "epoch_pct": 17.34, "eta": "46:45:31", "max_grad_norm": 1.0, "loss": 0.7957925796508789, "grad_norm": 0.47394421696662903, "learning_rate": 0.00019496516943205054}
+{"ts": "2025-12-27T04:06:01", "event": "train_log", "step": 2468, "epoch": 1.0413502109704642, "progress_pct": 17.36, "epoch_pct": 17.36, "eta": "46:43:55", "max_grad_norm": 1.0, "loss": 0.7964824438095093, "grad_norm": 0.5274425745010376, "learning_rate": 0.00019495011055252603}
+{"ts": "2025-12-27T04:06:16", "event": "train_log", "step": 2470, "epoch": 1.0421940928270041, "progress_pct": 17.37, "epoch_pct": 17.37, "eta": "46:42:22", "max_grad_norm": 1.0, "loss": 0.7478582262992859, "grad_norm": 0.44004881381988525, "learning_rate": 0.0001949350297698346}
+{"ts": "2025-12-27T04:06:32", "event": "train_log", "step": 2472, "epoch": 1.0430379746835443, "progress_pct": 17.38, "epoch_pct": 17.38, "eta": "46:40:52", "max_grad_norm": 1.0, "loss": 0.738339900970459, "grad_norm": 0.40662670135498047, "learning_rate": 0.000194919927087455}
+{"ts": "2025-12-27T04:06:47", "event": "train_log", "step": 2474, "epoch": 1.0438818565400845, "progress_pct": 17.4, "epoch_pct": 17.4, "eta": "46:39:19", "max_grad_norm": 1.0, "loss": 0.817171037197113, "grad_norm": 0.43934181332588196, "learning_rate": 0.00019490480250887123}
+{"ts": "2025-12-27T04:07:01", "event": "train_log", "step": 2476, "epoch": 1.0447257383966244, "progress_pct": 17.41, "epoch_pct": 17.41, "eta": "46:37:42", "max_grad_norm": 1.0, "loss": 0.7413169145584106, "grad_norm": 0.4111221432685852, "learning_rate": 0.00019488965603757223}
+{"ts": "2025-12-27T04:07:16", "event": "train_log", "step": 2478, "epoch": 1.0455696202531646, "progress_pct": 17.43, "epoch_pct": 17.43, "eta": "46:36:12", "max_grad_norm": 1.0, "loss": 0.7893102169036865, "grad_norm": 0.3770996630191803, "learning_rate": 0.00019487448767705194}
+{"ts": "2025-12-27T04:07:31", "event": "train_log", "step": 2480, "epoch": 1.0464135021097047, "progress_pct": 17.44, "epoch_pct": 17.44, "eta": "46:34:38", "max_grad_norm": 1.0, "loss": 0.7037841081619263, "grad_norm": 0.4524461627006531, "learning_rate": 0.00019485929743080945}
+{"ts": "2025-12-27T04:07:46", "event": "train_log", "step": 2482, "epoch": 1.0472573839662447, "progress_pct": 17.45, "epoch_pct": 17.45, "eta": "46:33:04", "max_grad_norm": 1.0, "loss": 0.7456525564193726, "grad_norm": 0.4232378900051117, "learning_rate": 0.00019484408530234885}
+{"ts": "2025-12-27T04:08:02", "event": "train_log", "step": 2484, "epoch": 1.0481012658227848, "progress_pct": 17.47, "epoch_pct": 17.47, "eta": "46:31:36", "max_grad_norm": 1.0, "loss": 0.7263808250427246, "grad_norm": 0.3842810392379761, "learning_rate": 0.00019482885129517929}
+{"ts": "2025-12-27T04:08:17", "event": "train_log", "step": 2486, "epoch": 1.048945147679325, "progress_pct": 17.48, "epoch_pct": 17.48, "eta": "46:30:03", "max_grad_norm": 1.0, "loss": 0.7101474404335022, "grad_norm": 0.43203726410865784, "learning_rate": 0.00019481359541281495}
+{"ts": "2025-12-27T04:08:32", "event": "train_log", "step": 2488, "epoch": 1.049789029535865, "progress_pct": 17.5, "epoch_pct": 17.5, "eta": "46:28:30", "max_grad_norm": 1.0, "loss": 0.8561921715736389, "grad_norm": 0.46944454312324524, "learning_rate": 0.00019479831765877509}
+{"ts": "2025-12-27T04:08:47", "event": "train_log", "step": 2490, "epoch": 1.0506329113924051, "progress_pct": 17.51, "epoch_pct": 17.51, "eta": "46:26:58", "max_grad_norm": 1.0, "loss": 0.7206679582595825, "grad_norm": 0.4455019533634186, "learning_rate": 0.00019478301803658397}
+{"ts": "2025-12-27T04:09:02", "event": "train_log", "step": 2492, "epoch": 1.051476793248945, "progress_pct": 17.52, "epoch_pct": 17.52, "eta": "46:25:26", "max_grad_norm": 1.0, "loss": 0.7271745800971985, "grad_norm": 0.4253813326358795, "learning_rate": 0.00019476769654977095}
+{"ts": "2025-12-27T04:09:16", "event": "train_log", "step": 2494, "epoch": 1.0523206751054852, "progress_pct": 17.54, "epoch_pct": 17.54, "eta": "46:23:52", "max_grad_norm": 1.0, "loss": 0.7358576059341431, "grad_norm": 0.42758429050445557, "learning_rate": 0.00019475235320187033}
+{"ts": "2025-12-27T04:09:33", "event": "train_log", "step": 2496, "epoch": 1.0531645569620254, "progress_pct": 17.55, "epoch_pct": 17.55, "eta": "46:22:27", "max_grad_norm": 1.0, "loss": 0.6617647409439087, "grad_norm": 0.400634765625, "learning_rate": 0.00019473698799642164}
+{"ts": "2025-12-27T04:09:48", "event": "train_log", "step": 2498, "epoch": 1.0540084388185653, "progress_pct": 17.57, "epoch_pct": 17.57, "eta": "46:20:55", "max_grad_norm": 1.0, "loss": 0.6900970935821533, "grad_norm": 0.4132104814052582, "learning_rate": 0.00019472160093696926}
+{"ts": "2025-12-27T04:10:02", "event": "train_log", "step": 2500, "epoch": 1.0548523206751055, "progress_pct": 17.58, "epoch_pct": 17.58, "eta": "46:19:22", "max_grad_norm": 1.0, "loss": 0.7352861762046814, "grad_norm": 0.5116495490074158, "learning_rate": 0.00019470619202706268}
+{"ts": "2025-12-27T04:21:22", "event": "train_log", "step": 2500, "epoch": 1.0548523206751055, "progress_pct": 17.58, "epoch_pct": 17.58, "eta": "47:12:27", "max_grad_norm": 1.0, "eval_loss": 0.8249453902244568, "eval_runtime": 679.4446, "eval_samples_per_second": 3.101, "eval_steps_per_second": 3.101}
+{"ts": "2025-12-27T04:21:37", "event": "train_log", "step": 2502, "epoch": 1.0556962025316456, "progress_pct": 17.59, "epoch_pct": 17.59, "eta": "47:10:55", "max_grad_norm": 1.0, "loss": 0.7583479881286621, "grad_norm": 0.42822444438934326, "learning_rate": 0.0001946907612702565}
+{"ts": "2025-12-27T04:21:53", "event": "train_log", "step": 2504, "epoch": 1.0565400843881856, "progress_pct": 17.61, "epoch_pct": 17.61, "eta": "47:09:22", "max_grad_norm": 1.0, "loss": 0.777511477470398, "grad_norm": 0.4788980782032013, "learning_rate": 0.00019467530867011025}
+{"ts": "2025-12-27T04:22:09", "event": "train_log", "step": 2506, "epoch": 1.0573839662447257, "progress_pct": 17.62, "epoch_pct": 17.62, "eta": "47:07:52", "max_grad_norm": 1.0, "loss": 0.7452852129936218, "grad_norm": 0.4210498034954071, "learning_rate": 0.00019465983423018856}
+{"ts": "2025-12-27T04:22:24", "event": "train_log", "step": 2508, "epoch": 1.058227848101266, "progress_pct": 17.64, "epoch_pct": 17.64, "eta": "47:06:19", "max_grad_norm": 1.0, "loss": 0.7540958523750305, "grad_norm": 0.43219906091690063, "learning_rate": 0.0001946443379540611}
+{"ts": "2025-12-27T04:22:39", "event": "train_log", "step": 2510, "epoch": 1.0590717299578059, "progress_pct": 17.65, "epoch_pct": 17.65, "eta": "47:04:47", "max_grad_norm": 1.0, "loss": 0.7175624966621399, "grad_norm": 0.4361002445220947, "learning_rate": 0.00019462881984530253}
+{"ts": "2025-12-27T04:22:55", "event": "train_log", "step": 2512, "epoch": 1.059915611814346, "progress_pct": 17.67, "epoch_pct": 17.67, "eta": "47:03:15", "max_grad_norm": 1.0, "loss": 0.7342076301574707, "grad_norm": 0.4575091600418091, "learning_rate": 0.00019461327990749264}
+{"ts": "2025-12-27T04:23:10", "event": "train_log", "step": 2514, "epoch": 1.0607594936708862, "progress_pct": 17.68, "epoch_pct": 17.68, "eta": "47:01:43", "max_grad_norm": 1.0, "loss": 0.8255488872528076, "grad_norm": 0.4823324978351593, "learning_rate": 0.00019459771814421615}
+{"ts": "2025-12-27T04:23:26", "event": "train_log", "step": 2516, "epoch": 1.0616033755274261, "progress_pct": 17.69, "epoch_pct": 17.69, "eta": "47:00:12", "max_grad_norm": 1.0, "loss": 0.6624729633331299, "grad_norm": 0.41693785786628723, "learning_rate": 0.00019458213455906284}
+{"ts": "2025-12-27T04:23:41", "event": "train_log", "step": 2518, "epoch": 1.0624472573839663, "progress_pct": 17.71, "epoch_pct": 17.71, "eta": "46:58:41", "max_grad_norm": 1.0, "loss": 0.8224266767501831, "grad_norm": 0.41319167613983154, "learning_rate": 0.00019456652915562762}
+{"ts": "2025-12-27T04:23:57", "event": "train_log", "step": 2520, "epoch": 1.0632911392405062, "progress_pct": 17.72, "epoch_pct": 17.72, "eta": "46:57:11", "max_grad_norm": 1.0, "loss": 0.7582900524139404, "grad_norm": 0.4405190050601959, "learning_rate": 0.00019455090193751023}
+{"ts": "2025-12-27T04:24:13", "event": "train_log", "step": 2522, "epoch": 1.0641350210970464, "progress_pct": 17.74, "epoch_pct": 17.74, "eta": "46:55:40", "max_grad_norm": 1.0, "loss": 0.7948101162910461, "grad_norm": 0.4564387798309326, "learning_rate": 0.0001945352529083157}
+{"ts": "2025-12-27T04:24:28", "event": "train_log", "step": 2524, "epoch": 1.0649789029535865, "progress_pct": 17.75, "epoch_pct": 17.75, "eta": "46:54:09", "max_grad_norm": 1.0, "loss": 0.7670508027076721, "grad_norm": 0.4247225224971771, "learning_rate": 0.00019451958207165394}
+{"ts": "2025-12-27T04:24:43", "event": "train_log", "step": 2526, "epoch": 1.0658227848101265, "progress_pct": 17.76, "epoch_pct": 17.76, "eta": "46:52:37", "max_grad_norm": 1.0, "loss": 0.7753698229789734, "grad_norm": 0.42070862650871277, "learning_rate": 0.00019450388943113983}
+{"ts": "2025-12-27T04:24:59", "event": "train_log", "step": 2528, "epoch": 1.0666666666666667, "progress_pct": 17.78, "epoch_pct": 17.78, "eta": "46:51:09", "max_grad_norm": 1.0, "loss": 0.710904598236084, "grad_norm": 0.3729179799556732, "learning_rate": 0.00019448817499039342}
+{"ts": "2025-12-27T04:25:16", "event": "train_log", "step": 2530, "epoch": 1.0675105485232068, "progress_pct": 17.79, "epoch_pct": 17.79, "eta": "46:49:44", "max_grad_norm": 1.0, "loss": 0.7000213265419006, "grad_norm": 0.40901753306388855, "learning_rate": 0.0001944724387530397}
+{"ts": "2025-12-27T04:25:32", "event": "train_log", "step": 2532, "epoch": 1.0683544303797468, "progress_pct": 17.81, "epoch_pct": 17.81, "eta": "46:48:14", "max_grad_norm": 1.0, "loss": 0.7374410629272461, "grad_norm": 0.4481421113014221, "learning_rate": 0.00019445668072270877}
+{"ts": "2025-12-27T04:25:47", "event": "train_log", "step": 2534, "epoch": 1.069198312236287, "progress_pct": 17.82, "epoch_pct": 17.82, "eta": "46:46:43", "max_grad_norm": 1.0, "loss": 0.746309220790863, "grad_norm": 0.4839915335178375, "learning_rate": 0.00019444090090303567}
+{"ts": "2025-12-27T04:26:03", "event": "train_log", "step": 2536, "epoch": 1.070042194092827, "progress_pct": 17.83, "epoch_pct": 17.83, "eta": "46:45:14", "max_grad_norm": 1.0, "loss": 0.7448083162307739, "grad_norm": 0.4982132613658905, "learning_rate": 0.00019442509929766048}
+{"ts": "2025-12-27T04:26:19", "event": "train_log", "step": 2538, "epoch": 1.070886075949367, "progress_pct": 17.85, "epoch_pct": 17.85, "eta": "46:43:48", "max_grad_norm": 1.0, "loss": 0.7052555680274963, "grad_norm": 0.3987475037574768, "learning_rate": 0.00019440927591022838}
+{"ts": "2025-12-27T04:26:34", "event": "train_log", "step": 2540, "epoch": 1.0717299578059072, "progress_pct": 17.86, "epoch_pct": 17.86, "eta": "46:42:17", "max_grad_norm": 1.0, "loss": 0.7090281248092651, "grad_norm": 0.4089896082878113, "learning_rate": 0.00019439343074438944}
+{"ts": "2025-12-27T04:26:50", "event": "train_log", "step": 2542, "epoch": 1.0725738396624473, "progress_pct": 17.88, "epoch_pct": 17.88, "eta": "46:40:47", "max_grad_norm": 1.0, "loss": 0.7530601024627686, "grad_norm": 0.3703136146068573, "learning_rate": 0.00019437756380379895}
+{"ts": "2025-12-27T04:27:05", "event": "train_log", "step": 2544, "epoch": 1.0734177215189873, "progress_pct": 17.89, "epoch_pct": 17.89, "eta": "46:39:15", "max_grad_norm": 1.0, "loss": 0.7597941160202026, "grad_norm": 0.4015701413154602, "learning_rate": 0.00019436167509211702}
+{"ts": "2025-12-27T04:27:23", "event": "train_log", "step": 2546, "epoch": 1.0742616033755275, "progress_pct": 17.9, "epoch_pct": 17.9, "eta": "46:37:56", "max_grad_norm": 1.0, "loss": 0.7168124318122864, "grad_norm": 0.3915148675441742, "learning_rate": 0.00019434576461300888}
+{"ts": "2025-12-27T04:27:38", "event": "train_log", "step": 2548, "epoch": 1.0751054852320676, "progress_pct": 17.92, "epoch_pct": 17.92, "eta": "46:36:28", "max_grad_norm": 1.0, "loss": 0.7762725949287415, "grad_norm": 0.44643113017082214, "learning_rate": 0.00019432983237014482}
+{"ts": "2025-12-27T04:27:54", "event": "train_log", "step": 2550, "epoch": 1.0759493670886076, "progress_pct": 17.93, "epoch_pct": 17.93, "eta": "46:34:59", "max_grad_norm": 1.0, "loss": 0.7879725098609924, "grad_norm": 0.46874719858169556, "learning_rate": 0.00019431387836720004}
+{"ts": "2025-12-27T04:28:10", "event": "train_log", "step": 2552, "epoch": 1.0767932489451477, "progress_pct": 17.95, "epoch_pct": 17.95, "eta": "46:33:32", "max_grad_norm": 1.0, "loss": 0.7351391315460205, "grad_norm": 0.4143001437187195, "learning_rate": 0.00019429790260785487}
+{"ts": "2025-12-27T04:28:27", "event": "train_log", "step": 2554, "epoch": 1.0776371308016879, "progress_pct": 17.96, "epoch_pct": 17.96, "eta": "46:32:09", "max_grad_norm": 1.0, "loss": 0.86613929271698, "grad_norm": 0.4569888412952423, "learning_rate": 0.0001942819050957946}
+{"ts": "2025-12-27T04:28:43", "event": "train_log", "step": 2556, "epoch": 1.0784810126582278, "progress_pct": 17.97, "epoch_pct": 17.97, "eta": "46:30:44", "max_grad_norm": 1.0, "loss": 0.7122719287872314, "grad_norm": 0.41729897260665894, "learning_rate": 0.00019426588583470954}
+{"ts": "2025-12-27T04:28:59", "event": "train_log", "step": 2558, "epoch": 1.079324894514768, "progress_pct": 17.99, "epoch_pct": 17.99, "eta": "46:29:18", "max_grad_norm": 1.0, "loss": 0.6605332493782043, "grad_norm": 0.42295604944229126, "learning_rate": 0.000194249844828295}
+{"ts": "2025-12-27T04:29:14", "event": "train_log", "step": 2560, "epoch": 1.080168776371308, "progress_pct": 18.0, "epoch_pct": 18.0, "eta": "46:27:43", "max_grad_norm": 1.0, "loss": 0.7839207649230957, "grad_norm": 0.45787641406059265, "learning_rate": 0.00019423378208025137}
+{"ts": "2025-12-27T04:29:29", "event": "train_log", "step": 2562, "epoch": 1.081012658227848, "progress_pct": 18.02, "epoch_pct": 18.02, "eta": "46:26:15", "max_grad_norm": 1.0, "loss": 0.7557987570762634, "grad_norm": 0.43370622396469116, "learning_rate": 0.00019421769759428404}
+{"ts": "2025-12-27T04:29:45", "event": "train_log", "step": 2564, "epoch": 1.0818565400843883, "progress_pct": 18.03, "epoch_pct": 18.03, "eta": "46:24:46", "max_grad_norm": 1.0, "loss": 0.7349345088005066, "grad_norm": 0.4371975362300873, "learning_rate": 0.00019420159137410335}
+{"ts": "2025-12-27T04:30:01", "event": "train_log", "step": 2566, "epoch": 1.0827004219409282, "progress_pct": 18.05, "epoch_pct": 18.05, "eta": "46:23:22", "max_grad_norm": 1.0, "loss": 0.7024962902069092, "grad_norm": 0.468730628490448, "learning_rate": 0.0001941854634234247}
+{"ts": "2025-12-27T04:30:18", "event": "train_log", "step": 2568, "epoch": 1.0835443037974684, "progress_pct": 18.06, "epoch_pct": 18.06, "eta": "46:22:00", "max_grad_norm": 1.0, "loss": 0.6981064081192017, "grad_norm": 0.4257284998893738, "learning_rate": 0.0001941693137459685}
+{"ts": "2025-12-27T04:30:34", "event": "train_log", "step": 2570, "epoch": 1.0843881856540085, "progress_pct": 18.07, "epoch_pct": 18.07, "eta": "46:20:32", "max_grad_norm": 1.0, "loss": 0.7899637818336487, "grad_norm": 0.4322902262210846, "learning_rate": 0.00019415314234546015}
+{"ts": "2025-12-27T04:30:49", "event": "train_log", "step": 2572, "epoch": 1.0852320675105485, "progress_pct": 18.09, "epoch_pct": 18.09, "eta": "46:19:02", "max_grad_norm": 1.0, "loss": 0.7964152693748474, "grad_norm": 0.44384217262268066, "learning_rate": 0.00019413694922563014}
+{"ts": "2025-12-27T04:31:05", "event": "train_log", "step": 2574, "epoch": 1.0860759493670886, "progress_pct": 18.1, "epoch_pct": 18.1, "eta": "46:17:36", "max_grad_norm": 1.0, "loss": 0.7902941107749939, "grad_norm": 0.47391360998153687, "learning_rate": 0.0001941207343902139}
+{"ts": "2025-12-27T04:31:20", "event": "train_log", "step": 2576, "epoch": 1.0869198312236288, "progress_pct": 18.12, "epoch_pct": 18.12, "eta": "46:16:08", "max_grad_norm": 1.0, "loss": 0.7818595170974731, "grad_norm": 0.4575883746147156, "learning_rate": 0.00019410449784295182}
+{"ts": "2025-12-27T04:31:36", "event": "train_log", "step": 2578, "epoch": 1.0877637130801687, "progress_pct": 18.13, "epoch_pct": 18.13, "eta": "46:14:40", "max_grad_norm": 1.0, "loss": 0.7516021728515625, "grad_norm": 0.41236087679862976, "learning_rate": 0.0001940882395875894}
+{"ts": "2025-12-27T04:31:53", "event": "train_log", "step": 2580, "epoch": 1.0886075949367089, "progress_pct": 18.14, "epoch_pct": 18.14, "eta": "46:13:21", "max_grad_norm": 1.0, "loss": 0.6870401501655579, "grad_norm": 0.3817494213581085, "learning_rate": 0.0001940719596278771}
+{"ts": "2025-12-27T04:32:08", "event": "train_log", "step": 2582, "epoch": 1.0894514767932488, "progress_pct": 18.16, "epoch_pct": 18.16, "eta": "46:11:52", "max_grad_norm": 1.0, "loss": 0.7437341809272766, "grad_norm": 0.4630168378353119, "learning_rate": 0.00019405565796757043}
+{"ts": "2025-12-27T04:32:24", "event": "train_log", "step": 2584, "epoch": 1.090295358649789, "progress_pct": 18.17, "epoch_pct": 18.17, "eta": "46:10:26", "max_grad_norm": 1.0, "loss": 0.7101680040359497, "grad_norm": 0.4063820540904999, "learning_rate": 0.00019403933461042982}
+{"ts": "2025-12-27T04:32:39", "event": "train_log", "step": 2586, "epoch": 1.0911392405063292, "progress_pct": 18.19, "epoch_pct": 18.19, "eta": "46:08:58", "max_grad_norm": 1.0, "loss": 0.6894780397415161, "grad_norm": 0.44423359632492065, "learning_rate": 0.0001940229895602208}
+{"ts": "2025-12-27T04:32:55", "event": "train_log", "step": 2588, "epoch": 1.091983122362869, "progress_pct": 18.2, "epoch_pct": 18.2, "eta": "46:07:31", "max_grad_norm": 1.0, "loss": 0.7277120351791382, "grad_norm": 0.4267353117465973, "learning_rate": 0.0001940066228207138}
+{"ts": "2025-12-27T04:33:12", "event": "train_log", "step": 2590, "epoch": 1.0928270042194093, "progress_pct": 18.21, "epoch_pct": 18.21, "eta": "46:06:12", "max_grad_norm": 1.0, "loss": 0.6512353420257568, "grad_norm": 0.44747647643089294, "learning_rate": 0.00019399023439568434}
+{"ts": "2025-12-27T04:33:27", "event": "train_log", "step": 2592, "epoch": 1.0936708860759494, "progress_pct": 18.23, "epoch_pct": 18.23, "eta": "46:04:41", "max_grad_norm": 1.0, "loss": 0.7850068807601929, "grad_norm": 0.4557100832462311, "learning_rate": 0.00019397382428891296}
+{"ts": "2025-12-27T04:33:42", "event": "train_log", "step": 2594, "epoch": 1.0945147679324894, "progress_pct": 18.24, "epoch_pct": 18.24, "eta": "46:03:12", "max_grad_norm": 1.0, "loss": 0.786868691444397, "grad_norm": 0.44980114698410034, "learning_rate": 0.0001939573925041851}
+{"ts": "2025-12-27T04:33:57", "event": "train_log", "step": 2596, "epoch": 1.0953586497890295, "progress_pct": 18.26, "epoch_pct": 18.26, "eta": "46:01:42", "max_grad_norm": 1.0, "loss": 0.8260899186134338, "grad_norm": 0.45374399423599243, "learning_rate": 0.00019394093904529126}
+{"ts": "2025-12-27T04:34:11", "event": "train_log", "step": 2598, "epoch": 1.0962025316455697, "progress_pct": 18.27, "epoch_pct": 18.27, "eta": "46:00:10", "max_grad_norm": 1.0, "loss": 0.7056689262390137, "grad_norm": 0.40681353211402893, "learning_rate": 0.00019392446391602695}
+{"ts": "2025-12-27T04:34:26", "event": "train_log", "step": 2600, "epoch": 1.0970464135021096, "progress_pct": 18.28, "epoch_pct": 18.28, "eta": "45:58:42", "max_grad_norm": 1.0, "loss": 0.7557252645492554, "grad_norm": 0.39278075098991394, "learning_rate": 0.0001939079671201927}
+{"ts": "2025-12-27T04:45:36", "event": "train_log", "step": 2600, "epoch": 1.0970464135021096, "progress_pct": 18.28, "epoch_pct": 18.28, "eta": "46:48:37", "max_grad_norm": 1.0, "eval_loss": 0.8211485743522644, "eval_runtime": 670.2276, "eval_samples_per_second": 3.144, "eval_steps_per_second": 3.144}
+{"ts": "2025-12-27T04:45:51", "event": "train_log", "step": 2602, "epoch": 1.0978902953586498, "progress_pct": 18.3, "epoch_pct": 18.3, "eta": "46:47:03", "max_grad_norm": 1.0, "loss": 0.7139819264411926, "grad_norm": 0.4261230230331421, "learning_rate": 0.00019389144866159398}
+{"ts": "2025-12-27T04:46:06", "event": "train_log", "step": 2604, "epoch": 1.09873417721519, "progress_pct": 18.31, "epoch_pct": 18.31, "eta": "46:45:31", "max_grad_norm": 1.0, "loss": 0.7274705767631531, "grad_norm": 0.4664439558982849, "learning_rate": 0.0001938749085440412}
+{"ts": "2025-12-27T04:46:20", "event": "train_log", "step": 2606, "epoch": 1.09957805907173, "progress_pct": 18.33, "epoch_pct": 18.33, "eta": "46:43:58", "max_grad_norm": 1.0, "loss": 0.7547051906585693, "grad_norm": 0.46306154131889343, "learning_rate": 0.00019385834677134998}
+{"ts": "2025-12-27T04:46:35", "event": "train_log", "step": 2608, "epoch": 1.10042194092827, "progress_pct": 18.34, "epoch_pct": 18.34, "eta": "46:42:27", "max_grad_norm": 1.0, "loss": 0.6942694187164307, "grad_norm": 0.39753687381744385, "learning_rate": 0.00019384176334734073}
+{"ts": "2025-12-27T04:46:50", "event": "train_log", "step": 2610, "epoch": 1.1012658227848102, "progress_pct": 18.35, "epoch_pct": 18.35, "eta": "46:40:54", "max_grad_norm": 1.0, "loss": 0.7161701321601868, "grad_norm": 0.4556522071361542, "learning_rate": 0.00019382515827583898}
+{"ts": "2025-12-27T04:47:06", "event": "train_log", "step": 2612, "epoch": 1.1021097046413502, "progress_pct": 18.37, "epoch_pct": 18.37, "eta": "46:39:29", "max_grad_norm": 1.0, "loss": 0.7165710926055908, "grad_norm": 0.3863290250301361, "learning_rate": 0.0001938085315606751}
+{"ts": "2025-12-27T04:47:21", "event": "train_log", "step": 2614, "epoch": 1.1029535864978903, "progress_pct": 18.38, "epoch_pct": 18.38, "eta": "46:37:58", "max_grad_norm": 1.0, "loss": 0.8080679774284363, "grad_norm": 0.49601292610168457, "learning_rate": 0.00019379188320568466}
+{"ts": "2025-12-27T04:47:37", "event": "train_log", "step": 2616, "epoch": 1.1037974683544305, "progress_pct": 18.4, "epoch_pct": 18.4, "eta": "46:36:31", "max_grad_norm": 1.0, "loss": 0.7032502889633179, "grad_norm": 0.4498465955257416, "learning_rate": 0.00019377521321470805}
+{"ts": "2025-12-27T04:47:54", "event": "train_log", "step": 2618, "epoch": 1.1046413502109704, "progress_pct": 18.41, "epoch_pct": 18.41, "eta": "46:35:11", "max_grad_norm": 1.0, "loss": 0.7076731324195862, "grad_norm": 0.4537963569164276, "learning_rate": 0.00019375852159159073}
+{"ts": "2025-12-27T04:48:12", "event": "train_log", "step": 2620, "epoch": 1.1054852320675106, "progress_pct": 18.42, "epoch_pct": 18.42, "eta": "46:33:50", "max_grad_norm": 1.0, "loss": 0.7020825743675232, "grad_norm": 0.4042968153953552, "learning_rate": 0.0001937418083401832}
+{"ts": "2025-12-27T04:48:27", "event": "train_log", "step": 2622, "epoch": 1.1063291139240505, "progress_pct": 18.44, "epoch_pct": 18.44, "eta": "46:32:20", "max_grad_norm": 1.0, "loss": 0.7222874760627747, "grad_norm": 0.44306516647338867, "learning_rate": 0.00019372507346434076}
+{"ts": "2025-12-27T04:48:44", "event": "train_log", "step": 2624, "epoch": 1.1071729957805907, "progress_pct": 18.45, "epoch_pct": 18.45, "eta": "46:30:58", "max_grad_norm": 1.0, "loss": 0.6136211156845093, "grad_norm": 0.3792516887187958, "learning_rate": 0.00019370831696792392}
+{"ts": "2025-12-27T04:49:00", "event": "train_log", "step": 2626, "epoch": 1.1080168776371309, "progress_pct": 18.47, "epoch_pct": 18.47, "eta": "46:29:33", "max_grad_norm": 1.0, "loss": 0.7358977198600769, "grad_norm": 0.46655353903770447, "learning_rate": 0.00019369153885479806}
+{"ts": "2025-12-27T04:49:15", "event": "train_log", "step": 2628, "epoch": 1.1088607594936708, "progress_pct": 18.48, "epoch_pct": 18.48, "eta": "46:28:05", "max_grad_norm": 1.0, "loss": 0.7484551072120667, "grad_norm": 0.4416999816894531, "learning_rate": 0.0001936747391288336}
+{"ts": "2025-12-27T04:49:30", "event": "train_log", "step": 2630, "epoch": 1.109704641350211, "progress_pct": 18.5, "epoch_pct": 18.5, "eta": "46:26:35", "max_grad_norm": 1.0, "loss": 0.6874802708625793, "grad_norm": 0.4620150327682495, "learning_rate": 0.00019365791779390584}
+{"ts": "2025-12-27T04:49:46", "event": "train_log", "step": 2632, "epoch": 1.1105485232067511, "progress_pct": 18.51, "epoch_pct": 18.51, "eta": "46:25:09", "max_grad_norm": 1.0, "loss": 0.7168865203857422, "grad_norm": 0.4357883334159851, "learning_rate": 0.00019364107485389518}
+{"ts": "2025-12-27T04:50:01", "event": "train_log", "step": 2634, "epoch": 1.111392405063291, "progress_pct": 18.52, "epoch_pct": 18.52, "eta": "46:23:41", "max_grad_norm": 1.0, "loss": 0.7533737421035767, "grad_norm": 0.4581143856048584, "learning_rate": 0.00019362421031268699}
+{"ts": "2025-12-27T04:50:17", "event": "train_log", "step": 2636, "epoch": 1.1122362869198312, "progress_pct": 18.54, "epoch_pct": 18.54, "eta": "46:22:13", "max_grad_norm": 1.0, "loss": 0.8245398998260498, "grad_norm": 0.46408653259277344, "learning_rate": 0.00019360732417417152}
+{"ts": "2025-12-27T04:50:32", "event": "train_log", "step": 2638, "epoch": 1.1130801687763714, "progress_pct": 18.55, "epoch_pct": 18.55, "eta": "46:20:45", "max_grad_norm": 1.0, "loss": 0.7095663547515869, "grad_norm": 0.4639282524585724, "learning_rate": 0.00019359041644224415}
+{"ts": "2025-12-27T04:50:49", "event": "train_log", "step": 2640, "epoch": 1.1139240506329113, "progress_pct": 18.57, "epoch_pct": 18.57, "eta": "46:19:22", "max_grad_norm": 1.0, "loss": 0.7030285596847534, "grad_norm": 0.4598273038864136, "learning_rate": 0.00019357348712080517}
+{"ts": "2025-12-27T04:51:04", "event": "train_log", "step": 2642, "epoch": 1.1147679324894515, "progress_pct": 18.58, "epoch_pct": 18.58, "eta": "46:17:55", "max_grad_norm": 1.0, "loss": 0.7158969640731812, "grad_norm": 0.42993804812431335, "learning_rate": 0.00019355653621375978}
+{"ts": "2025-12-27T04:51:22", "event": "train_log", "step": 2644, "epoch": 1.1156118143459917, "progress_pct": 18.59, "epoch_pct": 18.59, "eta": "46:16:38", "max_grad_norm": 1.0, "loss": 0.6800820231437683, "grad_norm": 0.36132168769836426, "learning_rate": 0.0001935395637250183}
+{"ts": "2025-12-27T04:51:37", "event": "train_log", "step": 2646, "epoch": 1.1164556962025316, "progress_pct": 18.61, "epoch_pct": 18.61, "eta": "46:15:09", "max_grad_norm": 1.0, "loss": 0.694972813129425, "grad_norm": 0.5004428029060364, "learning_rate": 0.00019352256965849592}
+{"ts": "2025-12-27T04:51:54", "event": "train_log", "step": 2648, "epoch": 1.1172995780590718, "progress_pct": 18.62, "epoch_pct": 18.62, "eta": "46:13:48", "max_grad_norm": 1.0, "loss": 0.7164262533187866, "grad_norm": 0.43715354800224304, "learning_rate": 0.00019350555401811286}
+{"ts": "2025-12-27T04:52:09", "event": "train_log", "step": 2650, "epoch": 1.1181434599156117, "progress_pct": 18.64, "epoch_pct": 18.64, "eta": "46:12:20", "max_grad_norm": 1.0, "loss": 0.7026737928390503, "grad_norm": 0.42316755652427673, "learning_rate": 0.00019348851680779433}
+{"ts": "2025-12-27T04:52:25", "event": "train_log", "step": 2652, "epoch": 1.1189873417721519, "progress_pct": 18.65, "epoch_pct": 18.65, "eta": "46:10:58", "max_grad_norm": 1.0, "loss": 0.6774232387542725, "grad_norm": 0.35848063230514526, "learning_rate": 0.0001934714580314704}
+{"ts": "2025-12-27T04:52:41", "event": "train_log", "step": 2654, "epoch": 1.119831223628692, "progress_pct": 18.66, "epoch_pct": 18.66, "eta": "46:09:34", "max_grad_norm": 1.0, "loss": 0.7789244055747986, "grad_norm": 0.46613892912864685, "learning_rate": 0.00019345437769307627}
+{"ts": "2025-12-27T04:52:57", "event": "train_log", "step": 2656, "epoch": 1.120675105485232, "progress_pct": 18.68, "epoch_pct": 18.68, "eta": "46:08:09", "max_grad_norm": 1.0, "loss": 0.8647177815437317, "grad_norm": 0.4298146367073059, "learning_rate": 0.00019343727579655204}
+{"ts": "2025-12-27T04:53:14", "event": "train_log", "step": 2658, "epoch": 1.1215189873417721, "progress_pct": 18.69, "epoch_pct": 18.69, "eta": "46:06:47", "max_grad_norm": 1.0, "loss": 0.6949124336242676, "grad_norm": 0.4638161361217499, "learning_rate": 0.00019342015234584276}
+{"ts": "2025-12-27T04:53:28", "event": "train_log", "step": 2660, "epoch": 1.1223628691983123, "progress_pct": 18.71, "epoch_pct": 18.71, "eta": "46:05:15", "max_grad_norm": 1.0, "loss": 0.6677125096321106, "grad_norm": 0.4079605042934418, "learning_rate": 0.00019340300734489854}
+{"ts": "2025-12-27T04:53:43", "event": "train_log", "step": 2662, "epoch": 1.1232067510548522, "progress_pct": 18.72, "epoch_pct": 18.72, "eta": "46:03:46", "max_grad_norm": 1.0, "loss": 0.7521067261695862, "grad_norm": 0.46079063415527344, "learning_rate": 0.00019338584079767434}
+{"ts": "2025-12-27T04:53:58", "event": "train_log", "step": 2664, "epoch": 1.1240506329113924, "progress_pct": 18.73, "epoch_pct": 18.73, "eta": "46:02:20", "max_grad_norm": 1.0, "loss": 0.6829184889793396, "grad_norm": 0.3935202956199646, "learning_rate": 0.0001933686527081302}
+{"ts": "2025-12-27T04:54:14", "event": "train_log", "step": 2666, "epoch": 1.1248945147679326, "progress_pct": 18.75, "epoch_pct": 18.75, "eta": "46:00:53", "max_grad_norm": 1.0, "loss": 0.7122842669487, "grad_norm": 0.45696479082107544, "learning_rate": 0.000193351443080231}
+{"ts": "2025-12-27T04:54:30", "event": "train_log", "step": 2668, "epoch": 1.1257383966244725, "progress_pct": 18.76, "epoch_pct": 18.76, "eta": "45:59:30", "max_grad_norm": 1.0, "loss": 0.7316977977752686, "grad_norm": 0.4078282415866852, "learning_rate": 0.0001933342119179468}
+{"ts": "2025-12-27T04:54:45", "event": "train_log", "step": 2670, "epoch": 1.1265822784810127, "progress_pct": 18.78, "epoch_pct": 18.78, "eta": "45:58:02", "max_grad_norm": 1.0, "loss": 0.7323395013809204, "grad_norm": 0.4282447099685669, "learning_rate": 0.0001933169592252524}
+{"ts": "2025-12-27T04:55:01", "event": "train_log", "step": 2672, "epoch": 1.1274261603375528, "progress_pct": 18.79, "epoch_pct": 18.79, "eta": "45:56:41", "max_grad_norm": 1.0, "loss": 0.7177348732948303, "grad_norm": 0.46398648619651794, "learning_rate": 0.00019329968500612766}
+{"ts": "2025-12-27T04:55:17", "event": "train_log", "step": 2674, "epoch": 1.1282700421940928, "progress_pct": 18.8, "epoch_pct": 18.8, "eta": "45:55:16", "max_grad_norm": 1.0, "loss": 0.7957183122634888, "grad_norm": 0.4560202658176422, "learning_rate": 0.0001932823892645575}
+{"ts": "2025-12-27T04:55:33", "event": "train_log", "step": 2676, "epoch": 1.129113924050633, "progress_pct": 18.82, "epoch_pct": 18.82, "eta": "45:53:53", "max_grad_norm": 1.0, "loss": 0.762060284614563, "grad_norm": 0.4790743887424469, "learning_rate": 0.00019326507200453162}
+{"ts": "2025-12-27T04:55:48", "event": "train_log", "step": 2678, "epoch": 1.129957805907173, "progress_pct": 18.83, "epoch_pct": 18.83, "eta": "45:52:29", "max_grad_norm": 1.0, "loss": 0.7227895259857178, "grad_norm": 0.46524202823638916, "learning_rate": 0.00019324773323004484}
+{"ts": "2025-12-27T04:56:04", "event": "train_log", "step": 2680, "epoch": 1.130801687763713, "progress_pct": 18.85, "epoch_pct": 18.85, "eta": "45:51:04", "max_grad_norm": 1.0, "loss": 0.6968042254447937, "grad_norm": 0.47912222146987915, "learning_rate": 0.00019323037294509683}
+{"ts": "2025-12-27T04:56:19", "event": "train_log", "step": 2682, "epoch": 1.1316455696202532, "progress_pct": 18.86, "epoch_pct": 18.86, "eta": "45:49:37", "max_grad_norm": 1.0, "loss": 0.6682016253471375, "grad_norm": 0.4412849247455597, "learning_rate": 0.00019321299115369233}
+{"ts": "2025-12-27T04:56:34", "event": "train_log", "step": 2684, "epoch": 1.1324894514767934, "progress_pct": 18.87, "epoch_pct": 18.87, "eta": "45:48:09", "max_grad_norm": 1.0, "loss": 0.7287812829017639, "grad_norm": 0.4138370752334595, "learning_rate": 0.00019319558785984095}
+{"ts": "2025-12-27T04:56:50", "event": "train_log", "step": 2686, "epoch": 1.1333333333333333, "progress_pct": 18.89, "epoch_pct": 18.89, "eta": "45:46:46", "max_grad_norm": 1.0, "loss": 0.6547526121139526, "grad_norm": 0.391170859336853, "learning_rate": 0.00019317816306755733}
+{"ts": "2025-12-27T04:57:06", "event": "train_log", "step": 2688, "epoch": 1.1341772151898735, "progress_pct": 18.9, "epoch_pct": 18.9, "eta": "45:45:23", "max_grad_norm": 1.0, "loss": 0.6942042708396912, "grad_norm": 0.45472806692123413, "learning_rate": 0.00019316071678086098}
+{"ts": "2025-12-27T04:57:23", "event": "train_log", "step": 2690, "epoch": 1.1350210970464134, "progress_pct": 18.92, "epoch_pct": 18.92, "eta": "45:44:07", "max_grad_norm": 1.0, "loss": 0.7417841553688049, "grad_norm": 0.40321481227874756, "learning_rate": 0.00019314324900377648}
+{"ts": "2025-12-27T04:57:39", "event": "train_log", "step": 2692, "epoch": 1.1358649789029536, "progress_pct": 18.93, "epoch_pct": 18.93, "eta": "45:42:46", "max_grad_norm": 1.0, "loss": 0.7780525088310242, "grad_norm": 0.3806784152984619, "learning_rate": 0.00019312575974033327}
+{"ts": "2025-12-27T04:57:55", "event": "train_log", "step": 2694, "epoch": 1.1367088607594937, "progress_pct": 18.95, "epoch_pct": 18.95, "eta": "45:41:21", "max_grad_norm": 1.0, "loss": 0.661472499370575, "grad_norm": 0.43763765692710876, "learning_rate": 0.0001931082489945658}
+{"ts": "2025-12-27T04:58:10", "event": "train_log", "step": 2696, "epoch": 1.1375527426160337, "progress_pct": 18.96, "epoch_pct": 18.96, "eta": "45:39:55", "max_grad_norm": 1.0, "loss": 0.7443127632141113, "grad_norm": 0.45410826802253723, "learning_rate": 0.00019309071677051348}
+{"ts": "2025-12-27T04:58:26", "event": "train_log", "step": 2698, "epoch": 1.1383966244725738, "progress_pct": 18.97, "epoch_pct": 18.97, "eta": "45:38:33", "max_grad_norm": 1.0, "loss": 0.744393527507782, "grad_norm": 0.47907447814941406, "learning_rate": 0.00019307316307222062}
+{"ts": "2025-12-27T04:58:41", "event": "train_log", "step": 2700, "epoch": 1.139240506329114, "progress_pct": 18.99, "epoch_pct": 18.99, "eta": "45:37:09", "max_grad_norm": 1.0, "loss": 0.7042794227600098, "grad_norm": 0.41723814606666565, "learning_rate": 0.00019305558790373654}
+{"ts": "2025-12-27T05:10:00", "event": "train_log", "step": 2700, "epoch": 1.139240506329114, "progress_pct": 18.99, "epoch_pct": 18.99, "eta": "46:25:23", "max_grad_norm": 1.0, "eval_loss": 0.8155058026313782, "eval_runtime": 678.284, "eval_samples_per_second": 3.106, "eval_steps_per_second": 3.106}
+{"ts": "2025-12-27T05:10:15", "event": "train_log", "step": 2702, "epoch": 1.140084388185654, "progress_pct": 19.0, "epoch_pct": 19.0, "eta": "46:23:54", "max_grad_norm": 1.0, "loss": 0.6715745329856873, "grad_norm": 0.3863268196582794, "learning_rate": 0.0001930379912691155}
+{"ts": "2025-12-27T05:10:30", "event": "train_log", "step": 2704, "epoch": 1.140928270042194, "progress_pct": 19.02, "epoch_pct": 19.02, "eta": "46:22:26", "max_grad_norm": 1.0, "loss": 0.6959030628204346, "grad_norm": 0.49994897842407227, "learning_rate": 0.00019302037317241673}
+{"ts": "2025-12-27T05:10:45", "event": "train_log", "step": 2706, "epoch": 1.1417721518987343, "progress_pct": 19.03, "epoch_pct": 19.03, "eta": "46:20:59", "max_grad_norm": 1.0, "loss": 0.6534222364425659, "grad_norm": 0.46377572417259216, "learning_rate": 0.00019300273361770432}
+{"ts": "2025-12-27T05:11:01", "event": "train_log", "step": 2708, "epoch": 1.1426160337552742, "progress_pct": 19.04, "epoch_pct": 19.04, "eta": "46:19:33", "max_grad_norm": 1.0, "loss": 0.6750475764274597, "grad_norm": 0.3841579556465149, "learning_rate": 0.00019298507260904745}
+{"ts": "2025-12-27T05:11:15", "event": "train_log", "step": 2710, "epoch": 1.1434599156118144, "progress_pct": 19.06, "epoch_pct": 19.06, "eta": "46:18:03", "max_grad_norm": 1.0, "loss": 0.7714900374412537, "grad_norm": 0.4171544313430786, "learning_rate": 0.00019296739015052015}
+{"ts": "2025-12-27T05:11:31", "event": "train_log", "step": 2712, "epoch": 1.1443037974683543, "progress_pct": 19.07, "epoch_pct": 19.07, "eta": "46:16:37", "max_grad_norm": 1.0, "loss": 0.7281306982040405, "grad_norm": 0.44384199380874634, "learning_rate": 0.00019294968624620137}
+{"ts": "2025-12-27T05:11:46", "event": "train_log", "step": 2714, "epoch": 1.1451476793248945, "progress_pct": 19.09, "epoch_pct": 19.09, "eta": "46:15:08", "max_grad_norm": 1.0, "loss": 0.7407262325286865, "grad_norm": 0.43345192074775696, "learning_rate": 0.00019293196090017513}
+{"ts": "2025-12-27T05:12:01", "event": "train_log", "step": 2716, "epoch": 1.1459915611814346, "progress_pct": 19.1, "epoch_pct": 19.1, "eta": "46:13:40", "max_grad_norm": 1.0, "loss": 0.6794347167015076, "grad_norm": 0.3897020220756531, "learning_rate": 0.00019291421411653034}
+{"ts": "2025-12-27T05:12:16", "event": "train_log", "step": 2718, "epoch": 1.1468354430379746, "progress_pct": 19.11, "epoch_pct": 19.11, "eta": "46:12:12", "max_grad_norm": 1.0, "loss": 0.7026289105415344, "grad_norm": 0.4289143681526184, "learning_rate": 0.00019289644589936075}
+{"ts": "2025-12-27T05:12:30", "event": "train_log", "step": 2720, "epoch": 1.1476793248945147, "progress_pct": 19.13, "epoch_pct": 19.13, "eta": "46:10:42", "max_grad_norm": 1.0, "loss": 0.7613762617111206, "grad_norm": 0.4783557057380676, "learning_rate": 0.0001928786562527652}
+{"ts": "2025-12-27T05:12:46", "event": "train_log", "step": 2722, "epoch": 1.148523206751055, "progress_pct": 19.14, "epoch_pct": 19.14, "eta": "46:09:18", "max_grad_norm": 1.0, "loss": 0.7123059630393982, "grad_norm": 0.4152112901210785, "learning_rate": 0.00019286084518084745}
+{"ts": "2025-12-27T05:13:01", "event": "train_log", "step": 2724, "epoch": 1.1493670886075948, "progress_pct": 19.16, "epoch_pct": 19.16, "eta": "46:07:49", "max_grad_norm": 1.0, "loss": 0.7198041677474976, "grad_norm": 0.4204908013343811, "learning_rate": 0.0001928430126877161}
+{"ts": "2025-12-27T05:13:16", "event": "train_log", "step": 2726, "epoch": 1.150210970464135, "progress_pct": 19.17, "epoch_pct": 19.17, "eta": "46:06:25", "max_grad_norm": 1.0, "loss": 0.6859839558601379, "grad_norm": 0.41333863139152527, "learning_rate": 0.00019282515877748485}
+{"ts": "2025-12-27T05:13:32", "event": "train_log", "step": 2728, "epoch": 1.1510548523206752, "progress_pct": 19.18, "epoch_pct": 19.18, "eta": "46:04:58", "max_grad_norm": 1.0, "loss": 0.7185758948326111, "grad_norm": 0.4041498005390167, "learning_rate": 0.00019280728345427218}
+{"ts": "2025-12-27T05:13:47", "event": "train_log", "step": 2730, "epoch": 1.1518987341772151, "progress_pct": 19.2, "epoch_pct": 19.2, "eta": "46:03:31", "max_grad_norm": 1.0, "loss": 0.7899971008300781, "grad_norm": 0.4487358629703522, "learning_rate": 0.00019278938672220167}
+{"ts": "2025-12-27T05:14:03", "event": "train_log", "step": 2732, "epoch": 1.1527426160337553, "progress_pct": 19.21, "epoch_pct": 19.21, "eta": "46:02:09", "max_grad_norm": 1.0, "loss": 0.6266586780548096, "grad_norm": 0.3510114252567291, "learning_rate": 0.00019277146858540166}
+{"ts": "2025-12-27T05:14:17", "event": "train_log", "step": 2734, "epoch": 1.1535864978902954, "progress_pct": 19.23, "epoch_pct": 19.23, "eta": "46:00:38", "max_grad_norm": 1.0, "loss": 0.7450110912322998, "grad_norm": 0.4295555055141449, "learning_rate": 0.00019275352904800553}
+{"ts": "2025-12-27T05:14:32", "event": "train_log", "step": 2736, "epoch": 1.1544303797468354, "progress_pct": 19.24, "epoch_pct": 19.24, "eta": "45:59:10", "max_grad_norm": 1.0, "loss": 0.6875549554824829, "grad_norm": 0.40629687905311584, "learning_rate": 0.00019273556811415165}
+{"ts": "2025-12-27T05:14:46", "event": "train_log", "step": 2738, "epoch": 1.1552742616033755, "progress_pct": 19.25, "epoch_pct": 19.25, "eta": "45:57:42", "max_grad_norm": 1.0, "loss": 0.8141522407531738, "grad_norm": 0.4243362247943878, "learning_rate": 0.00019271758578798322}
+{"ts": "2025-12-27T05:15:02", "event": "train_log", "step": 2740, "epoch": 1.1561181434599157, "progress_pct": 19.27, "epoch_pct": 19.27, "eta": "45:56:18", "max_grad_norm": 1.0, "loss": 0.6905186176300049, "grad_norm": 0.483306884765625, "learning_rate": 0.00019269958207364843}
+{"ts": "2025-12-27T05:15:17", "event": "train_log", "step": 2742, "epoch": 1.1569620253164556, "progress_pct": 19.28, "epoch_pct": 19.28, "eta": "45:54:53", "max_grad_norm": 1.0, "loss": 0.6709047555923462, "grad_norm": 0.4303925335407257, "learning_rate": 0.00019268155697530037}
+{"ts": "2025-12-27T05:15:32", "event": "train_log", "step": 2744, "epoch": 1.1578059071729958, "progress_pct": 19.3, "epoch_pct": 19.3, "eta": "45:53:24", "max_grad_norm": 1.0, "loss": 0.7107465863227844, "grad_norm": 0.4533381760120392, "learning_rate": 0.0001926635104970971}
+{"ts": "2025-12-27T05:15:46", "event": "train_log", "step": 2746, "epoch": 1.158649789029536, "progress_pct": 19.31, "epoch_pct": 19.31, "eta": "45:51:55", "max_grad_norm": 1.0, "loss": 0.8043986558914185, "grad_norm": 0.4823286533355713, "learning_rate": 0.0001926454426432016}
+{"ts": "2025-12-27T05:16:01", "event": "train_log", "step": 2748, "epoch": 1.159493670886076, "progress_pct": 19.32, "epoch_pct": 19.32, "eta": "45:50:26", "max_grad_norm": 1.0, "loss": 0.7812165021896362, "grad_norm": 0.3892867863178253, "learning_rate": 0.00019262735341778179}
+{"ts": "2025-12-27T05:16:15", "event": "train_log", "step": 2750, "epoch": 1.160337552742616, "progress_pct": 19.34, "epoch_pct": 19.34, "eta": "45:48:56", "max_grad_norm": 1.0, "loss": 0.7141886353492737, "grad_norm": 0.46737003326416016, "learning_rate": 0.00019260924282501045}
+{"ts": "2025-12-27T05:16:29", "event": "train_log", "step": 2752, "epoch": 1.1611814345991562, "progress_pct": 19.35, "epoch_pct": 19.35, "eta": "45:47:28", "max_grad_norm": 1.0, "loss": 0.7528460621833801, "grad_norm": 0.4335328936576843, "learning_rate": 0.0001925911108690654}
+{"ts": "2025-12-27T05:16:44", "event": "train_log", "step": 2754, "epoch": 1.1620253164556962, "progress_pct": 19.37, "epoch_pct": 19.37, "eta": "45:46:03", "max_grad_norm": 1.0, "loss": 0.7817611694335938, "grad_norm": 0.3792484700679779, "learning_rate": 0.00019257295755412932}
+{"ts": "2025-12-27T05:17:00", "event": "train_log", "step": 2756, "epoch": 1.1628691983122363, "progress_pct": 19.38, "epoch_pct": 19.38, "eta": "45:44:38", "max_grad_norm": 1.0, "loss": 0.6904958486557007, "grad_norm": 0.41246429085731506, "learning_rate": 0.00019255478288438983}
+{"ts": "2025-12-27T05:17:14", "event": "train_log", "step": 2758, "epoch": 1.1637130801687763, "progress_pct": 19.4, "epoch_pct": 19.4, "eta": "45:43:12", "max_grad_norm": 1.0, "loss": 0.7571630477905273, "grad_norm": 0.44368481636047363, "learning_rate": 0.00019253658686403948}
+{"ts": "2025-12-27T05:17:30", "event": "train_log", "step": 2760, "epoch": 1.1645569620253164, "progress_pct": 19.41, "epoch_pct": 19.41, "eta": "45:41:47", "max_grad_norm": 1.0, "loss": 0.6747061014175415, "grad_norm": 0.40121781826019287, "learning_rate": 0.00019251836949727574}
+{"ts": "2025-12-27T05:17:44", "event": "train_log", "step": 2762, "epoch": 1.1654008438818566, "progress_pct": 19.42, "epoch_pct": 19.42, "eta": "45:40:19", "max_grad_norm": 1.0, "loss": 0.7722339034080505, "grad_norm": 0.4660188853740692, "learning_rate": 0.000192500130788301}
+{"ts": "2025-12-27T05:17:59", "event": "train_log", "step": 2764, "epoch": 1.1662447257383965, "progress_pct": 19.44, "epoch_pct": 19.44, "eta": "45:38:54", "max_grad_norm": 1.0, "loss": 0.8081457018852234, "grad_norm": 0.41693782806396484, "learning_rate": 0.0001924818707413226}
+{"ts": "2025-12-27T05:18:14", "event": "train_log", "step": 2766, "epoch": 1.1670886075949367, "progress_pct": 19.45, "epoch_pct": 19.45, "eta": "45:37:28", "max_grad_norm": 1.0, "loss": 0.7745853066444397, "grad_norm": 0.4274570345878601, "learning_rate": 0.00019246358936055274}
+{"ts": "2025-12-27T05:18:29", "event": "train_log", "step": 2768, "epoch": 1.1679324894514769, "progress_pct": 19.47, "epoch_pct": 19.47, "eta": "45:36:04", "max_grad_norm": 1.0, "loss": 0.7918146252632141, "grad_norm": 0.4385969638824463, "learning_rate": 0.00019244528665020863}
+{"ts": "2025-12-27T05:18:45", "event": "train_log", "step": 2770, "epoch": 1.1687763713080168, "progress_pct": 19.48, "epoch_pct": 19.48, "eta": "45:34:42", "max_grad_norm": 1.0, "loss": 0.7382830381393433, "grad_norm": 0.42832866311073303, "learning_rate": 0.00019242696261451234}
+{"ts": "2025-12-27T05:19:02", "event": "train_log", "step": 2772, "epoch": 1.169620253164557, "progress_pct": 19.49, "epoch_pct": 19.49, "eta": "45:33:25", "max_grad_norm": 1.0, "loss": 0.7501721978187561, "grad_norm": 0.44177982211112976, "learning_rate": 0.00019240861725769085}
+{"ts": "2025-12-27T05:19:19", "event": "train_log", "step": 2774, "epoch": 1.1704641350210971, "progress_pct": 19.51, "epoch_pct": 19.51, "eta": "45:32:06", "max_grad_norm": 1.0, "loss": 0.6495388150215149, "grad_norm": 0.4193457365036011, "learning_rate": 0.00019239025058397611}
+{"ts": "2025-12-27T05:19:34", "event": "train_log", "step": 2776, "epoch": 1.171308016877637, "progress_pct": 19.52, "epoch_pct": 19.52, "eta": "45:30:42", "max_grad_norm": 1.0, "loss": 0.8291991353034973, "grad_norm": 0.5145925283432007, "learning_rate": 0.000192371862597605}
+{"ts": "2025-12-27T05:19:51", "event": "train_log", "step": 2778, "epoch": 1.1721518987341772, "progress_pct": 19.54, "epoch_pct": 19.54, "eta": "45:29:25", "max_grad_norm": 1.0, "loss": 0.7491340637207031, "grad_norm": 0.40911251306533813, "learning_rate": 0.00019235345330281914}
+{"ts": "2025-12-27T05:20:07", "event": "train_log", "step": 2780, "epoch": 1.1729957805907172, "progress_pct": 19.55, "epoch_pct": 19.55, "eta": "45:28:07", "max_grad_norm": 1.0, "loss": 0.6315004825592041, "grad_norm": 0.3920276165008545, "learning_rate": 0.0001923350227038653}
+{"ts": "2025-12-27T05:20:23", "event": "train_log", "step": 2782, "epoch": 1.1738396624472573, "progress_pct": 19.56, "epoch_pct": 19.56, "eta": "45:26:44", "max_grad_norm": 1.0, "loss": 0.7538990378379822, "grad_norm": 0.442303329706192, "learning_rate": 0.0001923165708049951}
+{"ts": "2025-12-27T05:20:39", "event": "train_log", "step": 2784, "epoch": 1.1746835443037975, "progress_pct": 19.58, "epoch_pct": 19.58, "eta": "45:25:26", "max_grad_norm": 1.0, "loss": 0.7440794110298157, "grad_norm": 0.4693523049354553, "learning_rate": 0.00019229809761046497}
+{"ts": "2025-12-27T05:20:55", "event": "train_log", "step": 2786, "epoch": 1.1755274261603375, "progress_pct": 19.59, "epoch_pct": 19.59, "eta": "45:24:06", "max_grad_norm": 1.0, "loss": 0.7342568635940552, "grad_norm": 0.4948059618473053, "learning_rate": 0.0001922796031245363}
+{"ts": "2025-12-27T05:21:13", "event": "train_log", "step": 2788, "epoch": 1.1763713080168776, "progress_pct": 19.61, "epoch_pct": 19.61, "eta": "45:22:52", "max_grad_norm": 1.0, "loss": 0.6969661116600037, "grad_norm": 0.3900465965270996, "learning_rate": 0.00019226108735147547}
+{"ts": "2025-12-27T05:21:29", "event": "train_log", "step": 2790, "epoch": 1.1772151898734178, "progress_pct": 19.62, "epoch_pct": 19.62, "eta": "45:21:34", "max_grad_norm": 1.0, "loss": 0.7097541689872742, "grad_norm": 0.36153602600097656, "learning_rate": 0.00019224255029555372}
+{"ts": "2025-12-27T05:21:46", "event": "train_log", "step": 2792, "epoch": 1.1780590717299577, "progress_pct": 19.63, "epoch_pct": 19.63, "eta": "45:20:16", "max_grad_norm": 1.0, "loss": 0.8141831755638123, "grad_norm": 0.41168832778930664, "learning_rate": 0.00019222399196104715}
+{"ts": "2025-12-27T05:22:02", "event": "train_log", "step": 2794, "epoch": 1.1789029535864979, "progress_pct": 19.65, "epoch_pct": 19.65, "eta": "45:18:56", "max_grad_norm": 1.0, "loss": 0.7458863258361816, "grad_norm": 0.4542821943759918, "learning_rate": 0.0001922054123522368}
+{"ts": "2025-12-27T05:22:17", "event": "train_log", "step": 2796, "epoch": 1.179746835443038, "progress_pct": 19.66, "epoch_pct": 19.66, "eta": "45:17:34", "max_grad_norm": 1.0, "loss": 0.6535642147064209, "grad_norm": 0.4026047885417938, "learning_rate": 0.0001921868114734087}
+{"ts": "2025-12-27T05:22:34", "event": "train_log", "step": 2798, "epoch": 1.180590717299578, "progress_pct": 19.68, "epoch_pct": 19.68, "eta": "45:16:16", "max_grad_norm": 1.0, "loss": 0.6248135566711426, "grad_norm": 0.4417034387588501, "learning_rate": 0.0001921681893288537}
+{"ts": "2025-12-27T05:22:50", "event": "train_log", "step": 2800, "epoch": 1.1814345991561181, "progress_pct": 19.69, "epoch_pct": 19.69, "eta": "45:14:59", "max_grad_norm": 1.0, "loss": 0.6777785420417786, "grad_norm": 0.45027074217796326, "learning_rate": 0.00019214954592286748}
+{"ts": "2025-12-27T05:34:19", "event": "train_log", "step": 2800, "epoch": 1.1814345991561181, "progress_pct": 19.69, "epoch_pct": 19.69, "eta": "46:01:47", "max_grad_norm": 1.0, "eval_loss": 0.8124309182167053, "eval_runtime": 688.4759, "eval_samples_per_second": 3.06, "eval_steps_per_second": 3.06}
+{"ts": "2025-12-27T05:34:35", "event": "train_log", "step": 2802, "epoch": 1.1822784810126583, "progress_pct": 19.7, "epoch_pct": 19.7, "eta": "46:00:24", "max_grad_norm": 1.0, "loss": 0.6460301876068115, "grad_norm": 0.406865656375885, "learning_rate": 0.00019213088125975082}
+{"ts": "2025-12-27T05:34:49", "event": "train_log", "step": 2804, "epoch": 1.1831223628691983, "progress_pct": 19.72, "epoch_pct": 19.72, "eta": "45:58:53", "max_grad_norm": 1.0, "loss": 0.7292830944061279, "grad_norm": 0.5052749514579773, "learning_rate": 0.00019211219534380928}
+{"ts": "2025-12-27T05:35:03", "event": "train_log", "step": 2806, "epoch": 1.1839662447257384, "progress_pct": 19.73, "epoch_pct": 19.73, "eta": "45:57:26", "max_grad_norm": 1.0, "loss": 0.7399409413337708, "grad_norm": 0.4028380513191223, "learning_rate": 0.00019209348817935328}
+{"ts": "2025-12-27T05:35:19", "event": "train_log", "step": 2808, "epoch": 1.1848101265822786, "progress_pct": 19.75, "epoch_pct": 19.75, "eta": "45:56:03", "max_grad_norm": 1.0, "loss": 0.6885398030281067, "grad_norm": 0.4522664248943329, "learning_rate": 0.00019207475977069826}
+{"ts": "2025-12-27T05:35:34", "event": "train_log", "step": 2810, "epoch": 1.1856540084388185, "progress_pct": 19.76, "epoch_pct": 19.76, "eta": "45:54:38", "max_grad_norm": 1.0, "loss": 0.6962382197380066, "grad_norm": 0.419546514749527, "learning_rate": 0.0001920560101221645}
+{"ts": "2025-12-27T05:35:49", "event": "train_log", "step": 2812, "epoch": 1.1864978902953587, "progress_pct": 19.77, "epoch_pct": 19.77, "eta": "45:53:11", "max_grad_norm": 1.0, "loss": 0.7380965352058411, "grad_norm": 0.48912352323532104, "learning_rate": 0.00019203723923807715}
+{"ts": "2025-12-27T05:36:03", "event": "train_log", "step": 2814, "epoch": 1.1873417721518988, "progress_pct": 19.79, "epoch_pct": 19.79, "eta": "45:51:43", "max_grad_norm": 1.0, "loss": 0.755382776260376, "grad_norm": 0.40905487537384033, "learning_rate": 0.00019201844712276635}
+{"ts": "2025-12-27T05:36:18", "event": "train_log", "step": 2816, "epoch": 1.1881856540084388, "progress_pct": 19.8, "epoch_pct": 19.8, "eta": "45:50:18", "max_grad_norm": 1.0, "loss": 0.7066168189048767, "grad_norm": 0.385623037815094, "learning_rate": 0.000191999633780567}
+{"ts": "2025-12-27T05:36:33", "event": "train_log", "step": 2818, "epoch": 1.189029535864979, "progress_pct": 19.82, "epoch_pct": 19.82, "eta": "45:48:53", "max_grad_norm": 1.0, "loss": 0.7172947525978088, "grad_norm": 0.4488723576068878, "learning_rate": 0.00019198079921581908}
+{"ts": "2025-12-27T05:36:48", "event": "train_log", "step": 2820, "epoch": 1.189873417721519, "progress_pct": 19.83, "epoch_pct": 19.83, "eta": "45:47:28", "max_grad_norm": 1.0, "loss": 0.6714195013046265, "grad_norm": 0.385277658700943, "learning_rate": 0.00019196194343286727}
+{"ts": "2025-12-27T05:37:03", "event": "train_log", "step": 2822, "epoch": 1.190717299578059, "progress_pct": 19.85, "epoch_pct": 19.85, "eta": "45:46:03", "max_grad_norm": 1.0, "loss": 0.7076966166496277, "grad_norm": 0.4280855655670166, "learning_rate": 0.00019194306643606127}
+{"ts": "2025-12-27T05:37:18", "event": "train_log", "step": 2824, "epoch": 1.1915611814345992, "progress_pct": 19.86, "epoch_pct": 19.86, "eta": "45:44:35", "max_grad_norm": 1.0, "loss": 0.7926747798919678, "grad_norm": 0.4626067578792572, "learning_rate": 0.00019192416822975567}
+{"ts": "2025-12-27T05:37:33", "event": "train_log", "step": 2826, "epoch": 1.1924050632911392, "progress_pct": 19.87, "epoch_pct": 19.87, "eta": "45:43:09", "max_grad_norm": 1.0, "loss": 0.7654290199279785, "grad_norm": 0.3994208574295044, "learning_rate": 0.00019190524881830987}
+{"ts": "2025-12-27T05:37:47", "event": "train_log", "step": 2828, "epoch": 1.1932489451476793, "progress_pct": 19.89, "epoch_pct": 19.89, "eta": "45:41:43", "max_grad_norm": 1.0, "loss": 0.7857998609542847, "grad_norm": 0.47346508502960205, "learning_rate": 0.00019188630820608825}
+{"ts": "2025-12-27T05:38:02", "event": "train_log", "step": 2830, "epoch": 1.1940928270042195, "progress_pct": 19.9, "epoch_pct": 19.9, "eta": "45:40:17", "max_grad_norm": 1.0, "loss": 0.7603738903999329, "grad_norm": 0.446077823638916, "learning_rate": 0.00019186734639746003}
+{"ts": "2025-12-27T05:38:17", "event": "train_log", "step": 2832, "epoch": 1.1949367088607594, "progress_pct": 19.92, "epoch_pct": 19.92, "eta": "45:38:54", "max_grad_norm": 1.0, "loss": 0.790582537651062, "grad_norm": 0.41503727436065674, "learning_rate": 0.00019184836339679936}
+{"ts": "2025-12-27T05:38:32", "event": "train_log", "step": 2834, "epoch": 1.1957805907172996, "progress_pct": 19.93, "epoch_pct": 19.93, "eta": "45:37:30", "max_grad_norm": 1.0, "loss": 0.7229864597320557, "grad_norm": 0.43651700019836426, "learning_rate": 0.00019182935920848522}
+{"ts": "2025-12-27T05:38:47", "event": "train_log", "step": 2836, "epoch": 1.1966244725738397, "progress_pct": 19.94, "epoch_pct": 19.94, "eta": "45:36:03", "max_grad_norm": 1.0, "loss": 0.7817287445068359, "grad_norm": 0.4409031867980957, "learning_rate": 0.00019181033383690153}
+{"ts": "2025-12-27T05:39:02", "event": "train_log", "step": 2838, "epoch": 1.1974683544303797, "progress_pct": 19.96, "epoch_pct": 19.96, "eta": "45:34:40", "max_grad_norm": 1.0, "loss": 0.6342297792434692, "grad_norm": 0.3562015891075134, "learning_rate": 0.0001917912872864371}
+{"ts": "2025-12-27T05:39:17", "event": "train_log", "step": 2840, "epoch": 1.1983122362869199, "progress_pct": 19.97, "epoch_pct": 19.97, "eta": "45:33:15", "max_grad_norm": 1.0, "loss": 0.7065272927284241, "grad_norm": 0.4767192602157593, "learning_rate": 0.00019177221956148555}
+{"ts": "2025-12-27T05:39:33", "event": "train_log", "step": 2842, "epoch": 1.1991561181434598, "progress_pct": 19.99, "epoch_pct": 19.99, "eta": "45:31:54", "max_grad_norm": 1.0, "loss": 0.777110755443573, "grad_norm": 0.45131945610046387, "learning_rate": 0.00019175313066644546}
+{"ts": "2025-12-27T05:39:48", "event": "train_log", "step": 2844, "epoch": 1.2, "progress_pct": 20.0, "epoch_pct": 20.0, "eta": "45:30:30", "max_grad_norm": 1.0, "loss": 0.7456457018852234, "grad_norm": 0.47566652297973633, "learning_rate": 0.00019173402060572027}
+{"ts": "2025-12-27T05:40:02", "event": "train_log", "step": 2846, "epoch": 1.2008438818565401, "progress_pct": 20.01, "epoch_pct": 20.01, "eta": "45:29:03", "max_grad_norm": 1.0, "loss": 0.8041547536849976, "grad_norm": 0.48909857869148254, "learning_rate": 0.00019171488938371834}
+{"ts": "2025-12-27T05:40:17", "event": "train_log", "step": 2848, "epoch": 1.20168776371308, "progress_pct": 20.03, "epoch_pct": 20.03, "eta": "45:27:40", "max_grad_norm": 1.0, "loss": 0.6786676049232483, "grad_norm": 0.436591774225235, "learning_rate": 0.00019169573700485284}
+{"ts": "2025-12-27T05:40:32", "event": "train_log", "step": 2850, "epoch": 1.2025316455696202, "progress_pct": 20.04, "epoch_pct": 20.04, "eta": "45:26:14", "max_grad_norm": 1.0, "loss": 0.7661880850791931, "grad_norm": 0.44230160117149353, "learning_rate": 0.00019167656347354183}
+{"ts": "2025-12-27T05:40:46", "event": "train_log", "step": 2852, "epoch": 1.2033755274261604, "progress_pct": 20.06, "epoch_pct": 20.06, "eta": "45:24:49", "max_grad_norm": 1.0, "loss": 0.7369645833969116, "grad_norm": 0.4844048321247101, "learning_rate": 0.00019165736879420836}
+{"ts": "2025-12-27T05:41:01", "event": "train_log", "step": 2854, "epoch": 1.2042194092827003, "progress_pct": 20.07, "epoch_pct": 20.07, "eta": "45:23:23", "max_grad_norm": 1.0, "loss": 0.7776681184768677, "grad_norm": 0.4272955060005188, "learning_rate": 0.0001916381529712802}
+{"ts": "2025-12-27T05:41:15", "event": "train_log", "step": 2856, "epoch": 1.2050632911392405, "progress_pct": 20.08, "epoch_pct": 20.08, "eta": "45:21:57", "max_grad_norm": 1.0, "loss": 0.8019667267799377, "grad_norm": 0.42392730712890625, "learning_rate": 0.00019161891600919008}
+{"ts": "2025-12-27T05:41:29", "event": "train_log", "step": 2858, "epoch": 1.2059071729957807, "progress_pct": 20.1, "epoch_pct": 20.1, "eta": "45:20:31", "max_grad_norm": 1.0, "loss": 0.7811952233314514, "grad_norm": 0.4528326988220215, "learning_rate": 0.00019159965791237565}
+{"ts": "2025-12-27T05:41:45", "event": "train_log", "step": 2860, "epoch": 1.2067510548523206, "progress_pct": 20.11, "epoch_pct": 20.11, "eta": "45:19:09", "max_grad_norm": 1.0, "loss": 0.7244279980659485, "grad_norm": 0.42840948700904846, "learning_rate": 0.00019158037868527932}
+{"ts": "2025-12-27T05:42:00", "event": "train_log", "step": 2862, "epoch": 1.2075949367088608, "progress_pct": 20.13, "epoch_pct": 20.13, "eta": "45:17:45", "max_grad_norm": 1.0, "loss": 0.7675760984420776, "grad_norm": 0.4291226863861084, "learning_rate": 0.00019156107833234845}
+{"ts": "2025-12-27T05:42:14", "event": "train_log", "step": 2864, "epoch": 1.208438818565401, "progress_pct": 20.14, "epoch_pct": 20.14, "eta": "45:16:21", "max_grad_norm": 1.0, "loss": 0.7071483135223389, "grad_norm": 0.4455093443393707, "learning_rate": 0.00019154175685803527}
+{"ts": "2025-12-27T05:42:29", "event": "train_log", "step": 2866, "epoch": 1.2092827004219409, "progress_pct": 20.15, "epoch_pct": 20.15, "eta": "45:14:58", "max_grad_norm": 1.0, "loss": 0.7554603815078735, "grad_norm": 0.41356268525123596, "learning_rate": 0.00019152241426679688}
+{"ts": "2025-12-27T05:42:44", "event": "train_log", "step": 2868, "epoch": 1.210126582278481, "progress_pct": 20.17, "epoch_pct": 20.17, "eta": "45:13:35", "max_grad_norm": 1.0, "loss": 0.7198356986045837, "grad_norm": 0.43875646591186523, "learning_rate": 0.00019150305056309525}
+{"ts": "2025-12-27T05:42:59", "event": "train_log", "step": 2870, "epoch": 1.2109704641350212, "progress_pct": 20.18, "epoch_pct": 20.18, "eta": "45:12:11", "max_grad_norm": 1.0, "loss": 0.7393195629119873, "grad_norm": 0.454356849193573, "learning_rate": 0.0001914836657513972}
+{"ts": "2025-12-27T05:43:14", "event": "train_log", "step": 2872, "epoch": 1.2118143459915611, "progress_pct": 20.2, "epoch_pct": 20.2, "eta": "45:10:48", "max_grad_norm": 1.0, "loss": 0.7277671694755554, "grad_norm": 0.37668535113334656, "learning_rate": 0.00019146425983617444}
+{"ts": "2025-12-27T05:43:30", "event": "train_log", "step": 2874, "epoch": 1.2126582278481013, "progress_pct": 20.21, "epoch_pct": 20.21, "eta": "45:09:28", "max_grad_norm": 1.0, "loss": 0.6783158183097839, "grad_norm": 0.37869662046432495, "learning_rate": 0.00019144483282190354}
+{"ts": "2025-12-27T05:43:46", "event": "train_log", "step": 2876, "epoch": 1.2135021097046415, "progress_pct": 20.23, "epoch_pct": 20.23, "eta": "45:08:10", "max_grad_norm": 1.0, "loss": 0.7024772763252258, "grad_norm": 0.4333525002002716, "learning_rate": 0.00019142538471306596}
+{"ts": "2025-12-27T05:44:00", "event": "train_log", "step": 2878, "epoch": 1.2143459915611814, "progress_pct": 20.24, "epoch_pct": 20.24, "eta": "45:06:46", "max_grad_norm": 1.0, "loss": 0.7387984395027161, "grad_norm": 0.4693842828273773, "learning_rate": 0.00019140591551414796}
+{"ts": "2025-12-27T05:44:16", "event": "train_log", "step": 2880, "epoch": 1.2151898734177216, "progress_pct": 20.25, "epoch_pct": 20.25, "eta": "45:05:24", "max_grad_norm": 1.0, "loss": 0.7597174644470215, "grad_norm": 0.432640939950943, "learning_rate": 0.00019138642522964074}
+{"ts": "2025-12-27T05:44:30", "event": "train_log", "step": 2882, "epoch": 1.2160337552742617, "progress_pct": 20.27, "epoch_pct": 20.27, "eta": "45:03:59", "max_grad_norm": 1.0, "loss": 0.7884587049484253, "grad_norm": 0.432477205991745, "learning_rate": 0.00019136691386404032}
+{"ts": "2025-12-27T05:44:46", "event": "train_log", "step": 2884, "epoch": 1.2168776371308017, "progress_pct": 20.28, "epoch_pct": 20.28, "eta": "45:02:40", "max_grad_norm": 1.0, "loss": 0.6873269081115723, "grad_norm": 0.39305904507637024, "learning_rate": 0.00019134738142184763}
+{"ts": "2025-12-27T05:45:01", "event": "train_log", "step": 2886, "epoch": 1.2177215189873418, "progress_pct": 20.3, "epoch_pct": 20.3, "eta": "45:01:18", "max_grad_norm": 1.0, "loss": 0.6523491144180298, "grad_norm": 0.4485352635383606, "learning_rate": 0.00019132782790756838}
+{"ts": "2025-12-27T05:45:15", "event": "train_log", "step": 2888, "epoch": 1.2185654008438818, "progress_pct": 20.31, "epoch_pct": 20.31, "eta": "44:59:54", "max_grad_norm": 1.0, "loss": 0.7383941411972046, "grad_norm": 0.4480868875980377, "learning_rate": 0.00019130825332571322}
+{"ts": "2025-12-27T05:45:30", "event": "train_log", "step": 2890, "epoch": 1.219409282700422, "progress_pct": 20.32, "epoch_pct": 20.32, "eta": "44:58:33", "max_grad_norm": 1.0, "loss": 0.7391335964202881, "grad_norm": 0.44085049629211426, "learning_rate": 0.00019128865768079764}
+{"ts": "2025-12-27T05:45:45", "event": "train_log", "step": 2892, "epoch": 1.220253164556962, "progress_pct": 20.34, "epoch_pct": 20.34, "eta": "44:57:10", "max_grad_norm": 1.0, "loss": 0.7282017469406128, "grad_norm": 0.4621105492115021, "learning_rate": 0.00019126904097734198}
+{"ts": "2025-12-27T05:46:01", "event": "train_log", "step": 2894, "epoch": 1.221097046413502, "progress_pct": 20.35, "epoch_pct": 20.35, "eta": "44:55:51", "max_grad_norm": 1.0, "loss": 0.7554038763046265, "grad_norm": 0.41478198766708374, "learning_rate": 0.00019124940321987136}
+{"ts": "2025-12-27T05:46:15", "event": "train_log", "step": 2896, "epoch": 1.2219409282700422, "progress_pct": 20.37, "epoch_pct": 20.37, "eta": "44:54:28", "max_grad_norm": 1.0, "loss": 0.6826799511909485, "grad_norm": 0.4143751263618469, "learning_rate": 0.00019122974441291595}
+{"ts": "2025-12-27T05:46:31", "event": "train_log", "step": 2898, "epoch": 1.2227848101265824, "progress_pct": 20.38, "epoch_pct": 20.38, "eta": "44:53:08", "max_grad_norm": 1.0, "loss": 0.751000702381134, "grad_norm": 0.45392894744873047, "learning_rate": 0.00019121006456101057}
+{"ts": "2025-12-27T05:46:45", "event": "train_log", "step": 2900, "epoch": 1.2236286919831223, "progress_pct": 20.39, "epoch_pct": 20.39, "eta": "44:51:45", "max_grad_norm": 1.0, "loss": 0.7598516941070557, "grad_norm": 0.46946296095848083, "learning_rate": 0.00019119036366869506}
+{"ts": "2025-12-27T05:58:11", "event": "train_log", "step": 2900, "epoch": 1.2236286919831223, "progress_pct": 20.39, "epoch_pct": 20.39, "eta": "45:36:22", "max_grad_norm": 1.0, "eval_loss": 0.8077136278152466, "eval_runtime": 685.6042, "eval_samples_per_second": 3.073, "eval_steps_per_second": 3.073}
+{"ts": "2025-12-27T05:58:26", "event": "train_log", "step": 2902, "epoch": 1.2244725738396625, "progress_pct": 20.41, "epoch_pct": 20.41, "eta": "45:34:58", "max_grad_norm": 1.0, "loss": 0.7518358826637268, "grad_norm": 0.4316656291484833, "learning_rate": 0.00019117064174051395}
+{"ts": "2025-12-27T05:58:42", "event": "train_log", "step": 2904, "epoch": 1.2253164556962026, "progress_pct": 20.42, "epoch_pct": 20.42, "eta": "45:33:37", "max_grad_norm": 1.0, "loss": 0.7308215498924255, "grad_norm": 0.47737377882003784, "learning_rate": 0.0001911508987810168}
+{"ts": "2025-12-27T05:58:57", "event": "train_log", "step": 2906, "epoch": 1.2261603375527426, "progress_pct": 20.44, "epoch_pct": 20.44, "eta": "45:32:17", "max_grad_norm": 1.0, "loss": 0.6996226906776428, "grad_norm": 0.40906062722206116, "learning_rate": 0.00019113113479475783}
+{"ts": "2025-12-27T05:59:13", "event": "train_log", "step": 2908, "epoch": 1.2270042194092827, "progress_pct": 20.45, "epoch_pct": 20.45, "eta": "45:30:54", "max_grad_norm": 1.0, "loss": 0.8364415168762207, "grad_norm": 0.517085075378418, "learning_rate": 0.00019111134978629633}
+{"ts": "2025-12-27T05:59:28", "event": "train_log", "step": 2910, "epoch": 1.2278481012658227, "progress_pct": 20.46, "epoch_pct": 20.46, "eta": "45:29:33", "max_grad_norm": 1.0, "loss": 0.722011923789978, "grad_norm": 0.425581693649292, "learning_rate": 0.00019109154376019624}
+{"ts": "2025-12-27T05:59:43", "event": "train_log", "step": 2912, "epoch": 1.2286919831223628, "progress_pct": 20.48, "epoch_pct": 20.48, "eta": "45:28:09", "max_grad_norm": 1.0, "loss": 0.7586172223091125, "grad_norm": 0.4651593565940857, "learning_rate": 0.00019107171672102643}
+{"ts": "2025-12-27T05:59:58", "event": "train_log", "step": 2914, "epoch": 1.229535864978903, "progress_pct": 20.49, "epoch_pct": 20.49, "eta": "45:26:46", "max_grad_norm": 1.0, "loss": 0.7387955188751221, "grad_norm": 0.4578891396522522, "learning_rate": 0.00019105186867336067}
+{"ts": "2025-12-27T06:00:14", "event": "train_log", "step": 2916, "epoch": 1.230379746835443, "progress_pct": 20.51, "epoch_pct": 20.51, "eta": "45:25:26", "max_grad_norm": 1.0, "loss": 0.8036285638809204, "grad_norm": 0.4312961995601654, "learning_rate": 0.00019103199962177748}
+{"ts": "2025-12-27T06:00:29", "event": "train_log", "step": 2918, "epoch": 1.231223628691983, "progress_pct": 20.52, "epoch_pct": 20.52, "eta": "45:24:06", "max_grad_norm": 1.0, "loss": 0.7356550097465515, "grad_norm": 0.4355425536632538, "learning_rate": 0.00019101210957086026}
+{"ts": "2025-12-27T06:00:45", "event": "train_log", "step": 2920, "epoch": 1.2320675105485233, "progress_pct": 20.53, "epoch_pct": 20.53, "eta": "45:22:44", "max_grad_norm": 1.0, "loss": 0.6730139255523682, "grad_norm": 0.424675852060318, "learning_rate": 0.00019099219852519733}
+{"ts": "2025-12-27T06:01:01", "event": "train_log", "step": 2922, "epoch": 1.2329113924050632, "progress_pct": 20.55, "epoch_pct": 20.55, "eta": "45:21:25", "max_grad_norm": 1.0, "loss": 0.6983892321586609, "grad_norm": 0.41102367639541626, "learning_rate": 0.0001909722664893817}
+{"ts": "2025-12-27T06:01:17", "event": "train_log", "step": 2924, "epoch": 1.2337552742616034, "progress_pct": 20.56, "epoch_pct": 20.56, "eta": "45:20:08", "max_grad_norm": 1.0, "loss": 0.6297144889831543, "grad_norm": 0.4088250398635864, "learning_rate": 0.00019095231346801132}
+{"ts": "2025-12-27T06:01:33", "event": "train_log", "step": 2926, "epoch": 1.2345991561181435, "progress_pct": 20.58, "epoch_pct": 20.58, "eta": "45:18:50", "max_grad_norm": 1.0, "loss": 0.7035690546035767, "grad_norm": 0.3685890734195709, "learning_rate": 0.00019093233946568906}
+{"ts": "2025-12-27T06:01:50", "event": "train_log", "step": 2928, "epoch": 1.2354430379746835, "progress_pct": 20.59, "epoch_pct": 20.59, "eta": "45:17:33", "max_grad_norm": 1.0, "loss": 0.820853054523468, "grad_norm": 0.39438629150390625, "learning_rate": 0.0001909123444870224}
+{"ts": "2025-12-27T06:02:04", "event": "train_log", "step": 2930, "epoch": 1.2362869198312236, "progress_pct": 20.6, "epoch_pct": 20.6, "eta": "45:16:09", "max_grad_norm": 1.0, "loss": 0.7884082794189453, "grad_norm": 0.5041676163673401, "learning_rate": 0.00019089232853662392}
+{"ts": "2025-12-27T06:02:19", "event": "train_log", "step": 2932, "epoch": 1.2371308016877638, "progress_pct": 20.62, "epoch_pct": 20.62, "eta": "45:14:47", "max_grad_norm": 1.0, "loss": 0.773978590965271, "grad_norm": 0.40592801570892334, "learning_rate": 0.00019087229161911086}
+{"ts": "2025-12-27T06:02:35", "event": "train_log", "step": 2934, "epoch": 1.2379746835443037, "progress_pct": 20.63, "epoch_pct": 20.63, "eta": "45:13:29", "max_grad_norm": 1.0, "loss": 0.6742141842842102, "grad_norm": 0.41828030347824097, "learning_rate": 0.00019085223373910536}
+{"ts": "2025-12-27T06:02:50", "event": "train_log", "step": 2936, "epoch": 1.238818565400844, "progress_pct": 20.65, "epoch_pct": 20.65, "eta": "45:12:05", "max_grad_norm": 1.0, "loss": 0.7106125950813293, "grad_norm": 0.39984026551246643, "learning_rate": 0.0001908321549012344}
+{"ts": "2025-12-27T06:03:05", "event": "train_log", "step": 2938, "epoch": 1.239662447257384, "progress_pct": 20.66, "epoch_pct": 20.66, "eta": "45:10:42", "max_grad_norm": 1.0, "loss": 0.7390915751457214, "grad_norm": 0.4841296970844269, "learning_rate": 0.00019081205511012973}
+{"ts": "2025-12-27T06:03:21", "event": "train_log", "step": 2940, "epoch": 1.240506329113924, "progress_pct": 20.68, "epoch_pct": 20.68, "eta": "45:09:27", "max_grad_norm": 1.0, "loss": 0.6527198553085327, "grad_norm": 0.39837637543678284, "learning_rate": 0.00019079193437042806}
+{"ts": "2025-12-27T06:03:37", "event": "train_log", "step": 2942, "epoch": 1.2413502109704642, "progress_pct": 20.69, "epoch_pct": 20.69, "eta": "45:08:07", "max_grad_norm": 1.0, "loss": 0.6939712762832642, "grad_norm": 0.40916863083839417, "learning_rate": 0.0001907717926867708}
+{"ts": "2025-12-27T06:03:52", "event": "train_log", "step": 2944, "epoch": 1.2421940928270043, "progress_pct": 20.7, "epoch_pct": 20.7, "eta": "45:06:46", "max_grad_norm": 1.0, "loss": 0.7566131353378296, "grad_norm": 0.4329306483268738, "learning_rate": 0.0001907516300638043}
+{"ts": "2025-12-27T06:04:08", "event": "train_log", "step": 2946, "epoch": 1.2430379746835443, "progress_pct": 20.72, "epoch_pct": 20.72, "eta": "45:05:26", "max_grad_norm": 1.0, "loss": 0.7893748879432678, "grad_norm": 0.47983601689338684, "learning_rate": 0.00019073144650617963}
+{"ts": "2025-12-27T06:04:23", "event": "train_log", "step": 2948, "epoch": 1.2438818565400844, "progress_pct": 20.73, "epoch_pct": 20.73, "eta": "45:04:07", "max_grad_norm": 1.0, "loss": 0.8259525299072266, "grad_norm": 0.4368246793746948, "learning_rate": 0.0001907112420185528}
+{"ts": "2025-12-27T06:04:38", "event": "train_log", "step": 2950, "epoch": 1.2447257383966246, "progress_pct": 20.75, "epoch_pct": 20.75, "eta": "45:02:45", "max_grad_norm": 1.0, "loss": 0.734692394733429, "grad_norm": 0.41272372007369995, "learning_rate": 0.0001906910166055846}
+{"ts": "2025-12-27T06:04:53", "event": "train_log", "step": 2952, "epoch": 1.2455696202531645, "progress_pct": 20.76, "epoch_pct": 20.76, "eta": "45:01:23", "max_grad_norm": 1.0, "loss": 0.7091750502586365, "grad_norm": 0.36929839849472046, "learning_rate": 0.00019067077027194057}
+{"ts": "2025-12-27T06:05:08", "event": "train_log", "step": 2954, "epoch": 1.2464135021097047, "progress_pct": 20.77, "epoch_pct": 20.77, "eta": "45:00:01", "max_grad_norm": 1.0, "loss": 0.6960210204124451, "grad_norm": 0.4395252466201782, "learning_rate": 0.00019065050302229125}
+{"ts": "2025-12-27T06:05:22", "event": "train_log", "step": 2956, "epoch": 1.2472573839662446, "progress_pct": 20.79, "epoch_pct": 20.79, "eta": "44:58:36", "max_grad_norm": 1.0, "loss": 0.7817115187644958, "grad_norm": 0.46784868836402893, "learning_rate": 0.00019063021486131186}
+{"ts": "2025-12-27T06:05:37", "event": "train_log", "step": 2958, "epoch": 1.2481012658227848, "progress_pct": 20.8, "epoch_pct": 20.8, "eta": "44:57:14", "max_grad_norm": 1.0, "loss": 0.7933676242828369, "grad_norm": 0.4191588759422302, "learning_rate": 0.00019060990579368244}
+{"ts": "2025-12-27T06:05:51", "event": "train_log", "step": 2960, "epoch": 1.248945147679325, "progress_pct": 20.82, "epoch_pct": 20.82, "eta": "44:55:53", "max_grad_norm": 1.0, "loss": 0.6856688857078552, "grad_norm": 0.3967524468898773, "learning_rate": 0.000190589575824088}
+{"ts": "2025-12-27T06:06:07", "event": "train_log", "step": 2962, "epoch": 1.249789029535865, "progress_pct": 20.83, "epoch_pct": 20.83, "eta": "44:54:33", "max_grad_norm": 1.0, "loss": 0.6783818602561951, "grad_norm": 0.40346091985702515, "learning_rate": 0.0001905692249572182}
+{"ts": "2025-12-27T06:06:22", "event": "train_log", "step": 2964, "epoch": 1.250632911392405, "progress_pct": 20.84, "epoch_pct": 20.84, "eta": "44:53:14", "max_grad_norm": 1.0, "loss": 0.6997019052505493, "grad_norm": 0.38349980115890503, "learning_rate": 0.0001905488531977676}
+{"ts": "2025-12-27T06:06:37", "event": "train_log", "step": 2966, "epoch": 1.2514767932489452, "progress_pct": 20.86, "epoch_pct": 20.86, "eta": "44:51:54", "max_grad_norm": 1.0, "loss": 0.6978800892829895, "grad_norm": 0.44021397829055786, "learning_rate": 0.0001905284605504356}
+{"ts": "2025-12-27T06:06:53", "event": "train_log", "step": 2968, "epoch": 1.2523206751054852, "progress_pct": 20.87, "epoch_pct": 20.87, "eta": "44:50:34", "max_grad_norm": 1.0, "loss": 0.7007693648338318, "grad_norm": 0.47491583228111267, "learning_rate": 0.0001905080470199264}
+{"ts": "2025-12-27T06:07:08", "event": "train_log", "step": 2970, "epoch": 1.2531645569620253, "progress_pct": 20.89, "epoch_pct": 20.89, "eta": "44:49:15", "max_grad_norm": 1.0, "loss": 0.7258896231651306, "grad_norm": 0.4011451005935669, "learning_rate": 0.00019048761261094898}
+{"ts": "2025-12-27T06:07:23", "event": "train_log", "step": 2972, "epoch": 1.2540084388185653, "progress_pct": 20.9, "epoch_pct": 20.9, "eta": "44:47:54", "max_grad_norm": 1.0, "loss": 0.6746299266815186, "grad_norm": 0.44206640124320984, "learning_rate": 0.0001904671573282172}
+{"ts": "2025-12-27T06:07:38", "event": "train_log", "step": 2974, "epoch": 1.2548523206751054, "progress_pct": 20.91, "epoch_pct": 20.91, "eta": "44:46:36", "max_grad_norm": 1.0, "loss": 0.7638527154922485, "grad_norm": 0.4532456696033478, "learning_rate": 0.00019044668117644965}
+{"ts": "2025-12-27T06:07:53", "event": "train_log", "step": 2976, "epoch": 1.2556962025316456, "progress_pct": 20.93, "epoch_pct": 20.93, "eta": "44:45:13", "max_grad_norm": 1.0, "loss": 0.659598171710968, "grad_norm": 0.44301095604896545, "learning_rate": 0.00019042618416036984}
+{"ts": "2025-12-27T06:08:08", "event": "train_log", "step": 2978, "epoch": 1.2565400843881855, "progress_pct": 20.94, "epoch_pct": 20.94, "eta": "44:43:52", "max_grad_norm": 1.0, "loss": 0.7132709622383118, "grad_norm": 0.37427952885627747, "learning_rate": 0.00019040566628470597}
+{"ts": "2025-12-27T06:08:22", "event": "train_log", "step": 2980, "epoch": 1.2573839662447257, "progress_pct": 20.96, "epoch_pct": 20.96, "eta": "44:42:30", "max_grad_norm": 1.0, "loss": 0.8163022398948669, "grad_norm": 0.4195408821105957, "learning_rate": 0.0001903851275541912}
+{"ts": "2025-12-27T06:08:37", "event": "train_log", "step": 2982, "epoch": 1.2582278481012659, "progress_pct": 20.97, "epoch_pct": 20.97, "eta": "44:41:10", "max_grad_norm": 1.0, "loss": 0.7481505870819092, "grad_norm": 0.3951880633831024, "learning_rate": 0.00019036456797356336}
+{"ts": "2025-12-27T06:08:53", "event": "train_log", "step": 2984, "epoch": 1.2590717299578058, "progress_pct": 20.98, "epoch_pct": 20.98, "eta": "44:39:53", "max_grad_norm": 1.0, "loss": 0.750126838684082, "grad_norm": 0.4239100217819214, "learning_rate": 0.00019034398754756522}
+{"ts": "2025-12-27T06:09:08", "event": "train_log", "step": 2986, "epoch": 1.259915611814346, "progress_pct": 21.0, "epoch_pct": 21.0, "eta": "44:38:32", "max_grad_norm": 1.0, "loss": 0.746433436870575, "grad_norm": 0.4209354817867279, "learning_rate": 0.0001903233862809442}
+{"ts": "2025-12-27T06:09:22", "event": "train_log", "step": 2988, "epoch": 1.2607594936708861, "progress_pct": 21.01, "epoch_pct": 21.01, "eta": "44:37:12", "max_grad_norm": 1.0, "loss": 0.7104454040527344, "grad_norm": 0.4188138544559479, "learning_rate": 0.00019030276417845266}
+{"ts": "2025-12-27T06:09:38", "event": "train_log", "step": 2990, "epoch": 1.261603375527426, "progress_pct": 21.03, "epoch_pct": 21.03, "eta": "44:35:52", "max_grad_norm": 1.0, "loss": 0.650738000869751, "grad_norm": 0.4531371593475342, "learning_rate": 0.00019028212124484772}
+{"ts": "2025-12-27T06:09:53", "event": "train_log", "step": 2992, "epoch": 1.2624472573839662, "progress_pct": 21.04, "epoch_pct": 21.04, "eta": "44:34:35", "max_grad_norm": 1.0, "loss": 0.6870551109313965, "grad_norm": 0.4672204256057739, "learning_rate": 0.00019026145748489133}
+{"ts": "2025-12-27T06:10:08", "event": "train_log", "step": 2994, "epoch": 1.2632911392405064, "progress_pct": 21.05, "epoch_pct": 21.05, "eta": "44:33:16", "max_grad_norm": 1.0, "loss": 0.6879403591156006, "grad_norm": 0.40681278705596924, "learning_rate": 0.00019024077290335018}
+{"ts": "2025-12-27T06:10:23", "event": "train_log", "step": 2996, "epoch": 1.2641350210970463, "progress_pct": 21.07, "epoch_pct": 21.07, "eta": "44:31:56", "max_grad_norm": 1.0, "loss": 0.6886372566223145, "grad_norm": 0.4491691291332245, "learning_rate": 0.00019022006750499585}
+{"ts": "2025-12-27T06:10:38", "event": "train_log", "step": 2998, "epoch": 1.2649789029535865, "progress_pct": 21.08, "epoch_pct": 21.08, "eta": "44:30:36", "max_grad_norm": 1.0, "loss": 0.7199324369430542, "grad_norm": 0.4144807457923889, "learning_rate": 0.0001901993412946046}
+{"ts": "2025-12-27T06:10:53", "event": "train_log", "step": 3000, "epoch": 1.2658227848101267, "progress_pct": 21.1, "epoch_pct": 21.1, "eta": "44:29:16", "max_grad_norm": 1.0, "loss": 0.7105490565299988, "grad_norm": 0.40370482206344604, "learning_rate": 0.0001901785942769577}
+{"ts": "2025-12-27T06:22:11", "event": "train_log", "step": 3000, "epoch": 1.2658227848101267, "progress_pct": 21.1, "epoch_pct": 21.1, "eta": "45:11:34", "max_grad_norm": 1.0, "eval_loss": 0.8033165335655212, "eval_runtime": 678.7554, "eval_samples_per_second": 3.104, "eval_steps_per_second": 3.104}
+{"ts": "2025-12-27T06:22:28", "event": "train_log", "step": 3002, "epoch": 1.2666666666666666, "progress_pct": 21.11, "epoch_pct": 21.11, "eta": "45:10:18", "max_grad_norm": 1.0, "loss": 0.7385403513908386, "grad_norm": 0.42637890577316284, "learning_rate": 0.00019015782645684097}
+{"ts": "2025-12-27T06:22:45", "event": "train_log", "step": 3004, "epoch": 1.2675105485232068, "progress_pct": 21.13, "epoch_pct": 21.13, "eta": "45:09:04", "max_grad_norm": 1.0, "loss": 0.6395129561424255, "grad_norm": 0.3709809482097626, "learning_rate": 0.00019013703783904518}
+{"ts": "2025-12-27T06:23:00", "event": "train_log", "step": 3006, "epoch": 1.268354430379747, "progress_pct": 21.14, "epoch_pct": 21.14, "eta": "45:07:45", "max_grad_norm": 1.0, "loss": 0.7123657464981079, "grad_norm": 0.4642501175403595, "learning_rate": 0.00019011622842836592}
+{"ts": "2025-12-27T06:23:15", "event": "train_log", "step": 3008, "epoch": 1.2691983122362869, "progress_pct": 21.15, "epoch_pct": 21.15, "eta": "45:06:23", "max_grad_norm": 1.0, "loss": 0.744048535823822, "grad_norm": 0.4553911089897156, "learning_rate": 0.00019009539822960342}
+{"ts": "2025-12-27T06:23:31", "event": "train_log", "step": 3010, "epoch": 1.270042194092827, "progress_pct": 21.17, "epoch_pct": 21.17, "eta": "45:05:07", "max_grad_norm": 1.0, "loss": 0.7181914448738098, "grad_norm": 0.39334386587142944, "learning_rate": 0.0001900745472475629}
+{"ts": "2025-12-27T06:23:47", "event": "train_log", "step": 3012, "epoch": 1.2708860759493672, "progress_pct": 21.18, "epoch_pct": 21.18, "eta": "45:03:50", "max_grad_norm": 1.0, "loss": 0.7238107919692993, "grad_norm": 0.40902766585350037, "learning_rate": 0.00019005367548705426}
+{"ts": "2025-12-27T06:24:03", "event": "train_log", "step": 3014, "epoch": 1.2717299578059071, "progress_pct": 21.2, "epoch_pct": 21.2, "eta": "45:02:30", "max_grad_norm": 1.0, "loss": 0.7479744553565979, "grad_norm": 0.48296019434928894, "learning_rate": 0.00019003278295289216}
+{"ts": "2025-12-27T06:24:18", "event": "train_log", "step": 3016, "epoch": 1.2725738396624473, "progress_pct": 21.21, "epoch_pct": 21.21, "eta": "45:01:11", "max_grad_norm": 1.0, "loss": 0.829924464225769, "grad_norm": 0.4807029962539673, "learning_rate": 0.00019001186964989617}
+{"ts": "2025-12-27T06:24:34", "event": "train_log", "step": 3018, "epoch": 1.2734177215189875, "progress_pct": 21.22, "epoch_pct": 21.22, "eta": "44:59:52", "max_grad_norm": 1.0, "loss": 0.7204707264900208, "grad_norm": 0.4239083528518677, "learning_rate": 0.00018999093558289055}
+{"ts": "2025-12-27T06:24:50", "event": "train_log", "step": 3020, "epoch": 1.2742616033755274, "progress_pct": 21.24, "epoch_pct": 21.24, "eta": "44:58:36", "max_grad_norm": 1.0, "loss": 0.7668164968490601, "grad_norm": 0.4687928259372711, "learning_rate": 0.0001899699807567044}
+{"ts": "2025-12-27T06:25:06", "event": "train_log", "step": 3022, "epoch": 1.2751054852320676, "progress_pct": 21.25, "epoch_pct": 21.25, "eta": "44:57:19", "max_grad_norm": 1.0, "loss": 0.766342043876648, "grad_norm": 0.41724732518196106, "learning_rate": 0.0001899490051761716}
+{"ts": "2025-12-27T06:25:22", "event": "train_log", "step": 3024, "epoch": 1.2759493670886077, "progress_pct": 21.27, "epoch_pct": 21.27, "eta": "44:56:02", "max_grad_norm": 1.0, "loss": 0.6595507860183716, "grad_norm": 0.4131613075733185, "learning_rate": 0.00018992800884613082}
+{"ts": "2025-12-27T06:25:38", "event": "train_log", "step": 3026, "epoch": 1.2767932489451477, "progress_pct": 21.28, "epoch_pct": 21.28, "eta": "44:54:47", "max_grad_norm": 1.0, "loss": 0.7419850826263428, "grad_norm": 0.4156053364276886, "learning_rate": 0.00018990699177142548}
+{"ts": "2025-12-27T06:25:56", "event": "train_log", "step": 3028, "epoch": 1.2776371308016878, "progress_pct": 21.29, "epoch_pct": 21.29, "eta": "44:53:37", "max_grad_norm": 1.0, "loss": 0.67264324426651, "grad_norm": 0.41845977306365967, "learning_rate": 0.00018988595395690388}
+{"ts": "2025-12-27T06:26:12", "event": "train_log", "step": 3030, "epoch": 1.2784810126582278, "progress_pct": 21.31, "epoch_pct": 21.31, "eta": "44:52:21", "max_grad_norm": 1.0, "loss": 0.6269281506538391, "grad_norm": 0.3826757073402405, "learning_rate": 0.00018986489540741893}
+{"ts": "2025-12-27T06:26:28", "event": "train_log", "step": 3032, "epoch": 1.279324894514768, "progress_pct": 21.32, "epoch_pct": 21.32, "eta": "44:51:05", "max_grad_norm": 1.0, "loss": 0.71211177110672, "grad_norm": 0.4531428813934326, "learning_rate": 0.00018984381612782854}
+{"ts": "2025-12-27T06:26:42", "event": "train_log", "step": 3034, "epoch": 1.2801687763713079, "progress_pct": 21.34, "epoch_pct": 21.34, "eta": "44:49:43", "max_grad_norm": 1.0, "loss": 0.635804295539856, "grad_norm": 0.3747255802154541, "learning_rate": 0.00018982271612299524}
+{"ts": "2025-12-27T06:26:59", "event": "train_log", "step": 3036, "epoch": 1.281012658227848, "progress_pct": 21.35, "epoch_pct": 21.35, "eta": "44:48:30", "max_grad_norm": 1.0, "loss": 0.7269464731216431, "grad_norm": 0.4079552888870239, "learning_rate": 0.00018980159539778638}
+{"ts": "2025-12-27T06:27:15", "event": "train_log", "step": 3038, "epoch": 1.2818565400843882, "progress_pct": 21.36, "epoch_pct": 21.36, "eta": "44:47:13", "max_grad_norm": 1.0, "loss": 0.7737077474594116, "grad_norm": 0.4193854033946991, "learning_rate": 0.00018978045395707418}
+{"ts": "2025-12-27T06:27:31", "event": "train_log", "step": 3040, "epoch": 1.2827004219409281, "progress_pct": 21.38, "epoch_pct": 21.38, "eta": "44:45:55", "max_grad_norm": 1.0, "loss": 0.7458174824714661, "grad_norm": 0.4182516038417816, "learning_rate": 0.0001897592918057355}
+{"ts": "2025-12-27T06:27:46", "event": "train_log", "step": 3042, "epoch": 1.2835443037974683, "progress_pct": 21.39, "epoch_pct": 21.39, "eta": "44:44:36", "max_grad_norm": 1.0, "loss": 0.6565808057785034, "grad_norm": 0.43343567848205566, "learning_rate": 0.00018973810894865205}
+{"ts": "2025-12-27T06:28:02", "event": "train_log", "step": 3044, "epoch": 1.2843881856540085, "progress_pct": 21.41, "epoch_pct": 21.41, "eta": "44:43:21", "max_grad_norm": 1.0, "loss": 0.736944317817688, "grad_norm": 0.4294503331184387, "learning_rate": 0.00018971690539071033}
+{"ts": "2025-12-27T06:28:18", "event": "train_log", "step": 3046, "epoch": 1.2852320675105484, "progress_pct": 21.42, "epoch_pct": 21.42, "eta": "44:42:06", "max_grad_norm": 1.0, "loss": 0.6921742558479309, "grad_norm": 0.41330763697624207, "learning_rate": 0.0001896956811368016}
+{"ts": "2025-12-27T06:28:33", "event": "train_log", "step": 3048, "epoch": 1.2860759493670886, "progress_pct": 21.43, "epoch_pct": 21.43, "eta": "44:40:47", "max_grad_norm": 1.0, "loss": 0.7004357576370239, "grad_norm": 0.44992703199386597, "learning_rate": 0.0001896744361918218}
+{"ts": "2025-12-27T06:28:50", "event": "train_log", "step": 3050, "epoch": 1.2869198312236287, "progress_pct": 21.45, "epoch_pct": 21.45, "eta": "44:39:35", "max_grad_norm": 1.0, "loss": 0.6767720580101013, "grad_norm": 0.43748870491981506, "learning_rate": 0.0001896531705606719}
+{"ts": "2025-12-27T06:29:06", "event": "train_log", "step": 3052, "epoch": 1.2877637130801687, "progress_pct": 21.46, "epoch_pct": 21.46, "eta": "44:38:18", "max_grad_norm": 1.0, "loss": 0.7596892714500427, "grad_norm": 0.47581973671913147, "learning_rate": 0.0001896318842482573}
+{"ts": "2025-12-27T06:29:22", "event": "train_log", "step": 3054, "epoch": 1.2886075949367088, "progress_pct": 21.48, "epoch_pct": 21.48, "eta": "44:37:02", "max_grad_norm": 1.0, "loss": 0.7650044560432434, "grad_norm": 0.45806005597114563, "learning_rate": 0.00018961057725948844}
+{"ts": "2025-12-27T06:29:36", "event": "train_log", "step": 3056, "epoch": 1.289451476793249, "progress_pct": 21.49, "epoch_pct": 21.49, "eta": "44:35:41", "max_grad_norm": 1.0, "loss": 0.6654610633850098, "grad_norm": 0.4291481375694275, "learning_rate": 0.00018958924959928042}
+{"ts": "2025-12-27T06:29:52", "event": "train_log", "step": 3058, "epoch": 1.290295358649789, "progress_pct": 21.5, "epoch_pct": 21.5, "eta": "44:34:26", "max_grad_norm": 1.0, "loss": 0.7079797387123108, "grad_norm": 0.42454418540000916, "learning_rate": 0.00018956790127255309}
+{"ts": "2025-12-27T06:30:07", "event": "train_log", "step": 3060, "epoch": 1.2911392405063291, "progress_pct": 21.52, "epoch_pct": 21.52, "eta": "44:33:07", "max_grad_norm": 1.0, "loss": 0.7724313735961914, "grad_norm": 0.46820637583732605, "learning_rate": 0.00018954653228423114}
+{"ts": "2025-12-27T06:30:23", "event": "train_log", "step": 3062, "epoch": 1.2919831223628693, "progress_pct": 21.53, "epoch_pct": 21.53, "eta": "44:31:52", "max_grad_norm": 1.0, "loss": 0.7746338248252869, "grad_norm": 0.4440016746520996, "learning_rate": 0.00018952514263924396}
+{"ts": "2025-12-27T06:30:39", "event": "train_log", "step": 3064, "epoch": 1.2928270042194092, "progress_pct": 21.55, "epoch_pct": 21.55, "eta": "44:30:36", "max_grad_norm": 1.0, "loss": 0.7570929527282715, "grad_norm": 0.4296419620513916, "learning_rate": 0.00018950373234252572}
+{"ts": "2025-12-27T06:30:55", "event": "train_log", "step": 3066, "epoch": 1.2936708860759494, "progress_pct": 21.56, "epoch_pct": 21.56, "eta": "44:29:20", "max_grad_norm": 1.0, "loss": 0.7274802923202515, "grad_norm": 0.428780734539032, "learning_rate": 0.0001894823013990154}
+{"ts": "2025-12-27T06:31:10", "event": "train_log", "step": 3068, "epoch": 1.2945147679324895, "progress_pct": 21.58, "epoch_pct": 21.58, "eta": "44:28:02", "max_grad_norm": 1.0, "loss": 0.6595024466514587, "grad_norm": 0.4111585021018982, "learning_rate": 0.0001894608498136567}
+{"ts": "2025-12-27T06:31:25", "event": "train_log", "step": 3070, "epoch": 1.2953586497890295, "progress_pct": 21.59, "epoch_pct": 21.59, "eta": "44:26:43", "max_grad_norm": 1.0, "loss": 0.7316440939903259, "grad_norm": 0.4373013973236084, "learning_rate": 0.00018943937759139802}
+{"ts": "2025-12-27T06:31:40", "event": "train_log", "step": 3072, "epoch": 1.2962025316455696, "progress_pct": 21.6, "epoch_pct": 21.6, "eta": "44:25:26", "max_grad_norm": 1.0, "loss": 0.7072293162345886, "grad_norm": 0.37705400586128235, "learning_rate": 0.0001894178847371927}
+{"ts": "2025-12-27T06:31:56", "event": "train_log", "step": 3074, "epoch": 1.2970464135021098, "progress_pct": 21.62, "epoch_pct": 21.62, "eta": "44:24:09", "max_grad_norm": 1.0, "loss": 0.7218101024627686, "grad_norm": 0.4171125888824463, "learning_rate": 0.00018939637125599864}
+{"ts": "2025-12-27T06:32:11", "event": "train_log", "step": 3076, "epoch": 1.2978902953586497, "progress_pct": 21.63, "epoch_pct": 21.63, "eta": "44:22:53", "max_grad_norm": 1.0, "loss": 0.725140392780304, "grad_norm": 0.4264235198497772, "learning_rate": 0.00018937483715277865}
+{"ts": "2025-12-27T06:32:27", "event": "train_log", "step": 3078, "epoch": 1.29873417721519, "progress_pct": 21.65, "epoch_pct": 21.65, "eta": "44:21:38", "max_grad_norm": 1.0, "loss": 0.7423256635665894, "grad_norm": 0.4251643121242523, "learning_rate": 0.0001893532824325002}
+{"ts": "2025-12-27T06:32:44", "event": "train_log", "step": 3080, "epoch": 1.29957805907173, "progress_pct": 21.66, "epoch_pct": 21.66, "eta": "44:20:26", "max_grad_norm": 1.0, "loss": 0.7058811187744141, "grad_norm": 0.43096452951431274, "learning_rate": 0.00018933170710013554}
+{"ts": "2025-12-27T06:33:00", "event": "train_log", "step": 3082, "epoch": 1.30042194092827, "progress_pct": 21.67, "epoch_pct": 21.67, "eta": "44:19:13", "max_grad_norm": 1.0, "loss": 0.7334373593330383, "grad_norm": 0.466804176568985, "learning_rate": 0.00018931011116066172}
+{"ts": "2025-12-27T06:33:18", "event": "train_log", "step": 3084, "epoch": 1.3012658227848102, "progress_pct": 21.69, "epoch_pct": 21.69, "eta": "44:18:03", "max_grad_norm": 1.0, "loss": 0.6033462285995483, "grad_norm": 0.36896654963493347, "learning_rate": 0.00018928849461906047}
+{"ts": "2025-12-27T06:33:34", "event": "train_log", "step": 3086, "epoch": 1.3021097046413503, "progress_pct": 21.7, "epoch_pct": 21.7, "eta": "44:16:50", "max_grad_norm": 1.0, "loss": 0.690375566482544, "grad_norm": 0.42875027656555176, "learning_rate": 0.00018926685748031833}
+{"ts": "2025-12-27T06:33:49", "event": "train_log", "step": 3088, "epoch": 1.3029535864978903, "progress_pct": 21.72, "epoch_pct": 21.72, "eta": "44:15:32", "max_grad_norm": 1.0, "loss": 0.7468435168266296, "grad_norm": 0.4041171967983246, "learning_rate": 0.00018924519974942658}
+{"ts": "2025-12-27T06:34:04", "event": "train_log", "step": 3090, "epoch": 1.3037974683544304, "progress_pct": 21.73, "epoch_pct": 21.73, "eta": "44:14:13", "max_grad_norm": 1.0, "loss": 0.7574344873428345, "grad_norm": 0.4704097807407379, "learning_rate": 0.00018922352143138126}
+{"ts": "2025-12-27T06:34:19", "event": "train_log", "step": 3092, "epoch": 1.3046413502109704, "progress_pct": 21.74, "epoch_pct": 21.74, "eta": "44:12:58", "max_grad_norm": 1.0, "loss": 0.756452739238739, "grad_norm": 0.4391418397426605, "learning_rate": 0.0001892018225311831}
+{"ts": "2025-12-27T06:34:34", "event": "train_log", "step": 3094, "epoch": 1.3054852320675105, "progress_pct": 21.76, "epoch_pct": 21.76, "eta": "44:11:42", "max_grad_norm": 1.0, "loss": 0.6620991230010986, "grad_norm": 0.42684388160705566, "learning_rate": 0.0001891801030538376}
+{"ts": "2025-12-27T06:34:50", "event": "train_log", "step": 3096, "epoch": 1.3063291139240507, "progress_pct": 21.77, "epoch_pct": 21.77, "eta": "44:10:27", "max_grad_norm": 1.0, "loss": 0.7764125466346741, "grad_norm": 0.4972277283668518, "learning_rate": 0.00018915836300435507}
+{"ts": "2025-12-27T06:35:06", "event": "train_log", "step": 3098, "epoch": 1.3071729957805907, "progress_pct": 21.79, "epoch_pct": 21.79, "eta": "44:09:14", "max_grad_norm": 1.0, "loss": 0.6557663083076477, "grad_norm": 0.3905445337295532, "learning_rate": 0.00018913660238775053}
+{"ts": "2025-12-27T06:35:21", "event": "train_log", "step": 3100, "epoch": 1.3080168776371308, "progress_pct": 21.8, "epoch_pct": 21.8, "eta": "44:07:56", "max_grad_norm": 1.0, "loss": 0.6746718883514404, "grad_norm": 0.4553062319755554, "learning_rate": 0.0001891148212090437}
+{"ts": "2025-12-27T06:46:29", "event": "train_log", "step": 3100, "epoch": 1.3080168776371308, "progress_pct": 21.8, "epoch_pct": 21.8, "eta": "44:47:53", "max_grad_norm": 1.0, "eval_loss": 0.8010181784629822, "eval_runtime": 668.1688, "eval_samples_per_second": 3.153, "eval_steps_per_second": 3.153}
+{"ts": "2025-12-27T06:46:45", "event": "train_log", "step": 3102, "epoch": 1.3088607594936708, "progress_pct": 21.81, "epoch_pct": 21.81, "eta": "44:46:35", "max_grad_norm": 1.0, "loss": 0.7916419506072998, "grad_norm": 0.4254429340362549, "learning_rate": 0.00018909301947325915}
+{"ts": "2025-12-27T06:46:59", "event": "train_log", "step": 3104, "epoch": 1.309704641350211, "progress_pct": 21.83, "epoch_pct": 21.83, "eta": "44:45:13", "max_grad_norm": 1.0, "loss": 0.883936882019043, "grad_norm": 0.6118508577346802, "learning_rate": 0.00018907119718542602}
+{"ts": "2025-12-27T06:47:14", "event": "train_log", "step": 3106, "epoch": 1.310548523206751, "progress_pct": 21.84, "epoch_pct": 21.84, "eta": "44:43:54", "max_grad_norm": 1.0, "loss": 0.7145928740501404, "grad_norm": 0.43206971883773804, "learning_rate": 0.00018904935435057837}
+{"ts": "2025-12-27T06:47:30", "event": "train_log", "step": 3108, "epoch": 1.311392405063291, "progress_pct": 21.86, "epoch_pct": 21.86, "eta": "44:42:37", "max_grad_norm": 1.0, "loss": 0.7442842125892639, "grad_norm": 0.39657703042030334, "learning_rate": 0.0001890274909737549}
+{"ts": "2025-12-27T06:47:45", "event": "train_log", "step": 3110, "epoch": 1.3122362869198312, "progress_pct": 21.87, "epoch_pct": 21.87, "eta": "44:41:20", "max_grad_norm": 1.0, "loss": 0.6815276145935059, "grad_norm": 0.4087927043437958, "learning_rate": 0.00018900560705999904}
+{"ts": "2025-12-27T06:48:00", "event": "train_log", "step": 3112, "epoch": 1.3130801687763713, "progress_pct": 21.88, "epoch_pct": 21.88, "eta": "44:39:59", "max_grad_norm": 1.0, "loss": 0.8153640627861023, "grad_norm": 0.430259108543396, "learning_rate": 0.00018898370261435904}
+{"ts": "2025-12-27T06:48:14", "event": "train_log", "step": 3114, "epoch": 1.3139240506329113, "progress_pct": 21.9, "epoch_pct": 21.9, "eta": "44:38:38", "max_grad_norm": 1.0, "loss": 0.7828266024589539, "grad_norm": 0.47486868500709534, "learning_rate": 0.0001889617776418878}
+{"ts": "2025-12-27T06:48:29", "event": "train_log", "step": 3116, "epoch": 1.3147679324894515, "progress_pct": 21.91, "epoch_pct": 21.91, "eta": "44:37:19", "max_grad_norm": 1.0, "loss": 0.7849065661430359, "grad_norm": 0.4929087162017822, "learning_rate": 0.000188939832147643}
+{"ts": "2025-12-27T06:48:44", "event": "train_log", "step": 3118, "epoch": 1.3156118143459916, "progress_pct": 21.93, "epoch_pct": 21.93, "eta": "44:36:03", "max_grad_norm": 1.0, "loss": 0.7262985706329346, "grad_norm": 0.39038679003715515, "learning_rate": 0.00018891786613668704}
+{"ts": "2025-12-27T06:48:59", "event": "train_log", "step": 3120, "epoch": 1.3164556962025316, "progress_pct": 21.94, "epoch_pct": 21.94, "eta": "44:34:42", "max_grad_norm": 1.0, "loss": 0.8121856451034546, "grad_norm": 0.4827214479446411, "learning_rate": 0.0001888958796140871}
+{"ts": "2025-12-27T06:49:13", "event": "train_log", "step": 3122, "epoch": 1.3172995780590717, "progress_pct": 21.95, "epoch_pct": 21.95, "eta": "44:33:22", "max_grad_norm": 1.0, "loss": 0.7165566086769104, "grad_norm": 0.47739601135253906, "learning_rate": 0.000188873872584915}
+{"ts": "2025-12-27T06:49:28", "event": "train_log", "step": 3124, "epoch": 1.3181434599156119, "progress_pct": 21.97, "epoch_pct": 21.97, "eta": "44:32:01", "max_grad_norm": 1.0, "loss": 0.6698412299156189, "grad_norm": 0.4274037778377533, "learning_rate": 0.0001888518450542473}
+{"ts": "2025-12-27T06:49:44", "event": "train_log", "step": 3126, "epoch": 1.3189873417721518, "progress_pct": 21.98, "epoch_pct": 21.98, "eta": "44:30:50", "max_grad_norm": 1.0, "loss": 0.7271128296852112, "grad_norm": 0.44825276732444763, "learning_rate": 0.00018882979702716544}
+{"ts": "2025-12-27T06:50:01", "event": "train_log", "step": 3128, "epoch": 1.319831223628692, "progress_pct": 22.0, "epoch_pct": 22.0, "eta": "44:29:38", "max_grad_norm": 1.0, "loss": 0.7924301624298096, "grad_norm": 0.42956918478012085, "learning_rate": 0.00018880772850875537}
+{"ts": "2025-12-27T06:50:16", "event": "train_log", "step": 3130, "epoch": 1.3206751054852321, "progress_pct": 22.01, "epoch_pct": 22.01, "eta": "44:28:21", "max_grad_norm": 1.0, "loss": 0.7846497893333435, "grad_norm": 0.4848136305809021, "learning_rate": 0.00018878563950410792}
+{"ts": "2025-12-27T06:50:32", "event": "train_log", "step": 3132, "epoch": 1.321518987341772, "progress_pct": 22.03, "epoch_pct": 22.03, "eta": "44:27:04", "max_grad_norm": 1.0, "loss": 0.7262214422225952, "grad_norm": 0.46651726961135864, "learning_rate": 0.00018876353001831864}
+{"ts": "2025-12-27T06:50:47", "event": "train_log", "step": 3134, "epoch": 1.3223628691983123, "progress_pct": 22.04, "epoch_pct": 22.04, "eta": "44:25:46", "max_grad_norm": 1.0, "loss": 0.7640091180801392, "grad_norm": 0.44598284363746643, "learning_rate": 0.0001887414000564877}
+{"ts": "2025-12-27T06:51:03", "event": "train_log", "step": 3136, "epoch": 1.3232067510548524, "progress_pct": 22.05, "epoch_pct": 22.05, "eta": "44:24:33", "max_grad_norm": 1.0, "loss": 0.6960080862045288, "grad_norm": 0.43744903802871704, "learning_rate": 0.00018871924962372006}
+{"ts": "2025-12-27T06:51:17", "event": "train_log", "step": 3138, "epoch": 1.3240506329113924, "progress_pct": 22.07, "epoch_pct": 22.07, "eta": "44:23:13", "max_grad_norm": 1.0, "loss": 0.7121244072914124, "grad_norm": 0.4737822413444519, "learning_rate": 0.00018869707872512544}
+{"ts": "2025-12-27T06:51:33", "event": "train_log", "step": 3140, "epoch": 1.3248945147679325, "progress_pct": 22.08, "epoch_pct": 22.08, "eta": "44:21:58", "max_grad_norm": 1.0, "loss": 0.713148295879364, "grad_norm": 0.44452038407325745, "learning_rate": 0.00018867488736581817}
+{"ts": "2025-12-27T06:51:48", "event": "train_log", "step": 3142, "epoch": 1.3257383966244727, "progress_pct": 22.1, "epoch_pct": 22.1, "eta": "44:20:40", "max_grad_norm": 1.0, "loss": 0.6904974579811096, "grad_norm": 0.47142693400382996, "learning_rate": 0.0001886526755509175}
+{"ts": "2025-12-27T06:52:05", "event": "train_log", "step": 3144, "epoch": 1.3265822784810126, "progress_pct": 22.11, "epoch_pct": 22.11, "eta": "44:19:27", "max_grad_norm": 1.0, "loss": 0.7319386005401611, "grad_norm": 0.40572887659072876, "learning_rate": 0.00018863044328554712}
+{"ts": "2025-12-27T06:52:20", "event": "train_log", "step": 3146, "epoch": 1.3274261603375528, "progress_pct": 22.12, "epoch_pct": 22.12, "eta": "44:18:12", "max_grad_norm": 1.0, "loss": 0.7268120646476746, "grad_norm": 0.44173532724380493, "learning_rate": 0.00018860819057483566}
+{"ts": "2025-12-27T06:52:36", "event": "train_log", "step": 3148, "epoch": 1.328270042194093, "progress_pct": 22.14, "epoch_pct": 22.14, "eta": "44:16:57", "max_grad_norm": 1.0, "loss": 0.7129524946212769, "grad_norm": 0.43025046586990356, "learning_rate": 0.00018858591742391642}
+{"ts": "2025-12-27T06:52:52", "event": "train_log", "step": 3150, "epoch": 1.3291139240506329, "progress_pct": 22.15, "epoch_pct": 22.15, "eta": "44:15:43", "max_grad_norm": 1.0, "loss": 0.6917402148246765, "grad_norm": 0.41418036818504333, "learning_rate": 0.00018856362383792732}
+{"ts": "2025-12-27T06:53:08", "event": "train_log", "step": 3152, "epoch": 1.329957805907173, "progress_pct": 22.17, "epoch_pct": 22.17, "eta": "44:14:29", "max_grad_norm": 1.0, "loss": 0.7653918862342834, "grad_norm": 0.46275854110717773, "learning_rate": 0.00018854130982201111}
+{"ts": "2025-12-27T06:53:25", "event": "train_log", "step": 3154, "epoch": 1.3308016877637132, "progress_pct": 22.18, "epoch_pct": 22.18, "eta": "44:13:20", "max_grad_norm": 1.0, "loss": 0.7311477065086365, "grad_norm": 0.38927653431892395, "learning_rate": 0.0001885189753813152}
+{"ts": "2025-12-27T06:53:41", "event": "train_log", "step": 3156, "epoch": 1.3316455696202532, "progress_pct": 22.19, "epoch_pct": 22.19, "eta": "44:12:05", "max_grad_norm": 1.0, "loss": 0.7978816628456116, "grad_norm": 0.44294974207878113, "learning_rate": 0.0001884966205209917}
+{"ts": "2025-12-27T06:53:57", "event": "train_log", "step": 3158, "epoch": 1.3324894514767933, "progress_pct": 22.21, "epoch_pct": 22.21, "eta": "44:10:53", "max_grad_norm": 1.0, "loss": 0.7497738599777222, "grad_norm": 0.3772844970226288, "learning_rate": 0.00018847424524619745}
+{"ts": "2025-12-27T06:54:13", "event": "train_log", "step": 3160, "epoch": 1.3333333333333333, "progress_pct": 22.22, "epoch_pct": 22.22, "eta": "44:09:38", "max_grad_norm": 1.0, "loss": 0.712806761264801, "grad_norm": 0.4538787603378296, "learning_rate": 0.000188451849562094}
+{"ts": "2025-12-27T06:54:28", "event": "train_log", "step": 3162, "epoch": 1.3341772151898734, "progress_pct": 22.24, "epoch_pct": 22.24, "eta": "44:08:23", "max_grad_norm": 1.0, "loss": 0.7282580137252808, "grad_norm": 0.4194336533546448, "learning_rate": 0.00018842943347384763}
+{"ts": "2025-12-27T06:54:44", "event": "train_log", "step": 3164, "epoch": 1.3350210970464136, "progress_pct": 22.25, "epoch_pct": 22.25, "eta": "44:07:08", "max_grad_norm": 1.0, "loss": 0.797752320766449, "grad_norm": 0.4186438024044037, "learning_rate": 0.00018840699698662925}
+{"ts": "2025-12-27T06:54:59", "event": "train_log", "step": 3166, "epoch": 1.3358649789029535, "progress_pct": 22.26, "epoch_pct": 22.26, "eta": "44:05:51", "max_grad_norm": 1.0, "loss": 0.8175379633903503, "grad_norm": 0.47711506485939026, "learning_rate": 0.00018838454010561457}
+{"ts": "2025-12-27T06:55:14", "event": "train_log", "step": 3168, "epoch": 1.3367088607594937, "progress_pct": 22.28, "epoch_pct": 22.28, "eta": "44:04:37", "max_grad_norm": 1.0, "loss": 0.659176766872406, "grad_norm": 0.4380176067352295, "learning_rate": 0.00018836206283598394}
+{"ts": "2025-12-27T06:55:32", "event": "train_log", "step": 3170, "epoch": 1.3375527426160336, "progress_pct": 22.29, "epoch_pct": 22.29, "eta": "44:03:29", "max_grad_norm": 1.0, "loss": 0.732314944267273, "grad_norm": 0.4505249559879303, "learning_rate": 0.00018833956518292245}
+{"ts": "2025-12-27T06:55:47", "event": "train_log", "step": 3172, "epoch": 1.3383966244725738, "progress_pct": 22.31, "epoch_pct": 22.31, "eta": "44:02:15", "max_grad_norm": 1.0, "loss": 0.6791377067565918, "grad_norm": 0.3832913935184479, "learning_rate": 0.00018831704715161984}
+{"ts": "2025-12-27T06:56:03", "event": "train_log", "step": 3174, "epoch": 1.339240506329114, "progress_pct": 22.32, "epoch_pct": 22.32, "eta": "44:00:59", "max_grad_norm": 1.0, "loss": 0.7395796775817871, "grad_norm": 0.42996498942375183, "learning_rate": 0.00018829450874727064}
+{"ts": "2025-12-27T06:56:21", "event": "train_log", "step": 3176, "epoch": 1.340084388185654, "progress_pct": 22.33, "epoch_pct": 22.33, "eta": "43:59:53", "max_grad_norm": 1.0, "loss": 0.6893631219863892, "grad_norm": 0.3926190435886383, "learning_rate": 0.000188271949975074}
+{"ts": "2025-12-27T06:56:37", "event": "train_log", "step": 3178, "epoch": 1.340928270042194, "progress_pct": 22.35, "epoch_pct": 22.35, "eta": "43:58:42", "max_grad_norm": 1.0, "loss": 0.6708382964134216, "grad_norm": 0.4353938102722168, "learning_rate": 0.0001882493708402338}
+{"ts": "2025-12-27T06:56:52", "event": "train_log", "step": 3180, "epoch": 1.3417721518987342, "progress_pct": 22.36, "epoch_pct": 22.36, "eta": "43:57:27", "max_grad_norm": 1.0, "loss": 0.8447695374488831, "grad_norm": 0.43043214082717896, "learning_rate": 0.0001882267713479586}
+{"ts": "2025-12-27T06:57:08", "event": "train_log", "step": 3182, "epoch": 1.3426160337552742, "progress_pct": 22.38, "epoch_pct": 22.38, "eta": "43:56:15", "max_grad_norm": 1.0, "loss": 0.7081091403961182, "grad_norm": 0.40693536400794983, "learning_rate": 0.00018820415150346163}
+{"ts": "2025-12-27T06:57:24", "event": "train_log", "step": 3184, "epoch": 1.3434599156118143, "progress_pct": 22.39, "epoch_pct": 22.39, "eta": "43:55:01", "max_grad_norm": 1.0, "loss": 0.6822336316108704, "grad_norm": 0.45037057995796204, "learning_rate": 0.00018818151131196097}
+{"ts": "2025-12-27T06:57:40", "event": "train_log", "step": 3186, "epoch": 1.3443037974683545, "progress_pct": 22.41, "epoch_pct": 22.41, "eta": "43:53:49", "max_grad_norm": 1.0, "loss": 0.6316909790039062, "grad_norm": 0.3731948137283325, "learning_rate": 0.00018815885077867915}
+{"ts": "2025-12-27T06:57:56", "event": "train_log", "step": 3188, "epoch": 1.3451476793248944, "progress_pct": 22.42, "epoch_pct": 22.42, "eta": "43:52:34", "max_grad_norm": 1.0, "loss": 0.7452375292778015, "grad_norm": 0.4248095750808716, "learning_rate": 0.00018813616990884363}
+{"ts": "2025-12-27T06:58:12", "event": "train_log", "step": 3190, "epoch": 1.3459915611814346, "progress_pct": 22.43, "epoch_pct": 22.43, "eta": "43:51:25", "max_grad_norm": 1.0, "loss": 0.6493782997131348, "grad_norm": 0.39229142665863037, "learning_rate": 0.00018811346870768638}
+{"ts": "2025-12-27T06:58:28", "event": "train_log", "step": 3192, "epoch": 1.3468354430379748, "progress_pct": 22.45, "epoch_pct": 22.45, "eta": "43:50:11", "max_grad_norm": 1.0, "loss": 0.735163688659668, "grad_norm": 0.45015549659729004, "learning_rate": 0.00018809074718044414}
+{"ts": "2025-12-27T06:58:44", "event": "train_log", "step": 3194, "epoch": 1.3476793248945147, "progress_pct": 22.46, "epoch_pct": 22.46, "eta": "43:49:01", "max_grad_norm": 1.0, "loss": 0.6656006574630737, "grad_norm": 0.39756354689598083, "learning_rate": 0.00018806800533235836}
+{"ts": "2025-12-27T06:59:00", "event": "train_log", "step": 3196, "epoch": 1.3485232067510549, "progress_pct": 22.48, "epoch_pct": 22.48, "eta": "43:47:48", "max_grad_norm": 1.0, "loss": 0.7040987014770508, "grad_norm": 0.4410010576248169, "learning_rate": 0.0001880452431686751}
+{"ts": "2025-12-27T06:59:16", "event": "train_log", "step": 3198, "epoch": 1.349367088607595, "progress_pct": 22.49, "epoch_pct": 22.49, "eta": "43:46:36", "max_grad_norm": 1.0, "loss": 0.7722331285476685, "grad_norm": 0.48144352436065674, "learning_rate": 0.0001880224606946452}
+{"ts": "2025-12-27T06:59:33", "event": "train_log", "step": 3200, "epoch": 1.350210970464135, "progress_pct": 22.5, "epoch_pct": 22.5, "eta": "43:45:27", "max_grad_norm": 1.0, "loss": 0.6681689023971558, "grad_norm": 0.38644957542419434, "learning_rate": 0.00018799965791552414}
+{"ts": "2025-12-27T07:10:54", "event": "train_log", "step": 3200, "epoch": 1.350210970464135, "progress_pct": 22.5, "epoch_pct": 22.5, "eta": "44:24:33", "max_grad_norm": 1.0, "eval_loss": 0.797160804271698, "eval_runtime": 680.976, "eval_samples_per_second": 3.094, "eval_steps_per_second": 3.094}
+{"ts": "2025-12-27T07:11:09", "event": "train_log", "step": 3202, "epoch": 1.3510548523206751, "progress_pct": 22.52, "epoch_pct": 22.52, "eta": "44:23:15", "max_grad_norm": 1.0, "loss": 0.6995527744293213, "grad_norm": 0.4471014440059662, "learning_rate": 0.00018797683483657201}
+{"ts": "2025-12-27T07:11:25", "event": "train_log", "step": 3204, "epoch": 1.3518987341772153, "progress_pct": 22.53, "epoch_pct": 22.53, "eta": "44:22:00", "max_grad_norm": 1.0, "loss": 0.659622848033905, "grad_norm": 0.4065002202987671, "learning_rate": 0.00018795399146305378}
+{"ts": "2025-12-27T07:11:39", "event": "train_log", "step": 3206, "epoch": 1.3527426160337552, "progress_pct": 22.55, "epoch_pct": 22.55, "eta": "44:20:40", "max_grad_norm": 1.0, "loss": 0.7927116751670837, "grad_norm": 0.4787910580635071, "learning_rate": 0.00018793112780023888}
+{"ts": "2025-12-27T07:11:54", "event": "train_log", "step": 3208, "epoch": 1.3535864978902954, "progress_pct": 22.56, "epoch_pct": 22.56, "eta": "44:19:23", "max_grad_norm": 1.0, "loss": 0.6738835573196411, "grad_norm": 0.43351876735687256, "learning_rate": 0.0001879082438534015}
+{"ts": "2025-12-27T07:12:09", "event": "train_log", "step": 3210, "epoch": 1.3544303797468356, "progress_pct": 22.57, "epoch_pct": 22.57, "eta": "44:18:06", "max_grad_norm": 1.0, "loss": 0.7785354852676392, "grad_norm": 0.4275745749473572, "learning_rate": 0.00018788533962782063}
+{"ts": "2025-12-27T07:12:24", "event": "train_log", "step": 3212, "epoch": 1.3552742616033755, "progress_pct": 22.59, "epoch_pct": 22.59, "eta": "44:16:49", "max_grad_norm": 1.0, "loss": 0.7325424551963806, "grad_norm": 0.4506145119667053, "learning_rate": 0.00018786241512877975}
+{"ts": "2025-12-27T07:12:39", "event": "train_log", "step": 3214, "epoch": 1.3561181434599157, "progress_pct": 22.6, "epoch_pct": 22.6, "eta": "44:15:31", "max_grad_norm": 1.0, "loss": 0.6898744702339172, "grad_norm": 0.42040735483169556, "learning_rate": 0.00018783947036156713}
+{"ts": "2025-12-27T07:12:54", "event": "train_log", "step": 3216, "epoch": 1.3569620253164558, "progress_pct": 22.62, "epoch_pct": 22.62, "eta": "44:14:15", "max_grad_norm": 1.0, "loss": 0.7498815655708313, "grad_norm": 0.4650438129901886, "learning_rate": 0.0001878165053314757}
+{"ts": "2025-12-27T07:13:09", "event": "train_log", "step": 3218, "epoch": 1.3578059071729958, "progress_pct": 22.63, "epoch_pct": 22.63, "eta": "44:13:00", "max_grad_norm": 1.0, "loss": 0.6891883611679077, "grad_norm": 0.4194130599498749, "learning_rate": 0.000187793520043803}
+{"ts": "2025-12-27T07:13:24", "event": "train_log", "step": 3220, "epoch": 1.358649789029536, "progress_pct": 22.64, "epoch_pct": 22.64, "eta": "44:11:44", "max_grad_norm": 1.0, "loss": 0.753805935382843, "grad_norm": 0.4447539746761322, "learning_rate": 0.0001877705145038514}
+{"ts": "2025-12-27T07:13:40", "event": "train_log", "step": 3222, "epoch": 1.3594936708860759, "progress_pct": 22.66, "epoch_pct": 22.66, "eta": "44:10:28", "max_grad_norm": 1.0, "loss": 0.6476293802261353, "grad_norm": 0.4047110974788666, "learning_rate": 0.00018774748871692773}
+{"ts": "2025-12-27T07:13:55", "event": "train_log", "step": 3224, "epoch": 1.360337552742616, "progress_pct": 22.67, "epoch_pct": 22.67, "eta": "44:09:12", "max_grad_norm": 1.0, "loss": 0.6765931844711304, "grad_norm": 0.39818957448005676, "learning_rate": 0.00018772444268834364}
+{"ts": "2025-12-27T07:14:10", "event": "train_log", "step": 3226, "epoch": 1.3611814345991562, "progress_pct": 22.69, "epoch_pct": 22.69, "eta": "44:07:55", "max_grad_norm": 1.0, "loss": 0.7360382676124573, "grad_norm": 0.41843733191490173, "learning_rate": 0.0001877013764234154}
+{"ts": "2025-12-27T07:14:24", "event": "train_log", "step": 3228, "epoch": 1.3620253164556961, "progress_pct": 22.7, "epoch_pct": 22.7, "eta": "44:06:39", "max_grad_norm": 1.0, "loss": 0.7149524688720703, "grad_norm": 0.41397789120674133, "learning_rate": 0.00018767828992746394}
+{"ts": "2025-12-27T07:14:39", "event": "train_log", "step": 3230, "epoch": 1.3628691983122363, "progress_pct": 22.71, "epoch_pct": 22.71, "eta": "44:05:23", "max_grad_norm": 1.0, "loss": 0.6932925581932068, "grad_norm": 0.43585047125816345, "learning_rate": 0.00018765518320581492}
+{"ts": "2025-12-27T07:14:55", "event": "train_log", "step": 3232, "epoch": 1.3637130801687762, "progress_pct": 22.73, "epoch_pct": 22.73, "eta": "44:04:09", "max_grad_norm": 1.0, "loss": 0.6561182141304016, "grad_norm": 0.3651057481765747, "learning_rate": 0.00018763205626379857}
+{"ts": "2025-12-27T07:15:10", "event": "train_log", "step": 3234, "epoch": 1.3645569620253164, "progress_pct": 22.74, "epoch_pct": 22.74, "eta": "44:02:52", "max_grad_norm": 1.0, "loss": 0.7436831593513489, "grad_norm": 0.44937077164649963, "learning_rate": 0.00018760890910674984}
+{"ts": "2025-12-27T07:15:24", "event": "train_log", "step": 3236, "epoch": 1.3654008438818566, "progress_pct": 22.76, "epoch_pct": 22.76, "eta": "44:01:32", "max_grad_norm": 1.0, "loss": 0.7849522233009338, "grad_norm": 0.5045920014381409, "learning_rate": 0.00018758574174000832}
+{"ts": "2025-12-27T07:15:38", "event": "train_log", "step": 3238, "epoch": 1.3662447257383965, "progress_pct": 22.77, "epoch_pct": 22.77, "eta": "44:00:13", "max_grad_norm": 1.0, "loss": 0.7501618266105652, "grad_norm": 0.4625580906867981, "learning_rate": 0.00018756255416891833}
+{"ts": "2025-12-27T07:15:53", "event": "train_log", "step": 3240, "epoch": 1.3670886075949367, "progress_pct": 22.78, "epoch_pct": 22.78, "eta": "43:58:57", "max_grad_norm": 1.0, "loss": 0.7656083106994629, "grad_norm": 0.3721269369125366, "learning_rate": 0.00018753934639882875}
+{"ts": "2025-12-27T07:16:07", "event": "train_log", "step": 3242, "epoch": 1.3679324894514768, "progress_pct": 22.8, "epoch_pct": 22.8, "eta": "43:57:40", "max_grad_norm": 1.0, "loss": 0.7816168069839478, "grad_norm": 0.41721680760383606, "learning_rate": 0.00018751611843509318}
+{"ts": "2025-12-27T07:16:22", "event": "train_log", "step": 3244, "epoch": 1.3687763713080168, "progress_pct": 22.81, "epoch_pct": 22.81, "eta": "43:56:25", "max_grad_norm": 1.0, "loss": 0.7002755403518677, "grad_norm": 0.8452199101448059, "learning_rate": 0.00018749287028306988}
+{"ts": "2025-12-27T07:16:37", "event": "train_log", "step": 3246, "epoch": 1.369620253164557, "progress_pct": 22.83, "epoch_pct": 22.83, "eta": "43:55:07", "max_grad_norm": 1.0, "loss": 0.7358270883560181, "grad_norm": 0.4320507347583771, "learning_rate": 0.00018746960194812172}
+{"ts": "2025-12-27T07:16:52", "event": "train_log", "step": 3248, "epoch": 1.370464135021097, "progress_pct": 22.84, "epoch_pct": 22.84, "eta": "43:53:53", "max_grad_norm": 1.0, "loss": 0.6528644561767578, "grad_norm": 0.43323323130607605, "learning_rate": 0.00018744631343561627}
+{"ts": "2025-12-27T07:17:07", "event": "train_log", "step": 3250, "epoch": 1.371308016877637, "progress_pct": 22.86, "epoch_pct": 22.86, "eta": "43:52:35", "max_grad_norm": 1.0, "loss": 0.7317517995834351, "grad_norm": 0.41487041115760803, "learning_rate": 0.0001874230047509258}
+{"ts": "2025-12-27T07:17:20", "event": "train_log", "step": 3252, "epoch": 1.3721518987341772, "progress_pct": 22.87, "epoch_pct": 22.87, "eta": "43:51:15", "max_grad_norm": 1.0, "loss": 0.8236643671989441, "grad_norm": 0.5085384249687195, "learning_rate": 0.00018739967589942707}
+{"ts": "2025-12-27T07:17:35", "event": "train_log", "step": 3254, "epoch": 1.3729957805907174, "progress_pct": 22.88, "epoch_pct": 22.88, "eta": "43:50:00", "max_grad_norm": 1.0, "loss": 0.730254054069519, "grad_norm": 0.4126952290534973, "learning_rate": 0.00018737632688650168}
+{"ts": "2025-12-27T07:17:50", "event": "train_log", "step": 3256, "epoch": 1.3738396624472573, "progress_pct": 22.9, "epoch_pct": 22.9, "eta": "43:48:43", "max_grad_norm": 1.0, "loss": 0.7637890577316284, "grad_norm": 0.5272142291069031, "learning_rate": 0.00018735295771753573}
+{"ts": "2025-12-27T07:18:04", "event": "train_log", "step": 3258, "epoch": 1.3746835443037975, "progress_pct": 22.91, "epoch_pct": 22.91, "eta": "43:47:25", "max_grad_norm": 1.0, "loss": 0.7906988263130188, "grad_norm": 0.4319530725479126, "learning_rate": 0.00018732956839792012}
+{"ts": "2025-12-27T07:18:19", "event": "train_log", "step": 3260, "epoch": 1.3755274261603376, "progress_pct": 22.93, "epoch_pct": 22.93, "eta": "43:46:10", "max_grad_norm": 1.0, "loss": 0.7261547446250916, "grad_norm": 0.4393143355846405, "learning_rate": 0.00018730615893305025}
+{"ts": "2025-12-27T07:18:34", "event": "train_log", "step": 3262, "epoch": 1.3763713080168776, "progress_pct": 22.94, "epoch_pct": 22.94, "eta": "43:44:55", "max_grad_norm": 1.0, "loss": 0.6857497692108154, "grad_norm": 0.4161636233329773, "learning_rate": 0.00018728272932832632}
+{"ts": "2025-12-27T07:18:48", "event": "train_log", "step": 3264, "epoch": 1.3772151898734177, "progress_pct": 22.95, "epoch_pct": 22.95, "eta": "43:43:37", "max_grad_norm": 1.0, "loss": 0.7488857507705688, "grad_norm": 0.48862120509147644, "learning_rate": 0.00018725927958915297}
+{"ts": "2025-12-27T07:19:03", "event": "train_log", "step": 3266, "epoch": 1.378059071729958, "progress_pct": 22.97, "epoch_pct": 22.97, "eta": "43:42:22", "max_grad_norm": 1.0, "loss": 0.7257721424102783, "grad_norm": 0.4422304928302765, "learning_rate": 0.0001872358097209397}
+{"ts": "2025-12-27T07:19:17", "event": "train_log", "step": 3268, "epoch": 1.3789029535864978, "progress_pct": 22.98, "epoch_pct": 22.98, "eta": "43:41:06", "max_grad_norm": 1.0, "loss": 0.746697187423706, "grad_norm": 0.4577067494392395, "learning_rate": 0.00018721231972910048}
+{"ts": "2025-12-27T07:19:32", "event": "train_log", "step": 3270, "epoch": 1.379746835443038, "progress_pct": 23.0, "epoch_pct": 23.0, "eta": "43:39:51", "max_grad_norm": 1.0, "loss": 0.768519401550293, "grad_norm": 0.42382001876831055, "learning_rate": 0.00018718880961905406}
+{"ts": "2025-12-27T07:19:47", "event": "train_log", "step": 3272, "epoch": 1.3805907172995782, "progress_pct": 23.01, "epoch_pct": 23.01, "eta": "43:38:36", "max_grad_norm": 1.0, "loss": 0.7426102757453918, "grad_norm": 0.43368014693260193, "learning_rate": 0.00018716527939622378}
+{"ts": "2025-12-27T07:20:03", "event": "train_log", "step": 3274, "epoch": 1.381434599156118, "progress_pct": 23.02, "epoch_pct": 23.02, "eta": "43:37:23", "max_grad_norm": 1.0, "loss": 0.7104197144508362, "grad_norm": 0.3777220547199249, "learning_rate": 0.00018714172906603757}
+{"ts": "2025-12-27T07:20:17", "event": "train_log", "step": 3276, "epoch": 1.3822784810126583, "progress_pct": 23.04, "epoch_pct": 23.04, "eta": "43:36:08", "max_grad_norm": 1.0, "loss": 0.7502968907356262, "grad_norm": 0.46071696281433105, "learning_rate": 0.00018711815863392802}
+{"ts": "2025-12-27T07:20:33", "event": "train_log", "step": 3278, "epoch": 1.3831223628691984, "progress_pct": 23.05, "epoch_pct": 23.05, "eta": "43:34:56", "max_grad_norm": 1.0, "loss": 0.6400540471076965, "grad_norm": 0.42510855197906494, "learning_rate": 0.00018709456810533245}
+{"ts": "2025-12-27T07:20:50", "event": "train_log", "step": 3280, "epoch": 1.3839662447257384, "progress_pct": 23.07, "epoch_pct": 23.07, "eta": "43:33:49", "max_grad_norm": 1.0, "loss": 0.6648687720298767, "grad_norm": 0.4217228293418884, "learning_rate": 0.00018707095748569268}
+{"ts": "2025-12-27T07:21:07", "event": "train_log", "step": 3282, "epoch": 1.3848101265822785, "progress_pct": 23.08, "epoch_pct": 23.08, "eta": "43:32:39", "max_grad_norm": 1.0, "loss": 0.7176171541213989, "grad_norm": 0.48267799615859985, "learning_rate": 0.00018704732678045526}
+{"ts": "2025-12-27T07:21:21", "event": "train_log", "step": 3284, "epoch": 1.3856540084388187, "progress_pct": 23.09, "epoch_pct": 23.09, "eta": "43:31:24", "max_grad_norm": 1.0, "loss": 0.8402457237243652, "grad_norm": 0.47114965319633484, "learning_rate": 0.00018702367599507134}
+{"ts": "2025-12-27T07:21:37", "event": "train_log", "step": 3286, "epoch": 1.3864978902953586, "progress_pct": 23.11, "epoch_pct": 23.11, "eta": "43:30:13", "max_grad_norm": 1.0, "loss": 0.8130836486816406, "grad_norm": 0.5054492354393005, "learning_rate": 0.00018700000513499667}
+{"ts": "2025-12-27T07:21:54", "event": "train_log", "step": 3288, "epoch": 1.3873417721518988, "progress_pct": 23.12, "epoch_pct": 23.12, "eta": "43:29:06", "max_grad_norm": 1.0, "loss": 0.6432579755783081, "grad_norm": 0.39435645937919617, "learning_rate": 0.00018697631420569168}
+{"ts": "2025-12-27T07:22:11", "event": "train_log", "step": 3290, "epoch": 1.3881856540084387, "progress_pct": 23.14, "epoch_pct": 23.14, "eta": "43:27:57", "max_grad_norm": 1.0, "loss": 0.6921643614768982, "grad_norm": 0.3997228443622589, "learning_rate": 0.00018695260321262142}
+{"ts": "2025-12-27T07:22:27", "event": "train_log", "step": 3292, "epoch": 1.389029535864979, "progress_pct": 23.15, "epoch_pct": 23.15, "eta": "43:26:46", "max_grad_norm": 1.0, "loss": 0.7162153124809265, "grad_norm": 0.37304237484931946, "learning_rate": 0.00018692887216125557}
+{"ts": "2025-12-27T07:22:43", "event": "train_log", "step": 3294, "epoch": 1.389873417721519, "progress_pct": 23.16, "epoch_pct": 23.16, "eta": "43:25:38", "max_grad_norm": 1.0, "loss": 0.7194803953170776, "grad_norm": 0.7997981309890747, "learning_rate": 0.00018690512105706838}
+{"ts": "2025-12-27T07:22:58", "event": "train_log", "step": 3296, "epoch": 1.390717299578059, "progress_pct": 23.18, "epoch_pct": 23.18, "eta": "43:24:23", "max_grad_norm": 1.0, "loss": 0.6888322830200195, "grad_norm": 0.41268959641456604, "learning_rate": 0.00018688134990553883}
+{"ts": "2025-12-27T07:23:14", "event": "train_log", "step": 3298, "epoch": 1.3915611814345992, "progress_pct": 23.19, "epoch_pct": 23.19, "eta": "43:23:13", "max_grad_norm": 1.0, "loss": 0.7387221455574036, "grad_norm": 0.4485826790332794, "learning_rate": 0.00018685755871215043}
+{"ts": "2025-12-27T07:23:29", "event": "train_log", "step": 3300, "epoch": 1.3924050632911391, "progress_pct": 23.21, "epoch_pct": 23.21, "eta": "43:21:58", "max_grad_norm": 1.0, "loss": 0.8061952590942383, "grad_norm": 0.47835296392440796, "learning_rate": 0.0001868337474823914}
+{"ts": "2025-12-27T07:35:01", "event": "train_log", "step": 3300, "epoch": 1.3924050632911391, "progress_pct": 23.21, "epoch_pct": 23.21, "eta": "44:00:10", "max_grad_norm": 1.0, "eval_loss": 0.795619547367096, "eval_runtime": 692.7157, "eval_samples_per_second": 3.042, "eval_steps_per_second": 3.042}
+{"ts": "2025-12-27T07:35:18", "event": "train_log", "step": 3302, "epoch": 1.3932489451476793, "progress_pct": 23.22, "epoch_pct": 23.22, "eta": "43:58:58", "max_grad_norm": 1.0, "loss": 0.7317537069320679, "grad_norm": 0.4024396240711212, "learning_rate": 0.00018680991622175446}
+{"ts": "2025-12-27T07:35:34", "event": "train_log", "step": 3304, "epoch": 1.3940928270042194, "progress_pct": 23.23, "epoch_pct": 23.23, "eta": "43:57:47", "max_grad_norm": 1.0, "loss": 0.7470992207527161, "grad_norm": 0.4526490271091461, "learning_rate": 0.00018678606493573708}
+{"ts": "2025-12-27T07:35:50", "event": "train_log", "step": 3306, "epoch": 1.3949367088607594, "progress_pct": 23.25, "epoch_pct": 23.25, "eta": "43:56:34", "max_grad_norm": 1.0, "loss": 0.7627943754196167, "grad_norm": 0.5055129528045654, "learning_rate": 0.00018676219362984126}
+{"ts": "2025-12-27T07:36:05", "event": "train_log", "step": 3308, "epoch": 1.3957805907172995, "progress_pct": 23.26, "epoch_pct": 23.26, "eta": "43:55:21", "max_grad_norm": 1.0, "loss": 0.684033989906311, "grad_norm": 0.4030216634273529, "learning_rate": 0.0001867383023095737}
+{"ts": "2025-12-27T07:36:21", "event": "train_log", "step": 3310, "epoch": 1.3966244725738397, "progress_pct": 23.28, "epoch_pct": 23.28, "eta": "43:54:09", "max_grad_norm": 1.0, "loss": 0.7366722822189331, "grad_norm": 0.4353308379650116, "learning_rate": 0.0001867143909804456}
+{"ts": "2025-12-27T07:36:37", "event": "train_log", "step": 3312, "epoch": 1.3974683544303796, "progress_pct": 23.29, "epoch_pct": 23.29, "eta": "43:52:57", "max_grad_norm": 1.0, "loss": 0.8050951957702637, "grad_norm": 0.5360553860664368, "learning_rate": 0.0001866904596479729}
+{"ts": "2025-12-27T07:36:51", "event": "train_log", "step": 3314, "epoch": 1.3983122362869198, "progress_pct": 23.31, "epoch_pct": 23.31, "eta": "43:51:40", "max_grad_norm": 1.0, "loss": 0.7227963805198669, "grad_norm": 0.46081677079200745, "learning_rate": 0.00018666650831767608}
+{"ts": "2025-12-27T07:37:05", "event": "train_log", "step": 3316, "epoch": 1.39915611814346, "progress_pct": 23.32, "epoch_pct": 23.32, "eta": "43:50:22", "max_grad_norm": 1.0, "loss": 0.714216411113739, "grad_norm": 0.4325845539569855, "learning_rate": 0.00018664253699508029}
+{"ts": "2025-12-27T07:37:21", "event": "train_log", "step": 3318, "epoch": 1.4, "progress_pct": 23.33, "epoch_pct": 23.33, "eta": "43:49:08", "max_grad_norm": 1.0, "loss": 0.7378593683242798, "grad_norm": 0.4373100996017456, "learning_rate": 0.0001866185456857152}
+{"ts": "2025-12-27T07:37:35", "event": "train_log", "step": 3320, "epoch": 1.40084388185654, "progress_pct": 23.35, "epoch_pct": 23.35, "eta": "43:47:53", "max_grad_norm": 1.0, "loss": 0.7565981149673462, "grad_norm": 0.43221279978752136, "learning_rate": 0.00018659453439511512}
+{"ts": "2025-12-27T07:37:51", "event": "train_log", "step": 3322, "epoch": 1.4016877637130802, "progress_pct": 23.36, "epoch_pct": 23.36, "eta": "43:46:40", "max_grad_norm": 1.0, "loss": 0.6409488916397095, "grad_norm": 0.3938370943069458, "learning_rate": 0.00018657050312881904}
+{"ts": "2025-12-27T07:38:06", "event": "train_log", "step": 3324, "epoch": 1.4025316455696202, "progress_pct": 23.38, "epoch_pct": 23.38, "eta": "43:45:24", "max_grad_norm": 1.0, "loss": 0.6911652088165283, "grad_norm": 0.43609440326690674, "learning_rate": 0.00018654645189237055}
+{"ts": "2025-12-27T07:38:20", "event": "train_log", "step": 3326, "epoch": 1.4033755274261603, "progress_pct": 23.39, "epoch_pct": 23.39, "eta": "43:44:08", "max_grad_norm": 1.0, "loss": 0.7275710701942444, "grad_norm": 0.4526350200176239, "learning_rate": 0.00018652238069131774}
+{"ts": "2025-12-27T07:38:35", "event": "train_log", "step": 3328, "epoch": 1.4042194092827005, "progress_pct": 23.4, "epoch_pct": 23.4, "eta": "43:42:54", "max_grad_norm": 1.0, "loss": 0.6466895341873169, "grad_norm": 0.3990516662597656, "learning_rate": 0.0001864982895312134}
+{"ts": "2025-12-27T07:38:51", "event": "train_log", "step": 3330, "epoch": 1.4050632911392404, "progress_pct": 23.42, "epoch_pct": 23.42, "eta": "43:41:41", "max_grad_norm": 1.0, "loss": 0.6520540714263916, "grad_norm": 0.44321122765541077, "learning_rate": 0.0001864741784176149}
+{"ts": "2025-12-27T07:39:05", "event": "train_log", "step": 3332, "epoch": 1.4059071729957806, "progress_pct": 23.43, "epoch_pct": 23.43, "eta": "43:40:25", "max_grad_norm": 1.0, "loss": 0.7289671301841736, "grad_norm": 0.44733941555023193, "learning_rate": 0.0001864500473560842}
+{"ts": "2025-12-27T07:39:19", "event": "train_log", "step": 3334, "epoch": 1.4067510548523208, "progress_pct": 23.45, "epoch_pct": 23.45, "eta": "43:39:09", "max_grad_norm": 1.0, "loss": 0.7664391398429871, "grad_norm": 0.4569414556026459, "learning_rate": 0.00018642589635218789}
+{"ts": "2025-12-27T07:39:35", "event": "train_log", "step": 3336, "epoch": 1.4075949367088607, "progress_pct": 23.46, "epoch_pct": 23.46, "eta": "43:37:55", "max_grad_norm": 1.0, "loss": 0.7747462391853333, "grad_norm": 0.4172186851501465, "learning_rate": 0.00018640172541149709}
+{"ts": "2025-12-27T07:39:49", "event": "train_log", "step": 3338, "epoch": 1.4084388185654009, "progress_pct": 23.47, "epoch_pct": 23.47, "eta": "43:36:39", "max_grad_norm": 1.0, "loss": 0.8337607383728027, "grad_norm": 0.4540993869304657, "learning_rate": 0.0001863775345395877}
+{"ts": "2025-12-27T07:40:04", "event": "train_log", "step": 3340, "epoch": 1.409282700421941, "progress_pct": 23.49, "epoch_pct": 23.49, "eta": "43:35:26", "max_grad_norm": 1.0, "loss": 0.6851577758789062, "grad_norm": 0.4306154251098633, "learning_rate": 0.00018635332374203993}
+{"ts": "2025-12-27T07:40:18", "event": "train_log", "step": 3342, "epoch": 1.410126582278481, "progress_pct": 23.5, "epoch_pct": 23.5, "eta": "43:34:09", "max_grad_norm": 1.0, "loss": 0.6841398477554321, "grad_norm": 0.45615455508232117, "learning_rate": 0.00018632909302443884}
+{"ts": "2025-12-27T07:40:34", "event": "train_log", "step": 3344, "epoch": 1.4109704641350211, "progress_pct": 23.52, "epoch_pct": 23.52, "eta": "43:32:59", "max_grad_norm": 1.0, "loss": 0.6672312021255493, "grad_norm": 0.35296666622161865, "learning_rate": 0.00018630484239237397}
+{"ts": "2025-12-27T07:40:48", "event": "train_log", "step": 3346, "epoch": 1.4118143459915613, "progress_pct": 23.53, "epoch_pct": 23.53, "eta": "43:31:42", "max_grad_norm": 1.0, "loss": 0.7027004361152649, "grad_norm": 0.4424770176410675, "learning_rate": 0.00018628057185143945}
+{"ts": "2025-12-27T07:41:04", "event": "train_log", "step": 3348, "epoch": 1.4126582278481012, "progress_pct": 23.54, "epoch_pct": 23.54, "eta": "43:30:29", "max_grad_norm": 1.0, "loss": 0.7397928237915039, "grad_norm": 0.429817259311676, "learning_rate": 0.0001862562814072341}
+{"ts": "2025-12-27T07:41:19", "event": "train_log", "step": 3350, "epoch": 1.4135021097046414, "progress_pct": 23.56, "epoch_pct": 23.56, "eta": "43:29:16", "max_grad_norm": 1.0, "loss": 0.8009215593338013, "grad_norm": 0.4720051884651184, "learning_rate": 0.00018623197106536118}
+{"ts": "2025-12-27T07:41:34", "event": "train_log", "step": 3352, "epoch": 1.4143459915611816, "progress_pct": 23.57, "epoch_pct": 23.57, "eta": "43:28:03", "max_grad_norm": 1.0, "loss": 0.6875046491622925, "grad_norm": 0.42197802662849426, "learning_rate": 0.0001862076408314287}
+{"ts": "2025-12-27T07:41:50", "event": "train_log", "step": 3354, "epoch": 1.4151898734177215, "progress_pct": 23.59, "epoch_pct": 23.59, "eta": "43:26:52", "max_grad_norm": 1.0, "loss": 0.7202261090278625, "grad_norm": 0.3674047589302063, "learning_rate": 0.00018618329071104907}
+{"ts": "2025-12-27T07:42:04", "event": "train_log", "step": 3356, "epoch": 1.4160337552742617, "progress_pct": 23.6, "epoch_pct": 23.6, "eta": "43:25:36", "max_grad_norm": 1.0, "loss": 0.6981248259544373, "grad_norm": 0.42327889800071716, "learning_rate": 0.0001861589207098395}
+{"ts": "2025-12-27T07:42:20", "event": "train_log", "step": 3358, "epoch": 1.4168776371308016, "progress_pct": 23.61, "epoch_pct": 23.61, "eta": "43:24:26", "max_grad_norm": 1.0, "loss": 0.7081356048583984, "grad_norm": 0.40644150972366333, "learning_rate": 0.00018613453083342162}
+{"ts": "2025-12-27T07:42:35", "event": "train_log", "step": 3360, "epoch": 1.4177215189873418, "progress_pct": 23.63, "epoch_pct": 23.63, "eta": "43:23:14", "max_grad_norm": 1.0, "loss": 0.6420091986656189, "grad_norm": 0.3948482275009155, "learning_rate": 0.00018611012108742168}
+{"ts": "2025-12-27T07:42:49", "event": "train_log", "step": 3362, "epoch": 1.4185654008438817, "progress_pct": 23.64, "epoch_pct": 23.64, "eta": "43:21:57", "max_grad_norm": 1.0, "loss": 0.7662023305892944, "grad_norm": 0.44622674584388733, "learning_rate": 0.00018608569147747064}
+{"ts": "2025-12-27T07:43:04", "event": "train_log", "step": 3364, "epoch": 1.4194092827004219, "progress_pct": 23.66, "epoch_pct": 23.66, "eta": "43:20:43", "max_grad_norm": 1.0, "loss": 0.7243877649307251, "grad_norm": 0.39772045612335205, "learning_rate": 0.00018606124200920386}
+{"ts": "2025-12-27T07:43:19", "event": "train_log", "step": 3366, "epoch": 1.420253164556962, "progress_pct": 23.67, "epoch_pct": 23.67, "eta": "43:19:29", "max_grad_norm": 1.0, "loss": 0.7399296164512634, "grad_norm": 0.4284326136112213, "learning_rate": 0.00018603677268826138}
+{"ts": "2025-12-27T07:43:34", "event": "train_log", "step": 3368, "epoch": 1.421097046413502, "progress_pct": 23.68, "epoch_pct": 23.68, "eta": "43:18:17", "max_grad_norm": 1.0, "loss": 0.6978716850280762, "grad_norm": 0.3749450445175171, "learning_rate": 0.0001860122835202878}
+{"ts": "2025-12-27T07:43:48", "event": "train_log", "step": 3370, "epoch": 1.4219409282700421, "progress_pct": 23.7, "epoch_pct": 23.7, "eta": "43:17:02", "max_grad_norm": 1.0, "loss": 0.7733059525489807, "grad_norm": 0.46721795201301575, "learning_rate": 0.00018598777451093234}
+{"ts": "2025-12-27T07:44:04", "event": "train_log", "step": 3372, "epoch": 1.4227848101265823, "progress_pct": 23.71, "epoch_pct": 23.71, "eta": "43:15:51", "max_grad_norm": 1.0, "loss": 0.6721867918968201, "grad_norm": 0.39606085419654846, "learning_rate": 0.0001859632456658487}
+{"ts": "2025-12-27T07:44:19", "event": "train_log", "step": 3374, "epoch": 1.4236286919831223, "progress_pct": 23.73, "epoch_pct": 23.73, "eta": "43:14:38", "max_grad_norm": 1.0, "loss": 0.715825080871582, "grad_norm": 0.3879879117012024, "learning_rate": 0.00018593869699069527}
+{"ts": "2025-12-27T07:44:35", "event": "train_log", "step": 3376, "epoch": 1.4244725738396624, "progress_pct": 23.74, "epoch_pct": 23.74, "eta": "43:13:27", "max_grad_norm": 1.0, "loss": 0.6226941347122192, "grad_norm": 0.3641676902770996, "learning_rate": 0.0001859141284911349}
+{"ts": "2025-12-27T07:44:49", "event": "train_log", "step": 3378, "epoch": 1.4253164556962026, "progress_pct": 23.76, "epoch_pct": 23.76, "eta": "43:12:14", "max_grad_norm": 1.0, "loss": 0.7972121238708496, "grad_norm": 0.4612879753112793, "learning_rate": 0.0001858895401728351}
+{"ts": "2025-12-27T07:45:04", "event": "train_log", "step": 3380, "epoch": 1.4261603375527425, "progress_pct": 23.77, "epoch_pct": 23.77, "eta": "43:11:00", "max_grad_norm": 1.0, "loss": 0.6663658022880554, "grad_norm": 0.4406103193759918, "learning_rate": 0.00018586493204146795}
+{"ts": "2025-12-27T07:45:20", "event": "train_log", "step": 3382, "epoch": 1.4270042194092827, "progress_pct": 23.78, "epoch_pct": 23.78, "eta": "43:09:49", "max_grad_norm": 1.0, "loss": 0.6867321133613586, "grad_norm": 0.420205295085907, "learning_rate": 0.00018584030410271}
+{"ts": "2025-12-27T07:45:35", "event": "train_log", "step": 3384, "epoch": 1.4278481012658228, "progress_pct": 23.8, "epoch_pct": 23.8, "eta": "43:08:39", "max_grad_norm": 1.0, "loss": 0.66688072681427, "grad_norm": 0.4499560594558716, "learning_rate": 0.00018581565636224253}
+{"ts": "2025-12-27T07:45:51", "event": "train_log", "step": 3386, "epoch": 1.4286919831223628, "progress_pct": 23.81, "epoch_pct": 23.81, "eta": "43:07:28", "max_grad_norm": 1.0, "loss": 0.6714702844619751, "grad_norm": 0.42507919669151306, "learning_rate": 0.00018579098882575123}
+{"ts": "2025-12-27T07:46:06", "event": "train_log", "step": 3388, "epoch": 1.429535864978903, "progress_pct": 23.83, "epoch_pct": 23.83, "eta": "43:06:17", "max_grad_norm": 1.0, "loss": 0.7219351530075073, "grad_norm": 0.43856513500213623, "learning_rate": 0.0001857663014989265}
+{"ts": "2025-12-27T07:46:21", "event": "train_log", "step": 3390, "epoch": 1.4303797468354431, "progress_pct": 23.84, "epoch_pct": 23.84, "eta": "43:05:05", "max_grad_norm": 1.0, "loss": 0.7157991528511047, "grad_norm": 0.5167419910430908, "learning_rate": 0.00018574159438746314}
+{"ts": "2025-12-27T07:46:36", "event": "train_log", "step": 3392, "epoch": 1.431223628691983, "progress_pct": 23.85, "epoch_pct": 23.85, "eta": "43:03:53", "max_grad_norm": 1.0, "loss": 0.7509840726852417, "grad_norm": 0.4443167746067047, "learning_rate": 0.00018571686749706067}
+{"ts": "2025-12-27T07:46:51", "event": "train_log", "step": 3394, "epoch": 1.4320675105485232, "progress_pct": 23.87, "epoch_pct": 23.87, "eta": "43:02:41", "max_grad_norm": 1.0, "loss": 0.756000816822052, "grad_norm": 0.41321757435798645, "learning_rate": 0.0001856921208334231}
+{"ts": "2025-12-27T07:47:06", "event": "train_log", "step": 3396, "epoch": 1.4329113924050634, "progress_pct": 23.88, "epoch_pct": 23.88, "eta": "43:01:29", "max_grad_norm": 1.0, "loss": 0.7047854065895081, "grad_norm": 0.4094153046607971, "learning_rate": 0.000185667354402259}
+{"ts": "2025-12-27T07:47:21", "event": "train_log", "step": 3398, "epoch": 1.4337552742616033, "progress_pct": 23.9, "epoch_pct": 23.9, "eta": "43:00:16", "max_grad_norm": 1.0, "loss": 0.7620865106582642, "grad_norm": 0.4248482286930084, "learning_rate": 0.0001856425682092815}
+{"ts": "2025-12-27T07:47:35", "event": "train_log", "step": 3400, "epoch": 1.4345991561181435, "progress_pct": 23.91, "epoch_pct": 23.91, "eta": "42:59:01", "max_grad_norm": 1.0, "loss": 0.7441845536231995, "grad_norm": 0.4190550148487091, "learning_rate": 0.0001856177622602083}
+{"ts": "2025-12-27T07:59:01", "event": "train_log", "step": 3400, "epoch": 1.4345991561181435, "progress_pct": 23.91, "epoch_pct": 23.91, "eta": "43:35:25", "max_grad_norm": 1.0, "eval_loss": 0.7917885780334473, "eval_runtime": 686.1689, "eval_samples_per_second": 3.071, "eval_steps_per_second": 3.071}
+{"ts": "2025-12-27T07:59:17", "event": "train_log", "step": 3402, "epoch": 1.4354430379746836, "progress_pct": 23.92, "epoch_pct": 23.92, "eta": "43:34:13", "max_grad_norm": 1.0, "loss": 0.7123498916625977, "grad_norm": 0.45602327585220337, "learning_rate": 0.00018559293656076166}
+{"ts": "2025-12-27T07:59:32", "event": "train_log", "step": 3404, "epoch": 1.4362869198312236, "progress_pct": 23.94, "epoch_pct": 23.94, "eta": "43:33:00", "max_grad_norm": 1.0, "loss": 0.7096527814865112, "grad_norm": 0.5209269523620605, "learning_rate": 0.00018556809111666843}
+{"ts": "2025-12-27T07:59:48", "event": "train_log", "step": 3406, "epoch": 1.4371308016877637, "progress_pct": 23.95, "epoch_pct": 23.95, "eta": "43:31:49", "max_grad_norm": 1.0, "loss": 0.7116838693618774, "grad_norm": 0.44521182775497437, "learning_rate": 0.00018554322593365995}
+{"ts": "2025-12-27T08:00:02", "event": "train_log", "step": 3408, "epoch": 1.437974683544304, "progress_pct": 23.97, "epoch_pct": 23.97, "eta": "43:30:34", "max_grad_norm": 1.0, "loss": 0.7756352424621582, "grad_norm": 0.44571003317832947, "learning_rate": 0.00018551834101747212}
+{"ts": "2025-12-27T08:00:18", "event": "train_log", "step": 3410, "epoch": 1.4388185654008439, "progress_pct": 23.98, "epoch_pct": 23.98, "eta": "43:29:23", "max_grad_norm": 1.0, "loss": 0.689633309841156, "grad_norm": 0.4030776023864746, "learning_rate": 0.00018549343637384543}
+{"ts": "2025-12-27T08:00:34", "event": "train_log", "step": 3412, "epoch": 1.439662447257384, "progress_pct": 23.99, "epoch_pct": 23.99, "eta": "43:28:14", "max_grad_norm": 1.0, "loss": 0.7173827290534973, "grad_norm": 0.4025530517101288, "learning_rate": 0.0001854685120085249}
+{"ts": "2025-12-27T08:00:50", "event": "train_log", "step": 3414, "epoch": 1.4405063291139242, "progress_pct": 24.01, "epoch_pct": 24.01, "eta": "43:27:04", "max_grad_norm": 1.0, "loss": 0.7212072610855103, "grad_norm": 0.4376870095729828, "learning_rate": 0.0001854435679272601}
+{"ts": "2025-12-27T08:01:05", "event": "train_log", "step": 3416, "epoch": 1.4413502109704641, "progress_pct": 24.02, "epoch_pct": 24.02, "eta": "43:25:50", "max_grad_norm": 1.0, "loss": 0.8260660171508789, "grad_norm": 0.4469084143638611, "learning_rate": 0.00018541860413580515}
+{"ts": "2025-12-27T08:01:21", "event": "train_log", "step": 3418, "epoch": 1.4421940928270043, "progress_pct": 24.04, "epoch_pct": 24.04, "eta": "43:24:40", "max_grad_norm": 1.0, "loss": 0.764930248260498, "grad_norm": 0.48033684492111206, "learning_rate": 0.00018539362063991872}
+{"ts": "2025-12-27T08:01:37", "event": "train_log", "step": 3420, "epoch": 1.4430379746835442, "progress_pct": 24.05, "epoch_pct": 24.05, "eta": "43:23:30", "max_grad_norm": 1.0, "loss": 0.7436494827270508, "grad_norm": 0.4428006708621979, "learning_rate": 0.00018536861744536403}
+{"ts": "2025-12-27T08:01:53", "event": "train_log", "step": 3422, "epoch": 1.4438818565400844, "progress_pct": 24.06, "epoch_pct": 24.06, "eta": "43:22:21", "max_grad_norm": 1.0, "loss": 0.7695552706718445, "grad_norm": 0.41812989115715027, "learning_rate": 0.00018534359455790885}
+{"ts": "2025-12-27T08:02:09", "event": "train_log", "step": 3424, "epoch": 1.4447257383966245, "progress_pct": 24.08, "epoch_pct": 24.08, "eta": "43:21:10", "max_grad_norm": 1.0, "loss": 0.7430856823921204, "grad_norm": 0.4243115186691284, "learning_rate": 0.00018531855198332544}
+{"ts": "2025-12-27T08:02:23", "event": "train_log", "step": 3426, "epoch": 1.4455696202531645, "progress_pct": 24.09, "epoch_pct": 24.09, "eta": "43:19:56", "max_grad_norm": 1.0, "loss": 0.7015851140022278, "grad_norm": 0.44387027621269226, "learning_rate": 0.00018529348972739067}
+{"ts": "2025-12-27T08:02:38", "event": "train_log", "step": 3428, "epoch": 1.4464135021097047, "progress_pct": 24.11, "epoch_pct": 24.11, "eta": "43:18:41", "max_grad_norm": 1.0, "loss": 0.7073075771331787, "grad_norm": 0.5105122923851013, "learning_rate": 0.0001852684077958859}
+{"ts": "2025-12-27T08:02:53", "event": "train_log", "step": 3430, "epoch": 1.4472573839662446, "progress_pct": 24.12, "epoch_pct": 24.12, "eta": "43:17:32", "max_grad_norm": 1.0, "loss": 0.7478930950164795, "grad_norm": 0.40193185210227966, "learning_rate": 0.00018524330619459708}
+{"ts": "2025-12-27T08:03:08", "event": "train_log", "step": 3432, "epoch": 1.4481012658227848, "progress_pct": 24.14, "epoch_pct": 24.14, "eta": "43:16:18", "max_grad_norm": 1.0, "loss": 0.6987152099609375, "grad_norm": 0.47935256361961365, "learning_rate": 0.00018521818492931463}
+{"ts": "2025-12-27T08:03:23", "event": "train_log", "step": 3434, "epoch": 1.448945147679325, "progress_pct": 24.15, "epoch_pct": 24.15, "eta": "43:15:06", "max_grad_norm": 1.0, "loss": 0.6677015423774719, "grad_norm": 0.42298686504364014, "learning_rate": 0.00018519304400583356}
+{"ts": "2025-12-27T08:03:39", "event": "train_log", "step": 3436, "epoch": 1.4497890295358649, "progress_pct": 24.16, "epoch_pct": 24.16, "eta": "43:13:55", "max_grad_norm": 1.0, "loss": 0.5727065205574036, "grad_norm": 0.37220221757888794, "learning_rate": 0.0001851678834299534}
+{"ts": "2025-12-27T08:03:55", "event": "train_log", "step": 3438, "epoch": 1.450632911392405, "progress_pct": 24.18, "epoch_pct": 24.18, "eta": "43:12:45", "max_grad_norm": 1.0, "loss": 0.7245276570320129, "grad_norm": 0.4990854561328888, "learning_rate": 0.00018514270320747823}
+{"ts": "2025-12-27T08:04:11", "event": "train_log", "step": 3440, "epoch": 1.4514767932489452, "progress_pct": 24.19, "epoch_pct": 24.19, "eta": "43:11:38", "max_grad_norm": 1.0, "loss": 0.5907649993896484, "grad_norm": 0.4366965889930725, "learning_rate": 0.00018511750334421665}
+{"ts": "2025-12-27T08:04:26", "event": "train_log", "step": 3442, "epoch": 1.4523206751054851, "progress_pct": 24.21, "epoch_pct": 24.21, "eta": "43:10:25", "max_grad_norm": 1.0, "loss": 0.6898093223571777, "grad_norm": 0.4355267286300659, "learning_rate": 0.00018509228384598172}
+{"ts": "2025-12-27T08:04:43", "event": "train_log", "step": 3444, "epoch": 1.4531645569620253, "progress_pct": 24.22, "epoch_pct": 24.22, "eta": "43:09:19", "max_grad_norm": 1.0, "loss": 0.6242932081222534, "grad_norm": 0.4055832028388977, "learning_rate": 0.00018506704471859116}
+{"ts": "2025-12-27T08:04:58", "event": "train_log", "step": 3446, "epoch": 1.4540084388185655, "progress_pct": 24.23, "epoch_pct": 24.23, "eta": "43:08:07", "max_grad_norm": 1.0, "loss": 0.7214970588684082, "grad_norm": 0.42599600553512573, "learning_rate": 0.00018504178596786712}
+{"ts": "2025-12-27T08:05:14", "event": "train_log", "step": 3448, "epoch": 1.4548523206751054, "progress_pct": 24.25, "epoch_pct": 24.25, "eta": "43:06:59", "max_grad_norm": 1.0, "loss": 0.7120789885520935, "grad_norm": 0.41228237748146057, "learning_rate": 0.0001850165075996363}
+{"ts": "2025-12-27T08:05:30", "event": "train_log", "step": 3450, "epoch": 1.4556962025316456, "progress_pct": 24.26, "epoch_pct": 24.26, "eta": "43:05:48", "max_grad_norm": 1.0, "loss": 0.7376183867454529, "grad_norm": 0.4529595375061035, "learning_rate": 0.00018499120961972996}
+{"ts": "2025-12-27T08:05:44", "event": "train_log", "step": 3452, "epoch": 1.4565400843881857, "progress_pct": 24.28, "epoch_pct": 24.28, "eta": "43:04:35", "max_grad_norm": 1.0, "loss": 0.7321018576622009, "grad_norm": 0.4260895550251007, "learning_rate": 0.00018496589203398386}
+{"ts": "2025-12-27T08:05:58", "event": "train_log", "step": 3454, "epoch": 1.4573839662447257, "progress_pct": 24.29, "epoch_pct": 24.29, "eta": "43:03:20", "max_grad_norm": 1.0, "loss": 0.7099342346191406, "grad_norm": 0.46325981616973877, "learning_rate": 0.00018494055484823824}
+{"ts": "2025-12-27T08:06:14", "event": "train_log", "step": 3456, "epoch": 1.4582278481012658, "progress_pct": 24.3, "epoch_pct": 24.3, "eta": "43:02:10", "max_grad_norm": 1.0, "loss": 0.6663834452629089, "grad_norm": 0.43322354555130005, "learning_rate": 0.00018491519806833795}
+{"ts": "2025-12-27T08:06:30", "event": "train_log", "step": 3458, "epoch": 1.459071729957806, "progress_pct": 24.32, "epoch_pct": 24.32, "eta": "43:01:03", "max_grad_norm": 1.0, "loss": 0.6634764671325684, "grad_norm": 0.421683132648468, "learning_rate": 0.0001848898217001323}
+{"ts": "2025-12-27T08:06:46", "event": "train_log", "step": 3460, "epoch": 1.459915611814346, "progress_pct": 24.33, "epoch_pct": 24.33, "eta": "42:59:53", "max_grad_norm": 1.0, "loss": 0.697095513343811, "grad_norm": 0.44365957379341125, "learning_rate": 0.00018486442574947511}
+{"ts": "2025-12-27T08:07:01", "event": "train_log", "step": 3462, "epoch": 1.460759493670886, "progress_pct": 24.35, "epoch_pct": 24.35, "eta": "42:58:43", "max_grad_norm": 1.0, "loss": 0.8009715676307678, "grad_norm": 0.4892813265323639, "learning_rate": 0.00018483901022222474}
+{"ts": "2025-12-27T08:07:17", "event": "train_log", "step": 3464, "epoch": 1.4616033755274263, "progress_pct": 24.36, "epoch_pct": 24.36, "eta": "42:57:33", "max_grad_norm": 1.0, "loss": 0.631499707698822, "grad_norm": 0.4230288863182068, "learning_rate": 0.00018481357512424408}
+{"ts": "2025-12-27T08:07:32", "event": "train_log", "step": 3466, "epoch": 1.4624472573839662, "progress_pct": 24.37, "epoch_pct": 24.37, "eta": "42:56:21", "max_grad_norm": 1.0, "loss": 0.8045525550842285, "grad_norm": 0.49490758776664734, "learning_rate": 0.00018478812046140056}
+{"ts": "2025-12-27T08:07:47", "event": "train_log", "step": 3468, "epoch": 1.4632911392405064, "progress_pct": 24.39, "epoch_pct": 24.39, "eta": "42:55:10", "max_grad_norm": 1.0, "loss": 0.7299133539199829, "grad_norm": 0.5035375952720642, "learning_rate": 0.00018476264623956597}
+{"ts": "2025-12-27T08:08:02", "event": "train_log", "step": 3470, "epoch": 1.4641350210970465, "progress_pct": 24.4, "epoch_pct": 24.4, "eta": "42:53:58", "max_grad_norm": 1.0, "loss": 0.7313228845596313, "grad_norm": 0.45048409700393677, "learning_rate": 0.00018473715246461684}
+{"ts": "2025-12-27T08:08:16", "event": "train_log", "step": 3472, "epoch": 1.4649789029535865, "progress_pct": 24.42, "epoch_pct": 24.42, "eta": "42:52:46", "max_grad_norm": 1.0, "loss": 0.7515901923179626, "grad_norm": 0.41414541006088257, "learning_rate": 0.00018471163914243405}
+{"ts": "2025-12-27T08:08:31", "event": "train_log", "step": 3474, "epoch": 1.4658227848101266, "progress_pct": 24.43, "epoch_pct": 24.43, "eta": "42:51:35", "max_grad_norm": 1.0, "loss": 0.7425532341003418, "grad_norm": 0.4440646767616272, "learning_rate": 0.00018468610627890298}
+{"ts": "2025-12-27T08:08:46", "event": "train_log", "step": 3476, "epoch": 1.4666666666666668, "progress_pct": 24.44, "epoch_pct": 24.44, "eta": "42:50:23", "max_grad_norm": 1.0, "loss": 0.7509122490882874, "grad_norm": 0.4749740660190582, "learning_rate": 0.00018466055387991362}
+{"ts": "2025-12-27T08:09:00", "event": "train_log", "step": 3478, "epoch": 1.4675105485232067, "progress_pct": 24.46, "epoch_pct": 24.46, "eta": "42:49:10", "max_grad_norm": 1.0, "loss": 0.7395774126052856, "grad_norm": 0.498951256275177, "learning_rate": 0.00018463498195136046}
+{"ts": "2025-12-27T08:09:15", "event": "train_log", "step": 3480, "epoch": 1.4683544303797469, "progress_pct": 24.47, "epoch_pct": 24.47, "eta": "42:47:58", "max_grad_norm": 1.0, "loss": 0.7860153317451477, "grad_norm": 0.45996859669685364, "learning_rate": 0.00018460939049914237}
+{"ts": "2025-12-27T08:09:30", "event": "train_log", "step": 3482, "epoch": 1.469198312236287, "progress_pct": 24.49, "epoch_pct": 24.49, "eta": "42:46:47", "max_grad_norm": 1.0, "loss": 0.8193565607070923, "grad_norm": 0.5989604592323303, "learning_rate": 0.0001845837795291629}
+{"ts": "2025-12-27T08:09:45", "event": "train_log", "step": 3484, "epoch": 1.470042194092827, "progress_pct": 24.5, "epoch_pct": 24.5, "eta": "42:45:35", "max_grad_norm": 1.0, "loss": 0.6569210290908813, "grad_norm": 0.3903721272945404, "learning_rate": 0.0001845581490473299}
+{"ts": "2025-12-27T08:09:59", "event": "train_log", "step": 3486, "epoch": 1.4708860759493672, "progress_pct": 24.51, "epoch_pct": 24.51, "eta": "42:44:24", "max_grad_norm": 1.0, "loss": 0.7226642370223999, "grad_norm": 0.4316142201423645, "learning_rate": 0.00018453249905955592}
+{"ts": "2025-12-27T08:10:14", "event": "train_log", "step": 3488, "epoch": 1.471729957805907, "progress_pct": 24.53, "epoch_pct": 24.53, "eta": "42:43:11", "max_grad_norm": 1.0, "loss": 0.8134398460388184, "grad_norm": 0.4447585344314575, "learning_rate": 0.00018450682957175786}
+{"ts": "2025-12-27T08:10:29", "event": "train_log", "step": 3490, "epoch": 1.4725738396624473, "progress_pct": 24.54, "epoch_pct": 24.54, "eta": "42:41:59", "max_grad_norm": 1.0, "loss": 0.7668985724449158, "grad_norm": 0.42515450716018677, "learning_rate": 0.0001844811405898572}
+{"ts": "2025-12-27T08:10:43", "event": "train_log", "step": 3492, "epoch": 1.4734177215189874, "progress_pct": 24.56, "epoch_pct": 24.56, "eta": "42:40:49", "max_grad_norm": 1.0, "loss": 0.7289790511131287, "grad_norm": 0.39883363246917725, "learning_rate": 0.0001844554321197799}
+{"ts": "2025-12-27T08:10:58", "event": "train_log", "step": 3494, "epoch": 1.4742616033755274, "progress_pct": 24.57, "epoch_pct": 24.57, "eta": "42:39:36", "max_grad_norm": 1.0, "loss": 0.7611824870109558, "grad_norm": 0.4419080317020416, "learning_rate": 0.00018442970416745645}
+{"ts": "2025-12-27T08:11:13", "event": "train_log", "step": 3496, "epoch": 1.4751054852320675, "progress_pct": 24.59, "epoch_pct": 24.59, "eta": "42:38:27", "max_grad_norm": 1.0, "loss": 0.6926230788230896, "grad_norm": 0.37460315227508545, "learning_rate": 0.0001844039567388217}
+{"ts": "2025-12-27T08:11:28", "event": "train_log", "step": 3498, "epoch": 1.4759493670886075, "progress_pct": 24.6, "epoch_pct": 24.6, "eta": "42:37:17", "max_grad_norm": 1.0, "loss": 0.7889325618743896, "grad_norm": 0.47706854343414307, "learning_rate": 0.00018437818983981513}
+{"ts": "2025-12-27T08:11:43", "event": "train_log", "step": 3500, "epoch": 1.4767932489451476, "progress_pct": 24.61, "epoch_pct": 24.61, "eta": "42:36:05", "max_grad_norm": 1.0, "loss": 0.7231656312942505, "grad_norm": 0.449028879404068, "learning_rate": 0.00018435240347638074}
+{"ts": "2025-12-27T08:22:55", "event": "train_log", "step": 3500, "epoch": 1.4767932489451476, "progress_pct": 24.61, "epoch_pct": 24.61, "eta": "43:10:24", "max_grad_norm": 1.0, "eval_loss": 0.7902651429176331, "eval_runtime": 672.312, "eval_samples_per_second": 3.134, "eval_steps_per_second": 3.134}
+{"ts": "2025-12-27T08:23:12", "event": "train_log", "step": 3502, "epoch": 1.4776371308016878, "progress_pct": 24.63, "epoch_pct": 24.63, "eta": "43:09:17", "max_grad_norm": 1.0, "loss": 0.7464927434921265, "grad_norm": 0.44085755944252014, "learning_rate": 0.00018432659765446686}
+{"ts": "2025-12-27T08:23:25", "event": "train_log", "step": 3504, "epoch": 1.4784810126582277, "progress_pct": 24.64, "epoch_pct": 24.64, "eta": "43:08:01", "max_grad_norm": 1.0, "loss": 0.7312393188476562, "grad_norm": 0.4243546724319458, "learning_rate": 0.0001843007723800264}
+{"ts": "2025-12-27T08:23:42", "event": "train_log", "step": 3506, "epoch": 1.479324894514768, "progress_pct": 24.66, "epoch_pct": 24.66, "eta": "43:06:54", "max_grad_norm": 1.0, "loss": 0.7169944643974304, "grad_norm": 0.4437529444694519, "learning_rate": 0.00018427492765901677}
+{"ts": "2025-12-27T08:23:57", "event": "train_log", "step": 3508, "epoch": 1.480168776371308, "progress_pct": 24.67, "epoch_pct": 24.67, "eta": "43:05:43", "max_grad_norm": 1.0, "loss": 0.6169725060462952, "grad_norm": 0.3639313280582428, "learning_rate": 0.0001842490634973999}
+{"ts": "2025-12-27T08:24:13", "event": "train_log", "step": 3510, "epoch": 1.481012658227848, "progress_pct": 24.68, "epoch_pct": 24.68, "eta": "43:04:33", "max_grad_norm": 1.0, "loss": 0.7034649848937988, "grad_norm": 0.4452592730522156, "learning_rate": 0.0001842231799011421}
+{"ts": "2025-12-27T08:24:27", "event": "train_log", "step": 3512, "epoch": 1.4818565400843882, "progress_pct": 24.7, "epoch_pct": 24.7, "eta": "43:03:19", "max_grad_norm": 1.0, "loss": 0.7535615563392639, "grad_norm": 0.5229591131210327, "learning_rate": 0.00018419727687621421}
+{"ts": "2025-12-27T08:24:43", "event": "train_log", "step": 3514, "epoch": 1.4827004219409283, "progress_pct": 24.71, "epoch_pct": 24.71, "eta": "43:02:11", "max_grad_norm": 1.0, "loss": 0.6858918070793152, "grad_norm": 0.40748512744903564, "learning_rate": 0.00018417135442859163}
+{"ts": "2025-12-27T08:24:58", "event": "train_log", "step": 3516, "epoch": 1.4835443037974683, "progress_pct": 24.73, "epoch_pct": 24.73, "eta": "43:01:01", "max_grad_norm": 1.0, "loss": 0.7120109796524048, "grad_norm": 0.4532773792743683, "learning_rate": 0.0001841454125642541}
+{"ts": "2025-12-27T08:25:13", "event": "train_log", "step": 3518, "epoch": 1.4843881856540084, "progress_pct": 24.74, "epoch_pct": 24.74, "eta": "42:59:50", "max_grad_norm": 1.0, "loss": 0.6837015748023987, "grad_norm": 0.4281020760536194, "learning_rate": 0.00018411945128918593}
+{"ts": "2025-12-27T08:25:29", "event": "train_log", "step": 3520, "epoch": 1.4852320675105486, "progress_pct": 24.75, "epoch_pct": 24.75, "eta": "42:58:41", "max_grad_norm": 1.0, "loss": 0.6639618873596191, "grad_norm": 0.4891088306903839, "learning_rate": 0.0001840934706093759}
+{"ts": "2025-12-27T08:25:44", "event": "train_log", "step": 3522, "epoch": 1.4860759493670885, "progress_pct": 24.77, "epoch_pct": 24.77, "eta": "42:57:30", "max_grad_norm": 1.0, "loss": 0.693483829498291, "grad_norm": 0.44358906149864197, "learning_rate": 0.00018406747053081724}
+{"ts": "2025-12-27T08:26:01", "event": "train_log", "step": 3524, "epoch": 1.4869198312236287, "progress_pct": 24.78, "epoch_pct": 24.78, "eta": "42:56:23", "max_grad_norm": 1.0, "loss": 0.7046290636062622, "grad_norm": 0.40134626626968384, "learning_rate": 0.00018404145105950767}
+{"ts": "2025-12-27T08:26:17", "event": "train_log", "step": 3526, "epoch": 1.4877637130801689, "progress_pct": 24.8, "epoch_pct": 24.8, "eta": "42:55:16", "max_grad_norm": 1.0, "loss": 0.6991957426071167, "grad_norm": 0.4601934552192688, "learning_rate": 0.0001840154122014494}
+{"ts": "2025-12-27T08:26:33", "event": "train_log", "step": 3528, "epoch": 1.4886075949367088, "progress_pct": 24.81, "epoch_pct": 24.81, "eta": "42:54:09", "max_grad_norm": 1.0, "loss": 0.667023241519928, "grad_norm": 0.39107242226600647, "learning_rate": 0.00018398935396264906}
+{"ts": "2025-12-27T08:26:50", "event": "train_log", "step": 3530, "epoch": 1.489451476793249, "progress_pct": 24.82, "epoch_pct": 24.82, "eta": "42:53:03", "max_grad_norm": 1.0, "loss": 0.7187332510948181, "grad_norm": 0.4498196542263031, "learning_rate": 0.00018396327634911784}
+{"ts": "2025-12-27T08:27:06", "event": "train_log", "step": 3532, "epoch": 1.4902953586497891, "progress_pct": 24.84, "epoch_pct": 24.84, "eta": "42:51:55", "max_grad_norm": 1.0, "loss": 0.6854014992713928, "grad_norm": 0.41769179701805115, "learning_rate": 0.0001839371793668713}
+{"ts": "2025-12-27T08:27:21", "event": "train_log", "step": 3534, "epoch": 1.491139240506329, "progress_pct": 24.85, "epoch_pct": 24.85, "eta": "42:50:44", "max_grad_norm": 1.0, "loss": 0.6643120050430298, "grad_norm": 0.40170612931251526, "learning_rate": 0.0001839110630219295}
+{"ts": "2025-12-27T08:27:36", "event": "train_log", "step": 3536, "epoch": 1.4919831223628692, "progress_pct": 24.87, "epoch_pct": 24.87, "eta": "42:49:33", "max_grad_norm": 1.0, "loss": 0.6822295188903809, "grad_norm": 0.43697425723075867, "learning_rate": 0.00018388492732031701}
+{"ts": "2025-12-27T08:27:52", "event": "train_log", "step": 3538, "epoch": 1.4928270042194094, "progress_pct": 24.88, "epoch_pct": 24.88, "eta": "42:48:25", "max_grad_norm": 1.0, "loss": 0.7500558495521545, "grad_norm": 0.4400080442428589, "learning_rate": 0.00018385877226806288}
+{"ts": "2025-12-27T08:28:08", "event": "train_log", "step": 3540, "epoch": 1.4936708860759493, "progress_pct": 24.89, "epoch_pct": 24.89, "eta": "42:47:19", "max_grad_norm": 1.0, "loss": 0.7742161750793457, "grad_norm": 0.5162575244903564, "learning_rate": 0.00018383259787120048}
+{"ts": "2025-12-27T08:28:24", "event": "train_log", "step": 3542, "epoch": 1.4945147679324895, "progress_pct": 24.91, "epoch_pct": 24.91, "eta": "42:46:11", "max_grad_norm": 1.0, "loss": 0.7334147095680237, "grad_norm": 0.4234789311885834, "learning_rate": 0.0001838064041357678}
+{"ts": "2025-12-27T08:28:41", "event": "train_log", "step": 3544, "epoch": 1.4953586497890297, "progress_pct": 24.92, "epoch_pct": 24.92, "eta": "42:45:05", "max_grad_norm": 1.0, "loss": 0.691303014755249, "grad_norm": 0.40562525391578674, "learning_rate": 0.0001837801910678072}
+{"ts": "2025-12-27T08:28:55", "event": "train_log", "step": 3546, "epoch": 1.4962025316455696, "progress_pct": 24.94, "epoch_pct": 24.94, "eta": "42:43:54", "max_grad_norm": 1.0, "loss": 0.6478174328804016, "grad_norm": 0.3651523292064667, "learning_rate": 0.00018375395867336556}
+{"ts": "2025-12-27T08:29:11", "event": "train_log", "step": 3548, "epoch": 1.4970464135021098, "progress_pct": 24.95, "epoch_pct": 24.95, "eta": "42:42:46", "max_grad_norm": 1.0, "loss": 0.6800187826156616, "grad_norm": 0.45428016781806946, "learning_rate": 0.00018372770695849417}
+{"ts": "2025-12-27T08:29:29", "event": "train_log", "step": 3550, "epoch": 1.49789029535865, "progress_pct": 24.96, "epoch_pct": 24.96, "eta": "42:41:43", "max_grad_norm": 1.0, "loss": 0.7386951446533203, "grad_norm": 0.4258574843406677, "learning_rate": 0.00018370143592924882}
+{"ts": "2025-12-27T08:29:43", "event": "train_log", "step": 3552, "epoch": 1.4987341772151899, "progress_pct": 24.98, "epoch_pct": 24.98, "eta": "42:40:31", "max_grad_norm": 1.0, "loss": 0.757290780544281, "grad_norm": 0.4260908365249634, "learning_rate": 0.00018367514559168972}
+{"ts": "2025-12-27T08:29:58", "event": "train_log", "step": 3554, "epoch": 1.49957805907173, "progress_pct": 24.99, "epoch_pct": 24.99, "eta": "42:39:22", "max_grad_norm": 1.0, "loss": 0.7254101037979126, "grad_norm": 0.4037017226219177, "learning_rate": 0.00018364883595188152}
+{"ts": "2025-12-27T08:30:13", "event": "train_log", "step": 3556, "epoch": 1.5004219409282702, "progress_pct": 25.01, "epoch_pct": 25.01, "eta": "42:38:11", "max_grad_norm": 1.0, "loss": 0.7342239618301392, "grad_norm": 0.4212391674518585, "learning_rate": 0.0001836225070158934}
+{"ts": "2025-12-27T08:30:28", "event": "train_log", "step": 3558, "epoch": 1.5012658227848101, "progress_pct": 25.02, "epoch_pct": 25.02, "eta": "42:36:59", "max_grad_norm": 1.0, "loss": 0.7147436738014221, "grad_norm": 0.4902292490005493, "learning_rate": 0.0001835961587897989}
+{"ts": "2025-12-27T08:30:44", "event": "train_log", "step": 3560, "epoch": 1.50210970464135, "progress_pct": 25.04, "epoch_pct": 25.04, "eta": "42:35:53", "max_grad_norm": 1.0, "loss": 0.7473883628845215, "grad_norm": 0.39230459928512573, "learning_rate": 0.00018356979127967605}
+{"ts": "2025-12-27T08:31:00", "event": "train_log", "step": 3562, "epoch": 1.5029535864978905, "progress_pct": 25.05, "epoch_pct": 25.05, "eta": "42:34:45", "max_grad_norm": 1.0, "loss": 0.7026845812797546, "grad_norm": 0.4203200340270996, "learning_rate": 0.00018354340449160735}
+{"ts": "2025-12-27T08:31:15", "event": "train_log", "step": 3564, "epoch": 1.5037974683544304, "progress_pct": 25.06, "epoch_pct": 25.06, "eta": "42:33:37", "max_grad_norm": 1.0, "loss": 0.7426526546478271, "grad_norm": 0.44363775849342346, "learning_rate": 0.00018351699843167972}
+{"ts": "2025-12-27T08:31:32", "event": "train_log", "step": 3566, "epoch": 1.5046413502109703, "progress_pct": 25.08, "epoch_pct": 25.08, "eta": "42:32:32", "max_grad_norm": 1.0, "loss": 0.6800932884216309, "grad_norm": 0.4479754865169525, "learning_rate": 0.0001834905731059845}
+{"ts": "2025-12-27T08:31:49", "event": "train_log", "step": 3568, "epoch": 1.5054852320675105, "progress_pct": 25.09, "epoch_pct": 25.09, "eta": "42:31:28", "max_grad_norm": 1.0, "loss": 0.6681315898895264, "grad_norm": 0.43898341059684753, "learning_rate": 0.0001834641285206176}
+{"ts": "2025-12-27T08:32:03", "event": "train_log", "step": 3570, "epoch": 1.5063291139240507, "progress_pct": 25.11, "epoch_pct": 25.11, "eta": "42:30:16", "max_grad_norm": 1.0, "loss": 0.7319603562355042, "grad_norm": 0.44789746403694153, "learning_rate": 0.00018343766468167916}
+{"ts": "2025-12-27T08:32:20", "event": "train_log", "step": 3572, "epoch": 1.5071729957805906, "progress_pct": 25.12, "epoch_pct": 25.12, "eta": "42:29:11", "max_grad_norm": 1.0, "loss": 0.7205765843391418, "grad_norm": 0.44450023770332336, "learning_rate": 0.00018341118159527392}
+{"ts": "2025-12-27T08:32:36", "event": "train_log", "step": 3574, "epoch": 1.5080168776371308, "progress_pct": 25.13, "epoch_pct": 25.13, "eta": "42:28:05", "max_grad_norm": 1.0, "loss": 0.8053180575370789, "grad_norm": 0.4461601674556732, "learning_rate": 0.00018338467926751103}
+{"ts": "2025-12-27T08:32:51", "event": "train_log", "step": 3576, "epoch": 1.508860759493671, "progress_pct": 25.15, "epoch_pct": 25.15, "eta": "42:26:57", "max_grad_norm": 1.0, "loss": 0.6944835186004639, "grad_norm": 0.415216863155365, "learning_rate": 0.00018335815770450408}
+{"ts": "2025-12-27T08:33:07", "event": "train_log", "step": 3578, "epoch": 1.5097046413502109, "progress_pct": 25.16, "epoch_pct": 25.16, "eta": "42:25:48", "max_grad_norm": 1.0, "loss": 0.7333522439002991, "grad_norm": 0.48354068398475647, "learning_rate": 0.00018333161691237105}
+{"ts": "2025-12-27T08:33:23", "event": "train_log", "step": 3580, "epoch": 1.510548523206751, "progress_pct": 25.18, "epoch_pct": 25.18, "eta": "42:24:43", "max_grad_norm": 1.0, "loss": 0.7036639451980591, "grad_norm": 0.44450995326042175, "learning_rate": 0.00018330505689723446}
+{"ts": "2025-12-27T08:33:39", "event": "train_log", "step": 3582, "epoch": 1.5113924050632912, "progress_pct": 25.19, "epoch_pct": 25.19, "eta": "42:23:36", "max_grad_norm": 1.0, "loss": 0.6589286923408508, "grad_norm": 0.355945348739624, "learning_rate": 0.00018327847766522111}
+{"ts": "2025-12-27T08:33:54", "event": "train_log", "step": 3584, "epoch": 1.5122362869198311, "progress_pct": 25.2, "epoch_pct": 25.2, "eta": "42:22:27", "max_grad_norm": 1.0, "loss": 0.7284913063049316, "grad_norm": 0.4532116651535034, "learning_rate": 0.00018325187922246236}
+{"ts": "2025-12-27T08:34:09", "event": "train_log", "step": 3586, "epoch": 1.5130801687763713, "progress_pct": 25.22, "epoch_pct": 25.22, "eta": "42:21:18", "max_grad_norm": 1.0, "loss": 0.7253853678703308, "grad_norm": 0.45910802483558655, "learning_rate": 0.00018322526157509396}
+{"ts": "2025-12-27T08:34:27", "event": "train_log", "step": 3588, "epoch": 1.5139240506329115, "progress_pct": 25.23, "epoch_pct": 25.23, "eta": "42:20:16", "max_grad_norm": 1.0, "loss": 0.7626010179519653, "grad_norm": 0.4729263186454773, "learning_rate": 0.0001831986247292561}
+{"ts": "2025-12-27T08:34:43", "event": "train_log", "step": 3590, "epoch": 1.5147679324894514, "progress_pct": 25.25, "epoch_pct": 25.25, "eta": "42:19:11", "max_grad_norm": 1.0, "loss": 0.705466091632843, "grad_norm": 0.42071840167045593, "learning_rate": 0.00018317196869109336}
+{"ts": "2025-12-27T08:34:59", "event": "train_log", "step": 3592, "epoch": 1.5156118143459916, "progress_pct": 25.26, "epoch_pct": 25.26, "eta": "42:18:07", "max_grad_norm": 1.0, "loss": 0.6900550723075867, "grad_norm": 0.40275856852531433, "learning_rate": 0.00018314529346675479}
+{"ts": "2025-12-27T08:35:15", "event": "train_log", "step": 3594, "epoch": 1.5164556962025317, "progress_pct": 25.27, "epoch_pct": 25.27, "eta": "42:16:59", "max_grad_norm": 1.0, "loss": 0.6951512694358826, "grad_norm": 0.43557414412498474, "learning_rate": 0.00018311859906239387}
+{"ts": "2025-12-27T08:35:30", "event": "train_log", "step": 3596, "epoch": 1.5172995780590717, "progress_pct": 25.29, "epoch_pct": 25.29, "eta": "42:15:51", "max_grad_norm": 1.0, "loss": 0.7453750967979431, "grad_norm": 0.4007355570793152, "learning_rate": 0.00018309188548416844}
+{"ts": "2025-12-27T08:35:46", "event": "train_log", "step": 3598, "epoch": 1.5181434599156118, "progress_pct": 25.3, "epoch_pct": 25.3, "eta": "42:14:46", "max_grad_norm": 1.0, "loss": 0.7549434900283813, "grad_norm": 0.4686451256275177, "learning_rate": 0.00018306515273824086}
+{"ts": "2025-12-27T08:36:01", "event": "train_log", "step": 3600, "epoch": 1.518987341772152, "progress_pct": 25.32, "epoch_pct": 25.32, "eta": "42:13:35", "max_grad_norm": 1.0, "loss": 0.7708864212036133, "grad_norm": 0.47308704257011414, "learning_rate": 0.0001830384008307778}
+{"ts": "2025-12-27T08:47:16", "event": "train_log", "step": 3600, "epoch": 1.518987341772152, "progress_pct": 25.32, "epoch_pct": 25.32, "eta": "42:46:47", "max_grad_norm": 1.0, "eval_loss": 0.785450279712677, "eval_runtime": 675.2312, "eval_samples_per_second": 3.12, "eval_steps_per_second": 3.12}
+{"ts": "2025-12-27T08:47:32", "event": "train_log", "step": 3602, "epoch": 1.519831223628692, "progress_pct": 25.33, "epoch_pct": 25.33, "eta": "42:45:39", "max_grad_norm": 1.0, "loss": 0.7501338124275208, "grad_norm": 0.4343389570713043, "learning_rate": 0.0001830116297679505}
+{"ts": "2025-12-27T08:47:46", "event": "train_log", "step": 3604, "epoch": 1.520675105485232, "progress_pct": 25.34, "epoch_pct": 25.34, "eta": "42:44:28", "max_grad_norm": 1.0, "loss": 0.7358015179634094, "grad_norm": 0.4971345067024231, "learning_rate": 0.00018298483955593445}
+{"ts": "2025-12-27T08:48:01", "event": "train_log", "step": 3606, "epoch": 1.5215189873417723, "progress_pct": 25.36, "epoch_pct": 25.36, "eta": "42:43:17", "max_grad_norm": 1.0, "loss": 0.7972728610038757, "grad_norm": 0.42013493180274963, "learning_rate": 0.0001829580302009097}
+{"ts": "2025-12-27T08:48:17", "event": "train_log", "step": 3608, "epoch": 1.5223628691983122, "progress_pct": 25.37, "epoch_pct": 25.37, "eta": "42:42:08", "max_grad_norm": 1.0, "loss": 0.5908644199371338, "grad_norm": 0.44829806685447693, "learning_rate": 0.00018293120170906062}
+{"ts": "2025-12-27T08:48:31", "event": "train_log", "step": 3610, "epoch": 1.5232067510548524, "progress_pct": 25.39, "epoch_pct": 25.39, "eta": "42:40:56", "max_grad_norm": 1.0, "loss": 0.7872797846794128, "grad_norm": 0.47412917017936707, "learning_rate": 0.000182904354086576}
+{"ts": "2025-12-27T08:48:46", "event": "train_log", "step": 3612, "epoch": 1.5240506329113925, "progress_pct": 25.4, "epoch_pct": 25.4, "eta": "42:39:46", "max_grad_norm": 1.0, "loss": 0.7327616214752197, "grad_norm": 0.4224211275577545, "learning_rate": 0.0001828774873396491}
+{"ts": "2025-12-27T08:49:01", "event": "train_log", "step": 3614, "epoch": 1.5248945147679325, "progress_pct": 25.41, "epoch_pct": 25.41, "eta": "42:38:37", "max_grad_norm": 1.0, "loss": 0.6649892330169678, "grad_norm": 0.4055147171020508, "learning_rate": 0.00018285060147447756}
+{"ts": "2025-12-27T08:49:17", "event": "train_log", "step": 3616, "epoch": 1.5257383966244724, "progress_pct": 25.43, "epoch_pct": 25.43, "eta": "42:37:28", "max_grad_norm": 1.0, "loss": 0.6939427852630615, "grad_norm": 0.4058035910129547, "learning_rate": 0.00018282369649726346}
+{"ts": "2025-12-27T08:49:32", "event": "train_log", "step": 3618, "epoch": 1.5265822784810128, "progress_pct": 25.44, "epoch_pct": 25.44, "eta": "42:36:18", "max_grad_norm": 1.0, "loss": 0.6351069211959839, "grad_norm": 0.4192068874835968, "learning_rate": 0.00018279677241421325}
+{"ts": "2025-12-27T08:49:46", "event": "train_log", "step": 3620, "epoch": 1.5274261603375527, "progress_pct": 25.46, "epoch_pct": 25.46, "eta": "42:35:08", "max_grad_norm": 1.0, "loss": 0.6963213682174683, "grad_norm": 0.4659916162490845, "learning_rate": 0.00018276982923153776}
+{"ts": "2025-12-27T08:50:00", "event": "train_log", "step": 3622, "epoch": 1.5282700421940927, "progress_pct": 25.47, "epoch_pct": 25.47, "eta": "42:33:55", "max_grad_norm": 1.0, "loss": 0.7463255524635315, "grad_norm": 0.4993051290512085, "learning_rate": 0.00018274286695545229}
+{"ts": "2025-12-27T08:50:15", "event": "train_log", "step": 3624, "epoch": 1.529113924050633, "progress_pct": 25.49, "epoch_pct": 25.49, "eta": "42:32:45", "max_grad_norm": 1.0, "loss": 0.6566499471664429, "grad_norm": 0.39232251048088074, "learning_rate": 0.00018271588559217654}
+{"ts": "2025-12-27T08:50:30", "event": "train_log", "step": 3626, "epoch": 1.529957805907173, "progress_pct": 25.5, "epoch_pct": 25.5, "eta": "42:31:34", "max_grad_norm": 1.0, "loss": 0.7695494890213013, "grad_norm": 0.4482063353061676, "learning_rate": 0.00018268888514793457}
+{"ts": "2025-12-27T08:50:45", "event": "train_log", "step": 3628, "epoch": 1.530801687763713, "progress_pct": 25.51, "epoch_pct": 25.51, "eta": "42:30:25", "max_grad_norm": 1.0, "loss": 0.7357239723205566, "grad_norm": 0.4000336825847626, "learning_rate": 0.00018266186562895483}
+{"ts": "2025-12-27T08:51:00", "event": "train_log", "step": 3630, "epoch": 1.5316455696202531, "progress_pct": 25.53, "epoch_pct": 25.53, "eta": "42:29:15", "max_grad_norm": 1.0, "loss": 0.7234170436859131, "grad_norm": 0.4220462441444397, "learning_rate": 0.0001826348270414703}
+{"ts": "2025-12-27T08:51:14", "event": "train_log", "step": 3632, "epoch": 1.5324894514767933, "progress_pct": 25.54, "epoch_pct": 25.54, "eta": "42:28:03", "max_grad_norm": 1.0, "loss": 0.7864061594009399, "grad_norm": 0.4787490665912628, "learning_rate": 0.00018260776939171814}
+{"ts": "2025-12-27T08:51:29", "event": "train_log", "step": 3634, "epoch": 1.5333333333333332, "progress_pct": 25.56, "epoch_pct": 25.56, "eta": "42:26:56", "max_grad_norm": 1.0, "loss": 0.670537531375885, "grad_norm": 0.3670683205127716, "learning_rate": 0.00018258069268594013}
+{"ts": "2025-12-27T08:51:45", "event": "train_log", "step": 3636, "epoch": 1.5341772151898734, "progress_pct": 25.57, "epoch_pct": 25.57, "eta": "42:25:48", "max_grad_norm": 1.0, "loss": 0.7535691261291504, "grad_norm": 0.41599538922309875, "learning_rate": 0.0001825535969303823}
+{"ts": "2025-12-27T08:51:59", "event": "train_log", "step": 3638, "epoch": 1.5350210970464135, "progress_pct": 25.58, "epoch_pct": 25.58, "eta": "42:24:36", "max_grad_norm": 1.0, "loss": 0.7553712725639343, "grad_norm": 0.5033245086669922, "learning_rate": 0.00018252648213129514}
+{"ts": "2025-12-27T08:52:15", "event": "train_log", "step": 3640, "epoch": 1.5358649789029535, "progress_pct": 25.6, "epoch_pct": 25.6, "eta": "42:23:30", "max_grad_norm": 1.0, "loss": 0.7149441242218018, "grad_norm": 0.4235823452472687, "learning_rate": 0.0001824993482949335}
+{"ts": "2025-12-27T08:52:30", "event": "train_log", "step": 3642, "epoch": 1.5367088607594936, "progress_pct": 25.61, "epoch_pct": 25.61, "eta": "42:22:22", "max_grad_norm": 1.0, "loss": 0.735568642616272, "grad_norm": 0.37993329763412476, "learning_rate": 0.00018247219542755664}
+{"ts": "2025-12-27T08:52:46", "event": "train_log", "step": 3644, "epoch": 1.5375527426160338, "progress_pct": 25.63, "epoch_pct": 25.63, "eta": "42:21:14", "max_grad_norm": 1.0, "loss": 0.693510115146637, "grad_norm": 0.39534425735473633, "learning_rate": 0.00018244502353542818}
+{"ts": "2025-12-27T08:53:01", "event": "train_log", "step": 3646, "epoch": 1.5383966244725737, "progress_pct": 25.64, "epoch_pct": 25.64, "eta": "42:20:05", "max_grad_norm": 1.0, "loss": 0.7261984348297119, "grad_norm": 0.45447441935539246, "learning_rate": 0.00018241783262481622}
+{"ts": "2025-12-27T08:53:18", "event": "train_log", "step": 3648, "epoch": 1.539240506329114, "progress_pct": 25.65, "epoch_pct": 25.65, "eta": "42:19:02", "max_grad_norm": 1.0, "loss": 0.7518824338912964, "grad_norm": 0.4131409823894501, "learning_rate": 0.0001823906227019931}
+{"ts": "2025-12-27T08:53:33", "event": "train_log", "step": 3650, "epoch": 1.540084388185654, "progress_pct": 25.67, "epoch_pct": 25.67, "eta": "42:17:54", "max_grad_norm": 1.0, "loss": 0.7650659680366516, "grad_norm": 0.46603885293006897, "learning_rate": 0.00018236339377323567}
+{"ts": "2025-12-27T08:53:51", "event": "train_log", "step": 3652, "epoch": 1.540928270042194, "progress_pct": 25.68, "epoch_pct": 25.68, "eta": "42:16:53", "max_grad_norm": 1.0, "loss": 0.6436064839363098, "grad_norm": 0.3998747766017914, "learning_rate": 0.00018233614584482513}
+{"ts": "2025-12-27T08:54:07", "event": "train_log", "step": 3654, "epoch": 1.5417721518987342, "progress_pct": 25.7, "epoch_pct": 25.7, "eta": "42:15:47", "max_grad_norm": 1.0, "loss": 0.7068451642990112, "grad_norm": 0.4314839839935303, "learning_rate": 0.00018230887892304703}
+{"ts": "2025-12-27T08:54:22", "event": "train_log", "step": 3656, "epoch": 1.5426160337552743, "progress_pct": 25.71, "epoch_pct": 25.71, "eta": "42:14:38", "max_grad_norm": 1.0, "loss": 0.7342977523803711, "grad_norm": 0.4324610233306885, "learning_rate": 0.0001822815930141913}
+{"ts": "2025-12-27T08:54:38", "event": "train_log", "step": 3658, "epoch": 1.5434599156118143, "progress_pct": 25.72, "epoch_pct": 25.72, "eta": "42:13:32", "max_grad_norm": 1.0, "loss": 0.6482785940170288, "grad_norm": 0.4385458827018738, "learning_rate": 0.0001822542881245524}
+{"ts": "2025-12-27T08:54:54", "event": "train_log", "step": 3660, "epoch": 1.5443037974683544, "progress_pct": 25.74, "epoch_pct": 25.74, "eta": "42:12:27", "max_grad_norm": 1.0, "loss": 0.6319621205329895, "grad_norm": 0.3736587166786194, "learning_rate": 0.0001822269642604289}
+{"ts": "2025-12-27T08:55:08", "event": "train_log", "step": 3662, "epoch": 1.5451476793248946, "progress_pct": 25.75, "epoch_pct": 25.75, "eta": "42:11:17", "max_grad_norm": 1.0, "loss": 0.6603307723999023, "grad_norm": 0.42113035917282104, "learning_rate": 0.00018219962142812393}
+{"ts": "2025-12-27T08:55:24", "event": "train_log", "step": 3664, "epoch": 1.5459915611814345, "progress_pct": 25.77, "epoch_pct": 25.77, "eta": "42:10:09", "max_grad_norm": 1.0, "loss": 0.6874772906303406, "grad_norm": 0.422158420085907, "learning_rate": 0.00018217225963394496}
+{"ts": "2025-12-27T08:55:40", "event": "train_log", "step": 3666, "epoch": 1.5468354430379747, "progress_pct": 25.78, "epoch_pct": 25.78, "eta": "42:09:06", "max_grad_norm": 1.0, "loss": 0.7231854796409607, "grad_norm": 0.42724940180778503, "learning_rate": 0.00018214487888420388}
+{"ts": "2025-12-27T08:55:58", "event": "train_log", "step": 3668, "epoch": 1.5476793248945149, "progress_pct": 25.79, "epoch_pct": 25.79, "eta": "42:08:06", "max_grad_norm": 1.0, "loss": 0.6418231129646301, "grad_norm": 0.3747844398021698, "learning_rate": 0.00018211747918521685}
+{"ts": "2025-12-27T08:56:15", "event": "train_log", "step": 3670, "epoch": 1.5485232067510548, "progress_pct": 25.81, "epoch_pct": 25.81, "eta": "42:07:03", "max_grad_norm": 1.0, "loss": 0.7019823789596558, "grad_norm": 0.37007269263267517, "learning_rate": 0.00018209006054330446}
+{"ts": "2025-12-27T08:56:30", "event": "train_log", "step": 3672, "epoch": 1.549367088607595, "progress_pct": 25.82, "epoch_pct": 25.82, "eta": "42:05:55", "max_grad_norm": 1.0, "loss": 0.6632764339447021, "grad_norm": 0.409701943397522, "learning_rate": 0.00018206262296479168}
+{"ts": "2025-12-27T08:56:45", "event": "train_log", "step": 3674, "epoch": 1.5502109704641351, "progress_pct": 25.84, "epoch_pct": 25.84, "eta": "42:04:47", "max_grad_norm": 1.0, "loss": 0.7327075600624084, "grad_norm": 0.49566221237182617, "learning_rate": 0.00018203516645600782}
+{"ts": "2025-12-27T08:57:01", "event": "train_log", "step": 3676, "epoch": 1.551054852320675, "progress_pct": 25.85, "epoch_pct": 25.85, "eta": "42:03:42", "max_grad_norm": 1.0, "loss": 0.6535149812698364, "grad_norm": 0.4549940526485443, "learning_rate": 0.0001820076910232866}
+{"ts": "2025-12-27T08:57:16", "event": "train_log", "step": 3678, "epoch": 1.5518987341772152, "progress_pct": 25.86, "epoch_pct": 25.86, "eta": "42:02:35", "max_grad_norm": 1.0, "loss": 0.7761320471763611, "grad_norm": 0.42619478702545166, "learning_rate": 0.000181980196672966}
+{"ts": "2025-12-27T08:57:31", "event": "train_log", "step": 3680, "epoch": 1.5527426160337554, "progress_pct": 25.88, "epoch_pct": 25.88, "eta": "42:01:26", "max_grad_norm": 1.0, "loss": 0.6728626489639282, "grad_norm": 0.4332887828350067, "learning_rate": 0.00018195268341138852}
+{"ts": "2025-12-27T08:57:47", "event": "train_log", "step": 3682, "epoch": 1.5535864978902953, "progress_pct": 25.89, "epoch_pct": 25.89, "eta": "42:00:19", "max_grad_norm": 1.0, "loss": 0.7369232177734375, "grad_norm": 0.4972975552082062, "learning_rate": 0.0001819251512449009}
+{"ts": "2025-12-27T08:58:03", "event": "train_log", "step": 3684, "epoch": 1.5544303797468353, "progress_pct": 25.91, "epoch_pct": 25.91, "eta": "41:59:15", "max_grad_norm": 1.0, "loss": 0.6703663468360901, "grad_norm": 0.3619556725025177, "learning_rate": 0.00018189760017985428}
+{"ts": "2025-12-27T08:58:19", "event": "train_log", "step": 3686, "epoch": 1.5552742616033757, "progress_pct": 25.92, "epoch_pct": 25.92, "eta": "41:58:11", "max_grad_norm": 1.0, "loss": 0.7183756828308105, "grad_norm": 0.46435829997062683, "learning_rate": 0.00018187003022260417}
+{"ts": "2025-12-27T08:58:36", "event": "train_log", "step": 3688, "epoch": 1.5561181434599156, "progress_pct": 25.94, "epoch_pct": 25.94, "eta": "41:57:08", "max_grad_norm": 1.0, "loss": 0.680042564868927, "grad_norm": 0.4605063498020172, "learning_rate": 0.00018184244137951045}
+{"ts": "2025-12-27T08:58:52", "event": "train_log", "step": 3690, "epoch": 1.5569620253164556, "progress_pct": 25.95, "epoch_pct": 25.95, "eta": "41:56:02", "max_grad_norm": 1.0, "loss": 0.679821789264679, "grad_norm": 0.4023124873638153, "learning_rate": 0.00018181483365693732}
+{"ts": "2025-12-27T08:59:07", "event": "train_log", "step": 3692, "epoch": 1.557805907172996, "progress_pct": 25.96, "epoch_pct": 25.96, "eta": "41:54:55", "max_grad_norm": 1.0, "loss": 0.7408921718597412, "grad_norm": 0.459499716758728, "learning_rate": 0.00018178720706125333}
+{"ts": "2025-12-27T08:59:22", "event": "train_log", "step": 3694, "epoch": 1.5586497890295359, "progress_pct": 25.98, "epoch_pct": 25.98, "eta": "41:53:48", "max_grad_norm": 1.0, "loss": 0.6926741003990173, "grad_norm": 0.4397135078907013, "learning_rate": 0.00018175956159883146}
+{"ts": "2025-12-27T08:59:38", "event": "train_log", "step": 3696, "epoch": 1.5594936708860758, "progress_pct": 25.99, "epoch_pct": 25.99, "eta": "41:52:43", "max_grad_norm": 1.0, "loss": 0.752724826335907, "grad_norm": 0.4476170837879181, "learning_rate": 0.0001817318972760489}
+{"ts": "2025-12-27T08:59:53", "event": "train_log", "step": 3698, "epoch": 1.560337552742616, "progress_pct": 26.01, "epoch_pct": 26.01, "eta": "41:51:36", "max_grad_norm": 1.0, "loss": 0.7217722535133362, "grad_norm": 0.43684154748916626, "learning_rate": 0.00018170421409928735}
+{"ts": "2025-12-27T09:00:10", "event": "train_log", "step": 3700, "epoch": 1.5611814345991561, "progress_pct": 26.02, "epoch_pct": 26.02, "eta": "41:50:33", "max_grad_norm": 1.0, "loss": 0.7264936566352844, "grad_norm": 0.4118131995201111, "learning_rate": 0.00018167651207493278}
+{"ts": "2025-12-27T09:11:38", "event": "train_log", "step": 3700, "epoch": 1.5611814345991561, "progress_pct": 26.02, "epoch_pct": 26.02, "eta": "42:23:09", "max_grad_norm": 1.0, "eval_loss": 0.7854447960853577, "eval_runtime": 687.7907, "eval_samples_per_second": 3.063, "eval_steps_per_second": 3.063}
+{"ts": "2025-12-27T09:11:53", "event": "train_log", "step": 3702, "epoch": 1.562025316455696, "progress_pct": 26.03, "epoch_pct": 26.03, "eta": "42:22:00", "max_grad_norm": 1.0, "loss": 0.701478123664856, "grad_norm": 0.40639030933380127, "learning_rate": 0.0001816487912093755}
+{"ts": "2025-12-27T09:12:08", "event": "train_log", "step": 3704, "epoch": 1.5628691983122363, "progress_pct": 26.05, "epoch_pct": 26.05, "eta": "42:20:52", "max_grad_norm": 1.0, "loss": 0.696711003780365, "grad_norm": 0.4205699563026428, "learning_rate": 0.00018162105150901017}
+{"ts": "2025-12-27T09:12:22", "event": "train_log", "step": 3706, "epoch": 1.5637130801687764, "progress_pct": 26.06, "epoch_pct": 26.06, "eta": "42:19:41", "max_grad_norm": 1.0, "loss": 0.7130293250083923, "grad_norm": 0.46437183022499084, "learning_rate": 0.0001815932929802358}
+{"ts": "2025-12-27T09:12:36", "event": "train_log", "step": 3708, "epoch": 1.5645569620253164, "progress_pct": 26.08, "epoch_pct": 26.08, "eta": "42:18:30", "max_grad_norm": 1.0, "loss": 0.7180310487747192, "grad_norm": 0.4413023591041565, "learning_rate": 0.0001815655156294558}
+{"ts": "2025-12-27T09:12:51", "event": "train_log", "step": 3710, "epoch": 1.5654008438818565, "progress_pct": 26.09, "epoch_pct": 26.09, "eta": "42:17:22", "max_grad_norm": 1.0, "loss": 0.6883325576782227, "grad_norm": 0.3964478075504303, "learning_rate": 0.00018153771946307783}
+{"ts": "2025-12-27T09:13:08", "event": "train_log", "step": 3712, "epoch": 1.5662447257383967, "progress_pct": 26.1, "epoch_pct": 26.1, "eta": "42:16:18", "max_grad_norm": 1.0, "loss": 0.6683162450790405, "grad_norm": 0.34486109018325806, "learning_rate": 0.00018150990448751394}
+{"ts": "2025-12-27T09:13:23", "event": "train_log", "step": 3714, "epoch": 1.5670886075949366, "progress_pct": 26.12, "epoch_pct": 26.12, "eta": "42:15:09", "max_grad_norm": 1.0, "loss": 0.6969789862632751, "grad_norm": 0.4209035336971283, "learning_rate": 0.00018148207070918052}
+{"ts": "2025-12-27T09:13:38", "event": "train_log", "step": 3716, "epoch": 1.5679324894514768, "progress_pct": 26.13, "epoch_pct": 26.13, "eta": "42:14:00", "max_grad_norm": 1.0, "loss": 0.714320182800293, "grad_norm": 0.41674816608428955, "learning_rate": 0.00018145421813449829}
+{"ts": "2025-12-27T09:13:53", "event": "train_log", "step": 3718, "epoch": 1.568776371308017, "progress_pct": 26.15, "epoch_pct": 26.15, "eta": "42:12:53", "max_grad_norm": 1.0, "loss": 0.7125279307365417, "grad_norm": 0.4365934729576111, "learning_rate": 0.00018142634676989228}
+{"ts": "2025-12-27T09:14:08", "event": "train_log", "step": 3720, "epoch": 1.5696202531645569, "progress_pct": 26.16, "epoch_pct": 26.16, "eta": "42:11:44", "max_grad_norm": 1.0, "loss": 0.6787398457527161, "grad_norm": 0.43599721789360046, "learning_rate": 0.00018139845662179188}
+{"ts": "2025-12-27T09:14:24", "event": "train_log", "step": 3722, "epoch": 1.570464135021097, "progress_pct": 26.17, "epoch_pct": 26.17, "eta": "42:10:38", "max_grad_norm": 1.0, "loss": 0.6401328444480896, "grad_norm": 0.36320003867149353, "learning_rate": 0.0001813705476966308}
+{"ts": "2025-12-27T09:14:39", "event": "train_log", "step": 3724, "epoch": 1.5713080168776372, "progress_pct": 26.19, "epoch_pct": 26.19, "eta": "42:09:31", "max_grad_norm": 1.0, "loss": 0.6875129342079163, "grad_norm": 0.43504419922828674, "learning_rate": 0.00018134262000084718}
+{"ts": "2025-12-27T09:14:54", "event": "train_log", "step": 3726, "epoch": 1.5721518987341772, "progress_pct": 26.2, "epoch_pct": 26.2, "eta": "42:08:23", "max_grad_norm": 1.0, "loss": 0.6903918385505676, "grad_norm": 0.476723313331604, "learning_rate": 0.00018131467354088332}
+{"ts": "2025-12-27T09:15:09", "event": "train_log", "step": 3728, "epoch": 1.5729957805907173, "progress_pct": 26.22, "epoch_pct": 26.22, "eta": "42:07:15", "max_grad_norm": 1.0, "loss": 0.690018892288208, "grad_norm": 0.4465550184249878, "learning_rate": 0.00018128670832318592}
+{"ts": "2025-12-27T09:15:24", "event": "train_log", "step": 3730, "epoch": 1.5738396624472575, "progress_pct": 26.23, "epoch_pct": 26.23, "eta": "42:06:07", "max_grad_norm": 1.0, "loss": 0.6912158727645874, "grad_norm": 0.43816477060317993, "learning_rate": 0.0001812587243542061}
+{"ts": "2025-12-27T09:15:39", "event": "train_log", "step": 3732, "epoch": 1.5746835443037974, "progress_pct": 26.24, "epoch_pct": 26.24, "eta": "42:04:59", "max_grad_norm": 1.0, "loss": 0.7415492534637451, "grad_norm": 0.40546250343322754, "learning_rate": 0.00018123072164039913}
+{"ts": "2025-12-27T09:15:54", "event": "train_log", "step": 3734, "epoch": 1.5755274261603376, "progress_pct": 26.26, "epoch_pct": 26.26, "eta": "42:03:52", "max_grad_norm": 1.0, "loss": 0.7664727568626404, "grad_norm": 0.4253178536891937, "learning_rate": 0.00018120270018822477}
+{"ts": "2025-12-27T09:16:08", "event": "train_log", "step": 3736, "epoch": 1.5763713080168777, "progress_pct": 26.27, "epoch_pct": 26.27, "eta": "42:02:41", "max_grad_norm": 1.0, "loss": 0.782779335975647, "grad_norm": 0.503524661064148, "learning_rate": 0.000181174660004147}
+{"ts": "2025-12-27T09:16:22", "event": "train_log", "step": 3738, "epoch": 1.5772151898734177, "progress_pct": 26.29, "epoch_pct": 26.29, "eta": "42:01:29", "max_grad_norm": 1.0, "loss": 0.806460976600647, "grad_norm": 0.5260395407676697, "learning_rate": 0.00018114660109463414}
+{"ts": "2025-12-27T09:16:37", "event": "train_log", "step": 3740, "epoch": 1.5780590717299579, "progress_pct": 26.3, "epoch_pct": 26.3, "eta": "42:00:23", "max_grad_norm": 1.0, "loss": 0.6859704852104187, "grad_norm": 0.3641359806060791, "learning_rate": 0.0001811185234661589}
+{"ts": "2025-12-27T09:16:52", "event": "train_log", "step": 3742, "epoch": 1.578902953586498, "progress_pct": 26.32, "epoch_pct": 26.32, "eta": "41:59:14", "max_grad_norm": 1.0, "loss": 0.7798553109169006, "grad_norm": 0.4468916058540344, "learning_rate": 0.00018109042712519818}
+{"ts": "2025-12-27T09:17:07", "event": "train_log", "step": 3744, "epoch": 1.579746835443038, "progress_pct": 26.33, "epoch_pct": 26.33, "eta": "41:58:06", "max_grad_norm": 1.0, "loss": 0.7086471915245056, "grad_norm": 0.4078207015991211, "learning_rate": 0.00018106231207823327}
+{"ts": "2025-12-27T09:17:23", "event": "train_log", "step": 3746, "epoch": 1.580590717299578, "progress_pct": 26.34, "epoch_pct": 26.34, "eta": "41:57:03", "max_grad_norm": 1.0, "loss": 0.6669760346412659, "grad_norm": 0.39144742488861084, "learning_rate": 0.00018103417833174981}
+{"ts": "2025-12-27T09:17:38", "event": "train_log", "step": 3748, "epoch": 1.5814345991561183, "progress_pct": 26.36, "epoch_pct": 26.36, "eta": "41:55:54", "max_grad_norm": 1.0, "loss": 0.7296954989433289, "grad_norm": 0.4366845190525055, "learning_rate": 0.0001810060258922377}
+{"ts": "2025-12-27T09:17:53", "event": "train_log", "step": 3750, "epoch": 1.5822784810126582, "progress_pct": 26.37, "epoch_pct": 26.37, "eta": "41:54:46", "max_grad_norm": 1.0, "loss": 0.7663856148719788, "grad_norm": 0.46795082092285156, "learning_rate": 0.00018097785476619118}
+{"ts": "2025-12-27T09:18:08", "event": "train_log", "step": 3752, "epoch": 1.5831223628691982, "progress_pct": 26.39, "epoch_pct": 26.39, "eta": "41:53:39", "max_grad_norm": 1.0, "loss": 0.7105234861373901, "grad_norm": 0.40331873297691345, "learning_rate": 0.00018094966496010878}
+{"ts": "2025-12-27T09:18:24", "event": "train_log", "step": 3754, "epoch": 1.5839662447257385, "progress_pct": 26.4, "epoch_pct": 26.4, "eta": "41:52:34", "max_grad_norm": 1.0, "loss": 0.666635274887085, "grad_norm": 0.38210469484329224, "learning_rate": 0.00018092145648049334}
+{"ts": "2025-12-27T09:18:38", "event": "train_log", "step": 3756, "epoch": 1.5848101265822785, "progress_pct": 26.41, "epoch_pct": 26.41, "eta": "41:51:24", "max_grad_norm": 1.0, "loss": 0.7287175059318542, "grad_norm": 0.45946040749549866, "learning_rate": 0.000180893229333852}
+{"ts": "2025-12-27T09:18:53", "event": "train_log", "step": 3758, "epoch": 1.5856540084388184, "progress_pct": 26.43, "epoch_pct": 26.43, "eta": "41:50:18", "max_grad_norm": 1.0, "loss": 0.7145618796348572, "grad_norm": 0.4035777747631073, "learning_rate": 0.00018086498352669624}
+{"ts": "2025-12-27T09:19:07", "event": "train_log", "step": 3760, "epoch": 1.5864978902953588, "progress_pct": 26.44, "epoch_pct": 26.44, "eta": "41:49:08", "max_grad_norm": 1.0, "loss": 0.6550979018211365, "grad_norm": 0.42843320965766907, "learning_rate": 0.0001808367190655418}
+{"ts": "2025-12-27T09:19:23", "event": "train_log", "step": 3762, "epoch": 1.5873417721518988, "progress_pct": 26.46, "epoch_pct": 26.46, "eta": "41:48:03", "max_grad_norm": 1.0, "loss": 0.7040430307388306, "grad_norm": 0.4016468822956085, "learning_rate": 0.0001808084359569088}
+{"ts": "2025-12-27T09:19:37", "event": "train_log", "step": 3764, "epoch": 1.5881856540084387, "progress_pct": 26.47, "epoch_pct": 26.47, "eta": "41:46:52", "max_grad_norm": 1.0, "loss": 0.7523179054260254, "grad_norm": 0.4126848578453064, "learning_rate": 0.00018078013420732157}
+{"ts": "2025-12-27T09:19:51", "event": "train_log", "step": 3766, "epoch": 1.5890295358649789, "progress_pct": 26.48, "epoch_pct": 26.48, "eta": "41:45:45", "max_grad_norm": 1.0, "loss": 0.7497967481613159, "grad_norm": 0.45878592133522034, "learning_rate": 0.00018075181382330878}
+{"ts": "2025-12-27T09:20:06", "event": "train_log", "step": 3768, "epoch": 1.589873417721519, "progress_pct": 26.5, "epoch_pct": 26.5, "eta": "41:44:37", "max_grad_norm": 1.0, "loss": 0.7694459557533264, "grad_norm": 0.44179272651672363, "learning_rate": 0.00018072347481140345}
+{"ts": "2025-12-27T09:20:21", "event": "train_log", "step": 3770, "epoch": 1.590717299578059, "progress_pct": 26.51, "epoch_pct": 26.51, "eta": "41:43:31", "max_grad_norm": 1.0, "loss": 0.6766634583473206, "grad_norm": 0.3706412613391876, "learning_rate": 0.00018069511717814276}
+{"ts": "2025-12-27T09:20:36", "event": "train_log", "step": 3772, "epoch": 1.5915611814345991, "progress_pct": 26.53, "epoch_pct": 26.53, "eta": "41:42:24", "max_grad_norm": 1.0, "loss": 0.6394028067588806, "grad_norm": 0.40555453300476074, "learning_rate": 0.0001806667409300683}
+{"ts": "2025-12-27T09:20:52", "event": "train_log", "step": 3774, "epoch": 1.5924050632911393, "progress_pct": 26.54, "epoch_pct": 26.54, "eta": "41:41:19", "max_grad_norm": 1.0, "loss": 0.6775637865066528, "grad_norm": 0.3752219080924988, "learning_rate": 0.000180638346073726}
+{"ts": "2025-12-27T09:21:08", "event": "train_log", "step": 3776, "epoch": 1.5932489451476792, "progress_pct": 26.55, "epoch_pct": 26.55, "eta": "41:40:16", "max_grad_norm": 1.0, "loss": 0.7200632095336914, "grad_norm": 0.40960127115249634, "learning_rate": 0.00018060993261566593}
+{"ts": "2025-12-27T09:21:23", "event": "train_log", "step": 3778, "epoch": 1.5940928270042194, "progress_pct": 26.57, "epoch_pct": 26.57, "eta": "41:39:09", "max_grad_norm": 1.0, "loss": 0.7314276695251465, "grad_norm": 0.4215535521507263, "learning_rate": 0.00018058150056244254}
+{"ts": "2025-12-27T09:21:38", "event": "train_log", "step": 3780, "epoch": 1.5949367088607596, "progress_pct": 26.58, "epoch_pct": 26.58, "eta": "41:38:02", "max_grad_norm": 1.0, "loss": 0.7954897284507751, "grad_norm": 0.42734605073928833, "learning_rate": 0.00018055304992061456}
+{"ts": "2025-12-27T09:21:54", "event": "train_log", "step": 3782, "epoch": 1.5957805907172995, "progress_pct": 26.6, "epoch_pct": 26.6, "eta": "41:36:57", "max_grad_norm": 1.0, "loss": 0.6552245020866394, "grad_norm": 0.3980397582054138, "learning_rate": 0.00018052458069674505}
+{"ts": "2025-12-27T09:22:07", "event": "train_log", "step": 3784, "epoch": 1.5966244725738397, "progress_pct": 26.61, "epoch_pct": 26.61, "eta": "41:35:48", "max_grad_norm": 1.0, "loss": 0.7638092041015625, "grad_norm": 0.44602933526039124, "learning_rate": 0.00018049609289740124}
+{"ts": "2025-12-27T09:22:23", "event": "train_log", "step": 3786, "epoch": 1.5974683544303798, "progress_pct": 26.62, "epoch_pct": 26.62, "eta": "41:34:43", "max_grad_norm": 1.0, "loss": 0.6886481046676636, "grad_norm": 0.43835216760635376, "learning_rate": 0.00018046758652915474}
+{"ts": "2025-12-27T09:22:39", "event": "train_log", "step": 3788, "epoch": 1.5983122362869198, "progress_pct": 26.64, "epoch_pct": 26.64, "eta": "41:33:39", "max_grad_norm": 1.0, "loss": 0.6726123690605164, "grad_norm": 0.38229402899742126, "learning_rate": 0.00018043906159858147}
+{"ts": "2025-12-27T09:22:54", "event": "train_log", "step": 3790, "epoch": 1.59915611814346, "progress_pct": 26.65, "epoch_pct": 26.65, "eta": "41:32:32", "max_grad_norm": 1.0, "loss": 0.7416896224021912, "grad_norm": 0.43049392104148865, "learning_rate": 0.00018041051811226155}
+{"ts": "2025-12-27T09:23:10", "event": "train_log", "step": 3792, "epoch": 1.6, "progress_pct": 26.67, "epoch_pct": 26.67, "eta": "41:31:28", "max_grad_norm": 1.0, "loss": 0.682691752910614, "grad_norm": 0.4027816951274872, "learning_rate": 0.00018038195607677943}
+{"ts": "2025-12-27T09:23:25", "event": "train_log", "step": 3794, "epoch": 1.60084388185654, "progress_pct": 26.68, "epoch_pct": 26.68, "eta": "41:30:21", "max_grad_norm": 1.0, "loss": 0.7199558615684509, "grad_norm": 0.4143599569797516, "learning_rate": 0.00018035337549872376}
+{"ts": "2025-12-27T09:23:40", "event": "train_log", "step": 3796, "epoch": 1.6016877637130802, "progress_pct": 26.69, "epoch_pct": 26.69, "eta": "41:29:16", "max_grad_norm": 1.0, "loss": 0.6586267948150635, "grad_norm": 0.41221415996551514, "learning_rate": 0.00018032477638468762}
+{"ts": "2025-12-27T09:23:56", "event": "train_log", "step": 3798, "epoch": 1.6025316455696204, "progress_pct": 26.71, "epoch_pct": 26.71, "eta": "41:28:15", "max_grad_norm": 1.0, "loss": 0.7579189538955688, "grad_norm": 0.40655994415283203, "learning_rate": 0.0001802961587412682}
+{"ts": "2025-12-27T09:24:12", "event": "train_log", "step": 3800, "epoch": 1.6033755274261603, "progress_pct": 26.72, "epoch_pct": 26.72, "eta": "41:27:09", "max_grad_norm": 1.0, "loss": 0.7246071100234985, "grad_norm": 0.4163990914821625, "learning_rate": 0.00018026752257506708}
+{"ts": "2025-12-27T09:35:44", "event": "train_log", "step": 3800, "epoch": 1.6033755274261603, "progress_pct": 26.72, "epoch_pct": 26.72, "eta": "41:58:48", "max_grad_norm": 1.0, "eval_loss": 0.778353214263916, "eval_runtime": 692.5522, "eval_samples_per_second": 3.042, "eval_steps_per_second": 3.042}
+{"ts": "2025-12-27T09:36:01", "event": "train_log", "step": 3802, "epoch": 1.6042194092827005, "progress_pct": 26.74, "epoch_pct": 26.74, "eta": "41:57:47", "max_grad_norm": 1.0, "loss": 0.7114223837852478, "grad_norm": 0.378628671169281, "learning_rate": 0.0001802388678926901}
+{"ts": "2025-12-27T09:36:16", "event": "train_log", "step": 3804, "epoch": 1.6050632911392406, "progress_pct": 26.75, "epoch_pct": 26.75, "eta": "41:56:37", "max_grad_norm": 1.0, "loss": 0.7519527077674866, "grad_norm": 0.5050750374794006, "learning_rate": 0.0001802101947007473}
+{"ts": "2025-12-27T09:36:30", "event": "train_log", "step": 3806, "epoch": 1.6059071729957806, "progress_pct": 26.77, "epoch_pct": 26.77, "eta": "41:55:27", "max_grad_norm": 1.0, "loss": 0.7553771138191223, "grad_norm": 0.5003950595855713, "learning_rate": 0.00018018150300585305}
+{"ts": "2025-12-27T09:36:45", "event": "train_log", "step": 3808, "epoch": 1.6067510548523207, "progress_pct": 26.78, "epoch_pct": 26.78, "eta": "41:54:20", "max_grad_norm": 1.0, "loss": 0.6584326028823853, "grad_norm": 0.3851728141307831, "learning_rate": 0.00018015279281462595}
+{"ts": "2025-12-27T09:37:01", "event": "train_log", "step": 3810, "epoch": 1.6075949367088609, "progress_pct": 26.79, "epoch_pct": 26.79, "eta": "41:53:17", "max_grad_norm": 1.0, "loss": 0.7215287089347839, "grad_norm": 0.41256240010261536, "learning_rate": 0.00018012406413368893}
+{"ts": "2025-12-27T09:37:17", "event": "train_log", "step": 3812, "epoch": 1.6084388185654008, "progress_pct": 26.81, "epoch_pct": 26.81, "eta": "41:52:13", "max_grad_norm": 1.0, "loss": 0.7753872275352478, "grad_norm": 0.41479718685150146, "learning_rate": 0.00018009531696966912}
+{"ts": "2025-12-27T09:37:32", "event": "train_log", "step": 3814, "epoch": 1.6092827004219408, "progress_pct": 26.82, "epoch_pct": 26.82, "eta": "41:51:04", "max_grad_norm": 1.0, "loss": 0.6984832882881165, "grad_norm": 0.398509681224823, "learning_rate": 0.00018006655132919797}
+{"ts": "2025-12-27T09:37:47", "event": "train_log", "step": 3816, "epoch": 1.6101265822784812, "progress_pct": 26.84, "epoch_pct": 26.84, "eta": "41:49:57", "max_grad_norm": 1.0, "loss": 0.8117644786834717, "grad_norm": 0.42210307717323303, "learning_rate": 0.0001800377672189111}
+{"ts": "2025-12-27T09:38:03", "event": "train_log", "step": 3818, "epoch": 1.610970464135021, "progress_pct": 26.85, "epoch_pct": 26.85, "eta": "41:48:54", "max_grad_norm": 1.0, "loss": 0.6302977204322815, "grad_norm": 0.3714556396007538, "learning_rate": 0.0001800089646454485}
+{"ts": "2025-12-27T09:38:18", "event": "train_log", "step": 3820, "epoch": 1.611814345991561, "progress_pct": 26.86, "epoch_pct": 26.86, "eta": "41:47:48", "max_grad_norm": 1.0, "loss": 0.7327325940132141, "grad_norm": 0.4478890299797058, "learning_rate": 0.00017998014361545438}
+{"ts": "2025-12-27T09:38:34", "event": "train_log", "step": 3822, "epoch": 1.6126582278481014, "progress_pct": 26.88, "epoch_pct": 26.88, "eta": "41:46:43", "max_grad_norm": 1.0, "loss": 0.6877713203430176, "grad_norm": 0.43516477942466736, "learning_rate": 0.00017995130413557713}
+{"ts": "2025-12-27T09:38:50", "event": "train_log", "step": 3824, "epoch": 1.6135021097046414, "progress_pct": 26.89, "epoch_pct": 26.89, "eta": "41:45:39", "max_grad_norm": 1.0, "loss": 0.6797093152999878, "grad_norm": 0.42912542819976807, "learning_rate": 0.0001799224462124695}
+{"ts": "2025-12-27T09:39:06", "event": "train_log", "step": 3826, "epoch": 1.6143459915611813, "progress_pct": 26.91, "epoch_pct": 26.91, "eta": "41:44:36", "max_grad_norm": 1.0, "loss": 0.7402470111846924, "grad_norm": 0.40646228194236755, "learning_rate": 0.00017989356985278853}
+{"ts": "2025-12-27T09:39:22", "event": "train_log", "step": 3828, "epoch": 1.6151898734177215, "progress_pct": 26.92, "epoch_pct": 26.92, "eta": "41:43:31", "max_grad_norm": 1.0, "loss": 0.6698583364486694, "grad_norm": 0.42576655745506287, "learning_rate": 0.00017986467506319535}
+{"ts": "2025-12-27T09:39:37", "event": "train_log", "step": 3830, "epoch": 1.6160337552742616, "progress_pct": 26.93, "epoch_pct": 26.93, "eta": "41:42:25", "max_grad_norm": 1.0, "loss": 0.6736086010932922, "grad_norm": 0.4298654794692993, "learning_rate": 0.0001798357618503555}
+{"ts": "2025-12-27T09:39:52", "event": "train_log", "step": 3832, "epoch": 1.6168776371308016, "progress_pct": 26.95, "epoch_pct": 26.95, "eta": "41:41:18", "max_grad_norm": 1.0, "loss": 0.7724249958992004, "grad_norm": 0.4205976128578186, "learning_rate": 0.0001798068302209386}
+{"ts": "2025-12-27T09:40:07", "event": "train_log", "step": 3834, "epoch": 1.6177215189873417, "progress_pct": 26.96, "epoch_pct": 26.96, "eta": "41:40:11", "max_grad_norm": 1.0, "loss": 0.6468521952629089, "grad_norm": 0.39669108390808105, "learning_rate": 0.00017977788018161872}
+{"ts": "2025-12-27T09:40:21", "event": "train_log", "step": 3836, "epoch": 1.618565400843882, "progress_pct": 26.98, "epoch_pct": 26.98, "eta": "41:39:02", "max_grad_norm": 1.0, "loss": 0.7086918950080872, "grad_norm": 0.428551584482193, "learning_rate": 0.00017974891173907405}
+{"ts": "2025-12-27T09:40:36", "event": "train_log", "step": 3838, "epoch": 1.6194092827004218, "progress_pct": 26.99, "epoch_pct": 26.99, "eta": "41:37:56", "max_grad_norm": 1.0, "loss": 0.6483711004257202, "grad_norm": 0.43773409724235535, "learning_rate": 0.00017971992489998703}
+{"ts": "2025-12-27T09:40:52", "event": "train_log", "step": 3840, "epoch": 1.620253164556962, "progress_pct": 27.0, "epoch_pct": 27.0, "eta": "41:36:51", "max_grad_norm": 1.0, "loss": 0.6947301030158997, "grad_norm": 0.40540891885757446, "learning_rate": 0.00017969091967104438}
+{"ts": "2025-12-27T09:41:08", "event": "train_log", "step": 3842, "epoch": 1.6210970464135022, "progress_pct": 27.02, "epoch_pct": 27.02, "eta": "41:35:48", "max_grad_norm": 1.0, "loss": 0.667517900466919, "grad_norm": 0.3685019016265869, "learning_rate": 0.00017966189605893708}
+{"ts": "2025-12-27T09:41:23", "event": "train_log", "step": 3844, "epoch": 1.621940928270042, "progress_pct": 27.03, "epoch_pct": 27.03, "eta": "41:34:41", "max_grad_norm": 1.0, "loss": 0.6957448124885559, "grad_norm": 0.44813284277915955, "learning_rate": 0.0001796328540703603}
+{"ts": "2025-12-27T09:41:38", "event": "train_log", "step": 3846, "epoch": 1.6227848101265823, "progress_pct": 27.05, "epoch_pct": 27.05, "eta": "41:33:34", "max_grad_norm": 1.0, "loss": 0.7735851407051086, "grad_norm": 0.4524349868297577, "learning_rate": 0.00017960379371201345}
+{"ts": "2025-12-27T09:41:54", "event": "train_log", "step": 3848, "epoch": 1.6236286919831224, "progress_pct": 27.06, "epoch_pct": 27.06, "eta": "41:32:31", "max_grad_norm": 1.0, "loss": 0.7052056193351746, "grad_norm": 0.42349138855934143, "learning_rate": 0.0001795747149906002}
+{"ts": "2025-12-27T09:42:08", "event": "train_log", "step": 3850, "epoch": 1.6244725738396624, "progress_pct": 27.07, "epoch_pct": 27.07, "eta": "41:31:23", "max_grad_norm": 1.0, "loss": 0.7402311563491821, "grad_norm": 0.47317537665367126, "learning_rate": 0.0001795456179128285}
+{"ts": "2025-12-27T09:42:23", "event": "train_log", "step": 3852, "epoch": 1.6253164556962025, "progress_pct": 27.09, "epoch_pct": 27.09, "eta": "41:30:18", "max_grad_norm": 1.0, "loss": 0.6743243932723999, "grad_norm": 0.38915711641311646, "learning_rate": 0.0001795165024854104}
+{"ts": "2025-12-27T09:42:38", "event": "train_log", "step": 3854, "epoch": 1.6261603375527427, "progress_pct": 27.1, "epoch_pct": 27.1, "eta": "41:29:11", "max_grad_norm": 1.0, "loss": 0.6983860731124878, "grad_norm": 0.4430672526359558, "learning_rate": 0.00017948736871506235}
+{"ts": "2025-12-27T09:42:53", "event": "train_log", "step": 3856, "epoch": 1.6270042194092826, "progress_pct": 27.12, "epoch_pct": 27.12, "eta": "41:28:04", "max_grad_norm": 1.0, "loss": 0.8012514710426331, "grad_norm": 0.44286689162254333, "learning_rate": 0.00017945821660850494}
+{"ts": "2025-12-27T09:43:07", "event": "train_log", "step": 3858, "epoch": 1.6278481012658228, "progress_pct": 27.13, "epoch_pct": 27.13, "eta": "41:26:57", "max_grad_norm": 1.0, "loss": 0.8064966201782227, "grad_norm": 0.43364086747169495, "learning_rate": 0.00017942904617246295}
+{"ts": "2025-12-27T09:43:22", "event": "train_log", "step": 3860, "epoch": 1.628691983122363, "progress_pct": 27.14, "epoch_pct": 27.14, "eta": "41:25:52", "max_grad_norm": 1.0, "loss": 0.7069969177246094, "grad_norm": 0.45848122239112854, "learning_rate": 0.00017939985741366547}
+{"ts": "2025-12-27T09:43:38", "event": "train_log", "step": 3862, "epoch": 1.629535864978903, "progress_pct": 27.16, "epoch_pct": 27.16, "eta": "41:24:47", "max_grad_norm": 1.0, "loss": 0.6722052097320557, "grad_norm": 0.4199717044830322, "learning_rate": 0.0001793706503388458}
+{"ts": "2025-12-27T09:43:53", "event": "train_log", "step": 3864, "epoch": 1.630379746835443, "progress_pct": 27.17, "epoch_pct": 27.17, "eta": "41:23:42", "max_grad_norm": 1.0, "loss": 0.663442075252533, "grad_norm": 0.3820365071296692, "learning_rate": 0.00017934142495474142}
+{"ts": "2025-12-27T09:44:09", "event": "train_log", "step": 3866, "epoch": 1.6312236286919832, "progress_pct": 27.19, "epoch_pct": 27.19, "eta": "41:22:39", "max_grad_norm": 1.0, "loss": 0.70677250623703, "grad_norm": 0.4470944106578827, "learning_rate": 0.00017931218126809412}
+{"ts": "2025-12-27T09:44:24", "event": "train_log", "step": 3868, "epoch": 1.6320675105485232, "progress_pct": 27.2, "epoch_pct": 27.2, "eta": "41:21:33", "max_grad_norm": 1.0, "loss": 0.7125341892242432, "grad_norm": 0.3941600024700165, "learning_rate": 0.0001792829192856498}
+{"ts": "2025-12-27T09:44:39", "event": "train_log", "step": 3870, "epoch": 1.6329113924050633, "progress_pct": 27.22, "epoch_pct": 27.22, "eta": "41:20:27", "max_grad_norm": 1.0, "loss": 0.7245699763298035, "grad_norm": 0.44924396276474, "learning_rate": 0.00017925363901415863}
+{"ts": "2025-12-27T09:44:54", "event": "train_log", "step": 3872, "epoch": 1.6337552742616035, "progress_pct": 27.23, "epoch_pct": 27.23, "eta": "41:19:23", "max_grad_norm": 1.0, "loss": 0.6977120637893677, "grad_norm": 0.4070027470588684, "learning_rate": 0.00017922434046037508}
+{"ts": "2025-12-27T09:45:09", "event": "train_log", "step": 3874, "epoch": 1.6345991561181434, "progress_pct": 27.24, "epoch_pct": 27.24, "eta": "41:18:16", "max_grad_norm": 1.0, "loss": 0.7631482481956482, "grad_norm": 0.45040908455848694, "learning_rate": 0.00017919502363105772}
+{"ts": "2025-12-27T09:45:25", "event": "train_log", "step": 3876, "epoch": 1.6354430379746834, "progress_pct": 27.26, "epoch_pct": 27.26, "eta": "41:17:13", "max_grad_norm": 1.0, "loss": 0.6656452417373657, "grad_norm": 0.4083724617958069, "learning_rate": 0.00017916568853296934}
+{"ts": "2025-12-27T09:45:42", "event": "train_log", "step": 3878, "epoch": 1.6362869198312238, "progress_pct": 27.27, "epoch_pct": 27.27, "eta": "41:16:13", "max_grad_norm": 1.0, "loss": 0.6302213072776794, "grad_norm": 0.3762308359146118, "learning_rate": 0.00017913633517287704}
+{"ts": "2025-12-27T09:45:57", "event": "train_log", "step": 3880, "epoch": 1.6371308016877637, "progress_pct": 27.29, "epoch_pct": 27.29, "eta": "41:15:10", "max_grad_norm": 1.0, "loss": 0.7119566202163696, "grad_norm": 0.36366912722587585, "learning_rate": 0.00017910696355755205}
+{"ts": "2025-12-27T09:46:12", "event": "train_log", "step": 3882, "epoch": 1.6379746835443036, "progress_pct": 27.3, "epoch_pct": 27.3, "eta": "41:14:04", "max_grad_norm": 1.0, "loss": 0.7271678447723389, "grad_norm": 0.43168532848358154, "learning_rate": 0.00017907757369376985}
+{"ts": "2025-12-27T09:46:27", "event": "train_log", "step": 3884, "epoch": 1.638818565400844, "progress_pct": 27.31, "epoch_pct": 27.31, "eta": "41:12:59", "max_grad_norm": 1.0, "loss": 0.6340095400810242, "grad_norm": 0.37720009684562683, "learning_rate": 0.00017904816558831014}
+{"ts": "2025-12-27T09:46:43", "event": "train_log", "step": 3886, "epoch": 1.639662447257384, "progress_pct": 27.33, "epoch_pct": 27.33, "eta": "41:11:55", "max_grad_norm": 1.0, "loss": 0.6910572052001953, "grad_norm": 0.36539995670318604, "learning_rate": 0.00017901873924795677}
+{"ts": "2025-12-27T09:46:57", "event": "train_log", "step": 3888, "epoch": 1.640506329113924, "progress_pct": 27.34, "epoch_pct": 27.34, "eta": "41:10:48", "max_grad_norm": 1.0, "loss": 0.7570282220840454, "grad_norm": 0.42437905073165894, "learning_rate": 0.00017898929467949782}
+{"ts": "2025-12-27T09:47:11", "event": "train_log", "step": 3890, "epoch": 1.6413502109704643, "progress_pct": 27.36, "epoch_pct": 27.36, "eta": "41:09:41", "max_grad_norm": 1.0, "loss": 0.7835264205932617, "grad_norm": 0.43482181429862976, "learning_rate": 0.00017895983188972558}
+{"ts": "2025-12-27T09:47:26", "event": "train_log", "step": 3892, "epoch": 1.6421940928270042, "progress_pct": 27.37, "epoch_pct": 27.37, "eta": "41:08:36", "max_grad_norm": 1.0, "loss": 0.7058968544006348, "grad_norm": 0.40460237860679626, "learning_rate": 0.00017893035088543662}
+{"ts": "2025-12-27T09:47:42", "event": "train_log", "step": 3894, "epoch": 1.6430379746835442, "progress_pct": 27.38, "epoch_pct": 27.38, "eta": "41:07:34", "max_grad_norm": 1.0, "loss": 0.723969042301178, "grad_norm": 0.39779701828956604, "learning_rate": 0.00017890085167343158}
+{"ts": "2025-12-27T09:47:58", "event": "train_log", "step": 3896, "epoch": 1.6438818565400843, "progress_pct": 27.4, "epoch_pct": 27.4, "eta": "41:06:30", "max_grad_norm": 1.0, "loss": 0.7773922681808472, "grad_norm": 0.46131494641304016, "learning_rate": 0.00017887133426051536}
+{"ts": "2025-12-27T09:48:12", "event": "train_log", "step": 3898, "epoch": 1.6447257383966245, "progress_pct": 27.41, "epoch_pct": 27.41, "eta": "41:05:23", "max_grad_norm": 1.0, "loss": 0.7742162942886353, "grad_norm": 0.48412877321243286, "learning_rate": 0.0001788417986534971}
+{"ts": "2025-12-27T09:48:27", "event": "train_log", "step": 3900, "epoch": 1.6455696202531644, "progress_pct": 27.43, "epoch_pct": 27.43, "eta": "41:04:18", "max_grad_norm": 1.0, "loss": 0.6674869656562805, "grad_norm": 0.40083935856819153, "learning_rate": 0.00017881224485919006}
+{"ts": "2025-12-27T09:59:49", "event": "train_log", "step": 3900, "epoch": 1.6455696202531644, "progress_pct": 27.43, "epoch_pct": 27.43, "eta": "41:34:23", "max_grad_norm": 1.0, "eval_loss": 0.7763351202011108, "eval_runtime": 682.0824, "eval_samples_per_second": 3.089, "eval_steps_per_second": 3.089}
+{"ts": "2025-12-27T10:00:05", "event": "train_log", "step": 3902, "epoch": 1.6464135021097046, "progress_pct": 27.44, "epoch_pct": 27.44, "eta": "41:33:19", "max_grad_norm": 1.0, "loss": 0.6777821183204651, "grad_norm": 0.3944476842880249, "learning_rate": 0.00017878267288441176}
+{"ts": "2025-12-27T10:00:21", "event": "train_log", "step": 3904, "epoch": 1.6472573839662448, "progress_pct": 27.45, "epoch_pct": 27.45, "eta": "41:32:15", "max_grad_norm": 1.0, "loss": 0.7398322224617004, "grad_norm": 0.44863224029541016, "learning_rate": 0.0001787530827359839}
+{"ts": "2025-12-27T10:00:36", "event": "train_log", "step": 3906, "epoch": 1.6481012658227847, "progress_pct": 27.47, "epoch_pct": 27.47, "eta": "41:31:10", "max_grad_norm": 1.0, "loss": 0.7203068733215332, "grad_norm": 0.4665529131889343, "learning_rate": 0.00017872347442073232}
+{"ts": "2025-12-27T10:00:52", "event": "train_log", "step": 3908, "epoch": 1.6489451476793249, "progress_pct": 27.48, "epoch_pct": 27.48, "eta": "41:30:07", "max_grad_norm": 1.0, "loss": 0.6232115030288696, "grad_norm": 0.3603050410747528, "learning_rate": 0.00017869384794548708}
+{"ts": "2025-12-27T10:01:08", "event": "train_log", "step": 3910, "epoch": 1.649789029535865, "progress_pct": 27.5, "epoch_pct": 27.5, "eta": "41:29:04", "max_grad_norm": 1.0, "loss": 0.6120040416717529, "grad_norm": 0.3545325696468353, "learning_rate": 0.0001786642033170825}
+{"ts": "2025-12-27T10:01:23", "event": "train_log", "step": 3912, "epoch": 1.650632911392405, "progress_pct": 27.51, "epoch_pct": 27.51, "eta": "41:27:58", "max_grad_norm": 1.0, "loss": 0.7285997867584229, "grad_norm": 0.3947487473487854, "learning_rate": 0.00017863454054235697}
+{"ts": "2025-12-27T10:01:41", "event": "train_log", "step": 3914, "epoch": 1.6514767932489451, "progress_pct": 27.52, "epoch_pct": 27.52, "eta": "41:27:01", "max_grad_norm": 1.0, "loss": 0.6486109495162964, "grad_norm": 0.33153054118156433, "learning_rate": 0.00017860485962815312}
+{"ts": "2025-12-27T10:01:57", "event": "train_log", "step": 3916, "epoch": 1.6523206751054853, "progress_pct": 27.54, "epoch_pct": 27.54, "eta": "41:25:56", "max_grad_norm": 1.0, "loss": 0.6870376467704773, "grad_norm": 0.3886185884475708, "learning_rate": 0.00017857516058131782}
+{"ts": "2025-12-27T10:02:12", "event": "train_log", "step": 3918, "epoch": 1.6531645569620252, "progress_pct": 27.55, "epoch_pct": 27.55, "eta": "41:24:52", "max_grad_norm": 1.0, "loss": 0.7351298928260803, "grad_norm": 0.46399059891700745, "learning_rate": 0.00017854544340870201}
+{"ts": "2025-12-27T10:02:27", "event": "train_log", "step": 3920, "epoch": 1.6540084388185654, "progress_pct": 27.57, "epoch_pct": 27.57, "eta": "41:23:47", "max_grad_norm": 1.0, "loss": 0.7283915281295776, "grad_norm": 0.49280229210853577, "learning_rate": 0.00017851570811716093}
+{"ts": "2025-12-27T10:02:42", "event": "train_log", "step": 3922, "epoch": 1.6548523206751056, "progress_pct": 27.58, "epoch_pct": 27.58, "eta": "41:22:41", "max_grad_norm": 1.0, "loss": 0.7436906099319458, "grad_norm": 0.41570642590522766, "learning_rate": 0.00017848595471355388}
+{"ts": "2025-12-27T10:02:57", "event": "train_log", "step": 3924, "epoch": 1.6556962025316455, "progress_pct": 27.59, "epoch_pct": 27.59, "eta": "41:21:35", "max_grad_norm": 1.0, "loss": 0.7427505254745483, "grad_norm": 0.44591766595840454, "learning_rate": 0.00017845618320474444}
+{"ts": "2025-12-27T10:03:13", "event": "train_log", "step": 3926, "epoch": 1.6565400843881857, "progress_pct": 27.61, "epoch_pct": 27.61, "eta": "41:20:32", "max_grad_norm": 1.0, "loss": 0.7642725706100464, "grad_norm": 0.4804318845272064, "learning_rate": 0.00017842639359760032}
+{"ts": "2025-12-27T10:03:29", "event": "train_log", "step": 3928, "epoch": 1.6573839662447258, "progress_pct": 27.62, "epoch_pct": 27.62, "eta": "41:19:29", "max_grad_norm": 1.0, "loss": 0.7097790837287903, "grad_norm": 0.4164349138736725, "learning_rate": 0.00017839658589899343}
+{"ts": "2025-12-27T10:03:43", "event": "train_log", "step": 3930, "epoch": 1.6582278481012658, "progress_pct": 27.64, "epoch_pct": 27.64, "eta": "41:18:22", "max_grad_norm": 1.0, "loss": 0.8346033096313477, "grad_norm": 0.5307883620262146, "learning_rate": 0.00017836676011579976}
+{"ts": "2025-12-27T10:03:59", "event": "train_log", "step": 3932, "epoch": 1.659071729957806, "progress_pct": 27.65, "epoch_pct": 27.65, "eta": "41:17:20", "max_grad_norm": 1.0, "loss": 0.6327160596847534, "grad_norm": 0.37231504917144775, "learning_rate": 0.0001783369162548996}
+{"ts": "2025-12-27T10:04:14", "event": "train_log", "step": 3934, "epoch": 1.659915611814346, "progress_pct": 27.67, "epoch_pct": 27.67, "eta": "41:16:14", "max_grad_norm": 1.0, "loss": 0.6441237330436707, "grad_norm": 0.4178757965564728, "learning_rate": 0.00017830705432317738}
+{"ts": "2025-12-27T10:04:30", "event": "train_log", "step": 3936, "epoch": 1.660759493670886, "progress_pct": 27.68, "epoch_pct": 27.68, "eta": "41:15:12", "max_grad_norm": 1.0, "loss": 0.7267951369285583, "grad_norm": 0.42698073387145996, "learning_rate": 0.00017827717432752162}
+{"ts": "2025-12-27T10:04:46", "event": "train_log", "step": 3938, "epoch": 1.6616033755274262, "progress_pct": 27.69, "epoch_pct": 27.69, "eta": "41:14:09", "max_grad_norm": 1.0, "loss": 0.7248380780220032, "grad_norm": 0.48763108253479004, "learning_rate": 0.0001782472762748251}
+{"ts": "2025-12-27T10:05:03", "event": "train_log", "step": 3940, "epoch": 1.6624472573839664, "progress_pct": 27.71, "epoch_pct": 27.71, "eta": "41:13:07", "max_grad_norm": 1.0, "loss": 0.6665275692939758, "grad_norm": 0.38882505893707275, "learning_rate": 0.0001782173601719847}
+{"ts": "2025-12-27T10:05:18", "event": "train_log", "step": 3942, "epoch": 1.6632911392405063, "progress_pct": 27.72, "epoch_pct": 27.72, "eta": "41:12:03", "max_grad_norm": 1.0, "loss": 0.8017846941947937, "grad_norm": 0.4684160649776459, "learning_rate": 0.0001781874260259015}
+{"ts": "2025-12-27T10:05:33", "event": "train_log", "step": 3944, "epoch": 1.6641350210970463, "progress_pct": 27.74, "epoch_pct": 27.74, "eta": "41:10:59", "max_grad_norm": 1.0, "loss": 0.6810902953147888, "grad_norm": 0.4200204908847809, "learning_rate": 0.00017815747384348075}
+{"ts": "2025-12-27T10:05:48", "event": "train_log", "step": 3946, "epoch": 1.6649789029535866, "progress_pct": 27.75, "epoch_pct": 27.75, "eta": "41:09:55", "max_grad_norm": 1.0, "loss": 0.714082658290863, "grad_norm": 0.45843279361724854, "learning_rate": 0.0001781275036316318}
+{"ts": "2025-12-27T10:06:03", "event": "train_log", "step": 3948, "epoch": 1.6658227848101266, "progress_pct": 27.76, "epoch_pct": 27.76, "eta": "41:08:50", "max_grad_norm": 1.0, "loss": 0.842085599899292, "grad_norm": 0.522654116153717, "learning_rate": 0.00017809751539726824}
+{"ts": "2025-12-27T10:06:19", "event": "train_log", "step": 3950, "epoch": 1.6666666666666665, "progress_pct": 27.78, "epoch_pct": 27.78, "eta": "41:07:46", "max_grad_norm": 1.0, "loss": 0.6917972564697266, "grad_norm": 0.43468809127807617, "learning_rate": 0.00017806750914730777}
+{"ts": "2025-12-27T10:06:35", "event": "train_log", "step": 3952, "epoch": 1.667510548523207, "progress_pct": 27.79, "epoch_pct": 27.79, "eta": "41:06:43", "max_grad_norm": 1.0, "loss": 0.6676424741744995, "grad_norm": 0.4071558117866516, "learning_rate": 0.00017803748488867224}
+{"ts": "2025-12-27T10:06:51", "event": "train_log", "step": 3954, "epoch": 1.6683544303797468, "progress_pct": 27.81, "epoch_pct": 27.81, "eta": "41:05:42", "max_grad_norm": 1.0, "loss": 0.7563748359680176, "grad_norm": 0.4478702247142792, "learning_rate": 0.00017800744262828772}
+{"ts": "2025-12-27T10:07:07", "event": "train_log", "step": 3956, "epoch": 1.6691983122362868, "progress_pct": 27.82, "epoch_pct": 27.82, "eta": "41:04:39", "max_grad_norm": 1.0, "loss": 0.685714066028595, "grad_norm": 0.42986053228378296, "learning_rate": 0.00017797738237308432}
+{"ts": "2025-12-27T10:07:23", "event": "train_log", "step": 3958, "epoch": 1.6700421940928272, "progress_pct": 27.83, "epoch_pct": 27.83, "eta": "41:03:37", "max_grad_norm": 1.0, "loss": 0.7753605842590332, "grad_norm": 0.4410569667816162, "learning_rate": 0.0001779473041299964}
+{"ts": "2025-12-27T10:07:37", "event": "train_log", "step": 3960, "epoch": 1.6708860759493671, "progress_pct": 27.85, "epoch_pct": 27.85, "eta": "41:02:32", "max_grad_norm": 1.0, "loss": 0.7826018333435059, "grad_norm": 0.44055965542793274, "learning_rate": 0.00017791720790596242}
+{"ts": "2025-12-27T10:07:53", "event": "train_log", "step": 3962, "epoch": 1.671729957805907, "progress_pct": 27.86, "epoch_pct": 27.86, "eta": "41:01:29", "max_grad_norm": 1.0, "loss": 0.7012629508972168, "grad_norm": 0.4660222828388214, "learning_rate": 0.00017788709370792502}
+{"ts": "2025-12-27T10:08:10", "event": "train_log", "step": 3964, "epoch": 1.6725738396624472, "progress_pct": 27.88, "epoch_pct": 27.88, "eta": "41:00:30", "max_grad_norm": 1.0, "loss": 0.6647007465362549, "grad_norm": 0.3655732274055481, "learning_rate": 0.00017785696154283093}
+{"ts": "2025-12-27T10:08:26", "event": "train_log", "step": 3966, "epoch": 1.6734177215189874, "progress_pct": 27.89, "epoch_pct": 27.89, "eta": "40:59:27", "max_grad_norm": 1.0, "loss": 0.7238346338272095, "grad_norm": 0.45735979080200195, "learning_rate": 0.0001778268114176311}
+{"ts": "2025-12-27T10:08:42", "event": "train_log", "step": 3968, "epoch": 1.6742616033755273, "progress_pct": 27.9, "epoch_pct": 27.9, "eta": "40:58:25", "max_grad_norm": 1.0, "loss": 0.7595300078392029, "grad_norm": 0.4894043207168579, "learning_rate": 0.00017779664333928053}
+{"ts": "2025-12-27T10:09:00", "event": "train_log", "step": 3970, "epoch": 1.6751054852320675, "progress_pct": 27.92, "epoch_pct": 27.92, "eta": "40:57:28", "max_grad_norm": 1.0, "loss": 0.6534504294395447, "grad_norm": 0.3613647222518921, "learning_rate": 0.00017776645731473848}
+{"ts": "2025-12-27T10:09:15", "event": "train_log", "step": 3972, "epoch": 1.6759493670886076, "progress_pct": 27.93, "epoch_pct": 27.93, "eta": "40:56:24", "max_grad_norm": 1.0, "loss": 0.7699148058891296, "grad_norm": 0.4402320683002472, "learning_rate": 0.00017773625335096828}
+{"ts": "2025-12-27T10:09:31", "event": "train_log", "step": 3974, "epoch": 1.6767932489451476, "progress_pct": 27.95, "epoch_pct": 27.95, "eta": "40:55:24", "max_grad_norm": 1.0, "loss": 0.771800696849823, "grad_norm": 0.4391469955444336, "learning_rate": 0.0001777060314549374}
+{"ts": "2025-12-27T10:09:47", "event": "train_log", "step": 3976, "epoch": 1.6776371308016877, "progress_pct": 27.96, "epoch_pct": 27.96, "eta": "40:54:22", "max_grad_norm": 1.0, "loss": 0.6957393884658813, "grad_norm": 0.41348180174827576, "learning_rate": 0.00017767579163361735}
+{"ts": "2025-12-27T10:10:03", "event": "train_log", "step": 3978, "epoch": 1.678481012658228, "progress_pct": 27.97, "epoch_pct": 27.97, "eta": "40:53:21", "max_grad_norm": 1.0, "loss": 0.6666664481163025, "grad_norm": 0.4115046560764313, "learning_rate": 0.00017764553389398407}
+{"ts": "2025-12-27T10:10:18", "event": "train_log", "step": 3980, "epoch": 1.6793248945147679, "progress_pct": 27.99, "epoch_pct": 27.99, "eta": "40:52:16", "max_grad_norm": 1.0, "loss": 0.7364352345466614, "grad_norm": 0.44647306203842163, "learning_rate": 0.00017761525824301725}
+{"ts": "2025-12-27T10:10:33", "event": "train_log", "step": 3982, "epoch": 1.680168776371308, "progress_pct": 28.0, "epoch_pct": 28.0, "eta": "40:51:12", "max_grad_norm": 1.0, "loss": 0.7229774594306946, "grad_norm": 0.424234002828598, "learning_rate": 0.00017758496468770103}
+{"ts": "2025-12-27T10:10:48", "event": "train_log", "step": 3984, "epoch": 1.6810126582278482, "progress_pct": 28.02, "epoch_pct": 28.02, "eta": "40:50:07", "max_grad_norm": 1.0, "loss": 0.6992678642272949, "grad_norm": 0.4850127696990967, "learning_rate": 0.00017755465323502346}
+{"ts": "2025-12-27T10:11:02", "event": "train_log", "step": 3986, "epoch": 1.6818565400843881, "progress_pct": 28.03, "epoch_pct": 28.03, "eta": "40:49:01", "max_grad_norm": 1.0, "loss": 0.6957634091377258, "grad_norm": 0.44637173414230347, "learning_rate": 0.00017752432389197688}
+{"ts": "2025-12-27T10:11:16", "event": "train_log", "step": 3988, "epoch": 1.6827004219409283, "progress_pct": 28.05, "epoch_pct": 28.05, "eta": "40:47:55", "max_grad_norm": 1.0, "loss": 0.6842859983444214, "grad_norm": 0.477393239736557, "learning_rate": 0.00017749397666555768}
+{"ts": "2025-12-27T10:11:32", "event": "train_log", "step": 3990, "epoch": 1.6835443037974684, "progress_pct": 28.06, "epoch_pct": 28.06, "eta": "40:46:54", "max_grad_norm": 1.0, "loss": 0.6478199362754822, "grad_norm": 0.3634146451950073, "learning_rate": 0.00017746361156276632}
+{"ts": "2025-12-27T10:11:47", "event": "train_log", "step": 3992, "epoch": 1.6843881856540084, "progress_pct": 28.07, "epoch_pct": 28.07, "eta": "40:45:50", "max_grad_norm": 1.0, "loss": 0.7166973352432251, "grad_norm": 0.4806389808654785, "learning_rate": 0.00017743322859060752}
+{"ts": "2025-12-27T10:12:02", "event": "train_log", "step": 3994, "epoch": 1.6852320675105485, "progress_pct": 28.09, "epoch_pct": 28.09, "eta": "40:44:47", "max_grad_norm": 1.0, "loss": 0.6624643206596375, "grad_norm": 0.4273380935192108, "learning_rate": 0.00017740282775609}
+{"ts": "2025-12-27T10:12:18", "event": "train_log", "step": 3996, "epoch": 1.6860759493670887, "progress_pct": 28.1, "epoch_pct": 28.1, "eta": "40:43:44", "max_grad_norm": 1.0, "loss": 0.6408489346504211, "grad_norm": 0.4340716600418091, "learning_rate": 0.0001773724090662267}
+{"ts": "2025-12-27T10:12:32", "event": "train_log", "step": 3998, "epoch": 1.6869198312236287, "progress_pct": 28.12, "epoch_pct": 28.12, "eta": "40:42:38", "max_grad_norm": 1.0, "loss": 0.8096797466278076, "grad_norm": 0.50480717420578, "learning_rate": 0.00017734197252803458}
+{"ts": "2025-12-27T10:12:46", "event": "train_log", "step": 4000, "epoch": 1.6877637130801688, "progress_pct": 28.13, "epoch_pct": 28.13, "eta": "40:41:34", "max_grad_norm": 1.0, "loss": 0.6902434825897217, "grad_norm": 0.4549327790737152, "learning_rate": 0.00017731151814853475}
+{"ts": "2025-12-27T10:23:55", "event": "train_log", "step": 4000, "epoch": 1.6877637130801688, "progress_pct": 28.13, "epoch_pct": 28.13, "eta": "41:10:01", "max_grad_norm": 1.0, "eval_loss": 0.7721371650695801, "eval_runtime": 668.395, "eval_samples_per_second": 3.152, "eval_steps_per_second": 3.152}
+{"ts": "2025-12-27T10:24:12", "event": "train_log", "step": 4002, "epoch": 1.688607594936709, "progress_pct": 28.14, "epoch_pct": 28.14, "eta": "41:09:02", "max_grad_norm": 1.0, "loss": 0.6510542035102844, "grad_norm": 0.38909077644348145, "learning_rate": 0.00017728104593475247}
+{"ts": "2025-12-27T10:24:27", "event": "train_log", "step": 4004, "epoch": 1.689451476793249, "progress_pct": 28.16, "epoch_pct": 28.16, "eta": "41:07:56", "max_grad_norm": 1.0, "loss": 0.7331066727638245, "grad_norm": 0.444608598947525, "learning_rate": 0.00017725055589371715}
+{"ts": "2025-12-27T10:24:43", "event": "train_log", "step": 4006, "epoch": 1.690295358649789, "progress_pct": 28.17, "epoch_pct": 28.17, "eta": "41:06:54", "max_grad_norm": 1.0, "loss": 0.5950609445571899, "grad_norm": 0.3364178538322449, "learning_rate": 0.0001772200480324622}
+{"ts": "2025-12-27T10:24:59", "event": "train_log", "step": 4008, "epoch": 1.6911392405063292, "progress_pct": 28.19, "epoch_pct": 28.19, "eta": "41:05:54", "max_grad_norm": 1.0, "loss": 0.7024590969085693, "grad_norm": 0.37264227867126465, "learning_rate": 0.0001771895223580252}
+{"ts": "2025-12-27T10:25:17", "event": "train_log", "step": 4010, "epoch": 1.6919831223628692, "progress_pct": 28.2, "epoch_pct": 28.2, "eta": "41:04:56", "max_grad_norm": 1.0, "loss": 0.6620895266532898, "grad_norm": 0.3576899766921997, "learning_rate": 0.00017715897887744784}
+{"ts": "2025-12-27T10:25:33", "event": "train_log", "step": 4012, "epoch": 1.6928270042194091, "progress_pct": 28.21, "epoch_pct": 28.21, "eta": "41:03:53", "max_grad_norm": 1.0, "loss": 0.6726759076118469, "grad_norm": 0.45855066180229187, "learning_rate": 0.00017712841759777591}
+{"ts": "2025-12-27T10:25:49", "event": "train_log", "step": 4014, "epoch": 1.6936708860759495, "progress_pct": 28.23, "epoch_pct": 28.23, "eta": "41:02:52", "max_grad_norm": 1.0, "loss": 0.6550953984260559, "grad_norm": 0.44610634446144104, "learning_rate": 0.00017709783852605933}
+{"ts": "2025-12-27T10:26:05", "event": "train_log", "step": 4016, "epoch": 1.6945147679324895, "progress_pct": 28.24, "epoch_pct": 28.24, "eta": "41:01:51", "max_grad_norm": 1.0, "loss": 0.7338476777076721, "grad_norm": 0.4262036681175232, "learning_rate": 0.00017706724166935208}
+{"ts": "2025-12-27T10:26:20", "event": "train_log", "step": 4018, "epoch": 1.6953586497890294, "progress_pct": 28.26, "epoch_pct": 28.26, "eta": "41:00:47", "max_grad_norm": 1.0, "loss": 0.6785868406295776, "grad_norm": 0.44584009051322937, "learning_rate": 0.0001770366270347123}
+{"ts": "2025-12-27T10:26:37", "event": "train_log", "step": 4020, "epoch": 1.6962025316455698, "progress_pct": 28.27, "epoch_pct": 28.27, "eta": "40:59:46", "max_grad_norm": 1.0, "loss": 0.7113003134727478, "grad_norm": 0.44828134775161743, "learning_rate": 0.0001770059946292022}
+{"ts": "2025-12-27T10:26:52", "event": "train_log", "step": 4022, "epoch": 1.6970464135021097, "progress_pct": 28.28, "epoch_pct": 28.28, "eta": "40:58:43", "max_grad_norm": 1.0, "loss": 0.6420866847038269, "grad_norm": 0.39716270565986633, "learning_rate": 0.00017697534445988803}
+{"ts": "2025-12-27T10:27:08", "event": "train_log", "step": 4024, "epoch": 1.6978902953586497, "progress_pct": 28.3, "epoch_pct": 28.3, "eta": "40:57:40", "max_grad_norm": 1.0, "loss": 0.8087385892868042, "grad_norm": 0.43546608090400696, "learning_rate": 0.00017694467653384024}
+{"ts": "2025-12-27T10:27:24", "event": "train_log", "step": 4026, "epoch": 1.6987341772151898, "progress_pct": 28.31, "epoch_pct": 28.31, "eta": "40:56:39", "max_grad_norm": 1.0, "loss": 0.6378461718559265, "grad_norm": 0.3698647618293762, "learning_rate": 0.00017691399085813334}
+{"ts": "2025-12-27T10:27:40", "event": "train_log", "step": 4028, "epoch": 1.69957805907173, "progress_pct": 28.33, "epoch_pct": 28.33, "eta": "40:55:37", "max_grad_norm": 1.0, "loss": 0.6504215002059937, "grad_norm": 0.4256688058376312, "learning_rate": 0.0001768832874398459}
+{"ts": "2025-12-27T10:27:55", "event": "train_log", "step": 4030, "epoch": 1.70042194092827, "progress_pct": 28.34, "epoch_pct": 28.34, "eta": "40:54:35", "max_grad_norm": 1.0, "loss": 0.6648795008659363, "grad_norm": 0.4014805555343628, "learning_rate": 0.00017685256628606063}
+{"ts": "2025-12-27T10:28:10", "event": "train_log", "step": 4032, "epoch": 1.70126582278481, "progress_pct": 28.35, "epoch_pct": 28.35, "eta": "40:53:31", "max_grad_norm": 1.0, "loss": 0.6905859112739563, "grad_norm": 0.43924233317375183, "learning_rate": 0.00017682182740386427}
+{"ts": "2025-12-27T10:28:27", "event": "train_log", "step": 4034, "epoch": 1.7021097046413503, "progress_pct": 28.37, "epoch_pct": 28.37, "eta": "40:52:31", "max_grad_norm": 1.0, "loss": 0.7800544500350952, "grad_norm": 0.43565094470977783, "learning_rate": 0.00017679107080034774}
+{"ts": "2025-12-27T10:28:42", "event": "train_log", "step": 4036, "epoch": 1.7029535864978902, "progress_pct": 28.38, "epoch_pct": 28.38, "eta": "40:51:27", "max_grad_norm": 1.0, "loss": 0.7487648725509644, "grad_norm": 0.4681089520454407, "learning_rate": 0.00017676029648260597}
+{"ts": "2025-12-27T10:28:59", "event": "train_log", "step": 4038, "epoch": 1.7037974683544304, "progress_pct": 28.4, "epoch_pct": 28.4, "eta": "40:50:28", "max_grad_norm": 1.0, "loss": 0.7374032735824585, "grad_norm": 0.41400185227394104, "learning_rate": 0.00017672950445773804}
+{"ts": "2025-12-27T10:29:14", "event": "train_log", "step": 4040, "epoch": 1.7046413502109705, "progress_pct": 28.41, "epoch_pct": 28.41, "eta": "40:49:25", "max_grad_norm": 1.0, "loss": 0.7539847493171692, "grad_norm": 0.5120159983634949, "learning_rate": 0.00017669869473284702}
+{"ts": "2025-12-27T10:29:30", "event": "train_log", "step": 4042, "epoch": 1.7054852320675105, "progress_pct": 28.42, "epoch_pct": 28.42, "eta": "40:48:24", "max_grad_norm": 1.0, "loss": 0.7029932737350464, "grad_norm": 0.4043315052986145, "learning_rate": 0.00017666786731504013}
+{"ts": "2025-12-27T10:29:47", "event": "train_log", "step": 4044, "epoch": 1.7063291139240506, "progress_pct": 28.44, "epoch_pct": 28.44, "eta": "40:47:23", "max_grad_norm": 1.0, "loss": 0.6464276313781738, "grad_norm": 0.462801992893219, "learning_rate": 0.00017663702221142874}
+{"ts": "2025-12-27T10:30:01", "event": "train_log", "step": 4046, "epoch": 1.7071729957805908, "progress_pct": 28.45, "epoch_pct": 28.45, "eta": "40:46:19", "max_grad_norm": 1.0, "loss": 0.7646113634109497, "grad_norm": 0.4751932621002197, "learning_rate": 0.00017660615942912813}
+{"ts": "2025-12-27T10:30:16", "event": "train_log", "step": 4048, "epoch": 1.7080168776371307, "progress_pct": 28.47, "epoch_pct": 28.47, "eta": "40:45:14", "max_grad_norm": 1.0, "loss": 0.7191748023033142, "grad_norm": 0.41720855236053467, "learning_rate": 0.0001765752789752578}
+{"ts": "2025-12-27T10:30:31", "event": "train_log", "step": 4050, "epoch": 1.7088607594936709, "progress_pct": 28.48, "epoch_pct": 28.48, "eta": "40:44:11", "max_grad_norm": 1.0, "loss": 0.6946488618850708, "grad_norm": 0.40235280990600586, "learning_rate": 0.00017654438085694128}
+{"ts": "2025-12-27T10:30:46", "event": "train_log", "step": 4052, "epoch": 1.709704641350211, "progress_pct": 28.5, "epoch_pct": 28.5, "eta": "40:43:06", "max_grad_norm": 1.0, "loss": 0.7655195593833923, "grad_norm": 0.4385916292667389, "learning_rate": 0.00017651346508130612}
+{"ts": "2025-12-27T10:31:01", "event": "train_log", "step": 4054, "epoch": 1.710548523206751, "progress_pct": 28.51, "epoch_pct": 28.51, "eta": "40:42:04", "max_grad_norm": 1.0, "loss": 0.7192418575286865, "grad_norm": 0.4192262887954712, "learning_rate": 0.00017648253165548406}
+{"ts": "2025-12-27T10:31:17", "event": "train_log", "step": 4056, "epoch": 1.7113924050632912, "progress_pct": 28.52, "epoch_pct": 28.52, "eta": "40:41:03", "max_grad_norm": 1.0, "loss": 0.6501081585884094, "grad_norm": 0.4348951280117035, "learning_rate": 0.00017645158058661082}
+{"ts": "2025-12-27T10:31:34", "event": "train_log", "step": 4058, "epoch": 1.7122362869198313, "progress_pct": 28.54, "epoch_pct": 28.54, "eta": "40:40:04", "max_grad_norm": 1.0, "loss": 0.741439163684845, "grad_norm": 0.4178772270679474, "learning_rate": 0.0001764206118818262}
+{"ts": "2025-12-27T10:31:49", "event": "train_log", "step": 4060, "epoch": 1.7130801687763713, "progress_pct": 28.55, "epoch_pct": 28.55, "eta": "40:39:01", "max_grad_norm": 1.0, "loss": 0.7087798714637756, "grad_norm": 0.46996670961380005, "learning_rate": 0.00017638962554827414}
+{"ts": "2025-12-27T10:32:05", "event": "train_log", "step": 4062, "epoch": 1.7139240506329114, "progress_pct": 28.57, "epoch_pct": 28.57, "eta": "40:38:00", "max_grad_norm": 1.0, "loss": 0.6716644167900085, "grad_norm": 0.4775761365890503, "learning_rate": 0.00017635862159310253}
+{"ts": "2025-12-27T10:32:21", "event": "train_log", "step": 4064, "epoch": 1.7147679324894516, "progress_pct": 28.58, "epoch_pct": 28.58, "eta": "40:36:58", "max_grad_norm": 1.0, "loss": 0.6630500555038452, "grad_norm": 0.4094069004058838, "learning_rate": 0.00017632760002346345}
+{"ts": "2025-12-27T10:32:36", "event": "train_log", "step": 4066, "epoch": 1.7156118143459915, "progress_pct": 28.59, "epoch_pct": 28.59, "eta": "40:35:56", "max_grad_norm": 1.0, "loss": 0.7678145170211792, "grad_norm": 0.3968847393989563, "learning_rate": 0.0001762965608465129}
+{"ts": "2025-12-27T10:32:52", "event": "train_log", "step": 4068, "epoch": 1.7164556962025317, "progress_pct": 28.61, "epoch_pct": 28.61, "eta": "40:34:54", "max_grad_norm": 1.0, "loss": 0.6981875896453857, "grad_norm": 0.43751946091651917, "learning_rate": 0.0001762655040694111}
+{"ts": "2025-12-27T10:33:06", "event": "train_log", "step": 4070, "epoch": 1.7172995780590719, "progress_pct": 28.62, "epoch_pct": 28.62, "eta": "40:33:50", "max_grad_norm": 1.0, "loss": 0.6919799447059631, "grad_norm": 0.44504284858703613, "learning_rate": 0.00017623442969932218}
+{"ts": "2025-12-27T10:33:23", "event": "train_log", "step": 4072, "epoch": 1.7181434599156118, "progress_pct": 28.64, "epoch_pct": 28.64, "eta": "40:32:52", "max_grad_norm": 1.0, "loss": 0.6310772895812988, "grad_norm": 0.36895114183425903, "learning_rate": 0.00017620333774341447}
+{"ts": "2025-12-27T10:33:38", "event": "train_log", "step": 4074, "epoch": 1.7189873417721517, "progress_pct": 28.65, "epoch_pct": 28.65, "eta": "40:31:49", "max_grad_norm": 1.0, "loss": 0.6954092383384705, "grad_norm": 0.4312762916088104, "learning_rate": 0.00017617222820886022}
+{"ts": "2025-12-27T10:33:54", "event": "train_log", "step": 4076, "epoch": 1.7198312236286921, "progress_pct": 28.66, "epoch_pct": 28.66, "eta": "40:30:48", "max_grad_norm": 1.0, "loss": 0.6973698139190674, "grad_norm": 0.40909886360168457, "learning_rate": 0.00017614110110283587}
+{"ts": "2025-12-27T10:34:10", "event": "train_log", "step": 4078, "epoch": 1.720675105485232, "progress_pct": 28.68, "epoch_pct": 28.68, "eta": "40:29:46", "max_grad_norm": 1.0, "loss": 0.6775513291358948, "grad_norm": 0.43111470341682434, "learning_rate": 0.0001761099564325218}
+{"ts": "2025-12-27T10:34:25", "event": "train_log", "step": 4080, "epoch": 1.721518987341772, "progress_pct": 28.69, "epoch_pct": 28.69, "eta": "40:28:45", "max_grad_norm": 1.0, "loss": 0.6889033913612366, "grad_norm": 0.45573189854621887, "learning_rate": 0.0001760787942051025}
+{"ts": "2025-12-27T10:34:40", "event": "train_log", "step": 4082, "epoch": 1.7223628691983124, "progress_pct": 28.71, "epoch_pct": 28.71, "eta": "40:27:41", "max_grad_norm": 1.0, "loss": 0.6894892454147339, "grad_norm": 0.44934743642807007, "learning_rate": 0.0001760476144277665}
+{"ts": "2025-12-27T10:34:57", "event": "train_log", "step": 4084, "epoch": 1.7232067510548523, "progress_pct": 28.72, "epoch_pct": 28.72, "eta": "40:26:43", "max_grad_norm": 1.0, "loss": 0.6841691732406616, "grad_norm": 0.39716506004333496, "learning_rate": 0.0001760164171077064}
+{"ts": "2025-12-27T10:35:12", "event": "train_log", "step": 4086, "epoch": 1.7240506329113923, "progress_pct": 28.73, "epoch_pct": 28.73, "eta": "40:25:40", "max_grad_norm": 1.0, "loss": 0.7197495102882385, "grad_norm": 0.4486255645751953, "learning_rate": 0.00017598520225211876}
+{"ts": "2025-12-27T10:35:28", "event": "train_log", "step": 4088, "epoch": 1.7248945147679327, "progress_pct": 28.75, "epoch_pct": 28.75, "eta": "40:24:41", "max_grad_norm": 1.0, "loss": 0.7388272881507874, "grad_norm": 0.40130025148391724, "learning_rate": 0.00017595396986820432}
+{"ts": "2025-12-27T10:35:43", "event": "train_log", "step": 4090, "epoch": 1.7257383966244726, "progress_pct": 28.76, "epoch_pct": 28.76, "eta": "40:23:38", "max_grad_norm": 1.0, "loss": 0.7447546720504761, "grad_norm": 0.40419334173202515, "learning_rate": 0.0001759227199631677}
+{"ts": "2025-12-27T10:35:59", "event": "train_log", "step": 4092, "epoch": 1.7265822784810125, "progress_pct": 28.78, "epoch_pct": 28.78, "eta": "40:22:39", "max_grad_norm": 1.0, "loss": 0.6348387598991394, "grad_norm": 0.39778804779052734, "learning_rate": 0.00017589145254421774}
+{"ts": "2025-12-27T10:36:14", "event": "train_log", "step": 4094, "epoch": 1.7274261603375527, "progress_pct": 28.79, "epoch_pct": 28.79, "eta": "40:21:36", "max_grad_norm": 1.0, "loss": 0.6813296675682068, "grad_norm": 0.3934648036956787, "learning_rate": 0.0001758601676185672}
+{"ts": "2025-12-27T10:36:30", "event": "train_log", "step": 4096, "epoch": 1.7282700421940929, "progress_pct": 28.8, "epoch_pct": 28.8, "eta": "40:20:36", "max_grad_norm": 1.0, "loss": 0.6369423270225525, "grad_norm": 0.4080725908279419, "learning_rate": 0.0001758288651934329}
+{"ts": "2025-12-27T10:36:46", "event": "train_log", "step": 4098, "epoch": 1.7291139240506328, "progress_pct": 28.82, "epoch_pct": 28.82, "eta": "40:19:34", "max_grad_norm": 1.0, "loss": 0.6664180159568787, "grad_norm": 0.5186997652053833, "learning_rate": 0.00017579754527603575}
+{"ts": "2025-12-27T10:37:01", "event": "train_log", "step": 4100, "epoch": 1.729957805907173, "progress_pct": 28.83, "epoch_pct": 28.83, "eta": "40:18:33", "max_grad_norm": 1.0, "loss": 0.7181394696235657, "grad_norm": 0.45290765166282654, "learning_rate": 0.0001757662078736006}
+{"ts": "2025-12-27T10:48:15", "event": "train_log", "step": 4100, "epoch": 1.729957805907173, "progress_pct": 28.83, "epoch_pct": 28.83, "eta": "40:46:16", "max_grad_norm": 1.0, "eval_loss": 0.7690847516059875, "eval_runtime": 673.6323, "eval_samples_per_second": 3.128, "eval_steps_per_second": 3.128}
+{"ts": "2025-12-27T10:48:29", "event": "train_log", "step": 4102, "epoch": 1.7308016877637131, "progress_pct": 28.85, "epoch_pct": 28.85, "eta": "40:45:11", "max_grad_norm": 1.0, "loss": 0.6850123405456543, "grad_norm": 0.45677125453948975, "learning_rate": 0.00017573485299335643}
+{"ts": "2025-12-27T10:48:44", "event": "train_log", "step": 4104, "epoch": 1.731645569620253, "progress_pct": 28.86, "epoch_pct": 28.86, "eta": "40:44:07", "max_grad_norm": 1.0, "loss": 0.7225809693336487, "grad_norm": 0.4854084253311157, "learning_rate": 0.0001757034806425362}
+{"ts": "2025-12-27T10:48:59", "event": "train_log", "step": 4106, "epoch": 1.7324894514767932, "progress_pct": 28.87, "epoch_pct": 28.87, "eta": "40:43:02", "max_grad_norm": 1.0, "loss": 0.7303882837295532, "grad_norm": 0.486583411693573, "learning_rate": 0.0001756720908283769}
+{"ts": "2025-12-27T10:49:14", "event": "train_log", "step": 4108, "epoch": 1.7333333333333334, "progress_pct": 28.89, "epoch_pct": 28.89, "eta": "40:42:00", "max_grad_norm": 1.0, "loss": 0.6789878010749817, "grad_norm": 0.4609545171260834, "learning_rate": 0.00017564068355811952}
+{"ts": "2025-12-27T10:49:29", "event": "train_log", "step": 4110, "epoch": 1.7341772151898733, "progress_pct": 28.9, "epoch_pct": 28.9, "eta": "40:40:56", "max_grad_norm": 1.0, "loss": 0.7311062812805176, "grad_norm": 0.3853047788143158, "learning_rate": 0.00017560925883900914}
+{"ts": "2025-12-27T10:49:44", "event": "train_log", "step": 4112, "epoch": 1.7350210970464135, "progress_pct": 28.92, "epoch_pct": 28.92, "eta": "40:39:52", "max_grad_norm": 1.0, "loss": 0.7106070518493652, "grad_norm": 0.41957035660743713, "learning_rate": 0.00017557781667829487}
+{"ts": "2025-12-27T10:49:59", "event": "train_log", "step": 4114, "epoch": 1.7358649789029537, "progress_pct": 28.93, "epoch_pct": 28.93, "eta": "40:38:49", "max_grad_norm": 1.0, "loss": 0.6345783472061157, "grad_norm": 0.37419262528419495, "learning_rate": 0.0001755463570832298}
+{"ts": "2025-12-27T10:50:13", "event": "train_log", "step": 4116, "epoch": 1.7367088607594936, "progress_pct": 28.95, "epoch_pct": 28.95, "eta": "40:37:44", "max_grad_norm": 1.0, "loss": 0.7923876047134399, "grad_norm": 0.5050614476203918, "learning_rate": 0.00017551488006107104}
+{"ts": "2025-12-27T10:50:28", "event": "train_log", "step": 4118, "epoch": 1.7375527426160338, "progress_pct": 28.96, "epoch_pct": 28.96, "eta": "40:36:41", "max_grad_norm": 1.0, "loss": 0.7502455115318298, "grad_norm": 0.49726125597953796, "learning_rate": 0.00017548338561907976}
+{"ts": "2025-12-27T10:50:44", "event": "train_log", "step": 4120, "epoch": 1.738396624472574, "progress_pct": 28.97, "epoch_pct": 28.97, "eta": "40:35:40", "max_grad_norm": 1.0, "loss": 0.6333624720573425, "grad_norm": 0.38956591486930847, "learning_rate": 0.00017545187376452105}
+{"ts": "2025-12-27T10:50:59", "event": "train_log", "step": 4122, "epoch": 1.7392405063291139, "progress_pct": 28.99, "epoch_pct": 28.99, "eta": "40:34:38", "max_grad_norm": 1.0, "loss": 0.6884907484054565, "grad_norm": 0.430318146944046, "learning_rate": 0.0001754203445046642}
+{"ts": "2025-12-27T10:51:15", "event": "train_log", "step": 4124, "epoch": 1.740084388185654, "progress_pct": 29.0, "epoch_pct": 29.0, "eta": "40:33:35", "max_grad_norm": 1.0, "loss": 0.7345985174179077, "grad_norm": 0.39826440811157227, "learning_rate": 0.0001753887978467823}
+{"ts": "2025-12-27T10:51:30", "event": "train_log", "step": 4126, "epoch": 1.7409282700421942, "progress_pct": 29.02, "epoch_pct": 29.02, "eta": "40:32:32", "max_grad_norm": 1.0, "loss": 0.6537752151489258, "grad_norm": 0.4263492822647095, "learning_rate": 0.00017535723379815267}
+{"ts": "2025-12-27T10:51:45", "event": "train_log", "step": 4128, "epoch": 1.7417721518987341, "progress_pct": 29.03, "epoch_pct": 29.03, "eta": "40:31:31", "max_grad_norm": 1.0, "loss": 0.6368000507354736, "grad_norm": 0.4012296199798584, "learning_rate": 0.0001753256523660564}
+{"ts": "2025-12-27T10:52:01", "event": "train_log", "step": 4130, "epoch": 1.7426160337552743, "progress_pct": 29.04, "epoch_pct": 29.04, "eta": "40:30:29", "max_grad_norm": 1.0, "loss": 0.6038334369659424, "grad_norm": 0.3881204426288605, "learning_rate": 0.0001752940535577788}
+{"ts": "2025-12-27T10:52:16", "event": "train_log", "step": 4132, "epoch": 1.7434599156118145, "progress_pct": 29.06, "epoch_pct": 29.06, "eta": "40:29:28", "max_grad_norm": 1.0, "loss": 0.6657052636146545, "grad_norm": 0.38410526514053345, "learning_rate": 0.0001752624373806091}
+{"ts": "2025-12-27T10:52:31", "event": "train_log", "step": 4134, "epoch": 1.7443037974683544, "progress_pct": 29.07, "epoch_pct": 29.07, "eta": "40:28:25", "max_grad_norm": 1.0, "loss": 0.6518114805221558, "grad_norm": 0.3970552980899811, "learning_rate": 0.00017523080384184057}
+{"ts": "2025-12-27T10:52:47", "event": "train_log", "step": 4136, "epoch": 1.7451476793248946, "progress_pct": 29.09, "epoch_pct": 29.09, "eta": "40:27:24", "max_grad_norm": 1.0, "loss": 0.609853208065033, "grad_norm": 0.37542012333869934, "learning_rate": 0.0001751991529487704}
+{"ts": "2025-12-27T10:53:02", "event": "train_log", "step": 4138, "epoch": 1.7459915611814347, "progress_pct": 29.1, "epoch_pct": 29.1, "eta": "40:26:21", "max_grad_norm": 1.0, "loss": 0.7356346845626831, "grad_norm": 0.49812132120132446, "learning_rate": 0.00017516748470869988}
+{"ts": "2025-12-27T10:53:17", "event": "train_log", "step": 4140, "epoch": 1.7468354430379747, "progress_pct": 29.11, "epoch_pct": 29.11, "eta": "40:25:18", "max_grad_norm": 1.0, "loss": 0.7198674082756042, "grad_norm": 0.48197710514068604, "learning_rate": 0.00017513579912893426}
+{"ts": "2025-12-27T10:53:31", "event": "train_log", "step": 4142, "epoch": 1.7476793248945146, "progress_pct": 29.13, "epoch_pct": 29.13, "eta": "40:24:14", "max_grad_norm": 1.0, "loss": 0.6841102242469788, "grad_norm": 0.38266444206237793, "learning_rate": 0.0001751040962167828}
+{"ts": "2025-12-27T10:53:46", "event": "train_log", "step": 4144, "epoch": 1.748523206751055, "progress_pct": 29.14, "epoch_pct": 29.14, "eta": "40:23:12", "max_grad_norm": 1.0, "loss": 0.7348533272743225, "grad_norm": 0.4165695309638977, "learning_rate": 0.0001750723759795587}
+{"ts": "2025-12-27T10:54:01", "event": "train_log", "step": 4146, "epoch": 1.749367088607595, "progress_pct": 29.16, "epoch_pct": 29.16, "eta": "40:22:10", "max_grad_norm": 1.0, "loss": 0.7344526648521423, "grad_norm": 0.4414960443973541, "learning_rate": 0.0001750406384245793}
+{"ts": "2025-12-27T10:54:16", "event": "train_log", "step": 4148, "epoch": 1.7502109704641349, "progress_pct": 29.17, "epoch_pct": 29.17, "eta": "40:21:07", "max_grad_norm": 1.0, "loss": 0.7047654986381531, "grad_norm": 0.4324318468570709, "learning_rate": 0.00017500888355916576}
+{"ts": "2025-12-27T10:54:31", "event": "train_log", "step": 4150, "epoch": 1.7510548523206753, "progress_pct": 29.18, "epoch_pct": 29.18, "eta": "40:20:04", "max_grad_norm": 1.0, "loss": 0.7317671179771423, "grad_norm": 0.44670364260673523, "learning_rate": 0.00017497711139064336}
+{"ts": "2025-12-27T10:54:46", "event": "train_log", "step": 4152, "epoch": 1.7518987341772152, "progress_pct": 29.2, "epoch_pct": 29.2, "eta": "40:19:01", "max_grad_norm": 1.0, "loss": 0.7447407245635986, "grad_norm": 0.43074679374694824, "learning_rate": 0.00017494532192634138}
+{"ts": "2025-12-27T10:55:01", "event": "train_log", "step": 4154, "epoch": 1.7527426160337551, "progress_pct": 29.21, "epoch_pct": 29.21, "eta": "40:17:59", "max_grad_norm": 1.0, "loss": 0.7352449893951416, "grad_norm": 0.43057894706726074, "learning_rate": 0.00017491351517359292}
+{"ts": "2025-12-27T10:55:17", "event": "train_log", "step": 4156, "epoch": 1.7535864978902953, "progress_pct": 29.23, "epoch_pct": 29.23, "eta": "40:16:58", "max_grad_norm": 1.0, "loss": 0.7244091033935547, "grad_norm": 0.49488556385040283, "learning_rate": 0.00017488169113973528}
+{"ts": "2025-12-27T10:55:31", "event": "train_log", "step": 4158, "epoch": 1.7544303797468355, "progress_pct": 29.24, "epoch_pct": 29.24, "eta": "40:15:54", "max_grad_norm": 1.0, "loss": 0.7515795230865479, "grad_norm": 0.4565380811691284, "learning_rate": 0.0001748498498321096}
+{"ts": "2025-12-27T10:55:47", "event": "train_log", "step": 4160, "epoch": 1.7552742616033754, "progress_pct": 29.25, "epoch_pct": 29.25, "eta": "40:14:54", "max_grad_norm": 1.0, "loss": 0.7451042532920837, "grad_norm": 0.45722341537475586, "learning_rate": 0.0001748179912580611}
+{"ts": "2025-12-27T10:56:02", "event": "train_log", "step": 4162, "epoch": 1.7561181434599156, "progress_pct": 29.27, "epoch_pct": 29.27, "eta": "40:13:53", "max_grad_norm": 1.0, "loss": 0.7133217453956604, "grad_norm": 0.47247326374053955, "learning_rate": 0.0001747861154249389}
+{"ts": "2025-12-27T10:56:17", "event": "train_log", "step": 4164, "epoch": 1.7569620253164557, "progress_pct": 29.28, "epoch_pct": 29.28, "eta": "40:12:51", "max_grad_norm": 1.0, "loss": 0.643742561340332, "grad_norm": 0.48307132720947266, "learning_rate": 0.00017475422234009623}
+{"ts": "2025-12-27T10:56:33", "event": "train_log", "step": 4166, "epoch": 1.7578059071729957, "progress_pct": 29.3, "epoch_pct": 29.3, "eta": "40:11:50", "max_grad_norm": 1.0, "loss": 0.7587857842445374, "grad_norm": 0.49163761734962463, "learning_rate": 0.00017472231201089012}
+{"ts": "2025-12-27T10:56:48", "event": "train_log", "step": 4168, "epoch": 1.7586497890295358, "progress_pct": 29.31, "epoch_pct": 29.31, "eta": "40:10:49", "max_grad_norm": 1.0, "loss": 0.6980416774749756, "grad_norm": 0.4726962149143219, "learning_rate": 0.00017469038444468165}
+{"ts": "2025-12-27T10:57:04", "event": "train_log", "step": 4170, "epoch": 1.759493670886076, "progress_pct": 29.32, "epoch_pct": 29.32, "eta": "40:09:48", "max_grad_norm": 1.0, "loss": 0.6929903626441956, "grad_norm": 0.4192392826080322, "learning_rate": 0.00017465843964883598}
+{"ts": "2025-12-27T10:57:19", "event": "train_log", "step": 4172, "epoch": 1.760337552742616, "progress_pct": 29.34, "epoch_pct": 29.34, "eta": "40:08:47", "max_grad_norm": 1.0, "loss": 0.7076222896575928, "grad_norm": 0.43000563979148865, "learning_rate": 0.00017462647763072216}
+{"ts": "2025-12-27T10:57:35", "event": "train_log", "step": 4174, "epoch": 1.761181434599156, "progress_pct": 29.35, "epoch_pct": 29.35, "eta": "40:07:47", "max_grad_norm": 1.0, "loss": 0.7260748147964478, "grad_norm": 0.45402681827545166, "learning_rate": 0.00017459449839771313}
+{"ts": "2025-12-27T10:57:50", "event": "train_log", "step": 4176, "epoch": 1.7620253164556963, "progress_pct": 29.37, "epoch_pct": 29.37, "eta": "40:06:46", "max_grad_norm": 1.0, "loss": 0.6901930570602417, "grad_norm": 0.4077165126800537, "learning_rate": 0.00017456250195718595}
+{"ts": "2025-12-27T10:58:06", "event": "train_log", "step": 4178, "epoch": 1.7628691983122362, "progress_pct": 29.38, "epoch_pct": 29.38, "eta": "40:05:46", "max_grad_norm": 1.0, "loss": 0.7085300087928772, "grad_norm": 0.4167994558811188, "learning_rate": 0.00017453048831652157}
+{"ts": "2025-12-27T10:58:22", "event": "train_log", "step": 4180, "epoch": 1.7637130801687764, "progress_pct": 29.4, "epoch_pct": 29.4, "eta": "40:04:48", "max_grad_norm": 1.0, "loss": 0.7646799087524414, "grad_norm": 0.45262131094932556, "learning_rate": 0.00017449845748310493}
+{"ts": "2025-12-27T10:58:37", "event": "train_log", "step": 4182, "epoch": 1.7645569620253165, "progress_pct": 29.41, "epoch_pct": 29.41, "eta": "40:03:46", "max_grad_norm": 1.0, "loss": 0.7847923040390015, "grad_norm": 0.4639943540096283, "learning_rate": 0.0001744664094643249}
+{"ts": "2025-12-27T10:58:54", "event": "train_log", "step": 4184, "epoch": 1.7654008438818565, "progress_pct": 29.42, "epoch_pct": 29.42, "eta": "40:02:49", "max_grad_norm": 1.0, "loss": 0.7192286252975464, "grad_norm": 0.33605465292930603, "learning_rate": 0.00017443434426757439}
+{"ts": "2025-12-27T10:59:11", "event": "train_log", "step": 4186, "epoch": 1.7662447257383966, "progress_pct": 29.44, "epoch_pct": 29.44, "eta": "40:01:51", "max_grad_norm": 1.0, "loss": 0.6554746627807617, "grad_norm": 0.4036388099193573, "learning_rate": 0.00017440226190025015}
+{"ts": "2025-12-27T10:59:27", "event": "train_log", "step": 4188, "epoch": 1.7670886075949368, "progress_pct": 29.45, "epoch_pct": 29.45, "eta": "40:00:52", "max_grad_norm": 1.0, "loss": 0.6433794498443604, "grad_norm": 0.45648789405822754, "learning_rate": 0.00017437016236975303}
+{"ts": "2025-12-27T10:59:42", "event": "train_log", "step": 4190, "epoch": 1.7679324894514767, "progress_pct": 29.47, "epoch_pct": 29.47, "eta": "39:59:52", "max_grad_norm": 1.0, "loss": 0.6608775854110718, "grad_norm": 0.38929006457328796, "learning_rate": 0.00017433804568348776}
+{"ts": "2025-12-27T11:00:00", "event": "train_log", "step": 4192, "epoch": 1.768776371308017, "progress_pct": 29.48, "epoch_pct": 29.48, "eta": "39:58:56", "max_grad_norm": 1.0, "loss": 0.6470816135406494, "grad_norm": 0.37158843874931335, "learning_rate": 0.00017430591184886308}
+{"ts": "2025-12-27T11:00:16", "event": "train_log", "step": 4194, "epoch": 1.769620253164557, "progress_pct": 29.49, "epoch_pct": 29.49, "eta": "39:57:56", "max_grad_norm": 1.0, "loss": 0.6825124025344849, "grad_norm": 0.39842307567596436, "learning_rate": 0.00017427376087329158}
+{"ts": "2025-12-27T11:00:31", "event": "train_log", "step": 4196, "epoch": 1.770464135021097, "progress_pct": 29.51, "epoch_pct": 29.51, "eta": "39:56:56", "max_grad_norm": 1.0, "loss": 0.7641696333885193, "grad_norm": 0.44220465421676636, "learning_rate": 0.00017424159276418987}
+{"ts": "2025-12-27T11:00:48", "event": "train_log", "step": 4198, "epoch": 1.7713080168776372, "progress_pct": 29.52, "epoch_pct": 29.52, "eta": "39:55:59", "max_grad_norm": 1.0, "loss": 0.7783681154251099, "grad_norm": 0.44530072808265686, "learning_rate": 0.0001742094075289786}
+{"ts": "2025-12-27T11:01:04", "event": "train_log", "step": 4200, "epoch": 1.7721518987341773, "progress_pct": 29.54, "epoch_pct": 29.54, "eta": "39:54:59", "max_grad_norm": 1.0, "loss": 0.6718224883079529, "grad_norm": 0.4497349262237549, "learning_rate": 0.00017417720517508216}
+{"ts": "2025-12-27T11:12:31", "event": "train_log", "step": 4200, "epoch": 1.7721518987341773, "progress_pct": 29.54, "epoch_pct": 29.54, "eta": "40:22:20", "max_grad_norm": 1.0, "eval_loss": 0.7676366567611694, "eval_runtime": 687.9619, "eval_samples_per_second": 3.063, "eval_steps_per_second": 3.063}
+{"ts": "2025-12-27T11:12:47", "event": "train_log", "step": 4202, "epoch": 1.7729957805907173, "progress_pct": 29.55, "epoch_pct": 29.55, "eta": "40:21:19", "max_grad_norm": 1.0, "loss": 0.7050814032554626, "grad_norm": 0.3857066333293915, "learning_rate": 0.00017414498570992914}
+{"ts": "2025-12-27T11:13:01", "event": "train_log", "step": 4204, "epoch": 1.7738396624472574, "progress_pct": 29.56, "epoch_pct": 29.56, "eta": "40:20:16", "max_grad_norm": 1.0, "loss": 0.7222742438316345, "grad_norm": 0.4533625543117523, "learning_rate": 0.00017411274914095188}
+{"ts": "2025-12-27T11:13:16", "event": "train_log", "step": 4206, "epoch": 1.7746835443037976, "progress_pct": 29.58, "epoch_pct": 29.58, "eta": "40:19:13", "max_grad_norm": 1.0, "loss": 0.6434906125068665, "grad_norm": 0.40805840492248535, "learning_rate": 0.00017408049547558676}
+{"ts": "2025-12-27T11:13:31", "event": "train_log", "step": 4208, "epoch": 1.7755274261603375, "progress_pct": 29.59, "epoch_pct": 29.59, "eta": "40:18:11", "max_grad_norm": 1.0, "loss": 0.686283528804779, "grad_norm": 0.5041740536689758, "learning_rate": 0.00017404822472127406}
+{"ts": "2025-12-27T11:13:47", "event": "train_log", "step": 4210, "epoch": 1.7763713080168775, "progress_pct": 29.61, "epoch_pct": 29.61, "eta": "40:17:10", "max_grad_norm": 1.0, "loss": 0.6785043478012085, "grad_norm": 0.41846802830696106, "learning_rate": 0.00017401593688545807}
+{"ts": "2025-12-27T11:14:02", "event": "train_log", "step": 4212, "epoch": 1.7772151898734179, "progress_pct": 29.62, "epoch_pct": 29.62, "eta": "40:16:08", "max_grad_norm": 1.0, "loss": 0.7566665410995483, "grad_norm": 0.4696303606033325, "learning_rate": 0.00017398363197558694}
+{"ts": "2025-12-27T11:14:18", "event": "train_log", "step": 4214, "epoch": 1.7780590717299578, "progress_pct": 29.63, "epoch_pct": 29.63, "eta": "40:15:08", "max_grad_norm": 1.0, "loss": 0.6178455948829651, "grad_norm": 0.3731461763381958, "learning_rate": 0.0001739513099991128}
+{"ts": "2025-12-27T11:14:33", "event": "train_log", "step": 4216, "epoch": 1.7789029535864977, "progress_pct": 29.65, "epoch_pct": 29.65, "eta": "40:14:05", "max_grad_norm": 1.0, "loss": 0.771265983581543, "grad_norm": 0.48098450899124146, "learning_rate": 0.00017391897096349167}
+{"ts": "2025-12-27T11:14:47", "event": "train_log", "step": 4218, "epoch": 1.7797468354430381, "progress_pct": 29.66, "epoch_pct": 29.66, "eta": "40:13:02", "max_grad_norm": 1.0, "loss": 0.6372597813606262, "grad_norm": 0.4164000451564789, "learning_rate": 0.0001738866148761836}
+{"ts": "2025-12-27T11:15:04", "event": "train_log", "step": 4220, "epoch": 1.780590717299578, "progress_pct": 29.68, "epoch_pct": 29.68, "eta": "40:12:03", "max_grad_norm": 1.0, "loss": 0.7213301658630371, "grad_norm": 0.4490415155887604, "learning_rate": 0.00017385424174465251}
+{"ts": "2025-12-27T11:15:18", "event": "train_log", "step": 4222, "epoch": 1.781434599156118, "progress_pct": 29.69, "epoch_pct": 29.69, "eta": "40:11:01", "max_grad_norm": 1.0, "loss": 0.7575726509094238, "grad_norm": 0.44518229365348816, "learning_rate": 0.00017382185157636623}
+{"ts": "2025-12-27T11:15:34", "event": "train_log", "step": 4224, "epoch": 1.7822784810126582, "progress_pct": 29.7, "epoch_pct": 29.7, "eta": "40:10:01", "max_grad_norm": 1.0, "loss": 0.6473780274391174, "grad_norm": 0.34899434447288513, "learning_rate": 0.00017378944437879658}
+{"ts": "2025-12-27T11:15:48", "event": "train_log", "step": 4226, "epoch": 1.7831223628691983, "progress_pct": 29.72, "epoch_pct": 29.72, "eta": "40:08:57", "max_grad_norm": 1.0, "loss": 0.7238336205482483, "grad_norm": 0.4068683087825775, "learning_rate": 0.00017375702015941924}
+{"ts": "2025-12-27T11:16:04", "event": "train_log", "step": 4228, "epoch": 1.7839662447257383, "progress_pct": 29.73, "epoch_pct": 29.73, "eta": "40:07:56", "max_grad_norm": 1.0, "loss": 0.7797979116439819, "grad_norm": 0.4025551974773407, "learning_rate": 0.0001737245789257139}
+{"ts": "2025-12-27T11:16:20", "event": "train_log", "step": 4230, "epoch": 1.7848101265822784, "progress_pct": 29.75, "epoch_pct": 29.75, "eta": "40:06:56", "max_grad_norm": 1.0, "loss": 0.6304885149002075, "grad_norm": 0.35764896869659424, "learning_rate": 0.00017369212068516412}
+{"ts": "2025-12-27T11:16:34", "event": "train_log", "step": 4232, "epoch": 1.7856540084388186, "progress_pct": 29.76, "epoch_pct": 29.76, "eta": "40:05:53", "max_grad_norm": 1.0, "loss": 0.6606637835502625, "grad_norm": 0.43337100744247437, "learning_rate": 0.00017365964544525738}
+{"ts": "2025-12-27T11:16:49", "event": "train_log", "step": 4234, "epoch": 1.7864978902953585, "progress_pct": 29.77, "epoch_pct": 29.77, "eta": "40:04:51", "max_grad_norm": 1.0, "loss": 0.712163507938385, "grad_norm": 0.4630814492702484, "learning_rate": 0.0001736271532134851}
+{"ts": "2025-12-27T11:17:04", "event": "train_log", "step": 4236, "epoch": 1.7873417721518987, "progress_pct": 29.79, "epoch_pct": 29.79, "eta": "40:03:49", "max_grad_norm": 1.0, "loss": 0.6649466753005981, "grad_norm": 0.38714799284935, "learning_rate": 0.00017359464399734255}
+{"ts": "2025-12-27T11:17:19", "event": "train_log", "step": 4238, "epoch": 1.7881856540084389, "progress_pct": 29.8, "epoch_pct": 29.8, "eta": "40:02:48", "max_grad_norm": 1.0, "loss": 0.6223326921463013, "grad_norm": 0.38999977707862854, "learning_rate": 0.0001735621178043291}
+{"ts": "2025-12-27T11:17:34", "event": "train_log", "step": 4240, "epoch": 1.7890295358649788, "progress_pct": 29.82, "epoch_pct": 29.82, "eta": "40:01:46", "max_grad_norm": 1.0, "loss": 0.7207517027854919, "grad_norm": 0.4736211597919464, "learning_rate": 0.00017352957464194786}
+{"ts": "2025-12-27T11:17:49", "event": "train_log", "step": 4242, "epoch": 1.789873417721519, "progress_pct": 29.83, "epoch_pct": 29.83, "eta": "40:00:44", "max_grad_norm": 1.0, "loss": 0.7203661203384399, "grad_norm": 0.423476904630661, "learning_rate": 0.00017349701451770588}
+{"ts": "2025-12-27T11:18:03", "event": "train_log", "step": 4244, "epoch": 1.7907172995780591, "progress_pct": 29.85, "epoch_pct": 29.85, "eta": "39:59:41", "max_grad_norm": 1.0, "loss": 0.709998607635498, "grad_norm": 0.4487479627132416, "learning_rate": 0.0001734644374391142}
+{"ts": "2025-12-27T11:18:17", "event": "train_log", "step": 4246, "epoch": 1.791561181434599, "progress_pct": 29.86, "epoch_pct": 29.86, "eta": "39:58:38", "max_grad_norm": 1.0, "loss": 0.7079582214355469, "grad_norm": 0.43721267580986023, "learning_rate": 0.00017343184341368773}
+{"ts": "2025-12-27T11:18:32", "event": "train_log", "step": 4248, "epoch": 1.7924050632911392, "progress_pct": 29.87, "epoch_pct": 29.87, "eta": "39:57:36", "max_grad_norm": 1.0, "loss": 0.6772971749305725, "grad_norm": 0.426864892244339, "learning_rate": 0.00017339923244894524}
+{"ts": "2025-12-27T11:18:48", "event": "train_log", "step": 4250, "epoch": 1.7932489451476794, "progress_pct": 29.89, "epoch_pct": 29.89, "eta": "39:56:37", "max_grad_norm": 1.0, "loss": 0.6882370710372925, "grad_norm": 0.3580208420753479, "learning_rate": 0.0001733666045524095}
+{"ts": "2025-12-27T11:19:02", "event": "train_log", "step": 4252, "epoch": 1.7940928270042193, "progress_pct": 29.9, "epoch_pct": 29.9, "eta": "39:55:34", "max_grad_norm": 1.0, "loss": 0.7534610629081726, "grad_norm": 0.487248957157135, "learning_rate": 0.00017333395973160714}
+{"ts": "2025-12-27T11:19:17", "event": "train_log", "step": 4254, "epoch": 1.7949367088607595, "progress_pct": 29.92, "epoch_pct": 29.92, "eta": "39:54:32", "max_grad_norm": 1.0, "loss": 0.6897269487380981, "grad_norm": 0.44908860325813293, "learning_rate": 0.00017330129799406866}
+{"ts": "2025-12-27T11:19:33", "event": "train_log", "step": 4256, "epoch": 1.7957805907172997, "progress_pct": 29.93, "epoch_pct": 29.93, "eta": "39:53:32", "max_grad_norm": 1.0, "loss": 0.7081538438796997, "grad_norm": 0.43617817759513855, "learning_rate": 0.00017326861934732852}
+{"ts": "2025-12-27T11:19:48", "event": "train_log", "step": 4258, "epoch": 1.7966244725738396, "progress_pct": 29.94, "epoch_pct": 29.94, "eta": "39:52:31", "max_grad_norm": 1.0, "loss": 0.6852753758430481, "grad_norm": 0.428666353225708, "learning_rate": 0.00017323592379892504}
+{"ts": "2025-12-27T11:20:03", "event": "train_log", "step": 4260, "epoch": 1.7974683544303798, "progress_pct": 29.96, "epoch_pct": 29.96, "eta": "39:51:31", "max_grad_norm": 1.0, "loss": 0.7463182210922241, "grad_norm": 0.4522199034690857, "learning_rate": 0.00017320321135640045}
+{"ts": "2025-12-27T11:20:19", "event": "train_log", "step": 4262, "epoch": 1.79831223628692, "progress_pct": 29.97, "epoch_pct": 29.97, "eta": "39:50:32", "max_grad_norm": 1.0, "loss": 0.6643774509429932, "grad_norm": 0.38632410764694214, "learning_rate": 0.00017317048202730088}
+{"ts": "2025-12-27T11:20:34", "event": "train_log", "step": 4264, "epoch": 1.7991561181434599, "progress_pct": 29.99, "epoch_pct": 29.99, "eta": "39:49:32", "max_grad_norm": 1.0, "loss": 0.6726668477058411, "grad_norm": 0.41171637177467346, "learning_rate": 0.00017313773581917642}
+{"ts": "2025-12-27T11:20:50", "event": "train_log", "step": 4266, "epoch": 1.8, "progress_pct": 30.0, "epoch_pct": 30.0, "eta": "39:48:33", "max_grad_norm": 1.0, "loss": 0.6278629302978516, "grad_norm": 0.398520290851593, "learning_rate": 0.0001731049727395809}
+{"ts": "2025-12-27T11:21:06", "event": "train_log", "step": 4268, "epoch": 1.8008438818565402, "progress_pct": 30.01, "epoch_pct": 30.01, "eta": "39:47:34", "max_grad_norm": 1.0, "loss": 0.5823352336883545, "grad_norm": 0.4000356197357178, "learning_rate": 0.0001730721927960722}
+{"ts": "2025-12-27T11:21:20", "event": "train_log", "step": 4270, "epoch": 1.8016877637130801, "progress_pct": 30.03, "epoch_pct": 30.03, "eta": "39:46:31", "max_grad_norm": 1.0, "loss": 0.7492824196815491, "grad_norm": 0.47376710176467896, "learning_rate": 0.000173039395996212}
+{"ts": "2025-12-27T11:21:35", "event": "train_log", "step": 4272, "epoch": 1.80253164556962, "progress_pct": 30.04, "epoch_pct": 30.04, "eta": "39:45:30", "max_grad_norm": 1.0, "loss": 0.7630324363708496, "grad_norm": 0.4624378979206085, "learning_rate": 0.00017300658234756588}
+{"ts": "2025-12-27T11:21:50", "event": "train_log", "step": 4274, "epoch": 1.8033755274261605, "progress_pct": 30.06, "epoch_pct": 30.06, "eta": "39:44:28", "max_grad_norm": 1.0, "loss": 0.6484803557395935, "grad_norm": 0.4178096354007721, "learning_rate": 0.00017297375185770336}
+{"ts": "2025-12-27T11:22:04", "event": "train_log", "step": 4276, "epoch": 1.8042194092827004, "progress_pct": 30.07, "epoch_pct": 30.07, "eta": "39:43:27", "max_grad_norm": 1.0, "loss": 0.776732325553894, "grad_norm": 0.511215329170227, "learning_rate": 0.00017294090453419774}
+{"ts": "2025-12-27T11:22:19", "event": "train_log", "step": 4278, "epoch": 1.8050632911392404, "progress_pct": 30.08, "epoch_pct": 30.08, "eta": "39:42:24", "max_grad_norm": 1.0, "loss": 0.6801077723503113, "grad_norm": 0.44643548130989075, "learning_rate": 0.00017290804038462633}
+{"ts": "2025-12-27T11:22:33", "event": "train_log", "step": 4280, "epoch": 1.8059071729957807, "progress_pct": 30.1, "epoch_pct": 30.1, "eta": "39:41:22", "max_grad_norm": 1.0, "loss": 0.8024200797080994, "grad_norm": 0.5090142488479614, "learning_rate": 0.0001728751594165702}
+{"ts": "2025-12-27T11:22:47", "event": "train_log", "step": 4282, "epoch": 1.8067510548523207, "progress_pct": 30.11, "epoch_pct": 30.11, "eta": "39:40:19", "max_grad_norm": 1.0, "loss": 0.7810231447219849, "grad_norm": 0.45888981223106384, "learning_rate": 0.00017284226163761436}
+{"ts": "2025-12-27T11:23:02", "event": "train_log", "step": 4284, "epoch": 1.8075949367088606, "progress_pct": 30.13, "epoch_pct": 30.13, "eta": "39:39:19", "max_grad_norm": 1.0, "loss": 0.6768189072608948, "grad_norm": 0.3985820710659027, "learning_rate": 0.0001728093470553478}
+{"ts": "2025-12-27T11:23:18", "event": "train_log", "step": 4286, "epoch": 1.808438818565401, "progress_pct": 30.14, "epoch_pct": 30.14, "eta": "39:38:19", "max_grad_norm": 1.0, "loss": 0.6093943119049072, "grad_norm": 0.3988003730773926, "learning_rate": 0.00017277641567736316}
+{"ts": "2025-12-27T11:23:33", "event": "train_log", "step": 4288, "epoch": 1.809282700421941, "progress_pct": 30.15, "epoch_pct": 30.15, "eta": "39:37:18", "max_grad_norm": 1.0, "loss": 0.6836619973182678, "grad_norm": 0.46131962537765503, "learning_rate": 0.0001727434675112571}
+{"ts": "2025-12-27T11:23:47", "event": "train_log", "step": 4290, "epoch": 1.810126582278481, "progress_pct": 30.17, "epoch_pct": 30.17, "eta": "39:36:17", "max_grad_norm": 1.0, "loss": 0.7144147157669067, "grad_norm": 0.42959487438201904, "learning_rate": 0.0001727105025646302}
+{"ts": "2025-12-27T11:24:02", "event": "train_log", "step": 4292, "epoch": 1.810970464135021, "progress_pct": 30.18, "epoch_pct": 30.18, "eta": "39:35:17", "max_grad_norm": 1.0, "loss": 0.7190625667572021, "grad_norm": 0.4110259413719177, "learning_rate": 0.00017267752084508675}
+{"ts": "2025-12-27T11:24:17", "event": "train_log", "step": 4294, "epoch": 1.8118143459915612, "progress_pct": 30.2, "epoch_pct": 30.2, "eta": "39:34:17", "max_grad_norm": 1.0, "loss": 0.7272495031356812, "grad_norm": 0.4501377046108246, "learning_rate": 0.00017264452236023505}
+{"ts": "2025-12-27T11:24:32", "event": "train_log", "step": 4296, "epoch": 1.8126582278481012, "progress_pct": 30.21, "epoch_pct": 30.21, "eta": "39:33:15", "max_grad_norm": 1.0, "loss": 0.758166491985321, "grad_norm": 0.467817097902298, "learning_rate": 0.0001726115071176872}
+{"ts": "2025-12-27T11:24:47", "event": "train_log", "step": 4298, "epoch": 1.8135021097046413, "progress_pct": 30.23, "epoch_pct": 30.23, "eta": "39:32:15", "max_grad_norm": 1.0, "loss": 0.6712231040000916, "grad_norm": 0.42394939064979553, "learning_rate": 0.0001725784751250592}
+{"ts": "2025-12-27T11:25:02", "event": "train_log", "step": 4300, "epoch": 1.8143459915611815, "progress_pct": 30.24, "epoch_pct": 30.24, "eta": "39:31:15", "max_grad_norm": 1.0, "loss": 0.5999719500541687, "grad_norm": 0.37668925523757935, "learning_rate": 0.00017254542638997088}
+{"ts": "2025-12-27T11:36:30", "event": "train_log", "step": 4300, "epoch": 1.8143459915611815, "progress_pct": 30.24, "epoch_pct": 30.24, "eta": "39:57:43", "max_grad_norm": 1.0, "eval_loss": 0.7672964930534363, "eval_runtime": 688.4249, "eval_samples_per_second": 3.061, "eval_steps_per_second": 3.061}
+{"ts": "2025-12-27T11:36:47", "event": "train_log", "step": 4302, "epoch": 1.8151898734177214, "progress_pct": 30.25, "epoch_pct": 30.25, "eta": "39:56:45", "max_grad_norm": 1.0, "loss": 0.7188607454299927, "grad_norm": 0.44964465498924255, "learning_rate": 0.00017251236092004596}
+{"ts": "2025-12-27T11:37:02", "event": "train_log", "step": 4304, "epoch": 1.8160337552742616, "progress_pct": 30.27, "epoch_pct": 30.27, "eta": "39:55:45", "max_grad_norm": 1.0, "loss": 0.707834005355835, "grad_norm": 0.4108492434024811, "learning_rate": 0.000172479278722912}
+{"ts": "2025-12-27T11:37:18", "event": "train_log", "step": 4306, "epoch": 1.8168776371308017, "progress_pct": 30.28, "epoch_pct": 30.28, "eta": "39:54:45", "max_grad_norm": 1.0, "loss": 0.6915075182914734, "grad_norm": 0.4455305337905884, "learning_rate": 0.00017244617980620043}
+{"ts": "2025-12-27T11:37:35", "event": "train_log", "step": 4308, "epoch": 1.8177215189873417, "progress_pct": 30.3, "epoch_pct": 30.3, "eta": "39:53:48", "max_grad_norm": 1.0, "loss": 0.6591874361038208, "grad_norm": 0.3977123200893402, "learning_rate": 0.00017241306417754655}
+{"ts": "2025-12-27T11:37:49", "event": "train_log", "step": 4310, "epoch": 1.8185654008438819, "progress_pct": 30.31, "epoch_pct": 30.31, "eta": "39:52:45", "max_grad_norm": 1.0, "loss": 0.6254785060882568, "grad_norm": 0.4028434157371521, "learning_rate": 0.00017237993184458946}
+{"ts": "2025-12-27T11:38:05", "event": "train_log", "step": 4312, "epoch": 1.819409282700422, "progress_pct": 30.32, "epoch_pct": 30.32, "eta": "39:51:46", "max_grad_norm": 1.0, "loss": 0.6968662142753601, "grad_norm": 0.41024482250213623, "learning_rate": 0.00017234678281497217}
+{"ts": "2025-12-27T11:38:22", "event": "train_log", "step": 4314, "epoch": 1.820253164556962, "progress_pct": 30.34, "epoch_pct": 30.34, "eta": "39:50:50", "max_grad_norm": 1.0, "loss": 0.6535647511482239, "grad_norm": 0.45941361784935, "learning_rate": 0.00017231361709634153}
+{"ts": "2025-12-27T11:38:38", "event": "train_log", "step": 4316, "epoch": 1.8210970464135021, "progress_pct": 30.35, "epoch_pct": 30.35, "eta": "39:49:52", "max_grad_norm": 1.0, "loss": 0.6378563642501831, "grad_norm": 0.39477071166038513, "learning_rate": 0.00017228043469634825}
+{"ts": "2025-12-27T11:38:54", "event": "train_log", "step": 4318, "epoch": 1.8219409282700423, "progress_pct": 30.37, "epoch_pct": 30.37, "eta": "39:48:53", "max_grad_norm": 1.0, "loss": 0.7021255493164062, "grad_norm": 0.403826504945755, "learning_rate": 0.00017224723562264687}
+{"ts": "2025-12-27T11:39:09", "event": "train_log", "step": 4320, "epoch": 1.8227848101265822, "progress_pct": 30.38, "epoch_pct": 30.38, "eta": "39:47:52", "max_grad_norm": 1.0, "loss": 0.7411118745803833, "grad_norm": 0.4179201126098633, "learning_rate": 0.00017221401988289575}
+{"ts": "2025-12-27T11:39:25", "event": "train_log", "step": 4322, "epoch": 1.8236286919831224, "progress_pct": 30.39, "epoch_pct": 30.39, "eta": "39:46:53", "max_grad_norm": 1.0, "loss": 0.7435210943222046, "grad_norm": 0.4719425141811371, "learning_rate": 0.0001721807874847571}
+{"ts": "2025-12-27T11:39:41", "event": "train_log", "step": 4324, "epoch": 1.8244725738396625, "progress_pct": 30.41, "epoch_pct": 30.41, "eta": "39:45:54", "max_grad_norm": 1.0, "loss": 0.6813794374465942, "grad_norm": 0.39946448802948, "learning_rate": 0.0001721475384358971}
+{"ts": "2025-12-27T11:39:57", "event": "train_log", "step": 4326, "epoch": 1.8253164556962025, "progress_pct": 30.42, "epoch_pct": 30.42, "eta": "39:44:57", "max_grad_norm": 1.0, "loss": 0.6468828320503235, "grad_norm": 0.4367005527019501, "learning_rate": 0.00017211427274398559}
+{"ts": "2025-12-27T11:40:14", "event": "train_log", "step": 4328, "epoch": 1.8261603375527427, "progress_pct": 30.44, "epoch_pct": 30.44, "eta": "39:43:59", "max_grad_norm": 1.0, "loss": 0.6628729701042175, "grad_norm": 0.4078138470649719, "learning_rate": 0.00017208099041669632}
+{"ts": "2025-12-27T11:40:29", "event": "train_log", "step": 4330, "epoch": 1.8270042194092828, "progress_pct": 30.45, "epoch_pct": 30.45, "eta": "39:42:59", "max_grad_norm": 1.0, "loss": 0.7028043866157532, "grad_norm": 0.4141235649585724, "learning_rate": 0.0001720476914617069}
+{"ts": "2025-12-27T11:40:44", "event": "train_log", "step": 4332, "epoch": 1.8278481012658228, "progress_pct": 30.46, "epoch_pct": 30.46, "eta": "39:41:59", "max_grad_norm": 1.0, "loss": 0.6710682511329651, "grad_norm": 0.448733925819397, "learning_rate": 0.00017201437588669878}
+{"ts": "2025-12-27T11:41:00", "event": "train_log", "step": 4334, "epoch": 1.828691983122363, "progress_pct": 30.48, "epoch_pct": 30.48, "eta": "39:41:01", "max_grad_norm": 1.0, "loss": 0.792453408241272, "grad_norm": 0.5084340572357178, "learning_rate": 0.00017198104369935718}
+{"ts": "2025-12-27T11:41:14", "event": "train_log", "step": 4336, "epoch": 1.829535864978903, "progress_pct": 30.49, "epoch_pct": 30.49, "eta": "39:39:58", "max_grad_norm": 1.0, "loss": 0.7536583542823792, "grad_norm": 0.43463611602783203, "learning_rate": 0.00017194769490737124}
+{"ts": "2025-12-27T11:41:30", "event": "train_log", "step": 4338, "epoch": 1.830379746835443, "progress_pct": 30.51, "epoch_pct": 30.51, "eta": "39:39:00", "max_grad_norm": 1.0, "loss": 0.7445555925369263, "grad_norm": 0.46039044857025146, "learning_rate": 0.00017191432951843387}
+{"ts": "2025-12-27T11:41:46", "event": "train_log", "step": 4340, "epoch": 1.831223628691983, "progress_pct": 30.52, "epoch_pct": 30.52, "eta": "39:38:00", "max_grad_norm": 1.0, "loss": 0.7003044486045837, "grad_norm": 0.3894030451774597, "learning_rate": 0.0001718809475402418}
+{"ts": "2025-12-27T11:42:00", "event": "train_log", "step": 4342, "epoch": 1.8320675105485233, "progress_pct": 30.53, "epoch_pct": 30.53, "eta": "39:36:57", "max_grad_norm": 1.0, "loss": 0.6948950886726379, "grad_norm": 0.5851747393608093, "learning_rate": 0.00017184754898049568}
+{"ts": "2025-12-27T11:42:15", "event": "train_log", "step": 4344, "epoch": 1.8329113924050633, "progress_pct": 30.55, "epoch_pct": 30.55, "eta": "39:35:57", "max_grad_norm": 1.0, "loss": 0.675811231136322, "grad_norm": 0.40906789898872375, "learning_rate": 0.00017181413384689982}
+{"ts": "2025-12-27T11:42:30", "event": "train_log", "step": 4346, "epoch": 1.8337552742616032, "progress_pct": 30.56, "epoch_pct": 30.56, "eta": "39:34:57", "max_grad_norm": 1.0, "loss": 0.7422580122947693, "grad_norm": 0.4235765039920807, "learning_rate": 0.0001717807021471625}
+{"ts": "2025-12-27T11:42:45", "event": "train_log", "step": 4348, "epoch": 1.8345991561181436, "progress_pct": 30.58, "epoch_pct": 30.58, "eta": "39:33:58", "max_grad_norm": 1.0, "loss": 0.7014731168746948, "grad_norm": 0.3898368775844574, "learning_rate": 0.0001717472538889958}
+{"ts": "2025-12-27T11:43:00", "event": "train_log", "step": 4350, "epoch": 1.8354430379746836, "progress_pct": 30.59, "epoch_pct": 30.59, "eta": "39:32:57", "max_grad_norm": 1.0, "loss": 0.6952776312828064, "grad_norm": 0.4122314751148224, "learning_rate": 0.00017171378908011553}
+{"ts": "2025-12-27T11:43:15", "event": "train_log", "step": 4352, "epoch": 1.8362869198312235, "progress_pct": 30.6, "epoch_pct": 30.6, "eta": "39:31:56", "max_grad_norm": 1.0, "loss": 0.6758598685264587, "grad_norm": 0.38085612654685974, "learning_rate": 0.00017168030772824144}
+{"ts": "2025-12-27T11:43:29", "event": "train_log", "step": 4354, "epoch": 1.8371308016877637, "progress_pct": 30.62, "epoch_pct": 30.62, "eta": "39:30:54", "max_grad_norm": 1.0, "loss": 0.7139424085617065, "grad_norm": 0.43519729375839233, "learning_rate": 0.00017164680984109695}
+{"ts": "2025-12-27T11:43:44", "event": "train_log", "step": 4356, "epoch": 1.8379746835443038, "progress_pct": 30.63, "epoch_pct": 30.63, "eta": "39:29:54", "max_grad_norm": 1.0, "loss": 0.7125198841094971, "grad_norm": 0.4404325783252716, "learning_rate": 0.0001716132954264095}
+{"ts": "2025-12-27T11:43:59", "event": "train_log", "step": 4358, "epoch": 1.8388185654008438, "progress_pct": 30.65, "epoch_pct": 30.65, "eta": "39:28:53", "max_grad_norm": 1.0, "loss": 0.7011314630508423, "grad_norm": 0.3965512812137604, "learning_rate": 0.00017157976449191012}
+{"ts": "2025-12-27T11:44:13", "event": "train_log", "step": 4360, "epoch": 1.839662447257384, "progress_pct": 30.66, "epoch_pct": 30.66, "eta": "39:27:52", "max_grad_norm": 1.0, "loss": 0.6485645174980164, "grad_norm": 0.4273630976676941, "learning_rate": 0.0001715462170453338}
+{"ts": "2025-12-27T11:44:28", "event": "train_log", "step": 4362, "epoch": 1.840506329113924, "progress_pct": 30.68, "epoch_pct": 30.68, "eta": "39:26:52", "max_grad_norm": 1.0, "loss": 0.6669548153877258, "grad_norm": 0.4639208912849426, "learning_rate": 0.00017151265309441925}
+{"ts": "2025-12-27T11:44:43", "event": "train_log", "step": 4364, "epoch": 1.841350210970464, "progress_pct": 30.69, "epoch_pct": 30.69, "eta": "39:25:52", "max_grad_norm": 1.0, "loss": 0.6276374459266663, "grad_norm": 0.4190906286239624, "learning_rate": 0.00017147907264690907}
+{"ts": "2025-12-27T11:44:59", "event": "train_log", "step": 4366, "epoch": 1.8421940928270042, "progress_pct": 30.7, "epoch_pct": 30.7, "eta": "39:24:53", "max_grad_norm": 1.0, "loss": 0.5975915193557739, "grad_norm": 0.4146515130996704, "learning_rate": 0.00017144547571054962}
+{"ts": "2025-12-27T11:45:14", "event": "train_log", "step": 4368, "epoch": 1.8430379746835444, "progress_pct": 30.72, "epoch_pct": 30.72, "eta": "39:23:53", "max_grad_norm": 1.0, "loss": 0.6536507606506348, "grad_norm": 0.38717755675315857, "learning_rate": 0.00017141186229309105}
+{"ts": "2025-12-27T11:45:29", "event": "train_log", "step": 4370, "epoch": 1.8438818565400843, "progress_pct": 30.73, "epoch_pct": 30.73, "eta": "39:22:54", "max_grad_norm": 1.0, "loss": 0.7078962326049805, "grad_norm": 0.3988541066646576, "learning_rate": 0.00017137823240228735}
+{"ts": "2025-12-27T11:45:44", "event": "train_log", "step": 4372, "epoch": 1.8447257383966245, "progress_pct": 30.75, "epoch_pct": 30.75, "eta": "39:21:54", "max_grad_norm": 1.0, "loss": 0.6795231699943542, "grad_norm": 0.43783870339393616, "learning_rate": 0.00017134458604589628}
+{"ts": "2025-12-27T11:45:59", "event": "train_log", "step": 4374, "epoch": 1.8455696202531646, "progress_pct": 30.76, "epoch_pct": 30.76, "eta": "39:20:53", "max_grad_norm": 1.0, "loss": 0.7030490040779114, "grad_norm": 0.439636766910553, "learning_rate": 0.00017131092323167938}
+{"ts": "2025-12-27T11:46:14", "event": "train_log", "step": 4376, "epoch": 1.8464135021097046, "progress_pct": 30.77, "epoch_pct": 30.77, "eta": "39:19:54", "max_grad_norm": 1.0, "loss": 0.7611671686172485, "grad_norm": 0.4426078498363495, "learning_rate": 0.00017127724396740206}
+{"ts": "2025-12-27T11:46:29", "event": "train_log", "step": 4378, "epoch": 1.8472573839662447, "progress_pct": 30.79, "epoch_pct": 30.79, "eta": "39:18:54", "max_grad_norm": 1.0, "loss": 0.737910270690918, "grad_norm": 0.40058812499046326, "learning_rate": 0.00017124354826083348}
+{"ts": "2025-12-27T11:46:44", "event": "train_log", "step": 4380, "epoch": 1.8481012658227849, "progress_pct": 30.8, "epoch_pct": 30.8, "eta": "39:17:54", "max_grad_norm": 1.0, "loss": 0.683937132358551, "grad_norm": 0.43037283420562744, "learning_rate": 0.00017120983611974654}
+{"ts": "2025-12-27T11:46:58", "event": "train_log", "step": 4382, "epoch": 1.8489451476793248, "progress_pct": 30.82, "epoch_pct": 30.82, "eta": "39:16:54", "max_grad_norm": 1.0, "loss": 0.6987335681915283, "grad_norm": 0.4479113221168518, "learning_rate": 0.00017117610755191807}
+{"ts": "2025-12-27T11:47:14", "event": "train_log", "step": 4384, "epoch": 1.849789029535865, "progress_pct": 30.83, "epoch_pct": 30.83, "eta": "39:15:55", "max_grad_norm": 1.0, "loss": 0.6964257955551147, "grad_norm": 0.4218229651451111, "learning_rate": 0.0001711423625651285}
+{"ts": "2025-12-27T11:47:29", "event": "train_log", "step": 4386, "epoch": 1.8506329113924052, "progress_pct": 30.84, "epoch_pct": 30.84, "eta": "39:14:56", "max_grad_norm": 1.0, "loss": 0.6983811855316162, "grad_norm": 0.397733211517334, "learning_rate": 0.00017110860116716222}
+{"ts": "2025-12-27T11:47:43", "event": "train_log", "step": 4388, "epoch": 1.851476793248945, "progress_pct": 30.86, "epoch_pct": 30.86, "eta": "39:13:55", "max_grad_norm": 1.0, "loss": 0.7579324245452881, "grad_norm": 0.4844493865966797, "learning_rate": 0.00017107482336580735}
+{"ts": "2025-12-27T11:47:58", "event": "train_log", "step": 4390, "epoch": 1.8523206751054853, "progress_pct": 30.87, "epoch_pct": 30.87, "eta": "39:12:54", "max_grad_norm": 1.0, "loss": 0.703558623790741, "grad_norm": 0.4494120478630066, "learning_rate": 0.0001710410291688557}
+{"ts": "2025-12-27T11:48:12", "event": "train_log", "step": 4392, "epoch": 1.8531645569620254, "progress_pct": 30.89, "epoch_pct": 30.89, "eta": "39:11:54", "max_grad_norm": 1.0, "loss": 0.7386981248855591, "grad_norm": 0.4721727967262268, "learning_rate": 0.000171007218584103}
+{"ts": "2025-12-27T11:48:26", "event": "train_log", "step": 4394, "epoch": 1.8540084388185654, "progress_pct": 30.9, "epoch_pct": 30.9, "eta": "39:10:52", "max_grad_norm": 1.0, "loss": 0.7769017219543457, "grad_norm": 0.4816451370716095, "learning_rate": 0.0001709733916193487}
+{"ts": "2025-12-27T11:48:41", "event": "train_log", "step": 4396, "epoch": 1.8548523206751055, "progress_pct": 30.91, "epoch_pct": 30.91, "eta": "39:09:52", "max_grad_norm": 1.0, "loss": 0.7315034866333008, "grad_norm": 0.43638876080513, "learning_rate": 0.00017093954828239602}
+{"ts": "2025-12-27T11:48:57", "event": "train_log", "step": 4398, "epoch": 1.8556962025316457, "progress_pct": 30.93, "epoch_pct": 30.93, "eta": "39:08:54", "max_grad_norm": 1.0, "loss": 0.7289890050888062, "grad_norm": 0.4035055935382843, "learning_rate": 0.000170905688581052}
+{"ts": "2025-12-27T11:49:11", "event": "train_log", "step": 4400, "epoch": 1.8565400843881856, "progress_pct": 30.94, "epoch_pct": 30.94, "eta": "39:07:54", "max_grad_norm": 1.0, "loss": 0.7190408110618591, "grad_norm": 0.44265708327293396, "learning_rate": 0.00017087181252312735}
+{"ts": "2025-12-27T12:00:29", "event": "train_log", "step": 4400, "epoch": 1.8565400843881856, "progress_pct": 30.94, "epoch_pct": 30.94, "eta": "39:33:08", "max_grad_norm": 1.0, "eval_loss": 0.7635221481323242, "eval_runtime": 678.243, "eval_samples_per_second": 3.107, "eval_steps_per_second": 3.107}
+{"ts": "2025-12-27T12:00:47", "event": "train_log", "step": 4402, "epoch": 1.8573839662447258, "progress_pct": 30.96, "epoch_pct": 30.96, "eta": "39:32:14", "max_grad_norm": 1.0, "loss": 0.6318185925483704, "grad_norm": 0.39691466093063354, "learning_rate": 0.0001708379201164367}
+{"ts": "2025-12-27T12:01:03", "event": "train_log", "step": 4404, "epoch": 1.858227848101266, "progress_pct": 30.97, "epoch_pct": 30.97, "eta": "39:31:14", "max_grad_norm": 1.0, "loss": 0.7040913105010986, "grad_norm": 0.47801393270492554, "learning_rate": 0.0001708040113687983}
+{"ts": "2025-12-27T12:01:18", "event": "train_log", "step": 4406, "epoch": 1.859071729957806, "progress_pct": 30.98, "epoch_pct": 30.98, "eta": "39:30:15", "max_grad_norm": 1.0, "loss": 0.7834659218788147, "grad_norm": 0.47457239031791687, "learning_rate": 0.00017077008628803436}
+{"ts": "2025-12-27T12:01:33", "event": "train_log", "step": 4408, "epoch": 1.8599156118143458, "progress_pct": 31.0, "epoch_pct": 31.0, "eta": "39:29:14", "max_grad_norm": 1.0, "loss": 0.7027983665466309, "grad_norm": 0.41362807154655457, "learning_rate": 0.00017073614488197065}
+{"ts": "2025-12-27T12:01:47", "event": "train_log", "step": 4410, "epoch": 1.8607594936708862, "progress_pct": 31.01, "epoch_pct": 31.01, "eta": "39:28:14", "max_grad_norm": 1.0, "loss": 0.7143591642379761, "grad_norm": 0.457927942276001, "learning_rate": 0.0001707021871584368}
+{"ts": "2025-12-27T12:02:03", "event": "train_log", "step": 4412, "epoch": 1.8616033755274262, "progress_pct": 31.03, "epoch_pct": 31.03, "eta": "39:27:16", "max_grad_norm": 1.0, "loss": 0.7328225374221802, "grad_norm": 0.42931967973709106, "learning_rate": 0.00017066821312526625}
+{"ts": "2025-12-27T12:02:18", "event": "train_log", "step": 4414, "epoch": 1.862447257383966, "progress_pct": 31.04, "epoch_pct": 31.04, "eta": "39:26:16", "max_grad_norm": 1.0, "loss": 0.7199252843856812, "grad_norm": 0.3821638524532318, "learning_rate": 0.00017063422279029616}
+{"ts": "2025-12-27T12:02:35", "event": "train_log", "step": 4416, "epoch": 1.8632911392405065, "progress_pct": 31.05, "epoch_pct": 31.05, "eta": "39:25:19", "max_grad_norm": 1.0, "loss": 0.7345157265663147, "grad_norm": 0.3954671621322632, "learning_rate": 0.00017060021616136742}
+{"ts": "2025-12-27T12:02:51", "event": "train_log", "step": 4418, "epoch": 1.8641350210970464, "progress_pct": 31.07, "epoch_pct": 31.07, "eta": "39:24:21", "max_grad_norm": 1.0, "loss": 0.587895393371582, "grad_norm": 0.38824960589408875, "learning_rate": 0.0001705661932463247}
+{"ts": "2025-12-27T12:03:06", "event": "train_log", "step": 4420, "epoch": 1.8649789029535864, "progress_pct": 31.08, "epoch_pct": 31.08, "eta": "39:23:22", "max_grad_norm": 1.0, "loss": 0.7227945327758789, "grad_norm": 0.510923445224762, "learning_rate": 0.00017053215405301648}
+{"ts": "2025-12-27T12:03:23", "event": "train_log", "step": 4422, "epoch": 1.8658227848101265, "progress_pct": 31.1, "epoch_pct": 31.1, "eta": "39:22:27", "max_grad_norm": 1.0, "loss": 0.6605637073516846, "grad_norm": 0.38688936829566956, "learning_rate": 0.0001704980985892949}
+{"ts": "2025-12-27T12:03:37", "event": "train_log", "step": 4424, "epoch": 1.8666666666666667, "progress_pct": 31.11, "epoch_pct": 31.11, "eta": "39:21:24", "max_grad_norm": 1.0, "loss": 0.7581667304039001, "grad_norm": 0.4886745512485504, "learning_rate": 0.0001704640268630159}
+{"ts": "2025-12-27T12:03:53", "event": "train_log", "step": 4426, "epoch": 1.8675105485232066, "progress_pct": 31.13, "epoch_pct": 31.13, "eta": "39:20:27", "max_grad_norm": 1.0, "loss": 0.7051677107810974, "grad_norm": 0.4214446246623993, "learning_rate": 0.0001704299388820392}
+{"ts": "2025-12-27T12:04:09", "event": "train_log", "step": 4428, "epoch": 1.8683544303797468, "progress_pct": 31.14, "epoch_pct": 31.14, "eta": "39:19:29", "max_grad_norm": 1.0, "loss": 0.6489346623420715, "grad_norm": 0.4189293384552002, "learning_rate": 0.00017039583465422824}
+{"ts": "2025-12-27T12:04:24", "event": "train_log", "step": 4430, "epoch": 1.869198312236287, "progress_pct": 31.15, "epoch_pct": 31.15, "eta": "39:18:30", "max_grad_norm": 1.0, "loss": 0.6896629929542542, "grad_norm": 0.4264903962612152, "learning_rate": 0.0001703617141874502}
+{"ts": "2025-12-27T12:04:40", "event": "train_log", "step": 4432, "epoch": 1.870042194092827, "progress_pct": 31.17, "epoch_pct": 31.17, "eta": "39:17:32", "max_grad_norm": 1.0, "loss": 0.7468283772468567, "grad_norm": 0.5351413488388062, "learning_rate": 0.00017032757748957603}
+{"ts": "2025-12-27T12:04:56", "event": "train_log", "step": 4434, "epoch": 1.870886075949367, "progress_pct": 31.18, "epoch_pct": 31.18, "eta": "39:16:35", "max_grad_norm": 1.0, "loss": 0.6029976010322571, "grad_norm": 0.40163540840148926, "learning_rate": 0.00017029342456848035}
+{"ts": "2025-12-27T12:05:11", "event": "train_log", "step": 4436, "epoch": 1.8717299578059072, "progress_pct": 31.2, "epoch_pct": 31.2, "eta": "39:15:37", "max_grad_norm": 1.0, "loss": 0.6891162991523743, "grad_norm": 0.3786795139312744, "learning_rate": 0.0001702592554320417}
+{"ts": "2025-12-27T12:05:27", "event": "train_log", "step": 4438, "epoch": 1.8725738396624472, "progress_pct": 31.21, "epoch_pct": 31.21, "eta": "39:14:37", "max_grad_norm": 1.0, "loss": 0.6189853549003601, "grad_norm": 0.4266849458217621, "learning_rate": 0.00017022507008814214}
+{"ts": "2025-12-27T12:05:41", "event": "train_log", "step": 4440, "epoch": 1.8734177215189873, "progress_pct": 31.22, "epoch_pct": 31.22, "eta": "39:13:36", "max_grad_norm": 1.0, "loss": 0.6858769655227661, "grad_norm": 0.4076935052871704, "learning_rate": 0.0001701908685446676}
+{"ts": "2025-12-27T12:05:57", "event": "train_log", "step": 4442, "epoch": 1.8742616033755275, "progress_pct": 31.24, "epoch_pct": 31.24, "eta": "39:12:39", "max_grad_norm": 1.0, "loss": 0.6650485396385193, "grad_norm": 0.4209558665752411, "learning_rate": 0.00017015665080950774}
+{"ts": "2025-12-27T12:06:14", "event": "train_log", "step": 4444, "epoch": 1.8751054852320674, "progress_pct": 31.25, "epoch_pct": 31.25, "eta": "39:11:44", "max_grad_norm": 1.0, "loss": 0.6887878179550171, "grad_norm": 0.4114036560058594, "learning_rate": 0.0001701224168905559}
+{"ts": "2025-12-27T12:06:30", "event": "train_log", "step": 4446, "epoch": 1.8759493670886076, "progress_pct": 31.27, "epoch_pct": 31.27, "eta": "39:10:46", "max_grad_norm": 1.0, "loss": 0.6950712203979492, "grad_norm": 0.34596455097198486, "learning_rate": 0.00017008816679570926}
+{"ts": "2025-12-27T12:06:46", "event": "train_log", "step": 4448, "epoch": 1.8767932489451478, "progress_pct": 31.28, "epoch_pct": 31.28, "eta": "39:09:50", "max_grad_norm": 1.0, "loss": 0.7360581159591675, "grad_norm": 0.45132747292518616, "learning_rate": 0.0001700539005328686}
+{"ts": "2025-12-27T12:07:03", "event": "train_log", "step": 4450, "epoch": 1.8776371308016877, "progress_pct": 31.29, "epoch_pct": 31.29, "eta": "39:08:55", "max_grad_norm": 1.0, "loss": 0.6672278046607971, "grad_norm": 0.4258243441581726, "learning_rate": 0.00017001961810993846}
+{"ts": "2025-12-27T12:07:18", "event": "train_log", "step": 4452, "epoch": 1.8784810126582279, "progress_pct": 31.31, "epoch_pct": 31.31, "eta": "39:07:55", "max_grad_norm": 1.0, "loss": 0.6740089654922485, "grad_norm": 0.39144474267959595, "learning_rate": 0.0001699853195348272}
+{"ts": "2025-12-27T12:07:33", "event": "train_log", "step": 4454, "epoch": 1.879324894514768, "progress_pct": 31.32, "epoch_pct": 31.32, "eta": "39:06:58", "max_grad_norm": 1.0, "loss": 0.7747141122817993, "grad_norm": 0.42183035612106323, "learning_rate": 0.00016995100481544681}
+{"ts": "2025-12-27T12:07:50", "event": "train_log", "step": 4456, "epoch": 1.880168776371308, "progress_pct": 31.34, "epoch_pct": 31.34, "eta": "39:06:02", "max_grad_norm": 1.0, "loss": 0.7189701199531555, "grad_norm": 0.4110005795955658, "learning_rate": 0.00016991667395971303}
+{"ts": "2025-12-27T12:08:06", "event": "train_log", "step": 4458, "epoch": 1.8810126582278481, "progress_pct": 31.35, "epoch_pct": 31.35, "eta": "39:05:05", "max_grad_norm": 1.0, "loss": 0.6880105137825012, "grad_norm": 0.4073067307472229, "learning_rate": 0.00016988232697554538}
+{"ts": "2025-12-27T12:08:22", "event": "train_log", "step": 4460, "epoch": 1.8818565400843883, "progress_pct": 31.36, "epoch_pct": 31.36, "eta": "39:04:07", "max_grad_norm": 1.0, "loss": 0.6994736790657043, "grad_norm": 0.467443585395813, "learning_rate": 0.00016984796387086698}
+{"ts": "2025-12-27T12:08:37", "event": "train_log", "step": 4462, "epoch": 1.8827004219409282, "progress_pct": 31.38, "epoch_pct": 31.38, "eta": "39:03:08", "max_grad_norm": 1.0, "loss": 0.7389267683029175, "grad_norm": 0.4769018292427063, "learning_rate": 0.00016981358465360482}
+{"ts": "2025-12-27T12:08:53", "event": "train_log", "step": 4464, "epoch": 1.8835443037974684, "progress_pct": 31.39, "epoch_pct": 31.39, "eta": "39:02:13", "max_grad_norm": 1.0, "loss": 0.6375740170478821, "grad_norm": 0.3834097981452942, "learning_rate": 0.00016977918933168938}
+{"ts": "2025-12-27T12:09:09", "event": "train_log", "step": 4466, "epoch": 1.8843881856540086, "progress_pct": 31.41, "epoch_pct": 31.41, "eta": "39:01:15", "max_grad_norm": 1.0, "loss": 0.6845629811286926, "grad_norm": 0.44548940658569336, "learning_rate": 0.00016974477791305517}
+{"ts": "2025-12-27T12:09:24", "event": "train_log", "step": 4468, "epoch": 1.8852320675105485, "progress_pct": 31.42, "epoch_pct": 31.42, "eta": "39:00:16", "max_grad_norm": 1.0, "loss": 0.7307721376419067, "grad_norm": 0.39344266057014465, "learning_rate": 0.00016971035040564016}
+{"ts": "2025-12-27T12:09:40", "event": "train_log", "step": 4470, "epoch": 1.8860759493670884, "progress_pct": 31.43, "epoch_pct": 31.43, "eta": "38:59:20", "max_grad_norm": 1.0, "loss": 0.6733962297439575, "grad_norm": 0.37901216745376587, "learning_rate": 0.0001696759068173861}
+{"ts": "2025-12-27T12:09:55", "event": "train_log", "step": 4472, "epoch": 1.8869198312236288, "progress_pct": 31.45, "epoch_pct": 31.45, "eta": "38:58:21", "max_grad_norm": 1.0, "loss": 0.7294619679450989, "grad_norm": 0.44465333223342896, "learning_rate": 0.00016964144715623852}
+{"ts": "2025-12-27T12:10:11", "event": "train_log", "step": 4474, "epoch": 1.8877637130801688, "progress_pct": 31.46, "epoch_pct": 31.46, "eta": "38:57:24", "max_grad_norm": 1.0, "loss": 0.6537288427352905, "grad_norm": 0.39566636085510254, "learning_rate": 0.00016960697143014657}
+{"ts": "2025-12-27T12:10:27", "event": "train_log", "step": 4476, "epoch": 1.8886075949367087, "progress_pct": 31.48, "epoch_pct": 31.48, "eta": "38:56:28", "max_grad_norm": 1.0, "loss": 0.623903751373291, "grad_norm": 0.4047028124332428, "learning_rate": 0.0001695724796470631}
+{"ts": "2025-12-27T12:10:42", "event": "train_log", "step": 4478, "epoch": 1.889451476793249, "progress_pct": 31.49, "epoch_pct": 31.49, "eta": "38:55:29", "max_grad_norm": 1.0, "loss": 0.661283552646637, "grad_norm": 0.4559679329395294, "learning_rate": 0.00016953797181494475}
+{"ts": "2025-12-27T12:10:59", "event": "train_log", "step": 4480, "epoch": 1.890295358649789, "progress_pct": 31.5, "epoch_pct": 31.5, "eta": "38:54:35", "max_grad_norm": 1.0, "loss": 0.6587702035903931, "grad_norm": 0.3980371952056885, "learning_rate": 0.00016950344794175183}
+{"ts": "2025-12-27T12:11:15", "event": "train_log", "step": 4482, "epoch": 1.891139240506329, "progress_pct": 31.52, "epoch_pct": 31.52, "eta": "38:53:38", "max_grad_norm": 1.0, "loss": 0.7898983955383301, "grad_norm": 0.41418227553367615, "learning_rate": 0.0001694689080354483}
+{"ts": "2025-12-27T12:11:30", "event": "train_log", "step": 4484, "epoch": 1.8919831223628694, "progress_pct": 31.53, "epoch_pct": 31.53, "eta": "38:52:39", "max_grad_norm": 1.0, "loss": 0.6213424205780029, "grad_norm": 0.41184720396995544, "learning_rate": 0.00016943435210400184}
+{"ts": "2025-12-27T12:11:46", "event": "train_log", "step": 4486, "epoch": 1.8928270042194093, "progress_pct": 31.55, "epoch_pct": 31.55, "eta": "38:51:43", "max_grad_norm": 1.0, "loss": 0.6624190807342529, "grad_norm": 0.39218711853027344, "learning_rate": 0.00016939978015538388}
+{"ts": "2025-12-27T12:12:02", "event": "train_log", "step": 4488, "epoch": 1.8936708860759492, "progress_pct": 31.56, "epoch_pct": 31.56, "eta": "38:50:47", "max_grad_norm": 1.0, "loss": 0.7063044309616089, "grad_norm": 0.4875730872154236, "learning_rate": 0.0001693651921975695}
+{"ts": "2025-12-27T12:12:18", "event": "train_log", "step": 4490, "epoch": 1.8945147679324894, "progress_pct": 31.58, "epoch_pct": 31.58, "eta": "38:49:50", "max_grad_norm": 1.0, "loss": 0.6471660137176514, "grad_norm": 0.38058432936668396, "learning_rate": 0.00016933058823853748}
+{"ts": "2025-12-27T12:12:32", "event": "train_log", "step": 4492, "epoch": 1.8953586497890296, "progress_pct": 31.59, "epoch_pct": 31.59, "eta": "38:48:51", "max_grad_norm": 1.0, "loss": 0.7694545388221741, "grad_norm": 0.5368654131889343, "learning_rate": 0.0001692959682862702}
+{"ts": "2025-12-27T12:12:47", "event": "train_log", "step": 4494, "epoch": 1.8962025316455695, "progress_pct": 31.6, "epoch_pct": 31.6, "eta": "38:47:52", "max_grad_norm": 1.0, "loss": 0.700016438961029, "grad_norm": 0.41736599802970886, "learning_rate": 0.00016926133234875396}
+{"ts": "2025-12-27T12:13:02", "event": "train_log", "step": 4496, "epoch": 1.8970464135021097, "progress_pct": 31.62, "epoch_pct": 31.62, "eta": "38:46:53", "max_grad_norm": 1.0, "loss": 0.6926955580711365, "grad_norm": 0.4086865484714508, "learning_rate": 0.0001692266804339785}
+{"ts": "2025-12-27T12:13:17", "event": "train_log", "step": 4498, "epoch": 1.8978902953586498, "progress_pct": 31.63, "epoch_pct": 31.63, "eta": "38:45:55", "max_grad_norm": 1.0, "loss": 0.6776167154312134, "grad_norm": 0.447516530752182, "learning_rate": 0.00016919201254993738}
+{"ts": "2025-12-27T12:13:33", "event": "train_log", "step": 4500, "epoch": 1.8987341772151898, "progress_pct": 31.65, "epoch_pct": 31.65, "eta": "38:44:58", "max_grad_norm": 1.0, "loss": 0.6860454678535461, "grad_norm": 0.43951377272605896, "learning_rate": 0.00016915732870462783}
+{"ts": "2025-12-27T12:24:47", "event": "train_log", "step": 4500, "epoch": 1.8987341772151898, "progress_pct": 31.65, "epoch_pct": 31.65, "eta": "39:09:15", "max_grad_norm": 1.0, "eval_loss": 0.7600579857826233, "eval_runtime": 674.2593, "eval_samples_per_second": 3.125, "eval_steps_per_second": 3.125}
+{"ts": "2025-12-27T12:25:04", "event": "train_log", "step": 4502, "epoch": 1.89957805907173, "progress_pct": 31.66, "epoch_pct": 31.66, "eta": "39:08:19", "max_grad_norm": 1.0, "loss": 0.6525024175643921, "grad_norm": 0.380329966545105, "learning_rate": 0.00016912262890605076}
+{"ts": "2025-12-27T12:25:19", "event": "train_log", "step": 4504, "epoch": 1.90042194092827, "progress_pct": 31.67, "epoch_pct": 31.67, "eta": "39:07:21", "max_grad_norm": 1.0, "loss": 0.6473777294158936, "grad_norm": 0.3461032807826996, "learning_rate": 0.00016908791316221073}
+{"ts": "2025-12-27T12:25:34", "event": "train_log", "step": 4506, "epoch": 1.90126582278481, "progress_pct": 31.69, "epoch_pct": 31.69, "eta": "39:06:21", "max_grad_norm": 1.0, "loss": 0.760007381439209, "grad_norm": 0.47325965762138367, "learning_rate": 0.00016905318148111595}
+{"ts": "2025-12-27T12:25:50", "event": "train_log", "step": 4508, "epoch": 1.9021097046413502, "progress_pct": 31.7, "epoch_pct": 31.7, "eta": "39:05:24", "max_grad_norm": 1.0, "loss": 0.659562349319458, "grad_norm": 0.39158931374549866, "learning_rate": 0.00016901843387077844}
+{"ts": "2025-12-27T12:26:05", "event": "train_log", "step": 4510, "epoch": 1.9029535864978904, "progress_pct": 31.72, "epoch_pct": 31.72, "eta": "39:04:25", "max_grad_norm": 1.0, "loss": 0.5873214602470398, "grad_norm": 0.36785584688186646, "learning_rate": 0.0001689836703392137}
+{"ts": "2025-12-27T12:26:21", "event": "train_log", "step": 4512, "epoch": 1.9037974683544303, "progress_pct": 31.73, "epoch_pct": 31.73, "eta": "39:03:29", "max_grad_norm": 1.0, "loss": 0.6236514449119568, "grad_norm": 0.3843834102153778, "learning_rate": 0.00016894889089444115}
+{"ts": "2025-12-27T12:26:37", "event": "train_log", "step": 4514, "epoch": 1.9046413502109705, "progress_pct": 31.74, "epoch_pct": 31.74, "eta": "39:02:31", "max_grad_norm": 1.0, "loss": 0.7129080891609192, "grad_norm": 0.41675102710723877, "learning_rate": 0.0001689140955444836}
+{"ts": "2025-12-27T12:26:52", "event": "train_log", "step": 4516, "epoch": 1.9054852320675106, "progress_pct": 31.76, "epoch_pct": 31.76, "eta": "39:01:32", "max_grad_norm": 1.0, "loss": 0.7348461747169495, "grad_norm": 0.46025362610816956, "learning_rate": 0.00016887928429736775}
+{"ts": "2025-12-27T12:27:08", "event": "train_log", "step": 4518, "epoch": 1.9063291139240506, "progress_pct": 31.77, "epoch_pct": 31.77, "eta": "39:00:34", "max_grad_norm": 1.0, "loss": 0.6797979474067688, "grad_norm": 0.43589887022972107, "learning_rate": 0.00016884445716112387}
+{"ts": "2025-12-27T12:27:23", "event": "train_log", "step": 4520, "epoch": 1.9071729957805907, "progress_pct": 31.79, "epoch_pct": 31.79, "eta": "38:59:36", "max_grad_norm": 1.0, "loss": 0.6754241585731506, "grad_norm": 0.4301862120628357, "learning_rate": 0.0001688096141437859}
+{"ts": "2025-12-27T12:27:38", "event": "train_log", "step": 4522, "epoch": 1.908016877637131, "progress_pct": 31.8, "epoch_pct": 31.8, "eta": "38:58:39", "max_grad_norm": 1.0, "loss": 0.681866466999054, "grad_norm": 0.41446763277053833, "learning_rate": 0.00016877475525339146}
+{"ts": "2025-12-27T12:27:56", "event": "train_log", "step": 4524, "epoch": 1.9088607594936708, "progress_pct": 31.81, "epoch_pct": 31.81, "eta": "38:57:45", "max_grad_norm": 1.0, "loss": 0.6626620888710022, "grad_norm": 0.4051108658313751, "learning_rate": 0.0001687398804979818}
+{"ts": "2025-12-27T12:28:11", "event": "train_log", "step": 4526, "epoch": 1.909704641350211, "progress_pct": 31.83, "epoch_pct": 31.83, "eta": "38:56:47", "max_grad_norm": 1.0, "loss": 0.6753385066986084, "grad_norm": 0.3896785080432892, "learning_rate": 0.00016870498988560192}
+{"ts": "2025-12-27T12:28:28", "event": "train_log", "step": 4528, "epoch": 1.9105485232067512, "progress_pct": 31.84, "epoch_pct": 31.84, "eta": "38:55:51", "max_grad_norm": 1.0, "loss": 0.6971913576126099, "grad_norm": 0.4307839274406433, "learning_rate": 0.00016867008342430037}
+{"ts": "2025-12-27T12:28:44", "event": "train_log", "step": 4530, "epoch": 1.9113924050632911, "progress_pct": 31.86, "epoch_pct": 31.86, "eta": "38:54:56", "max_grad_norm": 1.0, "loss": 0.7674229145050049, "grad_norm": 0.4120950698852539, "learning_rate": 0.0001686351611221294}
+{"ts": "2025-12-27T12:29:00", "event": "train_log", "step": 4532, "epoch": 1.9122362869198313, "progress_pct": 31.87, "epoch_pct": 31.87, "eta": "38:54:00", "max_grad_norm": 1.0, "loss": 0.7074664235115051, "grad_norm": 0.4321785569190979, "learning_rate": 0.00016860022298714492}
+{"ts": "2025-12-27T12:29:16", "event": "train_log", "step": 4534, "epoch": 1.9130801687763714, "progress_pct": 31.88, "epoch_pct": 31.88, "eta": "38:53:04", "max_grad_norm": 1.0, "loss": 0.6178744435310364, "grad_norm": 0.38313060998916626, "learning_rate": 0.00016856526902740652}
+{"ts": "2025-12-27T12:29:32", "event": "train_log", "step": 4536, "epoch": 1.9139240506329114, "progress_pct": 31.9, "epoch_pct": 31.9, "eta": "38:52:05", "max_grad_norm": 1.0, "loss": 0.7168449759483337, "grad_norm": 0.3940589129924774, "learning_rate": 0.0001685302992509773}
+{"ts": "2025-12-27T12:29:48", "event": "train_log", "step": 4538, "epoch": 1.9147679324894513, "progress_pct": 31.91, "epoch_pct": 31.91, "eta": "38:51:10", "max_grad_norm": 1.0, "loss": 0.6892635226249695, "grad_norm": 0.3719513416290283, "learning_rate": 0.0001684953136659243}
+{"ts": "2025-12-27T12:30:04", "event": "train_log", "step": 4540, "epoch": 1.9156118143459917, "progress_pct": 31.93, "epoch_pct": 31.93, "eta": "38:50:12", "max_grad_norm": 1.0, "loss": 0.7352524399757385, "grad_norm": 0.40150535106658936, "learning_rate": 0.00016846031228031784}
+{"ts": "2025-12-27T12:30:19", "event": "train_log", "step": 4542, "epoch": 1.9164556962025316, "progress_pct": 31.94, "epoch_pct": 31.94, "eta": "38:49:16", "max_grad_norm": 1.0, "loss": 0.6469122767448425, "grad_norm": 0.4212523102760315, "learning_rate": 0.00016842529510223213}
+{"ts": "2025-12-27T12:30:34", "event": "train_log", "step": 4544, "epoch": 1.9172995780590716, "progress_pct": 31.95, "epoch_pct": 31.95, "eta": "38:48:17", "max_grad_norm": 1.0, "loss": 0.6883940696716309, "grad_norm": 0.45365604758262634, "learning_rate": 0.00016839026213974503}
+{"ts": "2025-12-27T12:30:50", "event": "train_log", "step": 4546, "epoch": 1.918143459915612, "progress_pct": 31.97, "epoch_pct": 31.97, "eta": "38:47:20", "max_grad_norm": 1.0, "loss": 0.7378416657447815, "grad_norm": 0.436541885137558, "learning_rate": 0.00016835521340093787}
+{"ts": "2025-12-27T12:31:05", "event": "train_log", "step": 4548, "epoch": 1.918987341772152, "progress_pct": 31.98, "epoch_pct": 31.98, "eta": "38:46:23", "max_grad_norm": 1.0, "loss": 0.7432116270065308, "grad_norm": 0.41461536288261414, "learning_rate": 0.0001683201488938958}
+{"ts": "2025-12-27T12:31:21", "event": "train_log", "step": 4550, "epoch": 1.9198312236286919, "progress_pct": 32.0, "epoch_pct": 32.0, "eta": "38:45:26", "max_grad_norm": 1.0, "loss": 0.6801432967185974, "grad_norm": 0.40218585729599, "learning_rate": 0.00016828506862670746}
+{"ts": "2025-12-27T12:31:36", "event": "train_log", "step": 4552, "epoch": 1.920675105485232, "progress_pct": 32.01, "epoch_pct": 32.01, "eta": "38:44:28", "max_grad_norm": 1.0, "loss": 0.6835547685623169, "grad_norm": 0.3994768559932709, "learning_rate": 0.00016824997260746525}
+{"ts": "2025-12-27T12:31:52", "event": "train_log", "step": 4554, "epoch": 1.9215189873417722, "progress_pct": 32.03, "epoch_pct": 32.03, "eta": "38:43:32", "max_grad_norm": 1.0, "loss": 0.7723629474639893, "grad_norm": 0.460328072309494, "learning_rate": 0.00016821486084426512}
+{"ts": "2025-12-27T12:32:08", "event": "train_log", "step": 4556, "epoch": 1.9223628691983121, "progress_pct": 32.04, "epoch_pct": 32.04, "eta": "38:42:35", "max_grad_norm": 1.0, "loss": 0.7513449192047119, "grad_norm": 0.4739786386489868, "learning_rate": 0.00016817973334520668}
+{"ts": "2025-12-27T12:32:23", "event": "train_log", "step": 4558, "epoch": 1.9232067510548523, "progress_pct": 32.05, "epoch_pct": 32.05, "eta": "38:41:36", "max_grad_norm": 1.0, "loss": 0.7000700235366821, "grad_norm": 0.46618419885635376, "learning_rate": 0.00016814459011839315}
+{"ts": "2025-12-27T12:32:39", "event": "train_log", "step": 4560, "epoch": 1.9240506329113924, "progress_pct": 32.07, "epoch_pct": 32.07, "eta": "38:40:40", "max_grad_norm": 1.0, "loss": 0.6381427049636841, "grad_norm": 0.4187009632587433, "learning_rate": 0.00016810943117193146}
+{"ts": "2025-12-27T12:32:54", "event": "train_log", "step": 4562, "epoch": 1.9248945147679324, "progress_pct": 32.08, "epoch_pct": 32.08, "eta": "38:39:42", "max_grad_norm": 1.0, "loss": 0.7823911309242249, "grad_norm": 0.5057410597801208, "learning_rate": 0.00016807425651393207}
+{"ts": "2025-12-27T12:33:09", "event": "train_log", "step": 4564, "epoch": 1.9257383966244725, "progress_pct": 32.1, "epoch_pct": 32.1, "eta": "38:38:46", "max_grad_norm": 1.0, "loss": 0.7210549712181091, "grad_norm": 0.4235171973705292, "learning_rate": 0.00016803906615250907}
+{"ts": "2025-12-27T12:33:25", "event": "train_log", "step": 4566, "epoch": 1.9265822784810127, "progress_pct": 32.11, "epoch_pct": 32.11, "eta": "38:37:48", "max_grad_norm": 1.0, "loss": 0.794435441493988, "grad_norm": 0.45755431056022644, "learning_rate": 0.00016800386009578025}
+{"ts": "2025-12-27T12:33:40", "event": "train_log", "step": 4568, "epoch": 1.9274261603375527, "progress_pct": 32.12, "epoch_pct": 32.12, "eta": "38:36:51", "max_grad_norm": 1.0, "loss": 0.6490750312805176, "grad_norm": 0.4225042760372162, "learning_rate": 0.0001679686383518669}
+{"ts": "2025-12-27T12:33:56", "event": "train_log", "step": 4570, "epoch": 1.9282700421940928, "progress_pct": 32.14, "epoch_pct": 32.14, "eta": "38:35:55", "max_grad_norm": 1.0, "loss": 0.6976916790008545, "grad_norm": 0.41081616282463074, "learning_rate": 0.00016793340092889408}
+{"ts": "2025-12-27T12:34:12", "event": "train_log", "step": 4572, "epoch": 1.929113924050633, "progress_pct": 32.15, "epoch_pct": 32.15, "eta": "38:34:59", "max_grad_norm": 1.0, "loss": 0.6559295058250427, "grad_norm": 0.42354175448417664, "learning_rate": 0.00016789814783499033}
+{"ts": "2025-12-27T12:34:28", "event": "train_log", "step": 4574, "epoch": 1.929957805907173, "progress_pct": 32.17, "epoch_pct": 32.17, "eta": "38:34:02", "max_grad_norm": 1.0, "loss": 0.7900391817092896, "grad_norm": 0.39702117443084717, "learning_rate": 0.0001678628790782879}
+{"ts": "2025-12-27T12:34:42", "event": "train_log", "step": 4576, "epoch": 1.930801687763713, "progress_pct": 32.18, "epoch_pct": 32.18, "eta": "38:33:03", "max_grad_norm": 1.0, "loss": 0.7020596265792847, "grad_norm": 0.4400629997253418, "learning_rate": 0.00016782759466692256}
+{"ts": "2025-12-27T12:34:57", "event": "train_log", "step": 4578, "epoch": 1.9316455696202532, "progress_pct": 32.19, "epoch_pct": 32.19, "eta": "38:32:05", "max_grad_norm": 1.0, "loss": 0.7107458710670471, "grad_norm": 0.42563396692276, "learning_rate": 0.00016779229460903383}
+{"ts": "2025-12-27T12:35:13", "event": "train_log", "step": 4580, "epoch": 1.9324894514767932, "progress_pct": 32.21, "epoch_pct": 32.21, "eta": "38:31:10", "max_grad_norm": 1.0, "loss": 0.58088618516922, "grad_norm": 0.40650051832199097, "learning_rate": 0.0001677569789127647}
+{"ts": "2025-12-27T12:35:30", "event": "train_log", "step": 4582, "epoch": 1.9333333333333333, "progress_pct": 32.22, "epoch_pct": 32.22, "eta": "38:30:16", "max_grad_norm": 1.0, "loss": 0.6815688610076904, "grad_norm": 0.42461681365966797, "learning_rate": 0.00016772164758626184}
+{"ts": "2025-12-27T12:35:46", "event": "train_log", "step": 4584, "epoch": 1.9341772151898735, "progress_pct": 32.24, "epoch_pct": 32.24, "eta": "38:29:20", "max_grad_norm": 1.0, "loss": 0.6617820262908936, "grad_norm": 0.4248030483722687, "learning_rate": 0.0001676863006376755}
+{"ts": "2025-12-27T12:36:01", "event": "train_log", "step": 4586, "epoch": 1.9350210970464135, "progress_pct": 32.25, "epoch_pct": 32.25, "eta": "38:28:23", "max_grad_norm": 1.0, "loss": 0.7462296485900879, "grad_norm": 0.4336024522781372, "learning_rate": 0.0001676509380751595}
+{"ts": "2025-12-27T12:36:18", "event": "train_log", "step": 4588, "epoch": 1.9358649789029536, "progress_pct": 32.26, "epoch_pct": 32.26, "eta": "38:27:29", "max_grad_norm": 1.0, "loss": 0.7801896333694458, "grad_norm": 0.42542141675949097, "learning_rate": 0.00016761555990687136}
+{"ts": "2025-12-27T12:36:33", "event": "train_log", "step": 4590, "epoch": 1.9367088607594938, "progress_pct": 32.28, "epoch_pct": 32.28, "eta": "38:26:32", "max_grad_norm": 1.0, "loss": 0.7308001518249512, "grad_norm": 0.41240575909614563, "learning_rate": 0.00016758016614097217}
+{"ts": "2025-12-27T12:36:48", "event": "train_log", "step": 4592, "epoch": 1.9375527426160337, "progress_pct": 32.29, "epoch_pct": 32.29, "eta": "38:25:35", "max_grad_norm": 1.0, "loss": 0.7126680016517639, "grad_norm": 0.3811012804508209, "learning_rate": 0.00016754475678562654}
+{"ts": "2025-12-27T12:37:04", "event": "train_log", "step": 4594, "epoch": 1.9383966244725739, "progress_pct": 32.31, "epoch_pct": 32.31, "eta": "38:24:40", "max_grad_norm": 1.0, "loss": 0.679752767086029, "grad_norm": 0.4859951436519623, "learning_rate": 0.00016750933184900272}
+{"ts": "2025-12-27T12:37:20", "event": "train_log", "step": 4596, "epoch": 1.939240506329114, "progress_pct": 32.32, "epoch_pct": 32.32, "eta": "38:23:43", "max_grad_norm": 1.0, "loss": 0.6077587604522705, "grad_norm": 0.36403799057006836, "learning_rate": 0.00016747389133927261}
+{"ts": "2025-12-27T12:37:36", "event": "train_log", "step": 4598, "epoch": 1.940084388185654, "progress_pct": 32.33, "epoch_pct": 32.33, "eta": "38:22:49", "max_grad_norm": 1.0, "loss": 0.6919949650764465, "grad_norm": 0.3960701525211334, "learning_rate": 0.00016743843526461158}
+{"ts": "2025-12-27T12:37:51", "event": "train_log", "step": 4600, "epoch": 1.9409282700421941, "progress_pct": 32.35, "epoch_pct": 32.35, "eta": "38:21:52", "max_grad_norm": 1.0, "loss": 0.728607177734375, "grad_norm": 0.4058634042739868, "learning_rate": 0.00016740296363319877}
+{"ts": "2025-12-27T12:49:10", "event": "train_log", "step": 4600, "epoch": 1.9409282700421941, "progress_pct": 32.35, "epoch_pct": 32.35, "eta": "38:45:32", "max_grad_norm": 1.0, "eval_loss": 0.7585541009902954, "eval_runtime": 679.0866, "eval_samples_per_second": 3.103, "eval_steps_per_second": 3.103}
+{"ts": "2025-12-27T12:49:26", "event": "train_log", "step": 4602, "epoch": 1.9417721518987343, "progress_pct": 32.36, "epoch_pct": 32.36, "eta": "38:44:34", "max_grad_norm": 1.0, "loss": 0.669603168964386, "grad_norm": 0.45107272267341614, "learning_rate": 0.00016736747645321672}
+{"ts": "2025-12-27T12:49:42", "event": "train_log", "step": 4604, "epoch": 1.9426160337552743, "progress_pct": 32.38, "epoch_pct": 32.38, "eta": "38:43:38", "max_grad_norm": 1.0, "loss": 0.6975913643836975, "grad_norm": 0.4244096577167511, "learning_rate": 0.00016733197373285165}
+{"ts": "2025-12-27T12:49:58", "event": "train_log", "step": 4606, "epoch": 1.9434599156118142, "progress_pct": 32.39, "epoch_pct": 32.39, "eta": "38:42:42", "max_grad_norm": 1.0, "loss": 0.5990989804267883, "grad_norm": 0.37567681074142456, "learning_rate": 0.00016729645548029342}
+{"ts": "2025-12-27T12:50:13", "event": "train_log", "step": 4608, "epoch": 1.9443037974683546, "progress_pct": 32.41, "epoch_pct": 32.41, "eta": "38:41:44", "max_grad_norm": 1.0, "loss": 0.723430871963501, "grad_norm": 0.40256237983703613, "learning_rate": 0.00016726092170373533}
+{"ts": "2025-12-27T12:50:27", "event": "train_log", "step": 4610, "epoch": 1.9451476793248945, "progress_pct": 32.42, "epoch_pct": 32.42, "eta": "38:40:44", "max_grad_norm": 1.0, "loss": 0.7460100650787354, "grad_norm": 0.4884983003139496, "learning_rate": 0.0001672253724113744}
+{"ts": "2025-12-27T12:50:43", "event": "train_log", "step": 4612, "epoch": 1.9459915611814345, "progress_pct": 32.43, "epoch_pct": 32.43, "eta": "38:39:48", "max_grad_norm": 1.0, "loss": 0.5577017664909363, "grad_norm": 0.3629630208015442, "learning_rate": 0.0001671898076114111}
+{"ts": "2025-12-27T12:50:57", "event": "train_log", "step": 4614, "epoch": 1.9468354430379748, "progress_pct": 32.45, "epoch_pct": 32.45, "eta": "38:38:49", "max_grad_norm": 1.0, "loss": 0.7394112944602966, "grad_norm": 0.42751777172088623, "learning_rate": 0.00016715422731204967}
+{"ts": "2025-12-27T12:51:13", "event": "train_log", "step": 4616, "epoch": 1.9476793248945148, "progress_pct": 32.46, "epoch_pct": 32.46, "eta": "38:37:51", "max_grad_norm": 1.0, "loss": 0.5905179381370544, "grad_norm": 0.35809609293937683, "learning_rate": 0.00016711863152149765}
+{"ts": "2025-12-27T12:51:28", "event": "train_log", "step": 4618, "epoch": 1.9485232067510547, "progress_pct": 32.48, "epoch_pct": 32.48, "eta": "38:36:55", "max_grad_norm": 1.0, "loss": 0.7035338878631592, "grad_norm": 0.4111083745956421, "learning_rate": 0.0001670830202479664}
+{"ts": "2025-12-27T12:51:43", "event": "train_log", "step": 4620, "epoch": 1.9493670886075949, "progress_pct": 32.49, "epoch_pct": 32.49, "eta": "38:35:55", "max_grad_norm": 1.0, "loss": 0.7048936486244202, "grad_norm": 0.4530862867832184, "learning_rate": 0.0001670473934996707}
+{"ts": "2025-12-27T12:51:57", "event": "train_log", "step": 4622, "epoch": 1.950210970464135, "progress_pct": 32.5, "epoch_pct": 32.5, "eta": "38:34:57", "max_grad_norm": 1.0, "loss": 0.6594944596290588, "grad_norm": 0.41778260469436646, "learning_rate": 0.00016701175128482903}
+{"ts": "2025-12-27T12:52:12", "event": "train_log", "step": 4624, "epoch": 1.951054852320675, "progress_pct": 32.52, "epoch_pct": 32.52, "eta": "38:33:58", "max_grad_norm": 1.0, "loss": 0.6902912259101868, "grad_norm": 0.40512707829475403, "learning_rate": 0.0001669760936116633}
+{"ts": "2025-12-27T12:52:27", "event": "train_log", "step": 4626, "epoch": 1.9518987341772152, "progress_pct": 32.53, "epoch_pct": 32.53, "eta": "38:33:00", "max_grad_norm": 1.0, "loss": 0.7086893320083618, "grad_norm": 0.4226403832435608, "learning_rate": 0.0001669404204883991}
+{"ts": "2025-12-27T12:52:42", "event": "train_log", "step": 4628, "epoch": 1.9527426160337553, "progress_pct": 32.55, "epoch_pct": 32.55, "eta": "38:32:03", "max_grad_norm": 1.0, "loss": 0.6647822856903076, "grad_norm": 0.397930383682251, "learning_rate": 0.00016690473192326549}
+{"ts": "2025-12-27T12:52:56", "event": "train_log", "step": 4630, "epoch": 1.9535864978902953, "progress_pct": 32.56, "epoch_pct": 32.56, "eta": "38:31:04", "max_grad_norm": 1.0, "loss": 0.747227668762207, "grad_norm": 0.3985617458820343, "learning_rate": 0.00016686902792449515}
+{"ts": "2025-12-27T12:53:11", "event": "train_log", "step": 4632, "epoch": 1.9544303797468354, "progress_pct": 32.57, "epoch_pct": 32.57, "eta": "38:30:05", "max_grad_norm": 1.0, "loss": 0.7526347637176514, "grad_norm": 0.4174083173274994, "learning_rate": 0.00016683330850032432}
+{"ts": "2025-12-27T12:53:26", "event": "train_log", "step": 4634, "epoch": 1.9552742616033756, "progress_pct": 32.59, "epoch_pct": 32.59, "eta": "38:29:06", "max_grad_norm": 1.0, "loss": 0.6674112677574158, "grad_norm": 0.425070583820343, "learning_rate": 0.00016679757365899275}
+{"ts": "2025-12-27T12:53:41", "event": "train_log", "step": 4636, "epoch": 1.9561181434599155, "progress_pct": 32.6, "epoch_pct": 32.6, "eta": "38:28:09", "max_grad_norm": 1.0, "loss": 0.6515055298805237, "grad_norm": 0.4049135148525238, "learning_rate": 0.00016676182340874386}
+{"ts": "2025-12-27T12:53:56", "event": "train_log", "step": 4638, "epoch": 1.9569620253164557, "progress_pct": 32.62, "epoch_pct": 32.62, "eta": "38:27:11", "max_grad_norm": 1.0, "loss": 0.645137369632721, "grad_norm": 0.4312315285205841, "learning_rate": 0.00016672605775782447}
+{"ts": "2025-12-27T12:54:10", "event": "train_log", "step": 4640, "epoch": 1.9578059071729959, "progress_pct": 32.63, "epoch_pct": 32.63, "eta": "38:26:13", "max_grad_norm": 1.0, "loss": 0.6519104838371277, "grad_norm": 0.39540281891822815, "learning_rate": 0.00016669027671448503}
+{"ts": "2025-12-27T12:54:25", "event": "train_log", "step": 4642, "epoch": 1.9586497890295358, "progress_pct": 32.64, "epoch_pct": 32.64, "eta": "38:25:14", "max_grad_norm": 1.0, "loss": 0.6815812587738037, "grad_norm": 0.5028185844421387, "learning_rate": 0.00016665448028697961}
+{"ts": "2025-12-27T12:54:39", "event": "train_log", "step": 4644, "epoch": 1.959493670886076, "progress_pct": 32.66, "epoch_pct": 32.66, "eta": "38:24:16", "max_grad_norm": 1.0, "loss": 0.6713913083076477, "grad_norm": 0.4395635724067688, "learning_rate": 0.00016661866848356567}
+{"ts": "2025-12-27T12:54:55", "event": "train_log", "step": 4646, "epoch": 1.9603375527426161, "progress_pct": 32.67, "epoch_pct": 32.67, "eta": "38:23:20", "max_grad_norm": 1.0, "loss": 0.6989195942878723, "grad_norm": 0.42733073234558105, "learning_rate": 0.00016658284131250436}
+{"ts": "2025-12-27T12:55:10", "event": "train_log", "step": 4648, "epoch": 1.961181434599156, "progress_pct": 32.69, "epoch_pct": 32.69, "eta": "38:22:22", "max_grad_norm": 1.0, "loss": 0.6501366496086121, "grad_norm": 0.41415879130363464, "learning_rate": 0.00016654699878206031}
+{"ts": "2025-12-27T12:55:25", "event": "train_log", "step": 4650, "epoch": 1.9620253164556962, "progress_pct": 32.7, "epoch_pct": 32.7, "eta": "38:21:26", "max_grad_norm": 1.0, "loss": 0.6437955498695374, "grad_norm": 0.39072170853614807, "learning_rate": 0.0001665111409005017}
+{"ts": "2025-12-27T12:55:40", "event": "train_log", "step": 4652, "epoch": 1.9628691983122364, "progress_pct": 32.71, "epoch_pct": 32.71, "eta": "38:20:27", "max_grad_norm": 1.0, "loss": 0.6521314382553101, "grad_norm": 0.41802850365638733, "learning_rate": 0.00016647526767610025}
+{"ts": "2025-12-27T12:55:54", "event": "train_log", "step": 4654, "epoch": 1.9637130801687763, "progress_pct": 32.73, "epoch_pct": 32.73, "eta": "38:19:29", "max_grad_norm": 1.0, "loss": 0.7300742864608765, "grad_norm": 0.41475021839141846, "learning_rate": 0.00016643937911713124}
+{"ts": "2025-12-27T12:56:09", "event": "train_log", "step": 4656, "epoch": 1.9645569620253165, "progress_pct": 32.74, "epoch_pct": 32.74, "eta": "38:18:32", "max_grad_norm": 1.0, "loss": 0.6217485070228577, "grad_norm": 0.41763070225715637, "learning_rate": 0.00016640347523187344}
+{"ts": "2025-12-27T12:56:24", "event": "train_log", "step": 4658, "epoch": 1.9654008438818567, "progress_pct": 32.76, "epoch_pct": 32.76, "eta": "38:17:33", "max_grad_norm": 1.0, "loss": 0.7974472641944885, "grad_norm": 0.4488297402858734, "learning_rate": 0.00016636755602860922}
+{"ts": "2025-12-27T12:56:39", "event": "train_log", "step": 4660, "epoch": 1.9662447257383966, "progress_pct": 32.77, "epoch_pct": 32.77, "eta": "38:16:36", "max_grad_norm": 1.0, "loss": 0.728827714920044, "grad_norm": 0.37688007950782776, "learning_rate": 0.00016633162151562445}
+{"ts": "2025-12-27T12:56:53", "event": "train_log", "step": 4662, "epoch": 1.9670886075949368, "progress_pct": 32.78, "epoch_pct": 32.78, "eta": "38:15:38", "max_grad_norm": 1.0, "loss": 0.7563744783401489, "grad_norm": 0.4636184275150299, "learning_rate": 0.0001662956717012085}
+{"ts": "2025-12-27T12:57:08", "event": "train_log", "step": 4664, "epoch": 1.967932489451477, "progress_pct": 32.8, "epoch_pct": 32.8, "eta": "38:14:41", "max_grad_norm": 1.0, "loss": 0.7408145666122437, "grad_norm": 0.432020902633667, "learning_rate": 0.00016625970659365434}
+{"ts": "2025-12-27T12:57:23", "event": "train_log", "step": 4666, "epoch": 1.9687763713080169, "progress_pct": 32.81, "epoch_pct": 32.81, "eta": "38:13:43", "max_grad_norm": 1.0, "loss": 0.6961510181427002, "grad_norm": 0.4098186790943146, "learning_rate": 0.0001662237262012584}
+{"ts": "2025-12-27T12:57:39", "event": "train_log", "step": 4668, "epoch": 1.9696202531645568, "progress_pct": 32.83, "epoch_pct": 32.83, "eta": "38:12:47", "max_grad_norm": 1.0, "loss": 0.6959825754165649, "grad_norm": 0.44124510884284973, "learning_rate": 0.00016618773053232068}
+{"ts": "2025-12-27T12:57:54", "event": "train_log", "step": 4670, "epoch": 1.9704641350210972, "progress_pct": 32.84, "epoch_pct": 32.84, "eta": "38:11:51", "max_grad_norm": 1.0, "loss": 0.6693721413612366, "grad_norm": 0.3729988634586334, "learning_rate": 0.00016615171959514466}
+{"ts": "2025-12-27T12:58:10", "event": "train_log", "step": 4672, "epoch": 1.9713080168776371, "progress_pct": 32.86, "epoch_pct": 32.86, "eta": "38:10:55", "max_grad_norm": 1.0, "loss": 0.6943060159683228, "grad_norm": 0.4506371021270752, "learning_rate": 0.00016611569339803744}
+{"ts": "2025-12-27T12:58:25", "event": "train_log", "step": 4674, "epoch": 1.972151898734177, "progress_pct": 32.87, "epoch_pct": 32.87, "eta": "38:09:59", "max_grad_norm": 1.0, "loss": 0.6915813684463501, "grad_norm": 0.43713298439979553, "learning_rate": 0.00016607965194930947}
+{"ts": "2025-12-27T12:58:41", "event": "train_log", "step": 4676, "epoch": 1.9729957805907175, "progress_pct": 32.88, "epoch_pct": 32.88, "eta": "38:09:04", "max_grad_norm": 1.0, "loss": 0.7430137991905212, "grad_norm": 0.4375544786453247, "learning_rate": 0.00016604359525727492}
+{"ts": "2025-12-27T12:58:56", "event": "train_log", "step": 4678, "epoch": 1.9738396624472574, "progress_pct": 32.9, "epoch_pct": 32.9, "eta": "38:08:08", "max_grad_norm": 1.0, "loss": 0.7260952591896057, "grad_norm": 0.4393104314804077, "learning_rate": 0.00016600752333025134}
+{"ts": "2025-12-27T12:59:11", "event": "train_log", "step": 4680, "epoch": 1.9746835443037973, "progress_pct": 32.91, "epoch_pct": 32.91, "eta": "38:07:10", "max_grad_norm": 1.0, "loss": 0.7350760102272034, "grad_norm": 0.4471568167209625, "learning_rate": 0.00016597143617655982}
+{"ts": "2025-12-27T12:59:27", "event": "train_log", "step": 4682, "epoch": 1.9755274261603377, "progress_pct": 32.93, "epoch_pct": 32.93, "eta": "38:06:15", "max_grad_norm": 1.0, "loss": 0.7295693159103394, "grad_norm": 0.39821380376815796, "learning_rate": 0.00016593533380452499}
+{"ts": "2025-12-27T12:59:41", "event": "train_log", "step": 4684, "epoch": 1.9763713080168777, "progress_pct": 32.94, "epoch_pct": 32.94, "eta": "38:05:18", "max_grad_norm": 1.0, "loss": 0.6461828947067261, "grad_norm": 0.40367990732192993, "learning_rate": 0.00016589921622247493}
+{"ts": "2025-12-27T12:59:57", "event": "train_log", "step": 4686, "epoch": 1.9772151898734176, "progress_pct": 32.95, "epoch_pct": 32.95, "eta": "38:04:22", "max_grad_norm": 1.0, "loss": 0.699313223361969, "grad_norm": 0.4723254442214966, "learning_rate": 0.00016586308343874136}
+{"ts": "2025-12-27T13:00:14", "event": "train_log", "step": 4688, "epoch": 1.9780590717299578, "progress_pct": 32.97, "epoch_pct": 32.97, "eta": "38:03:29", "max_grad_norm": 1.0, "loss": 0.7299131751060486, "grad_norm": 0.45741328597068787, "learning_rate": 0.0001658269354616593}
+{"ts": "2025-12-27T13:00:29", "event": "train_log", "step": 4690, "epoch": 1.978902953586498, "progress_pct": 32.98, "epoch_pct": 32.98, "eta": "38:02:32", "max_grad_norm": 1.0, "loss": 0.7643240690231323, "grad_norm": 0.4011725187301636, "learning_rate": 0.0001657907722995675}
+{"ts": "2025-12-27T13:00:45", "event": "train_log", "step": 4692, "epoch": 1.9797468354430379, "progress_pct": 33.0, "epoch_pct": 33.0, "eta": "38:01:37", "max_grad_norm": 1.0, "loss": 0.6945428848266602, "grad_norm": 0.3904820382595062, "learning_rate": 0.00016575459396080805}
+{"ts": "2025-12-27T13:01:00", "event": "train_log", "step": 4694, "epoch": 1.980590717299578, "progress_pct": 33.01, "epoch_pct": 33.01, "eta": "38:00:42", "max_grad_norm": 1.0, "loss": 0.5986754298210144, "grad_norm": 0.3643582761287689, "learning_rate": 0.00016571840045372663}
+{"ts": "2025-12-27T13:01:17", "event": "train_log", "step": 4696, "epoch": 1.9814345991561182, "progress_pct": 33.02, "epoch_pct": 33.02, "eta": "37:59:49", "max_grad_norm": 1.0, "loss": 0.7462178468704224, "grad_norm": 0.4371087849140167, "learning_rate": 0.00016568219178667235}
+{"ts": "2025-12-27T13:01:32", "event": "train_log", "step": 4698, "epoch": 1.9822784810126581, "progress_pct": 33.04, "epoch_pct": 33.04, "eta": "37:58:53", "max_grad_norm": 1.0, "loss": 0.7553940415382385, "grad_norm": 0.4440254867076874, "learning_rate": 0.0001656459679679979}
+{"ts": "2025-12-27T13:01:48", "event": "train_log", "step": 4700, "epoch": 1.9831223628691983, "progress_pct": 33.05, "epoch_pct": 33.05, "eta": "37:57:59", "max_grad_norm": 1.0, "loss": 0.7119964361190796, "grad_norm": 0.41370099782943726, "learning_rate": 0.00016560972900605937}
+{"ts": "2025-12-27T13:13:19", "event": "train_log", "step": 4700, "epoch": 1.9831223628691983, "progress_pct": 33.05, "epoch_pct": 33.05, "eta": "38:21:17", "max_grad_norm": 1.0, "eval_loss": 0.7582268714904785, "eval_runtime": 690.081, "eval_samples_per_second": 3.053, "eval_steps_per_second": 3.053}
+{"ts": "2025-12-27T13:13:34", "event": "train_log", "step": 4702, "epoch": 1.9839662447257385, "progress_pct": 33.07, "epoch_pct": 33.07, "eta": "38:20:20", "max_grad_norm": 1.0, "loss": 0.6260318160057068, "grad_norm": 0.4408273696899414, "learning_rate": 0.0001655734749092164}
+{"ts": "2025-12-27T13:13:49", "event": "train_log", "step": 4704, "epoch": 1.9848101265822784, "progress_pct": 33.08, "epoch_pct": 33.08, "eta": "38:19:22", "max_grad_norm": 1.0, "loss": 0.7382199168205261, "grad_norm": 0.4697865843772888, "learning_rate": 0.0001655372056858322}
+{"ts": "2025-12-27T13:14:04", "event": "train_log", "step": 4706, "epoch": 1.9856540084388186, "progress_pct": 33.09, "epoch_pct": 33.09, "eta": "38:18:25", "max_grad_norm": 1.0, "loss": 0.7037063241004944, "grad_norm": 0.423520028591156, "learning_rate": 0.00016550092134427327}
+{"ts": "2025-12-27T13:14:20", "event": "train_log", "step": 4708, "epoch": 1.9864978902953587, "progress_pct": 33.11, "epoch_pct": 33.11, "eta": "38:17:29", "max_grad_norm": 1.0, "loss": 0.7173316478729248, "grad_norm": 0.38633009791374207, "learning_rate": 0.00016546462189290974}
+{"ts": "2025-12-27T13:14:35", "event": "train_log", "step": 4710, "epoch": 1.9873417721518987, "progress_pct": 33.12, "epoch_pct": 33.12, "eta": "38:16:33", "max_grad_norm": 1.0, "loss": 0.6884669065475464, "grad_norm": 0.410230815410614, "learning_rate": 0.00016542830734011523}
+{"ts": "2025-12-27T13:14:49", "event": "train_log", "step": 4712, "epoch": 1.9881856540084388, "progress_pct": 33.14, "epoch_pct": 33.14, "eta": "38:15:34", "max_grad_norm": 1.0, "loss": 0.7181811332702637, "grad_norm": 0.4486774802207947, "learning_rate": 0.00016539197769426678}
+{"ts": "2025-12-27T13:15:03", "event": "train_log", "step": 4714, "epoch": 1.989029535864979, "progress_pct": 33.15, "epoch_pct": 33.15, "eta": "38:14:36", "max_grad_norm": 1.0, "loss": 0.7737504243850708, "grad_norm": 0.47757431864738464, "learning_rate": 0.00016535563296374495}
+{"ts": "2025-12-27T13:15:20", "event": "train_log", "step": 4716, "epoch": 1.989873417721519, "progress_pct": 33.16, "epoch_pct": 33.16, "eta": "38:13:42", "max_grad_norm": 1.0, "loss": 0.683630645275116, "grad_norm": 0.40281420946121216, "learning_rate": 0.00016531927315693382}
+{"ts": "2025-12-27T13:15:34", "event": "train_log", "step": 4718, "epoch": 1.990717299578059, "progress_pct": 33.18, "epoch_pct": 33.18, "eta": "38:12:43", "max_grad_norm": 1.0, "loss": 0.7447960376739502, "grad_norm": 0.42349720001220703, "learning_rate": 0.00016528289828222082}
+{"ts": "2025-12-27T13:15:49", "event": "train_log", "step": 4720, "epoch": 1.9915611814345993, "progress_pct": 33.19, "epoch_pct": 33.19, "eta": "38:11:45", "max_grad_norm": 1.0, "loss": 0.6800103783607483, "grad_norm": 0.40754765272140503, "learning_rate": 0.00016524650834799694}
+{"ts": "2025-12-27T13:16:04", "event": "train_log", "step": 4722, "epoch": 1.9924050632911392, "progress_pct": 33.21, "epoch_pct": 33.21, "eta": "38:10:49", "max_grad_norm": 1.0, "loss": 0.6609475612640381, "grad_norm": 0.3766736090183258, "learning_rate": 0.0001652101033626567}
+{"ts": "2025-12-27T13:16:20", "event": "train_log", "step": 4724, "epoch": 1.9932489451476794, "progress_pct": 33.22, "epoch_pct": 33.22, "eta": "38:09:54", "max_grad_norm": 1.0, "loss": 0.6496920585632324, "grad_norm": 0.3600929081439972, "learning_rate": 0.00016517368333459798}
+{"ts": "2025-12-27T13:16:35", "event": "train_log", "step": 4726, "epoch": 1.9940928270042195, "progress_pct": 33.23, "epoch_pct": 33.23, "eta": "38:08:56", "max_grad_norm": 1.0, "loss": 0.6305265426635742, "grad_norm": 0.39590612053871155, "learning_rate": 0.00016513724827222227}
+{"ts": "2025-12-27T13:16:49", "event": "train_log", "step": 4728, "epoch": 1.9949367088607595, "progress_pct": 33.25, "epoch_pct": 33.25, "eta": "38:07:59", "max_grad_norm": 1.0, "loss": 0.712282657623291, "grad_norm": 0.41002732515335083, "learning_rate": 0.00016510079818393426}
+{"ts": "2025-12-27T13:17:05", "event": "train_log", "step": 4730, "epoch": 1.9957805907172996, "progress_pct": 33.26, "epoch_pct": 33.26, "eta": "38:07:03", "max_grad_norm": 1.0, "loss": 0.6974884271621704, "grad_norm": 0.41948413848876953, "learning_rate": 0.00016506433307814246}
+{"ts": "2025-12-27T13:17:20", "event": "train_log", "step": 4732, "epoch": 1.9966244725738398, "progress_pct": 33.28, "epoch_pct": 33.28, "eta": "38:06:05", "max_grad_norm": 1.0, "loss": 0.6844019889831543, "grad_norm": 0.4283764958381653, "learning_rate": 0.00016502785296325858}
+{"ts": "2025-12-27T13:17:34", "event": "train_log", "step": 4734, "epoch": 1.9974683544303797, "progress_pct": 33.29, "epoch_pct": 33.29, "eta": "38:05:07", "max_grad_norm": 1.0, "loss": 0.7453671097755432, "grad_norm": 0.39830246567726135, "learning_rate": 0.0001649913578476979}
+{"ts": "2025-12-27T13:17:49", "event": "train_log", "step": 4736, "epoch": 1.9983122362869197, "progress_pct": 33.31, "epoch_pct": 33.31, "eta": "38:04:10", "max_grad_norm": 1.0, "loss": 0.6571418046951294, "grad_norm": 0.3875660300254822, "learning_rate": 0.00016495484773987917}
+{"ts": "2025-12-27T13:18:04", "event": "train_log", "step": 4738, "epoch": 1.99915611814346, "progress_pct": 33.32, "epoch_pct": 33.32, "eta": "38:03:14", "max_grad_norm": 1.0, "loss": 0.6844568848609924, "grad_norm": 0.379986971616745, "learning_rate": 0.00016491832264822453}
+{"ts": "2025-12-27T13:18:19", "event": "train_log", "step": 4740, "epoch": 2.0, "progress_pct": 33.33, "epoch_pct": 33.33, "eta": "38:02:18", "max_grad_norm": 1.0, "loss": 0.6115277409553528, "grad_norm": 0.38605886697769165, "learning_rate": 0.00016488178258115965}
+{"ts": "2025-12-27T13:18:35", "event": "train_log", "step": 4742, "epoch": 2.00084388185654, "progress_pct": 33.35, "epoch_pct": 33.35, "eta": "38:01:22", "max_grad_norm": 1.0, "loss": 0.6192594170570374, "grad_norm": 0.3991186022758484, "learning_rate": 0.0001648452275471136}
+{"ts": "2025-12-27T13:18:49", "event": "train_log", "step": 4744, "epoch": 2.0016877637130803, "progress_pct": 33.36, "epoch_pct": 33.36, "eta": "38:00:24", "max_grad_norm": 1.0, "loss": 0.6478689312934875, "grad_norm": 0.41378065943717957, "learning_rate": 0.00016480865755451893}
+{"ts": "2025-12-27T13:19:04", "event": "train_log", "step": 4746, "epoch": 2.0025316455696203, "progress_pct": 33.38, "epoch_pct": 33.38, "eta": "37:59:26", "max_grad_norm": 1.0, "loss": 0.6162620186805725, "grad_norm": 0.41938909888267517, "learning_rate": 0.00016477207261181162}
+{"ts": "2025-12-27T13:19:20", "event": "train_log", "step": 4748, "epoch": 2.00337552742616, "progress_pct": 33.39, "epoch_pct": 33.39, "eta": "37:58:33", "max_grad_norm": 1.0, "loss": 0.5328736901283264, "grad_norm": 0.40436387062072754, "learning_rate": 0.00016473547272743115}
+{"ts": "2025-12-27T13:19:35", "event": "train_log", "step": 4750, "epoch": 2.0042194092827006, "progress_pct": 33.4, "epoch_pct": 33.4, "eta": "37:57:36", "max_grad_norm": 1.0, "loss": 0.547049880027771, "grad_norm": 0.4148593544960022, "learning_rate": 0.00016469885790982037}
+{"ts": "2025-12-27T13:19:51", "event": "train_log", "step": 4752, "epoch": 2.0050632911392405, "progress_pct": 33.42, "epoch_pct": 33.42, "eta": "37:56:41", "max_grad_norm": 1.0, "loss": 0.5509368181228638, "grad_norm": 0.42139601707458496, "learning_rate": 0.00016466222816742564}
+{"ts": "2025-12-27T13:20:06", "event": "train_log", "step": 4754, "epoch": 2.0059071729957805, "progress_pct": 33.43, "epoch_pct": 33.43, "eta": "37:55:46", "max_grad_norm": 1.0, "loss": 0.5994526147842407, "grad_norm": 0.4509783089160919, "learning_rate": 0.00016462558350869677}
+{"ts": "2025-12-27T13:20:21", "event": "train_log", "step": 4756, "epoch": 2.006751054852321, "progress_pct": 33.45, "epoch_pct": 33.45, "eta": "37:54:49", "max_grad_norm": 1.0, "loss": 0.6348381638526917, "grad_norm": 0.4232040345668793, "learning_rate": 0.0001645889239420869}
+{"ts": "2025-12-27T13:20:36", "event": "train_log", "step": 4758, "epoch": 2.007594936708861, "progress_pct": 33.46, "epoch_pct": 33.46, "eta": "37:53:53", "max_grad_norm": 1.0, "loss": 0.5487880110740662, "grad_norm": 0.4259628355503082, "learning_rate": 0.00016455224947605274}
+{"ts": "2025-12-27T13:20:51", "event": "train_log", "step": 4760, "epoch": 2.0084388185654007, "progress_pct": 33.47, "epoch_pct": 33.47, "eta": "37:52:56", "max_grad_norm": 1.0, "loss": 0.5468270778656006, "grad_norm": 0.5485594868659973, "learning_rate": 0.0001645155601190544}
+{"ts": "2025-12-27T13:21:06", "event": "train_log", "step": 4762, "epoch": 2.009282700421941, "progress_pct": 33.49, "epoch_pct": 33.49, "eta": "37:51:59", "max_grad_norm": 1.0, "loss": 0.6165785789489746, "grad_norm": 0.4243668019771576, "learning_rate": 0.00016447885587955537}
+{"ts": "2025-12-27T13:21:20", "event": "train_log", "step": 4764, "epoch": 2.010126582278481, "progress_pct": 33.5, "epoch_pct": 33.5, "eta": "37:51:02", "max_grad_norm": 1.0, "loss": 0.5868126749992371, "grad_norm": 0.46513810753822327, "learning_rate": 0.00016444213676602263}
+{"ts": "2025-12-27T13:21:35", "event": "train_log", "step": 4766, "epoch": 2.010970464135021, "progress_pct": 33.52, "epoch_pct": 33.52, "eta": "37:50:06", "max_grad_norm": 1.0, "loss": 0.548373818397522, "grad_norm": 0.4555842876434326, "learning_rate": 0.00016440540278692654}
+{"ts": "2025-12-27T13:21:51", "event": "train_log", "step": 4768, "epoch": 2.0118143459915614, "progress_pct": 33.53, "epoch_pct": 33.53, "eta": "37:49:11", "max_grad_norm": 1.0, "loss": 0.6254384517669678, "grad_norm": 0.4402005672454834, "learning_rate": 0.000164368653950741}
+{"ts": "2025-12-27T13:22:05", "event": "train_log", "step": 4770, "epoch": 2.0126582278481013, "progress_pct": 33.54, "epoch_pct": 33.54, "eta": "37:48:14", "max_grad_norm": 1.0, "loss": 0.6335576772689819, "grad_norm": 0.5100982785224915, "learning_rate": 0.0001643318902659432}
+{"ts": "2025-12-27T13:22:20", "event": "train_log", "step": 4772, "epoch": 2.0135021097046413, "progress_pct": 33.56, "epoch_pct": 33.56, "eta": "37:47:17", "max_grad_norm": 1.0, "loss": 0.6601388454437256, "grad_norm": 0.508076548576355, "learning_rate": 0.0001642951117410138}
+{"ts": "2025-12-27T13:22:36", "event": "train_log", "step": 4774, "epoch": 2.014345991561181, "progress_pct": 33.57, "epoch_pct": 33.57, "eta": "37:46:22", "max_grad_norm": 1.0, "loss": 0.6183612942695618, "grad_norm": 0.47249099612236023, "learning_rate": 0.000164258318384437}
+{"ts": "2025-12-27T13:22:50", "event": "train_log", "step": 4776, "epoch": 2.0151898734177216, "progress_pct": 33.59, "epoch_pct": 33.59, "eta": "37:45:25", "max_grad_norm": 1.0, "loss": 0.6172632575035095, "grad_norm": 0.459249883890152, "learning_rate": 0.00016422151020470022}
+{"ts": "2025-12-27T13:23:05", "event": "train_log", "step": 4778, "epoch": 2.0160337552742615, "progress_pct": 33.6, "epoch_pct": 33.6, "eta": "37:44:28", "max_grad_norm": 1.0, "loss": 0.6392593383789062, "grad_norm": 0.44374513626098633, "learning_rate": 0.00016418468721029442}
+{"ts": "2025-12-27T13:23:19", "event": "train_log", "step": 4780, "epoch": 2.0168776371308015, "progress_pct": 33.61, "epoch_pct": 33.61, "eta": "37:43:31", "max_grad_norm": 1.0, "loss": 0.6331546902656555, "grad_norm": 0.4633971154689789, "learning_rate": 0.000164147849409714}
+{"ts": "2025-12-27T13:23:35", "event": "train_log", "step": 4782, "epoch": 2.017721518987342, "progress_pct": 33.63, "epoch_pct": 33.63, "eta": "37:42:36", "max_grad_norm": 1.0, "loss": 0.5368412137031555, "grad_norm": 0.4080882668495178, "learning_rate": 0.00016411099681145667}
+{"ts": "2025-12-27T13:23:51", "event": "train_log", "step": 4784, "epoch": 2.018565400843882, "progress_pct": 33.64, "epoch_pct": 33.64, "eta": "37:41:42", "max_grad_norm": 1.0, "loss": 0.4763215184211731, "grad_norm": 0.3852706253528595, "learning_rate": 0.00016407412942402366}
+{"ts": "2025-12-27T13:24:06", "event": "train_log", "step": 4786, "epoch": 2.0194092827004217, "progress_pct": 33.66, "epoch_pct": 33.66, "eta": "37:40:46", "max_grad_norm": 1.0, "loss": 0.6510329842567444, "grad_norm": 0.45222651958465576, "learning_rate": 0.00016403724725591958}
+{"ts": "2025-12-27T13:24:21", "event": "train_log", "step": 4788, "epoch": 2.020253164556962, "progress_pct": 33.67, "epoch_pct": 33.67, "eta": "37:39:50", "max_grad_norm": 1.0, "loss": 0.5856698751449585, "grad_norm": 0.4027794897556305, "learning_rate": 0.00016400035031565238}
+{"ts": "2025-12-27T13:24:36", "event": "train_log", "step": 4790, "epoch": 2.021097046413502, "progress_pct": 33.68, "epoch_pct": 33.68, "eta": "37:38:56", "max_grad_norm": 1.0, "loss": 0.5960753560066223, "grad_norm": 0.3780474066734314, "learning_rate": 0.00016396343861173356}
+{"ts": "2025-12-27T13:24:52", "event": "train_log", "step": 4792, "epoch": 2.021940928270042, "progress_pct": 33.7, "epoch_pct": 33.7, "eta": "37:38:02", "max_grad_norm": 1.0, "loss": 0.5696688890457153, "grad_norm": 0.4297144412994385, "learning_rate": 0.00016392651215267785}
+{"ts": "2025-12-27T13:25:07", "event": "train_log", "step": 4794, "epoch": 2.0227848101265824, "progress_pct": 33.71, "epoch_pct": 33.71, "eta": "37:37:06", "max_grad_norm": 1.0, "loss": 0.6610676050186157, "grad_norm": 0.4521540701389313, "learning_rate": 0.00016388957094700356}
+{"ts": "2025-12-27T13:25:22", "event": "train_log", "step": 4796, "epoch": 2.0236286919831223, "progress_pct": 33.73, "epoch_pct": 33.73, "eta": "37:36:11", "max_grad_norm": 1.0, "loss": 0.5387522578239441, "grad_norm": 0.41184982657432556, "learning_rate": 0.00016385261500323224}
+{"ts": "2025-12-27T13:25:37", "event": "train_log", "step": 4798, "epoch": 2.0244725738396623, "progress_pct": 33.74, "epoch_pct": 33.74, "eta": "37:35:13", "max_grad_norm": 1.0, "loss": 0.6614878177642822, "grad_norm": 0.489391565322876, "learning_rate": 0.00016381564432988897}
+{"ts": "2025-12-27T13:25:52", "event": "train_log", "step": 4800, "epoch": 2.0253164556962027, "progress_pct": 33.76, "epoch_pct": 33.76, "eta": "37:34:19", "max_grad_norm": 1.0, "loss": 0.5513694286346436, "grad_norm": 0.3741590678691864, "learning_rate": 0.00016377865893550214}
+{"ts": "2025-12-27T13:37:21", "event": "train_log", "step": 4800, "epoch": 2.0253164556962027, "progress_pct": 33.76, "epoch_pct": 33.76, "eta": "37:56:51", "max_grad_norm": 1.0, "eval_loss": 0.7633068561553955, "eval_runtime": 688.8684, "eval_samples_per_second": 3.059, "eval_steps_per_second": 3.059}
+{"ts": "2025-12-27T13:37:37", "event": "train_log", "step": 4802, "epoch": 2.0261603375527426, "progress_pct": 33.77, "epoch_pct": 33.77, "eta": "37:55:56", "max_grad_norm": 1.0, "loss": 0.5732831358909607, "grad_norm": 0.45757874846458435, "learning_rate": 0.0001637416588286036}
+{"ts": "2025-12-27T13:37:53", "event": "train_log", "step": 4804, "epoch": 2.0270042194092825, "progress_pct": 33.78, "epoch_pct": 33.78, "eta": "37:55:01", "max_grad_norm": 1.0, "loss": 0.5994226932525635, "grad_norm": 0.4009305238723755, "learning_rate": 0.00016370464401772853}
+{"ts": "2025-12-27T13:38:09", "event": "train_log", "step": 4806, "epoch": 2.027848101265823, "progress_pct": 33.8, "epoch_pct": 33.8, "eta": "37:54:08", "max_grad_norm": 1.0, "loss": 0.6229469776153564, "grad_norm": 0.44907137751579285, "learning_rate": 0.0001636676145114156}
+{"ts": "2025-12-27T13:38:26", "event": "train_log", "step": 4808, "epoch": 2.028691983122363, "progress_pct": 33.81, "epoch_pct": 33.81, "eta": "37:53:14", "max_grad_norm": 1.0, "loss": 0.5126086473464966, "grad_norm": 0.4153282344341278, "learning_rate": 0.00016363057031820673}
+{"ts": "2025-12-27T13:38:42", "event": "train_log", "step": 4810, "epoch": 2.029535864978903, "progress_pct": 33.83, "epoch_pct": 33.83, "eta": "37:52:21", "max_grad_norm": 1.0, "loss": 0.5779650211334229, "grad_norm": 0.41859957575798035, "learning_rate": 0.0001635935114466473}
+{"ts": "2025-12-27T13:38:59", "event": "train_log", "step": 4812, "epoch": 2.030379746835443, "progress_pct": 33.84, "epoch_pct": 33.84, "eta": "37:51:27", "max_grad_norm": 1.0, "loss": 0.5561378598213196, "grad_norm": 0.44624394178390503, "learning_rate": 0.00016355643790528617}
+{"ts": "2025-12-27T13:39:14", "event": "train_log", "step": 4814, "epoch": 2.031223628691983, "progress_pct": 33.85, "epoch_pct": 33.85, "eta": "37:50:32", "max_grad_norm": 1.0, "loss": 0.5381612181663513, "grad_norm": 0.44554266333580017, "learning_rate": 0.00016351934970267542}
+{"ts": "2025-12-27T13:39:30", "event": "train_log", "step": 4816, "epoch": 2.032067510548523, "progress_pct": 33.87, "epoch_pct": 33.87, "eta": "37:49:37", "max_grad_norm": 1.0, "loss": 0.6434502601623535, "grad_norm": 0.4529551565647125, "learning_rate": 0.00016348224684737065}
+{"ts": "2025-12-27T13:39:45", "event": "train_log", "step": 4818, "epoch": 2.0329113924050635, "progress_pct": 33.88, "epoch_pct": 33.88, "eta": "37:48:42", "max_grad_norm": 1.0, "loss": 0.6316392421722412, "grad_norm": 0.891188383102417, "learning_rate": 0.00016344512934793066}
+{"ts": "2025-12-27T13:40:02", "event": "train_log", "step": 4820, "epoch": 2.0337552742616034, "progress_pct": 33.9, "epoch_pct": 33.9, "eta": "37:47:50", "max_grad_norm": 1.0, "loss": 0.5569570064544678, "grad_norm": 0.4621472656726837, "learning_rate": 0.00016340799721291783}
+{"ts": "2025-12-27T13:40:19", "event": "train_log", "step": 4822, "epoch": 2.0345991561181433, "progress_pct": 33.91, "epoch_pct": 33.91, "eta": "37:46:56", "max_grad_norm": 1.0, "loss": 0.5403776168823242, "grad_norm": 0.38671162724494934, "learning_rate": 0.00016337085045089786}
+{"ts": "2025-12-27T13:40:34", "event": "train_log", "step": 4824, "epoch": 2.0354430379746837, "progress_pct": 33.92, "epoch_pct": 33.92, "eta": "37:46:01", "max_grad_norm": 1.0, "loss": 0.5629472732543945, "grad_norm": 0.3896256685256958, "learning_rate": 0.00016333368907043972}
+{"ts": "2025-12-27T13:40:49", "event": "train_log", "step": 4826, "epoch": 2.0362869198312237, "progress_pct": 33.94, "epoch_pct": 33.94, "eta": "37:45:06", "max_grad_norm": 1.0, "loss": 0.5740808844566345, "grad_norm": 0.4890784025192261, "learning_rate": 0.00016329651308011583}
+{"ts": "2025-12-27T13:41:05", "event": "train_log", "step": 4828, "epoch": 2.0371308016877636, "progress_pct": 33.95, "epoch_pct": 33.95, "eta": "37:44:11", "max_grad_norm": 1.0, "loss": 0.6362084746360779, "grad_norm": 0.493118554353714, "learning_rate": 0.00016325932248850207}
+{"ts": "2025-12-27T13:41:21", "event": "train_log", "step": 4830, "epoch": 2.037974683544304, "progress_pct": 33.97, "epoch_pct": 33.97, "eta": "37:43:16", "max_grad_norm": 1.0, "loss": 0.5512531399726868, "grad_norm": 0.4112105965614319, "learning_rate": 0.0001632221173041775}
+{"ts": "2025-12-27T13:41:36", "event": "train_log", "step": 4832, "epoch": 2.038818565400844, "progress_pct": 33.98, "epoch_pct": 33.98, "eta": "37:42:21", "max_grad_norm": 1.0, "loss": 0.5507554411888123, "grad_norm": 0.48197677731513977, "learning_rate": 0.00016318489753572464}
+{"ts": "2025-12-27T13:41:52", "event": "train_log", "step": 4834, "epoch": 2.039662447257384, "progress_pct": 33.99, "epoch_pct": 33.99, "eta": "37:41:27", "max_grad_norm": 1.0, "loss": 0.7607873678207397, "grad_norm": 0.5130576491355896, "learning_rate": 0.00016314766319172947}
+{"ts": "2025-12-27T13:42:08", "event": "train_log", "step": 4836, "epoch": 2.0405063291139243, "progress_pct": 34.01, "epoch_pct": 34.01, "eta": "37:40:32", "max_grad_norm": 1.0, "loss": 0.6044831275939941, "grad_norm": 0.4454704523086548, "learning_rate": 0.00016311041428078112}
+{"ts": "2025-12-27T13:42:24", "event": "train_log", "step": 4838, "epoch": 2.041350210970464, "progress_pct": 34.02, "epoch_pct": 34.02, "eta": "37:39:39", "max_grad_norm": 1.0, "loss": 0.587941586971283, "grad_norm": 0.4688933789730072, "learning_rate": 0.00016307315081147227}
+{"ts": "2025-12-27T13:42:39", "event": "train_log", "step": 4840, "epoch": 2.042194092827004, "progress_pct": 34.04, "epoch_pct": 34.04, "eta": "37:38:42", "max_grad_norm": 1.0, "loss": 0.5822482109069824, "grad_norm": 0.45465248823165894, "learning_rate": 0.0001630358727923989}
+{"ts": "2025-12-27T13:42:54", "event": "train_log", "step": 4842, "epoch": 2.043037974683544, "progress_pct": 34.05, "epoch_pct": 34.05, "eta": "37:37:48", "max_grad_norm": 1.0, "loss": 0.584092378616333, "grad_norm": 0.4670163691043854, "learning_rate": 0.00016299858023216028}
+{"ts": "2025-12-27T13:43:11", "event": "train_log", "step": 4844, "epoch": 2.0438818565400845, "progress_pct": 34.06, "epoch_pct": 34.06, "eta": "37:36:56", "max_grad_norm": 1.0, "loss": 0.5876650214195251, "grad_norm": 0.43894922733306885, "learning_rate": 0.0001629612731393591}
+{"ts": "2025-12-27T13:43:26", "event": "train_log", "step": 4846, "epoch": 2.0447257383966244, "progress_pct": 34.08, "epoch_pct": 34.08, "eta": "37:36:00", "max_grad_norm": 1.0, "loss": 0.5628184676170349, "grad_norm": 0.4144253730773926, "learning_rate": 0.0001629239515226014}
+{"ts": "2025-12-27T13:43:42", "event": "train_log", "step": 4848, "epoch": 2.0455696202531644, "progress_pct": 34.09, "epoch_pct": 34.09, "eta": "37:35:06", "max_grad_norm": 1.0, "loss": 0.6156699657440186, "grad_norm": 0.47422146797180176, "learning_rate": 0.0001628866153904966}
+{"ts": "2025-12-27T13:43:59", "event": "train_log", "step": 4850, "epoch": 2.0464135021097047, "progress_pct": 34.11, "epoch_pct": 34.11, "eta": "37:34:14", "max_grad_norm": 1.0, "loss": 0.5761739015579224, "grad_norm": 0.46102508902549744, "learning_rate": 0.00016284926475165735}
+{"ts": "2025-12-27T13:44:14", "event": "train_log", "step": 4852, "epoch": 2.0472573839662447, "progress_pct": 34.12, "epoch_pct": 34.12, "eta": "37:33:17", "max_grad_norm": 1.0, "loss": 0.6326378583908081, "grad_norm": 0.4987565577030182, "learning_rate": 0.00016281189961469977}
+{"ts": "2025-12-27T13:44:28", "event": "train_log", "step": 4854, "epoch": 2.0481012658227846, "progress_pct": 34.14, "epoch_pct": 34.14, "eta": "37:32:21", "max_grad_norm": 1.0, "loss": 0.6072084903717041, "grad_norm": 0.46640047430992126, "learning_rate": 0.0001627745199882433}
+{"ts": "2025-12-27T13:44:43", "event": "train_log", "step": 4856, "epoch": 2.048945147679325, "progress_pct": 34.15, "epoch_pct": 34.15, "eta": "37:31:25", "max_grad_norm": 1.0, "loss": 0.6507929563522339, "grad_norm": 0.5362635254859924, "learning_rate": 0.00016273712588091066}
+{"ts": "2025-12-27T13:44:57", "event": "train_log", "step": 4858, "epoch": 2.049789029535865, "progress_pct": 34.16, "epoch_pct": 34.16, "eta": "37:30:28", "max_grad_norm": 1.0, "loss": 0.6264472007751465, "grad_norm": 0.459285706281662, "learning_rate": 0.00016269971730132796}
+{"ts": "2025-12-27T13:45:11", "event": "train_log", "step": 4860, "epoch": 2.050632911392405, "progress_pct": 34.18, "epoch_pct": 34.18, "eta": "37:29:31", "max_grad_norm": 1.0, "loss": 0.6185184717178345, "grad_norm": 0.4467318058013916, "learning_rate": 0.00016266229425812467}
+{"ts": "2025-12-27T13:45:27", "event": "train_log", "step": 4862, "epoch": 2.0514767932489453, "progress_pct": 34.19, "epoch_pct": 34.19, "eta": "37:28:36", "max_grad_norm": 1.0, "loss": 0.5723293423652649, "grad_norm": 0.4965224266052246, "learning_rate": 0.0001626248567599335}
+{"ts": "2025-12-27T13:45:42", "event": "train_log", "step": 4864, "epoch": 2.052320675105485, "progress_pct": 34.21, "epoch_pct": 34.21, "eta": "37:27:40", "max_grad_norm": 1.0, "loss": 0.578831672668457, "grad_norm": 0.46917736530303955, "learning_rate": 0.0001625874048153907}
+{"ts": "2025-12-27T13:45:56", "event": "train_log", "step": 4866, "epoch": 2.053164556962025, "progress_pct": 34.22, "epoch_pct": 34.22, "eta": "37:26:43", "max_grad_norm": 1.0, "loss": 0.6448227763175964, "grad_norm": 0.498439222574234, "learning_rate": 0.00016254993843313554}
+{"ts": "2025-12-27T13:46:11", "event": "train_log", "step": 4868, "epoch": 2.0540084388185655, "progress_pct": 34.23, "epoch_pct": 34.23, "eta": "37:25:47", "max_grad_norm": 1.0, "loss": 0.6240531802177429, "grad_norm": 0.4801621735095978, "learning_rate": 0.00016251245762181088}
+{"ts": "2025-12-27T13:46:26", "event": "train_log", "step": 4870, "epoch": 2.0548523206751055, "progress_pct": 34.25, "epoch_pct": 34.25, "eta": "37:24:53", "max_grad_norm": 1.0, "loss": 0.4955635070800781, "grad_norm": 0.3652440309524536, "learning_rate": 0.00016247496239006287}
+{"ts": "2025-12-27T13:46:41", "event": "train_log", "step": 4872, "epoch": 2.0556962025316454, "progress_pct": 34.26, "epoch_pct": 34.26, "eta": "37:23:58", "max_grad_norm": 1.0, "loss": 0.6184053421020508, "grad_norm": 0.504094123840332, "learning_rate": 0.00016243745274654084}
+{"ts": "2025-12-27T13:46:56", "event": "train_log", "step": 4874, "epoch": 2.056540084388186, "progress_pct": 34.28, "epoch_pct": 34.28, "eta": "37:23:02", "max_grad_norm": 1.0, "loss": 0.5592952370643616, "grad_norm": 0.4311911463737488, "learning_rate": 0.0001623999286998976}
+{"ts": "2025-12-27T13:47:11", "event": "train_log", "step": 4876, "epoch": 2.0573839662447257, "progress_pct": 34.29, "epoch_pct": 34.29, "eta": "37:22:06", "max_grad_norm": 1.0, "loss": 0.6024063229560852, "grad_norm": 0.5173681974411011, "learning_rate": 0.00016236239025878928}
+{"ts": "2025-12-27T13:47:26", "event": "train_log", "step": 4878, "epoch": 2.0582278481012657, "progress_pct": 34.3, "epoch_pct": 34.3, "eta": "37:21:12", "max_grad_norm": 1.0, "loss": 0.5788836479187012, "grad_norm": 0.41021528840065, "learning_rate": 0.0001623248374318752}
+{"ts": "2025-12-27T13:47:41", "event": "train_log", "step": 4880, "epoch": 2.059071729957806, "progress_pct": 34.32, "epoch_pct": 34.32, "eta": "37:20:16", "max_grad_norm": 1.0, "loss": 0.6215962767601013, "grad_norm": 0.5167528986930847, "learning_rate": 0.00016228727022781807}
+{"ts": "2025-12-27T13:47:55", "event": "train_log", "step": 4882, "epoch": 2.059915611814346, "progress_pct": 34.33, "epoch_pct": 34.33, "eta": "37:19:20", "max_grad_norm": 1.0, "loss": 0.5767302513122559, "grad_norm": 0.5254554152488708, "learning_rate": 0.00016224968865528395}
+{"ts": "2025-12-27T13:48:11", "event": "train_log", "step": 4884, "epoch": 2.060759493670886, "progress_pct": 34.35, "epoch_pct": 34.35, "eta": "37:18:27", "max_grad_norm": 1.0, "loss": 0.5494593381881714, "grad_norm": 0.3826330602169037, "learning_rate": 0.00016221209272294218}
+{"ts": "2025-12-27T13:48:26", "event": "train_log", "step": 4886, "epoch": 2.0616033755274263, "progress_pct": 34.36, "epoch_pct": 34.36, "eta": "37:17:32", "max_grad_norm": 1.0, "loss": 0.5395565032958984, "grad_norm": 0.48100414872169495, "learning_rate": 0.00016217448243946542}
+{"ts": "2025-12-27T13:48:42", "event": "train_log", "step": 4888, "epoch": 2.0624472573839663, "progress_pct": 34.37, "epoch_pct": 34.37, "eta": "37:16:38", "max_grad_norm": 1.0, "loss": 0.5737000107765198, "grad_norm": 0.4237285852432251, "learning_rate": 0.00016213685781352966}
+{"ts": "2025-12-27T13:48:57", "event": "train_log", "step": 4890, "epoch": 2.0632911392405062, "progress_pct": 34.39, "epoch_pct": 34.39, "eta": "37:15:44", "max_grad_norm": 1.0, "loss": 0.6150708794593811, "grad_norm": 0.46409112215042114, "learning_rate": 0.0001620992188538142}
+{"ts": "2025-12-27T13:49:12", "event": "train_log", "step": 4892, "epoch": 2.0641350210970466, "progress_pct": 34.4, "epoch_pct": 34.4, "eta": "37:14:48", "max_grad_norm": 1.0, "loss": 0.5868361592292786, "grad_norm": 0.5053852200508118, "learning_rate": 0.0001620615655690015}
+{"ts": "2025-12-27T13:49:27", "event": "train_log", "step": 4894, "epoch": 2.0649789029535865, "progress_pct": 34.42, "epoch_pct": 34.42, "eta": "37:13:54", "max_grad_norm": 1.0, "loss": 0.543509304523468, "grad_norm": 0.4552644193172455, "learning_rate": 0.00016202389796777763}
+{"ts": "2025-12-27T13:49:42", "event": "train_log", "step": 4896, "epoch": 2.0658227848101265, "progress_pct": 34.43, "epoch_pct": 34.43, "eta": "37:12:59", "max_grad_norm": 1.0, "loss": 0.6006189584732056, "grad_norm": 0.49399474263191223, "learning_rate": 0.00016198621605883164}
+{"ts": "2025-12-27T13:49:58", "event": "train_log", "step": 4898, "epoch": 2.066666666666667, "progress_pct": 34.44, "epoch_pct": 34.44, "eta": "37:12:05", "max_grad_norm": 1.0, "loss": 0.6238788962364197, "grad_norm": 0.4846300184726715, "learning_rate": 0.00016194851985085608}
+{"ts": "2025-12-27T13:50:14", "event": "train_log", "step": 4900, "epoch": 2.067510548523207, "progress_pct": 34.46, "epoch_pct": 34.46, "eta": "37:11:12", "max_grad_norm": 1.0, "loss": 0.5164834856987, "grad_norm": 0.4207721948623657, "learning_rate": 0.00016191080935254673}
+{"ts": "2025-12-27T14:01:30", "event": "train_log", "step": 4900, "epoch": 2.067510548523207, "progress_pct": 34.46, "epoch_pct": 34.46, "eta": "37:32:37", "max_grad_norm": 1.0, "eval_loss": 0.7676681280136108, "eval_runtime": 676.0104, "eval_samples_per_second": 3.117, "eval_steps_per_second": 3.117}
+{"ts": "2025-12-27T14:01:46", "event": "train_log", "step": 4902, "epoch": 2.0683544303797468, "progress_pct": 34.47, "epoch_pct": 34.47, "eta": "37:31:44", "max_grad_norm": 1.0, "loss": 0.6363024115562439, "grad_norm": 0.49213138222694397, "learning_rate": 0.00016187308457260268}
+{"ts": "2025-12-27T14:02:04", "event": "train_log", "step": 4904, "epoch": 2.0691983122362867, "progress_pct": 34.49, "epoch_pct": 34.49, "eta": "37:30:55", "max_grad_norm": 1.0, "loss": 0.5472462177276611, "grad_norm": 0.4619614779949188, "learning_rate": 0.00016183534551972634}
+{"ts": "2025-12-27T14:02:20", "event": "train_log", "step": 4906, "epoch": 2.070042194092827, "progress_pct": 34.5, "epoch_pct": 34.5, "eta": "37:30:02", "max_grad_norm": 1.0, "loss": 0.5588397979736328, "grad_norm": 0.45133984088897705, "learning_rate": 0.00016179759220262335}
+{"ts": "2025-12-27T14:02:35", "event": "train_log", "step": 4908, "epoch": 2.070886075949367, "progress_pct": 34.51, "epoch_pct": 34.51, "eta": "37:29:06", "max_grad_norm": 1.0, "loss": 0.6439877152442932, "grad_norm": 0.43284907937049866, "learning_rate": 0.00016175982463000266}
+{"ts": "2025-12-27T14:02:51", "event": "train_log", "step": 4910, "epoch": 2.071729957805907, "progress_pct": 34.53, "epoch_pct": 34.53, "eta": "37:28:11", "max_grad_norm": 1.0, "loss": 0.5987089276313782, "grad_norm": 0.4439920485019684, "learning_rate": 0.00016172204281057656}
+{"ts": "2025-12-27T14:03:08", "event": "train_log", "step": 4912, "epoch": 2.0725738396624473, "progress_pct": 34.54, "epoch_pct": 34.54, "eta": "37:27:20", "max_grad_norm": 1.0, "loss": 0.53819739818573, "grad_norm": 0.376004695892334, "learning_rate": 0.0001616842467530606}
+{"ts": "2025-12-27T14:03:24", "event": "train_log", "step": 4914, "epoch": 2.0734177215189873, "progress_pct": 34.56, "epoch_pct": 34.56, "eta": "37:26:27", "max_grad_norm": 1.0, "loss": 0.5739918947219849, "grad_norm": 0.42577606439590454, "learning_rate": 0.00016164643646617358}
+{"ts": "2025-12-27T14:03:40", "event": "train_log", "step": 4916, "epoch": 2.0742616033755272, "progress_pct": 34.57, "epoch_pct": 34.57, "eta": "37:25:32", "max_grad_norm": 1.0, "loss": 0.580957293510437, "grad_norm": 0.4607972502708435, "learning_rate": 0.00016160861195863756}
+{"ts": "2025-12-27T14:03:56", "event": "train_log", "step": 4918, "epoch": 2.0751054852320676, "progress_pct": 34.59, "epoch_pct": 34.59, "eta": "37:24:39", "max_grad_norm": 1.0, "loss": 0.6063762903213501, "grad_norm": 0.4114130735397339, "learning_rate": 0.00016157077323917802}
+{"ts": "2025-12-27T14:04:12", "event": "train_log", "step": 4920, "epoch": 2.0759493670886076, "progress_pct": 34.6, "epoch_pct": 34.6, "eta": "37:23:46", "max_grad_norm": 1.0, "loss": 0.5996072292327881, "grad_norm": 0.4415706694126129, "learning_rate": 0.00016153292031652358}
+{"ts": "2025-12-27T14:04:28", "event": "train_log", "step": 4922, "epoch": 2.0767932489451475, "progress_pct": 34.61, "epoch_pct": 34.61, "eta": "37:22:53", "max_grad_norm": 1.0, "loss": 0.6442107558250427, "grad_norm": 0.5539509654045105, "learning_rate": 0.00016149505319940617}
+{"ts": "2025-12-27T14:04:44", "event": "train_log", "step": 4924, "epoch": 2.077637130801688, "progress_pct": 34.63, "epoch_pct": 34.63, "eta": "37:21:58", "max_grad_norm": 1.0, "loss": 0.6175990104675293, "grad_norm": 0.510313868522644, "learning_rate": 0.00016145717189656105}
+{"ts": "2025-12-27T14:05:00", "event": "train_log", "step": 4926, "epoch": 2.078481012658228, "progress_pct": 34.64, "epoch_pct": 34.64, "eta": "37:21:07", "max_grad_norm": 1.0, "loss": 0.6108829975128174, "grad_norm": 0.442271888256073, "learning_rate": 0.00016141927641672666}
+{"ts": "2025-12-27T14:05:16", "event": "train_log", "step": 4928, "epoch": 2.0793248945147678, "progress_pct": 34.66, "epoch_pct": 34.66, "eta": "37:20:13", "max_grad_norm": 1.0, "loss": 0.6193580031394958, "grad_norm": 0.4499345123767853, "learning_rate": 0.00016138136676864477}
+{"ts": "2025-12-27T14:05:33", "event": "train_log", "step": 4930, "epoch": 2.080168776371308, "progress_pct": 34.67, "epoch_pct": 34.67, "eta": "37:19:22", "max_grad_norm": 1.0, "loss": 0.5283842086791992, "grad_norm": 0.36362606287002563, "learning_rate": 0.00016134344296106043}
+{"ts": "2025-12-27T14:05:49", "event": "train_log", "step": 4932, "epoch": 2.081012658227848, "progress_pct": 34.68, "epoch_pct": 34.68, "eta": "37:18:28", "max_grad_norm": 1.0, "loss": 0.5338225960731506, "grad_norm": 0.460282564163208, "learning_rate": 0.00016130550500272195}
+{"ts": "2025-12-27T14:06:05", "event": "train_log", "step": 4934, "epoch": 2.081856540084388, "progress_pct": 34.7, "epoch_pct": 34.7, "eta": "37:17:35", "max_grad_norm": 1.0, "loss": 0.6192039251327515, "grad_norm": 0.4316645860671997, "learning_rate": 0.0001612675529023808}
+{"ts": "2025-12-27T14:06:21", "event": "train_log", "step": 4936, "epoch": 2.0827004219409284, "progress_pct": 34.71, "epoch_pct": 34.71, "eta": "37:16:42", "max_grad_norm": 1.0, "loss": 0.5953296422958374, "grad_norm": 0.4333813488483429, "learning_rate": 0.0001612295866687919}
+{"ts": "2025-12-27T14:06:37", "event": "train_log", "step": 4938, "epoch": 2.0835443037974684, "progress_pct": 34.73, "epoch_pct": 34.73, "eta": "37:15:48", "max_grad_norm": 1.0, "loss": 0.6765958070755005, "grad_norm": 0.560342013835907, "learning_rate": 0.0001611916063107133}
+{"ts": "2025-12-27T14:06:54", "event": "train_log", "step": 4940, "epoch": 2.0843881856540083, "progress_pct": 34.74, "epoch_pct": 34.74, "eta": "37:14:57", "max_grad_norm": 1.0, "loss": 0.5475618243217468, "grad_norm": 0.4263869822025299, "learning_rate": 0.00016115361183690631}
+{"ts": "2025-12-27T14:07:10", "event": "train_log", "step": 4942, "epoch": 2.0852320675105487, "progress_pct": 34.75, "epoch_pct": 34.75, "eta": "37:14:05", "max_grad_norm": 1.0, "loss": 0.570278525352478, "grad_norm": 0.41631078720092773, "learning_rate": 0.00016111560325613553}
+{"ts": "2025-12-27T14:07:27", "event": "train_log", "step": 4944, "epoch": 2.0860759493670886, "progress_pct": 34.77, "epoch_pct": 34.77, "eta": "37:13:12", "max_grad_norm": 1.0, "loss": 0.5691710710525513, "grad_norm": 0.3922317326068878, "learning_rate": 0.00016107758057716884}
+{"ts": "2025-12-27T14:07:42", "event": "train_log", "step": 4946, "epoch": 2.0869198312236286, "progress_pct": 34.78, "epoch_pct": 34.78, "eta": "37:12:19", "max_grad_norm": 1.0, "loss": 0.6014654040336609, "grad_norm": 0.518889307975769, "learning_rate": 0.00016103954380877736}
+{"ts": "2025-12-27T14:07:58", "event": "train_log", "step": 4948, "epoch": 2.087763713080169, "progress_pct": 34.8, "epoch_pct": 34.8, "eta": "37:11:25", "max_grad_norm": 1.0, "loss": 0.6084938049316406, "grad_norm": 0.560624897480011, "learning_rate": 0.00016100149295973537}
+{"ts": "2025-12-27T14:08:14", "event": "train_log", "step": 4950, "epoch": 2.088607594936709, "progress_pct": 34.81, "epoch_pct": 34.81, "eta": "37:10:31", "max_grad_norm": 1.0, "loss": 0.6078231334686279, "grad_norm": 0.46847012639045715, "learning_rate": 0.00016096342803882053}
+{"ts": "2025-12-27T14:08:29", "event": "train_log", "step": 4952, "epoch": 2.089451476793249, "progress_pct": 34.82, "epoch_pct": 34.82, "eta": "37:09:38", "max_grad_norm": 1.0, "loss": 0.5664147734642029, "grad_norm": 0.43910443782806396, "learning_rate": 0.00016092534905481367}
+{"ts": "2025-12-27T14:08:45", "event": "train_log", "step": 4954, "epoch": 2.090295358649789, "progress_pct": 34.84, "epoch_pct": 34.84, "eta": "37:08:45", "max_grad_norm": 1.0, "loss": 0.5793747901916504, "grad_norm": 0.40413206815719604, "learning_rate": 0.00016088725601649887}
+{"ts": "2025-12-27T14:09:01", "event": "train_log", "step": 4956, "epoch": 2.091139240506329, "progress_pct": 34.85, "epoch_pct": 34.85, "eta": "37:07:51", "max_grad_norm": 1.0, "loss": 0.6121629476547241, "grad_norm": 0.47664782404899597, "learning_rate": 0.00016084914893266347}
+{"ts": "2025-12-27T14:09:16", "event": "train_log", "step": 4958, "epoch": 2.091983122362869, "progress_pct": 34.87, "epoch_pct": 34.87, "eta": "37:06:56", "max_grad_norm": 1.0, "loss": 0.6086400747299194, "grad_norm": 0.4575684070587158, "learning_rate": 0.00016081102781209805}
+{"ts": "2025-12-27T14:09:32", "event": "train_log", "step": 4960, "epoch": 2.0928270042194095, "progress_pct": 34.88, "epoch_pct": 34.88, "eta": "37:06:03", "max_grad_norm": 1.0, "loss": 0.5971894860267639, "grad_norm": 0.4286920726299286, "learning_rate": 0.0001607728926635964}
+{"ts": "2025-12-27T14:09:47", "event": "train_log", "step": 4962, "epoch": 2.0936708860759494, "progress_pct": 34.89, "epoch_pct": 34.89, "eta": "37:05:09", "max_grad_norm": 1.0, "loss": 0.6195604801177979, "grad_norm": 0.5314067006111145, "learning_rate": 0.00016073474349595555}
+{"ts": "2025-12-27T14:10:02", "event": "train_log", "step": 4964, "epoch": 2.0945147679324894, "progress_pct": 34.91, "epoch_pct": 34.91, "eta": "37:04:15", "max_grad_norm": 1.0, "loss": 0.6141531467437744, "grad_norm": 0.5221946239471436, "learning_rate": 0.00016069658031797585}
+{"ts": "2025-12-27T14:10:19", "event": "train_log", "step": 4966, "epoch": 2.0953586497890297, "progress_pct": 34.92, "epoch_pct": 34.92, "eta": "37:03:23", "max_grad_norm": 1.0, "loss": 0.6354720592498779, "grad_norm": 0.46874696016311646, "learning_rate": 0.00016065840313846076}
+{"ts": "2025-12-27T14:10:35", "event": "train_log", "step": 4968, "epoch": 2.0962025316455697, "progress_pct": 34.94, "epoch_pct": 34.94, "eta": "37:02:30", "max_grad_norm": 1.0, "loss": 0.5759334564208984, "grad_norm": 0.44518616795539856, "learning_rate": 0.00016062021196621703}
+{"ts": "2025-12-27T14:10:52", "event": "train_log", "step": 4970, "epoch": 2.0970464135021096, "progress_pct": 34.95, "epoch_pct": 34.95, "eta": "37:01:40", "max_grad_norm": 1.0, "loss": 0.5387811064720154, "grad_norm": 0.44092273712158203, "learning_rate": 0.00016058200681005462}
+{"ts": "2025-12-27T14:11:08", "event": "train_log", "step": 4972, "epoch": 2.09789029535865, "progress_pct": 34.96, "epoch_pct": 34.96, "eta": "37:00:47", "max_grad_norm": 1.0, "loss": 0.5712288022041321, "grad_norm": 0.4952426552772522, "learning_rate": 0.00016054378767878678}
+{"ts": "2025-12-27T14:11:25", "event": "train_log", "step": 4974, "epoch": 2.09873417721519, "progress_pct": 34.98, "epoch_pct": 34.98, "eta": "36:59:56", "max_grad_norm": 1.0, "loss": 0.6320934891700745, "grad_norm": 0.4890255331993103, "learning_rate": 0.00016050555458122984}
+{"ts": "2025-12-27T14:11:39", "event": "train_log", "step": 4976, "epoch": 2.09957805907173, "progress_pct": 34.99, "epoch_pct": 34.99, "eta": "36:59:01", "max_grad_norm": 1.0, "loss": 0.6141278147697449, "grad_norm": 0.5182738304138184, "learning_rate": 0.00016046730752620352}
+{"ts": "2025-12-27T14:11:54", "event": "train_log", "step": 4978, "epoch": 2.10042194092827, "progress_pct": 35.01, "epoch_pct": 35.01, "eta": "36:58:06", "max_grad_norm": 1.0, "loss": 0.6079437732696533, "grad_norm": 0.5344594717025757, "learning_rate": 0.00016042904652253064}
+{"ts": "2025-12-27T14:12:10", "event": "train_log", "step": 4980, "epoch": 2.1012658227848102, "progress_pct": 35.02, "epoch_pct": 35.02, "eta": "36:57:13", "max_grad_norm": 1.0, "loss": 0.5634552836418152, "grad_norm": 0.436641126871109, "learning_rate": 0.0001603907715790373}
+{"ts": "2025-12-27T14:12:26", "event": "train_log", "step": 4982, "epoch": 2.10210970464135, "progress_pct": 35.04, "epoch_pct": 35.04, "eta": "36:56:20", "max_grad_norm": 1.0, "loss": 0.5707821846008301, "grad_norm": 0.44045236706733704, "learning_rate": 0.00016035248270455273}
+{"ts": "2025-12-27T14:12:42", "event": "train_log", "step": 4984, "epoch": 2.10295358649789, "progress_pct": 35.05, "epoch_pct": 35.05, "eta": "36:55:28", "max_grad_norm": 1.0, "loss": 0.5635452270507812, "grad_norm": 0.43434852361679077, "learning_rate": 0.00016031417990790953}
+{"ts": "2025-12-27T14:12:57", "event": "train_log", "step": 4986, "epoch": 2.1037974683544305, "progress_pct": 35.06, "epoch_pct": 35.06, "eta": "36:54:33", "max_grad_norm": 1.0, "loss": 0.632486879825592, "grad_norm": 0.5538138747215271, "learning_rate": 0.00016027586319794341}
+{"ts": "2025-12-27T14:13:12", "event": "train_log", "step": 4988, "epoch": 2.1046413502109704, "progress_pct": 35.08, "epoch_pct": 35.08, "eta": "36:53:39", "max_grad_norm": 1.0, "loss": 0.6509168744087219, "grad_norm": 0.5425865054130554, "learning_rate": 0.00016023753258349323}
+{"ts": "2025-12-27T14:13:27", "event": "train_log", "step": 4990, "epoch": 2.1054852320675104, "progress_pct": 35.09, "epoch_pct": 35.09, "eta": "36:52:45", "max_grad_norm": 1.0, "loss": 0.615444540977478, "grad_norm": 0.4603687524795532, "learning_rate": 0.0001601991880734012}
+{"ts": "2025-12-27T14:13:42", "event": "train_log", "step": 4992, "epoch": 2.1063291139240508, "progress_pct": 35.11, "epoch_pct": 35.11, "eta": "36:51:52", "max_grad_norm": 1.0, "loss": 0.684465765953064, "grad_norm": 0.4866391718387604, "learning_rate": 0.0001601608296765126}
+{"ts": "2025-12-27T14:13:57", "event": "train_log", "step": 4994, "epoch": 2.1071729957805907, "progress_pct": 35.12, "epoch_pct": 35.12, "eta": "36:50:57", "max_grad_norm": 1.0, "loss": 0.5981772541999817, "grad_norm": 0.47539305686950684, "learning_rate": 0.000160122457401676}
+{"ts": "2025-12-27T14:14:14", "event": "train_log", "step": 4996, "epoch": 2.1080168776371306, "progress_pct": 35.13, "epoch_pct": 35.13, "eta": "36:50:06", "max_grad_norm": 1.0, "loss": 0.6041272878646851, "grad_norm": 0.4135447144508362, "learning_rate": 0.00016008407125774315}
+{"ts": "2025-12-27T14:14:30", "event": "train_log", "step": 4998, "epoch": 2.108860759493671, "progress_pct": 35.15, "epoch_pct": 35.15, "eta": "36:49:14", "max_grad_norm": 1.0, "loss": 0.6491074562072754, "grad_norm": 0.5324689745903015, "learning_rate": 0.00016004567125356905}
+{"ts": "2025-12-27T14:14:47", "event": "train_log", "step": 5000, "epoch": 2.109704641350211, "progress_pct": 35.16, "epoch_pct": 35.16, "eta": "36:48:24", "max_grad_norm": 1.0, "loss": 0.6178454756736755, "grad_norm": 0.4004698395729065, "learning_rate": 0.00016000725739801172}
+{"ts": "2025-12-27T14:25:57", "event": "train_log", "step": 5000, "epoch": 2.109704641350211, "progress_pct": 35.16, "epoch_pct": 35.16, "eta": "37:08:58", "max_grad_norm": 1.0, "eval_loss": 0.7635271549224854, "eval_runtime": 669.5049, "eval_samples_per_second": 3.147, "eval_steps_per_second": 3.147}
+{"ts": "2025-12-27T14:26:13", "event": "train_log", "step": 5002, "epoch": 2.110548523206751, "progress_pct": 35.18, "epoch_pct": 35.18, "eta": "37:08:07", "max_grad_norm": 1.0, "loss": 0.6120179295539856, "grad_norm": 0.37075328826904297, "learning_rate": 0.00015996882969993262}
+{"ts": "2025-12-27T14:26:29", "event": "train_log", "step": 5004, "epoch": 2.1113924050632913, "progress_pct": 35.19, "epoch_pct": 35.19, "eta": "37:07:12", "max_grad_norm": 1.0, "loss": 0.5534689426422119, "grad_norm": 0.40976548194885254, "learning_rate": 0.0001599303881681962}
+{"ts": "2025-12-27T14:26:43", "event": "train_log", "step": 5006, "epoch": 2.1122362869198312, "progress_pct": 35.2, "epoch_pct": 35.2, "eta": "37:06:17", "max_grad_norm": 1.0, "loss": 0.6165696978569031, "grad_norm": 0.5793944597244263, "learning_rate": 0.00015989193281167022}
+{"ts": "2025-12-27T14:26:59", "event": "train_log", "step": 5008, "epoch": 2.113080168776371, "progress_pct": 35.22, "epoch_pct": 35.22, "eta": "37:05:24", "max_grad_norm": 1.0, "loss": 0.5813543796539307, "grad_norm": 0.501341700553894, "learning_rate": 0.00015985346363922562}
+{"ts": "2025-12-27T14:27:15", "event": "train_log", "step": 5010, "epoch": 2.1139240506329116, "progress_pct": 35.23, "epoch_pct": 35.23, "eta": "37:04:31", "max_grad_norm": 1.0, "loss": 0.6293289661407471, "grad_norm": 0.41892293095588684, "learning_rate": 0.00015981498065973642}
+{"ts": "2025-12-27T14:27:30", "event": "train_log", "step": 5012, "epoch": 2.1147679324894515, "progress_pct": 35.25, "epoch_pct": 35.25, "eta": "37:03:35", "max_grad_norm": 1.0, "loss": 0.621943473815918, "grad_norm": 0.5439590215682983, "learning_rate": 0.00015977648388208}
+{"ts": "2025-12-27T14:27:44", "event": "train_log", "step": 5014, "epoch": 2.1156118143459914, "progress_pct": 35.26, "epoch_pct": 35.26, "eta": "37:02:40", "max_grad_norm": 1.0, "loss": 0.6048483848571777, "grad_norm": 0.4917672276496887, "learning_rate": 0.00015973797331513673}
+{"ts": "2025-12-27T14:27:59", "event": "train_log", "step": 5016, "epoch": 2.116455696202532, "progress_pct": 35.27, "epoch_pct": 35.27, "eta": "37:01:44", "max_grad_norm": 1.0, "loss": 0.5886700749397278, "grad_norm": 0.5026940107345581, "learning_rate": 0.00015969944896779034}
+{"ts": "2025-12-27T14:28:14", "event": "train_log", "step": 5018, "epoch": 2.1172995780590718, "progress_pct": 35.29, "epoch_pct": 35.29, "eta": "37:00:51", "max_grad_norm": 1.0, "loss": 0.6189847588539124, "grad_norm": 0.4289521872997284, "learning_rate": 0.0001596609108489276}
+{"ts": "2025-12-27T14:28:29", "event": "train_log", "step": 5020, "epoch": 2.1181434599156117, "progress_pct": 35.3, "epoch_pct": 35.3, "eta": "36:59:56", "max_grad_norm": 1.0, "loss": 0.6217852830886841, "grad_norm": 0.4428614675998688, "learning_rate": 0.00015962235896743854}
+{"ts": "2025-12-27T14:28:44", "event": "train_log", "step": 5022, "epoch": 2.118987341772152, "progress_pct": 35.32, "epoch_pct": 35.32, "eta": "36:59:01", "max_grad_norm": 1.0, "loss": 0.6325339078903198, "grad_norm": 0.46286383271217346, "learning_rate": 0.00015958379333221636}
+{"ts": "2025-12-27T14:28:59", "event": "train_log", "step": 5024, "epoch": 2.119831223628692, "progress_pct": 35.33, "epoch_pct": 35.33, "eta": "36:58:07", "max_grad_norm": 1.0, "loss": 0.6053416728973389, "grad_norm": 0.4348789155483246, "learning_rate": 0.00015954521395215735}
+{"ts": "2025-12-27T14:29:14", "event": "train_log", "step": 5026, "epoch": 2.120675105485232, "progress_pct": 35.34, "epoch_pct": 35.34, "eta": "36:57:12", "max_grad_norm": 1.0, "loss": 0.5830395817756653, "grad_norm": 0.43629688024520874, "learning_rate": 0.00015950662083616109}
+{"ts": "2025-12-27T14:29:29", "event": "train_log", "step": 5028, "epoch": 2.1215189873417724, "progress_pct": 35.36, "epoch_pct": 35.36, "eta": "36:56:18", "max_grad_norm": 1.0, "loss": 0.6273260116577148, "grad_norm": 0.45533162355422974, "learning_rate": 0.00015946801399313023}
+{"ts": "2025-12-27T14:29:44", "event": "train_log", "step": 5030, "epoch": 2.1223628691983123, "progress_pct": 35.37, "epoch_pct": 35.37, "eta": "36:55:24", "max_grad_norm": 1.0, "loss": 0.6343342661857605, "grad_norm": 0.5445238351821899, "learning_rate": 0.00015942939343197064}
+{"ts": "2025-12-27T14:29:59", "event": "train_log", "step": 5032, "epoch": 2.1232067510548522, "progress_pct": 35.39, "epoch_pct": 35.39, "eta": "36:54:30", "max_grad_norm": 1.0, "loss": 0.6357948780059814, "grad_norm": 0.4698406457901001, "learning_rate": 0.0001593907591615914}
+{"ts": "2025-12-27T14:30:14", "event": "train_log", "step": 5034, "epoch": 2.124050632911392, "progress_pct": 35.4, "epoch_pct": 35.4, "eta": "36:53:35", "max_grad_norm": 1.0, "loss": 0.6478350162506104, "grad_norm": 0.5374915599822998, "learning_rate": 0.0001593521111909046}
+{"ts": "2025-12-27T14:30:30", "event": "train_log", "step": 5036, "epoch": 2.1248945147679326, "progress_pct": 35.41, "epoch_pct": 35.41, "eta": "36:52:42", "max_grad_norm": 1.0, "loss": 0.5516082048416138, "grad_norm": 0.4432879388332367, "learning_rate": 0.00015931344952882563}
+{"ts": "2025-12-27T14:30:45", "event": "train_log", "step": 5038, "epoch": 2.1257383966244725, "progress_pct": 35.43, "epoch_pct": 35.43, "eta": "36:51:48", "max_grad_norm": 1.0, "loss": 0.751211404800415, "grad_norm": 0.5592076778411865, "learning_rate": 0.00015927477418427302}
+{"ts": "2025-12-27T14:31:00", "event": "train_log", "step": 5040, "epoch": 2.1265822784810124, "progress_pct": 35.44, "epoch_pct": 35.44, "eta": "36:50:55", "max_grad_norm": 1.0, "loss": 0.6150388717651367, "grad_norm": 0.47873231768608093, "learning_rate": 0.00015923608516616838}
+{"ts": "2025-12-27T14:31:16", "event": "train_log", "step": 5042, "epoch": 2.127426160337553, "progress_pct": 35.46, "epoch_pct": 35.46, "eta": "36:50:01", "max_grad_norm": 1.0, "loss": 0.6196672320365906, "grad_norm": 0.4760212302207947, "learning_rate": 0.00015919738248343653}
+{"ts": "2025-12-27T14:31:30", "event": "train_log", "step": 5044, "epoch": 2.1282700421940928, "progress_pct": 35.47, "epoch_pct": 35.47, "eta": "36:49:06", "max_grad_norm": 1.0, "loss": 0.6550422310829163, "grad_norm": 0.4995400011539459, "learning_rate": 0.00015915866614500546}
+{"ts": "2025-12-27T14:31:47", "event": "train_log", "step": 5046, "epoch": 2.1291139240506327, "progress_pct": 35.49, "epoch_pct": 35.49, "eta": "36:48:14", "max_grad_norm": 1.0, "loss": 0.6072712540626526, "grad_norm": 0.495466023683548, "learning_rate": 0.00015911993615980627}
+{"ts": "2025-12-27T14:32:01", "event": "train_log", "step": 5048, "epoch": 2.129957805907173, "progress_pct": 35.5, "epoch_pct": 35.5, "eta": "36:47:20", "max_grad_norm": 1.0, "loss": 0.6082064509391785, "grad_norm": 0.4815806448459625, "learning_rate": 0.00015908119253677322}
+{"ts": "2025-12-27T14:32:16", "event": "train_log", "step": 5050, "epoch": 2.130801687763713, "progress_pct": 35.51, "epoch_pct": 35.51, "eta": "36:46:25", "max_grad_norm": 1.0, "loss": 0.5966012477874756, "grad_norm": 0.43178048729896545, "learning_rate": 0.00015904243528484373}
+{"ts": "2025-12-27T14:32:33", "event": "train_log", "step": 5052, "epoch": 2.131645569620253, "progress_pct": 35.53, "epoch_pct": 35.53, "eta": "36:45:34", "max_grad_norm": 1.0, "loss": 0.5763883590698242, "grad_norm": 0.4170011281967163, "learning_rate": 0.00015900366441295836}
+{"ts": "2025-12-27T14:32:47", "event": "train_log", "step": 5054, "epoch": 2.1324894514767934, "progress_pct": 35.54, "epoch_pct": 35.54, "eta": "36:44:40", "max_grad_norm": 1.0, "loss": 0.6827293634414673, "grad_norm": 0.48954638838768005, "learning_rate": 0.0001589648799300608}
+{"ts": "2025-12-27T14:33:04", "event": "train_log", "step": 5056, "epoch": 2.1333333333333333, "progress_pct": 35.56, "epoch_pct": 35.56, "eta": "36:43:48", "max_grad_norm": 1.0, "loss": 0.5683901309967041, "grad_norm": 0.42273828387260437, "learning_rate": 0.00015892608184509788}
+{"ts": "2025-12-27T14:33:19", "event": "train_log", "step": 5058, "epoch": 2.1341772151898732, "progress_pct": 35.57, "epoch_pct": 35.57, "eta": "36:42:55", "max_grad_norm": 1.0, "loss": 0.5256756544113159, "grad_norm": 0.43793967366218567, "learning_rate": 0.00015888727016701956}
+{"ts": "2025-12-27T14:33:34", "event": "train_log", "step": 5060, "epoch": 2.1350210970464136, "progress_pct": 35.58, "epoch_pct": 35.58, "eta": "36:42:01", "max_grad_norm": 1.0, "loss": 0.5872990489006042, "grad_norm": 0.49315446615219116, "learning_rate": 0.000158848444904779}
+{"ts": "2025-12-27T14:33:50", "event": "train_log", "step": 5062, "epoch": 2.1358649789029536, "progress_pct": 35.6, "epoch_pct": 35.6, "eta": "36:41:07", "max_grad_norm": 1.0, "loss": 0.6437247395515442, "grad_norm": 0.47955402731895447, "learning_rate": 0.00015880960606733235}
+{"ts": "2025-12-27T14:34:05", "event": "train_log", "step": 5064, "epoch": 2.1367088607594935, "progress_pct": 35.61, "epoch_pct": 35.61, "eta": "36:40:14", "max_grad_norm": 1.0, "loss": 0.6244118809700012, "grad_norm": 0.49625128507614136, "learning_rate": 0.0001587707536636391}
+{"ts": "2025-12-27T14:34:20", "event": "train_log", "step": 5066, "epoch": 2.137552742616034, "progress_pct": 35.63, "epoch_pct": 35.63, "eta": "36:39:20", "max_grad_norm": 1.0, "loss": 0.6575133800506592, "grad_norm": 0.5150440335273743, "learning_rate": 0.00015873188770266166}
+{"ts": "2025-12-27T14:34:35", "event": "train_log", "step": 5068, "epoch": 2.138396624472574, "progress_pct": 35.64, "epoch_pct": 35.64, "eta": "36:38:27", "max_grad_norm": 1.0, "loss": 0.6086075305938721, "grad_norm": 0.5271725654602051, "learning_rate": 0.0001586930081933657}
+{"ts": "2025-12-27T14:34:51", "event": "train_log", "step": 5070, "epoch": 2.1392405063291138, "progress_pct": 35.65, "epoch_pct": 35.65, "eta": "36:37:34", "max_grad_norm": 1.0, "loss": 0.6213793158531189, "grad_norm": 0.4663483202457428, "learning_rate": 0.00015865411514471997}
+{"ts": "2025-12-27T14:35:06", "event": "train_log", "step": 5072, "epoch": 2.140084388185654, "progress_pct": 35.67, "epoch_pct": 35.67, "eta": "36:36:41", "max_grad_norm": 1.0, "loss": 0.5942427515983582, "grad_norm": 0.44225719571113586, "learning_rate": 0.00015861520856569634}
+{"ts": "2025-12-27T14:35:23", "event": "train_log", "step": 5074, "epoch": 2.140928270042194, "progress_pct": 35.68, "epoch_pct": 35.68, "eta": "36:35:51", "max_grad_norm": 1.0, "loss": 0.5674160122871399, "grad_norm": 0.41664719581604004, "learning_rate": 0.00015857628846526985}
+{"ts": "2025-12-27T14:35:37", "event": "train_log", "step": 5076, "epoch": 2.141772151898734, "progress_pct": 35.7, "epoch_pct": 35.7, "eta": "36:34:56", "max_grad_norm": 1.0, "loss": 0.6162048578262329, "grad_norm": 0.47769808769226074, "learning_rate": 0.00015853735485241857}
+{"ts": "2025-12-27T14:35:53", "event": "train_log", "step": 5078, "epoch": 2.1426160337552744, "progress_pct": 35.71, "epoch_pct": 35.71, "eta": "36:34:04", "max_grad_norm": 1.0, "loss": 0.619512677192688, "grad_norm": 0.46560928225517273, "learning_rate": 0.00015849840773612375}
+{"ts": "2025-12-27T14:36:09", "event": "train_log", "step": 5080, "epoch": 2.1434599156118144, "progress_pct": 35.72, "epoch_pct": 35.72, "eta": "36:33:12", "max_grad_norm": 1.0, "loss": 0.6183180809020996, "grad_norm": 0.4625875949859619, "learning_rate": 0.00015845944712536975}
+{"ts": "2025-12-27T14:36:26", "event": "train_log", "step": 5082, "epoch": 2.1443037974683543, "progress_pct": 35.74, "epoch_pct": 35.74, "eta": "36:32:21", "max_grad_norm": 1.0, "loss": 0.548591673374176, "grad_norm": 0.3870140314102173, "learning_rate": 0.00015842047302914406}
+{"ts": "2025-12-27T14:36:43", "event": "train_log", "step": 5084, "epoch": 2.1451476793248947, "progress_pct": 35.75, "epoch_pct": 35.75, "eta": "36:31:32", "max_grad_norm": 1.0, "loss": 0.6416457295417786, "grad_norm": 0.48335012793540955, "learning_rate": 0.0001583814854564372}
+{"ts": "2025-12-27T14:36:58", "event": "train_log", "step": 5086, "epoch": 2.1459915611814346, "progress_pct": 35.77, "epoch_pct": 35.77, "eta": "36:30:39", "max_grad_norm": 1.0, "loss": 0.6126701235771179, "grad_norm": 0.5226625204086304, "learning_rate": 0.0001583424844162429}
+{"ts": "2025-12-27T14:37:15", "event": "train_log", "step": 5088, "epoch": 2.1468354430379746, "progress_pct": 35.78, "epoch_pct": 35.78, "eta": "36:29:48", "max_grad_norm": 1.0, "loss": 0.6021472215652466, "grad_norm": 0.5319521427154541, "learning_rate": 0.0001583034699175579}
+{"ts": "2025-12-27T14:37:31", "event": "train_log", "step": 5090, "epoch": 2.147679324894515, "progress_pct": 35.79, "epoch_pct": 35.79, "eta": "36:28:57", "max_grad_norm": 1.0, "loss": 0.6050764918327332, "grad_norm": 0.49504420161247253, "learning_rate": 0.00015826444196938218}
+{"ts": "2025-12-27T14:37:47", "event": "train_log", "step": 5092, "epoch": 2.148523206751055, "progress_pct": 35.81, "epoch_pct": 35.81, "eta": "36:28:06", "max_grad_norm": 1.0, "loss": 0.592272162437439, "grad_norm": 0.5531564950942993, "learning_rate": 0.00015822540058071864}
+{"ts": "2025-12-27T14:38:03", "event": "train_log", "step": 5094, "epoch": 2.149367088607595, "progress_pct": 35.82, "epoch_pct": 35.82, "eta": "36:27:14", "max_grad_norm": 1.0, "loss": 0.6607300043106079, "grad_norm": 0.44624611735343933, "learning_rate": 0.00015818634576057346}
+{"ts": "2025-12-27T14:38:20", "event": "train_log", "step": 5096, "epoch": 2.1502109704641352, "progress_pct": 35.84, "epoch_pct": 35.84, "eta": "36:26:23", "max_grad_norm": 1.0, "loss": 0.6294105648994446, "grad_norm": 0.46433159708976746, "learning_rate": 0.00015814727751795578}
+{"ts": "2025-12-27T14:38:35", "event": "train_log", "step": 5098, "epoch": 2.151054852320675, "progress_pct": 35.85, "epoch_pct": 35.85, "eta": "36:25:30", "max_grad_norm": 1.0, "loss": 0.5308064818382263, "grad_norm": 0.46241268515586853, "learning_rate": 0.00015810819586187792}
+{"ts": "2025-12-27T14:38:51", "event": "train_log", "step": 5100, "epoch": 2.151898734177215, "progress_pct": 35.86, "epoch_pct": 35.86, "eta": "36:24:38", "max_grad_norm": 1.0, "loss": 0.6138747930526733, "grad_norm": 0.48821282386779785, "learning_rate": 0.00015806910080135527}
+{"ts": "2025-12-27T14:50:12", "event": "train_log", "step": 5100, "epoch": 2.151898734177215, "progress_pct": 35.86, "epoch_pct": 35.86, "eta": "36:44:57", "max_grad_norm": 1.0, "eval_loss": 0.7654595971107483, "eval_runtime": 681.4966, "eval_samples_per_second": 3.092, "eval_steps_per_second": 3.092}
+{"ts": "2025-12-27T14:50:12", "event": "train_log", "step": 5100, "epoch": 2.151898734177215, "progress_pct": 35.86, "epoch_pct": 35.86, "eta": "36:44:57", "max_grad_norm": 1.0, "train_runtime": 73982.3221, "train_samples_per_second": 1.538, "train_steps_per_second": 0.192, "total_flos": 3.049019256156893e+18, "train_loss": 0.8076860591944526}
+{"ts": "2025-12-27T15:01:27", "event": "train_log", "step": 5100, "epoch": 2.151898734177215, "progress_pct": 35.86, "epoch_pct": 35.86, "eta": "37:05:04", "max_grad_norm": 1.0, "eval_loss": 0.7600579857826233, "eval_runtime": 674.048, "eval_samples_per_second": 3.126, "eval_steps_per_second": 3.126}
diff --git a/sft_qwen_14B_v2/wandb/debug-internal.log b/sft_qwen_14B_v2/wandb/debug-internal.log
new file mode 100644
index 0000000000000000000000000000000000000000..15b2d6d11efb0a8c278e35049a993476ffa1baaf
--- /dev/null
+++ b/sft_qwen_14B_v2/wandb/debug-internal.log
@@ -0,0 +1,12 @@
+{"time":"2025-12-26T18:15:45.044149374Z","level":"INFO","msg":"stream: starting","core version":"0.23.1"}
+{"time":"2025-12-26T18:15:45.212148231Z","level":"INFO","msg":"stream: created new stream","id":"upub1jan"}
+{"time":"2025-12-26T18:15:45.212312297Z","level":"INFO","msg":"handler: started","stream_id":"upub1jan"}
+{"time":"2025-12-26T18:15:45.212463318Z","level":"INFO","msg":"stream: started","id":"upub1jan"}
+{"time":"2025-12-26T18:15:45.212498387Z","level":"INFO","msg":"writer: started","stream_id":"upub1jan"}
+{"time":"2025-12-26T18:15:45.212503642Z","level":"INFO","msg":"sender: started","stream_id":"upub1jan"}
+{"time":"2025-12-27T15:02:32.177112089Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
+{"time":"2025-12-27T15:02:32.275585886Z","level":"INFO","msg":"handler: operation stats","stats":{}}
+{"time":"2025-12-27T15:02:32.280819216Z","level":"INFO","msg":"stream: closing","id":"upub1jan"}
+{"time":"2025-12-27T15:02:32.280845154Z","level":"INFO","msg":"handler: closed","stream_id":"upub1jan"}
+{"time":"2025-12-27T15:02:32.280896164Z","level":"INFO","msg":"sender: closed","stream_id":"upub1jan"}
+{"time":"2025-12-27T15:02:32.280915121Z","level":"INFO","msg":"stream: closed","id":"upub1jan"}
diff --git a/sft_qwen_14B_v2/wandb/debug.log b/sft_qwen_14B_v2/wandb/debug.log
new file mode 100644
index 0000000000000000000000000000000000000000..87293499cc1e7f690ab9960d95a2b5911253be42
--- /dev/null
+++ b/sft_qwen_14B_v2/wandb/debug.log
@@ -0,0 +1,29 @@
+2025-12-26 18:15:44,766 INFO    MainThread:194421 [wandb_setup.py:_flush():80] Current SDK version is 0.23.1
+2025-12-26 18:15:44,767 INFO    MainThread:194421 [wandb_setup.py:_flush():80] Configure stats pid to 194421
+2025-12-26 18:15:44,767 INFO    MainThread:194421 [wandb_setup.py:_flush():80] Loading settings from /root/.config/wandb/settings
+2025-12-26 18:15:44,767 INFO    MainThread:194421 [wandb_setup.py:_flush():80] Loading settings from /workspace/wandb/settings
+2025-12-26 18:15:44,767 INFO    MainThread:194421 [wandb_setup.py:_flush():80] Loading settings from environment variables
+2025-12-26 18:15:44,767 INFO    MainThread:194421 [wandb_init.py:setup_run_log_directory():714] Logging user logs to task2file/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/logs/debug.log
+2025-12-26 18:15:44,767 INFO    MainThread:194421 [wandb_init.py:setup_run_log_directory():715] Logging internal logs to task2file/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/logs/debug-internal.log
+2025-12-26 18:15:44,767 INFO    MainThread:194421 [wandb_init.py:init():841] calling init triggers
+2025-12-26 18:15:44,767 INFO    MainThread:194421 [wandb_init.py:init():846] wandb.init called with sweep_config: {}
+config: {'model': {'repo_id': './Models/Qwen2.5-Coder-14B-CPT', 'revision': None, 'base_local_dir': 'base_model', 'trust_remote_code': True, 'tokenizer_use_fast': True, 'device_map': 'auto', 'torch_dtype': 'bfloat16', 'use_4bit': False, 'bnb_4bit_quant_type': 'nf4', 'bnb_4bit_use_double_quant': False, 'bnb_4bit_compute_dtype': 'bfloat16', 'attn_implementation': None}, 'data': {'train_jsonl': 'sft_dataset.jsonl', 'eval_jsonl': None, 'eval_split_ratio': 0.1, 'instruction_field': 'instruction', 'input_field': 'input', 'output_field': 'output', 'format_type': 'custom', 'system_prompt': 'You are a Hyperswitch Rust code analyzer. Identify functions/structs that need modification for a given task.\n\n## Output Format\n\n##OUTPUT\nExplain the data flow and why each component must change:\n- Flow: [Input → Processing → Output with arrows]\n- For each component: "The [ComponentName] ([path]) must [action] because [reason]—without this, [consequence]"\n- Explain coupling between components\n\n##SELECT\nmodify::crates/path/to/file.rs::impl::ComponentName\nadd::crates/another/file.rs::function::AnotherComponent\n<EOS>\n\n## Rules\n\n1. Use full paths: `remove::crates/folder/file.rs::Type::Name`\n2. Use `::` for nested items: `status::StructName::Type::Name`\n3. Always explain "must change because" and "without this"\n3. Types of components: function, struct, enum, impl, trait\n4. If there is extra information (e.g., enum variants), include that too.\n5. Start with ##OUTPUT, end with ##SELECT, terminate with <EOS>\n\n## Example\n\n##TASK\nAdd webhook subscription support\n\n##OUTPUT\nThe webhook system routes events via EventClass enum. Flow: webhook → EventClass → handler → processing. The EventClass enum (crates/common_enums/src/enums.rs::EventClass) must add Subscriptions variant because it defines event routing—without this, subscription events cannot be processed. The SubscriptionStatus impl (crates/common_enums/src/transformers.rs::SubscriptionStatus) must map to EventType because it converts status to events—without this, status changes don\'t trigger webhooks. These are coupled: EventClass routes to handlers that use SubscriptionStatus mappings.\n\n##SELECT\ncrates/common_enums/src/enums.rs::EventClass\ncrates/common_enums/src/transformers.rs::SubscriptionStatus\n<EOS>\n', 'custom_template': '##INSTRUCTION\n{instruction}<|im_end|>\n{input}<|im_end|>\n{output}<|im_end|>', 'max_length': 2048, 'shuffle': True, 'num_proc': 4}, 'peft': {'enabled': True, 'r': 32, 'lora_alpha': 64, 'lora_dropout': 0.05, 'bias': 'none', 'target_modules': 'auto'}, 'train': {'num_train_epochs': 6, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'gradient_accumulation_steps': 8, 'learning_rate': '2e-4', 'weight_decay': 0.0, 'warmup_ratio': 0.08, 'lr_scheduler_type': 'cosine', 'optim': 'adamw_torch', 'max_grad_norm': 1.0, 'gradient_checkpointing': True, 'logging_steps': 2, 'save_strategy': 'steps', 'save_steps': 500, 'save_total_limit': 20, 'evaluation_strategy': 'steps', 'eval_steps': 100, 'load_best_model_at_end': True, 'early_stopping': {'enabled': True, 'patience': 5, 'min_delta': 0.001, 'metric': 'eval_loss', 'mode': 'min'}, 'resume_from_checkpoint': 'auto'}, 'run_dir': 'task2file/sft_qwen_14B_v2', '_wandb': {}}
+2025-12-26 18:15:44,767 INFO    MainThread:194421 [wandb_init.py:init():889] starting backend
+2025-12-26 18:15:45,035 INFO    MainThread:194421 [wandb_init.py:init():892] sending inform_init request
+2025-12-26 18:15:45,040 INFO    MainThread:194421 [wandb_init.py:init():900] backend started and connected
+2025-12-26 18:15:45,044 INFO    MainThread:194421 [wandb_init.py:init():970] updated telemetry
+2025-12-26 18:15:45,045 INFO    MainThread:194421 [wandb_init.py:init():994] communicating run to backend with 90.0 second timeout
+2025-12-26 18:15:45,420 INFO    MainThread:194421 [wandb_init.py:init():1041] starting run threads in backend
+2025-12-26 18:15:45,537 INFO    MainThread:194421 [wandb_run.py:_console_start():2521] atexit reg
+2025-12-26 18:15:45,537 INFO    MainThread:194421 [wandb_run.py:_redirect():2369] redirect: wrap_raw
+2025-12-26 18:15:45,537 INFO    MainThread:194421 [wandb_run.py:_redirect():2438] Wrapping output streams.
+2025-12-26 18:15:45,537 INFO    MainThread:194421 [wandb_run.py:_redirect():2461] Redirects installed.
+2025-12-26 18:15:45,542 INFO    MainThread:194421 [wandb_init.py:init():1081] run started, returning control to user process
+2025-12-26 18:17:10,652 INFO    MainThread:194421 [wandb_run.py:_config_callback():1396] config_cb None None {'peft_config': {'default': {'task_type': 'CAUSAL_LM', 'peft_type': 'LORA', 'auto_mapping': None, 'peft_version': '0.18.0', 'base_model_name_or_path': 'Models/Qwen2.5-Coder-14B-CPT', 'revision': None, 'inference_mode': False, 'r': 32, 'target_modules': ['o_proj', 'v_proj', 'k_proj', 'q_proj'], 'exclude_modules': None, 'lora_alpha': 64, 'lora_dropout': 0.05, 'fan_in_fan_out': False, 'bias': 'none', 'use_rslora': False, 'modules_to_save': None, 'init_lora_weights': True, 'layers_to_transform': None, 'layers_pattern': None, 'rank_pattern': {}, 'alpha_pattern': {}, 'megatron_config': None, 'megatron_core': 'megatron.core', 'trainable_token_indices': None, 'loftq_config': {}, 'eva_config': None, 'corda_config': None, 'use_dora': False, 'alora_invocation_tokens': None, 'use_qalora': False, 'qalora_group_size': 16, 'layer_replication': None, 'runtime_config': {'ephemeral_gpu_offload': False}, 'lora_bias': False, 'target_parameters': None, 'arrow_config': None, 'ensure_weight_tying': False}}, 'vocab_size': 152064, 'max_position_embeddings': 32768, 'hidden_size': 5120, 'intermediate_size': 13824, 'num_hidden_layers': 48, 'num_attention_heads': 40, 'use_sliding_window': False, 'sliding_window': None, 'max_window_layers': 48, 'num_key_value_heads': 8, 'hidden_act': 'silu', 'initializer_range': 0.02, 'rms_norm_eps': 1e-06, 'use_cache': False, 'attention_dropout': 0.0, 'layer_types': ['full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention'], 'rope_parameters': {'rope_theta': 1000000.0, 'rope_type': 'default'}, 'return_dict': True, 'output_hidden_states': False, 'dtype': 'bfloat16', 'tie_word_embeddings': False, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'architectures': ['Qwen2ForCausalLM'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'task_specific_params': None, 'problem_type': None, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 151643, 'pad_token_id': None, 'eos_token_id': 151643, 'sep_token_id': None, 'decoder_start_token_id': None, '_name_or_path': 'Models/Qwen2.5-Coder-14B-CPT', 'transformers_version': '5.0.0.dev0', 'model_type': 'qwen2', 'output_attentions': False, 'output_dir': 'task2file/sft_qwen_14B_v2/checkpoints', 'do_train': False, 'do_eval': True, 'do_predict': False, 'eval_strategy': 'steps', 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'gradient_accumulation_steps': 8, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 0.0002, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 6.0, 'max_steps': -1, 'lr_scheduler_type': 'cosine', 'lr_scheduler_kwargs': None, 'warmup_ratio': 0.08, 'warmup_steps': 0.08, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': None, 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 2, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 500, 'save_total_limit': 20, 'enable_jit_checkpoint': False, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'use_cpu': False, 'seed': 42, 'data_seed': None, 'bf16': True, 'fp16': False, 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': -1, 'ddp_backend': None, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': 100, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'run_name': None, 'disable_tqdm': False, 'remove_unused_columns': False, 'label_names': None, 'load_best_model_at_end': True, 'metric_for_best_model': 'eval_loss', 'greater_is_better': False, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'parallelism_config': None, 'deepspeed': None, 'label_smoothing_factor': 0.0, 'optim': 'adamw_torch', 'optim_args': None, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'project': 'huggingface', 'trackio_space_id': 'trackio', 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'push_to_hub': False, 'resume_from_checkpoint': None, 'hub_model_id': None, 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_always_push': False, 'hub_revision': None, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'auto_find_batch_size': False, 'full_determinism': False, 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_num_input_tokens_seen': 'no', 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': False, 'liger_kernel_config': None, 'eval_use_gather_object': False, 'average_tokens_across_devices': True}
+2025-12-26 18:17:10,660 INFO    MainThread:194421 [wandb_config.py:__setitem__():154] [no run ID] config set model/num_parameters = 14820365312 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x7ed24f0556c0>>
+2025-12-26 18:17:10,660 INFO    MainThread:194421 [wandb_run.py:_config_callback():1396] config_cb model/num_parameters 14820365312 None
+2025-12-27 15:02:31,903 INFO    MainThread:194421 [wandb_run.py:_finish():2287] finishing run sirajuddin-shaik-007/sft-training/upub1jan
+2025-12-27 15:02:31,905 INFO    MainThread:194421 [wandb_run.py:_atexit_cleanup():2486] got exitcode: 0
+2025-12-27 15:02:31,906 INFO    MainThread:194421 [wandb_run.py:_restore():2468] restore
+2025-12-27 15:02:31,906 INFO    MainThread:194421 [wandb_run.py:_restore():2474] restore done
+2025-12-27 15:02:32,279 INFO    MainThread:194421 [wandb_run.py:_footer_sync_info():3862] logging synced files
diff --git a/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/files/config.yaml b/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/files/config.yaml
new file mode 100644
index 0000000000000000000000000000000000000000..e440583ecf0f59607b4d1b21da17683f6689cb69
--- /dev/null
+++ b/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/files/config.yaml
@@ -0,0 +1,601 @@
+_name_or_path:
+    value: Models/Qwen2.5-Coder-14B-CPT
+_wandb:
+    value:
+        cli_version: 0.23.1
+        e:
+            ba9dsvvs7npkm5vvx6733495pp0yghz4:
+                args:
+                    - --config
+                    - trainer-kit/SFT-14b/config_instruct.yaml
+                codePath: trainer-kit/SFT-14b/run_instruct.py
+                codePathLocal: trainer-kit/SFT-14b/run_instruct.py
+                cpu_count: 12
+                cpu_count_logical: 24
+                cudaVersion: "13.0"
+                disk:
+                    /:
+                        total: "791251738624"
+                        used: "392925650944"
+                email: shaiksirajuddin9949@gmail.com
+                executable: /workspace/llm_finetuning_env/bin/python
+                gpu: NVIDIA A100-SXM4-80GB
+                gpu_count: 2
+                gpu_nvidia:
+                    - architecture: Ampere
+                      cudaCores: 6912
+                      memoryTotal: "85899345920"
+                      name: NVIDIA A100-SXM4-80GB
+                      uuid: GPU-989794b0-ec3b-13bf-db9f-3fbe341497ba
+                    - architecture: Ampere
+                      cudaCores: 6912
+                      memoryTotal: "85899345920"
+                      name: NVIDIA A100-SXM4-80GB
+                      uuid: GPU-3790aa64-60ef-9eac-b0b1-b278ee8c0d40
+                host: a100-2gpu-shell-session-757d587799-mfdvv
+                memory:
+                    total: "359047892992"
+                os: Linux-6.12.46+-x86_64-with-glibc2.35
+                program: /workspace/trainer-kit/SFT-14b/run_instruct.py
+                python: CPython 3.10.12
+                root: task2file/sft_qwen_14B_v2
+                startedAt: "2025-12-26T18:15:44.765252Z"
+                writerId: ba9dsvvs7npkm5vvx6733495pp0yghz4
+        m:
+            - "1": train/global_step
+              "6":
+                - 3
+              "7": []
+            - "2": '*'
+              "5": 1
+              "6":
+                - 1
+              "7": []
+        python_version: 3.10.12
+        t:
+            "1":
+                - 1
+                - 11
+                - 41
+                - 49
+                - 51
+                - 71
+                - 98
+            "2":
+                - 1
+                - 11
+                - 41
+                - 49
+                - 51
+                - 71
+                - 98
+            "3":
+                - 2
+                - 7
+                - 15
+                - 16
+                - 19
+                - 62
+                - 66
+            "4": 3.10.12
+            "5": 0.23.1
+            "6": 5.0.0.dev0
+            "9":
+                "1": transformers_trainer
+            "12": 0.23.1
+            "13": linux-x86_64
+accelerator_config:
+    value:
+        dispatch_batches: null
+        even_batches: true
+        gradient_accumulation_kwargs: null
+        non_blocking: false
+        split_batches: false
+        use_seedable_sampler: true
+adam_beta1:
+    value: 0.9
+adam_beta2:
+    value: 0.999
+adam_epsilon:
+    value: 1e-08
+add_cross_attention:
+    value: false
+architectures:
+    value:
+        - Qwen2ForCausalLM
+attention_dropout:
+    value: 0
+auto_find_batch_size:
+    value: false
+average_tokens_across_devices:
+    value: true
+batch_eval_metrics:
+    value: false
+bf16:
+    value: true
+bf16_full_eval:
+    value: false
+bos_token_id:
+    value: 151643
+chunk_size_feed_forward:
+    value: 0
+cross_attention_hidden_size:
+    value: null
+data:
+    value:
+        custom_template: |-
+            ##INSTRUCTION
+            {instruction}<|im_end|>
+            {input}<|im_end|>
+            {output}<|im_end|>
+        eval_jsonl: null
+        eval_split_ratio: 0.1
+        format_type: custom
+        input_field: input
+        instruction_field: instruction
+        max_length: 2048
+        num_proc: 4
+        output_field: output
+        shuffle: true
+        system_prompt: |
+            You are a Hyperswitch Rust code analyzer. Identify functions/structs that need modification for a given task.
+
+            ## Output Format
+
+            ##OUTPUT
+            Explain the data flow and why each component must change:
+            - Flow: [Input → Processing → Output with arrows]
+            - For each component: "The [ComponentName] ([path]) must [action] because [reason]—without this, [consequence]"
+            - Explain coupling between components
+
+            ##SELECT
+            modify::crates/path/to/file.rs::impl::ComponentName
+            add::crates/another/file.rs::function::AnotherComponent
+            <EOS>
+
+            ## Rules
+
+            1. Use full paths: `remove::crates/folder/file.rs::Type::Name`
+            2. Use `::` for nested items: `status::StructName::Type::Name`
+            3. Always explain "must change because" and "without this"
+            3. Types of components: function, struct, enum, impl, trait
+            4. If there is extra information (e.g., enum variants), include that too.
+            5. Start with ##OUTPUT, end with ##SELECT, terminate with <EOS>
+
+            ## Example
+
+            ##TASK
+            Add webhook subscription support
+
+            ##OUTPUT
+            The webhook system routes events via EventClass enum. Flow: webhook → EventClass → handler → processing. The EventClass enum (crates/common_enums/src/enums.rs::EventClass) must add Subscriptions variant because it defines event routing—without this, subscription events cannot be processed. The SubscriptionStatus impl (crates/common_enums/src/transformers.rs::SubscriptionStatus) must map to EventType because it converts status to events—without this, status changes don't trigger webhooks. These are coupled: EventClass routes to handlers that use SubscriptionStatus mappings.
+
+            ##SELECT
+            crates/common_enums/src/enums.rs::EventClass
+            crates/common_enums/src/transformers.rs::SubscriptionStatus
+            <EOS>
+        train_jsonl: sft_dataset.jsonl
+data_seed:
+    value: null
+dataloader_drop_last:
+    value: false
+dataloader_num_workers:
+    value: 0
+dataloader_persistent_workers:
+    value: false
+dataloader_pin_memory:
+    value: true
+dataloader_prefetch_factor:
+    value: null
+ddp_backend:
+    value: null
+ddp_broadcast_buffers:
+    value: null
+ddp_bucket_cap_mb:
+    value: null
+ddp_find_unused_parameters:
+    value: null
+ddp_timeout:
+    value: 1800
+debug:
+    value: []
+decoder_start_token_id:
+    value: null
+deepspeed:
+    value: null
+disable_tqdm:
+    value: false
+do_eval:
+    value: true
+do_predict:
+    value: false
+do_train:
+    value: false
+dtype:
+    value: bfloat16
+enable_jit_checkpoint:
+    value: false
+eos_token_id:
+    value: 151643
+eval_accumulation_steps:
+    value: null
+eval_delay:
+    value: 0
+eval_do_concat_batches:
+    value: true
+eval_on_start:
+    value: false
+eval_steps:
+    value: 100
+eval_strategy:
+    value: steps
+eval_use_gather_object:
+    value: false
+finetuning_task:
+    value: null
+fp16:
+    value: false
+fp16_full_eval:
+    value: false
+fsdp:
+    value: []
+fsdp_config:
+    value:
+        min_num_params: 0
+        xla: false
+        xla_fsdp_grad_ckpt: false
+        xla_fsdp_v2: false
+full_determinism:
+    value: false
+gradient_accumulation_steps:
+    value: 8
+gradient_checkpointing:
+    value: false
+gradient_checkpointing_kwargs:
+    value: null
+greater_is_better:
+    value: false
+group_by_length:
+    value: false
+hidden_act:
+    value: silu
+hidden_size:
+    value: 5120
+hub_always_push:
+    value: false
+hub_model_id:
+    value: null
+hub_private_repo:
+    value: null
+hub_revision:
+    value: null
+hub_strategy:
+    value: every_save
+hub_token:
+    value: <HUB_TOKEN>
+id2label:
+    value:
+        "0": LABEL_0
+        "1": LABEL_1
+ignore_data_skip:
+    value: false
+include_for_metrics:
+    value: []
+include_num_input_tokens_seen:
+    value: "no"
+initializer_range:
+    value: 0.02
+intermediate_size:
+    value: 13824
+is_decoder:
+    value: false
+is_encoder_decoder:
+    value: false
+label_names:
+    value: null
+label_smoothing_factor:
+    value: 0
+label2id:
+    value:
+        LABEL_0: 0
+        LABEL_1: 1
+layer_types:
+    value:
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+        - full_attention
+learning_rate:
+    value: 0.0002
+length_column_name:
+    value: length
+liger_kernel_config:
+    value: null
+load_best_model_at_end:
+    value: true
+local_rank:
+    value: -1
+log_level:
+    value: passive
+log_level_replica:
+    value: warning
+log_on_each_node:
+    value: true
+logging_dir:
+    value: null
+logging_first_step:
+    value: false
+logging_nan_inf_filter:
+    value: true
+logging_steps:
+    value: 2
+logging_strategy:
+    value: steps
+lr_scheduler_kwargs:
+    value: null
+lr_scheduler_type:
+    value: cosine
+max_grad_norm:
+    value: 1
+max_position_embeddings:
+    value: 32768
+max_steps:
+    value: -1
+max_window_layers:
+    value: 48
+metric_for_best_model:
+    value: eval_loss
+model:
+    value:
+        attn_implementation: null
+        base_local_dir: base_model
+        bnb_4bit_compute_dtype: bfloat16
+        bnb_4bit_quant_type: nf4
+        bnb_4bit_use_double_quant: false
+        device_map: auto
+        repo_id: ./Models/Qwen2.5-Coder-14B-CPT
+        revision: null
+        tokenizer_use_fast: true
+        torch_dtype: bfloat16
+        trust_remote_code: true
+        use_4bit: false
+model/num_parameters:
+    value: 14820365312
+model_type:
+    value: qwen2
+neftune_noise_alpha:
+    value: null
+num_attention_heads:
+    value: 40
+num_hidden_layers:
+    value: 48
+num_key_value_heads:
+    value: 8
+num_train_epochs:
+    value: 6
+optim:
+    value: adamw_torch
+optim_args:
+    value: null
+optim_target_modules:
+    value: null
+output_attentions:
+    value: false
+output_dir:
+    value: task2file/sft_qwen_14B_v2/checkpoints
+output_hidden_states:
+    value: false
+pad_token_id:
+    value: null
+parallelism_config:
+    value: null
+peft:
+    value:
+        bias: none
+        enabled: true
+        lora_alpha: 64
+        lora_dropout: 0.05
+        r: 32
+        target_modules: auto
+peft_config:
+    value:
+        default:
+            alora_invocation_tokens: null
+            arrow_config: null
+            auto_mapping: null
+            base_model_name_or_path: Models/Qwen2.5-Coder-14B-CPT
+            bias: none
+            corda_config: null
+            ensure_weight_tying: false
+            eva_config: null
+            exclude_modules: null
+            fan_in_fan_out: false
+            inference_mode: false
+            init_lora_weights: true
+            layer_replication: null
+            layers_pattern: null
+            layers_to_transform: null
+            lora_alpha: 64
+            lora_bias: false
+            lora_dropout: 0.05
+            megatron_config: null
+            megatron_core: megatron.core
+            modules_to_save: null
+            peft_type: LORA
+            peft_version: 0.18.0
+            qalora_group_size: 16
+            r: 32
+            revision: null
+            runtime_config:
+                ephemeral_gpu_offload: false
+            target_modules:
+                - o_proj
+                - v_proj
+                - k_proj
+                - q_proj
+            target_parameters: null
+            task_type: CAUSAL_LM
+            trainable_token_indices: null
+            use_dora: false
+            use_qalora: false
+            use_rslora: false
+per_device_eval_batch_size:
+    value: 1
+per_device_train_batch_size:
+    value: 1
+prediction_loss_only:
+    value: false
+prefix:
+    value: null
+problem_type:
+    value: null
+project:
+    value: huggingface
+push_to_hub:
+    value: false
+remove_unused_columns:
+    value: false
+report_to:
+    value:
+        - wandb
+restore_callback_states_from_checkpoint:
+    value: false
+resume_from_checkpoint:
+    value: null
+return_dict:
+    value: true
+rms_norm_eps:
+    value: 1e-06
+rope_parameters:
+    value:
+        rope_theta: 1e+06
+        rope_type: default
+run_dir:
+    value: task2file/sft_qwen_14B_v2
+run_name:
+    value: null
+save_on_each_node:
+    value: false
+save_only_model:
+    value: false
+save_steps:
+    value: 500
+save_strategy:
+    value: steps
+save_total_limit:
+    value: 20
+seed:
+    value: 42
+sep_token_id:
+    value: null
+skip_memory_metrics:
+    value: true
+sliding_window:
+    value: null
+task_specific_params:
+    value: null
+tf32:
+    value: null
+tie_word_embeddings:
+    value: false
+tokenizer_class:
+    value: null
+torch_compile:
+    value: false
+torch_compile_backend:
+    value: null
+torch_compile_mode:
+    value: null
+torch_empty_cache_steps:
+    value: null
+trackio_space_id:
+    value: trackio
+train:
+    value:
+        early_stopping:
+            enabled: true
+            metric: eval_loss
+            min_delta: 0.001
+            mode: min
+            patience: 5
+        eval_steps: 100
+        evaluation_strategy: steps
+        gradient_accumulation_steps: 8
+        gradient_checkpointing: true
+        learning_rate: "2e-4"
+        load_best_model_at_end: true
+        logging_steps: 2
+        lr_scheduler_type: cosine
+        max_grad_norm: 1
+        num_train_epochs: 6
+        optim: adamw_torch
+        per_device_eval_batch_size: 1
+        per_device_train_batch_size: 1
+        resume_from_checkpoint: auto
+        save_steps: 500
+        save_strategy: steps
+        save_total_limit: 20
+        warmup_ratio: 0.08
+        weight_decay: 0
+transformers_version:
+    value: 5.0.0.dev0
+use_cache:
+    value: false
+use_cpu:
+    value: false
+use_liger_kernel:
+    value: false
+use_sliding_window:
+    value: false
+vocab_size:
+    value: 152064
+warmup_ratio:
+    value: 0.08
+warmup_steps:
+    value: 0.08
+weight_decay:
+    value: 0
diff --git a/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/files/output.log b/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/files/output.log
new file mode 100644
index 0000000000000000000000000000000000000000..b2b64125a2824854fdee2126558654b4c9df8156
--- /dev/null
+++ b/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/files/output.log
@@ -0,0 +1,2620 @@
+Wandb initialized: project='sft-training', name='auto-generated'
+`torch_dtype` is deprecated! Use `dtype` instead!
+Loading weights: 100%|█| 579/579 [00:09<00:00, 57.95it/s, Materializing param=model.n
+Formatting train instructions (num_proc=4): 100%|█| 18960/18960 [00:10<00:00, 1774.11
+Formatting eval instructions (num_proc=4): 100%|█| 2107/2107 [00:10<00:00, 204.45 exa
+Tokenizing and masking train (num_proc=4): 100%|█| 18960/18960 [00:24<00:00, 780.24 e
+Tokenizing and masking eval (num_proc=4): 100%|█| 2107/2107 [00:11<00:00, 177.77 exam
+warmup_ratio is deprecated and will be removed in v5.2. Use `warmup_steps` instead.
+Early stopping enabled: patience=5, min_delta=0.001
+Starting instruction fine-tuning...
+                                                                                     
+{'loss': '1.694', 'grad_norm': '0.5387', 'learning_rate': '1.757e-07', 'epoch': '0.0008439'}
+{'loss': '1.713', 'grad_norm': '0.5477', 'learning_rate': '5.272e-07', 'epoch': '0.001688'}
+{'loss': '1.641', 'grad_norm': '0.5391', 'learning_rate': '8.787e-07', 'epoch': '0.002532'}
+{'loss': '1.562', 'grad_norm': '0.5024', 'learning_rate': '1.23e-06', 'epoch': '0.003376'}
+{'loss': '1.572', 'grad_norm': '0.4899', 'learning_rate': '1.582e-06', 'epoch': '0.004219'}
+{'loss': '1.624', 'grad_norm': '0.524', 'learning_rate': '1.933e-06', 'epoch': '0.005063'}
+{'loss': '1.68', 'grad_norm': '0.5173', 'learning_rate': '2.285e-06', 'epoch': '0.005907'}
+{'loss': '1.645', 'grad_norm': '0.5539', 'learning_rate': '2.636e-06', 'epoch': '0.006751'}
+{'loss': '1.667', 'grad_norm': '0.5255', 'learning_rate': '2.988e-06', 'epoch': '0.007595'}
+{'loss': '1.532', 'grad_norm': '0.5075', 'learning_rate': '3.339e-06', 'epoch': '0.008439'}
+{'loss': '1.599', 'grad_norm': '0.416', 'learning_rate': '3.691e-06', 'epoch': '0.009283'}
+{'loss': '1.67', 'grad_norm': '0.5716', 'learning_rate': '4.042e-06', 'epoch': '0.01013'}
+{'loss': '1.662', 'grad_norm': '0.5149', 'learning_rate': '4.394e-06', 'epoch': '0.01097'}
+{'loss': '1.669', 'grad_norm': '0.5757', 'learning_rate': '4.745e-06', 'epoch': '0.01181'}
+{'loss': '1.552', 'grad_norm': '0.5346', 'learning_rate': '5.097e-06', 'epoch': '0.01266'}
+{'loss': '1.593', 'grad_norm': '0.4462', 'learning_rate': '5.448e-06', 'epoch': '0.0135'}
+{'loss': '1.607', 'grad_norm': '0.512', 'learning_rate': '5.8e-06', 'epoch': '0.01435'}
+{'loss': '1.584', 'grad_norm': '0.5329', 'learning_rate': '6.151e-06', 'epoch': '0.01519'}
+{'loss': '1.608', 'grad_norm': '0.5066', 'learning_rate': '6.503e-06', 'epoch': '0.01603'}
+{'loss': '1.594', 'grad_norm': '0.4479', 'learning_rate': '6.854e-06', 'epoch': '0.01688'}
+{'loss': '1.644', 'grad_norm': '0.4284', 'learning_rate': '7.206e-06', 'epoch': '0.01772'}
+{'loss': '1.607', 'grad_norm': '0.4766', 'learning_rate': '7.557e-06', 'epoch': '0.01857'}
+{'loss': '1.662', 'grad_norm': '0.4533', 'learning_rate': '7.909e-06', 'epoch': '0.01941'}
+{'loss': '1.473', 'grad_norm': '0.3701', 'learning_rate': '8.26e-06', 'epoch': '0.02025'}
+{'loss': '1.483', 'grad_norm': '0.3847', 'learning_rate': '8.612e-06', 'epoch': '0.0211'}
+{'loss': '1.388', 'grad_norm': '0.3602', 'learning_rate': '8.963e-06', 'epoch': '0.02194'}
+{'loss': '1.491', 'grad_norm': '0.4032', 'learning_rate': '9.315e-06', 'epoch': '0.02278'}
+{'loss': '1.491', 'grad_norm': '0.3224', 'learning_rate': '9.666e-06', 'epoch': '0.02363'}
+{'loss': '1.527', 'grad_norm': '0.3873', 'learning_rate': '1.002e-05', 'epoch': '0.02447'}
+{'loss': '1.448', 'grad_norm': '0.4102', 'learning_rate': '1.037e-05', 'epoch': '0.02532'}
+{'loss': '1.422', 'grad_norm': '0.3632', 'learning_rate': '1.072e-05', 'epoch': '0.02616'}
+{'loss': '1.387', 'grad_norm': '0.3312', 'learning_rate': '1.107e-05', 'epoch': '0.027'}
+{'loss': '1.352', 'grad_norm': '0.3302', 'learning_rate': '1.142e-05', 'epoch': '0.02785'}
+{'loss': '1.452', 'grad_norm': '0.3434', 'learning_rate': '1.178e-05', 'epoch': '0.02869'}
+{'loss': '1.351', 'grad_norm': '0.3243', 'learning_rate': '1.213e-05', 'epoch': '0.02954'}
+{'loss': '1.494', 'grad_norm': '0.345', 'learning_rate': '1.248e-05', 'epoch': '0.03038'}
+{'loss': '1.342', 'grad_norm': '0.3891', 'learning_rate': '1.283e-05', 'epoch': '0.03122'}
+{'loss': '1.428', 'grad_norm': '0.3019', 'learning_rate': '1.318e-05', 'epoch': '0.03207'}
+{'loss': '1.426', 'grad_norm': '0.378', 'learning_rate': '1.353e-05', 'epoch': '0.03291'}
+{'loss': '1.326', 'grad_norm': '0.3419', 'learning_rate': '1.388e-05', 'epoch': '0.03376'}
+{'loss': '1.271', 'grad_norm': '0.3524', 'learning_rate': '1.424e-05', 'epoch': '0.0346'}
+{'loss': '1.253', 'grad_norm': '0.3809', 'learning_rate': '1.459e-05', 'epoch': '0.03544'}
+{'loss': '1.306', 'grad_norm': '0.3684', 'learning_rate': '1.494e-05', 'epoch': '0.03629'}
+{'loss': '1.294', 'grad_norm': '0.3444', 'learning_rate': '1.529e-05', 'epoch': '0.03713'}
+{'loss': '1.338', 'grad_norm': '0.3799', 'learning_rate': '1.564e-05', 'epoch': '0.03797'}
+{'loss': '1.393', 'grad_norm': '0.405', 'learning_rate': '1.599e-05', 'epoch': '0.03882'}
+{'loss': '1.357', 'grad_norm': '0.442', 'learning_rate': '1.634e-05', 'epoch': '0.03966'}
+{'loss': '1.303', 'grad_norm': '0.362', 'learning_rate': '1.67e-05', 'epoch': '0.04051'}
+{'loss': '1.35', 'grad_norm': '0.4851', 'learning_rate': '1.705e-05', 'epoch': '0.04135'}
+{'loss': '1.185', 'grad_norm': '0.4187', 'learning_rate': '1.74e-05', 'epoch': '0.04219'}
+                                                                                     
+{'eval_loss': '1.298', 'eval_runtime': '682.2', 'eval_samples_per_second': '3.089', 'eval_steps_per_second': '3.089', 'epoch': '0.04219'}
+{'loss': '1.222', 'grad_norm': '0.4464', 'learning_rate': '1.775e-05', 'epoch': '0.04304'}
+{'loss': '1.205', 'grad_norm': '0.4706', 'learning_rate': '1.81e-05', 'epoch': '0.04388'}
+{'loss': '1.221', 'grad_norm': '0.4639', 'learning_rate': '1.845e-05', 'epoch': '0.04473'}
+{'loss': '1.239', 'grad_norm': '0.4727', 'learning_rate': '1.88e-05', 'epoch': '0.04557'}
+{'loss': '1.285', 'grad_norm': '0.4213', 'learning_rate': '1.916e-05', 'epoch': '0.04641'}
+{'loss': '1.275', 'grad_norm': '0.4505', 'learning_rate': '1.951e-05', 'epoch': '0.04726'}
+{'loss': '1.243', 'grad_norm': '0.3961', 'learning_rate': '1.986e-05', 'epoch': '0.0481'}
+{'loss': '1.294', 'grad_norm': '0.5106', 'learning_rate': '2.021e-05', 'epoch': '0.04895'}
+{'loss': '1.263', 'grad_norm': '0.4235', 'learning_rate': '2.056e-05', 'epoch': '0.04979'}
+{'loss': '1.265', 'grad_norm': '0.4404', 'learning_rate': '2.091e-05', 'epoch': '0.05063'}
+{'loss': '1.235', 'grad_norm': '0.5261', 'learning_rate': '2.127e-05', 'epoch': '0.05148'}
+{'loss': '1.038', 'grad_norm': '0.4498', 'learning_rate': '2.162e-05', 'epoch': '0.05232'}
+{'loss': '1.152', 'grad_norm': '0.4773', 'learning_rate': '2.197e-05', 'epoch': '0.05316'}
+{'loss': '1.226', 'grad_norm': '0.5474', 'learning_rate': '2.232e-05', 'epoch': '0.05401'}
+{'loss': '1.162', 'grad_norm': '0.4474', 'learning_rate': '2.267e-05', 'epoch': '0.05485'}
+{'loss': '1.133', 'grad_norm': '0.598', 'learning_rate': '2.302e-05', 'epoch': '0.0557'}
+{'loss': '1.155', 'grad_norm': '0.5988', 'learning_rate': '2.337e-05', 'epoch': '0.05654'}
+{'loss': '1.195', 'grad_norm': '0.4536', 'learning_rate': '2.373e-05', 'epoch': '0.05738'}
+{'loss': '1.107', 'grad_norm': '0.5374', 'learning_rate': '2.408e-05', 'epoch': '0.05823'}
+{'loss': '1.188', 'grad_norm': '0.4873', 'learning_rate': '2.443e-05', 'epoch': '0.05907'}
+{'loss': '1.162', 'grad_norm': '0.4692', 'learning_rate': '2.478e-05', 'epoch': '0.05992'}
+{'loss': '1.098', 'grad_norm': '0.5326', 'learning_rate': '2.513e-05', 'epoch': '0.06076'}
+{'loss': '1.203', 'grad_norm': '0.5854', 'learning_rate': '2.548e-05', 'epoch': '0.0616'}
+{'loss': '1.082', 'grad_norm': '0.5062', 'learning_rate': '2.583e-05', 'epoch': '0.06245'}
+{'loss': '1.106', 'grad_norm': '0.4943', 'learning_rate': '2.619e-05', 'epoch': '0.06329'}
+{'loss': '1.199', 'grad_norm': '0.5847', 'learning_rate': '2.654e-05', 'epoch': '0.06414'}
+{'loss': '1.176', 'grad_norm': '0.5518', 'learning_rate': '2.689e-05', 'epoch': '0.06498'}
+{'loss': '1.055', 'grad_norm': '0.5667', 'learning_rate': '2.724e-05', 'epoch': '0.06582'}
+{'loss': '1.184', 'grad_norm': '0.676', 'learning_rate': '2.759e-05', 'epoch': '0.06667'}
+{'loss': '1.195', 'grad_norm': '0.5261', 'learning_rate': '2.794e-05', 'epoch': '0.06751'}
+{'loss': '1.202', 'grad_norm': '0.6155', 'learning_rate': '2.83e-05', 'epoch': '0.06835'}
+{'loss': '1.144', 'grad_norm': '0.6132', 'learning_rate': '2.865e-05', 'epoch': '0.0692'}
+{'loss': '1.133', 'grad_norm': '0.5749', 'learning_rate': '2.9e-05', 'epoch': '0.07004'}
+{'loss': '1.089', 'grad_norm': '0.5243', 'learning_rate': '2.935e-05', 'epoch': '0.07089'}
+{'loss': '1.163', 'grad_norm': '0.719', 'learning_rate': '2.97e-05', 'epoch': '0.07173'}
+{'loss': '1.088', 'grad_norm': '0.5487', 'learning_rate': '3.005e-05', 'epoch': '0.07257'}
+{'loss': '1.043', 'grad_norm': '0.5021', 'learning_rate': '3.04e-05', 'epoch': '0.07342'}
+{'loss': '1.053', 'grad_norm': '0.4733', 'learning_rate': '3.076e-05', 'epoch': '0.07426'}
+{'loss': '1.163', 'grad_norm': '0.6636', 'learning_rate': '3.111e-05', 'epoch': '0.07511'}
+{'loss': '1.085', 'grad_norm': '0.5625', 'learning_rate': '3.146e-05', 'epoch': '0.07595'}
+{'loss': '1.223', 'grad_norm': '0.603', 'learning_rate': '3.181e-05', 'epoch': '0.07679'}
+{'loss': '1.096', 'grad_norm': '0.931', 'learning_rate': '3.216e-05', 'epoch': '0.07764'}
+{'loss': '1.064', 'grad_norm': '0.5327', 'learning_rate': '3.251e-05', 'epoch': '0.07848'}
+{'loss': '1.07', 'grad_norm': '0.5485', 'learning_rate': '3.286e-05', 'epoch': '0.07932'}
+{'loss': '1.159', 'grad_norm': '0.6053', 'learning_rate': '3.322e-05', 'epoch': '0.08017'}
+{'loss': '1.162', 'grad_norm': '0.5704', 'learning_rate': '3.357e-05', 'epoch': '0.08101'}
+{'loss': '1.135', 'grad_norm': '0.5929', 'learning_rate': '3.392e-05', 'epoch': '0.08186'}
+{'loss': '1.093', 'grad_norm': '0.5624', 'learning_rate': '3.427e-05', 'epoch': '0.0827'}
+{'loss': '1.181', 'grad_norm': '0.6717', 'learning_rate': '3.462e-05', 'epoch': '0.08354'}
+{'loss': '1.153', 'grad_norm': '0.512', 'learning_rate': '3.497e-05', 'epoch': '0.08439'}
+{'eval_loss': '1.142', 'eval_runtime': '668.2', 'eval_samples_per_second': '3.153', 'eval_steps_per_second': '3.153', 'epoch': '0.08439'}
+{'loss': '1.024', 'grad_norm': '0.5144', 'learning_rate': '3.533e-05', 'epoch': '0.08523'}
+{'loss': '1.119', 'grad_norm': '0.6325', 'learning_rate': '3.568e-05', 'epoch': '0.08608'}
+{'loss': '1.038', 'grad_norm': '0.5502', 'learning_rate': '3.603e-05', 'epoch': '0.08692'}
+{'loss': '0.9838', 'grad_norm': '0.6134', 'learning_rate': '3.638e-05', 'epoch': '0.08776'}
+{'loss': '1.091', 'grad_norm': '0.58', 'learning_rate': '3.673e-05', 'epoch': '0.08861'}
+{'loss': '1.12', 'grad_norm': '0.604', 'learning_rate': '3.708e-05', 'epoch': '0.08945'}
+{'loss': '1.097', 'grad_norm': '0.5983', 'learning_rate': '3.743e-05', 'epoch': '0.0903'}
+{'loss': '1.123', 'grad_norm': '0.5641', 'learning_rate': '3.779e-05', 'epoch': '0.09114'}
+{'loss': '1.126', 'grad_norm': '0.6557', 'learning_rate': '3.814e-05', 'epoch': '0.09198'}
+{'loss': '1.078', 'grad_norm': '0.6112', 'learning_rate': '3.849e-05', 'epoch': '0.09283'}
+{'loss': '1.106', 'grad_norm': '0.6821', 'learning_rate': '3.884e-05', 'epoch': '0.09367'}
+{'loss': '1.15', 'grad_norm': '0.6932', 'learning_rate': '3.919e-05', 'epoch': '0.09451'}
+{'loss': '1.059', 'grad_norm': '0.5288', 'learning_rate': '3.954e-05', 'epoch': '0.09536'}
+{'loss': '1.111', 'grad_norm': '0.6883', 'learning_rate': '3.989e-05', 'epoch': '0.0962'}
+{'loss': '1.021', 'grad_norm': '0.5834', 'learning_rate': '4.025e-05', 'epoch': '0.09705'}
+{'loss': '1.064', 'grad_norm': '0.7157', 'learning_rate': '4.06e-05', 'epoch': '0.09789'}
+{'loss': '1.036', 'grad_norm': '0.6531', 'learning_rate': '4.095e-05', 'epoch': '0.09873'}
+{'loss': '1.051', 'grad_norm': '0.633', 'learning_rate': '4.13e-05', 'epoch': '0.09958'}
+{'loss': '1.262', 'grad_norm': '0.6597', 'learning_rate': '4.165e-05', 'epoch': '0.1004'}
+{'loss': '1.122', 'grad_norm': '0.6195', 'learning_rate': '4.2e-05', 'epoch': '0.1013'}
+{'loss': '1.125', 'grad_norm': '0.6764', 'learning_rate': '4.236e-05', 'epoch': '0.1021'}
+{'loss': '1.028', 'grad_norm': '0.5524', 'learning_rate': '4.271e-05', 'epoch': '0.103'}
+{'loss': '1.043', 'grad_norm': '0.562', 'learning_rate': '4.306e-05', 'epoch': '0.1038'}
+{'loss': '1.144', 'grad_norm': '0.686', 'learning_rate': '4.341e-05', 'epoch': '0.1046'}
+{'loss': '1.122', 'grad_norm': '0.6033', 'learning_rate': '4.376e-05', 'epoch': '0.1055'}
+{'loss': '1.161', 'grad_norm': '0.6292', 'learning_rate': '4.411e-05', 'epoch': '0.1063'}
+{'loss': '1.064', 'grad_norm': '0.7983', 'learning_rate': '4.446e-05', 'epoch': '0.1072'}
+{'loss': '1.072', 'grad_norm': '0.7719', 'learning_rate': '4.482e-05', 'epoch': '0.108'}
+{'loss': '1.078', 'grad_norm': '0.6101', 'learning_rate': '4.517e-05', 'epoch': '0.1089'}
+{'loss': '1.213', 'grad_norm': '0.7301', 'learning_rate': '4.552e-05', 'epoch': '0.1097'}
+{'loss': '1.222', 'grad_norm': '0.8349', 'learning_rate': '4.587e-05', 'epoch': '0.1105'}
+{'loss': '1.019', 'grad_norm': '0.5446', 'learning_rate': '4.622e-05', 'epoch': '0.1114'}
+{'loss': '1.097', 'grad_norm': '0.623', 'learning_rate': '4.657e-05', 'epoch': '0.1122'}
+{'loss': '1.09', 'grad_norm': '0.6231', 'learning_rate': '4.692e-05', 'epoch': '0.1131'}
+{'loss': '1.07', 'grad_norm': '0.7179', 'learning_rate': '4.728e-05', 'epoch': '0.1139'}
+{'loss': '1.051', 'grad_norm': '0.6896', 'learning_rate': '4.763e-05', 'epoch': '0.1148'}
+{'loss': '1.137', 'grad_norm': '0.6047', 'learning_rate': '4.798e-05', 'epoch': '0.1156'}
+{'loss': '1.073', 'grad_norm': '0.6525', 'learning_rate': '4.833e-05', 'epoch': '0.1165'}
+{'loss': '1.124', 'grad_norm': '0.6331', 'learning_rate': '4.868e-05', 'epoch': '0.1173'}
+{'loss': '1.064', 'grad_norm': '0.5919', 'learning_rate': '4.903e-05', 'epoch': '0.1181'}
+{'loss': '1.043', 'grad_norm': '0.6068', 'learning_rate': '4.938e-05', 'epoch': '0.119'}
+{'loss': '1.14', 'grad_norm': '0.6584', 'learning_rate': '4.974e-05', 'epoch': '0.1198'}
+{'loss': '1.059', 'grad_norm': '0.62', 'learning_rate': '5.009e-05', 'epoch': '0.1207'}
+{'loss': '1.132', 'grad_norm': '0.6799', 'learning_rate': '5.044e-05', 'epoch': '0.1215'}
+{'loss': '1.093', 'grad_norm': '0.7509', 'learning_rate': '5.079e-05', 'epoch': '0.1224'}
+{'loss': '1.163', 'grad_norm': '0.6901', 'learning_rate': '5.114e-05', 'epoch': '0.1232'}
+{'loss': '1.091', 'grad_norm': '0.6423', 'learning_rate': '5.149e-05', 'epoch': '0.1241'}
+{'loss': '1.069', 'grad_norm': '0.6839', 'learning_rate': '5.185e-05', 'epoch': '0.1249'}
+{'loss': '0.9823', 'grad_norm': '0.635', 'learning_rate': '5.22e-05', 'epoch': '0.1257'}
+{'loss': '1.113', 'grad_norm': '0.7137', 'learning_rate': '5.255e-05', 'epoch': '0.1266'}
+{'eval_loss': '1.095', 'eval_runtime': '677.1', 'eval_samples_per_second': '3.112', 'eval_steps_per_second': '3.112', 'epoch': '0.1266'}
+{'loss': '0.9736', 'grad_norm': '0.734', 'learning_rate': '5.29e-05', 'epoch': '0.1274'}
+{'loss': '0.982', 'grad_norm': '0.5941', 'learning_rate': '5.325e-05', 'epoch': '0.1283'}
+{'loss': '1.032', 'grad_norm': '0.7154', 'learning_rate': '5.36e-05', 'epoch': '0.1291'}
+{'loss': '0.9492', 'grad_norm': '0.5167', 'learning_rate': '5.395e-05', 'epoch': '0.13'}
+{'loss': '1.016', 'grad_norm': '0.6056', 'learning_rate': '5.431e-05', 'epoch': '0.1308'}
+{'loss': '1.179', 'grad_norm': '0.7662', 'learning_rate': '5.466e-05', 'epoch': '0.1316'}
+{'loss': '1.001', 'grad_norm': '0.6066', 'learning_rate': '5.501e-05', 'epoch': '0.1325'}
+{'loss': '1.021', 'grad_norm': '0.6042', 'learning_rate': '5.536e-05', 'epoch': '0.1333'}
+{'loss': '0.9968', 'grad_norm': '0.6187', 'learning_rate': '5.571e-05', 'epoch': '0.1342'}
+{'loss': '0.954', 'grad_norm': '0.71', 'learning_rate': '5.606e-05', 'epoch': '0.135'}
+{'loss': '1.063', 'grad_norm': '0.698', 'learning_rate': '5.641e-05', 'epoch': '0.1359'}
+{'loss': '1.017', 'grad_norm': '0.6237', 'learning_rate': '5.677e-05', 'epoch': '0.1367'}
+{'loss': '1.072', 'grad_norm': '0.6526', 'learning_rate': '5.712e-05', 'epoch': '0.1376'}
+{'loss': '1.011', 'grad_norm': '0.6869', 'learning_rate': '5.747e-05', 'epoch': '0.1384'}
+{'loss': '1.031', 'grad_norm': '0.5532', 'learning_rate': '5.782e-05', 'epoch': '0.1392'}
+{'loss': '1.021', 'grad_norm': '0.6761', 'learning_rate': '5.817e-05', 'epoch': '0.1401'}
+{'loss': '0.9749', 'grad_norm': '0.5907', 'learning_rate': '5.852e-05', 'epoch': '0.1409'}
+{'loss': '1.059', 'grad_norm': '0.7045', 'learning_rate': '5.888e-05', 'epoch': '0.1418'}
+{'loss': '1.134', 'grad_norm': '0.6791', 'learning_rate': '5.923e-05', 'epoch': '0.1426'}
+{'loss': '0.9541', 'grad_norm': '0.7676', 'learning_rate': '5.958e-05', 'epoch': '0.1435'}
+{'loss': '1.059', 'grad_norm': '0.6314', 'learning_rate': '5.993e-05', 'epoch': '0.1443'}
+{'loss': '1.076', 'grad_norm': '0.7123', 'learning_rate': '6.028e-05', 'epoch': '0.1451'}
+{'loss': '1.037', 'grad_norm': '0.5859', 'learning_rate': '6.063e-05', 'epoch': '0.146'}
+{'loss': '1.024', 'grad_norm': '0.5643', 'learning_rate': '6.098e-05', 'epoch': '0.1468'}
+{'loss': '1.05', 'grad_norm': '0.6268', 'learning_rate': '6.134e-05', 'epoch': '0.1477'}
+{'loss': '0.9812', 'grad_norm': '0.5107', 'learning_rate': '6.169e-05', 'epoch': '0.1485'}
+{'loss': '1.072', 'grad_norm': '0.6948', 'learning_rate': '6.204e-05', 'epoch': '0.1494'}
+{'loss': '1.041', 'grad_norm': '0.6997', 'learning_rate': '6.239e-05', 'epoch': '0.1502'}
+{'loss': '1.049', 'grad_norm': '0.6438', 'learning_rate': '6.274e-05', 'epoch': '0.1511'}
+{'loss': '0.9268', 'grad_norm': '0.6953', 'learning_rate': '6.309e-05', 'epoch': '0.1519'}
+{'loss': '1.043', 'grad_norm': '0.6713', 'learning_rate': '6.344e-05', 'epoch': '0.1527'}
+{'loss': '1.049', 'grad_norm': '0.675', 'learning_rate': '6.38e-05', 'epoch': '0.1536'}
+{'loss': '1.016', 'grad_norm': '0.6053', 'learning_rate': '6.415e-05', 'epoch': '0.1544'}
+{'loss': '1.002', 'grad_norm': '0.8064', 'learning_rate': '6.45e-05', 'epoch': '0.1553'}
+{'loss': '1.056', 'grad_norm': '0.8027', 'learning_rate': '6.485e-05', 'epoch': '0.1561'}
+{'loss': '1.015', 'grad_norm': '0.658', 'learning_rate': '6.52e-05', 'epoch': '0.157'}
+{'loss': '0.977', 'grad_norm': '0.6561', 'learning_rate': '6.555e-05', 'epoch': '0.1578'}
+{'loss': '0.9099', 'grad_norm': '0.6444', 'learning_rate': '6.591e-05', 'epoch': '0.1586'}
+{'loss': '1.08', 'grad_norm': '0.5879', 'learning_rate': '6.626e-05', 'epoch': '0.1595'}
+{'loss': '1.034', 'grad_norm': '0.6994', 'learning_rate': '6.661e-05', 'epoch': '0.1603'}
+{'loss': '1.018', 'grad_norm': '0.6129', 'learning_rate': '6.696e-05', 'epoch': '0.1612'}
+{'loss': '1.141', 'grad_norm': '1.024', 'learning_rate': '6.731e-05', 'epoch': '0.162'}
+{'loss': '1.031', 'grad_norm': '0.6092', 'learning_rate': '6.766e-05', 'epoch': '0.1629'}
+{'loss': '0.9564', 'grad_norm': '0.5916', 'learning_rate': '6.801e-05', 'epoch': '0.1637'}
+{'loss': '1.002', 'grad_norm': '0.7717', 'learning_rate': '6.837e-05', 'epoch': '0.1646'}
+{'loss': '1.07', 'grad_norm': '0.6547', 'learning_rate': '6.872e-05', 'epoch': '0.1654'}
+{'loss': '1.055', 'grad_norm': '0.6612', 'learning_rate': '6.907e-05', 'epoch': '0.1662'}
+{'loss': '1.056', 'grad_norm': '0.6956', 'learning_rate': '6.942e-05', 'epoch': '0.1671'}
+{'loss': '0.9851', 'grad_norm': '0.6666', 'learning_rate': '6.977e-05', 'epoch': '0.1679'}
+{'loss': '1.008', 'grad_norm': '0.6132', 'learning_rate': '7.012e-05', 'epoch': '0.1688'}
+{'eval_loss': '1.063', 'eval_runtime': '691', 'eval_samples_per_second': '3.049', 'eval_steps_per_second': '3.049', 'epoch': '0.1688'}
+{'loss': '1.101', 'grad_norm': '0.6286', 'learning_rate': '7.047e-05', 'epoch': '0.1696'}
+{'loss': '1.038', 'grad_norm': '0.6639', 'learning_rate': '7.083e-05', 'epoch': '0.1705'}
+{'loss': '1.032', 'grad_norm': '0.7504', 'learning_rate': '7.118e-05', 'epoch': '0.1713'}
+{'loss': '1.049', 'grad_norm': '0.8085', 'learning_rate': '7.153e-05', 'epoch': '0.1722'}
+{'loss': '1.101', 'grad_norm': '0.7448', 'learning_rate': '7.188e-05', 'epoch': '0.173'}
+{'loss': '0.9927', 'grad_norm': '0.5842', 'learning_rate': '7.223e-05', 'epoch': '0.1738'}
+{'loss': '1.059', 'grad_norm': '0.7126', 'learning_rate': '7.258e-05', 'epoch': '0.1747'}
+{'loss': '1.038', 'grad_norm': '0.5751', 'learning_rate': '7.293e-05', 'epoch': '0.1755'}
+{'loss': '0.9843', 'grad_norm': '0.5654', 'learning_rate': '7.329e-05', 'epoch': '0.1764'}
+{'loss': '1.073', 'grad_norm': '0.5826', 'learning_rate': '7.364e-05', 'epoch': '0.1772'}
+{'loss': '1.006', 'grad_norm': '0.6267', 'learning_rate': '7.399e-05', 'epoch': '0.1781'}
+{'loss': '1.046', 'grad_norm': '0.6621', 'learning_rate': '7.434e-05', 'epoch': '0.1789'}
+{'loss': '0.9789', 'grad_norm': '0.6635', 'learning_rate': '7.469e-05', 'epoch': '0.1797'}
+{'loss': '1.053', 'grad_norm': '0.6342', 'learning_rate': '7.504e-05', 'epoch': '0.1806'}
+{'loss': '0.9709', 'grad_norm': '0.6762', 'learning_rate': '7.54e-05', 'epoch': '0.1814'}
+{'loss': '1.051', 'grad_norm': '0.7073', 'learning_rate': '7.575e-05', 'epoch': '0.1823'}
+{'loss': '1.013', 'grad_norm': '0.6918', 'learning_rate': '7.61e-05', 'epoch': '0.1831'}
+{'loss': '0.9894', 'grad_norm': '0.5575', 'learning_rate': '7.645e-05', 'epoch': '0.184'}
+{'loss': '0.9846', 'grad_norm': '0.6408', 'learning_rate': '7.68e-05', 'epoch': '0.1848'}
+{'loss': '1.034', 'grad_norm': '0.6699', 'learning_rate': '7.715e-05', 'epoch': '0.1857'}
+{'loss': '1.018', 'grad_norm': '0.6165', 'learning_rate': '7.75e-05', 'epoch': '0.1865'}
+{'loss': '1.052', 'grad_norm': '0.6881', 'learning_rate': '7.786e-05', 'epoch': '0.1873'}
+{'loss': '1.044', 'grad_norm': '0.6501', 'learning_rate': '7.821e-05', 'epoch': '0.1882'}
+{'loss': '0.9828', 'grad_norm': '0.6076', 'learning_rate': '7.856e-05', 'epoch': '0.189'}
+{'loss': '1.057', 'grad_norm': '0.6418', 'learning_rate': '7.891e-05', 'epoch': '0.1899'}
+{'loss': '1.022', 'grad_norm': '0.7055', 'learning_rate': '7.926e-05', 'epoch': '0.1907'}
+{'loss': '0.9838', 'grad_norm': '0.5668', 'learning_rate': '7.961e-05', 'epoch': '0.1916'}
+{'loss': '1', 'grad_norm': '0.6419', 'learning_rate': '7.996e-05', 'epoch': '0.1924'}
+{'loss': '1.061', 'grad_norm': '0.5949', 'learning_rate': '8.032e-05', 'epoch': '0.1932'}
+{'loss': '1.054', 'grad_norm': '0.7032', 'learning_rate': '8.067e-05', 'epoch': '0.1941'}
+{'loss': '0.9819', 'grad_norm': '0.5776', 'learning_rate': '8.102e-05', 'epoch': '0.1949'}
+{'loss': '1.054', 'grad_norm': '0.9301', 'learning_rate': '8.137e-05', 'epoch': '0.1958'}
+{'loss': '0.9966', 'grad_norm': '0.6193', 'learning_rate': '8.172e-05', 'epoch': '0.1966'}
+{'loss': '1.047', 'grad_norm': '0.6286', 'learning_rate': '8.207e-05', 'epoch': '0.1975'}
+{'loss': '0.9549', 'grad_norm': '0.7419', 'learning_rate': '8.243e-05', 'epoch': '0.1983'}
+{'loss': '1.019', 'grad_norm': '0.6123', 'learning_rate': '8.278e-05', 'epoch': '0.1992'}
+{'loss': '1.099', 'grad_norm': '0.6375', 'learning_rate': '8.313e-05', 'epoch': '0.2'}
+{'loss': '1.037', 'grad_norm': '0.646', 'learning_rate': '8.348e-05', 'epoch': '0.2008'}
+{'loss': '1.058', 'grad_norm': '0.703', 'learning_rate': '8.383e-05', 'epoch': '0.2017'}
+{'loss': '1.023', 'grad_norm': '0.6345', 'learning_rate': '8.418e-05', 'epoch': '0.2025'}
+{'loss': '0.973', 'grad_norm': '0.5765', 'learning_rate': '8.453e-05', 'epoch': '0.2034'}
+{'loss': '1.029', 'grad_norm': '0.5884', 'learning_rate': '8.489e-05', 'epoch': '0.2042'}
+{'loss': '1.008', 'grad_norm': '0.6164', 'learning_rate': '8.524e-05', 'epoch': '0.2051'}
+{'loss': '0.9937', 'grad_norm': '0.7671', 'learning_rate': '8.559e-05', 'epoch': '0.2059'}
+{'loss': '1.052', 'grad_norm': '0.6197', 'learning_rate': '8.594e-05', 'epoch': '0.2068'}
+{'loss': '0.9475', 'grad_norm': '0.6913', 'learning_rate': '8.629e-05', 'epoch': '0.2076'}
+{'loss': '1.067', 'grad_norm': '0.6942', 'learning_rate': '8.664e-05', 'epoch': '0.2084'}
+{'loss': '1.002', 'grad_norm': '0.589', 'learning_rate': '8.699e-05', 'epoch': '0.2093'}
+{'loss': '1.017', 'grad_norm': '0.6479', 'learning_rate': '8.735e-05', 'epoch': '0.2101'}
+{'loss': '0.8798', 'grad_norm': '0.6018', 'learning_rate': '8.77e-05', 'epoch': '0.211'}
+{'eval_loss': '1.042', 'eval_runtime': '692.4', 'eval_samples_per_second': '3.043', 'eval_steps_per_second': '3.043', 'epoch': '0.211'}
+{'loss': '1.092', 'grad_norm': '0.579', 'learning_rate': '8.805e-05', 'epoch': '0.2118'}
+{'loss': '1.041', 'grad_norm': '0.6598', 'learning_rate': '8.84e-05', 'epoch': '0.2127'}
+{'loss': '1.007', 'grad_norm': '0.666', 'learning_rate': '8.875e-05', 'epoch': '0.2135'}
+{'loss': '1.032', 'grad_norm': '0.6283', 'learning_rate': '8.91e-05', 'epoch': '0.2143'}
+{'loss': '1.005', 'grad_norm': '0.6257', 'learning_rate': '8.946e-05', 'epoch': '0.2152'}
+{'loss': '0.9169', 'grad_norm': '0.5531', 'learning_rate': '8.981e-05', 'epoch': '0.216'}
+{'loss': '0.9082', 'grad_norm': '0.537', 'learning_rate': '9.016e-05', 'epoch': '0.2169'}
+{'loss': '0.9165', 'grad_norm': '0.6618', 'learning_rate': '9.051e-05', 'epoch': '0.2177'}
+{'loss': '0.9135', 'grad_norm': '0.6007', 'learning_rate': '9.086e-05', 'epoch': '0.2186'}
+{'loss': '1.007', 'grad_norm': '0.592', 'learning_rate': '9.121e-05', 'epoch': '0.2194'}
+{'loss': '1.001', 'grad_norm': '0.6881', 'learning_rate': '9.156e-05', 'epoch': '0.2203'}
+{'loss': '1.091', 'grad_norm': '0.6721', 'learning_rate': '9.192e-05', 'epoch': '0.2211'}
+{'loss': '1.012', 'grad_norm': '0.5987', 'learning_rate': '9.227e-05', 'epoch': '0.2219'}
+{'loss': '1.015', 'grad_norm': '0.6191', 'learning_rate': '9.262e-05', 'epoch': '0.2228'}
+{'loss': '0.965', 'grad_norm': '0.6822', 'learning_rate': '9.297e-05', 'epoch': '0.2236'}
+{'loss': '0.8673', 'grad_norm': '1.76', 'learning_rate': '9.332e-05', 'epoch': '0.2245'}
+{'loss': '0.9942', 'grad_norm': '0.667', 'learning_rate': '9.367e-05', 'epoch': '0.2253'}
+{'loss': '1.036', 'grad_norm': '0.7346', 'learning_rate': '9.402e-05', 'epoch': '0.2262'}
+{'loss': '0.9315', 'grad_norm': '0.5946', 'learning_rate': '9.438e-05', 'epoch': '0.227'}
+{'loss': '1.069', 'grad_norm': '0.5801', 'learning_rate': '9.473e-05', 'epoch': '0.2278'}
+{'loss': '1.035', 'grad_norm': '0.6712', 'learning_rate': '9.508e-05', 'epoch': '0.2287'}
+{'loss': '0.9795', 'grad_norm': '0.5641', 'learning_rate': '9.543e-05', 'epoch': '0.2295'}
+{'loss': '0.9641', 'grad_norm': '0.5041', 'learning_rate': '9.578e-05', 'epoch': '0.2304'}
+{'loss': '0.9375', 'grad_norm': '0.5791', 'learning_rate': '9.613e-05', 'epoch': '0.2312'}
+{'loss': '0.9414', 'grad_norm': '0.5888', 'learning_rate': '9.649e-05', 'epoch': '0.2321'}
+{'loss': '0.9006', 'grad_norm': '0.5595', 'learning_rate': '9.684e-05', 'epoch': '0.2329'}
+{'loss': '1.028', 'grad_norm': '0.6605', 'learning_rate': '9.719e-05', 'epoch': '0.2338'}
+{'loss': '1.006', 'grad_norm': '0.5657', 'learning_rate': '9.754e-05', 'epoch': '0.2346'}
+{'loss': '0.9835', 'grad_norm': '0.5433', 'learning_rate': '9.789e-05', 'epoch': '0.2354'}
+{'loss': '1.055', 'grad_norm': '0.613', 'learning_rate': '9.824e-05', 'epoch': '0.2363'}
+{'loss': '1.024', 'grad_norm': '0.6496', 'learning_rate': '9.859e-05', 'epoch': '0.2371'}
+{'loss': '1.023', 'grad_norm': '0.6381', 'learning_rate': '9.895e-05', 'epoch': '0.238'}
+{'loss': '1.011', 'grad_norm': '0.6152', 'learning_rate': '9.93e-05', 'epoch': '0.2388'}
+{'loss': '0.9451', 'grad_norm': '0.5728', 'learning_rate': '9.965e-05', 'epoch': '0.2397'}
+{'loss': '0.98', 'grad_norm': '0.6621', 'learning_rate': '0.0001', 'epoch': '0.2405'}
+{'loss': '0.9757', 'grad_norm': '0.6151', 'learning_rate': '0.0001004', 'epoch': '0.2414'}
+{'loss': '0.9105', 'grad_norm': '0.5672', 'learning_rate': '0.0001007', 'epoch': '0.2422'}
+{'loss': '1.103', 'grad_norm': '0.5697', 'learning_rate': '0.0001011', 'epoch': '0.243'}
+{'loss': '1.006', 'grad_norm': '0.559', 'learning_rate': '0.0001014', 'epoch': '0.2439'}
+{'loss': '1.021', 'grad_norm': '0.5519', 'learning_rate': '0.0001018', 'epoch': '0.2447'}
+{'loss': '0.9528', 'grad_norm': '0.6338', 'learning_rate': '0.0001021', 'epoch': '0.2456'}
+{'loss': '1.009', 'grad_norm': '0.6497', 'learning_rate': '0.0001025', 'epoch': '0.2464'}
+{'loss': '0.9961', 'grad_norm': '0.4989', 'learning_rate': '0.0001028', 'epoch': '0.2473'}
+{'loss': '0.9636', 'grad_norm': '0.568', 'learning_rate': '0.0001032', 'epoch': '0.2481'}
+{'loss': '0.9392', 'grad_norm': '0.6582', 'learning_rate': '0.0001035', 'epoch': '0.2489'}
+{'loss': '0.9601', 'grad_norm': '0.6183', 'learning_rate': '0.0001039', 'epoch': '0.2498'}
+{'loss': '1.005', 'grad_norm': '0.6004', 'learning_rate': '0.0001042', 'epoch': '0.2506'}
+{'loss': '0.9523', 'grad_norm': '0.5437', 'learning_rate': '0.0001046', 'epoch': '0.2515'}
+{'loss': '0.9527', 'grad_norm': '0.5525', 'learning_rate': '0.0001049', 'epoch': '0.2523'}
+{'loss': '1.02', 'grad_norm': '0.6795', 'learning_rate': '0.0001053', 'epoch': '0.2532'}
+{'eval_loss': '1.019', 'eval_runtime': '678', 'eval_samples_per_second': '3.108', 'eval_steps_per_second': '3.108', 'epoch': '0.2532'}
+{'loss': '0.9911', 'grad_norm': '0.5646', 'learning_rate': '0.0001056', 'epoch': '0.254'}
+{'loss': '0.9616', 'grad_norm': '0.5238', 'learning_rate': '0.000106', 'epoch': '0.2549'}
+{'loss': '1.008', 'grad_norm': '0.7484', 'learning_rate': '0.0001063', 'epoch': '0.2557'}
+{'loss': '0.9827', 'grad_norm': '0.5789', 'learning_rate': '0.0001067', 'epoch': '0.2565'}
+{'loss': '1.042', 'grad_norm': '0.5526', 'learning_rate': '0.000107', 'epoch': '0.2574'}
+{'loss': '0.9562', 'grad_norm': '0.6721', 'learning_rate': '0.0001074', 'epoch': '0.2582'}
+{'loss': '0.9603', 'grad_norm': '0.5701', 'learning_rate': '0.0001077', 'epoch': '0.2591'}
+{'loss': '1.142', 'grad_norm': '0.635', 'learning_rate': '0.0001081', 'epoch': '0.2599'}
+{'loss': '0.9763', 'grad_norm': '0.5496', 'learning_rate': '0.0001084', 'epoch': '0.2608'}
+{'loss': '1.013', 'grad_norm': '0.6081', 'learning_rate': '0.0001088', 'epoch': '0.2616'}
+{'loss': '0.9386', 'grad_norm': '0.545', 'learning_rate': '0.0001091', 'epoch': '0.2624'}
+{'loss': '0.936', 'grad_norm': '0.5921', 'learning_rate': '0.0001095', 'epoch': '0.2633'}
+{'loss': '1.027', 'grad_norm': '0.6449', 'learning_rate': '0.0001098', 'epoch': '0.2641'}
+{'loss': '0.9303', 'grad_norm': '0.5383', 'learning_rate': '0.0001102', 'epoch': '0.265'}
+{'loss': '0.9369', 'grad_norm': '0.5463', 'learning_rate': '0.0001105', 'epoch': '0.2658'}
+{'loss': '0.9573', 'grad_norm': '0.5819', 'learning_rate': '0.0001109', 'epoch': '0.2667'}
+{'loss': '0.9345', 'grad_norm': '0.5958', 'learning_rate': '0.0001112', 'epoch': '0.2675'}
+{'loss': '1.091', 'grad_norm': '0.6259', 'learning_rate': '0.0001116', 'epoch': '0.2684'}
+{'loss': '1.076', 'grad_norm': '0.5897', 'learning_rate': '0.000112', 'epoch': '0.2692'}
+{'loss': '0.9311', 'grad_norm': '0.5714', 'learning_rate': '0.0001123', 'epoch': '0.27'}
+{'loss': '0.9276', 'grad_norm': '0.4534', 'learning_rate': '0.0001127', 'epoch': '0.2709'}
+{'loss': '1.085', 'grad_norm': '0.6387', 'learning_rate': '0.000113', 'epoch': '0.2717'}
+{'loss': '0.9764', 'grad_norm': '0.6446', 'learning_rate': '0.0001134', 'epoch': '0.2726'}
+{'loss': '0.9518', 'grad_norm': '0.5926', 'learning_rate': '0.0001137', 'epoch': '0.2734'}
+{'loss': '1.059', 'grad_norm': '0.54', 'learning_rate': '0.0001141', 'epoch': '0.2743'}
+{'loss': '0.8833', 'grad_norm': '0.5872', 'learning_rate': '0.0001144', 'epoch': '0.2751'}
+{'loss': '1.031', 'grad_norm': '0.5575', 'learning_rate': '0.0001148', 'epoch': '0.2759'}
+{'loss': '0.8814', 'grad_norm': '0.4779', 'learning_rate': '0.0001151', 'epoch': '0.2768'}
+{'loss': '0.9461', 'grad_norm': '0.5566', 'learning_rate': '0.0001155', 'epoch': '0.2776'}
+{'loss': '0.9475', 'grad_norm': '0.53', 'learning_rate': '0.0001158', 'epoch': '0.2785'}
+{'loss': '1.063', 'grad_norm': '0.6503', 'learning_rate': '0.0001162', 'epoch': '0.2793'}
+{'loss': '0.9389', 'grad_norm': '0.5795', 'learning_rate': '0.0001165', 'epoch': '0.2802'}
+{'loss': '0.8934', 'grad_norm': '0.5763', 'learning_rate': '0.0001169', 'epoch': '0.281'}
+{'loss': '1.107', 'grad_norm': '0.6565', 'learning_rate': '0.0001172', 'epoch': '0.2819'}
+{'loss': '1.05', 'grad_norm': '0.582', 'learning_rate': '0.0001176', 'epoch': '0.2827'}
+{'loss': '0.9881', 'grad_norm': '0.6071', 'learning_rate': '0.0001179', 'epoch': '0.2835'}
+{'loss': '0.9671', 'grad_norm': '0.5765', 'learning_rate': '0.0001183', 'epoch': '0.2844'}
+{'loss': '0.9602', 'grad_norm': '0.5245', 'learning_rate': '0.0001186', 'epoch': '0.2852'}
+{'loss': '0.9685', 'grad_norm': '0.619', 'learning_rate': '0.000119', 'epoch': '0.2861'}
+{'loss': '0.9466', 'grad_norm': '0.6139', 'learning_rate': '0.0001193', 'epoch': '0.2869'}
+{'loss': '0.8496', 'grad_norm': '0.5372', 'learning_rate': '0.0001197', 'epoch': '0.2878'}
+{'loss': '0.9663', 'grad_norm': '0.555', 'learning_rate': '0.00012', 'epoch': '0.2886'}
+{'loss': '0.9736', 'grad_norm': '0.6484', 'learning_rate': '0.0001204', 'epoch': '0.2895'}
+{'loss': '1.059', 'grad_norm': '0.5404', 'learning_rate': '0.0001207', 'epoch': '0.2903'}
+{'loss': '0.9449', 'grad_norm': '0.5658', 'learning_rate': '0.0001211', 'epoch': '0.2911'}
+{'loss': '1.002', 'grad_norm': '0.6041', 'learning_rate': '0.0001214', 'epoch': '0.292'}
+{'loss': '1.035', 'grad_norm': '0.5435', 'learning_rate': '0.0001218', 'epoch': '0.2928'}
+{'loss': '0.9936', 'grad_norm': '0.5713', 'learning_rate': '0.0001221', 'epoch': '0.2937'}
+{'loss': '1.017', 'grad_norm': '0.6138', 'learning_rate': '0.0001225', 'epoch': '0.2945'}
+{'loss': '0.9334', 'grad_norm': '0.4495', 'learning_rate': '0.0001228', 'epoch': '0.2954'}
+{'eval_loss': '0.9969', 'eval_runtime': '668.6', 'eval_samples_per_second': '3.151', 'eval_steps_per_second': '3.151', 'epoch': '0.2954'}
+{'loss': '1.052', 'grad_norm': '0.5322', 'learning_rate': '0.0001232', 'epoch': '0.2962'}
+{'loss': '0.9387', 'grad_norm': '0.5717', 'learning_rate': '0.0001236', 'epoch': '0.297'}
+{'loss': '0.9974', 'grad_norm': '0.5618', 'learning_rate': '0.0001239', 'epoch': '0.2979'}
+{'loss': '0.9955', 'grad_norm': '0.4923', 'learning_rate': '0.0001243', 'epoch': '0.2987'}
+{'loss': '0.9124', 'grad_norm': '0.5435', 'learning_rate': '0.0001246', 'epoch': '0.2996'}
+{'loss': '0.982', 'grad_norm': '0.4959', 'learning_rate': '0.000125', 'epoch': '0.3004'}
+{'loss': '1.015', 'grad_norm': '0.5984', 'learning_rate': '0.0001253', 'epoch': '0.3013'}
+{'loss': '1.058', 'grad_norm': '0.9344', 'learning_rate': '0.0001257', 'epoch': '0.3021'}
+{'loss': '0.983', 'grad_norm': '0.5119', 'learning_rate': '0.000126', 'epoch': '0.303'}
+{'loss': '0.9466', 'grad_norm': '0.494', 'learning_rate': '0.0001264', 'epoch': '0.3038'}
+{'loss': '1.015', 'grad_norm': '0.5966', 'learning_rate': '0.0001267', 'epoch': '0.3046'}
+{'loss': '0.9704', 'grad_norm': '0.5021', 'learning_rate': '0.0001271', 'epoch': '0.3055'}
+{'loss': '1.026', 'grad_norm': '0.6067', 'learning_rate': '0.0001274', 'epoch': '0.3063'}
+{'loss': '0.9336', 'grad_norm': '0.5932', 'learning_rate': '0.0001278', 'epoch': '0.3072'}
+{'loss': '1.092', 'grad_norm': '0.5939', 'learning_rate': '0.0001281', 'epoch': '0.308'}
+{'loss': '0.9631', 'grad_norm': '0.4973', 'learning_rate': '0.0001285', 'epoch': '0.3089'}
+{'loss': '0.9513', 'grad_norm': '0.5452', 'learning_rate': '0.0001288', 'epoch': '0.3097'}
+{'loss': '0.8576', 'grad_norm': '0.5498', 'learning_rate': '0.0001292', 'epoch': '0.3105'}
+{'loss': '0.9543', 'grad_norm': '0.558', 'learning_rate': '0.0001295', 'epoch': '0.3114'}
+{'loss': '0.8993', 'grad_norm': '0.589', 'learning_rate': '0.0001299', 'epoch': '0.3122'}
+{'loss': '1.032', 'grad_norm': '0.5943', 'learning_rate': '0.0001302', 'epoch': '0.3131'}
+{'loss': '0.9217', 'grad_norm': '0.6342', 'learning_rate': '0.0001306', 'epoch': '0.3139'}
+{'loss': '0.9518', 'grad_norm': '0.5294', 'learning_rate': '0.0001309', 'epoch': '0.3148'}
+{'loss': '0.946', 'grad_norm': '0.6372', 'learning_rate': '0.0001313', 'epoch': '0.3156'}
+{'loss': '0.9626', 'grad_norm': '0.598', 'learning_rate': '0.0001316', 'epoch': '0.3165'}
+{'loss': '1.026', 'grad_norm': '0.5682', 'learning_rate': '0.000132', 'epoch': '0.3173'}
+{'loss': '0.932', 'grad_norm': '0.5349', 'learning_rate': '0.0001323', 'epoch': '0.3181'}
+{'loss': '0.9217', 'grad_norm': '0.6094', 'learning_rate': '0.0001327', 'epoch': '0.319'}
+{'loss': '0.9017', 'grad_norm': '0.5189', 'learning_rate': '0.000133', 'epoch': '0.3198'}
+{'loss': '1.036', 'grad_norm': '0.5877', 'learning_rate': '0.0001334', 'epoch': '0.3207'}
+{'loss': '0.8787', 'grad_norm': '0.5543', 'learning_rate': '0.0001337', 'epoch': '0.3215'}
+{'loss': '0.9237', 'grad_norm': '0.5085', 'learning_rate': '0.0001341', 'epoch': '0.3224'}
+{'loss': '1.015', 'grad_norm': '0.5462', 'learning_rate': '0.0001344', 'epoch': '0.3232'}
+{'loss': '0.8985', 'grad_norm': '0.5348', 'learning_rate': '0.0001348', 'epoch': '0.3241'}
+{'loss': '1.023', 'grad_norm': '0.5581', 'learning_rate': '0.0001351', 'epoch': '0.3249'}
+{'loss': '0.9571', 'grad_norm': '0.5203', 'learning_rate': '0.0001355', 'epoch': '0.3257'}
+{'loss': '1.047', 'grad_norm': '0.505', 'learning_rate': '0.0001359', 'epoch': '0.3266'}
+{'loss': '0.7743', 'grad_norm': '0.4723', 'learning_rate': '0.0001362', 'epoch': '0.3274'}
+{'loss': '1.022', 'grad_norm': '0.6311', 'learning_rate': '0.0001366', 'epoch': '0.3283'}
+{'loss': '0.9134', 'grad_norm': '0.4707', 'learning_rate': '0.0001369', 'epoch': '0.3291'}
+{'loss': '0.9798', 'grad_norm': '0.5725', 'learning_rate': '0.0001373', 'epoch': '0.33'}
+{'loss': '0.9373', 'grad_norm': '0.514', 'learning_rate': '0.0001376', 'epoch': '0.3308'}
+{'loss': '0.9528', 'grad_norm': '0.5276', 'learning_rate': '0.000138', 'epoch': '0.3316'}
+{'loss': '0.975', 'grad_norm': '0.5703', 'learning_rate': '0.0001383', 'epoch': '0.3325'}
+{'loss': '0.9412', 'grad_norm': '0.5281', 'learning_rate': '0.0001387', 'epoch': '0.3333'}
+{'loss': '0.9976', 'grad_norm': '0.5579', 'learning_rate': '0.000139', 'epoch': '0.3342'}
+{'loss': '0.9082', 'grad_norm': '0.5046', 'learning_rate': '0.0001394', 'epoch': '0.335'}
+{'loss': '1.008', 'grad_norm': '0.5259', 'learning_rate': '0.0001397', 'epoch': '0.3359'}
+{'loss': '0.9729', 'grad_norm': '0.5548', 'learning_rate': '0.0001401', 'epoch': '0.3367'}
+{'loss': '0.91', 'grad_norm': '0.486', 'learning_rate': '0.0001404', 'epoch': '0.3376'}
+{'eval_loss': '0.98', 'eval_runtime': '678.8', 'eval_samples_per_second': '3.104', 'eval_steps_per_second': '3.104', 'epoch': '0.3376'}
+{'loss': '0.8611', 'grad_norm': '0.5413', 'learning_rate': '0.0001408', 'epoch': '0.3384'}
+{'loss': '0.9585', 'grad_norm': '0.5147', 'learning_rate': '0.0001411', 'epoch': '0.3392'}
+{'loss': '1.014', 'grad_norm': '0.5932', 'learning_rate': '0.0001415', 'epoch': '0.3401'}
+{'loss': '1.008', 'grad_norm': '0.5178', 'learning_rate': '0.0001418', 'epoch': '0.3409'}
+{'loss': '0.9107', 'grad_norm': '0.5453', 'learning_rate': '0.0001422', 'epoch': '0.3418'}
+{'loss': '0.9981', 'grad_norm': '0.5886', 'learning_rate': '0.0001425', 'epoch': '0.3426'}
+{'loss': '0.9636', 'grad_norm': '0.4804', 'learning_rate': '0.0001429', 'epoch': '0.3435'}
+{'loss': '0.9591', 'grad_norm': '0.5012', 'learning_rate': '0.0001432', 'epoch': '0.3443'}
+{'loss': '0.9782', 'grad_norm': '0.5786', 'learning_rate': '0.0001436', 'epoch': '0.3451'}
+{'loss': '0.855', 'grad_norm': '0.4609', 'learning_rate': '0.0001439', 'epoch': '0.346'}
+{'loss': '0.9771', 'grad_norm': '0.5756', 'learning_rate': '0.0001443', 'epoch': '0.3468'}
+{'loss': '0.874', 'grad_norm': '0.5502', 'learning_rate': '0.0001446', 'epoch': '0.3477'}
+{'loss': '0.9129', 'grad_norm': '0.5452', 'learning_rate': '0.000145', 'epoch': '0.3485'}
+{'loss': '0.9197', 'grad_norm': '0.4625', 'learning_rate': '0.0001453', 'epoch': '0.3494'}
+{'loss': '0.9977', 'grad_norm': '0.5209', 'learning_rate': '0.0001457', 'epoch': '0.3502'}
+{'loss': '0.7753', 'grad_norm': '0.4469', 'learning_rate': '0.000146', 'epoch': '0.3511'}
+{'loss': '0.9585', 'grad_norm': '0.5105', 'learning_rate': '0.0001464', 'epoch': '0.3519'}
+{'loss': '0.9014', 'grad_norm': '0.5044', 'learning_rate': '0.0001467', 'epoch': '0.3527'}
+{'loss': '0.9776', 'grad_norm': '0.5759', 'learning_rate': '0.0001471', 'epoch': '0.3536'}
+{'loss': '0.9279', 'grad_norm': '0.5129', 'learning_rate': '0.0001475', 'epoch': '0.3544'}
+{'loss': '1.009', 'grad_norm': '0.5758', 'learning_rate': '0.0001478', 'epoch': '0.3553'}
+{'loss': '0.9287', 'grad_norm': '0.5009', 'learning_rate': '0.0001482', 'epoch': '0.3561'}
+{'loss': '0.9484', 'grad_norm': '0.5221', 'learning_rate': '0.0001485', 'epoch': '0.357'}
+{'loss': '1.056', 'grad_norm': '0.5495', 'learning_rate': '0.0001489', 'epoch': '0.3578'}
+{'loss': '0.9331', 'grad_norm': '0.4663', 'learning_rate': '0.0001492', 'epoch': '0.3586'}
+{'loss': '1.002', 'grad_norm': '0.4865', 'learning_rate': '0.0001496', 'epoch': '0.3595'}
+{'loss': '0.9085', 'grad_norm': '0.5033', 'learning_rate': '0.0001499', 'epoch': '0.3603'}
+{'loss': '0.9235', 'grad_norm': '0.5425', 'learning_rate': '0.0001503', 'epoch': '0.3612'}
+{'loss': '0.9192', 'grad_norm': '0.5182', 'learning_rate': '0.0001506', 'epoch': '0.362'}
+{'loss': '0.852', 'grad_norm': '0.5129', 'learning_rate': '0.000151', 'epoch': '0.3629'}
+{'loss': '0.8634', 'grad_norm': '0.4954', 'learning_rate': '0.0001513', 'epoch': '0.3637'}
+{'loss': '0.9474', 'grad_norm': '0.4707', 'learning_rate': '0.0001517', 'epoch': '0.3646'}
+{'loss': '0.9424', 'grad_norm': '0.5877', 'learning_rate': '0.000152', 'epoch': '0.3654'}
+{'loss': '0.9811', 'grad_norm': '0.5851', 'learning_rate': '0.0001524', 'epoch': '0.3662'}
+{'loss': '0.8151', 'grad_norm': '0.4636', 'learning_rate': '0.0001527', 'epoch': '0.3671'}
+{'loss': '0.9023', 'grad_norm': '0.4725', 'learning_rate': '0.0001531', 'epoch': '0.3679'}
+{'loss': '0.8758', 'grad_norm': '0.4751', 'learning_rate': '0.0001534', 'epoch': '0.3688'}
+{'loss': '0.9837', 'grad_norm': '0.5687', 'learning_rate': '0.0001538', 'epoch': '0.3696'}
+{'loss': '0.9084', 'grad_norm': '0.4906', 'learning_rate': '0.0001541', 'epoch': '0.3705'}
+{'loss': '0.8426', 'grad_norm': '0.5288', 'learning_rate': '0.0001545', 'epoch': '0.3713'}
+{'loss': '1.006', 'grad_norm': '0.5447', 'learning_rate': '0.0001548', 'epoch': '0.3722'}
+{'loss': '0.9779', 'grad_norm': '0.4687', 'learning_rate': '0.0001552', 'epoch': '0.373'}
+{'loss': '0.91', 'grad_norm': '0.4853', 'learning_rate': '0.0001555', 'epoch': '0.3738'}
+{'loss': '0.9731', 'grad_norm': '0.522', 'learning_rate': '0.0001559', 'epoch': '0.3747'}
+{'loss': '0.9305', 'grad_norm': '0.5259', 'learning_rate': '0.0001562', 'epoch': '0.3755'}
+{'loss': '0.8681', 'grad_norm': '0.4267', 'learning_rate': '0.0001566', 'epoch': '0.3764'}
+{'loss': '0.8428', 'grad_norm': '0.4636', 'learning_rate': '0.0001569', 'epoch': '0.3772'}
+{'loss': '0.8466', 'grad_norm': '0.5314', 'learning_rate': '0.0001573', 'epoch': '0.3781'}
+{'loss': '0.912', 'grad_norm': '0.4717', 'learning_rate': '0.0001576', 'epoch': '0.3789'}
+{'loss': '0.895', 'grad_norm': '0.4647', 'learning_rate': '0.000158', 'epoch': '0.3797'}
+{'eval_loss': '0.9643', 'eval_runtime': '691.8', 'eval_samples_per_second': '3.046', 'eval_steps_per_second': '3.046', 'epoch': '0.3797'}
+{'loss': '0.9082', 'grad_norm': '0.5445', 'learning_rate': '0.0001583', 'epoch': '0.3806'}
+{'loss': '0.8946', 'grad_norm': '0.4312', 'learning_rate': '0.0001587', 'epoch': '0.3814'}
+{'loss': '0.9478', 'grad_norm': '0.5591', 'learning_rate': '0.0001591', 'epoch': '0.3823'}
+{'loss': '0.926', 'grad_norm': '0.547', 'learning_rate': '0.0001594', 'epoch': '0.3831'}
+{'loss': '0.8825', 'grad_norm': '0.5499', 'learning_rate': '0.0001598', 'epoch': '0.384'}
+{'loss': '0.8842', 'grad_norm': '0.4779', 'learning_rate': '0.0001601', 'epoch': '0.3848'}
+{'loss': '0.9295', 'grad_norm': '0.5391', 'learning_rate': '0.0001605', 'epoch': '0.3857'}
+{'loss': '0.8421', 'grad_norm': '0.5055', 'learning_rate': '0.0001608', 'epoch': '0.3865'}
+{'loss': '0.8257', 'grad_norm': '0.4464', 'learning_rate': '0.0001612', 'epoch': '0.3873'}
+{'loss': '0.8701', 'grad_norm': '0.5321', 'learning_rate': '0.0001615', 'epoch': '0.3882'}
+{'loss': '0.9019', 'grad_norm': '0.4343', 'learning_rate': '0.0001619', 'epoch': '0.389'}
+{'loss': '0.8561', 'grad_norm': '0.5312', 'learning_rate': '0.0001622', 'epoch': '0.3899'}
+{'loss': '0.97', 'grad_norm': '0.5887', 'learning_rate': '0.0001626', 'epoch': '0.3907'}
+{'loss': '1.044', 'grad_norm': '0.5426', 'learning_rate': '0.0001629', 'epoch': '0.3916'}
+{'loss': '0.9502', 'grad_norm': '0.5229', 'learning_rate': '0.0001633', 'epoch': '0.3924'}
+{'loss': '0.9223', 'grad_norm': '0.4896', 'learning_rate': '0.0001636', 'epoch': '0.3932'}
+{'loss': '0.9693', 'grad_norm': '0.452', 'learning_rate': '0.000164', 'epoch': '0.3941'}
+{'loss': '0.9467', 'grad_norm': '0.5299', 'learning_rate': '0.0001643', 'epoch': '0.3949'}
+{'loss': '0.9118', 'grad_norm': '0.5608', 'learning_rate': '0.0001647', 'epoch': '0.3958'}
+{'loss': '0.9131', 'grad_norm': '0.5272', 'learning_rate': '0.000165', 'epoch': '0.3966'}
+{'loss': '0.9358', 'grad_norm': '0.5075', 'learning_rate': '0.0001654', 'epoch': '0.3975'}
+{'loss': '0.8908', 'grad_norm': '0.5157', 'learning_rate': '0.0001657', 'epoch': '0.3983'}
+{'loss': '0.9775', 'grad_norm': '0.4857', 'learning_rate': '0.0001661', 'epoch': '0.3992'}
+{'loss': '0.8957', 'grad_norm': '0.4885', 'learning_rate': '0.0001664', 'epoch': '0.4'}
+{'loss': '0.9036', 'grad_norm': '0.4299', 'learning_rate': '0.0001668', 'epoch': '0.4008'}
+{'loss': '0.991', 'grad_norm': '0.4955', 'learning_rate': '0.0001671', 'epoch': '0.4017'}
+{'loss': '0.8231', 'grad_norm': '0.4565', 'learning_rate': '0.0001675', 'epoch': '0.4025'}
+{'loss': '0.8785', 'grad_norm': '0.429', 'learning_rate': '0.0001678', 'epoch': '0.4034'}
+{'loss': '0.8764', 'grad_norm': '0.542', 'learning_rate': '0.0001682', 'epoch': '0.4042'}
+{'loss': '0.9471', 'grad_norm': '0.5178', 'learning_rate': '0.0001685', 'epoch': '0.4051'}
+{'loss': '0.9235', 'grad_norm': '0.5397', 'learning_rate': '0.0001689', 'epoch': '0.4059'}
+{'loss': '1.025', 'grad_norm': '0.5325', 'learning_rate': '0.0001692', 'epoch': '0.4068'}
+{'loss': '0.9077', 'grad_norm': '0.4936', 'learning_rate': '0.0001696', 'epoch': '0.4076'}
+{'loss': '0.9163', 'grad_norm': '0.4961', 'learning_rate': '0.0001699', 'epoch': '0.4084'}
+{'loss': '0.9041', 'grad_norm': '0.4509', 'learning_rate': '0.0001703', 'epoch': '0.4093'}
+{'loss': '0.8633', 'grad_norm': '0.456', 'learning_rate': '0.0001707', 'epoch': '0.4101'}
+{'loss': '0.8454', 'grad_norm': '0.4489', 'learning_rate': '0.000171', 'epoch': '0.411'}
+{'loss': '0.8247', 'grad_norm': '0.4725', 'learning_rate': '0.0001714', 'epoch': '0.4118'}
+{'loss': '0.9497', 'grad_norm': '0.6598', 'learning_rate': '0.0001717', 'epoch': '0.4127'}
+{'loss': '0.9419', 'grad_norm': '0.3981', 'learning_rate': '0.0001721', 'epoch': '0.4135'}
+{'loss': '0.9183', 'grad_norm': '0.4823', 'learning_rate': '0.0001724', 'epoch': '0.4143'}
+{'loss': '0.9387', 'grad_norm': '0.5439', 'learning_rate': '0.0001728', 'epoch': '0.4152'}
+{'loss': '0.8978', 'grad_norm': '0.496', 'learning_rate': '0.0001731', 'epoch': '0.416'}
+{'loss': '0.8777', 'grad_norm': '0.4353', 'learning_rate': '0.0001735', 'epoch': '0.4169'}
+{'loss': '0.9163', 'grad_norm': '0.4751', 'learning_rate': '0.0001738', 'epoch': '0.4177'}
+{'loss': '0.9483', 'grad_norm': '0.4907', 'learning_rate': '0.0001742', 'epoch': '0.4186'}
+{'loss': '0.8813', 'grad_norm': '0.445', 'learning_rate': '0.0001745', 'epoch': '0.4194'}
+{'loss': '0.944', 'grad_norm': '0.4536', 'learning_rate': '0.0001749', 'epoch': '0.4203'}
+{'loss': '0.8818', 'grad_norm': '0.5241', 'learning_rate': '0.0001752', 'epoch': '0.4211'}
+{'loss': '0.8862', 'grad_norm': '0.5301', 'learning_rate': '0.0001756', 'epoch': '0.4219'}
+{'eval_loss': '0.9487', 'eval_runtime': '689.4', 'eval_samples_per_second': '3.056', 'eval_steps_per_second': '3.056', 'epoch': '0.4219'}
+{'loss': '0.8814', 'grad_norm': '0.4788', 'learning_rate': '0.0001759', 'epoch': '0.4228'}
+{'loss': '0.8295', 'grad_norm': '0.493', 'learning_rate': '0.0001763', 'epoch': '0.4236'}
+{'loss': '0.8365', 'grad_norm': '0.4189', 'learning_rate': '0.0001766', 'epoch': '0.4245'}
+{'loss': '1.017', 'grad_norm': '0.5175', 'learning_rate': '0.000177', 'epoch': '0.4253'}
+{'loss': '0.8461', 'grad_norm': '0.4815', 'learning_rate': '0.0001773', 'epoch': '0.4262'}
+{'loss': '0.8596', 'grad_norm': '0.5211', 'learning_rate': '0.0001777', 'epoch': '0.427'}
+{'loss': '0.8765', 'grad_norm': '0.44', 'learning_rate': '0.000178', 'epoch': '0.4278'}
+{'loss': '0.9707', 'grad_norm': '0.5282', 'learning_rate': '0.0001784', 'epoch': '0.4287'}
+{'loss': '0.8912', 'grad_norm': '0.4752', 'learning_rate': '0.0001787', 'epoch': '0.4295'}
+{'loss': '0.8836', 'grad_norm': '0.4964', 'learning_rate': '0.0001791', 'epoch': '0.4304'}
+{'loss': '0.8712', 'grad_norm': '0.5253', 'learning_rate': '0.0001794', 'epoch': '0.4312'}
+{'loss': '0.9065', 'grad_norm': '0.4929', 'learning_rate': '0.0001798', 'epoch': '0.4321'}
+{'loss': '0.8993', 'grad_norm': '0.4876', 'learning_rate': '0.0001801', 'epoch': '0.4329'}
+{'loss': '0.8748', 'grad_norm': '0.4815', 'learning_rate': '0.0001805', 'epoch': '0.4338'}
+{'loss': '0.8879', 'grad_norm': '0.493', 'learning_rate': '0.0001808', 'epoch': '0.4346'}
+{'loss': '1.002', 'grad_norm': '0.493', 'learning_rate': '0.0001812', 'epoch': '0.4354'}
+{'loss': '0.9387', 'grad_norm': '0.5312', 'learning_rate': '0.0001815', 'epoch': '0.4363'}
+{'loss': '0.8542', 'grad_norm': '0.5175', 'learning_rate': '0.0001819', 'epoch': '0.4371'}
+{'loss': '0.9059', 'grad_norm': '0.5453', 'learning_rate': '0.0001822', 'epoch': '0.438'}
+{'loss': '0.9709', 'grad_norm': '0.5481', 'learning_rate': '0.0001826', 'epoch': '0.4388'}
+{'loss': '0.8561', 'grad_norm': '0.4469', 'learning_rate': '0.000183', 'epoch': '0.4397'}
+{'loss': '0.8812', 'grad_norm': '0.4024', 'learning_rate': '0.0001833', 'epoch': '0.4405'}
+{'loss': '0.9781', 'grad_norm': '0.5586', 'learning_rate': '0.0001837', 'epoch': '0.4414'}
+{'loss': '0.8641', 'grad_norm': '0.4919', 'learning_rate': '0.000184', 'epoch': '0.4422'}
+{'loss': '0.9078', 'grad_norm': '0.4335', 'learning_rate': '0.0001844', 'epoch': '0.443'}
+{'loss': '0.9607', 'grad_norm': '0.5275', 'learning_rate': '0.0001847', 'epoch': '0.4439'}
+{'loss': '0.855', 'grad_norm': '0.5144', 'learning_rate': '0.0001851', 'epoch': '0.4447'}
+{'loss': '0.9321', 'grad_norm': '0.5167', 'learning_rate': '0.0001854', 'epoch': '0.4456'}
+{'loss': '0.9632', 'grad_norm': '0.5012', 'learning_rate': '0.0001858', 'epoch': '0.4464'}
+{'loss': '0.9443', 'grad_norm': '0.4931', 'learning_rate': '0.0001861', 'epoch': '0.4473'}
+{'loss': '0.9528', 'grad_norm': '0.4996', 'learning_rate': '0.0001865', 'epoch': '0.4481'}
+{'loss': '0.9631', 'grad_norm': '0.4983', 'learning_rate': '0.0001868', 'epoch': '0.4489'}
+{'loss': '0.9081', 'grad_norm': '0.524', 'learning_rate': '0.0001872', 'epoch': '0.4498'}
+{'loss': '0.9399', 'grad_norm': '0.5109', 'learning_rate': '0.0001875', 'epoch': '0.4506'}
+{'loss': '0.9359', 'grad_norm': '0.4454', 'learning_rate': '0.0001879', 'epoch': '0.4515'}
+{'loss': '0.888', 'grad_norm': '0.4676', 'learning_rate': '0.0001882', 'epoch': '0.4523'}
+{'loss': '0.9966', 'grad_norm': '0.474', 'learning_rate': '0.0001886', 'epoch': '0.4532'}
+{'loss': '0.9184', 'grad_norm': '0.4605', 'learning_rate': '0.0001889', 'epoch': '0.454'}
+{'loss': '0.9435', 'grad_norm': '0.5601', 'learning_rate': '0.0001893', 'epoch': '0.4549'}
+{'loss': '0.9522', 'grad_norm': '0.4842', 'learning_rate': '0.0001896', 'epoch': '0.4557'}
+{'loss': '0.9315', 'grad_norm': '0.4628', 'learning_rate': '0.00019', 'epoch': '0.4565'}
+{'loss': '0.8166', 'grad_norm': '0.4659', 'learning_rate': '0.0001903', 'epoch': '0.4574'}
+{'loss': '0.8024', 'grad_norm': '0.4192', 'learning_rate': '0.0001907', 'epoch': '0.4582'}
+{'loss': '0.8621', 'grad_norm': '0.4179', 'learning_rate': '0.000191', 'epoch': '0.4591'}
+{'loss': '0.9122', 'grad_norm': '0.4489', 'learning_rate': '0.0001914', 'epoch': '0.4599'}
+{'loss': '0.9497', 'grad_norm': '0.5165', 'learning_rate': '0.0001917', 'epoch': '0.4608'}
+{'loss': '0.8758', 'grad_norm': '0.4156', 'learning_rate': '0.0001921', 'epoch': '0.4616'}
+{'loss': '0.8407', 'grad_norm': '2.322', 'learning_rate': '0.0001924', 'epoch': '0.4624'}
+{'loss': '0.9459', 'grad_norm': '0.4814', 'learning_rate': '0.0001928', 'epoch': '0.4633'}
+{'loss': '0.9377', 'grad_norm': '0.6234', 'learning_rate': '0.0001931', 'epoch': '0.4641'}
+{'eval_loss': '0.9358', 'eval_runtime': '677', 'eval_samples_per_second': '3.112', 'eval_steps_per_second': '3.112', 'epoch': '0.4641'}
+{'loss': '0.8999', 'grad_norm': '0.5126', 'learning_rate': '0.0001935', 'epoch': '0.465'}
+{'loss': '0.9214', 'grad_norm': '0.4534', 'learning_rate': '0.0001938', 'epoch': '0.4658'}
+{'loss': '0.974', 'grad_norm': '0.4715', 'learning_rate': '0.0001942', 'epoch': '0.4667'}
+{'loss': '0.9113', 'grad_norm': '0.4553', 'learning_rate': '0.0001946', 'epoch': '0.4675'}
+{'loss': '0.9253', 'grad_norm': '0.482', 'learning_rate': '0.0001949', 'epoch': '0.4684'}
+{'loss': '0.8613', 'grad_norm': '0.4425', 'learning_rate': '0.0001953', 'epoch': '0.4692'}
+{'loss': '0.8902', 'grad_norm': '0.4312', 'learning_rate': '0.0001956', 'epoch': '0.47'}
+{'loss': '0.9272', 'grad_norm': '0.4978', 'learning_rate': '0.000196', 'epoch': '0.4709'}
+{'loss': '0.8741', 'grad_norm': '0.4715', 'learning_rate': '0.0001963', 'epoch': '0.4717'}
+{'loss': '0.919', 'grad_norm': '0.503', 'learning_rate': '0.0001967', 'epoch': '0.4726'}
+{'loss': '1.004', 'grad_norm': '0.4797', 'learning_rate': '0.000197', 'epoch': '0.4734'}
+{'loss': '0.9473', 'grad_norm': '0.451', 'learning_rate': '0.0001974', 'epoch': '0.4743'}
+{'loss': '0.8043', 'grad_norm': '0.4599', 'learning_rate': '0.0001977', 'epoch': '0.4751'}
+{'loss': '0.9662', 'grad_norm': '0.5313', 'learning_rate': '0.0001981', 'epoch': '0.4759'}
+{'loss': '0.8691', 'grad_norm': '0.4347', 'learning_rate': '0.0001984', 'epoch': '0.4768'}
+{'loss': '0.9024', 'grad_norm': '0.5232', 'learning_rate': '0.0001988', 'epoch': '0.4776'}
+{'loss': '0.8997', 'grad_norm': '0.4767', 'learning_rate': '0.0001991', 'epoch': '0.4785'}
+{'loss': '0.8337', 'grad_norm': '0.4229', 'learning_rate': '0.0001995', 'epoch': '0.4793'}
+{'loss': '0.8943', 'grad_norm': '0.4701', 'learning_rate': '0.0001998', 'epoch': '0.4802'}
+{'loss': '0.934', 'grad_norm': '0.5341', 'learning_rate': '0.0002', 'epoch': '0.481'}
+{'loss': '0.8242', 'grad_norm': '0.4617', 'learning_rate': '0.0002', 'epoch': '0.4819'}
+{'loss': '1.01', 'grad_norm': '0.5475', 'learning_rate': '0.0002', 'epoch': '0.4827'}
+{'loss': '0.965', 'grad_norm': '0.5246', 'learning_rate': '0.0002', 'epoch': '0.4835'}
+{'loss': '0.9003', 'grad_norm': '0.4326', 'learning_rate': '0.0002', 'epoch': '0.4844'}
+{'loss': '0.9793', 'grad_norm': '0.4903', 'learning_rate': '0.0002', 'epoch': '0.4852'}
+{'loss': '0.8793', 'grad_norm': '0.476', 'learning_rate': '0.0002', 'epoch': '0.4861'}
+{'loss': '0.8886', 'grad_norm': '0.4461', 'learning_rate': '0.0002', 'epoch': '0.4869'}
+{'loss': '0.966', 'grad_norm': '0.503', 'learning_rate': '0.0002', 'epoch': '0.4878'}
+{'loss': '0.8247', 'grad_norm': '0.4296', 'learning_rate': '0.0002', 'epoch': '0.4886'}
+{'loss': '0.816', 'grad_norm': '0.3922', 'learning_rate': '0.0002', 'epoch': '0.4895'}
+{'loss': '0.9631', 'grad_norm': '0.4846', 'learning_rate': '0.0002', 'epoch': '0.4903'}
+{'loss': '0.8396', 'grad_norm': '0.4589', 'learning_rate': '0.0002', 'epoch': '0.4911'}
+{'loss': '0.8195', 'grad_norm': '0.5553', 'learning_rate': '0.0002', 'epoch': '0.492'}
+{'loss': '0.9269', 'grad_norm': '0.39', 'learning_rate': '0.0002', 'epoch': '0.4928'}
+{'loss': '0.9205', 'grad_norm': '0.4545', 'learning_rate': '0.0002', 'epoch': '0.4937'}
+{'loss': '0.927', 'grad_norm': '0.4842', 'learning_rate': '0.0002', 'epoch': '0.4945'}
+{'loss': '0.8783', 'grad_norm': '0.4522', 'learning_rate': '0.0002', 'epoch': '0.4954'}
+{'loss': '0.8731', 'grad_norm': '0.4158', 'learning_rate': '0.0002', 'epoch': '0.4962'}
+{'loss': '0.9593', 'grad_norm': '0.4633', 'learning_rate': '0.0002', 'epoch': '0.497'}
+{'loss': '0.9216', 'grad_norm': '0.5008', 'learning_rate': '0.0002', 'epoch': '0.4979'}
+{'loss': '0.8964', 'grad_norm': '0.5202', 'learning_rate': '0.0002', 'epoch': '0.4987'}
+{'loss': '0.8529', 'grad_norm': '0.5054', 'learning_rate': '0.0002', 'epoch': '0.4996'}
+{'loss': '0.8215', 'grad_norm': '0.444', 'learning_rate': '0.0002', 'epoch': '0.5004'}
+{'loss': '0.8882', 'grad_norm': '0.4563', 'learning_rate': '0.0002', 'epoch': '0.5013'}
+{'loss': '0.8883', 'grad_norm': '0.4667', 'learning_rate': '0.0002', 'epoch': '0.5021'}
+{'loss': '0.8354', 'grad_norm': '0.4424', 'learning_rate': '0.0002', 'epoch': '0.503'}
+{'loss': '0.8721', 'grad_norm': '0.4515', 'learning_rate': '0.0002', 'epoch': '0.5038'}
+{'loss': '0.9028', 'grad_norm': '0.4847', 'learning_rate': '0.0002', 'epoch': '0.5046'}
+{'loss': '0.8843', 'grad_norm': '0.4932', 'learning_rate': '0.0002', 'epoch': '0.5055'}
+{'loss': '0.9179', 'grad_norm': '0.5074', 'learning_rate': '0.0002', 'epoch': '0.5063'}
+{'eval_loss': '0.9224', 'eval_runtime': '669.8', 'eval_samples_per_second': '3.146', 'eval_steps_per_second': '3.146', 'epoch': '0.5063'}
+{'loss': '0.9079', 'grad_norm': '0.4024', 'learning_rate': '0.0002', 'epoch': '0.5072'}
+{'loss': '0.8089', 'grad_norm': '0.6559', 'learning_rate': '0.0002', 'epoch': '0.508'}
+{'loss': '0.9316', 'grad_norm': '0.4513', 'learning_rate': '0.0002', 'epoch': '0.5089'}
+{'loss': '0.9048', 'grad_norm': '0.4734', 'learning_rate': '0.0002', 'epoch': '0.5097'}
+{'loss': '0.8507', 'grad_norm': '0.4026', 'learning_rate': '0.0002', 'epoch': '0.5105'}
+{'loss': '0.849', 'grad_norm': '0.4342', 'learning_rate': '0.0002', 'epoch': '0.5114'}
+{'loss': '0.773', 'grad_norm': '0.3678', 'learning_rate': '0.0002', 'epoch': '0.5122'}
+{'loss': '0.9625', 'grad_norm': '0.4904', 'learning_rate': '0.0002', 'epoch': '0.5131'}
+{'loss': '0.8962', 'grad_norm': '0.4577', 'learning_rate': '0.0002', 'epoch': '0.5139'}
+{'loss': '0.8509', 'grad_norm': '0.4466', 'learning_rate': '0.0002', 'epoch': '0.5148'}
+{'loss': '0.8665', 'grad_norm': '0.4268', 'learning_rate': '0.0002', 'epoch': '0.5156'}
+{'loss': '0.8841', 'grad_norm': '0.3994', 'learning_rate': '0.0002', 'epoch': '0.5165'}
+{'loss': '0.8601', 'grad_norm': '0.4442', 'learning_rate': '0.0002', 'epoch': '0.5173'}
+{'loss': '0.8377', 'grad_norm': '0.4397', 'learning_rate': '0.0002', 'epoch': '0.5181'}
+{'loss': '0.9854', 'grad_norm': '0.5328', 'learning_rate': '0.0002', 'epoch': '0.519'}
+{'loss': '0.9236', 'grad_norm': '0.4093', 'learning_rate': '0.0002', 'epoch': '0.5198'}
+{'loss': '0.8684', 'grad_norm': '0.4599', 'learning_rate': '0.0002', 'epoch': '0.5207'}
+{'loss': '0.9947', 'grad_norm': '0.4814', 'learning_rate': '0.0002', 'epoch': '0.5215'}
+{'loss': '0.8609', 'grad_norm': '0.4352', 'learning_rate': '0.0002', 'epoch': '0.5224'}
+{'loss': '0.8194', 'grad_norm': '0.4565', 'learning_rate': '0.0002', 'epoch': '0.5232'}
+{'loss': '0.8935', 'grad_norm': '0.4683', 'learning_rate': '0.0002', 'epoch': '0.5241'}
+{'loss': '0.8625', 'grad_norm': '0.4607', 'learning_rate': '0.0002', 'epoch': '0.5249'}
+{'loss': '0.8541', 'grad_norm': '0.4507', 'learning_rate': '0.0002', 'epoch': '0.5257'}
+{'loss': '0.9003', 'grad_norm': '0.4551', 'learning_rate': '0.0002', 'epoch': '0.5266'}
+{'loss': '0.849', 'grad_norm': '0.4098', 'learning_rate': '0.0002', 'epoch': '0.5274'}
+{'loss': '0.9205', 'grad_norm': '0.4376', 'learning_rate': '0.0002', 'epoch': '0.5283'}
+{'loss': '0.8832', 'grad_norm': '0.4335', 'learning_rate': '0.0002', 'epoch': '0.5291'}
+{'loss': '0.8578', 'grad_norm': '0.4841', 'learning_rate': '0.0002', 'epoch': '0.53'}
+{'loss': '0.9235', 'grad_norm': '0.4891', 'learning_rate': '0.0002', 'epoch': '0.5308'}
+{'loss': '0.7898', 'grad_norm': '0.4111', 'learning_rate': '0.0002', 'epoch': '0.5316'}
+{'loss': '0.9393', 'grad_norm': '0.4223', 'learning_rate': '0.0002', 'epoch': '0.5325'}
+{'loss': '0.913', 'grad_norm': '0.3915', 'learning_rate': '0.0002', 'epoch': '0.5333'}
+{'loss': '0.9691', 'grad_norm': '0.4278', 'learning_rate': '0.0002', 'epoch': '0.5342'}
+{'loss': '0.999', 'grad_norm': '0.4823', 'learning_rate': '0.0002', 'epoch': '0.535'}
+{'loss': '0.8343', 'grad_norm': '0.467', 'learning_rate': '0.0002', 'epoch': '0.5359'}
+{'loss': '0.9606', 'grad_norm': '0.4025', 'learning_rate': '0.0001999', 'epoch': '0.5367'}
+{'loss': '0.8164', 'grad_norm': '0.4206', 'learning_rate': '0.0001999', 'epoch': '0.5376'}
+{'loss': '0.8798', 'grad_norm': '0.4795', 'learning_rate': '0.0001999', 'epoch': '0.5384'}
+{'loss': '0.7659', 'grad_norm': '0.3967', 'learning_rate': '0.0001999', 'epoch': '0.5392'}
+{'loss': '0.9477', 'grad_norm': '0.5136', 'learning_rate': '0.0001999', 'epoch': '0.5401'}
+{'loss': '0.7884', 'grad_norm': '0.4039', 'learning_rate': '0.0001999', 'epoch': '0.5409'}
+{'loss': '0.8526', 'grad_norm': '0.4425', 'learning_rate': '0.0001999', 'epoch': '0.5418'}
+{'loss': '0.8695', 'grad_norm': '0.4545', 'learning_rate': '0.0001999', 'epoch': '0.5426'}
+{'loss': '0.8212', 'grad_norm': '0.4287', 'learning_rate': '0.0001999', 'epoch': '0.5435'}
+{'loss': '0.7968', 'grad_norm': '0.4155', 'learning_rate': '0.0001999', 'epoch': '0.5443'}
+{'loss': '0.8292', 'grad_norm': '0.4289', 'learning_rate': '0.0001999', 'epoch': '0.5451'}
+{'loss': '0.8788', 'grad_norm': '0.4499', 'learning_rate': '0.0001999', 'epoch': '0.546'}
+{'loss': '0.8376', 'grad_norm': '0.4333', 'learning_rate': '0.0001999', 'epoch': '0.5468'}
+{'loss': '0.8538', 'grad_norm': '0.4066', 'learning_rate': '0.0001999', 'epoch': '0.5477'}
+{'loss': '0.9722', 'grad_norm': '0.4609', 'learning_rate': '0.0001999', 'epoch': '0.5485'}
+{'eval_loss': '0.9069', 'eval_runtime': '680.8', 'eval_samples_per_second': '3.095', 'eval_steps_per_second': '3.095', 'epoch': '0.5485'}
+{'loss': '0.8474', 'grad_norm': '0.4493', 'learning_rate': '0.0001999', 'epoch': '0.5494'}
+{'loss': '0.8441', 'grad_norm': '0.5185', 'learning_rate': '0.0001999', 'epoch': '0.5502'}
+{'loss': '0.8848', 'grad_norm': '0.4381', 'learning_rate': '0.0001999', 'epoch': '0.5511'}
+{'loss': '0.8265', 'grad_norm': '0.4862', 'learning_rate': '0.0001999', 'epoch': '0.5519'}
+{'loss': '0.9328', 'grad_norm': '0.5069', 'learning_rate': '0.0001999', 'epoch': '0.5527'}
+{'loss': '0.9315', 'grad_norm': '0.4903', 'learning_rate': '0.0001999', 'epoch': '0.5536'}
+{'loss': '0.9345', 'grad_norm': '0.4531', 'learning_rate': '0.0001999', 'epoch': '0.5544'}
+{'loss': '0.8326', 'grad_norm': '0.4107', 'learning_rate': '0.0001999', 'epoch': '0.5553'}
+{'loss': '0.8935', 'grad_norm': '0.444', 'learning_rate': '0.0001999', 'epoch': '0.5561'}
+{'loss': '0.6691', 'grad_norm': '0.4025', 'learning_rate': '0.0001999', 'epoch': '0.557'}
+{'loss': '0.856', 'grad_norm': '0.439', 'learning_rate': '0.0001999', 'epoch': '0.5578'}
+{'loss': '0.9024', 'grad_norm': '0.5132', 'learning_rate': '0.0001999', 'epoch': '0.5586'}
+{'loss': '0.8675', 'grad_norm': '0.454', 'learning_rate': '0.0001999', 'epoch': '0.5595'}
+{'loss': '0.8853', 'grad_norm': '0.4447', 'learning_rate': '0.0001999', 'epoch': '0.5603'}
+{'loss': '0.9675', 'grad_norm': '0.4486', 'learning_rate': '0.0001999', 'epoch': '0.5612'}
+{'loss': '0.8257', 'grad_norm': '0.4332', 'learning_rate': '0.0001999', 'epoch': '0.562'}
+{'loss': '0.7728', 'grad_norm': '0.4316', 'learning_rate': '0.0001999', 'epoch': '0.5629'}
+{'loss': '0.9241', 'grad_norm': '0.4515', 'learning_rate': '0.0001999', 'epoch': '0.5637'}
+{'loss': '0.8006', 'grad_norm': '0.4607', 'learning_rate': '0.0001999', 'epoch': '0.5646'}
+{'loss': '0.9043', 'grad_norm': '0.4893', 'learning_rate': '0.0001999', 'epoch': '0.5654'}
+{'loss': '0.8725', 'grad_norm': '0.4602', 'learning_rate': '0.0001999', 'epoch': '0.5662'}
+{'loss': '0.8653', 'grad_norm': '0.4415', 'learning_rate': '0.0001999', 'epoch': '0.5671'}
+{'loss': '0.8844', 'grad_norm': '0.449', 'learning_rate': '0.0001999', 'epoch': '0.5679'}
+{'loss': '0.8385', 'grad_norm': '0.4769', 'learning_rate': '0.0001999', 'epoch': '0.5688'}
+{'loss': '0.7749', 'grad_norm': '0.4141', 'learning_rate': '0.0001999', 'epoch': '0.5696'}
+{'loss': '0.8369', 'grad_norm': '0.4152', 'learning_rate': '0.0001999', 'epoch': '0.5705'}
+{'loss': '0.8332', 'grad_norm': '0.4105', 'learning_rate': '0.0001999', 'epoch': '0.5713'}
+{'loss': '0.8042', 'grad_norm': '0.447', 'learning_rate': '0.0001999', 'epoch': '0.5722'}
+{'loss': '0.9053', 'grad_norm': '0.4622', 'learning_rate': '0.0001999', 'epoch': '0.573'}
+{'loss': '0.8036', 'grad_norm': '0.4935', 'learning_rate': '0.0001999', 'epoch': '0.5738'}
+{'loss': '0.834', 'grad_norm': '0.4557', 'learning_rate': '0.0001999', 'epoch': '0.5747'}
+{'loss': '0.8704', 'grad_norm': '0.5048', 'learning_rate': '0.0001999', 'epoch': '0.5755'}
+{'loss': '0.9116', 'grad_norm': '0.4708', 'learning_rate': '0.0001999', 'epoch': '0.5764'}
+{'loss': '0.8603', 'grad_norm': '0.4502', 'learning_rate': '0.0001998', 'epoch': '0.5772'}
+{'loss': '0.7482', 'grad_norm': '0.4229', 'learning_rate': '0.0001998', 'epoch': '0.5781'}
+{'loss': '0.9071', 'grad_norm': '0.4921', 'learning_rate': '0.0001998', 'epoch': '0.5789'}
+{'loss': '0.9371', 'grad_norm': '0.4739', 'learning_rate': '0.0001998', 'epoch': '0.5797'}
+{'loss': '0.8536', 'grad_norm': '0.4483', 'learning_rate': '0.0001998', 'epoch': '0.5806'}
+{'loss': '0.736', 'grad_norm': '0.4164', 'learning_rate': '0.0001998', 'epoch': '0.5814'}
+{'loss': '0.8163', 'grad_norm': '0.4415', 'learning_rate': '0.0001998', 'epoch': '0.5823'}
+{'loss': '0.9169', 'grad_norm': '0.4444', 'learning_rate': '0.0001998', 'epoch': '0.5831'}
+{'loss': '0.8876', 'grad_norm': '0.4642', 'learning_rate': '0.0001998', 'epoch': '0.584'}
+{'loss': '0.7747', 'grad_norm': '0.4328', 'learning_rate': '0.0001998', 'epoch': '0.5848'}
+{'loss': '0.849', 'grad_norm': '0.442', 'learning_rate': '0.0001998', 'epoch': '0.5857'}
+{'loss': '0.8015', 'grad_norm': '0.3908', 'learning_rate': '0.0001998', 'epoch': '0.5865'}
+{'loss': '0.8651', 'grad_norm': '0.3691', 'learning_rate': '0.0001998', 'epoch': '0.5873'}
+{'loss': '0.7713', 'grad_norm': '0.48', 'learning_rate': '0.0001998', 'epoch': '0.5882'}
+{'loss': '0.9054', 'grad_norm': '0.4397', 'learning_rate': '0.0001998', 'epoch': '0.589'}
+{'loss': '0.8738', 'grad_norm': '0.4264', 'learning_rate': '0.0001998', 'epoch': '0.5899'}
+{'loss': '0.9397', 'grad_norm': '0.4469', 'learning_rate': '0.0001998', 'epoch': '0.5907'}
+{'eval_loss': '0.8972', 'eval_runtime': '692.8', 'eval_samples_per_second': '3.041', 'eval_steps_per_second': '3.041', 'epoch': '0.5907'}
+{'loss': '0.9261', 'grad_norm': '0.4128', 'learning_rate': '0.0001998', 'epoch': '0.5916'}
+{'loss': '0.8264', 'grad_norm': '0.4552', 'learning_rate': '0.0001998', 'epoch': '0.5924'}
+{'loss': '0.8148', 'grad_norm': '0.4252', 'learning_rate': '0.0001998', 'epoch': '0.5932'}
+{'loss': '0.7712', 'grad_norm': '0.4173', 'learning_rate': '0.0001998', 'epoch': '0.5941'}
+{'loss': '0.7876', 'grad_norm': '0.4619', 'learning_rate': '0.0001998', 'epoch': '0.5949'}
+{'loss': '0.859', 'grad_norm': '0.4456', 'learning_rate': '0.0001998', 'epoch': '0.5958'}
+{'loss': '0.9318', 'grad_norm': '0.4267', 'learning_rate': '0.0001998', 'epoch': '0.5966'}
+{'loss': '0.8644', 'grad_norm': '0.4263', 'learning_rate': '0.0001998', 'epoch': '0.5975'}
+{'loss': '0.776', 'grad_norm': '0.3939', 'learning_rate': '0.0001998', 'epoch': '0.5983'}
+{'loss': '0.8072', 'grad_norm': '0.4433', 'learning_rate': '0.0001998', 'epoch': '0.5992'}
+{'loss': '0.9077', 'grad_norm': '0.3939', 'learning_rate': '0.0001998', 'epoch': '0.6'}
+{'loss': '0.8534', 'grad_norm': '0.4675', 'learning_rate': '0.0001998', 'epoch': '0.6008'}
+{'loss': '0.905', 'grad_norm': '0.4841', 'learning_rate': '0.0001998', 'epoch': '0.6017'}
+{'loss': '0.8521', 'grad_norm': '0.4741', 'learning_rate': '0.0001998', 'epoch': '0.6025'}
+{'loss': '0.8221', 'grad_norm': '0.465', 'learning_rate': '0.0001998', 'epoch': '0.6034'}
+{'loss': '0.8132', 'grad_norm': '0.4089', 'learning_rate': '0.0001998', 'epoch': '0.6042'}
+{'loss': '0.8705', 'grad_norm': '0.4376', 'learning_rate': '0.0001997', 'epoch': '0.6051'}
+{'loss': '0.8826', 'grad_norm': '0.4781', 'learning_rate': '0.0001997', 'epoch': '0.6059'}
+{'loss': '0.8188', 'grad_norm': '0.4624', 'learning_rate': '0.0001997', 'epoch': '0.6068'}
+{'loss': '0.857', 'grad_norm': '0.4217', 'learning_rate': '0.0001997', 'epoch': '0.6076'}
+{'loss': '0.9174', 'grad_norm': '0.4939', 'learning_rate': '0.0001997', 'epoch': '0.6084'}
+{'loss': '0.9445', 'grad_norm': '0.4712', 'learning_rate': '0.0001997', 'epoch': '0.6093'}
+{'loss': '0.9537', 'grad_norm': '0.4832', 'learning_rate': '0.0001997', 'epoch': '0.6101'}
+{'loss': '0.8481', 'grad_norm': '0.4446', 'learning_rate': '0.0001997', 'epoch': '0.611'}
+{'loss': '0.8572', 'grad_norm': '0.4315', 'learning_rate': '0.0001997', 'epoch': '0.6118'}
+{'loss': '0.7626', 'grad_norm': '0.3856', 'learning_rate': '0.0001997', 'epoch': '0.6127'}
+{'loss': '0.788', 'grad_norm': '0.382', 'learning_rate': '0.0001997', 'epoch': '0.6135'}
+{'loss': '0.8697', 'grad_norm': '0.4251', 'learning_rate': '0.0001997', 'epoch': '0.6143'}
+{'loss': '0.8426', 'grad_norm': '0.4675', 'learning_rate': '0.0001997', 'epoch': '0.6152'}
+{'loss': '0.8029', 'grad_norm': '0.3937', 'learning_rate': '0.0001997', 'epoch': '0.616'}
+{'loss': '0.8663', 'grad_norm': '0.4361', 'learning_rate': '0.0001997', 'epoch': '0.6169'}
+{'loss': '0.8963', 'grad_norm': '0.4738', 'learning_rate': '0.0001997', 'epoch': '0.6177'}
+{'loss': '0.8631', 'grad_norm': '0.3953', 'learning_rate': '0.0001997', 'epoch': '0.6186'}
+{'loss': '0.8759', 'grad_norm': '0.4169', 'learning_rate': '0.0001997', 'epoch': '0.6194'}
+{'loss': '0.8459', 'grad_norm': '0.4599', 'learning_rate': '0.0001997', 'epoch': '0.6203'}
+{'loss': '0.8845', 'grad_norm': '0.5477', 'learning_rate': '0.0001997', 'epoch': '0.6211'}
+{'loss': '0.8609', 'grad_norm': '0.462', 'learning_rate': '0.0001997', 'epoch': '0.6219'}
+{'loss': '0.7588', 'grad_norm': '0.3945', 'learning_rate': '0.0001997', 'epoch': '0.6228'}
+{'loss': '0.9143', 'grad_norm': '0.45', 'learning_rate': '0.0001997', 'epoch': '0.6236'}
+{'loss': '0.7866', 'grad_norm': '0.4204', 'learning_rate': '0.0001997', 'epoch': '0.6245'}
+{'loss': '0.9193', 'grad_norm': '0.4562', 'learning_rate': '0.0001997', 'epoch': '0.6253'}
+{'loss': '0.8459', 'grad_norm': '0.414', 'learning_rate': '0.0001997', 'epoch': '0.6262'}
+{'loss': '0.8159', 'grad_norm': '0.4411', 'learning_rate': '0.0001997', 'epoch': '0.627'}
+{'loss': '0.8845', 'grad_norm': '0.4382', 'learning_rate': '0.0001996', 'epoch': '0.6278'}
+{'loss': '0.8694', 'grad_norm': '0.4692', 'learning_rate': '0.0001996', 'epoch': '0.6287'}
+{'loss': '0.8898', 'grad_norm': '0.4072', 'learning_rate': '0.0001996', 'epoch': '0.6295'}
+{'loss': '0.8275', 'grad_norm': '0.4123', 'learning_rate': '0.0001996', 'epoch': '0.6304'}
+{'loss': '0.8159', 'grad_norm': '0.4435', 'learning_rate': '0.0001996', 'epoch': '0.6312'}
+{'loss': '0.844', 'grad_norm': '0.47', 'learning_rate': '0.0001996', 'epoch': '0.6321'}
+{'loss': '0.8317', 'grad_norm': '0.4642', 'learning_rate': '0.0001996', 'epoch': '0.6329'}
+{'eval_loss': '0.8875', 'eval_runtime': '686.3', 'eval_samples_per_second': '3.07', 'eval_steps_per_second': '3.07', 'epoch': '0.6329'}
+{'loss': '0.9122', 'grad_norm': '0.4418', 'learning_rate': '0.0001996', 'epoch': '0.6338'}
+{'loss': '0.916', 'grad_norm': '0.4279', 'learning_rate': '0.0001996', 'epoch': '0.6346'}
+{'loss': '0.8617', 'grad_norm': '0.3967', 'learning_rate': '0.0001996', 'epoch': '0.6354'}
+{'loss': '0.8478', 'grad_norm': '0.4796', 'learning_rate': '0.0001996', 'epoch': '0.6363'}
+{'loss': '0.8739', 'grad_norm': '0.3861', 'learning_rate': '0.0001996', 'epoch': '0.6371'}
+{'loss': '0.7694', 'grad_norm': '0.4568', 'learning_rate': '0.0001996', 'epoch': '0.638'}
+{'loss': '0.8482', 'grad_norm': '0.4457', 'learning_rate': '0.0001996', 'epoch': '0.6388'}
+{'loss': '0.8627', 'grad_norm': '0.4392', 'learning_rate': '0.0001996', 'epoch': '0.6397'}
+{'loss': '0.8342', 'grad_norm': '0.4096', 'learning_rate': '0.0001996', 'epoch': '0.6405'}
+{'loss': '0.8794', 'grad_norm': '0.4181', 'learning_rate': '0.0001996', 'epoch': '0.6414'}
+{'loss': '0.7788', 'grad_norm': '0.4751', 'learning_rate': '0.0001996', 'epoch': '0.6422'}
+{'loss': '0.9084', 'grad_norm': '0.5129', 'learning_rate': '0.0001996', 'epoch': '0.643'}
+{'loss': '0.9067', 'grad_norm': '0.4829', 'learning_rate': '0.0001996', 'epoch': '0.6439'}
+{'loss': '0.8128', 'grad_norm': '0.4225', 'learning_rate': '0.0001996', 'epoch': '0.6447'}
+{'loss': '0.8782', 'grad_norm': '0.4232', 'learning_rate': '0.0001996', 'epoch': '0.6456'}
+{'loss': '0.8137', 'grad_norm': '0.4436', 'learning_rate': '0.0001996', 'epoch': '0.6464'}
+{'loss': '0.8004', 'grad_norm': '0.4018', 'learning_rate': '0.0001996', 'epoch': '0.6473'}
+{'loss': '0.8996', 'grad_norm': '0.4002', 'learning_rate': '0.0001995', 'epoch': '0.6481'}
+{'loss': '0.8611', 'grad_norm': '0.4285', 'learning_rate': '0.0001995', 'epoch': '0.6489'}
+{'loss': '0.8992', 'grad_norm': '0.4258', 'learning_rate': '0.0001995', 'epoch': '0.6498'}
+{'loss': '0.815', 'grad_norm': '0.4483', 'learning_rate': '0.0001995', 'epoch': '0.6506'}
+{'loss': '0.8573', 'grad_norm': '0.4321', 'learning_rate': '0.0001995', 'epoch': '0.6515'}
+{'loss': '0.7766', 'grad_norm': '0.3995', 'learning_rate': '0.0001995', 'epoch': '0.6523'}
+{'loss': '0.9032', 'grad_norm': '0.413', 'learning_rate': '0.0001995', 'epoch': '0.6532'}
+{'loss': '0.8361', 'grad_norm': '0.4437', 'learning_rate': '0.0001995', 'epoch': '0.654'}
+{'loss': '0.8662', 'grad_norm': '0.4229', 'learning_rate': '0.0001995', 'epoch': '0.6549'}
+{'loss': '0.9738', 'grad_norm': '0.484', 'learning_rate': '0.0001995', 'epoch': '0.6557'}
+{'loss': '0.7542', 'grad_norm': '0.3859', 'learning_rate': '0.0001995', 'epoch': '0.6565'}
+{'loss': '0.9226', 'grad_norm': '0.3964', 'learning_rate': '0.0001995', 'epoch': '0.6574'}
+{'loss': '0.8376', 'grad_norm': '0.4261', 'learning_rate': '0.0001995', 'epoch': '0.6582'}
+{'loss': '0.7847', 'grad_norm': '0.4459', 'learning_rate': '0.0001995', 'epoch': '0.6591'}
+{'loss': '0.8512', 'grad_norm': '0.468', 'learning_rate': '0.0001995', 'epoch': '0.6599'}
+{'loss': '0.826', 'grad_norm': '0.6826', 'learning_rate': '0.0001995', 'epoch': '0.6608'}
+{'loss': '0.879', 'grad_norm': '0.4737', 'learning_rate': '0.0001995', 'epoch': '0.6616'}
+{'loss': '0.9296', 'grad_norm': '0.4111', 'learning_rate': '0.0001995', 'epoch': '0.6624'}
+{'loss': '0.8738', 'grad_norm': '0.4298', 'learning_rate': '0.0001995', 'epoch': '0.6633'}
+{'loss': '0.9372', 'grad_norm': '0.4812', 'learning_rate': '0.0001995', 'epoch': '0.6641'}
+{'loss': '0.8399', 'grad_norm': '0.4048', 'learning_rate': '0.0001994', 'epoch': '0.665'}
+{'loss': '0.827', 'grad_norm': '0.4024', 'learning_rate': '0.0001994', 'epoch': '0.6658'}
+{'loss': '0.8238', 'grad_norm': '0.3784', 'learning_rate': '0.0001994', 'epoch': '0.6667'}
+{'loss': '0.8469', 'grad_norm': '0.4449', 'learning_rate': '0.0001994', 'epoch': '0.6675'}
+{'loss': '0.8631', 'grad_norm': '0.4691', 'learning_rate': '0.0001994', 'epoch': '0.6684'}
+{'loss': '0.8904', 'grad_norm': '0.4367', 'learning_rate': '0.0001994', 'epoch': '0.6692'}
+{'loss': '0.8356', 'grad_norm': '0.4237', 'learning_rate': '0.0001994', 'epoch': '0.67'}
+{'loss': '0.8898', 'grad_norm': '0.4862', 'learning_rate': '0.0001994', 'epoch': '0.6709'}
+{'loss': '0.9', 'grad_norm': '0.4454', 'learning_rate': '0.0001994', 'epoch': '0.6717'}
+{'loss': '0.8171', 'grad_norm': '0.4417', 'learning_rate': '0.0001994', 'epoch': '0.6726'}
+{'loss': '0.8704', 'grad_norm': '0.4506', 'learning_rate': '0.0001994', 'epoch': '0.6734'}
+{'loss': '0.8677', 'grad_norm': '0.4232', 'learning_rate': '0.0001994', 'epoch': '0.6743'}
+{'loss': '0.7314', 'grad_norm': '0.3936', 'learning_rate': '0.0001994', 'epoch': '0.6751'}
+{'eval_loss': '0.877', 'eval_runtime': '677.9', 'eval_samples_per_second': '3.108', 'eval_steps_per_second': '3.108', 'epoch': '0.6751'}
+{'loss': '0.8075', 'grad_norm': '0.4197', 'learning_rate': '0.0001994', 'epoch': '0.6759'}
+{'loss': '0.7886', 'grad_norm': '0.4523', 'learning_rate': '0.0001994', 'epoch': '0.6768'}
+{'loss': '0.8683', 'grad_norm': '0.4566', 'learning_rate': '0.0001994', 'epoch': '0.6776'}
+{'loss': '0.8307', 'grad_norm': '0.3955', 'learning_rate': '0.0001994', 'epoch': '0.6785'}
+{'loss': '0.7505', 'grad_norm': '0.3651', 'learning_rate': '0.0001994', 'epoch': '0.6793'}
+{'loss': '0.8793', 'grad_norm': '0.4241', 'learning_rate': '0.0001994', 'epoch': '0.6802'}
+{'loss': '0.7328', 'grad_norm': '0.4143', 'learning_rate': '0.0001994', 'epoch': '0.681'}
+{'loss': '0.8415', 'grad_norm': '0.4834', 'learning_rate': '0.0001993', 'epoch': '0.6819'}
+{'loss': '0.8589', 'grad_norm': '0.4312', 'learning_rate': '0.0001993', 'epoch': '0.6827'}
+{'loss': '0.8465', 'grad_norm': '0.4385', 'learning_rate': '0.0001993', 'epoch': '0.6835'}
+{'loss': '0.9352', 'grad_norm': '0.4577', 'learning_rate': '0.0001993', 'epoch': '0.6844'}
+{'loss': '0.8885', 'grad_norm': '0.4384', 'learning_rate': '0.0001993', 'epoch': '0.6852'}
+{'loss': '0.774', 'grad_norm': '0.4589', 'learning_rate': '0.0001993', 'epoch': '0.6861'}
+{'loss': '0.7948', 'grad_norm': '0.3863', 'learning_rate': '0.0001993', 'epoch': '0.6869'}
+{'loss': '0.7505', 'grad_norm': '0.3824', 'learning_rate': '0.0001993', 'epoch': '0.6878'}
+{'loss': '0.9007', 'grad_norm': '0.3981', 'learning_rate': '0.0001993', 'epoch': '0.6886'}
+{'loss': '0.6997', 'grad_norm': '0.4142', 'learning_rate': '0.0001993', 'epoch': '0.6895'}
+{'loss': '0.8523', 'grad_norm': '0.4378', 'learning_rate': '0.0001993', 'epoch': '0.6903'}
+{'loss': '0.8886', 'grad_norm': '0.4378', 'learning_rate': '0.0001993', 'epoch': '0.6911'}
+{'loss': '0.8564', 'grad_norm': '0.4276', 'learning_rate': '0.0001993', 'epoch': '0.692'}
+{'loss': '0.8541', 'grad_norm': '0.46', 'learning_rate': '0.0001993', 'epoch': '0.6928'}
+{'loss': '0.8558', 'grad_norm': '0.5058', 'learning_rate': '0.0001993', 'epoch': '0.6937'}
+{'loss': '0.8623', 'grad_norm': '0.4947', 'learning_rate': '0.0001993', 'epoch': '0.6945'}
+{'loss': '0.7837', 'grad_norm': '0.4256', 'learning_rate': '0.0001993', 'epoch': '0.6954'}
+{'loss': '0.7878', 'grad_norm': '0.4756', 'learning_rate': '0.0001992', 'epoch': '0.6962'}
+{'loss': '0.7757', 'grad_norm': '0.3917', 'learning_rate': '0.0001992', 'epoch': '0.697'}
+{'loss': '0.881', 'grad_norm': '0.4637', 'learning_rate': '0.0001992', 'epoch': '0.6979'}
+{'loss': '0.8513', 'grad_norm': '0.441', 'learning_rate': '0.0001992', 'epoch': '0.6987'}
+{'loss': '0.831', 'grad_norm': '0.4782', 'learning_rate': '0.0001992', 'epoch': '0.6996'}
+{'loss': '0.8818', 'grad_norm': '0.4193', 'learning_rate': '0.0001992', 'epoch': '0.7004'}
+{'loss': '0.8316', 'grad_norm': '0.4529', 'learning_rate': '0.0001992', 'epoch': '0.7013'}
+{'loss': '0.8698', 'grad_norm': '0.4712', 'learning_rate': '0.0001992', 'epoch': '0.7021'}
+{'loss': '0.9536', 'grad_norm': '0.4769', 'learning_rate': '0.0001992', 'epoch': '0.703'}
+{'loss': '0.8537', 'grad_norm': '0.394', 'learning_rate': '0.0001992', 'epoch': '0.7038'}
+{'loss': '0.816', 'grad_norm': '0.366', 'learning_rate': '0.0001992', 'epoch': '0.7046'}
+{'loss': '0.7825', 'grad_norm': '0.4382', 'learning_rate': '0.0001992', 'epoch': '0.7055'}
+{'loss': '0.9314', 'grad_norm': '0.4475', 'learning_rate': '0.0001992', 'epoch': '0.7063'}
+{'loss': '0.8608', 'grad_norm': '0.4515', 'learning_rate': '0.0001992', 'epoch': '0.7072'}
+{'loss': '0.8008', 'grad_norm': '0.4274', 'learning_rate': '0.0001992', 'epoch': '0.708'}
+{'loss': '0.7797', 'grad_norm': '0.4183', 'learning_rate': '0.0001992', 'epoch': '0.7089'}
+{'loss': '0.9007', 'grad_norm': '0.4435', 'learning_rate': '0.0001992', 'epoch': '0.7097'}
+{'loss': '0.775', 'grad_norm': '0.3933', 'learning_rate': '0.0001991', 'epoch': '0.7105'}
+{'loss': '0.7508', 'grad_norm': '0.4811', 'learning_rate': '0.0001991', 'epoch': '0.7114'}
+{'loss': '0.7927', 'grad_norm': '0.4167', 'learning_rate': '0.0001991', 'epoch': '0.7122'}
+{'loss': '0.8364', 'grad_norm': '0.479', 'learning_rate': '0.0001991', 'epoch': '0.7131'}
+{'loss': '0.865', 'grad_norm': '0.4351', 'learning_rate': '0.0001991', 'epoch': '0.7139'}
+{'loss': '0.8196', 'grad_norm': '0.407', 'learning_rate': '0.0001991', 'epoch': '0.7148'}
+{'loss': '0.9157', 'grad_norm': '0.4718', 'learning_rate': '0.0001991', 'epoch': '0.7156'}
+{'loss': '0.788', 'grad_norm': '0.3943', 'learning_rate': '0.0001991', 'epoch': '0.7165'}
+{'loss': '0.8089', 'grad_norm': '0.4038', 'learning_rate': '0.0001991', 'epoch': '0.7173'}
+{'eval_loss': '0.8708', 'eval_runtime': '670.3', 'eval_samples_per_second': '3.143', 'eval_steps_per_second': '3.143', 'epoch': '0.7173'}
+{'loss': '0.874', 'grad_norm': '0.4444', 'learning_rate': '0.0001991', 'epoch': '0.7181'}
+{'loss': '0.8305', 'grad_norm': '0.4509', 'learning_rate': '0.0001991', 'epoch': '0.719'}
+{'loss': '0.7718', 'grad_norm': '0.3903', 'learning_rate': '0.0001991', 'epoch': '0.7198'}
+{'loss': '0.8433', 'grad_norm': '0.4299', 'learning_rate': '0.0001991', 'epoch': '0.7207'}
+{'loss': '0.8332', 'grad_norm': '0.4288', 'learning_rate': '0.0001991', 'epoch': '0.7215'}
+{'loss': '0.7842', 'grad_norm': '0.4557', 'learning_rate': '0.0001991', 'epoch': '0.7224'}
+{'loss': '0.7808', 'grad_norm': '0.4002', 'learning_rate': '0.000199', 'epoch': '0.7232'}
+{'loss': '0.7936', 'grad_norm': '0.5209', 'learning_rate': '0.000199', 'epoch': '0.7241'}
+{'loss': '0.7965', 'grad_norm': '0.4647', 'learning_rate': '0.000199', 'epoch': '0.7249'}
+{'loss': '0.8302', 'grad_norm': '0.4058', 'learning_rate': '0.000199', 'epoch': '0.7257'}
+{'loss': '0.8487', 'grad_norm': '0.3875', 'learning_rate': '0.000199', 'epoch': '0.7266'}
+{'loss': '0.8378', 'grad_norm': '0.4687', 'learning_rate': '0.000199', 'epoch': '0.7274'}
+{'loss': '0.7641', 'grad_norm': '0.4139', 'learning_rate': '0.000199', 'epoch': '0.7283'}
+{'loss': '0.8236', 'grad_norm': '0.5003', 'learning_rate': '0.000199', 'epoch': '0.7291'}
+{'loss': '0.7793', 'grad_norm': '0.385', 'learning_rate': '0.000199', 'epoch': '0.73'}
+{'loss': '0.7734', 'grad_norm': '0.4463', 'learning_rate': '0.000199', 'epoch': '0.7308'}
+{'loss': '0.8933', 'grad_norm': '0.3982', 'learning_rate': '0.000199', 'epoch': '0.7316'}
+{'loss': '0.8198', 'grad_norm': '0.4231', 'learning_rate': '0.000199', 'epoch': '0.7325'}
+{'loss': '0.8266', 'grad_norm': '0.4364', 'learning_rate': '0.000199', 'epoch': '0.7333'}
+{'loss': '0.784', 'grad_norm': '0.4236', 'learning_rate': '0.000199', 'epoch': '0.7342'}
+{'loss': '0.7154', 'grad_norm': '0.4033', 'learning_rate': '0.000199', 'epoch': '0.735'}
+{'loss': '0.9079', 'grad_norm': '0.4428', 'learning_rate': '0.0001989', 'epoch': '0.7359'}
+{'loss': '0.8645', 'grad_norm': '0.4883', 'learning_rate': '0.0001989', 'epoch': '0.7367'}
+{'loss': '0.8134', 'grad_norm': '0.3486', 'learning_rate': '0.0001989', 'epoch': '0.7376'}
+{'loss': '0.8143', 'grad_norm': '0.4273', 'learning_rate': '0.0001989', 'epoch': '0.7384'}
+{'loss': '0.7958', 'grad_norm': '0.429', 'learning_rate': '0.0001989', 'epoch': '0.7392'}
+{'loss': '0.9', 'grad_norm': '0.4812', 'learning_rate': '0.0001989', 'epoch': '0.7401'}
+{'loss': '0.772', 'grad_norm': '0.4078', 'learning_rate': '0.0001989', 'epoch': '0.7409'}
+{'loss': '0.7682', 'grad_norm': '0.4017', 'learning_rate': '0.0001989', 'epoch': '0.7418'}
+{'loss': '0.7537', 'grad_norm': '0.3714', 'learning_rate': '0.0001989', 'epoch': '0.7426'}
+{'loss': '0.9', 'grad_norm': '0.4184', 'learning_rate': '0.0001989', 'epoch': '0.7435'}
+{'loss': '0.8047', 'grad_norm': '0.4242', 'learning_rate': '0.0001989', 'epoch': '0.7443'}
+{'loss': '0.8933', 'grad_norm': '0.4249', 'learning_rate': '0.0001989', 'epoch': '0.7451'}
+{'loss': '0.7685', 'grad_norm': '0.3582', 'learning_rate': '0.0001989', 'epoch': '0.746'}
+{'loss': '0.8644', 'grad_norm': '0.4104', 'learning_rate': '0.0001989', 'epoch': '0.7468'}
+{'loss': '0.8036', 'grad_norm': '0.545', 'learning_rate': '0.0001988', 'epoch': '0.7477'}
+{'loss': '0.8861', 'grad_norm': '0.5281', 'learning_rate': '0.0001988', 'epoch': '0.7485'}
+{'loss': '0.7507', 'grad_norm': '0.4267', 'learning_rate': '0.0001988', 'epoch': '0.7494'}
+{'loss': '0.7662', 'grad_norm': '0.3903', 'learning_rate': '0.0001988', 'epoch': '0.7502'}
+{'loss': '0.759', 'grad_norm': '0.4387', 'learning_rate': '0.0001988', 'epoch': '0.7511'}
+{'loss': '0.934', 'grad_norm': '0.4195', 'learning_rate': '0.0001988', 'epoch': '0.7519'}
+{'loss': '0.7969', 'grad_norm': '0.4236', 'learning_rate': '0.0001988', 'epoch': '0.7527'}
+{'loss': '0.7892', 'grad_norm': '0.4274', 'learning_rate': '0.0001988', 'epoch': '0.7536'}
+{'loss': '0.8753', 'grad_norm': '0.4623', 'learning_rate': '0.0001988', 'epoch': '0.7544'}
+{'loss': '0.813', 'grad_norm': '0.3649', 'learning_rate': '0.0001988', 'epoch': '0.7553'}
+{'loss': '0.731', 'grad_norm': '0.4136', 'learning_rate': '0.0001988', 'epoch': '0.7561'}
+{'loss': '0.8584', 'grad_norm': '0.4113', 'learning_rate': '0.0001988', 'epoch': '0.757'}
+{'loss': '0.8377', 'grad_norm': '0.4295', 'learning_rate': '0.0001988', 'epoch': '0.7578'}
+{'loss': '0.9139', 'grad_norm': '0.4228', 'learning_rate': '0.0001988', 'epoch': '0.7586'}
+{'loss': '0.6995', 'grad_norm': '0.3638', 'learning_rate': '0.0001987', 'epoch': '0.7595'}
+{'eval_loss': '0.8626', 'eval_runtime': '686.4', 'eval_samples_per_second': '3.07', 'eval_steps_per_second': '3.07', 'epoch': '0.7595'}
+{'loss': '0.8681', 'grad_norm': '0.4695', 'learning_rate': '0.0001987', 'epoch': '0.7603'}
+{'loss': '0.7421', 'grad_norm': '0.4196', 'learning_rate': '0.0001987', 'epoch': '0.7612'}
+{'loss': '0.8019', 'grad_norm': '0.4312', 'learning_rate': '0.0001987', 'epoch': '0.762'}
+{'loss': '0.8794', 'grad_norm': '0.4513', 'learning_rate': '0.0001987', 'epoch': '0.7629'}
+{'loss': '0.8417', 'grad_norm': '0.4428', 'learning_rate': '0.0001987', 'epoch': '0.7637'}
+{'loss': '0.8226', 'grad_norm': '0.4689', 'learning_rate': '0.0001987', 'epoch': '0.7646'}
+{'loss': '0.7872', 'grad_norm': '0.4413', 'learning_rate': '0.0001987', 'epoch': '0.7654'}
+{'loss': '0.8616', 'grad_norm': '0.484', 'learning_rate': '0.0001987', 'epoch': '0.7662'}
+{'loss': '0.7243', 'grad_norm': '0.3703', 'learning_rate': '0.0001987', 'epoch': '0.7671'}
+{'loss': '0.8624', 'grad_norm': '0.4522', 'learning_rate': '0.0001987', 'epoch': '0.7679'}
+{'loss': '0.7766', 'grad_norm': '0.4382', 'learning_rate': '0.0001987', 'epoch': '0.7688'}
+{'loss': '0.8723', 'grad_norm': '0.3826', 'learning_rate': '0.0001987', 'epoch': '0.7696'}
+{'loss': '0.9078', 'grad_norm': '0.4368', 'learning_rate': '0.0001986', 'epoch': '0.7705'}
+{'loss': '0.8506', 'grad_norm': '0.4129', 'learning_rate': '0.0001986', 'epoch': '0.7713'}
+{'loss': '0.8327', 'grad_norm': '0.4254', 'learning_rate': '0.0001986', 'epoch': '0.7722'}
+{'loss': '0.8172', 'grad_norm': '0.4137', 'learning_rate': '0.0001986', 'epoch': '0.773'}
+{'loss': '0.8611', 'grad_norm': '0.4388', 'learning_rate': '0.0001986', 'epoch': '0.7738'}
+{'loss': '0.7961', 'grad_norm': '0.399', 'learning_rate': '0.0001986', 'epoch': '0.7747'}
+{'loss': '0.8078', 'grad_norm': '0.408', 'learning_rate': '0.0001986', 'epoch': '0.7755'}
+{'loss': '0.8744', 'grad_norm': '0.4701', 'learning_rate': '0.0001986', 'epoch': '0.7764'}
+{'loss': '0.7682', 'grad_norm': '0.5391', 'learning_rate': '0.0001986', 'epoch': '0.7772'}
+{'loss': '0.7818', 'grad_norm': '0.3625', 'learning_rate': '0.0001986', 'epoch': '0.7781'}
+{'loss': '0.7615', 'grad_norm': '0.4124', 'learning_rate': '0.0001986', 'epoch': '0.7789'}
+{'loss': '0.7492', 'grad_norm': '0.3969', 'learning_rate': '0.0001986', 'epoch': '0.7797'}
+{'loss': '0.8616', 'grad_norm': '0.4415', 'learning_rate': '0.0001985', 'epoch': '0.7806'}
+{'loss': '0.8417', 'grad_norm': '0.4486', 'learning_rate': '0.0001985', 'epoch': '0.7814'}
+{'loss': '0.8988', 'grad_norm': '0.4515', 'learning_rate': '0.0001985', 'epoch': '0.7823'}
+{'loss': '0.7598', 'grad_norm': '0.3933', 'learning_rate': '0.0001985', 'epoch': '0.7831'}
+{'loss': '0.843', 'grad_norm': '0.426', 'learning_rate': '0.0001985', 'epoch': '0.784'}
+{'loss': '0.7444', 'grad_norm': '0.3752', 'learning_rate': '0.0001985', 'epoch': '0.7848'}
+{'loss': '0.9342', 'grad_norm': '0.4991', 'learning_rate': '0.0001985', 'epoch': '0.7857'}
+{'loss': '0.8939', 'grad_norm': '0.4711', 'learning_rate': '0.0001985', 'epoch': '0.7865'}
+{'loss': '0.8746', 'grad_norm': '0.4145', 'learning_rate': '0.0001985', 'epoch': '0.7873'}
+{'loss': '0.7179', 'grad_norm': '0.3444', 'learning_rate': '0.0001985', 'epoch': '0.7882'}
+{'loss': '0.9114', 'grad_norm': '0.3463', 'learning_rate': '0.0001985', 'epoch': '0.789'}
+{'loss': '0.8363', 'grad_norm': '0.3863', 'learning_rate': '0.0001985', 'epoch': '0.7899'}
+{'loss': '0.8248', 'grad_norm': '0.4095', 'learning_rate': '0.0001984', 'epoch': '0.7907'}
+{'loss': '0.8446', 'grad_norm': '0.4588', 'learning_rate': '0.0001984', 'epoch': '0.7916'}
+{'loss': '0.8568', 'grad_norm': '0.372', 'learning_rate': '0.0001984', 'epoch': '0.7924'}
+{'loss': '0.8134', 'grad_norm': '0.3839', 'learning_rate': '0.0001984', 'epoch': '0.7932'}
+{'loss': '0.867', 'grad_norm': '0.4019', 'learning_rate': '0.0001984', 'epoch': '0.7941'}
+{'loss': '0.7975', 'grad_norm': '0.3721', 'learning_rate': '0.0001984', 'epoch': '0.7949'}
+{'loss': '0.8756', 'grad_norm': '0.4493', 'learning_rate': '0.0001984', 'epoch': '0.7958'}
+{'loss': '0.868', 'grad_norm': '0.4511', 'learning_rate': '0.0001984', 'epoch': '0.7966'}
+{'loss': '0.7991', 'grad_norm': '0.4037', 'learning_rate': '0.0001984', 'epoch': '0.7975'}
+{'loss': '0.8599', 'grad_norm': '0.4613', 'learning_rate': '0.0001984', 'epoch': '0.7983'}
+{'loss': '0.7707', 'grad_norm': '0.4459', 'learning_rate': '0.0001984', 'epoch': '0.7992'}
+{'loss': '0.807', 'grad_norm': '0.445', 'learning_rate': '0.0001984', 'epoch': '0.8'}
+{'loss': '0.8623', 'grad_norm': '0.4008', 'learning_rate': '0.0001983', 'epoch': '0.8008'}
+{'loss': '0.8125', 'grad_norm': '0.3919', 'learning_rate': '0.0001983', 'epoch': '0.8017'}
+{'eval_loss': '0.8546', 'eval_runtime': '688.5', 'eval_samples_per_second': '3.06', 'eval_steps_per_second': '3.06', 'epoch': '0.8017'}
+{'loss': '0.8246', 'grad_norm': '0.4411', 'learning_rate': '0.0001983', 'epoch': '0.8025'}
+{'loss': '0.8388', 'grad_norm': '0.4089', 'learning_rate': '0.0001983', 'epoch': '0.8034'}
+{'loss': '0.7891', 'grad_norm': '0.4872', 'learning_rate': '0.0001983', 'epoch': '0.8042'}
+{'loss': '0.8009', 'grad_norm': '0.4009', 'learning_rate': '0.0001983', 'epoch': '0.8051'}
+{'loss': '0.7922', 'grad_norm': '0.3878', 'learning_rate': '0.0001983', 'epoch': '0.8059'}
+{'loss': '0.7753', 'grad_norm': '0.3783', 'learning_rate': '0.0001983', 'epoch': '0.8068'}
+{'loss': '0.7958', 'grad_norm': '0.4114', 'learning_rate': '0.0001983', 'epoch': '0.8076'}
+{'loss': '0.8177', 'grad_norm': '0.4872', 'learning_rate': '0.0001983', 'epoch': '0.8084'}
+{'loss': '0.7727', 'grad_norm': '0.4157', 'learning_rate': '0.0001983', 'epoch': '0.8093'}
+{'loss': '0.6601', 'grad_norm': '0.3234', 'learning_rate': '0.0001982', 'epoch': '0.8101'}
+{'loss': '0.7867', 'grad_norm': '0.4643', 'learning_rate': '0.0001982', 'epoch': '0.811'}
+{'loss': '0.7858', 'grad_norm': '0.4356', 'learning_rate': '0.0001982', 'epoch': '0.8118'}
+{'loss': '0.8408', 'grad_norm': '0.3997', 'learning_rate': '0.0001982', 'epoch': '0.8127'}
+{'loss': '0.8434', 'grad_norm': '0.4415', 'learning_rate': '0.0001982', 'epoch': '0.8135'}
+{'loss': '0.6941', 'grad_norm': '0.3733', 'learning_rate': '0.0001982', 'epoch': '0.8143'}
+{'loss': '0.7756', 'grad_norm': '0.4483', 'learning_rate': '0.0001982', 'epoch': '0.8152'}
+{'loss': '0.8496', 'grad_norm': '0.4262', 'learning_rate': '0.0001982', 'epoch': '0.816'}
+{'loss': '0.7768', 'grad_norm': '0.4238', 'learning_rate': '0.0001982', 'epoch': '0.8169'}
+{'loss': '0.7348', 'grad_norm': '0.415', 'learning_rate': '0.0001982', 'epoch': '0.8177'}
+{'loss': '0.784', 'grad_norm': '0.3927', 'learning_rate': '0.0001982', 'epoch': '0.8186'}
+{'loss': '0.8871', 'grad_norm': '0.4483', 'learning_rate': '0.0001981', 'epoch': '0.8194'}
+{'loss': '0.8647', 'grad_norm': '0.44', 'learning_rate': '0.0001981', 'epoch': '0.8203'}
+{'loss': '0.9577', 'grad_norm': '0.47', 'learning_rate': '0.0001981', 'epoch': '0.8211'}
+{'loss': '0.7956', 'grad_norm': '0.3881', 'learning_rate': '0.0001981', 'epoch': '0.8219'}
+{'loss': '0.844', 'grad_norm': '0.4271', 'learning_rate': '0.0001981', 'epoch': '0.8228'}
+{'loss': '0.8589', 'grad_norm': '0.4098', 'learning_rate': '0.0001981', 'epoch': '0.8236'}
+{'loss': '0.8607', 'grad_norm': '0.4155', 'learning_rate': '0.0001981', 'epoch': '0.8245'}
+{'loss': '0.7582', 'grad_norm': '0.3681', 'learning_rate': '0.0001981', 'epoch': '0.8253'}
+{'loss': '0.8345', 'grad_norm': '0.5733', 'learning_rate': '0.0001981', 'epoch': '0.8262'}
+{'loss': '0.8897', 'grad_norm': '0.4363', 'learning_rate': '0.0001981', 'epoch': '0.827'}
+{'loss': '0.8093', 'grad_norm': '0.4425', 'learning_rate': '0.0001981', 'epoch': '0.8278'}
+{'loss': '0.892', 'grad_norm': '0.4502', 'learning_rate': '0.000198', 'epoch': '0.8287'}
+{'loss': '0.7652', 'grad_norm': '0.4313', 'learning_rate': '0.000198', 'epoch': '0.8295'}
+{'loss': '0.8512', 'grad_norm': '0.4461', 'learning_rate': '0.000198', 'epoch': '0.8304'}
+{'loss': '0.8681', 'grad_norm': '0.4677', 'learning_rate': '0.000198', 'epoch': '0.8312'}
+{'loss': '0.7477', 'grad_norm': '0.3838', 'learning_rate': '0.000198', 'epoch': '0.8321'}
+{'loss': '0.8142', 'grad_norm': '0.3717', 'learning_rate': '0.000198', 'epoch': '0.8329'}
+{'loss': '0.7669', 'grad_norm': '0.4345', 'learning_rate': '0.000198', 'epoch': '0.8338'}
+{'loss': '0.8322', 'grad_norm': '0.4587', 'learning_rate': '0.000198', 'epoch': '0.8346'}
+{'loss': '0.7824', 'grad_norm': '0.4062', 'learning_rate': '0.000198', 'epoch': '0.8354'}
+{'loss': '0.8752', 'grad_norm': '0.3838', 'learning_rate': '0.000198', 'epoch': '0.8363'}
+{'loss': '0.667', 'grad_norm': '0.3634', 'learning_rate': '0.0001979', 'epoch': '0.8371'}
+{'loss': '0.8582', 'grad_norm': '0.4638', 'learning_rate': '0.0001979', 'epoch': '0.838'}
+{'loss': '0.7725', 'grad_norm': '0.3836', 'learning_rate': '0.0001979', 'epoch': '0.8388'}
+{'loss': '0.8542', 'grad_norm': '0.4397', 'learning_rate': '0.0001979', 'epoch': '0.8397'}
+{'loss': '0.8424', 'grad_norm': '0.4623', 'learning_rate': '0.0001979', 'epoch': '0.8405'}
+{'loss': '0.8401', 'grad_norm': '0.4492', 'learning_rate': '0.0001979', 'epoch': '0.8414'}
+{'loss': '0.7265', 'grad_norm': '0.3723', 'learning_rate': '0.0001979', 'epoch': '0.8422'}
+{'loss': '0.8458', 'grad_norm': '0.4329', 'learning_rate': '0.0001979', 'epoch': '0.843'}
+{'loss': '0.8517', 'grad_norm': '0.4225', 'learning_rate': '0.0001979', 'epoch': '0.8439'}
+{'eval_loss': '0.8461', 'eval_runtime': '685.3', 'eval_samples_per_second': '3.075', 'eval_steps_per_second': '3.075', 'epoch': '0.8439'}
+{'loss': '0.8045', 'grad_norm': '0.3927', 'learning_rate': '0.0001979', 'epoch': '0.8447'}
+{'loss': '0.751', 'grad_norm': '0.3819', 'learning_rate': '0.0001979', 'epoch': '0.8456'}
+{'loss': '0.773', 'grad_norm': '0.4038', 'learning_rate': '0.0001978', 'epoch': '0.8464'}
+{'loss': '0.7097', 'grad_norm': '0.3665', 'learning_rate': '0.0001978', 'epoch': '0.8473'}
+{'loss': '0.8186', 'grad_norm': '0.3662', 'learning_rate': '0.0001978', 'epoch': '0.8481'}
+{'loss': '0.8503', 'grad_norm': '0.4773', 'learning_rate': '0.0001978', 'epoch': '0.8489'}
+{'loss': '0.9011', 'grad_norm': '0.4446', 'learning_rate': '0.0001978', 'epoch': '0.8498'}
+{'loss': '0.7742', 'grad_norm': '0.4238', 'learning_rate': '0.0001978', 'epoch': '0.8506'}
+{'loss': '0.8488', 'grad_norm': '0.4302', 'learning_rate': '0.0001978', 'epoch': '0.8515'}
+{'loss': '0.8005', 'grad_norm': '0.3819', 'learning_rate': '0.0001978', 'epoch': '0.8523'}
+{'loss': '0.7681', 'grad_norm': '0.3684', 'learning_rate': '0.0001978', 'epoch': '0.8532'}
+{'loss': '0.899', 'grad_norm': '0.4448', 'learning_rate': '0.0001978', 'epoch': '0.854'}
+{'loss': '0.7676', 'grad_norm': '0.3432', 'learning_rate': '0.0001977', 'epoch': '0.8549'}
+{'loss': '0.8547', 'grad_norm': '0.3729', 'learning_rate': '0.0001977', 'epoch': '0.8557'}
+{'loss': '0.801', 'grad_norm': '0.4277', 'learning_rate': '0.0001977', 'epoch': '0.8565'}
+{'loss': '0.8424', 'grad_norm': '0.4126', 'learning_rate': '0.0001977', 'epoch': '0.8574'}
+{'loss': '0.7941', 'grad_norm': '0.4399', 'learning_rate': '0.0001977', 'epoch': '0.8582'}
+{'loss': '0.9424', 'grad_norm': '0.418', 'learning_rate': '0.0001977', 'epoch': '0.8591'}
+{'loss': '0.9103', 'grad_norm': '0.5245', 'learning_rate': '0.0001977', 'epoch': '0.8599'}
+{'loss': '0.7435', 'grad_norm': '0.4093', 'learning_rate': '0.0001977', 'epoch': '0.8608'}
+{'loss': '0.7442', 'grad_norm': '0.4079', 'learning_rate': '0.0001977', 'epoch': '0.8616'}
+{'loss': '0.8318', 'grad_norm': '0.4244', 'learning_rate': '0.0001976', 'epoch': '0.8624'}
+{'loss': '0.7645', 'grad_norm': '0.3826', 'learning_rate': '0.0001976', 'epoch': '0.8633'}
+{'loss': '0.8666', 'grad_norm': '0.433', 'learning_rate': '0.0001976', 'epoch': '0.8641'}
+{'loss': '0.7607', 'grad_norm': '0.3638', 'learning_rate': '0.0001976', 'epoch': '0.865'}
+{'loss': '0.8275', 'grad_norm': '0.4297', 'learning_rate': '0.0001976', 'epoch': '0.8658'}
+{'loss': '0.7419', 'grad_norm': '0.3831', 'learning_rate': '0.0001976', 'epoch': '0.8667'}
+{'loss': '0.8603', 'grad_norm': '0.4471', 'learning_rate': '0.0001976', 'epoch': '0.8675'}
+{'loss': '0.9007', 'grad_norm': '0.4401', 'learning_rate': '0.0001976', 'epoch': '0.8684'}
+{'loss': '0.8416', 'grad_norm': '0.4217', 'learning_rate': '0.0001976', 'epoch': '0.8692'}
+{'loss': '0.8281', 'grad_norm': '0.4082', 'learning_rate': '0.0001976', 'epoch': '0.87'}
+{'loss': '0.8387', 'grad_norm': '0.4272', 'learning_rate': '0.0001975', 'epoch': '0.8709'}
+{'loss': '0.8338', 'grad_norm': '0.3733', 'learning_rate': '0.0001975', 'epoch': '0.8717'}
+{'loss': '0.7677', 'grad_norm': '0.4048', 'learning_rate': '0.0001975', 'epoch': '0.8726'}
+{'loss': '0.8321', 'grad_norm': '0.4363', 'learning_rate': '0.0001975', 'epoch': '0.8734'}
+{'loss': '0.848', 'grad_norm': '0.4367', 'learning_rate': '0.0001975', 'epoch': '0.8743'}
+{'loss': '0.7599', 'grad_norm': '0.3679', 'learning_rate': '0.0001975', 'epoch': '0.8751'}
+{'loss': '0.7907', 'grad_norm': '0.4316', 'learning_rate': '0.0001975', 'epoch': '0.8759'}
+{'loss': '0.8288', 'grad_norm': '0.4265', 'learning_rate': '0.0001975', 'epoch': '0.8768'}
+{'loss': '0.8465', 'grad_norm': '0.4473', 'learning_rate': '0.0001975', 'epoch': '0.8776'}
+{'loss': '0.7383', 'grad_norm': '0.3788', 'learning_rate': '0.0001974', 'epoch': '0.8785'}
+{'loss': '0.7292', 'grad_norm': '0.3915', 'learning_rate': '0.0001974', 'epoch': '0.8793'}
+{'loss': '0.6923', 'grad_norm': '0.3977', 'learning_rate': '0.0001974', 'epoch': '0.8802'}
+{'loss': '0.854', 'grad_norm': '0.4397', 'learning_rate': '0.0001974', 'epoch': '0.881'}
+{'loss': '0.8571', 'grad_norm': '0.4968', 'learning_rate': '0.0001974', 'epoch': '0.8819'}
+{'loss': '0.7841', 'grad_norm': '0.4169', 'learning_rate': '0.0001974', 'epoch': '0.8827'}
+{'loss': '0.8521', 'grad_norm': '0.3879', 'learning_rate': '0.0001974', 'epoch': '0.8835'}
+{'loss': '0.8458', 'grad_norm': '0.3837', 'learning_rate': '0.0001974', 'epoch': '0.8844'}
+{'loss': '0.7426', 'grad_norm': '0.3887', 'learning_rate': '0.0001974', 'epoch': '0.8852'}
+{'loss': '0.7742', 'grad_norm': '0.4068', 'learning_rate': '0.0001973', 'epoch': '0.8861'}
+{'eval_loss': '0.8401', 'eval_runtime': '669.1', 'eval_samples_per_second': '3.149', 'eval_steps_per_second': '3.149', 'epoch': '0.8861'}
+{'loss': '0.91', 'grad_norm': '0.4282', 'learning_rate': '0.0001973', 'epoch': '0.8869'}
+{'loss': '0.7459', 'grad_norm': '0.3388', 'learning_rate': '0.0001973', 'epoch': '0.8878'}
+{'loss': '0.783', 'grad_norm': '0.4396', 'learning_rate': '0.0001973', 'epoch': '0.8886'}
+{'loss': '0.7899', 'grad_norm': '0.4005', 'learning_rate': '0.0001973', 'epoch': '0.8895'}
+{'loss': '0.7874', 'grad_norm': '0.4057', 'learning_rate': '0.0001973', 'epoch': '0.8903'}
+{'loss': '0.85', 'grad_norm': '0.4209', 'learning_rate': '0.0001973', 'epoch': '0.8911'}
+{'loss': '0.794', 'grad_norm': '0.3885', 'learning_rate': '0.0001973', 'epoch': '0.892'}
+{'loss': '0.8279', 'grad_norm': '0.4158', 'learning_rate': '0.0001973', 'epoch': '0.8928'}
+{'loss': '0.8218', 'grad_norm': '0.3762', 'learning_rate': '0.0001972', 'epoch': '0.8937'}
+{'loss': '0.7638', 'grad_norm': '0.9318', 'learning_rate': '0.0001972', 'epoch': '0.8945'}
+{'loss': '0.8702', 'grad_norm': '0.4195', 'learning_rate': '0.0001972', 'epoch': '0.8954'}
+{'loss': '0.8147', 'grad_norm': '0.4718', 'learning_rate': '0.0001972', 'epoch': '0.8962'}
+{'loss': '0.8646', 'grad_norm': '0.4471', 'learning_rate': '0.0001972', 'epoch': '0.897'}
+{'loss': '0.7926', 'grad_norm': '0.4177', 'learning_rate': '0.0001972', 'epoch': '0.8979'}
+{'loss': '0.8142', 'grad_norm': '0.4855', 'learning_rate': '0.0001972', 'epoch': '0.8987'}
+{'loss': '0.8051', 'grad_norm': '0.427', 'learning_rate': '0.0001972', 'epoch': '0.8996'}
+{'loss': '0.7888', 'grad_norm': '0.3771', 'learning_rate': '0.0001972', 'epoch': '0.9004'}
+{'loss': '0.7348', 'grad_norm': '0.4181', 'learning_rate': '0.0001971', 'epoch': '0.9013'}
+{'loss': '0.773', 'grad_norm': '0.4475', 'learning_rate': '0.0001971', 'epoch': '0.9021'}
+{'loss': '0.8066', 'grad_norm': '0.4052', 'learning_rate': '0.0001971', 'epoch': '0.903'}
+{'loss': '0.8053', 'grad_norm': '0.4567', 'learning_rate': '0.0001971', 'epoch': '0.9038'}
+{'loss': '0.7243', 'grad_norm': '0.4304', 'learning_rate': '0.0001971', 'epoch': '0.9046'}
+{'loss': '0.8475', 'grad_norm': '0.4785', 'learning_rate': '0.0001971', 'epoch': '0.9055'}
+{'loss': '0.9285', 'grad_norm': '0.4523', 'learning_rate': '0.0001971', 'epoch': '0.9063'}
+{'loss': '0.8136', 'grad_norm': '0.3794', 'learning_rate': '0.0001971', 'epoch': '0.9072'}
+{'loss': '0.7808', 'grad_norm': '0.4005', 'learning_rate': '0.0001971', 'epoch': '0.908'}
+{'loss': '0.8293', 'grad_norm': '0.3793', 'learning_rate': '0.000197', 'epoch': '0.9089'}
+{'loss': '0.7729', 'grad_norm': '0.4459', 'learning_rate': '0.000197', 'epoch': '0.9097'}
+{'loss': '0.8204', 'grad_norm': '0.4191', 'learning_rate': '0.000197', 'epoch': '0.9105'}
+{'loss': '0.8361', 'grad_norm': '0.4061', 'learning_rate': '0.000197', 'epoch': '0.9114'}
+{'loss': '0.8456', 'grad_norm': '0.4213', 'learning_rate': '0.000197', 'epoch': '0.9122'}
+{'loss': '0.8732', 'grad_norm': '0.4926', 'learning_rate': '0.000197', 'epoch': '0.9131'}
+{'loss': '0.815', 'grad_norm': '0.42', 'learning_rate': '0.000197', 'epoch': '0.9139'}
+{'loss': '0.7509', 'grad_norm': '0.375', 'learning_rate': '0.000197', 'epoch': '0.9148'}
+{'loss': '0.8303', 'grad_norm': '0.3706', 'learning_rate': '0.000197', 'epoch': '0.9156'}
+{'loss': '0.7297', 'grad_norm': '0.3873', 'learning_rate': '0.0001969', 'epoch': '0.9165'}
+{'loss': '0.7924', 'grad_norm': '0.4288', 'learning_rate': '0.0001969', 'epoch': '0.9173'}
+{'loss': '0.8098', 'grad_norm': '0.4259', 'learning_rate': '0.0001969', 'epoch': '0.9181'}
+{'loss': '0.6789', 'grad_norm': '0.3815', 'learning_rate': '0.0001969', 'epoch': '0.919'}
+{'loss': '0.8583', 'grad_norm': '0.4135', 'learning_rate': '0.0001969', 'epoch': '0.9198'}
+{'loss': '0.806', 'grad_norm': '0.3819', 'learning_rate': '0.0001969', 'epoch': '0.9207'}
+{'loss': '0.7432', 'grad_norm': '0.3902', 'learning_rate': '0.0001969', 'epoch': '0.9215'}
+{'loss': '0.7561', 'grad_norm': '0.3786', 'learning_rate': '0.0001969', 'epoch': '0.9224'}
+{'loss': '0.7553', 'grad_norm': '0.4508', 'learning_rate': '0.0001968', 'epoch': '0.9232'}
+{'loss': '0.6905', 'grad_norm': '0.3791', 'learning_rate': '0.0001968', 'epoch': '0.9241'}
+{'loss': '0.7711', 'grad_norm': '0.4439', 'learning_rate': '0.0001968', 'epoch': '0.9249'}
+{'loss': '0.8213', 'grad_norm': '0.3908', 'learning_rate': '0.0001968', 'epoch': '0.9257'}
+{'loss': '0.7514', 'grad_norm': '0.3528', 'learning_rate': '0.0001968', 'epoch': '0.9266'}
+{'loss': '0.8204', 'grad_norm': '0.4351', 'learning_rate': '0.0001968', 'epoch': '0.9274'}
+{'loss': '0.7638', 'grad_norm': '0.4557', 'learning_rate': '0.0001968', 'epoch': '0.9283'}
+{'eval_loss': '0.8337', 'eval_runtime': '674.5', 'eval_samples_per_second': '3.124', 'eval_steps_per_second': '3.124', 'epoch': '0.9283'}
+{'loss': '0.8617', 'grad_norm': '0.4516', 'learning_rate': '0.0001968', 'epoch': '0.9291'}
+{'loss': '0.8063', 'grad_norm': '0.4342', 'learning_rate': '0.0001967', 'epoch': '0.93'}
+{'loss': '0.807', 'grad_norm': '0.4244', 'learning_rate': '0.0001967', 'epoch': '0.9308'}
+{'loss': '0.7336', 'grad_norm': '0.3976', 'learning_rate': '0.0001967', 'epoch': '0.9316'}
+{'loss': '0.756', 'grad_norm': '0.4131', 'learning_rate': '0.0001967', 'epoch': '0.9325'}
+{'loss': '0.8447', 'grad_norm': '0.4599', 'learning_rate': '0.0001967', 'epoch': '0.9333'}
+{'loss': '0.8329', 'grad_norm': '0.4291', 'learning_rate': '0.0001967', 'epoch': '0.9342'}
+{'loss': '0.7492', 'grad_norm': '0.4227', 'learning_rate': '0.0001967', 'epoch': '0.935'}
+{'loss': '0.747', 'grad_norm': '0.416', 'learning_rate': '0.0001967', 'epoch': '0.9359'}
+{'loss': '0.8094', 'grad_norm': '0.4161', 'learning_rate': '0.0001966', 'epoch': '0.9367'}
+{'loss': '0.8398', 'grad_norm': '0.4528', 'learning_rate': '0.0001966', 'epoch': '0.9376'}
+{'loss': '0.7652', 'grad_norm': '0.3961', 'learning_rate': '0.0001966', 'epoch': '0.9384'}
+{'loss': '0.7909', 'grad_norm': '0.4269', 'learning_rate': '0.0001966', 'epoch': '0.9392'}
+{'loss': '0.7822', 'grad_norm': '0.3819', 'learning_rate': '0.0001966', 'epoch': '0.9401'}
+{'loss': '0.7793', 'grad_norm': '0.459', 'learning_rate': '0.0001966', 'epoch': '0.9409'}
+{'loss': '0.8857', 'grad_norm': '0.4741', 'learning_rate': '0.0001966', 'epoch': '0.9418'}
+{'loss': '0.7898', 'grad_norm': '0.3913', 'learning_rate': '0.0001966', 'epoch': '0.9426'}
+{'loss': '0.8646', 'grad_norm': '0.4552', 'learning_rate': '0.0001965', 'epoch': '0.9435'}
+{'loss': '0.8068', 'grad_norm': '0.4179', 'learning_rate': '0.0001965', 'epoch': '0.9443'}
+{'loss': '0.7953', 'grad_norm': '0.3865', 'learning_rate': '0.0001965', 'epoch': '0.9451'}
+{'loss': '0.8423', 'grad_norm': '0.3979', 'learning_rate': '0.0001965', 'epoch': '0.946'}
+{'loss': '0.8884', 'grad_norm': '0.4746', 'learning_rate': '0.0001965', 'epoch': '0.9468'}
+{'loss': '0.7233', 'grad_norm': '0.3867', 'learning_rate': '0.0001965', 'epoch': '0.9477'}
+{'loss': '0.7265', 'grad_norm': '0.3951', 'learning_rate': '0.0001965', 'epoch': '0.9485'}
+{'loss': '0.8', 'grad_norm': '0.4316', 'learning_rate': '0.0001965', 'epoch': '0.9494'}
+{'loss': '0.8231', 'grad_norm': '0.404', 'learning_rate': '0.0001964', 'epoch': '0.9502'}
+{'loss': '0.8458', 'grad_norm': '0.4311', 'learning_rate': '0.0001964', 'epoch': '0.9511'}
+{'loss': '0.7866', 'grad_norm': '0.3885', 'learning_rate': '0.0001964', 'epoch': '0.9519'}
+{'loss': '0.7096', 'grad_norm': '0.3384', 'learning_rate': '0.0001964', 'epoch': '0.9527'}
+{'loss': '0.8272', 'grad_norm': '0.4043', 'learning_rate': '0.0001964', 'epoch': '0.9536'}
+{'loss': '0.8205', 'grad_norm': '0.455', 'learning_rate': '0.0001964', 'epoch': '0.9544'}
+{'loss': '0.8848', 'grad_norm': '0.472', 'learning_rate': '0.0001964', 'epoch': '0.9553'}
+{'loss': '0.7631', 'grad_norm': '0.3959', 'learning_rate': '0.0001964', 'epoch': '0.9561'}
+{'loss': '0.7979', 'grad_norm': '0.4197', 'learning_rate': '0.0001963', 'epoch': '0.957'}
+{'loss': '0.7756', 'grad_norm': '0.3813', 'learning_rate': '0.0001963', 'epoch': '0.9578'}
+{'loss': '0.7553', 'grad_norm': '0.3868', 'learning_rate': '0.0001963', 'epoch': '0.9586'}
+{'loss': '0.7792', 'grad_norm': '0.4351', 'learning_rate': '0.0001963', 'epoch': '0.9595'}
+{'loss': '0.8782', 'grad_norm': '0.4717', 'learning_rate': '0.0001963', 'epoch': '0.9603'}
+{'loss': '0.8891', 'grad_norm': '0.4555', 'learning_rate': '0.0001963', 'epoch': '0.9612'}
+{'loss': '0.8025', 'grad_norm': '0.4421', 'learning_rate': '0.0001963', 'epoch': '0.962'}
+{'loss': '0.8036', 'grad_norm': '0.4039', 'learning_rate': '0.0001963', 'epoch': '0.9629'}
+{'loss': '0.8072', 'grad_norm': '0.3867', 'learning_rate': '0.0001962', 'epoch': '0.9637'}
+{'loss': '0.8046', 'grad_norm': '0.4113', 'learning_rate': '0.0001962', 'epoch': '0.9646'}
+{'loss': '0.8434', 'grad_norm': '0.4072', 'learning_rate': '0.0001962', 'epoch': '0.9654'}
+{'loss': '0.7832', 'grad_norm': '0.3576', 'learning_rate': '0.0001962', 'epoch': '0.9662'}
+{'loss': '0.7976', 'grad_norm': '0.417', 'learning_rate': '0.0001962', 'epoch': '0.9671'}
+{'loss': '0.8554', 'grad_norm': '0.4127', 'learning_rate': '0.0001962', 'epoch': '0.9679'}
+{'loss': '0.8245', 'grad_norm': '0.4232', 'learning_rate': '0.0001962', 'epoch': '0.9688'}
+{'loss': '0.7008', 'grad_norm': '0.3556', 'learning_rate': '0.0001962', 'epoch': '0.9696'}
+{'loss': '0.8193', 'grad_norm': '0.437', 'learning_rate': '0.0001961', 'epoch': '0.9705'}
+{'eval_loss': '0.8282', 'eval_runtime': '688.6', 'eval_samples_per_second': '3.06', 'eval_steps_per_second': '3.06', 'epoch': '0.9705'}
+{'loss': '0.8789', 'grad_norm': '0.4208', 'learning_rate': '0.0001961', 'epoch': '0.9713'}
+{'loss': '0.7362', 'grad_norm': '0.3761', 'learning_rate': '0.0001961', 'epoch': '0.9722'}
+{'loss': '0.8404', 'grad_norm': '0.4331', 'learning_rate': '0.0001961', 'epoch': '0.973'}
+{'loss': '0.8415', 'grad_norm': '0.4533', 'learning_rate': '0.0001961', 'epoch': '0.9738'}
+{'loss': '0.8106', 'grad_norm': '0.4026', 'learning_rate': '0.0001961', 'epoch': '0.9747'}
+{'loss': '0.8054', 'grad_norm': '0.3789', 'learning_rate': '0.0001961', 'epoch': '0.9755'}
+{'loss': '0.8093', 'grad_norm': '0.4627', 'learning_rate': '0.000196', 'epoch': '0.9764'}
+{'loss': '0.7456', 'grad_norm': '0.377', 'learning_rate': '0.000196', 'epoch': '0.9772'}
+{'loss': '0.774', 'grad_norm': '0.4408', 'learning_rate': '0.000196', 'epoch': '0.9781'}
+{'loss': '0.6961', 'grad_norm': '0.388', 'learning_rate': '0.000196', 'epoch': '0.9789'}
+{'loss': '0.8448', 'grad_norm': '0.468', 'learning_rate': '0.000196', 'epoch': '0.9797'}
+{'loss': '0.8156', 'grad_norm': '0.4124', 'learning_rate': '0.000196', 'epoch': '0.9806'}
+{'loss': '0.7991', 'grad_norm': '0.3828', 'learning_rate': '0.000196', 'epoch': '0.9814'}
+{'loss': '0.89', 'grad_norm': '0.4644', 'learning_rate': '0.000196', 'epoch': '0.9823'}
+{'loss': '0.7863', 'grad_norm': '0.437', 'learning_rate': '0.0001959', 'epoch': '0.9831'}
+{'loss': '0.7961', 'grad_norm': '0.4071', 'learning_rate': '0.0001959', 'epoch': '0.984'}
+{'loss': '0.7141', 'grad_norm': '0.3736', 'learning_rate': '0.0001959', 'epoch': '0.9848'}
+{'loss': '0.7312', 'grad_norm': '0.4095', 'learning_rate': '0.0001959', 'epoch': '0.9857'}
+{'loss': '0.8191', 'grad_norm': '0.4275', 'learning_rate': '0.0001959', 'epoch': '0.9865'}
+{'loss': '0.7476', 'grad_norm': '0.4628', 'learning_rate': '0.0001959', 'epoch': '0.9873'}
+{'loss': '0.8258', 'grad_norm': '0.4468', 'learning_rate': '0.0001959', 'epoch': '0.9882'}
+{'loss': '0.7922', 'grad_norm': '0.4067', 'learning_rate': '0.0001958', 'epoch': '0.989'}
+{'loss': '0.7892', 'grad_norm': '0.4087', 'learning_rate': '0.0001958', 'epoch': '0.9899'}
+{'loss': '0.8767', 'grad_norm': '0.4331', 'learning_rate': '0.0001958', 'epoch': '0.9907'}
+{'loss': '0.9183', 'grad_norm': '0.4156', 'learning_rate': '0.0001958', 'epoch': '0.9916'}
+{'loss': '0.7906', 'grad_norm': '0.4327', 'learning_rate': '0.0001958', 'epoch': '0.9924'}
+{'loss': '0.828', 'grad_norm': '0.4117', 'learning_rate': '0.0001958', 'epoch': '0.9932'}
+{'loss': '0.7451', 'grad_norm': '0.3927', 'learning_rate': '0.0001958', 'epoch': '0.9941'}
+{'loss': '0.6769', 'grad_norm': '0.339', 'learning_rate': '0.0001957', 'epoch': '0.9949'}
+{'loss': '0.7591', 'grad_norm': '0.4221', 'learning_rate': '0.0001957', 'epoch': '0.9958'}
+{'loss': '0.8301', 'grad_norm': '0.43', 'learning_rate': '0.0001957', 'epoch': '0.9966'}
+{'loss': '0.8019', 'grad_norm': '0.4032', 'learning_rate': '0.0001957', 'epoch': '0.9975'}
+{'loss': '0.7427', 'grad_norm': '0.3958', 'learning_rate': '0.0001957', 'epoch': '0.9983'}
+{'loss': '0.8222', 'grad_norm': '0.4174', 'learning_rate': '0.0001957', 'epoch': '0.9992'}
+{'loss': '0.8143', 'grad_norm': '0.4712', 'learning_rate': '0.0001957', 'epoch': '1'}
+{'loss': '0.6972', 'grad_norm': '0.3855', 'learning_rate': '0.0001956', 'epoch': '1.001'}
+{'loss': '0.6719', 'grad_norm': '0.4029', 'learning_rate': '0.0001956', 'epoch': '1.002'}
+{'loss': '0.7162', 'grad_norm': '0.4014', 'learning_rate': '0.0001956', 'epoch': '1.003'}
+{'loss': '0.717', 'grad_norm': '0.4069', 'learning_rate': '0.0001956', 'epoch': '1.003'}
+{'loss': '0.803', 'grad_norm': '0.4615', 'learning_rate': '0.0001956', 'epoch': '1.004'}
+{'loss': '0.7666', 'grad_norm': '0.4505', 'learning_rate': '0.0001956', 'epoch': '1.005'}
+{'loss': '0.7481', 'grad_norm': '0.426', 'learning_rate': '0.0001956', 'epoch': '1.006'}
+{'loss': '0.7944', 'grad_norm': '0.482', 'learning_rate': '0.0001955', 'epoch': '1.007'}
+{'loss': '0.6402', 'grad_norm': '0.4579', 'learning_rate': '0.0001955', 'epoch': '1.008'}
+{'loss': '0.7059', 'grad_norm': '0.3702', 'learning_rate': '0.0001955', 'epoch': '1.008'}
+{'loss': '0.7631', 'grad_norm': '0.4442', 'learning_rate': '0.0001955', 'epoch': '1.009'}
+{'loss': '0.7874', 'grad_norm': '0.4478', 'learning_rate': '0.0001955', 'epoch': '1.01'}
+{'loss': '0.7882', 'grad_norm': '0.4596', 'learning_rate': '0.0001955', 'epoch': '1.011'}
+{'loss': '0.6359', 'grad_norm': '0.4063', 'learning_rate': '0.0001955', 'epoch': '1.012'}
+{'loss': '0.7804', 'grad_norm': '0.4763', 'learning_rate': '0.0001954', 'epoch': '1.013'}
+{'eval_loss': '0.8251', 'eval_runtime': '691.6', 'eval_samples_per_second': '3.047', 'eval_steps_per_second': '3.047', 'epoch': '1.013'}
+{'loss': '0.7395', 'grad_norm': '0.4069', 'learning_rate': '0.0001954', 'epoch': '1.014'}
+{'loss': '0.7097', 'grad_norm': '0.4554', 'learning_rate': '0.0001954', 'epoch': '1.014'}
+{'loss': '0.7962', 'grad_norm': '0.4367', 'learning_rate': '0.0001954', 'epoch': '1.015'}
+{'loss': '0.7438', 'grad_norm': '0.4458', 'learning_rate': '0.0001954', 'epoch': '1.016'}
+{'loss': '0.7274', 'grad_norm': '0.3742', 'learning_rate': '0.0001954', 'epoch': '1.017'}
+{'loss': '0.7923', 'grad_norm': '0.434', 'learning_rate': '0.0001954', 'epoch': '1.018'}
+{'loss': '0.6492', 'grad_norm': '0.4518', 'learning_rate': '0.0001953', 'epoch': '1.019'}
+{'loss': '0.7423', 'grad_norm': '0.4161', 'learning_rate': '0.0001953', 'epoch': '1.019'}
+{'loss': '0.6489', 'grad_norm': '0.3595', 'learning_rate': '0.0001953', 'epoch': '1.02'}
+{'loss': '0.7449', 'grad_norm': '0.3877', 'learning_rate': '0.0001953', 'epoch': '1.021'}
+{'loss': '0.8481', 'grad_norm': '0.4819', 'learning_rate': '0.0001953', 'epoch': '1.022'}
+{'loss': '0.7442', 'grad_norm': '0.4098', 'learning_rate': '0.0001953', 'epoch': '1.023'}
+{'loss': '0.7564', 'grad_norm': '0.4128', 'learning_rate': '0.0001953', 'epoch': '1.024'}
+{'loss': '0.7235', 'grad_norm': '0.4495', 'learning_rate': '0.0001952', 'epoch': '1.024'}
+{'loss': '0.7639', 'grad_norm': '0.3994', 'learning_rate': '0.0001952', 'epoch': '1.025'}
+{'loss': '0.6923', 'grad_norm': '0.4081', 'learning_rate': '0.0001952', 'epoch': '1.026'}
+{'loss': '0.763', 'grad_norm': '0.4574', 'learning_rate': '0.0001952', 'epoch': '1.027'}
+{'loss': '0.7812', 'grad_norm': '0.428', 'learning_rate': '0.0001952', 'epoch': '1.028'}
+{'loss': '0.7278', 'grad_norm': '0.4196', 'learning_rate': '0.0001952', 'epoch': '1.029'}
+{'loss': '0.8075', 'grad_norm': '0.4833', 'learning_rate': '0.0001952', 'epoch': '1.03'}
+{'loss': '0.7421', 'grad_norm': '0.4645', 'learning_rate': '0.0001951', 'epoch': '1.03'}
+{'loss': '0.6892', 'grad_norm': '0.4136', 'learning_rate': '0.0001951', 'epoch': '1.031'}
+{'loss': '0.7305', 'grad_norm': '0.4158', 'learning_rate': '0.0001951', 'epoch': '1.032'}
+{'loss': '0.7683', 'grad_norm': '0.4725', 'learning_rate': '0.0001951', 'epoch': '1.033'}
+{'loss': '0.747', 'grad_norm': '0.4341', 'learning_rate': '0.0001951', 'epoch': '1.034'}
+{'loss': '0.7312', 'grad_norm': '0.4452', 'learning_rate': '0.0001951', 'epoch': '1.035'}
+{'loss': '0.7861', 'grad_norm': '0.5314', 'learning_rate': '0.0001951', 'epoch': '1.035'}
+{'loss': '0.8458', 'grad_norm': '0.6094', 'learning_rate': '0.000195', 'epoch': '1.036'}
+{'loss': '0.6091', 'grad_norm': '0.3642', 'learning_rate': '0.000195', 'epoch': '1.037'}
+{'loss': '0.6703', 'grad_norm': '0.3993', 'learning_rate': '0.000195', 'epoch': '1.038'}
+{'loss': '0.7221', 'grad_norm': '0.3856', 'learning_rate': '0.000195', 'epoch': '1.039'}
+{'loss': '0.7501', 'grad_norm': '0.442', 'learning_rate': '0.000195', 'epoch': '1.04'}
+{'loss': '0.7958', 'grad_norm': '0.4739', 'learning_rate': '0.000195', 'epoch': '1.041'}
+{'loss': '0.7965', 'grad_norm': '0.5274', 'learning_rate': '0.000195', 'epoch': '1.041'}
+{'loss': '0.7479', 'grad_norm': '0.44', 'learning_rate': '0.0001949', 'epoch': '1.042'}
+{'loss': '0.7383', 'grad_norm': '0.4066', 'learning_rate': '0.0001949', 'epoch': '1.043'}
+{'loss': '0.8172', 'grad_norm': '0.4393', 'learning_rate': '0.0001949', 'epoch': '1.044'}
+{'loss': '0.7413', 'grad_norm': '0.4111', 'learning_rate': '0.0001949', 'epoch': '1.045'}
+{'loss': '0.7893', 'grad_norm': '0.3771', 'learning_rate': '0.0001949', 'epoch': '1.046'}
+{'loss': '0.7038', 'grad_norm': '0.4524', 'learning_rate': '0.0001949', 'epoch': '1.046'}
+{'loss': '0.7457', 'grad_norm': '0.4232', 'learning_rate': '0.0001948', 'epoch': '1.047'}
+{'loss': '0.7264', 'grad_norm': '0.3843', 'learning_rate': '0.0001948', 'epoch': '1.048'}
+{'loss': '0.7101', 'grad_norm': '0.432', 'learning_rate': '0.0001948', 'epoch': '1.049'}
+{'loss': '0.8562', 'grad_norm': '0.4694', 'learning_rate': '0.0001948', 'epoch': '1.05'}
+{'loss': '0.7207', 'grad_norm': '0.4455', 'learning_rate': '0.0001948', 'epoch': '1.051'}
+{'loss': '0.7272', 'grad_norm': '0.4254', 'learning_rate': '0.0001948', 'epoch': '1.051'}
+{'loss': '0.7359', 'grad_norm': '0.4276', 'learning_rate': '0.0001948', 'epoch': '1.052'}
+{'loss': '0.6618', 'grad_norm': '0.4006', 'learning_rate': '0.0001947', 'epoch': '1.053'}
+{'loss': '0.6901', 'grad_norm': '0.4132', 'learning_rate': '0.0001947', 'epoch': '1.054'}
+{'loss': '0.7353', 'grad_norm': '0.5116', 'learning_rate': '0.0001947', 'epoch': '1.055'}
+{'eval_loss': '0.8249', 'eval_runtime': '679.4', 'eval_samples_per_second': '3.101', 'eval_steps_per_second': '3.101', 'epoch': '1.055'}
+{'loss': '0.7583', 'grad_norm': '0.4282', 'learning_rate': '0.0001947', 'epoch': '1.056'}
+{'loss': '0.7775', 'grad_norm': '0.4789', 'learning_rate': '0.0001947', 'epoch': '1.057'}
+{'loss': '0.7453', 'grad_norm': '0.421', 'learning_rate': '0.0001947', 'epoch': '1.057'}
+{'loss': '0.7541', 'grad_norm': '0.4322', 'learning_rate': '0.0001946', 'epoch': '1.058'}
+{'loss': '0.7176', 'grad_norm': '0.4361', 'learning_rate': '0.0001946', 'epoch': '1.059'}
+{'loss': '0.7342', 'grad_norm': '0.4575', 'learning_rate': '0.0001946', 'epoch': '1.06'}
+{'loss': '0.8255', 'grad_norm': '0.4823', 'learning_rate': '0.0001946', 'epoch': '1.061'}
+{'loss': '0.6625', 'grad_norm': '0.4169', 'learning_rate': '0.0001946', 'epoch': '1.062'}
+{'loss': '0.8224', 'grad_norm': '0.4132', 'learning_rate': '0.0001946', 'epoch': '1.062'}
+{'loss': '0.7583', 'grad_norm': '0.4405', 'learning_rate': '0.0001946', 'epoch': '1.063'}
+{'loss': '0.7948', 'grad_norm': '0.4564', 'learning_rate': '0.0001945', 'epoch': '1.064'}
+{'loss': '0.7671', 'grad_norm': '0.4247', 'learning_rate': '0.0001945', 'epoch': '1.065'}
+{'loss': '0.7754', 'grad_norm': '0.4207', 'learning_rate': '0.0001945', 'epoch': '1.066'}
+{'loss': '0.7109', 'grad_norm': '0.3729', 'learning_rate': '0.0001945', 'epoch': '1.067'}
+{'loss': '0.7', 'grad_norm': '0.409', 'learning_rate': '0.0001945', 'epoch': '1.068'}
+{'loss': '0.7374', 'grad_norm': '0.4481', 'learning_rate': '0.0001945', 'epoch': '1.068'}
+{'loss': '0.7463', 'grad_norm': '0.484', 'learning_rate': '0.0001944', 'epoch': '1.069'}
+{'loss': '0.7448', 'grad_norm': '0.4982', 'learning_rate': '0.0001944', 'epoch': '1.07'}
+{'loss': '0.7053', 'grad_norm': '0.3987', 'learning_rate': '0.0001944', 'epoch': '1.071'}
+{'loss': '0.709', 'grad_norm': '0.409', 'learning_rate': '0.0001944', 'epoch': '1.072'}
+{'loss': '0.7531', 'grad_norm': '0.3703', 'learning_rate': '0.0001944', 'epoch': '1.073'}
+{'loss': '0.7598', 'grad_norm': '0.4016', 'learning_rate': '0.0001944', 'epoch': '1.073'}
+{'loss': '0.7168', 'grad_norm': '0.3915', 'learning_rate': '0.0001943', 'epoch': '1.074'}
+{'loss': '0.7763', 'grad_norm': '0.4464', 'learning_rate': '0.0001943', 'epoch': '1.075'}
+{'loss': '0.788', 'grad_norm': '0.4687', 'learning_rate': '0.0001943', 'epoch': '1.076'}
+{'loss': '0.7351', 'grad_norm': '0.4143', 'learning_rate': '0.0001943', 'epoch': '1.077'}
+{'loss': '0.8661', 'grad_norm': '0.457', 'learning_rate': '0.0001943', 'epoch': '1.078'}
+{'loss': '0.7123', 'grad_norm': '0.4173', 'learning_rate': '0.0001943', 'epoch': '1.078'}
+{'loss': '0.6605', 'grad_norm': '0.423', 'learning_rate': '0.0001942', 'epoch': '1.079'}
+{'loss': '0.7839', 'grad_norm': '0.4579', 'learning_rate': '0.0001942', 'epoch': '1.08'}
+{'loss': '0.7558', 'grad_norm': '0.4337', 'learning_rate': '0.0001942', 'epoch': '1.081'}
+{'loss': '0.7349', 'grad_norm': '0.4372', 'learning_rate': '0.0001942', 'epoch': '1.082'}
+{'loss': '0.7025', 'grad_norm': '0.4687', 'learning_rate': '0.0001942', 'epoch': '1.083'}
+{'loss': '0.6981', 'grad_norm': '0.4257', 'learning_rate': '0.0001942', 'epoch': '1.084'}
+{'loss': '0.79', 'grad_norm': '0.4323', 'learning_rate': '0.0001942', 'epoch': '1.084'}
+{'loss': '0.7964', 'grad_norm': '0.4438', 'learning_rate': '0.0001941', 'epoch': '1.085'}
+{'loss': '0.7903', 'grad_norm': '0.4739', 'learning_rate': '0.0001941', 'epoch': '1.086'}
+{'loss': '0.7819', 'grad_norm': '0.4576', 'learning_rate': '0.0001941', 'epoch': '1.087'}
+{'loss': '0.7516', 'grad_norm': '0.4124', 'learning_rate': '0.0001941', 'epoch': '1.088'}
+{'loss': '0.687', 'grad_norm': '0.3817', 'learning_rate': '0.0001941', 'epoch': '1.089'}
+{'loss': '0.7437', 'grad_norm': '0.463', 'learning_rate': '0.0001941', 'epoch': '1.089'}
+{'loss': '0.7102', 'grad_norm': '0.4064', 'learning_rate': '0.000194', 'epoch': '1.09'}
+{'loss': '0.6895', 'grad_norm': '0.4442', 'learning_rate': '0.000194', 'epoch': '1.091'}
+{'loss': '0.7277', 'grad_norm': '0.4267', 'learning_rate': '0.000194', 'epoch': '1.092'}
+{'loss': '0.6512', 'grad_norm': '0.4475', 'learning_rate': '0.000194', 'epoch': '1.093'}
+{'loss': '0.785', 'grad_norm': '0.4557', 'learning_rate': '0.000194', 'epoch': '1.094'}
+{'loss': '0.7869', 'grad_norm': '0.4498', 'learning_rate': '0.000194', 'epoch': '1.095'}
+{'loss': '0.8261', 'grad_norm': '0.4537', 'learning_rate': '0.0001939', 'epoch': '1.095'}
+{'loss': '0.7057', 'grad_norm': '0.4068', 'learning_rate': '0.0001939', 'epoch': '1.096'}
+{'loss': '0.7557', 'grad_norm': '0.3928', 'learning_rate': '0.0001939', 'epoch': '1.097'}
+{'eval_loss': '0.8211', 'eval_runtime': '670.2', 'eval_samples_per_second': '3.144', 'eval_steps_per_second': '3.144', 'epoch': '1.097'}
+{'loss': '0.714', 'grad_norm': '0.4261', 'learning_rate': '0.0001939', 'epoch': '1.098'}
+{'loss': '0.7275', 'grad_norm': '0.4664', 'learning_rate': '0.0001939', 'epoch': '1.099'}
+{'loss': '0.7547', 'grad_norm': '0.4631', 'learning_rate': '0.0001939', 'epoch': '1.1'}
+{'loss': '0.6943', 'grad_norm': '0.3975', 'learning_rate': '0.0001938', 'epoch': '1.1'}
+{'loss': '0.7162', 'grad_norm': '0.4557', 'learning_rate': '0.0001938', 'epoch': '1.101'}
+{'loss': '0.7166', 'grad_norm': '0.3863', 'learning_rate': '0.0001938', 'epoch': '1.102'}
+{'loss': '0.8081', 'grad_norm': '0.496', 'learning_rate': '0.0001938', 'epoch': '1.103'}
+{'loss': '0.7033', 'grad_norm': '0.4498', 'learning_rate': '0.0001938', 'epoch': '1.104'}
+{'loss': '0.7077', 'grad_norm': '0.4538', 'learning_rate': '0.0001938', 'epoch': '1.105'}
+{'loss': '0.7021', 'grad_norm': '0.4043', 'learning_rate': '0.0001937', 'epoch': '1.105'}
+{'loss': '0.7223', 'grad_norm': '0.4431', 'learning_rate': '0.0001937', 'epoch': '1.106'}
+{'loss': '0.6136', 'grad_norm': '0.3793', 'learning_rate': '0.0001937', 'epoch': '1.107'}
+{'loss': '0.7359', 'grad_norm': '0.4666', 'learning_rate': '0.0001937', 'epoch': '1.108'}
+{'loss': '0.7485', 'grad_norm': '0.4417', 'learning_rate': '0.0001937', 'epoch': '1.109'}
+{'loss': '0.6875', 'grad_norm': '0.462', 'learning_rate': '0.0001937', 'epoch': '1.11'}
+{'loss': '0.7169', 'grad_norm': '0.4358', 'learning_rate': '0.0001936', 'epoch': '1.111'}
+{'loss': '0.7534', 'grad_norm': '0.4581', 'learning_rate': '0.0001936', 'epoch': '1.111'}
+{'loss': '0.8245', 'grad_norm': '0.4641', 'learning_rate': '0.0001936', 'epoch': '1.112'}
+{'loss': '0.7096', 'grad_norm': '0.4639', 'learning_rate': '0.0001936', 'epoch': '1.113'}
+{'loss': '0.703', 'grad_norm': '0.4598', 'learning_rate': '0.0001936', 'epoch': '1.114'}
+{'loss': '0.7159', 'grad_norm': '0.4299', 'learning_rate': '0.0001936', 'epoch': '1.115'}
+{'loss': '0.6801', 'grad_norm': '0.3613', 'learning_rate': '0.0001935', 'epoch': '1.116'}
+{'loss': '0.695', 'grad_norm': '0.5004', 'learning_rate': '0.0001935', 'epoch': '1.116'}
+{'loss': '0.7164', 'grad_norm': '0.4372', 'learning_rate': '0.0001935', 'epoch': '1.117'}
+{'loss': '0.7027', 'grad_norm': '0.4232', 'learning_rate': '0.0001935', 'epoch': '1.118'}
+{'loss': '0.6774', 'grad_norm': '0.3585', 'learning_rate': '0.0001935', 'epoch': '1.119'}
+{'loss': '0.7789', 'grad_norm': '0.4661', 'learning_rate': '0.0001935', 'epoch': '1.12'}
+{'loss': '0.8647', 'grad_norm': '0.4298', 'learning_rate': '0.0001934', 'epoch': '1.121'}
+{'loss': '0.6949', 'grad_norm': '0.4638', 'learning_rate': '0.0001934', 'epoch': '1.122'}
+{'loss': '0.6677', 'grad_norm': '0.408', 'learning_rate': '0.0001934', 'epoch': '1.122'}
+{'loss': '0.7521', 'grad_norm': '0.4608', 'learning_rate': '0.0001934', 'epoch': '1.123'}
+{'loss': '0.6829', 'grad_norm': '0.3935', 'learning_rate': '0.0001934', 'epoch': '1.124'}
+{'loss': '0.7123', 'grad_norm': '0.457', 'learning_rate': '0.0001934', 'epoch': '1.125'}
+{'loss': '0.7317', 'grad_norm': '0.4078', 'learning_rate': '0.0001933', 'epoch': '1.126'}
+{'loss': '0.7323', 'grad_norm': '0.4282', 'learning_rate': '0.0001933', 'epoch': '1.127'}
+{'loss': '0.7177', 'grad_norm': '0.464', 'learning_rate': '0.0001933', 'epoch': '1.127'}
+{'loss': '0.7957', 'grad_norm': '0.456', 'learning_rate': '0.0001933', 'epoch': '1.128'}
+{'loss': '0.7621', 'grad_norm': '0.4791', 'learning_rate': '0.0001933', 'epoch': '1.129'}
+{'loss': '0.7228', 'grad_norm': '0.4652', 'learning_rate': '0.0001932', 'epoch': '1.13'}
+{'loss': '0.6968', 'grad_norm': '0.4791', 'learning_rate': '0.0001932', 'epoch': '1.131'}
+{'loss': '0.6682', 'grad_norm': '0.4413', 'learning_rate': '0.0001932', 'epoch': '1.132'}
+{'loss': '0.7288', 'grad_norm': '0.4138', 'learning_rate': '0.0001932', 'epoch': '1.132'}
+{'loss': '0.6548', 'grad_norm': '0.3912', 'learning_rate': '0.0001932', 'epoch': '1.133'}
+{'loss': '0.6942', 'grad_norm': '0.4547', 'learning_rate': '0.0001932', 'epoch': '1.134'}
+{'loss': '0.7418', 'grad_norm': '0.4032', 'learning_rate': '0.0001931', 'epoch': '1.135'}
+{'loss': '0.7781', 'grad_norm': '0.3807', 'learning_rate': '0.0001931', 'epoch': '1.136'}
+{'loss': '0.6615', 'grad_norm': '0.4376', 'learning_rate': '0.0001931', 'epoch': '1.137'}
+{'loss': '0.7443', 'grad_norm': '0.4541', 'learning_rate': '0.0001931', 'epoch': '1.138'}
+{'loss': '0.7444', 'grad_norm': '0.4791', 'learning_rate': '0.0001931', 'epoch': '1.138'}
+{'loss': '0.7043', 'grad_norm': '0.4172', 'learning_rate': '0.0001931', 'epoch': '1.139'}
+{'eval_loss': '0.8155', 'eval_runtime': '678.3', 'eval_samples_per_second': '3.106', 'eval_steps_per_second': '3.106', 'epoch': '1.139'}
+{'loss': '0.6716', 'grad_norm': '0.3863', 'learning_rate': '0.000193', 'epoch': '1.14'}
+{'loss': '0.6959', 'grad_norm': '0.4999', 'learning_rate': '0.000193', 'epoch': '1.141'}
+{'loss': '0.6534', 'grad_norm': '0.4638', 'learning_rate': '0.000193', 'epoch': '1.142'}
+{'loss': '0.675', 'grad_norm': '0.3842', 'learning_rate': '0.000193', 'epoch': '1.143'}
+{'loss': '0.7715', 'grad_norm': '0.4172', 'learning_rate': '0.000193', 'epoch': '1.143'}
+{'loss': '0.7281', 'grad_norm': '0.4438', 'learning_rate': '0.0001929', 'epoch': '1.144'}
+{'loss': '0.7407', 'grad_norm': '0.4335', 'learning_rate': '0.0001929', 'epoch': '1.145'}
+{'loss': '0.6794', 'grad_norm': '0.3897', 'learning_rate': '0.0001929', 'epoch': '1.146'}
+{'loss': '0.7026', 'grad_norm': '0.4289', 'learning_rate': '0.0001929', 'epoch': '1.147'}
+{'loss': '0.7614', 'grad_norm': '0.4784', 'learning_rate': '0.0001929', 'epoch': '1.148'}
+{'loss': '0.7123', 'grad_norm': '0.4152', 'learning_rate': '0.0001929', 'epoch': '1.149'}
+{'loss': '0.7198', 'grad_norm': '0.4205', 'learning_rate': '0.0001928', 'epoch': '1.149'}
+{'loss': '0.686', 'grad_norm': '0.4133', 'learning_rate': '0.0001928', 'epoch': '1.15'}
+{'loss': '0.7186', 'grad_norm': '0.4041', 'learning_rate': '0.0001928', 'epoch': '1.151'}
+{'loss': '0.79', 'grad_norm': '0.4487', 'learning_rate': '0.0001928', 'epoch': '1.152'}
+{'loss': '0.6267', 'grad_norm': '0.351', 'learning_rate': '0.0001928', 'epoch': '1.153'}
+{'loss': '0.745', 'grad_norm': '0.4296', 'learning_rate': '0.0001928', 'epoch': '1.154'}
+{'loss': '0.6876', 'grad_norm': '0.4063', 'learning_rate': '0.0001927', 'epoch': '1.154'}
+{'loss': '0.8142', 'grad_norm': '0.4243', 'learning_rate': '0.0001927', 'epoch': '1.155'}
+{'loss': '0.6905', 'grad_norm': '0.4833', 'learning_rate': '0.0001927', 'epoch': '1.156'}
+{'loss': '0.6709', 'grad_norm': '0.4304', 'learning_rate': '0.0001927', 'epoch': '1.157'}
+{'loss': '0.7107', 'grad_norm': '0.4533', 'learning_rate': '0.0001927', 'epoch': '1.158'}
+{'loss': '0.8044', 'grad_norm': '0.4823', 'learning_rate': '0.0001926', 'epoch': '1.159'}
+{'loss': '0.7812', 'grad_norm': '0.3893', 'learning_rate': '0.0001926', 'epoch': '1.159'}
+{'loss': '0.7142', 'grad_norm': '0.4674', 'learning_rate': '0.0001926', 'epoch': '1.16'}
+{'loss': '0.7528', 'grad_norm': '0.4335', 'learning_rate': '0.0001926', 'epoch': '1.161'}
+{'loss': '0.7818', 'grad_norm': '0.3792', 'learning_rate': '0.0001926', 'epoch': '1.162'}
+{'loss': '0.6905', 'grad_norm': '0.4125', 'learning_rate': '0.0001926', 'epoch': '1.163'}
+{'loss': '0.7572', 'grad_norm': '0.4437', 'learning_rate': '0.0001925', 'epoch': '1.164'}
+{'loss': '0.6747', 'grad_norm': '0.4012', 'learning_rate': '0.0001925', 'epoch': '1.165'}
+{'loss': '0.7722', 'grad_norm': '0.466', 'learning_rate': '0.0001925', 'epoch': '1.165'}
+{'loss': '0.8081', 'grad_norm': '0.4169', 'learning_rate': '0.0001925', 'epoch': '1.166'}
+{'loss': '0.7746', 'grad_norm': '0.4275', 'learning_rate': '0.0001925', 'epoch': '1.167'}
+{'loss': '0.7918', 'grad_norm': '0.4386', 'learning_rate': '0.0001924', 'epoch': '1.168'}
+{'loss': '0.7383', 'grad_norm': '0.4283', 'learning_rate': '0.0001924', 'epoch': '1.169'}
+{'loss': '0.7502', 'grad_norm': '0.4418', 'learning_rate': '0.0001924', 'epoch': '1.17'}
+{'loss': '0.6495', 'grad_norm': '0.4193', 'learning_rate': '0.0001924', 'epoch': '1.17'}
+{'loss': '0.8292', 'grad_norm': '0.5146', 'learning_rate': '0.0001924', 'epoch': '1.171'}
+{'loss': '0.7491', 'grad_norm': '0.4091', 'learning_rate': '0.0001924', 'epoch': '1.172'}
+{'loss': '0.6315', 'grad_norm': '0.392', 'learning_rate': '0.0001923', 'epoch': '1.173'}
+{'loss': '0.7539', 'grad_norm': '0.4423', 'learning_rate': '0.0001923', 'epoch': '1.174'}
+{'loss': '0.7441', 'grad_norm': '0.4694', 'learning_rate': '0.0001923', 'epoch': '1.175'}
+{'loss': '0.7343', 'grad_norm': '0.4948', 'learning_rate': '0.0001923', 'epoch': '1.176'}
+{'loss': '0.697', 'grad_norm': '0.39', 'learning_rate': '0.0001923', 'epoch': '1.176'}
+{'loss': '0.7098', 'grad_norm': '0.3615', 'learning_rate': '0.0001922', 'epoch': '1.177'}
+{'loss': '0.8142', 'grad_norm': '0.4117', 'learning_rate': '0.0001922', 'epoch': '1.178'}
+{'loss': '0.7459', 'grad_norm': '0.4543', 'learning_rate': '0.0001922', 'epoch': '1.179'}
+{'loss': '0.6536', 'grad_norm': '0.4026', 'learning_rate': '0.0001922', 'epoch': '1.18'}
+{'loss': '0.6248', 'grad_norm': '0.4417', 'learning_rate': '0.0001922', 'epoch': '1.181'}
+{'loss': '0.6778', 'grad_norm': '0.4503', 'learning_rate': '0.0001921', 'epoch': '1.181'}
+{'eval_loss': '0.8124', 'eval_runtime': '688.5', 'eval_samples_per_second': '3.06', 'eval_steps_per_second': '3.06', 'epoch': '1.181'}
+{'loss': '0.646', 'grad_norm': '0.4069', 'learning_rate': '0.0001921', 'epoch': '1.182'}
+{'loss': '0.7293', 'grad_norm': '0.5053', 'learning_rate': '0.0001921', 'epoch': '1.183'}
+{'loss': '0.7399', 'grad_norm': '0.4028', 'learning_rate': '0.0001921', 'epoch': '1.184'}
+{'loss': '0.6885', 'grad_norm': '0.4523', 'learning_rate': '0.0001921', 'epoch': '1.185'}
+{'loss': '0.6962', 'grad_norm': '0.4195', 'learning_rate': '0.0001921', 'epoch': '1.186'}
+{'loss': '0.7381', 'grad_norm': '0.4891', 'learning_rate': '0.000192', 'epoch': '1.186'}
+{'loss': '0.7554', 'grad_norm': '0.4091', 'learning_rate': '0.000192', 'epoch': '1.187'}
+{'loss': '0.7066', 'grad_norm': '0.3856', 'learning_rate': '0.000192', 'epoch': '1.188'}
+{'loss': '0.7173', 'grad_norm': '0.4489', 'learning_rate': '0.000192', 'epoch': '1.189'}
+{'loss': '0.6714', 'grad_norm': '0.3853', 'learning_rate': '0.000192', 'epoch': '1.19'}
+{'loss': '0.7077', 'grad_norm': '0.4281', 'learning_rate': '0.0001919', 'epoch': '1.191'}
+{'loss': '0.7927', 'grad_norm': '0.4626', 'learning_rate': '0.0001919', 'epoch': '1.192'}
+{'loss': '0.7654', 'grad_norm': '0.3994', 'learning_rate': '0.0001919', 'epoch': '1.192'}
+{'loss': '0.7858', 'grad_norm': '0.4735', 'learning_rate': '0.0001919', 'epoch': '1.193'}
+{'loss': '0.7604', 'grad_norm': '0.4461', 'learning_rate': '0.0001919', 'epoch': '1.194'}
+{'loss': '0.7906', 'grad_norm': '0.415', 'learning_rate': '0.0001918', 'epoch': '1.195'}
+{'loss': '0.723', 'grad_norm': '0.4365', 'learning_rate': '0.0001918', 'epoch': '1.196'}
+{'loss': '0.7817', 'grad_norm': '0.4409', 'learning_rate': '0.0001918', 'epoch': '1.197'}
+{'loss': '0.6342', 'grad_norm': '0.3562', 'learning_rate': '0.0001918', 'epoch': '1.197'}
+{'loss': '0.7065', 'grad_norm': '0.4767', 'learning_rate': '0.0001918', 'epoch': '1.198'}
+{'loss': '0.7771', 'grad_norm': '0.4513', 'learning_rate': '0.0001918', 'epoch': '1.199'}
+{'loss': '0.7456', 'grad_norm': '0.4757', 'learning_rate': '0.0001917', 'epoch': '1.2'}
+{'loss': '0.8042', 'grad_norm': '0.4891', 'learning_rate': '0.0001917', 'epoch': '1.201'}
+{'loss': '0.6787', 'grad_norm': '0.4366', 'learning_rate': '0.0001917', 'epoch': '1.202'}
+{'loss': '0.7662', 'grad_norm': '0.4423', 'learning_rate': '0.0001917', 'epoch': '1.203'}
+{'loss': '0.737', 'grad_norm': '0.4844', 'learning_rate': '0.0001917', 'epoch': '1.203'}
+{'loss': '0.7777', 'grad_norm': '0.4273', 'learning_rate': '0.0001916', 'epoch': '1.204'}
+{'loss': '0.802', 'grad_norm': '0.4239', 'learning_rate': '0.0001916', 'epoch': '1.205'}
+{'loss': '0.7812', 'grad_norm': '0.4528', 'learning_rate': '0.0001916', 'epoch': '1.206'}
+{'loss': '0.7244', 'grad_norm': '0.4284', 'learning_rate': '0.0001916', 'epoch': '1.207'}
+{'loss': '0.7676', 'grad_norm': '0.4291', 'learning_rate': '0.0001916', 'epoch': '1.208'}
+{'loss': '0.7071', 'grad_norm': '0.4455', 'learning_rate': '0.0001915', 'epoch': '1.208'}
+{'loss': '0.7555', 'grad_norm': '0.4136', 'learning_rate': '0.0001915', 'epoch': '1.209'}
+{'loss': '0.7198', 'grad_norm': '0.4388', 'learning_rate': '0.0001915', 'epoch': '1.21'}
+{'loss': '0.7393', 'grad_norm': '0.4544', 'learning_rate': '0.0001915', 'epoch': '1.211'}
+{'loss': '0.7278', 'grad_norm': '0.3767', 'learning_rate': '0.0001915', 'epoch': '1.212'}
+{'loss': '0.6783', 'grad_norm': '0.3787', 'learning_rate': '0.0001914', 'epoch': '1.213'}
+{'loss': '0.7025', 'grad_norm': '0.4334', 'learning_rate': '0.0001914', 'epoch': '1.214'}
+{'loss': '0.7388', 'grad_norm': '0.4694', 'learning_rate': '0.0001914', 'epoch': '1.214'}
+{'loss': '0.7597', 'grad_norm': '0.4326', 'learning_rate': '0.0001914', 'epoch': '1.215'}
+{'loss': '0.7885', 'grad_norm': '0.4325', 'learning_rate': '0.0001914', 'epoch': '1.216'}
+{'loss': '0.6873', 'grad_norm': '0.3931', 'learning_rate': '0.0001913', 'epoch': '1.217'}
+{'loss': '0.6523', 'grad_norm': '0.4485', 'learning_rate': '0.0001913', 'epoch': '1.218'}
+{'loss': '0.7384', 'grad_norm': '0.4481', 'learning_rate': '0.0001913', 'epoch': '1.219'}
+{'loss': '0.7391', 'grad_norm': '0.4409', 'learning_rate': '0.0001913', 'epoch': '1.219'}
+{'loss': '0.7282', 'grad_norm': '0.4621', 'learning_rate': '0.0001913', 'epoch': '1.22'}
+{'loss': '0.7554', 'grad_norm': '0.4148', 'learning_rate': '0.0001912', 'epoch': '1.221'}
+{'loss': '0.6827', 'grad_norm': '0.4144', 'learning_rate': '0.0001912', 'epoch': '1.222'}
+{'loss': '0.751', 'grad_norm': '0.4539', 'learning_rate': '0.0001912', 'epoch': '1.223'}
+{'loss': '0.7599', 'grad_norm': '0.4695', 'learning_rate': '0.0001912', 'epoch': '1.224'}
+{'eval_loss': '0.8077', 'eval_runtime': '685.6', 'eval_samples_per_second': '3.073', 'eval_steps_per_second': '3.073', 'epoch': '1.224'}
+{'loss': '0.7518', 'grad_norm': '0.4317', 'learning_rate': '0.0001912', 'epoch': '1.224'}
+{'loss': '0.7308', 'grad_norm': '0.4774', 'learning_rate': '0.0001912', 'epoch': '1.225'}
+{'loss': '0.6996', 'grad_norm': '0.4091', 'learning_rate': '0.0001911', 'epoch': '1.226'}
+{'loss': '0.8364', 'grad_norm': '0.5171', 'learning_rate': '0.0001911', 'epoch': '1.227'}
+{'loss': '0.722', 'grad_norm': '0.4256', 'learning_rate': '0.0001911', 'epoch': '1.228'}
+{'loss': '0.7586', 'grad_norm': '0.4652', 'learning_rate': '0.0001911', 'epoch': '1.229'}
+{'loss': '0.7388', 'grad_norm': '0.4579', 'learning_rate': '0.0001911', 'epoch': '1.23'}
+{'loss': '0.8036', 'grad_norm': '0.4313', 'learning_rate': '0.000191', 'epoch': '1.23'}
+{'loss': '0.7357', 'grad_norm': '0.4355', 'learning_rate': '0.000191', 'epoch': '1.231'}
+{'loss': '0.673', 'grad_norm': '0.4247', 'learning_rate': '0.000191', 'epoch': '1.232'}
+{'loss': '0.6984', 'grad_norm': '0.411', 'learning_rate': '0.000191', 'epoch': '1.233'}
+{'loss': '0.6297', 'grad_norm': '0.4088', 'learning_rate': '0.000191', 'epoch': '1.234'}
+{'loss': '0.7036', 'grad_norm': '0.3686', 'learning_rate': '0.0001909', 'epoch': '1.235'}
+{'loss': '0.8209', 'grad_norm': '0.3944', 'learning_rate': '0.0001909', 'epoch': '1.235'}
+{'loss': '0.7884', 'grad_norm': '0.5042', 'learning_rate': '0.0001909', 'epoch': '1.236'}
+{'loss': '0.774', 'grad_norm': '0.4059', 'learning_rate': '0.0001909', 'epoch': '1.237'}
+{'loss': '0.6742', 'grad_norm': '0.4183', 'learning_rate': '0.0001909', 'epoch': '1.238'}
+{'loss': '0.7106', 'grad_norm': '0.3998', 'learning_rate': '0.0001908', 'epoch': '1.239'}
+{'loss': '0.7391', 'grad_norm': '0.4841', 'learning_rate': '0.0001908', 'epoch': '1.24'}
+{'loss': '0.6527', 'grad_norm': '0.3984', 'learning_rate': '0.0001908', 'epoch': '1.241'}
+{'loss': '0.694', 'grad_norm': '0.4092', 'learning_rate': '0.0001908', 'epoch': '1.241'}
+{'loss': '0.7566', 'grad_norm': '0.4329', 'learning_rate': '0.0001908', 'epoch': '1.242'}
+{'loss': '0.7894', 'grad_norm': '0.4798', 'learning_rate': '0.0001907', 'epoch': '1.243'}
+{'loss': '0.826', 'grad_norm': '0.4368', 'learning_rate': '0.0001907', 'epoch': '1.244'}
+{'loss': '0.7347', 'grad_norm': '0.4127', 'learning_rate': '0.0001907', 'epoch': '1.245'}
+{'loss': '0.7092', 'grad_norm': '0.3693', 'learning_rate': '0.0001907', 'epoch': '1.246'}
+{'loss': '0.696', 'grad_norm': '0.4395', 'learning_rate': '0.0001907', 'epoch': '1.246'}
+{'loss': '0.7817', 'grad_norm': '0.4678', 'learning_rate': '0.0001906', 'epoch': '1.247'}
+{'loss': '0.7934', 'grad_norm': '0.4192', 'learning_rate': '0.0001906', 'epoch': '1.248'}
+{'loss': '0.6857', 'grad_norm': '0.3968', 'learning_rate': '0.0001906', 'epoch': '1.249'}
+{'loss': '0.6784', 'grad_norm': '0.4035', 'learning_rate': '0.0001906', 'epoch': '1.25'}
+{'loss': '0.6997', 'grad_norm': '0.3835', 'learning_rate': '0.0001905', 'epoch': '1.251'}
+{'loss': '0.6979', 'grad_norm': '0.4402', 'learning_rate': '0.0001905', 'epoch': '1.251'}
+{'loss': '0.7008', 'grad_norm': '0.4749', 'learning_rate': '0.0001905', 'epoch': '1.252'}
+{'loss': '0.7259', 'grad_norm': '0.4011', 'learning_rate': '0.0001905', 'epoch': '1.253'}
+{'loss': '0.6746', 'grad_norm': '0.4421', 'learning_rate': '0.0001905', 'epoch': '1.254'}
+{'loss': '0.7639', 'grad_norm': '0.4532', 'learning_rate': '0.0001904', 'epoch': '1.255'}
+{'loss': '0.6596', 'grad_norm': '0.443', 'learning_rate': '0.0001904', 'epoch': '1.256'}
+{'loss': '0.7133', 'grad_norm': '0.3743', 'learning_rate': '0.0001904', 'epoch': '1.257'}
+{'loss': '0.8163', 'grad_norm': '0.4195', 'learning_rate': '0.0001904', 'epoch': '1.257'}
+{'loss': '0.7482', 'grad_norm': '0.3952', 'learning_rate': '0.0001904', 'epoch': '1.258'}
+{'loss': '0.7501', 'grad_norm': '0.4239', 'learning_rate': '0.0001903', 'epoch': '1.259'}
+{'loss': '0.7464', 'grad_norm': '0.4209', 'learning_rate': '0.0001903', 'epoch': '1.26'}
+{'loss': '0.7104', 'grad_norm': '0.4188', 'learning_rate': '0.0001903', 'epoch': '1.261'}
+{'loss': '0.6507', 'grad_norm': '0.4531', 'learning_rate': '0.0001903', 'epoch': '1.262'}
+{'loss': '0.6871', 'grad_norm': '0.4672', 'learning_rate': '0.0001903', 'epoch': '1.262'}
+{'loss': '0.6879', 'grad_norm': '0.4068', 'learning_rate': '0.0001902', 'epoch': '1.263'}
+{'loss': '0.6886', 'grad_norm': '0.4492', 'learning_rate': '0.0001902', 'epoch': '1.264'}
+{'loss': '0.7199', 'grad_norm': '0.4145', 'learning_rate': '0.0001902', 'epoch': '1.265'}
+{'loss': '0.7105', 'grad_norm': '0.4037', 'learning_rate': '0.0001902', 'epoch': '1.266'}
+{'eval_loss': '0.8033', 'eval_runtime': '678.8', 'eval_samples_per_second': '3.104', 'eval_steps_per_second': '3.104', 'epoch': '1.266'}
+{'loss': '0.7385', 'grad_norm': '0.4264', 'learning_rate': '0.0001902', 'epoch': '1.267'}
+{'loss': '0.6395', 'grad_norm': '0.371', 'learning_rate': '0.0001901', 'epoch': '1.268'}
+{'loss': '0.7124', 'grad_norm': '0.4643', 'learning_rate': '0.0001901', 'epoch': '1.268'}
+{'loss': '0.744', 'grad_norm': '0.4554', 'learning_rate': '0.0001901', 'epoch': '1.269'}
+{'loss': '0.7182', 'grad_norm': '0.3933', 'learning_rate': '0.0001901', 'epoch': '1.27'}
+{'loss': '0.7238', 'grad_norm': '0.409', 'learning_rate': '0.0001901', 'epoch': '1.271'}
+{'loss': '0.748', 'grad_norm': '0.483', 'learning_rate': '0.00019', 'epoch': '1.272'}
+{'loss': '0.8299', 'grad_norm': '0.4807', 'learning_rate': '0.00019', 'epoch': '1.273'}
+{'loss': '0.7205', 'grad_norm': '0.4239', 'learning_rate': '0.00019', 'epoch': '1.273'}
+{'loss': '0.7668', 'grad_norm': '0.4688', 'learning_rate': '0.00019', 'epoch': '1.274'}
+{'loss': '0.7663', 'grad_norm': '0.4172', 'learning_rate': '0.0001899', 'epoch': '1.275'}
+{'loss': '0.6596', 'grad_norm': '0.4132', 'learning_rate': '0.0001899', 'epoch': '1.276'}
+{'loss': '0.742', 'grad_norm': '0.4156', 'learning_rate': '0.0001899', 'epoch': '1.277'}
+{'loss': '0.6726', 'grad_norm': '0.4185', 'learning_rate': '0.0001899', 'epoch': '1.278'}
+{'loss': '0.6269', 'grad_norm': '0.3827', 'learning_rate': '0.0001899', 'epoch': '1.278'}
+{'loss': '0.7121', 'grad_norm': '0.4531', 'learning_rate': '0.0001898', 'epoch': '1.279'}
+{'loss': '0.6358', 'grad_norm': '0.3747', 'learning_rate': '0.0001898', 'epoch': '1.28'}
+{'loss': '0.7269', 'grad_norm': '0.408', 'learning_rate': '0.0001898', 'epoch': '1.281'}
+{'loss': '0.7737', 'grad_norm': '0.4194', 'learning_rate': '0.0001898', 'epoch': '1.282'}
+{'loss': '0.7458', 'grad_norm': '0.4183', 'learning_rate': '0.0001898', 'epoch': '1.283'}
+{'loss': '0.6566', 'grad_norm': '0.4334', 'learning_rate': '0.0001897', 'epoch': '1.284'}
+{'loss': '0.7369', 'grad_norm': '0.4295', 'learning_rate': '0.0001897', 'epoch': '1.284'}
+{'loss': '0.6922', 'grad_norm': '0.4133', 'learning_rate': '0.0001897', 'epoch': '1.285'}
+{'loss': '0.7004', 'grad_norm': '0.4499', 'learning_rate': '0.0001897', 'epoch': '1.286'}
+{'loss': '0.6768', 'grad_norm': '0.4375', 'learning_rate': '0.0001897', 'epoch': '1.287'}
+{'loss': '0.7597', 'grad_norm': '0.4758', 'learning_rate': '0.0001896', 'epoch': '1.288'}
+{'loss': '0.765', 'grad_norm': '0.4581', 'learning_rate': '0.0001896', 'epoch': '1.289'}
+{'loss': '0.6655', 'grad_norm': '0.4291', 'learning_rate': '0.0001896', 'epoch': '1.289'}
+{'loss': '0.708', 'grad_norm': '0.4245', 'learning_rate': '0.0001896', 'epoch': '1.29'}
+{'loss': '0.7724', 'grad_norm': '0.4682', 'learning_rate': '0.0001895', 'epoch': '1.291'}
+{'loss': '0.7746', 'grad_norm': '0.444', 'learning_rate': '0.0001895', 'epoch': '1.292'}
+{'loss': '0.7571', 'grad_norm': '0.4296', 'learning_rate': '0.0001895', 'epoch': '1.293'}
+{'loss': '0.7275', 'grad_norm': '0.4288', 'learning_rate': '0.0001895', 'epoch': '1.294'}
+{'loss': '0.6595', 'grad_norm': '0.4112', 'learning_rate': '0.0001895', 'epoch': '1.295'}
+{'loss': '0.7316', 'grad_norm': '0.4373', 'learning_rate': '0.0001894', 'epoch': '1.295'}
+{'loss': '0.7072', 'grad_norm': '0.3771', 'learning_rate': '0.0001894', 'epoch': '1.296'}
+{'loss': '0.7218', 'grad_norm': '0.4171', 'learning_rate': '0.0001894', 'epoch': '1.297'}
+{'loss': '0.7251', 'grad_norm': '0.4264', 'learning_rate': '0.0001894', 'epoch': '1.298'}
+{'loss': '0.7423', 'grad_norm': '0.4252', 'learning_rate': '0.0001894', 'epoch': '1.299'}
+{'loss': '0.7059', 'grad_norm': '0.431', 'learning_rate': '0.0001893', 'epoch': '1.3'}
+{'loss': '0.7334', 'grad_norm': '0.4668', 'learning_rate': '0.0001893', 'epoch': '1.3'}
+{'loss': '0.6033', 'grad_norm': '0.369', 'learning_rate': '0.0001893', 'epoch': '1.301'}
+{'loss': '0.6904', 'grad_norm': '0.4288', 'learning_rate': '0.0001893', 'epoch': '1.302'}
+{'loss': '0.7468', 'grad_norm': '0.4041', 'learning_rate': '0.0001892', 'epoch': '1.303'}
+{'loss': '0.7574', 'grad_norm': '0.4704', 'learning_rate': '0.0001892', 'epoch': '1.304'}
+{'loss': '0.7565', 'grad_norm': '0.4391', 'learning_rate': '0.0001892', 'epoch': '1.305'}
+{'loss': '0.6621', 'grad_norm': '0.4268', 'learning_rate': '0.0001892', 'epoch': '1.305'}
+{'loss': '0.7764', 'grad_norm': '0.4972', 'learning_rate': '0.0001892', 'epoch': '1.306'}
+{'loss': '0.6558', 'grad_norm': '0.3905', 'learning_rate': '0.0001891', 'epoch': '1.307'}
+{'loss': '0.6747', 'grad_norm': '0.4553', 'learning_rate': '0.0001891', 'epoch': '1.308'}
+{'eval_loss': '0.801', 'eval_runtime': '668.2', 'eval_samples_per_second': '3.153', 'eval_steps_per_second': '3.153', 'epoch': '1.308'}
+{'loss': '0.7916', 'grad_norm': '0.4254', 'learning_rate': '0.0001891', 'epoch': '1.309'}
+{'loss': '0.8839', 'grad_norm': '0.6119', 'learning_rate': '0.0001891', 'epoch': '1.31'}
+{'loss': '0.7146', 'grad_norm': '0.4321', 'learning_rate': '0.000189', 'epoch': '1.311'}
+{'loss': '0.7443', 'grad_norm': '0.3966', 'learning_rate': '0.000189', 'epoch': '1.311'}
+{'loss': '0.6815', 'grad_norm': '0.4088', 'learning_rate': '0.000189', 'epoch': '1.312'}
+{'loss': '0.8154', 'grad_norm': '0.4303', 'learning_rate': '0.000189', 'epoch': '1.313'}
+{'loss': '0.7828', 'grad_norm': '0.4749', 'learning_rate': '0.000189', 'epoch': '1.314'}
+{'loss': '0.7849', 'grad_norm': '0.4929', 'learning_rate': '0.0001889', 'epoch': '1.315'}
+{'loss': '0.7263', 'grad_norm': '0.3904', 'learning_rate': '0.0001889', 'epoch': '1.316'}
+{'loss': '0.8122', 'grad_norm': '0.4827', 'learning_rate': '0.0001889', 'epoch': '1.316'}
+{'loss': '0.7166', 'grad_norm': '0.4774', 'learning_rate': '0.0001889', 'epoch': '1.317'}
+{'loss': '0.6698', 'grad_norm': '0.4274', 'learning_rate': '0.0001889', 'epoch': '1.318'}
+{'loss': '0.7271', 'grad_norm': '0.4483', 'learning_rate': '0.0001888', 'epoch': '1.319'}
+{'loss': '0.7924', 'grad_norm': '0.4296', 'learning_rate': '0.0001888', 'epoch': '1.32'}
+{'loss': '0.7846', 'grad_norm': '0.4848', 'learning_rate': '0.0001888', 'epoch': '1.321'}
+{'loss': '0.7262', 'grad_norm': '0.4665', 'learning_rate': '0.0001888', 'epoch': '1.322'}
+{'loss': '0.764', 'grad_norm': '0.446', 'learning_rate': '0.0001887', 'epoch': '1.322'}
+{'loss': '0.696', 'grad_norm': '0.4374', 'learning_rate': '0.0001887', 'epoch': '1.323'}
+{'loss': '0.7121', 'grad_norm': '0.4738', 'learning_rate': '0.0001887', 'epoch': '1.324'}
+{'loss': '0.7131', 'grad_norm': '0.4445', 'learning_rate': '0.0001887', 'epoch': '1.325'}
+{'loss': '0.6905', 'grad_norm': '0.4714', 'learning_rate': '0.0001887', 'epoch': '1.326'}
+{'loss': '0.7319', 'grad_norm': '0.4057', 'learning_rate': '0.0001886', 'epoch': '1.327'}
+{'loss': '0.7268', 'grad_norm': '0.4417', 'learning_rate': '0.0001886', 'epoch': '1.327'}
+{'loss': '0.713', 'grad_norm': '0.4303', 'learning_rate': '0.0001886', 'epoch': '1.328'}
+{'loss': '0.6917', 'grad_norm': '0.4142', 'learning_rate': '0.0001886', 'epoch': '1.329'}
+{'loss': '0.7654', 'grad_norm': '0.4628', 'learning_rate': '0.0001885', 'epoch': '1.33'}
+{'loss': '0.7311', 'grad_norm': '0.3893', 'learning_rate': '0.0001885', 'epoch': '1.331'}
+{'loss': '0.7979', 'grad_norm': '0.4429', 'learning_rate': '0.0001885', 'epoch': '1.332'}
+{'loss': '0.7498', 'grad_norm': '0.3773', 'learning_rate': '0.0001885', 'epoch': '1.332'}
+{'loss': '0.7128', 'grad_norm': '0.4539', 'learning_rate': '0.0001885', 'epoch': '1.333'}
+{'loss': '0.7283', 'grad_norm': '0.4194', 'learning_rate': '0.0001884', 'epoch': '1.334'}
+{'loss': '0.7978', 'grad_norm': '0.4186', 'learning_rate': '0.0001884', 'epoch': '1.335'}
+{'loss': '0.8175', 'grad_norm': '0.4771', 'learning_rate': '0.0001884', 'epoch': '1.336'}
+{'loss': '0.6592', 'grad_norm': '0.438', 'learning_rate': '0.0001884', 'epoch': '1.337'}
+{'loss': '0.7323', 'grad_norm': '0.4505', 'learning_rate': '0.0001883', 'epoch': '1.338'}
+{'loss': '0.6791', 'grad_norm': '0.3833', 'learning_rate': '0.0001883', 'epoch': '1.338'}
+{'loss': '0.7396', 'grad_norm': '0.43', 'learning_rate': '0.0001883', 'epoch': '1.339'}
+{'loss': '0.6894', 'grad_norm': '0.3926', 'learning_rate': '0.0001883', 'epoch': '1.34'}
+{'loss': '0.6708', 'grad_norm': '0.4354', 'learning_rate': '0.0001882', 'epoch': '1.341'}
+{'loss': '0.8448', 'grad_norm': '0.4304', 'learning_rate': '0.0001882', 'epoch': '1.342'}
+{'loss': '0.7081', 'grad_norm': '0.4069', 'learning_rate': '0.0001882', 'epoch': '1.343'}
+{'loss': '0.6822', 'grad_norm': '0.4504', 'learning_rate': '0.0001882', 'epoch': '1.343'}
+{'loss': '0.6317', 'grad_norm': '0.3732', 'learning_rate': '0.0001882', 'epoch': '1.344'}
+{'loss': '0.7452', 'grad_norm': '0.4248', 'learning_rate': '0.0001881', 'epoch': '1.345'}
+{'loss': '0.6494', 'grad_norm': '0.3923', 'learning_rate': '0.0001881', 'epoch': '1.346'}
+{'loss': '0.7352', 'grad_norm': '0.4502', 'learning_rate': '0.0001881', 'epoch': '1.347'}
+{'loss': '0.6656', 'grad_norm': '0.3976', 'learning_rate': '0.0001881', 'epoch': '1.348'}
+{'loss': '0.7041', 'grad_norm': '0.441', 'learning_rate': '0.000188', 'epoch': '1.349'}
+{'loss': '0.7722', 'grad_norm': '0.4814', 'learning_rate': '0.000188', 'epoch': '1.349'}
+{'loss': '0.6682', 'grad_norm': '0.3864', 'learning_rate': '0.000188', 'epoch': '1.35'}
+{'eval_loss': '0.7972', 'eval_runtime': '681', 'eval_samples_per_second': '3.094', 'eval_steps_per_second': '3.094', 'epoch': '1.35'}
+{'loss': '0.6996', 'grad_norm': '0.4471', 'learning_rate': '0.000188', 'epoch': '1.351'}
+{'loss': '0.6596', 'grad_norm': '0.4065', 'learning_rate': '0.000188', 'epoch': '1.352'}
+{'loss': '0.7927', 'grad_norm': '0.4788', 'learning_rate': '0.0001879', 'epoch': '1.353'}
+{'loss': '0.6739', 'grad_norm': '0.4335', 'learning_rate': '0.0001879', 'epoch': '1.354'}
+{'loss': '0.7785', 'grad_norm': '0.4276', 'learning_rate': '0.0001879', 'epoch': '1.354'}
+{'loss': '0.7325', 'grad_norm': '0.4506', 'learning_rate': '0.0001879', 'epoch': '1.355'}
+{'loss': '0.6899', 'grad_norm': '0.4204', 'learning_rate': '0.0001878', 'epoch': '1.356'}
+{'loss': '0.7499', 'grad_norm': '0.465', 'learning_rate': '0.0001878', 'epoch': '1.357'}
+{'loss': '0.6892', 'grad_norm': '0.4194', 'learning_rate': '0.0001878', 'epoch': '1.358'}
+{'loss': '0.7538', 'grad_norm': '0.4448', 'learning_rate': '0.0001878', 'epoch': '1.359'}
+{'loss': '0.6476', 'grad_norm': '0.4047', 'learning_rate': '0.0001877', 'epoch': '1.359'}
+{'loss': '0.6766', 'grad_norm': '0.3982', 'learning_rate': '0.0001877', 'epoch': '1.36'}
+{'loss': '0.736', 'grad_norm': '0.4184', 'learning_rate': '0.0001877', 'epoch': '1.361'}
+{'loss': '0.715', 'grad_norm': '0.414', 'learning_rate': '0.0001877', 'epoch': '1.362'}
+{'loss': '0.6933', 'grad_norm': '0.4359', 'learning_rate': '0.0001877', 'epoch': '1.363'}
+{'loss': '0.6561', 'grad_norm': '0.3651', 'learning_rate': '0.0001876', 'epoch': '1.364'}
+{'loss': '0.7437', 'grad_norm': '0.4494', 'learning_rate': '0.0001876', 'epoch': '1.365'}
+{'loss': '0.785', 'grad_norm': '0.5046', 'learning_rate': '0.0001876', 'epoch': '1.365'}
+{'loss': '0.7502', 'grad_norm': '0.4626', 'learning_rate': '0.0001876', 'epoch': '1.366'}
+{'loss': '0.7656', 'grad_norm': '0.3721', 'learning_rate': '0.0001875', 'epoch': '1.367'}
+{'loss': '0.7816', 'grad_norm': '0.4172', 'learning_rate': '0.0001875', 'epoch': '1.368'}
+{'loss': '0.7003', 'grad_norm': '0.8452', 'learning_rate': '0.0001875', 'epoch': '1.369'}
+{'loss': '0.7358', 'grad_norm': '0.4321', 'learning_rate': '0.0001875', 'epoch': '1.37'}
+{'loss': '0.6529', 'grad_norm': '0.4332', 'learning_rate': '0.0001874', 'epoch': '1.37'}
+{'loss': '0.7318', 'grad_norm': '0.4149', 'learning_rate': '0.0001874', 'epoch': '1.371'}
+{'loss': '0.8237', 'grad_norm': '0.5085', 'learning_rate': '0.0001874', 'epoch': '1.372'}
+{'loss': '0.7303', 'grad_norm': '0.4127', 'learning_rate': '0.0001874', 'epoch': '1.373'}
+{'loss': '0.7638', 'grad_norm': '0.5272', 'learning_rate': '0.0001874', 'epoch': '1.374'}
+{'loss': '0.7907', 'grad_norm': '0.432', 'learning_rate': '0.0001873', 'epoch': '1.375'}
+{'loss': '0.7262', 'grad_norm': '0.4393', 'learning_rate': '0.0001873', 'epoch': '1.376'}
+{'loss': '0.6857', 'grad_norm': '0.4162', 'learning_rate': '0.0001873', 'epoch': '1.376'}
+{'loss': '0.7489', 'grad_norm': '0.4886', 'learning_rate': '0.0001873', 'epoch': '1.377'}
+{'loss': '0.7258', 'grad_norm': '0.4422', 'learning_rate': '0.0001872', 'epoch': '1.378'}
+{'loss': '0.7467', 'grad_norm': '0.4577', 'learning_rate': '0.0001872', 'epoch': '1.379'}
+{'loss': '0.7685', 'grad_norm': '0.4238', 'learning_rate': '0.0001872', 'epoch': '1.38'}
+{'loss': '0.7426', 'grad_norm': '0.4337', 'learning_rate': '0.0001872', 'epoch': '1.381'}
+{'loss': '0.7104', 'grad_norm': '0.3777', 'learning_rate': '0.0001871', 'epoch': '1.381'}
+{'loss': '0.7503', 'grad_norm': '0.4607', 'learning_rate': '0.0001871', 'epoch': '1.382'}
+{'loss': '0.6401', 'grad_norm': '0.4251', 'learning_rate': '0.0001871', 'epoch': '1.383'}
+{'loss': '0.6649', 'grad_norm': '0.4217', 'learning_rate': '0.0001871', 'epoch': '1.384'}
+{'loss': '0.7176', 'grad_norm': '0.4827', 'learning_rate': '0.000187', 'epoch': '1.385'}
+{'loss': '0.8402', 'grad_norm': '0.4711', 'learning_rate': '0.000187', 'epoch': '1.386'}
+{'loss': '0.8131', 'grad_norm': '0.5054', 'learning_rate': '0.000187', 'epoch': '1.386'}
+{'loss': '0.6433', 'grad_norm': '0.3944', 'learning_rate': '0.000187', 'epoch': '1.387'}
+{'loss': '0.6922', 'grad_norm': '0.3997', 'learning_rate': '0.000187', 'epoch': '1.388'}
+{'loss': '0.7162', 'grad_norm': '0.373', 'learning_rate': '0.0001869', 'epoch': '1.389'}
+{'loss': '0.7195', 'grad_norm': '0.7998', 'learning_rate': '0.0001869', 'epoch': '1.39'}
+{'loss': '0.6888', 'grad_norm': '0.4127', 'learning_rate': '0.0001869', 'epoch': '1.391'}
+{'loss': '0.7387', 'grad_norm': '0.4486', 'learning_rate': '0.0001869', 'epoch': '1.392'}
+{'loss': '0.8062', 'grad_norm': '0.4784', 'learning_rate': '0.0001868', 'epoch': '1.392'}
+{'eval_loss': '0.7956', 'eval_runtime': '692.7', 'eval_samples_per_second': '3.042', 'eval_steps_per_second': '3.042', 'epoch': '1.392'}
+{'loss': '0.7318', 'grad_norm': '0.4024', 'learning_rate': '0.0001868', 'epoch': '1.393'}
+{'loss': '0.7471', 'grad_norm': '0.4526', 'learning_rate': '0.0001868', 'epoch': '1.394'}
+{'loss': '0.7628', 'grad_norm': '0.5055', 'learning_rate': '0.0001868', 'epoch': '1.395'}
+{'loss': '0.684', 'grad_norm': '0.403', 'learning_rate': '0.0001867', 'epoch': '1.396'}
+{'loss': '0.7367', 'grad_norm': '0.4353', 'learning_rate': '0.0001867', 'epoch': '1.397'}
+{'loss': '0.8051', 'grad_norm': '0.5361', 'learning_rate': '0.0001867', 'epoch': '1.397'}
+{'loss': '0.7228', 'grad_norm': '0.4608', 'learning_rate': '0.0001867', 'epoch': '1.398'}
+{'loss': '0.7142', 'grad_norm': '0.4326', 'learning_rate': '0.0001866', 'epoch': '1.399'}
+{'loss': '0.7379', 'grad_norm': '0.4373', 'learning_rate': '0.0001866', 'epoch': '1.4'}
+{'loss': '0.7566', 'grad_norm': '0.4322', 'learning_rate': '0.0001866', 'epoch': '1.401'}
+{'loss': '0.6409', 'grad_norm': '0.3938', 'learning_rate': '0.0001866', 'epoch': '1.402'}
+{'loss': '0.6912', 'grad_norm': '0.4361', 'learning_rate': '0.0001865', 'epoch': '1.403'}
+{'loss': '0.7276', 'grad_norm': '0.4526', 'learning_rate': '0.0001865', 'epoch': '1.403'}
+{'loss': '0.6467', 'grad_norm': '0.3991', 'learning_rate': '0.0001865', 'epoch': '1.404'}
+{'loss': '0.6521', 'grad_norm': '0.4432', 'learning_rate': '0.0001865', 'epoch': '1.405'}
+{'loss': '0.729', 'grad_norm': '0.4473', 'learning_rate': '0.0001865', 'epoch': '1.406'}
+{'loss': '0.7664', 'grad_norm': '0.4569', 'learning_rate': '0.0001864', 'epoch': '1.407'}
+{'loss': '0.7747', 'grad_norm': '0.4172', 'learning_rate': '0.0001864', 'epoch': '1.408'}
+{'loss': '0.8338', 'grad_norm': '0.4541', 'learning_rate': '0.0001864', 'epoch': '1.408'}
+{'loss': '0.6852', 'grad_norm': '0.4306', 'learning_rate': '0.0001864', 'epoch': '1.409'}
+{'loss': '0.6841', 'grad_norm': '0.4562', 'learning_rate': '0.0001863', 'epoch': '1.41'}
+{'loss': '0.6672', 'grad_norm': '0.353', 'learning_rate': '0.0001863', 'epoch': '1.411'}
+{'loss': '0.7027', 'grad_norm': '0.4425', 'learning_rate': '0.0001863', 'epoch': '1.412'}
+{'loss': '0.7398', 'grad_norm': '0.4298', 'learning_rate': '0.0001863', 'epoch': '1.413'}
+{'loss': '0.8009', 'grad_norm': '0.472', 'learning_rate': '0.0001862', 'epoch': '1.414'}
+{'loss': '0.6875', 'grad_norm': '0.422', 'learning_rate': '0.0001862', 'epoch': '1.414'}
+{'loss': '0.7202', 'grad_norm': '0.3674', 'learning_rate': '0.0001862', 'epoch': '1.415'}
+{'loss': '0.6981', 'grad_norm': '0.4233', 'learning_rate': '0.0001862', 'epoch': '1.416'}
+{'loss': '0.7081', 'grad_norm': '0.4064', 'learning_rate': '0.0001861', 'epoch': '1.417'}
+{'loss': '0.642', 'grad_norm': '0.3948', 'learning_rate': '0.0001861', 'epoch': '1.418'}
+{'loss': '0.7662', 'grad_norm': '0.4462', 'learning_rate': '0.0001861', 'epoch': '1.419'}
+{'loss': '0.7244', 'grad_norm': '0.3977', 'learning_rate': '0.0001861', 'epoch': '1.419'}
+{'loss': '0.7399', 'grad_norm': '0.4284', 'learning_rate': '0.000186', 'epoch': '1.42'}
+{'loss': '0.6979', 'grad_norm': '0.3749', 'learning_rate': '0.000186', 'epoch': '1.421'}
+{'loss': '0.7733', 'grad_norm': '0.4672', 'learning_rate': '0.000186', 'epoch': '1.422'}
+{'loss': '0.6722', 'grad_norm': '0.3961', 'learning_rate': '0.000186', 'epoch': '1.423'}
+{'loss': '0.7158', 'grad_norm': '0.388', 'learning_rate': '0.0001859', 'epoch': '1.424'}
+{'loss': '0.6227', 'grad_norm': '0.3642', 'learning_rate': '0.0001859', 'epoch': '1.424'}
+{'loss': '0.7972', 'grad_norm': '0.4613', 'learning_rate': '0.0001859', 'epoch': '1.425'}
+{'loss': '0.6664', 'grad_norm': '0.4406', 'learning_rate': '0.0001859', 'epoch': '1.426'}
+{'loss': '0.6867', 'grad_norm': '0.4202', 'learning_rate': '0.0001858', 'epoch': '1.427'}
+{'loss': '0.6669', 'grad_norm': '0.45', 'learning_rate': '0.0001858', 'epoch': '1.428'}
+{'loss': '0.6715', 'grad_norm': '0.4251', 'learning_rate': '0.0001858', 'epoch': '1.429'}
+{'loss': '0.7219', 'grad_norm': '0.4386', 'learning_rate': '0.0001858', 'epoch': '1.43'}
+{'loss': '0.7158', 'grad_norm': '0.5167', 'learning_rate': '0.0001857', 'epoch': '1.43'}
+{'loss': '0.751', 'grad_norm': '0.4443', 'learning_rate': '0.0001857', 'epoch': '1.431'}
+{'loss': '0.756', 'grad_norm': '0.4132', 'learning_rate': '0.0001857', 'epoch': '1.432'}
+{'loss': '0.7048', 'grad_norm': '0.4094', 'learning_rate': '0.0001857', 'epoch': '1.433'}
+{'loss': '0.7621', 'grad_norm': '0.4248', 'learning_rate': '0.0001856', 'epoch': '1.434'}
+{'loss': '0.7442', 'grad_norm': '0.4191', 'learning_rate': '0.0001856', 'epoch': '1.435'}
+{'eval_loss': '0.7918', 'eval_runtime': '686.2', 'eval_samples_per_second': '3.071', 'eval_steps_per_second': '3.071', 'epoch': '1.435'}
+{'loss': '0.7123', 'grad_norm': '0.456', 'learning_rate': '0.0001856', 'epoch': '1.435'}
+{'loss': '0.7097', 'grad_norm': '0.5209', 'learning_rate': '0.0001856', 'epoch': '1.436'}
+{'loss': '0.7117', 'grad_norm': '0.4452', 'learning_rate': '0.0001855', 'epoch': '1.437'}
+{'loss': '0.7756', 'grad_norm': '0.4457', 'learning_rate': '0.0001855', 'epoch': '1.438'}
+{'loss': '0.6896', 'grad_norm': '0.4031', 'learning_rate': '0.0001855', 'epoch': '1.439'}
+{'loss': '0.7174', 'grad_norm': '0.4026', 'learning_rate': '0.0001855', 'epoch': '1.44'}
+{'loss': '0.7212', 'grad_norm': '0.4377', 'learning_rate': '0.0001854', 'epoch': '1.441'}
+{'loss': '0.8261', 'grad_norm': '0.4469', 'learning_rate': '0.0001854', 'epoch': '1.441'}
+{'loss': '0.7649', 'grad_norm': '0.4803', 'learning_rate': '0.0001854', 'epoch': '1.442'}
+{'loss': '0.7436', 'grad_norm': '0.4428', 'learning_rate': '0.0001854', 'epoch': '1.443'}
+{'loss': '0.7696', 'grad_norm': '0.4181', 'learning_rate': '0.0001853', 'epoch': '1.444'}
+{'loss': '0.7431', 'grad_norm': '0.4243', 'learning_rate': '0.0001853', 'epoch': '1.445'}
+{'loss': '0.7016', 'grad_norm': '0.4439', 'learning_rate': '0.0001853', 'epoch': '1.446'}
+{'loss': '0.7073', 'grad_norm': '0.5105', 'learning_rate': '0.0001853', 'epoch': '1.446'}
+{'loss': '0.7479', 'grad_norm': '0.4019', 'learning_rate': '0.0001852', 'epoch': '1.447'}
+{'loss': '0.6987', 'grad_norm': '0.4794', 'learning_rate': '0.0001852', 'epoch': '1.448'}
+{'loss': '0.6677', 'grad_norm': '0.423', 'learning_rate': '0.0001852', 'epoch': '1.449'}
+{'loss': '0.5727', 'grad_norm': '0.3722', 'learning_rate': '0.0001852', 'epoch': '1.45'}
+{'loss': '0.7245', 'grad_norm': '0.4991', 'learning_rate': '0.0001851', 'epoch': '1.451'}
+{'loss': '0.5908', 'grad_norm': '0.4367', 'learning_rate': '0.0001851', 'epoch': '1.451'}
+{'loss': '0.6898', 'grad_norm': '0.4355', 'learning_rate': '0.0001851', 'epoch': '1.452'}
+{'loss': '0.6243', 'grad_norm': '0.4056', 'learning_rate': '0.0001851', 'epoch': '1.453'}
+{'loss': '0.7215', 'grad_norm': '0.426', 'learning_rate': '0.000185', 'epoch': '1.454'}
+{'loss': '0.7121', 'grad_norm': '0.4123', 'learning_rate': '0.000185', 'epoch': '1.455'}
+{'loss': '0.7376', 'grad_norm': '0.453', 'learning_rate': '0.000185', 'epoch': '1.456'}
+{'loss': '0.7321', 'grad_norm': '0.4261', 'learning_rate': '0.000185', 'epoch': '1.457'}
+{'loss': '0.7099', 'grad_norm': '0.4633', 'learning_rate': '0.0001849', 'epoch': '1.457'}
+{'loss': '0.6664', 'grad_norm': '0.4332', 'learning_rate': '0.0001849', 'epoch': '1.458'}
+{'loss': '0.6635', 'grad_norm': '0.4217', 'learning_rate': '0.0001849', 'epoch': '1.459'}
+{'loss': '0.6971', 'grad_norm': '0.4437', 'learning_rate': '0.0001849', 'epoch': '1.46'}
+{'loss': '0.801', 'grad_norm': '0.4893', 'learning_rate': '0.0001848', 'epoch': '1.461'}
+{'loss': '0.6315', 'grad_norm': '0.423', 'learning_rate': '0.0001848', 'epoch': '1.462'}
+{'loss': '0.8046', 'grad_norm': '0.4949', 'learning_rate': '0.0001848', 'epoch': '1.462'}
+{'loss': '0.7299', 'grad_norm': '0.5035', 'learning_rate': '0.0001848', 'epoch': '1.463'}
+{'loss': '0.7313', 'grad_norm': '0.4505', 'learning_rate': '0.0001847', 'epoch': '1.464'}
+{'loss': '0.7516', 'grad_norm': '0.4141', 'learning_rate': '0.0001847', 'epoch': '1.465'}
+{'loss': '0.7426', 'grad_norm': '0.4441', 'learning_rate': '0.0001847', 'epoch': '1.466'}
+{'loss': '0.7509', 'grad_norm': '0.475', 'learning_rate': '0.0001847', 'epoch': '1.467'}
+{'loss': '0.7396', 'grad_norm': '0.499', 'learning_rate': '0.0001846', 'epoch': '1.468'}
+{'loss': '0.786', 'grad_norm': '0.46', 'learning_rate': '0.0001846', 'epoch': '1.468'}
+{'loss': '0.8194', 'grad_norm': '0.599', 'learning_rate': '0.0001846', 'epoch': '1.469'}
+{'loss': '0.6569', 'grad_norm': '0.3904', 'learning_rate': '0.0001846', 'epoch': '1.47'}
+{'loss': '0.7227', 'grad_norm': '0.4316', 'learning_rate': '0.0001845', 'epoch': '1.471'}
+{'loss': '0.8134', 'grad_norm': '0.4448', 'learning_rate': '0.0001845', 'epoch': '1.472'}
+{'loss': '0.7669', 'grad_norm': '0.4252', 'learning_rate': '0.0001845', 'epoch': '1.473'}
+{'loss': '0.729', 'grad_norm': '0.3988', 'learning_rate': '0.0001845', 'epoch': '1.473'}
+{'loss': '0.7612', 'grad_norm': '0.4419', 'learning_rate': '0.0001844', 'epoch': '1.474'}
+{'loss': '0.6926', 'grad_norm': '0.3746', 'learning_rate': '0.0001844', 'epoch': '1.475'}
+{'loss': '0.7889', 'grad_norm': '0.4771', 'learning_rate': '0.0001844', 'epoch': '1.476'}
+{'loss': '0.7232', 'grad_norm': '0.449', 'learning_rate': '0.0001844', 'epoch': '1.477'}
+{'eval_loss': '0.7903', 'eval_runtime': '672.3', 'eval_samples_per_second': '3.134', 'eval_steps_per_second': '3.134', 'epoch': '1.477'}
+{'loss': '0.7465', 'grad_norm': '0.4409', 'learning_rate': '0.0001843', 'epoch': '1.478'}
+{'loss': '0.7312', 'grad_norm': '0.4244', 'learning_rate': '0.0001843', 'epoch': '1.478'}
+{'loss': '0.717', 'grad_norm': '0.4438', 'learning_rate': '0.0001843', 'epoch': '1.479'}
+{'loss': '0.617', 'grad_norm': '0.3639', 'learning_rate': '0.0001842', 'epoch': '1.48'}
+{'loss': '0.7035', 'grad_norm': '0.4453', 'learning_rate': '0.0001842', 'epoch': '1.481'}
+{'loss': '0.7536', 'grad_norm': '0.523', 'learning_rate': '0.0001842', 'epoch': '1.482'}
+{'loss': '0.6859', 'grad_norm': '0.4075', 'learning_rate': '0.0001842', 'epoch': '1.483'}
+{'loss': '0.712', 'grad_norm': '0.4533', 'learning_rate': '0.0001841', 'epoch': '1.484'}
+{'loss': '0.6837', 'grad_norm': '0.4281', 'learning_rate': '0.0001841', 'epoch': '1.484'}
+{'loss': '0.664', 'grad_norm': '0.4891', 'learning_rate': '0.0001841', 'epoch': '1.485'}
+{'loss': '0.6935', 'grad_norm': '0.4436', 'learning_rate': '0.0001841', 'epoch': '1.486'}
+{'loss': '0.7046', 'grad_norm': '0.4013', 'learning_rate': '0.000184', 'epoch': '1.487'}
+{'loss': '0.6992', 'grad_norm': '0.4602', 'learning_rate': '0.000184', 'epoch': '1.488'}
+{'loss': '0.667', 'grad_norm': '0.3911', 'learning_rate': '0.000184', 'epoch': '1.489'}
+{'loss': '0.7187', 'grad_norm': '0.4498', 'learning_rate': '0.000184', 'epoch': '1.489'}
+{'loss': '0.6854', 'grad_norm': '0.4177', 'learning_rate': '0.0001839', 'epoch': '1.49'}
+{'loss': '0.6643', 'grad_norm': '0.4017', 'learning_rate': '0.0001839', 'epoch': '1.491'}
+{'loss': '0.6822', 'grad_norm': '0.437', 'learning_rate': '0.0001839', 'epoch': '1.492'}
+{'loss': '0.7501', 'grad_norm': '0.44', 'learning_rate': '0.0001839', 'epoch': '1.493'}
+{'loss': '0.7742', 'grad_norm': '0.5163', 'learning_rate': '0.0001838', 'epoch': '1.494'}
+{'loss': '0.7334', 'grad_norm': '0.4235', 'learning_rate': '0.0001838', 'epoch': '1.495'}
+{'loss': '0.6913', 'grad_norm': '0.4056', 'learning_rate': '0.0001838', 'epoch': '1.495'}
+{'loss': '0.6478', 'grad_norm': '0.3652', 'learning_rate': '0.0001838', 'epoch': '1.496'}
+{'loss': '0.68', 'grad_norm': '0.4543', 'learning_rate': '0.0001837', 'epoch': '1.497'}
+{'loss': '0.7387', 'grad_norm': '0.4259', 'learning_rate': '0.0001837', 'epoch': '1.498'}
+{'loss': '0.7573', 'grad_norm': '0.4261', 'learning_rate': '0.0001837', 'epoch': '1.499'}
+{'loss': '0.7254', 'grad_norm': '0.4037', 'learning_rate': '0.0001836', 'epoch': '1.5'}
+{'loss': '0.7342', 'grad_norm': '0.4212', 'learning_rate': '0.0001836', 'epoch': '1.5'}
+{'loss': '0.7147', 'grad_norm': '0.4902', 'learning_rate': '0.0001836', 'epoch': '1.501'}
+{'loss': '0.7474', 'grad_norm': '0.3923', 'learning_rate': '0.0001836', 'epoch': '1.502'}
+{'loss': '0.7027', 'grad_norm': '0.4203', 'learning_rate': '0.0001835', 'epoch': '1.503'}
+{'loss': '0.7427', 'grad_norm': '0.4436', 'learning_rate': '0.0001835', 'epoch': '1.504'}
+{'loss': '0.6801', 'grad_norm': '0.448', 'learning_rate': '0.0001835', 'epoch': '1.505'}
+{'loss': '0.6681', 'grad_norm': '0.439', 'learning_rate': '0.0001835', 'epoch': '1.505'}
+{'loss': '0.732', 'grad_norm': '0.4479', 'learning_rate': '0.0001834', 'epoch': '1.506'}
+{'loss': '0.7206', 'grad_norm': '0.4445', 'learning_rate': '0.0001834', 'epoch': '1.507'}
+{'loss': '0.8053', 'grad_norm': '0.4462', 'learning_rate': '0.0001834', 'epoch': '1.508'}
+{'loss': '0.6945', 'grad_norm': '0.4152', 'learning_rate': '0.0001834', 'epoch': '1.509'}
+{'loss': '0.7334', 'grad_norm': '0.4835', 'learning_rate': '0.0001833', 'epoch': '1.51'}
+{'loss': '0.7037', 'grad_norm': '0.4445', 'learning_rate': '0.0001833', 'epoch': '1.511'}
+{'loss': '0.6589', 'grad_norm': '0.3559', 'learning_rate': '0.0001833', 'epoch': '1.511'}
+{'loss': '0.7285', 'grad_norm': '0.4532', 'learning_rate': '0.0001833', 'epoch': '1.512'}
+{'loss': '0.7254', 'grad_norm': '0.4591', 'learning_rate': '0.0001832', 'epoch': '1.513'}
+{'loss': '0.7626', 'grad_norm': '0.4729', 'learning_rate': '0.0001832', 'epoch': '1.514'}
+{'loss': '0.7055', 'grad_norm': '0.4207', 'learning_rate': '0.0001832', 'epoch': '1.515'}
+{'loss': '0.6901', 'grad_norm': '0.4028', 'learning_rate': '0.0001831', 'epoch': '1.516'}
+{'loss': '0.6952', 'grad_norm': '0.4356', 'learning_rate': '0.0001831', 'epoch': '1.516'}
+{'loss': '0.7454', 'grad_norm': '0.4007', 'learning_rate': '0.0001831', 'epoch': '1.517'}
+{'loss': '0.7549', 'grad_norm': '0.4686', 'learning_rate': '0.0001831', 'epoch': '1.518'}
+{'loss': '0.7709', 'grad_norm': '0.4731', 'learning_rate': '0.000183', 'epoch': '1.519'}
+{'eval_loss': '0.7855', 'eval_runtime': '675.2', 'eval_samples_per_second': '3.12', 'eval_steps_per_second': '3.12', 'epoch': '1.519'}
+{'loss': '0.7501', 'grad_norm': '0.4343', 'learning_rate': '0.000183', 'epoch': '1.52'}
+{'loss': '0.7358', 'grad_norm': '0.4971', 'learning_rate': '0.000183', 'epoch': '1.521'}
+{'loss': '0.7973', 'grad_norm': '0.4201', 'learning_rate': '0.000183', 'epoch': '1.522'}
+{'loss': '0.5909', 'grad_norm': '0.4483', 'learning_rate': '0.0001829', 'epoch': '1.522'}
+{'loss': '0.7873', 'grad_norm': '0.4741', 'learning_rate': '0.0001829', 'epoch': '1.523'}
+{'loss': '0.7328', 'grad_norm': '0.4224', 'learning_rate': '0.0001829', 'epoch': '1.524'}
+{'loss': '0.665', 'grad_norm': '0.4055', 'learning_rate': '0.0001829', 'epoch': '1.525'}
+{'loss': '0.6939', 'grad_norm': '0.4058', 'learning_rate': '0.0001828', 'epoch': '1.526'}
+{'loss': '0.6351', 'grad_norm': '0.4192', 'learning_rate': '0.0001828', 'epoch': '1.527'}
+{'loss': '0.6963', 'grad_norm': '0.466', 'learning_rate': '0.0001828', 'epoch': '1.527'}
+{'loss': '0.7463', 'grad_norm': '0.4993', 'learning_rate': '0.0001827', 'epoch': '1.528'}
+{'loss': '0.6566', 'grad_norm': '0.3923', 'learning_rate': '0.0001827', 'epoch': '1.529'}
+{'loss': '0.7695', 'grad_norm': '0.4482', 'learning_rate': '0.0001827', 'epoch': '1.53'}
+{'loss': '0.7357', 'grad_norm': '0.4', 'learning_rate': '0.0001827', 'epoch': '1.531'}
+{'loss': '0.7234', 'grad_norm': '0.422', 'learning_rate': '0.0001826', 'epoch': '1.532'}
+{'loss': '0.7864', 'grad_norm': '0.4787', 'learning_rate': '0.0001826', 'epoch': '1.532'}
+{'loss': '0.6705', 'grad_norm': '0.3671', 'learning_rate': '0.0001826', 'epoch': '1.533'}
+{'loss': '0.7536', 'grad_norm': '0.416', 'learning_rate': '0.0001826', 'epoch': '1.534'}
+{'loss': '0.7554', 'grad_norm': '0.5033', 'learning_rate': '0.0001825', 'epoch': '1.535'}
+{'loss': '0.7149', 'grad_norm': '0.4236', 'learning_rate': '0.0001825', 'epoch': '1.536'}
+{'loss': '0.7356', 'grad_norm': '0.3799', 'learning_rate': '0.0001825', 'epoch': '1.537'}
+{'loss': '0.6935', 'grad_norm': '0.3953', 'learning_rate': '0.0001824', 'epoch': '1.538'}
+{'loss': '0.7262', 'grad_norm': '0.4545', 'learning_rate': '0.0001824', 'epoch': '1.538'}
+{'loss': '0.7519', 'grad_norm': '0.4131', 'learning_rate': '0.0001824', 'epoch': '1.539'}
+{'loss': '0.7651', 'grad_norm': '0.466', 'learning_rate': '0.0001824', 'epoch': '1.54'}
+{'loss': '0.6436', 'grad_norm': '0.3999', 'learning_rate': '0.0001823', 'epoch': '1.541'}
+{'loss': '0.7068', 'grad_norm': '0.4315', 'learning_rate': '0.0001823', 'epoch': '1.542'}
+{'loss': '0.7343', 'grad_norm': '0.4325', 'learning_rate': '0.0001823', 'epoch': '1.543'}
+{'loss': '0.6483', 'grad_norm': '0.4385', 'learning_rate': '0.0001823', 'epoch': '1.543'}
+{'loss': '0.632', 'grad_norm': '0.3737', 'learning_rate': '0.0001822', 'epoch': '1.544'}
+{'loss': '0.6603', 'grad_norm': '0.4211', 'learning_rate': '0.0001822', 'epoch': '1.545'}
+{'loss': '0.6875', 'grad_norm': '0.4222', 'learning_rate': '0.0001822', 'epoch': '1.546'}
+{'loss': '0.7232', 'grad_norm': '0.4272', 'learning_rate': '0.0001821', 'epoch': '1.547'}
+{'loss': '0.6418', 'grad_norm': '0.3748', 'learning_rate': '0.0001821', 'epoch': '1.548'}
+{'loss': '0.702', 'grad_norm': '0.3701', 'learning_rate': '0.0001821', 'epoch': '1.549'}
+{'loss': '0.6633', 'grad_norm': '0.4097', 'learning_rate': '0.0001821', 'epoch': '1.549'}
+{'loss': '0.7327', 'grad_norm': '0.4957', 'learning_rate': '0.000182', 'epoch': '1.55'}
+{'loss': '0.6535', 'grad_norm': '0.455', 'learning_rate': '0.000182', 'epoch': '1.551'}
+{'loss': '0.7761', 'grad_norm': '0.4262', 'learning_rate': '0.000182', 'epoch': '1.552'}
+{'loss': '0.6729', 'grad_norm': '0.4333', 'learning_rate': '0.000182', 'epoch': '1.553'}
+{'loss': '0.7369', 'grad_norm': '0.4973', 'learning_rate': '0.0001819', 'epoch': '1.554'}
+{'loss': '0.6704', 'grad_norm': '0.362', 'learning_rate': '0.0001819', 'epoch': '1.554'}
+{'loss': '0.7184', 'grad_norm': '0.4644', 'learning_rate': '0.0001819', 'epoch': '1.555'}
+{'loss': '0.68', 'grad_norm': '0.4605', 'learning_rate': '0.0001818', 'epoch': '1.556'}
+{'loss': '0.6798', 'grad_norm': '0.4023', 'learning_rate': '0.0001818', 'epoch': '1.557'}
+{'loss': '0.7409', 'grad_norm': '0.4595', 'learning_rate': '0.0001818', 'epoch': '1.558'}
+{'loss': '0.6927', 'grad_norm': '0.4397', 'learning_rate': '0.0001818', 'epoch': '1.559'}
+{'loss': '0.7527', 'grad_norm': '0.4476', 'learning_rate': '0.0001817', 'epoch': '1.559'}
+{'loss': '0.7218', 'grad_norm': '0.4368', 'learning_rate': '0.0001817', 'epoch': '1.56'}
+{'loss': '0.7265', 'grad_norm': '0.4118', 'learning_rate': '0.0001817', 'epoch': '1.561'}
+{'eval_loss': '0.7854', 'eval_runtime': '687.8', 'eval_samples_per_second': '3.063', 'eval_steps_per_second': '3.063', 'epoch': '1.561'}
+{'loss': '0.7015', 'grad_norm': '0.4064', 'learning_rate': '0.0001816', 'epoch': '1.562'}
+{'loss': '0.6967', 'grad_norm': '0.4206', 'learning_rate': '0.0001816', 'epoch': '1.563'}
+{'loss': '0.713', 'grad_norm': '0.4644', 'learning_rate': '0.0001816', 'epoch': '1.564'}
+{'loss': '0.718', 'grad_norm': '0.4413', 'learning_rate': '0.0001816', 'epoch': '1.565'}
+{'loss': '0.6883', 'grad_norm': '0.3964', 'learning_rate': '0.0001815', 'epoch': '1.565'}
+{'loss': '0.6683', 'grad_norm': '0.3449', 'learning_rate': '0.0001815', 'epoch': '1.566'}
+{'loss': '0.697', 'grad_norm': '0.4209', 'learning_rate': '0.0001815', 'epoch': '1.567'}
+{'loss': '0.7143', 'grad_norm': '0.4167', 'learning_rate': '0.0001815', 'epoch': '1.568'}
+{'loss': '0.7125', 'grad_norm': '0.4366', 'learning_rate': '0.0001814', 'epoch': '1.569'}
+{'loss': '0.6787', 'grad_norm': '0.436', 'learning_rate': '0.0001814', 'epoch': '1.57'}
+{'loss': '0.6401', 'grad_norm': '0.3632', 'learning_rate': '0.0001814', 'epoch': '1.57'}
+{'loss': '0.6875', 'grad_norm': '0.435', 'learning_rate': '0.0001813', 'epoch': '1.571'}
+{'loss': '0.6904', 'grad_norm': '0.4767', 'learning_rate': '0.0001813', 'epoch': '1.572'}
+{'loss': '0.69', 'grad_norm': '0.4466', 'learning_rate': '0.0001813', 'epoch': '1.573'}
+{'loss': '0.6912', 'grad_norm': '0.4382', 'learning_rate': '0.0001813', 'epoch': '1.574'}
+{'loss': '0.7415', 'grad_norm': '0.4055', 'learning_rate': '0.0001812', 'epoch': '1.575'}
+{'loss': '0.7665', 'grad_norm': '0.4253', 'learning_rate': '0.0001812', 'epoch': '1.576'}
+{'loss': '0.7828', 'grad_norm': '0.5035', 'learning_rate': '0.0001812', 'epoch': '1.576'}
+{'loss': '0.8065', 'grad_norm': '0.526', 'learning_rate': '0.0001811', 'epoch': '1.577'}
+{'loss': '0.686', 'grad_norm': '0.3641', 'learning_rate': '0.0001811', 'epoch': '1.578'}
+{'loss': '0.7799', 'grad_norm': '0.4469', 'learning_rate': '0.0001811', 'epoch': '1.579'}
+{'loss': '0.7086', 'grad_norm': '0.4078', 'learning_rate': '0.0001811', 'epoch': '1.58'}
+{'loss': '0.667', 'grad_norm': '0.3914', 'learning_rate': '0.000181', 'epoch': '1.581'}
+{'loss': '0.7297', 'grad_norm': '0.4367', 'learning_rate': '0.000181', 'epoch': '1.581'}
+{'loss': '0.7664', 'grad_norm': '0.468', 'learning_rate': '0.000181', 'epoch': '1.582'}
+{'loss': '0.7105', 'grad_norm': '0.4033', 'learning_rate': '0.0001809', 'epoch': '1.583'}
+{'loss': '0.6666', 'grad_norm': '0.3821', 'learning_rate': '0.0001809', 'epoch': '1.584'}
+{'loss': '0.7287', 'grad_norm': '0.4595', 'learning_rate': '0.0001809', 'epoch': '1.585'}
+{'loss': '0.7146', 'grad_norm': '0.4036', 'learning_rate': '0.0001809', 'epoch': '1.586'}
+{'loss': '0.6551', 'grad_norm': '0.4284', 'learning_rate': '0.0001808', 'epoch': '1.586'}
+{'loss': '0.704', 'grad_norm': '0.4016', 'learning_rate': '0.0001808', 'epoch': '1.587'}
+{'loss': '0.7523', 'grad_norm': '0.4127', 'learning_rate': '0.0001808', 'epoch': '1.588'}
+{'loss': '0.7498', 'grad_norm': '0.4588', 'learning_rate': '0.0001808', 'epoch': '1.589'}
+{'loss': '0.7694', 'grad_norm': '0.4418', 'learning_rate': '0.0001807', 'epoch': '1.59'}
+{'loss': '0.6767', 'grad_norm': '0.3706', 'learning_rate': '0.0001807', 'epoch': '1.591'}
+{'loss': '0.6394', 'grad_norm': '0.4056', 'learning_rate': '0.0001807', 'epoch': '1.592'}
+{'loss': '0.6776', 'grad_norm': '0.3752', 'learning_rate': '0.0001806', 'epoch': '1.592'}
+{'loss': '0.7201', 'grad_norm': '0.4096', 'learning_rate': '0.0001806', 'epoch': '1.593'}
+{'loss': '0.7314', 'grad_norm': '0.4216', 'learning_rate': '0.0001806', 'epoch': '1.594'}
+{'loss': '0.7955', 'grad_norm': '0.4273', 'learning_rate': '0.0001806', 'epoch': '1.595'}
+{'loss': '0.6552', 'grad_norm': '0.398', 'learning_rate': '0.0001805', 'epoch': '1.596'}
+{'loss': '0.7638', 'grad_norm': '0.446', 'learning_rate': '0.0001805', 'epoch': '1.597'}
+{'loss': '0.6886', 'grad_norm': '0.4384', 'learning_rate': '0.0001805', 'epoch': '1.597'}
+{'loss': '0.6726', 'grad_norm': '0.3823', 'learning_rate': '0.0001804', 'epoch': '1.598'}
+{'loss': '0.7417', 'grad_norm': '0.4305', 'learning_rate': '0.0001804', 'epoch': '1.599'}
+{'loss': '0.6827', 'grad_norm': '0.4028', 'learning_rate': '0.0001804', 'epoch': '1.6'}
+{'loss': '0.72', 'grad_norm': '0.4144', 'learning_rate': '0.0001804', 'epoch': '1.601'}
+{'loss': '0.6586', 'grad_norm': '0.4122', 'learning_rate': '0.0001803', 'epoch': '1.602'}
+{'loss': '0.7579', 'grad_norm': '0.4066', 'learning_rate': '0.0001803', 'epoch': '1.603'}
+{'loss': '0.7246', 'grad_norm': '0.4164', 'learning_rate': '0.0001803', 'epoch': '1.603'}
+{'eval_loss': '0.7784', 'eval_runtime': '692.6', 'eval_samples_per_second': '3.042', 'eval_steps_per_second': '3.042', 'epoch': '1.603'}
+{'loss': '0.7114', 'grad_norm': '0.3786', 'learning_rate': '0.0001802', 'epoch': '1.604'}
+{'loss': '0.752', 'grad_norm': '0.5051', 'learning_rate': '0.0001802', 'epoch': '1.605'}
+{'loss': '0.7554', 'grad_norm': '0.5004', 'learning_rate': '0.0001802', 'epoch': '1.606'}
+{'loss': '0.6584', 'grad_norm': '0.3852', 'learning_rate': '0.0001802', 'epoch': '1.607'}
+{'loss': '0.7215', 'grad_norm': '0.4126', 'learning_rate': '0.0001801', 'epoch': '1.608'}
+{'loss': '0.7754', 'grad_norm': '0.4148', 'learning_rate': '0.0001801', 'epoch': '1.608'}
+{'loss': '0.6985', 'grad_norm': '0.3985', 'learning_rate': '0.0001801', 'epoch': '1.609'}
+{'loss': '0.8118', 'grad_norm': '0.4221', 'learning_rate': '0.00018', 'epoch': '1.61'}
+{'loss': '0.6303', 'grad_norm': '0.3715', 'learning_rate': '0.00018', 'epoch': '1.611'}
+{'loss': '0.7327', 'grad_norm': '0.4479', 'learning_rate': '0.00018', 'epoch': '1.612'}
+{'loss': '0.6878', 'grad_norm': '0.4352', 'learning_rate': '0.00018', 'epoch': '1.613'}
+{'loss': '0.6797', 'grad_norm': '0.4291', 'learning_rate': '0.0001799', 'epoch': '1.614'}
+{'loss': '0.7402', 'grad_norm': '0.4065', 'learning_rate': '0.0001799', 'epoch': '1.614'}
+{'loss': '0.6699', 'grad_norm': '0.4258', 'learning_rate': '0.0001799', 'epoch': '1.615'}
+{'loss': '0.6736', 'grad_norm': '0.4299', 'learning_rate': '0.0001798', 'epoch': '1.616'}
+{'loss': '0.7724', 'grad_norm': '0.4206', 'learning_rate': '0.0001798', 'epoch': '1.617'}
+{'loss': '0.6469', 'grad_norm': '0.3967', 'learning_rate': '0.0001798', 'epoch': '1.618'}
+{'loss': '0.7087', 'grad_norm': '0.4286', 'learning_rate': '0.0001797', 'epoch': '1.619'}
+{'loss': '0.6484', 'grad_norm': '0.4377', 'learning_rate': '0.0001797', 'epoch': '1.619'}
+{'loss': '0.6947', 'grad_norm': '0.4054', 'learning_rate': '0.0001797', 'epoch': '1.62'}
+{'loss': '0.6675', 'grad_norm': '0.3685', 'learning_rate': '0.0001797', 'epoch': '1.621'}
+{'loss': '0.6957', 'grad_norm': '0.4481', 'learning_rate': '0.0001796', 'epoch': '1.622'}
+{'loss': '0.7736', 'grad_norm': '0.4524', 'learning_rate': '0.0001796', 'epoch': '1.623'}
+{'loss': '0.7052', 'grad_norm': '0.4235', 'learning_rate': '0.0001796', 'epoch': '1.624'}
+{'loss': '0.7402', 'grad_norm': '0.4732', 'learning_rate': '0.0001795', 'epoch': '1.624'}
+{'loss': '0.6743', 'grad_norm': '0.3892', 'learning_rate': '0.0001795', 'epoch': '1.625'}
+{'loss': '0.6984', 'grad_norm': '0.4431', 'learning_rate': '0.0001795', 'epoch': '1.626'}
+{'loss': '0.8013', 'grad_norm': '0.4429', 'learning_rate': '0.0001795', 'epoch': '1.627'}
+{'loss': '0.8065', 'grad_norm': '0.4336', 'learning_rate': '0.0001794', 'epoch': '1.628'}
+{'loss': '0.707', 'grad_norm': '0.4585', 'learning_rate': '0.0001794', 'epoch': '1.629'}
+{'loss': '0.6722', 'grad_norm': '0.42', 'learning_rate': '0.0001794', 'epoch': '1.63'}
+{'loss': '0.6634', 'grad_norm': '0.382', 'learning_rate': '0.0001793', 'epoch': '1.63'}
+{'loss': '0.7068', 'grad_norm': '0.4471', 'learning_rate': '0.0001793', 'epoch': '1.631'}
+{'loss': '0.7125', 'grad_norm': '0.3942', 'learning_rate': '0.0001793', 'epoch': '1.632'}
+{'loss': '0.7246', 'grad_norm': '0.4492', 'learning_rate': '0.0001793', 'epoch': '1.633'}
+{'loss': '0.6977', 'grad_norm': '0.407', 'learning_rate': '0.0001792', 'epoch': '1.634'}
+{'loss': '0.7631', 'grad_norm': '0.4504', 'learning_rate': '0.0001792', 'epoch': '1.635'}
+{'loss': '0.6656', 'grad_norm': '0.4084', 'learning_rate': '0.0001792', 'epoch': '1.635'}
+{'loss': '0.6302', 'grad_norm': '0.3762', 'learning_rate': '0.0001791', 'epoch': '1.636'}
+{'loss': '0.712', 'grad_norm': '0.3637', 'learning_rate': '0.0001791', 'epoch': '1.637'}
+{'loss': '0.7272', 'grad_norm': '0.4317', 'learning_rate': '0.0001791', 'epoch': '1.638'}
+{'loss': '0.634', 'grad_norm': '0.3772', 'learning_rate': '0.000179', 'epoch': '1.639'}
+{'loss': '0.6911', 'grad_norm': '0.3654', 'learning_rate': '0.000179', 'epoch': '1.64'}
+{'loss': '0.757', 'grad_norm': '0.4244', 'learning_rate': '0.000179', 'epoch': '1.641'}
+{'loss': '0.7835', 'grad_norm': '0.4348', 'learning_rate': '0.000179', 'epoch': '1.641'}
+{'loss': '0.7059', 'grad_norm': '0.4046', 'learning_rate': '0.0001789', 'epoch': '1.642'}
+{'loss': '0.724', 'grad_norm': '0.3978', 'learning_rate': '0.0001789', 'epoch': '1.643'}
+{'loss': '0.7774', 'grad_norm': '0.4613', 'learning_rate': '0.0001789', 'epoch': '1.644'}
+{'loss': '0.7742', 'grad_norm': '0.4841', 'learning_rate': '0.0001788', 'epoch': '1.645'}
+{'loss': '0.6675', 'grad_norm': '0.4008', 'learning_rate': '0.0001788', 'epoch': '1.646'}
+{'eval_loss': '0.7763', 'eval_runtime': '682.1', 'eval_samples_per_second': '3.089', 'eval_steps_per_second': '3.089', 'epoch': '1.646'}
+{'loss': '0.6778', 'grad_norm': '0.3944', 'learning_rate': '0.0001788', 'epoch': '1.646'}
+{'loss': '0.7398', 'grad_norm': '0.4486', 'learning_rate': '0.0001788', 'epoch': '1.647'}
+{'loss': '0.7203', 'grad_norm': '0.4666', 'learning_rate': '0.0001787', 'epoch': '1.648'}
+{'loss': '0.6232', 'grad_norm': '0.3603', 'learning_rate': '0.0001787', 'epoch': '1.649'}
+{'loss': '0.612', 'grad_norm': '0.3545', 'learning_rate': '0.0001787', 'epoch': '1.65'}
+{'loss': '0.7286', 'grad_norm': '0.3947', 'learning_rate': '0.0001786', 'epoch': '1.651'}
+{'loss': '0.6486', 'grad_norm': '0.3315', 'learning_rate': '0.0001786', 'epoch': '1.651'}
+{'loss': '0.687', 'grad_norm': '0.3886', 'learning_rate': '0.0001786', 'epoch': '1.652'}
+{'loss': '0.7351', 'grad_norm': '0.464', 'learning_rate': '0.0001785', 'epoch': '1.653'}
+{'loss': '0.7284', 'grad_norm': '0.4928', 'learning_rate': '0.0001785', 'epoch': '1.654'}
+{'loss': '0.7437', 'grad_norm': '0.4157', 'learning_rate': '0.0001785', 'epoch': '1.655'}
+{'loss': '0.7428', 'grad_norm': '0.4459', 'learning_rate': '0.0001785', 'epoch': '1.656'}
+{'loss': '0.7643', 'grad_norm': '0.4804', 'learning_rate': '0.0001784', 'epoch': '1.657'}
+{'loss': '0.7098', 'grad_norm': '0.4164', 'learning_rate': '0.0001784', 'epoch': '1.657'}
+{'loss': '0.8346', 'grad_norm': '0.5308', 'learning_rate': '0.0001784', 'epoch': '1.658'}
+{'loss': '0.6327', 'grad_norm': '0.3723', 'learning_rate': '0.0001783', 'epoch': '1.659'}
+{'loss': '0.6441', 'grad_norm': '0.4179', 'learning_rate': '0.0001783', 'epoch': '1.66'}
+{'loss': '0.7268', 'grad_norm': '0.427', 'learning_rate': '0.0001783', 'epoch': '1.661'}
+{'loss': '0.7248', 'grad_norm': '0.4876', 'learning_rate': '0.0001782', 'epoch': '1.662'}
+{'loss': '0.6665', 'grad_norm': '0.3888', 'learning_rate': '0.0001782', 'epoch': '1.662'}
+{'loss': '0.8018', 'grad_norm': '0.4684', 'learning_rate': '0.0001782', 'epoch': '1.663'}
+{'loss': '0.6811', 'grad_norm': '0.42', 'learning_rate': '0.0001782', 'epoch': '1.664'}
+{'loss': '0.7141', 'grad_norm': '0.4584', 'learning_rate': '0.0001781', 'epoch': '1.665'}
+{'loss': '0.8421', 'grad_norm': '0.5227', 'learning_rate': '0.0001781', 'epoch': '1.666'}
+{'loss': '0.6918', 'grad_norm': '0.4347', 'learning_rate': '0.0001781', 'epoch': '1.667'}
+{'loss': '0.6676', 'grad_norm': '0.4072', 'learning_rate': '0.000178', 'epoch': '1.668'}
+{'loss': '0.7564', 'grad_norm': '0.4479', 'learning_rate': '0.000178', 'epoch': '1.668'}
+{'loss': '0.6857', 'grad_norm': '0.4299', 'learning_rate': '0.000178', 'epoch': '1.669'}
+{'loss': '0.7754', 'grad_norm': '0.4411', 'learning_rate': '0.0001779', 'epoch': '1.67'}
+{'loss': '0.7826', 'grad_norm': '0.4406', 'learning_rate': '0.0001779', 'epoch': '1.671'}
+{'loss': '0.7013', 'grad_norm': '0.466', 'learning_rate': '0.0001779', 'epoch': '1.672'}
+{'loss': '0.6647', 'grad_norm': '0.3656', 'learning_rate': '0.0001779', 'epoch': '1.673'}
+{'loss': '0.7238', 'grad_norm': '0.4574', 'learning_rate': '0.0001778', 'epoch': '1.673'}
+{'loss': '0.7595', 'grad_norm': '0.4894', 'learning_rate': '0.0001778', 'epoch': '1.674'}
+{'loss': '0.6535', 'grad_norm': '0.3614', 'learning_rate': '0.0001778', 'epoch': '1.675'}
+{'loss': '0.7699', 'grad_norm': '0.4402', 'learning_rate': '0.0001777', 'epoch': '1.676'}
+{'loss': '0.7718', 'grad_norm': '0.4391', 'learning_rate': '0.0001777', 'epoch': '1.677'}
+{'loss': '0.6957', 'grad_norm': '0.4135', 'learning_rate': '0.0001777', 'epoch': '1.678'}
+{'loss': '0.6667', 'grad_norm': '0.4115', 'learning_rate': '0.0001776', 'epoch': '1.678'}
+{'loss': '0.7364', 'grad_norm': '0.4465', 'learning_rate': '0.0001776', 'epoch': '1.679'}
+{'loss': '0.723', 'grad_norm': '0.4242', 'learning_rate': '0.0001776', 'epoch': '1.68'}
+{'loss': '0.6993', 'grad_norm': '0.485', 'learning_rate': '0.0001776', 'epoch': '1.681'}
+{'loss': '0.6958', 'grad_norm': '0.4464', 'learning_rate': '0.0001775', 'epoch': '1.682'}
+{'loss': '0.6843', 'grad_norm': '0.4774', 'learning_rate': '0.0001775', 'epoch': '1.683'}
+{'loss': '0.6478', 'grad_norm': '0.3634', 'learning_rate': '0.0001775', 'epoch': '1.684'}
+{'loss': '0.7167', 'grad_norm': '0.4806', 'learning_rate': '0.0001774', 'epoch': '1.684'}
+{'loss': '0.6625', 'grad_norm': '0.4273', 'learning_rate': '0.0001774', 'epoch': '1.685'}
+{'loss': '0.6408', 'grad_norm': '0.4341', 'learning_rate': '0.0001774', 'epoch': '1.686'}
+{'loss': '0.8097', 'grad_norm': '0.5048', 'learning_rate': '0.0001773', 'epoch': '1.687'}
+{'loss': '0.6902', 'grad_norm': '0.4549', 'learning_rate': '0.0001773', 'epoch': '1.688'}
+{'eval_loss': '0.7721', 'eval_runtime': '668.4', 'eval_samples_per_second': '3.152', 'eval_steps_per_second': '3.152', 'epoch': '1.688'}
+{'loss': '0.6511', 'grad_norm': '0.3891', 'learning_rate': '0.0001773', 'epoch': '1.689'}
+{'loss': '0.7331', 'grad_norm': '0.4446', 'learning_rate': '0.0001773', 'epoch': '1.689'}
+{'loss': '0.5951', 'grad_norm': '0.3364', 'learning_rate': '0.0001772', 'epoch': '1.69'}
+{'loss': '0.7025', 'grad_norm': '0.3726', 'learning_rate': '0.0001772', 'epoch': '1.691'}
+{'loss': '0.6621', 'grad_norm': '0.3577', 'learning_rate': '0.0001772', 'epoch': '1.692'}
+{'loss': '0.6727', 'grad_norm': '0.4586', 'learning_rate': '0.0001771', 'epoch': '1.693'}
+{'loss': '0.6551', 'grad_norm': '0.4461', 'learning_rate': '0.0001771', 'epoch': '1.694'}
+{'loss': '0.7338', 'grad_norm': '0.4262', 'learning_rate': '0.0001771', 'epoch': '1.695'}
+{'loss': '0.6786', 'grad_norm': '0.4458', 'learning_rate': '0.000177', 'epoch': '1.695'}
+{'loss': '0.7113', 'grad_norm': '0.4483', 'learning_rate': '0.000177', 'epoch': '1.696'}
+{'loss': '0.6421', 'grad_norm': '0.3972', 'learning_rate': '0.000177', 'epoch': '1.697'}
+{'loss': '0.8087', 'grad_norm': '0.4355', 'learning_rate': '0.0001769', 'epoch': '1.698'}
+{'loss': '0.6378', 'grad_norm': '0.3699', 'learning_rate': '0.0001769', 'epoch': '1.699'}
+{'loss': '0.6504', 'grad_norm': '0.4257', 'learning_rate': '0.0001769', 'epoch': '1.7'}
+{'loss': '0.6649', 'grad_norm': '0.4015', 'learning_rate': '0.0001769', 'epoch': '1.7'}
+{'loss': '0.6906', 'grad_norm': '0.4392', 'learning_rate': '0.0001768', 'epoch': '1.701'}
+{'loss': '0.7801', 'grad_norm': '0.4357', 'learning_rate': '0.0001768', 'epoch': '1.702'}
+{'loss': '0.7488', 'grad_norm': '0.4681', 'learning_rate': '0.0001768', 'epoch': '1.703'}
+{'loss': '0.7374', 'grad_norm': '0.414', 'learning_rate': '0.0001767', 'epoch': '1.704'}
+{'loss': '0.754', 'grad_norm': '0.512', 'learning_rate': '0.0001767', 'epoch': '1.705'}
+{'loss': '0.703', 'grad_norm': '0.4043', 'learning_rate': '0.0001767', 'epoch': '1.705'}
+{'loss': '0.6464', 'grad_norm': '0.4628', 'learning_rate': '0.0001766', 'epoch': '1.706'}
+{'loss': '0.7646', 'grad_norm': '0.4752', 'learning_rate': '0.0001766', 'epoch': '1.707'}
+{'loss': '0.7192', 'grad_norm': '0.4172', 'learning_rate': '0.0001766', 'epoch': '1.708'}
+{'loss': '0.6946', 'grad_norm': '0.4024', 'learning_rate': '0.0001765', 'epoch': '1.709'}
+{'loss': '0.7655', 'grad_norm': '0.4386', 'learning_rate': '0.0001765', 'epoch': '1.71'}
+{'loss': '0.7192', 'grad_norm': '0.4192', 'learning_rate': '0.0001765', 'epoch': '1.711'}
+{'loss': '0.6501', 'grad_norm': '0.4349', 'learning_rate': '0.0001765', 'epoch': '1.711'}
+{'loss': '0.7414', 'grad_norm': '0.4179', 'learning_rate': '0.0001764', 'epoch': '1.712'}
+{'loss': '0.7088', 'grad_norm': '0.47', 'learning_rate': '0.0001764', 'epoch': '1.713'}
+{'loss': '0.6717', 'grad_norm': '0.4776', 'learning_rate': '0.0001764', 'epoch': '1.714'}
+{'loss': '0.6631', 'grad_norm': '0.4094', 'learning_rate': '0.0001763', 'epoch': '1.715'}
+{'loss': '0.7678', 'grad_norm': '0.3969', 'learning_rate': '0.0001763', 'epoch': '1.716'}
+{'loss': '0.6982', 'grad_norm': '0.4375', 'learning_rate': '0.0001763', 'epoch': '1.716'}
+{'loss': '0.692', 'grad_norm': '0.445', 'learning_rate': '0.0001762', 'epoch': '1.717'}
+{'loss': '0.6311', 'grad_norm': '0.369', 'learning_rate': '0.0001762', 'epoch': '1.718'}
+{'loss': '0.6954', 'grad_norm': '0.4313', 'learning_rate': '0.0001762', 'epoch': '1.719'}
+{'loss': '0.6974', 'grad_norm': '0.4091', 'learning_rate': '0.0001761', 'epoch': '1.72'}
+{'loss': '0.6776', 'grad_norm': '0.4311', 'learning_rate': '0.0001761', 'epoch': '1.721'}
+{'loss': '0.6889', 'grad_norm': '0.4557', 'learning_rate': '0.0001761', 'epoch': '1.722'}
+{'loss': '0.6895', 'grad_norm': '0.4493', 'learning_rate': '0.000176', 'epoch': '1.722'}
+{'loss': '0.6842', 'grad_norm': '0.3972', 'learning_rate': '0.000176', 'epoch': '1.723'}
+{'loss': '0.7197', 'grad_norm': '0.4486', 'learning_rate': '0.000176', 'epoch': '1.724'}
+{'loss': '0.7388', 'grad_norm': '0.4013', 'learning_rate': '0.000176', 'epoch': '1.725'}
+{'loss': '0.7448', 'grad_norm': '0.4042', 'learning_rate': '0.0001759', 'epoch': '1.726'}
+{'loss': '0.6348', 'grad_norm': '0.3978', 'learning_rate': '0.0001759', 'epoch': '1.727'}
+{'loss': '0.6813', 'grad_norm': '0.3935', 'learning_rate': '0.0001759', 'epoch': '1.727'}
+{'loss': '0.6369', 'grad_norm': '0.4081', 'learning_rate': '0.0001758', 'epoch': '1.728'}
+{'loss': '0.6664', 'grad_norm': '0.5187', 'learning_rate': '0.0001758', 'epoch': '1.729'}
+{'loss': '0.7181', 'grad_norm': '0.4529', 'learning_rate': '0.0001758', 'epoch': '1.73'}
+{'eval_loss': '0.7691', 'eval_runtime': '673.6', 'eval_samples_per_second': '3.128', 'eval_steps_per_second': '3.128', 'epoch': '1.73'}
+{'loss': '0.685', 'grad_norm': '0.4568', 'learning_rate': '0.0001757', 'epoch': '1.731'}
+{'loss': '0.7226', 'grad_norm': '0.4854', 'learning_rate': '0.0001757', 'epoch': '1.732'}
+{'loss': '0.7304', 'grad_norm': '0.4866', 'learning_rate': '0.0001757', 'epoch': '1.732'}
+{'loss': '0.679', 'grad_norm': '0.461', 'learning_rate': '0.0001756', 'epoch': '1.733'}
+{'loss': '0.7311', 'grad_norm': '0.3853', 'learning_rate': '0.0001756', 'epoch': '1.734'}
+{'loss': '0.7106', 'grad_norm': '0.4196', 'learning_rate': '0.0001756', 'epoch': '1.735'}
+{'loss': '0.6346', 'grad_norm': '0.3742', 'learning_rate': '0.0001755', 'epoch': '1.736'}
+{'loss': '0.7924', 'grad_norm': '0.5051', 'learning_rate': '0.0001755', 'epoch': '1.737'}
+{'loss': '0.7502', 'grad_norm': '0.4973', 'learning_rate': '0.0001755', 'epoch': '1.738'}
+{'loss': '0.6334', 'grad_norm': '0.3896', 'learning_rate': '0.0001755', 'epoch': '1.738'}
+{'loss': '0.6885', 'grad_norm': '0.4303', 'learning_rate': '0.0001754', 'epoch': '1.739'}
+{'loss': '0.7346', 'grad_norm': '0.3983', 'learning_rate': '0.0001754', 'epoch': '1.74'}
+{'loss': '0.6538', 'grad_norm': '0.4263', 'learning_rate': '0.0001754', 'epoch': '1.741'}
+{'loss': '0.6368', 'grad_norm': '0.4012', 'learning_rate': '0.0001753', 'epoch': '1.742'}
+{'loss': '0.6038', 'grad_norm': '0.3881', 'learning_rate': '0.0001753', 'epoch': '1.743'}
+{'loss': '0.6657', 'grad_norm': '0.3841', 'learning_rate': '0.0001753', 'epoch': '1.743'}
+{'loss': '0.6518', 'grad_norm': '0.3971', 'learning_rate': '0.0001752', 'epoch': '1.744'}
+{'loss': '0.6099', 'grad_norm': '0.3754', 'learning_rate': '0.0001752', 'epoch': '1.745'}
+{'loss': '0.7356', 'grad_norm': '0.4981', 'learning_rate': '0.0001752', 'epoch': '1.746'}
+{'loss': '0.7199', 'grad_norm': '0.482', 'learning_rate': '0.0001751', 'epoch': '1.747'}
+{'loss': '0.6841', 'grad_norm': '0.3827', 'learning_rate': '0.0001751', 'epoch': '1.748'}
+{'loss': '0.7349', 'grad_norm': '0.4166', 'learning_rate': '0.0001751', 'epoch': '1.749'}
+{'loss': '0.7345', 'grad_norm': '0.4415', 'learning_rate': '0.000175', 'epoch': '1.749'}
+{'loss': '0.7048', 'grad_norm': '0.4324', 'learning_rate': '0.000175', 'epoch': '1.75'}
+{'loss': '0.7318', 'grad_norm': '0.4467', 'learning_rate': '0.000175', 'epoch': '1.751'}
+{'loss': '0.7447', 'grad_norm': '0.4307', 'learning_rate': '0.0001749', 'epoch': '1.752'}
+{'loss': '0.7352', 'grad_norm': '0.4306', 'learning_rate': '0.0001749', 'epoch': '1.753'}
+{'loss': '0.7244', 'grad_norm': '0.4949', 'learning_rate': '0.0001749', 'epoch': '1.754'}
+{'loss': '0.7516', 'grad_norm': '0.4565', 'learning_rate': '0.0001748', 'epoch': '1.754'}
+{'loss': '0.7451', 'grad_norm': '0.4572', 'learning_rate': '0.0001748', 'epoch': '1.755'}
+{'loss': '0.7133', 'grad_norm': '0.4725', 'learning_rate': '0.0001748', 'epoch': '1.756'}
+{'loss': '0.6437', 'grad_norm': '0.4831', 'learning_rate': '0.0001748', 'epoch': '1.757'}
+{'loss': '0.7588', 'grad_norm': '0.4916', 'learning_rate': '0.0001747', 'epoch': '1.758'}
+{'loss': '0.698', 'grad_norm': '0.4727', 'learning_rate': '0.0001747', 'epoch': '1.759'}
+{'loss': '0.693', 'grad_norm': '0.4192', 'learning_rate': '0.0001747', 'epoch': '1.759'}
+{'loss': '0.7076', 'grad_norm': '0.43', 'learning_rate': '0.0001746', 'epoch': '1.76'}
+{'loss': '0.7261', 'grad_norm': '0.454', 'learning_rate': '0.0001746', 'epoch': '1.761'}
+{'loss': '0.6902', 'grad_norm': '0.4077', 'learning_rate': '0.0001746', 'epoch': '1.762'}
+{'loss': '0.7085', 'grad_norm': '0.4168', 'learning_rate': '0.0001745', 'epoch': '1.763'}
+{'loss': '0.7647', 'grad_norm': '0.4526', 'learning_rate': '0.0001745', 'epoch': '1.764'}
+{'loss': '0.7848', 'grad_norm': '0.464', 'learning_rate': '0.0001745', 'epoch': '1.765'}
+{'loss': '0.7192', 'grad_norm': '0.3361', 'learning_rate': '0.0001744', 'epoch': '1.765'}
+{'loss': '0.6555', 'grad_norm': '0.4036', 'learning_rate': '0.0001744', 'epoch': '1.766'}
+{'loss': '0.6434', 'grad_norm': '0.4565', 'learning_rate': '0.0001744', 'epoch': '1.767'}
+{'loss': '0.6609', 'grad_norm': '0.3893', 'learning_rate': '0.0001743', 'epoch': '1.768'}
+{'loss': '0.6471', 'grad_norm': '0.3716', 'learning_rate': '0.0001743', 'epoch': '1.769'}
+{'loss': '0.6825', 'grad_norm': '0.3984', 'learning_rate': '0.0001743', 'epoch': '1.77'}
+{'loss': '0.7642', 'grad_norm': '0.4422', 'learning_rate': '0.0001742', 'epoch': '1.77'}
+{'loss': '0.7784', 'grad_norm': '0.4453', 'learning_rate': '0.0001742', 'epoch': '1.771'}
+{'loss': '0.6718', 'grad_norm': '0.4497', 'learning_rate': '0.0001742', 'epoch': '1.772'}
+{'eval_loss': '0.7676', 'eval_runtime': '688', 'eval_samples_per_second': '3.063', 'eval_steps_per_second': '3.063', 'epoch': '1.772'}
+{'loss': '0.7051', 'grad_norm': '0.3857', 'learning_rate': '0.0001741', 'epoch': '1.773'}
+{'loss': '0.7223', 'grad_norm': '0.4534', 'learning_rate': '0.0001741', 'epoch': '1.774'}
+{'loss': '0.6435', 'grad_norm': '0.4081', 'learning_rate': '0.0001741', 'epoch': '1.775'}
+{'loss': '0.6863', 'grad_norm': '0.5042', 'learning_rate': '0.000174', 'epoch': '1.776'}
+{'loss': '0.6785', 'grad_norm': '0.4185', 'learning_rate': '0.000174', 'epoch': '1.776'}
+{'loss': '0.7567', 'grad_norm': '0.4696', 'learning_rate': '0.000174', 'epoch': '1.777'}
+{'loss': '0.6178', 'grad_norm': '0.3731', 'learning_rate': '0.000174', 'epoch': '1.778'}
+{'loss': '0.7713', 'grad_norm': '0.481', 'learning_rate': '0.0001739', 'epoch': '1.779'}
+{'loss': '0.6373', 'grad_norm': '0.4164', 'learning_rate': '0.0001739', 'epoch': '1.78'}
+{'loss': '0.7213', 'grad_norm': '0.449', 'learning_rate': '0.0001739', 'epoch': '1.781'}
+{'loss': '0.7576', 'grad_norm': '0.4452', 'learning_rate': '0.0001738', 'epoch': '1.781'}
+{'loss': '0.6474', 'grad_norm': '0.349', 'learning_rate': '0.0001738', 'epoch': '1.782'}
+{'loss': '0.7238', 'grad_norm': '0.4069', 'learning_rate': '0.0001738', 'epoch': '1.783'}
+{'loss': '0.7798', 'grad_norm': '0.4026', 'learning_rate': '0.0001737', 'epoch': '1.784'}
+{'loss': '0.6305', 'grad_norm': '0.3576', 'learning_rate': '0.0001737', 'epoch': '1.785'}
+{'loss': '0.6607', 'grad_norm': '0.4334', 'learning_rate': '0.0001737', 'epoch': '1.786'}
+{'loss': '0.7122', 'grad_norm': '0.4631', 'learning_rate': '0.0001736', 'epoch': '1.786'}
+{'loss': '0.6649', 'grad_norm': '0.3871', 'learning_rate': '0.0001736', 'epoch': '1.787'}
+{'loss': '0.6223', 'grad_norm': '0.39', 'learning_rate': '0.0001736', 'epoch': '1.788'}
+{'loss': '0.7208', 'grad_norm': '0.4736', 'learning_rate': '0.0001735', 'epoch': '1.789'}
+{'loss': '0.7204', 'grad_norm': '0.4235', 'learning_rate': '0.0001735', 'epoch': '1.79'}
+{'loss': '0.71', 'grad_norm': '0.4487', 'learning_rate': '0.0001735', 'epoch': '1.791'}
+{'loss': '0.708', 'grad_norm': '0.4372', 'learning_rate': '0.0001734', 'epoch': '1.792'}
+{'loss': '0.6773', 'grad_norm': '0.4269', 'learning_rate': '0.0001734', 'epoch': '1.792'}
+{'loss': '0.6882', 'grad_norm': '0.358', 'learning_rate': '0.0001734', 'epoch': '1.793'}
+{'loss': '0.7535', 'grad_norm': '0.4872', 'learning_rate': '0.0001733', 'epoch': '1.794'}
+{'loss': '0.6897', 'grad_norm': '0.4491', 'learning_rate': '0.0001733', 'epoch': '1.795'}
+{'loss': '0.7082', 'grad_norm': '0.4362', 'learning_rate': '0.0001733', 'epoch': '1.796'}
+{'loss': '0.6853', 'grad_norm': '0.4287', 'learning_rate': '0.0001732', 'epoch': '1.797'}
+{'loss': '0.7463', 'grad_norm': '0.4522', 'learning_rate': '0.0001732', 'epoch': '1.797'}
+{'loss': '0.6644', 'grad_norm': '0.3863', 'learning_rate': '0.0001732', 'epoch': '1.798'}
+{'loss': '0.6727', 'grad_norm': '0.4117', 'learning_rate': '0.0001731', 'epoch': '1.799'}
+{'loss': '0.6279', 'grad_norm': '0.3985', 'learning_rate': '0.0001731', 'epoch': '1.8'}
+{'loss': '0.5823', 'grad_norm': '0.4', 'learning_rate': '0.0001731', 'epoch': '1.801'}
+{'loss': '0.7493', 'grad_norm': '0.4738', 'learning_rate': '0.000173', 'epoch': '1.802'}
+{'loss': '0.763', 'grad_norm': '0.4624', 'learning_rate': '0.000173', 'epoch': '1.803'}
+{'loss': '0.6485', 'grad_norm': '0.4178', 'learning_rate': '0.000173', 'epoch': '1.803'}
+{'loss': '0.7767', 'grad_norm': '0.5112', 'learning_rate': '0.0001729', 'epoch': '1.804'}
+{'loss': '0.6801', 'grad_norm': '0.4464', 'learning_rate': '0.0001729', 'epoch': '1.805'}
+{'loss': '0.8024', 'grad_norm': '0.509', 'learning_rate': '0.0001729', 'epoch': '1.806'}
+{'loss': '0.781', 'grad_norm': '0.4589', 'learning_rate': '0.0001728', 'epoch': '1.807'}
+{'loss': '0.6768', 'grad_norm': '0.3986', 'learning_rate': '0.0001728', 'epoch': '1.808'}
+{'loss': '0.6094', 'grad_norm': '0.3988', 'learning_rate': '0.0001728', 'epoch': '1.808'}
+{'loss': '0.6837', 'grad_norm': '0.4613', 'learning_rate': '0.0001727', 'epoch': '1.809'}
+{'loss': '0.7144', 'grad_norm': '0.4296', 'learning_rate': '0.0001727', 'epoch': '1.81'}
+{'loss': '0.7191', 'grad_norm': '0.411', 'learning_rate': '0.0001727', 'epoch': '1.811'}
+{'loss': '0.7272', 'grad_norm': '0.4501', 'learning_rate': '0.0001726', 'epoch': '1.812'}
+{'loss': '0.7582', 'grad_norm': '0.4678', 'learning_rate': '0.0001726', 'epoch': '1.813'}
+{'loss': '0.6712', 'grad_norm': '0.4239', 'learning_rate': '0.0001726', 'epoch': '1.814'}
+{'loss': '0.6', 'grad_norm': '0.3767', 'learning_rate': '0.0001725', 'epoch': '1.814'}
+{'eval_loss': '0.7673', 'eval_runtime': '688.4', 'eval_samples_per_second': '3.061', 'eval_steps_per_second': '3.061', 'epoch': '1.814'}
+{'loss': '0.7189', 'grad_norm': '0.4496', 'learning_rate': '0.0001725', 'epoch': '1.815'}
+{'loss': '0.7078', 'grad_norm': '0.4108', 'learning_rate': '0.0001725', 'epoch': '1.816'}
+{'loss': '0.6915', 'grad_norm': '0.4455', 'learning_rate': '0.0001724', 'epoch': '1.817'}
+{'loss': '0.6592', 'grad_norm': '0.3977', 'learning_rate': '0.0001724', 'epoch': '1.818'}
+{'loss': '0.6255', 'grad_norm': '0.4028', 'learning_rate': '0.0001724', 'epoch': '1.819'}
+{'loss': '0.6969', 'grad_norm': '0.4102', 'learning_rate': '0.0001723', 'epoch': '1.819'}
+{'loss': '0.6536', 'grad_norm': '0.4594', 'learning_rate': '0.0001723', 'epoch': '1.82'}
+{'loss': '0.6379', 'grad_norm': '0.3948', 'learning_rate': '0.0001723', 'epoch': '1.821'}
+{'loss': '0.7021', 'grad_norm': '0.4038', 'learning_rate': '0.0001722', 'epoch': '1.822'}
+{'loss': '0.7411', 'grad_norm': '0.4179', 'learning_rate': '0.0001722', 'epoch': '1.823'}
+{'loss': '0.7435', 'grad_norm': '0.4719', 'learning_rate': '0.0001722', 'epoch': '1.824'}
+{'loss': '0.6814', 'grad_norm': '0.3995', 'learning_rate': '0.0001721', 'epoch': '1.824'}
+{'loss': '0.6469', 'grad_norm': '0.4367', 'learning_rate': '0.0001721', 'epoch': '1.825'}
+{'loss': '0.6629', 'grad_norm': '0.4078', 'learning_rate': '0.0001721', 'epoch': '1.826'}
+{'loss': '0.7028', 'grad_norm': '0.4141', 'learning_rate': '0.000172', 'epoch': '1.827'}
+{'loss': '0.6711', 'grad_norm': '0.4487', 'learning_rate': '0.000172', 'epoch': '1.828'}
+{'loss': '0.7925', 'grad_norm': '0.5084', 'learning_rate': '0.000172', 'epoch': '1.829'}
+{'loss': '0.7537', 'grad_norm': '0.4346', 'learning_rate': '0.0001719', 'epoch': '1.83'}
+{'loss': '0.7446', 'grad_norm': '0.4604', 'learning_rate': '0.0001719', 'epoch': '1.83'}
+{'loss': '0.7003', 'grad_norm': '0.3894', 'learning_rate': '0.0001719', 'epoch': '1.831'}
+{'loss': '0.6949', 'grad_norm': '0.5852', 'learning_rate': '0.0001718', 'epoch': '1.832'}
+{'loss': '0.6758', 'grad_norm': '0.4091', 'learning_rate': '0.0001718', 'epoch': '1.833'}
+{'loss': '0.7423', 'grad_norm': '0.4236', 'learning_rate': '0.0001718', 'epoch': '1.834'}
+{'loss': '0.7015', 'grad_norm': '0.3898', 'learning_rate': '0.0001717', 'epoch': '1.835'}
+{'loss': '0.6953', 'grad_norm': '0.4122', 'learning_rate': '0.0001717', 'epoch': '1.835'}
+{'loss': '0.6759', 'grad_norm': '0.3809', 'learning_rate': '0.0001717', 'epoch': '1.836'}
+{'loss': '0.7139', 'grad_norm': '0.4352', 'learning_rate': '0.0001716', 'epoch': '1.837'}
+{'loss': '0.7125', 'grad_norm': '0.4404', 'learning_rate': '0.0001716', 'epoch': '1.838'}
+{'loss': '0.7011', 'grad_norm': '0.3966', 'learning_rate': '0.0001716', 'epoch': '1.839'}
+{'loss': '0.6486', 'grad_norm': '0.4274', 'learning_rate': '0.0001715', 'epoch': '1.84'}
+{'loss': '0.667', 'grad_norm': '0.4639', 'learning_rate': '0.0001715', 'epoch': '1.841'}
+{'loss': '0.6276', 'grad_norm': '0.4191', 'learning_rate': '0.0001715', 'epoch': '1.841'}
+{'loss': '0.5976', 'grad_norm': '0.4147', 'learning_rate': '0.0001714', 'epoch': '1.842'}
+{'loss': '0.6537', 'grad_norm': '0.3872', 'learning_rate': '0.0001714', 'epoch': '1.843'}
+{'loss': '0.7079', 'grad_norm': '0.3989', 'learning_rate': '0.0001714', 'epoch': '1.844'}
+{'loss': '0.6795', 'grad_norm': '0.4378', 'learning_rate': '0.0001713', 'epoch': '1.845'}
+{'loss': '0.703', 'grad_norm': '0.4396', 'learning_rate': '0.0001713', 'epoch': '1.846'}
+{'loss': '0.7612', 'grad_norm': '0.4426', 'learning_rate': '0.0001713', 'epoch': '1.846'}
+{'loss': '0.7379', 'grad_norm': '0.4006', 'learning_rate': '0.0001712', 'epoch': '1.847'}
+{'loss': '0.6839', 'grad_norm': '0.4304', 'learning_rate': '0.0001712', 'epoch': '1.848'}
+{'loss': '0.6987', 'grad_norm': '0.4479', 'learning_rate': '0.0001712', 'epoch': '1.849'}
+{'loss': '0.6964', 'grad_norm': '0.4218', 'learning_rate': '0.0001711', 'epoch': '1.85'}
+{'loss': '0.6984', 'grad_norm': '0.3977', 'learning_rate': '0.0001711', 'epoch': '1.851'}
+{'loss': '0.7579', 'grad_norm': '0.4844', 'learning_rate': '0.0001711', 'epoch': '1.851'}
+{'loss': '0.7036', 'grad_norm': '0.4494', 'learning_rate': '0.000171', 'epoch': '1.852'}
+{'loss': '0.7387', 'grad_norm': '0.4722', 'learning_rate': '0.000171', 'epoch': '1.853'}
+{'loss': '0.7769', 'grad_norm': '0.4816', 'learning_rate': '0.000171', 'epoch': '1.854'}
+{'loss': '0.7315', 'grad_norm': '0.4364', 'learning_rate': '0.0001709', 'epoch': '1.855'}
+{'loss': '0.729', 'grad_norm': '0.4035', 'learning_rate': '0.0001709', 'epoch': '1.856'}
+{'loss': '0.719', 'grad_norm': '0.4427', 'learning_rate': '0.0001709', 'epoch': '1.857'}
+{'eval_loss': '0.7635', 'eval_runtime': '678.2', 'eval_samples_per_second': '3.107', 'eval_steps_per_second': '3.107', 'epoch': '1.857'}
+{'loss': '0.6318', 'grad_norm': '0.3969', 'learning_rate': '0.0001708', 'epoch': '1.857'}
+{'loss': '0.7041', 'grad_norm': '0.478', 'learning_rate': '0.0001708', 'epoch': '1.858'}
+{'loss': '0.7835', 'grad_norm': '0.4746', 'learning_rate': '0.0001708', 'epoch': '1.859'}
+{'loss': '0.7028', 'grad_norm': '0.4136', 'learning_rate': '0.0001707', 'epoch': '1.86'}
+{'loss': '0.7144', 'grad_norm': '0.4579', 'learning_rate': '0.0001707', 'epoch': '1.861'}
+{'loss': '0.7328', 'grad_norm': '0.4293', 'learning_rate': '0.0001707', 'epoch': '1.862'}
+{'loss': '0.7199', 'grad_norm': '0.3822', 'learning_rate': '0.0001706', 'epoch': '1.862'}
+{'loss': '0.7345', 'grad_norm': '0.3955', 'learning_rate': '0.0001706', 'epoch': '1.863'}
+{'loss': '0.5879', 'grad_norm': '0.3882', 'learning_rate': '0.0001706', 'epoch': '1.864'}
+{'loss': '0.7228', 'grad_norm': '0.5109', 'learning_rate': '0.0001705', 'epoch': '1.865'}
+{'loss': '0.6606', 'grad_norm': '0.3869', 'learning_rate': '0.0001705', 'epoch': '1.866'}
+{'loss': '0.7582', 'grad_norm': '0.4887', 'learning_rate': '0.0001705', 'epoch': '1.867'}
+{'loss': '0.7052', 'grad_norm': '0.4214', 'learning_rate': '0.0001704', 'epoch': '1.868'}
+{'loss': '0.6489', 'grad_norm': '0.4189', 'learning_rate': '0.0001704', 'epoch': '1.868'}
+{'loss': '0.6897', 'grad_norm': '0.4265', 'learning_rate': '0.0001704', 'epoch': '1.869'}
+{'loss': '0.7468', 'grad_norm': '0.5351', 'learning_rate': '0.0001703', 'epoch': '1.87'}
+{'loss': '0.603', 'grad_norm': '0.4016', 'learning_rate': '0.0001703', 'epoch': '1.871'}
+{'loss': '0.6891', 'grad_norm': '0.3787', 'learning_rate': '0.0001703', 'epoch': '1.872'}
+{'loss': '0.619', 'grad_norm': '0.4267', 'learning_rate': '0.0001702', 'epoch': '1.873'}
+{'loss': '0.6859', 'grad_norm': '0.4077', 'learning_rate': '0.0001702', 'epoch': '1.873'}
+{'loss': '0.665', 'grad_norm': '0.421', 'learning_rate': '0.0001702', 'epoch': '1.874'}
+{'loss': '0.6888', 'grad_norm': '0.4114', 'learning_rate': '0.0001701', 'epoch': '1.875'}
+{'loss': '0.6951', 'grad_norm': '0.346', 'learning_rate': '0.0001701', 'epoch': '1.876'}
+{'loss': '0.7361', 'grad_norm': '0.4513', 'learning_rate': '0.0001701', 'epoch': '1.877'}
+{'loss': '0.6672', 'grad_norm': '0.4258', 'learning_rate': '0.00017', 'epoch': '1.878'}
+{'loss': '0.674', 'grad_norm': '0.3914', 'learning_rate': '0.00017', 'epoch': '1.878'}
+{'loss': '0.7747', 'grad_norm': '0.4218', 'learning_rate': '0.00017', 'epoch': '1.879'}
+{'loss': '0.719', 'grad_norm': '0.411', 'learning_rate': '0.0001699', 'epoch': '1.88'}
+{'loss': '0.688', 'grad_norm': '0.4073', 'learning_rate': '0.0001699', 'epoch': '1.881'}
+{'loss': '0.6995', 'grad_norm': '0.4674', 'learning_rate': '0.0001698', 'epoch': '1.882'}
+{'loss': '0.7389', 'grad_norm': '0.4769', 'learning_rate': '0.0001698', 'epoch': '1.883'}
+{'loss': '0.6376', 'grad_norm': '0.3834', 'learning_rate': '0.0001698', 'epoch': '1.884'}
+{'loss': '0.6846', 'grad_norm': '0.4455', 'learning_rate': '0.0001697', 'epoch': '1.884'}
+{'loss': '0.7308', 'grad_norm': '0.3934', 'learning_rate': '0.0001697', 'epoch': '1.885'}
+{'loss': '0.6734', 'grad_norm': '0.379', 'learning_rate': '0.0001697', 'epoch': '1.886'}
+{'loss': '0.7295', 'grad_norm': '0.4447', 'learning_rate': '0.0001696', 'epoch': '1.887'}
+{'loss': '0.6537', 'grad_norm': '0.3957', 'learning_rate': '0.0001696', 'epoch': '1.888'}
+{'loss': '0.6239', 'grad_norm': '0.4047', 'learning_rate': '0.0001696', 'epoch': '1.889'}
+{'loss': '0.6613', 'grad_norm': '0.456', 'learning_rate': '0.0001695', 'epoch': '1.889'}
+{'loss': '0.6588', 'grad_norm': '0.398', 'learning_rate': '0.0001695', 'epoch': '1.89'}
+{'loss': '0.7899', 'grad_norm': '0.4142', 'learning_rate': '0.0001695', 'epoch': '1.891'}
+{'loss': '0.6213', 'grad_norm': '0.4118', 'learning_rate': '0.0001694', 'epoch': '1.892'}
+{'loss': '0.6624', 'grad_norm': '0.3922', 'learning_rate': '0.0001694', 'epoch': '1.893'}
+{'loss': '0.7063', 'grad_norm': '0.4876', 'learning_rate': '0.0001694', 'epoch': '1.894'}
+{'loss': '0.6472', 'grad_norm': '0.3806', 'learning_rate': '0.0001693', 'epoch': '1.895'}
+{'loss': '0.7695', 'grad_norm': '0.5369', 'learning_rate': '0.0001693', 'epoch': '1.895'}
+{'loss': '0.7', 'grad_norm': '0.4174', 'learning_rate': '0.0001693', 'epoch': '1.896'}
+{'loss': '0.6927', 'grad_norm': '0.4087', 'learning_rate': '0.0001692', 'epoch': '1.897'}
+{'loss': '0.6776', 'grad_norm': '0.4475', 'learning_rate': '0.0001692', 'epoch': '1.898'}
+{'loss': '0.686', 'grad_norm': '0.4395', 'learning_rate': '0.0001692', 'epoch': '1.899'}
+{'eval_loss': '0.7601', 'eval_runtime': '674.3', 'eval_samples_per_second': '3.125', 'eval_steps_per_second': '3.125', 'epoch': '1.899'}
+{'loss': '0.6525', 'grad_norm': '0.3803', 'learning_rate': '0.0001691', 'epoch': '1.9'}
+{'loss': '0.6474', 'grad_norm': '0.3461', 'learning_rate': '0.0001691', 'epoch': '1.9'}
+{'loss': '0.76', 'grad_norm': '0.4733', 'learning_rate': '0.0001691', 'epoch': '1.901'}
+{'loss': '0.6596', 'grad_norm': '0.3916', 'learning_rate': '0.000169', 'epoch': '1.902'}
+{'loss': '0.5873', 'grad_norm': '0.3679', 'learning_rate': '0.000169', 'epoch': '1.903'}
+{'loss': '0.6237', 'grad_norm': '0.3844', 'learning_rate': '0.0001689', 'epoch': '1.904'}
+{'loss': '0.7129', 'grad_norm': '0.4168', 'learning_rate': '0.0001689', 'epoch': '1.905'}
+{'loss': '0.7348', 'grad_norm': '0.4603', 'learning_rate': '0.0001689', 'epoch': '1.905'}
+{'loss': '0.6798', 'grad_norm': '0.4359', 'learning_rate': '0.0001688', 'epoch': '1.906'}
+{'loss': '0.6754', 'grad_norm': '0.4302', 'learning_rate': '0.0001688', 'epoch': '1.907'}
+{'loss': '0.6819', 'grad_norm': '0.4145', 'learning_rate': '0.0001688', 'epoch': '1.908'}
+{'loss': '0.6627', 'grad_norm': '0.4051', 'learning_rate': '0.0001687', 'epoch': '1.909'}
+{'loss': '0.6753', 'grad_norm': '0.3897', 'learning_rate': '0.0001687', 'epoch': '1.91'}
+{'loss': '0.6972', 'grad_norm': '0.4308', 'learning_rate': '0.0001687', 'epoch': '1.911'}
+{'loss': '0.7674', 'grad_norm': '0.4121', 'learning_rate': '0.0001686', 'epoch': '1.911'}
+{'loss': '0.7075', 'grad_norm': '0.4322', 'learning_rate': '0.0001686', 'epoch': '1.912'}
+{'loss': '0.6179', 'grad_norm': '0.3831', 'learning_rate': '0.0001686', 'epoch': '1.913'}
+{'loss': '0.7168', 'grad_norm': '0.3941', 'learning_rate': '0.0001685', 'epoch': '1.914'}
+{'loss': '0.6893', 'grad_norm': '0.372', 'learning_rate': '0.0001685', 'epoch': '1.915'}
+{'loss': '0.7353', 'grad_norm': '0.4015', 'learning_rate': '0.0001685', 'epoch': '1.916'}
+{'loss': '0.6469', 'grad_norm': '0.4213', 'learning_rate': '0.0001684', 'epoch': '1.916'}
+{'loss': '0.6884', 'grad_norm': '0.4537', 'learning_rate': '0.0001684', 'epoch': '1.917'}
+{'loss': '0.7378', 'grad_norm': '0.4365', 'learning_rate': '0.0001684', 'epoch': '1.918'}
+{'loss': '0.7432', 'grad_norm': '0.4146', 'learning_rate': '0.0001683', 'epoch': '1.919'}
+{'loss': '0.6801', 'grad_norm': '0.4022', 'learning_rate': '0.0001683', 'epoch': '1.92'}
+{'loss': '0.6836', 'grad_norm': '0.3995', 'learning_rate': '0.0001682', 'epoch': '1.921'}
+{'loss': '0.7724', 'grad_norm': '0.4603', 'learning_rate': '0.0001682', 'epoch': '1.922'}
+{'loss': '0.7513', 'grad_norm': '0.474', 'learning_rate': '0.0001682', 'epoch': '1.922'}
+{'loss': '0.7001', 'grad_norm': '0.4662', 'learning_rate': '0.0001681', 'epoch': '1.923'}
+{'loss': '0.6381', 'grad_norm': '0.4187', 'learning_rate': '0.0001681', 'epoch': '1.924'}
+{'loss': '0.7824', 'grad_norm': '0.5057', 'learning_rate': '0.0001681', 'epoch': '1.925'}
+{'loss': '0.7211', 'grad_norm': '0.4235', 'learning_rate': '0.000168', 'epoch': '1.926'}
+{'loss': '0.7944', 'grad_norm': '0.4576', 'learning_rate': '0.000168', 'epoch': '1.927'}
+{'loss': '0.6491', 'grad_norm': '0.4225', 'learning_rate': '0.000168', 'epoch': '1.927'}
+{'loss': '0.6977', 'grad_norm': '0.4108', 'learning_rate': '0.0001679', 'epoch': '1.928'}
+{'loss': '0.6559', 'grad_norm': '0.4235', 'learning_rate': '0.0001679', 'epoch': '1.929'}
+{'loss': '0.79', 'grad_norm': '0.397', 'learning_rate': '0.0001679', 'epoch': '1.93'}
+{'loss': '0.7021', 'grad_norm': '0.4401', 'learning_rate': '0.0001678', 'epoch': '1.931'}
+{'loss': '0.7107', 'grad_norm': '0.4256', 'learning_rate': '0.0001678', 'epoch': '1.932'}
+{'loss': '0.5809', 'grad_norm': '0.4065', 'learning_rate': '0.0001678', 'epoch': '1.932'}
+{'loss': '0.6816', 'grad_norm': '0.4246', 'learning_rate': '0.0001677', 'epoch': '1.933'}
+{'loss': '0.6618', 'grad_norm': '0.4248', 'learning_rate': '0.0001677', 'epoch': '1.934'}
+{'loss': '0.7462', 'grad_norm': '0.4336', 'learning_rate': '0.0001677', 'epoch': '1.935'}
+{'loss': '0.7802', 'grad_norm': '0.4254', 'learning_rate': '0.0001676', 'epoch': '1.936'}
+{'loss': '0.7308', 'grad_norm': '0.4124', 'learning_rate': '0.0001676', 'epoch': '1.937'}
+{'loss': '0.7127', 'grad_norm': '0.3811', 'learning_rate': '0.0001675', 'epoch': '1.938'}
+{'loss': '0.6798', 'grad_norm': '0.486', 'learning_rate': '0.0001675', 'epoch': '1.938'}
+{'loss': '0.6078', 'grad_norm': '0.364', 'learning_rate': '0.0001675', 'epoch': '1.939'}
+{'loss': '0.692', 'grad_norm': '0.3961', 'learning_rate': '0.0001674', 'epoch': '1.94'}
+{'loss': '0.7286', 'grad_norm': '0.4059', 'learning_rate': '0.0001674', 'epoch': '1.941'}
+{'eval_loss': '0.7586', 'eval_runtime': '679.1', 'eval_samples_per_second': '3.103', 'eval_steps_per_second': '3.103', 'epoch': '1.941'}
+{'loss': '0.6696', 'grad_norm': '0.4511', 'learning_rate': '0.0001674', 'epoch': '1.942'}
+{'loss': '0.6976', 'grad_norm': '0.4244', 'learning_rate': '0.0001673', 'epoch': '1.943'}
+{'loss': '0.5991', 'grad_norm': '0.3757', 'learning_rate': '0.0001673', 'epoch': '1.943'}
+{'loss': '0.7234', 'grad_norm': '0.4026', 'learning_rate': '0.0001673', 'epoch': '1.944'}
+{'loss': '0.746', 'grad_norm': '0.4885', 'learning_rate': '0.0001672', 'epoch': '1.945'}
+{'loss': '0.5577', 'grad_norm': '0.363', 'learning_rate': '0.0001672', 'epoch': '1.946'}
+{'loss': '0.7394', 'grad_norm': '0.4275', 'learning_rate': '0.0001672', 'epoch': '1.947'}
+{'loss': '0.5905', 'grad_norm': '0.3581', 'learning_rate': '0.0001671', 'epoch': '1.948'}
+{'loss': '0.7035', 'grad_norm': '0.4111', 'learning_rate': '0.0001671', 'epoch': '1.949'}
+{'loss': '0.7049', 'grad_norm': '0.4531', 'learning_rate': '0.000167', 'epoch': '1.949'}
+{'loss': '0.6595', 'grad_norm': '0.4178', 'learning_rate': '0.000167', 'epoch': '1.95'}
+{'loss': '0.6903', 'grad_norm': '0.4051', 'learning_rate': '0.000167', 'epoch': '1.951'}
+{'loss': '0.7087', 'grad_norm': '0.4226', 'learning_rate': '0.0001669', 'epoch': '1.952'}
+{'loss': '0.6648', 'grad_norm': '0.3979', 'learning_rate': '0.0001669', 'epoch': '1.953'}
+{'loss': '0.7472', 'grad_norm': '0.3986', 'learning_rate': '0.0001669', 'epoch': '1.954'}
+{'loss': '0.7526', 'grad_norm': '0.4174', 'learning_rate': '0.0001668', 'epoch': '1.954'}
+{'loss': '0.6674', 'grad_norm': '0.4251', 'learning_rate': '0.0001668', 'epoch': '1.955'}
+{'loss': '0.6515', 'grad_norm': '0.4049', 'learning_rate': '0.0001668', 'epoch': '1.956'}
+{'loss': '0.6451', 'grad_norm': '0.4312', 'learning_rate': '0.0001667', 'epoch': '1.957'}
+{'loss': '0.6519', 'grad_norm': '0.3954', 'learning_rate': '0.0001667', 'epoch': '1.958'}
+{'loss': '0.6816', 'grad_norm': '0.5028', 'learning_rate': '0.0001667', 'epoch': '1.959'}
+{'loss': '0.6714', 'grad_norm': '0.4396', 'learning_rate': '0.0001666', 'epoch': '1.959'}
+{'loss': '0.6989', 'grad_norm': '0.4273', 'learning_rate': '0.0001666', 'epoch': '1.96'}
+{'loss': '0.6501', 'grad_norm': '0.4142', 'learning_rate': '0.0001665', 'epoch': '1.961'}
+{'loss': '0.6438', 'grad_norm': '0.3907', 'learning_rate': '0.0001665', 'epoch': '1.962'}
+{'loss': '0.6521', 'grad_norm': '0.418', 'learning_rate': '0.0001665', 'epoch': '1.963'}
+{'loss': '0.7301', 'grad_norm': '0.4148', 'learning_rate': '0.0001664', 'epoch': '1.964'}
+{'loss': '0.6217', 'grad_norm': '0.4176', 'learning_rate': '0.0001664', 'epoch': '1.965'}
+{'loss': '0.7974', 'grad_norm': '0.4488', 'learning_rate': '0.0001664', 'epoch': '1.965'}
+{'loss': '0.7288', 'grad_norm': '0.3769', 'learning_rate': '0.0001663', 'epoch': '1.966'}
+{'loss': '0.7564', 'grad_norm': '0.4636', 'learning_rate': '0.0001663', 'epoch': '1.967'}
+{'loss': '0.7408', 'grad_norm': '0.432', 'learning_rate': '0.0001663', 'epoch': '1.968'}
+{'loss': '0.6962', 'grad_norm': '0.4098', 'learning_rate': '0.0001662', 'epoch': '1.969'}
+{'loss': '0.696', 'grad_norm': '0.4412', 'learning_rate': '0.0001662', 'epoch': '1.97'}
+{'loss': '0.6694', 'grad_norm': '0.373', 'learning_rate': '0.0001662', 'epoch': '1.97'}
+{'loss': '0.6943', 'grad_norm': '0.4506', 'learning_rate': '0.0001661', 'epoch': '1.971'}
+{'loss': '0.6916', 'grad_norm': '0.4371', 'learning_rate': '0.0001661', 'epoch': '1.972'}
+{'loss': '0.743', 'grad_norm': '0.4376', 'learning_rate': '0.000166', 'epoch': '1.973'}
+{'loss': '0.7261', 'grad_norm': '0.4393', 'learning_rate': '0.000166', 'epoch': '1.974'}
+{'loss': '0.7351', 'grad_norm': '0.4472', 'learning_rate': '0.000166', 'epoch': '1.975'}
+{'loss': '0.7296', 'grad_norm': '0.3982', 'learning_rate': '0.0001659', 'epoch': '1.976'}
+{'loss': '0.6462', 'grad_norm': '0.4037', 'learning_rate': '0.0001659', 'epoch': '1.976'}
+{'loss': '0.6993', 'grad_norm': '0.4723', 'learning_rate': '0.0001659', 'epoch': '1.977'}
+{'loss': '0.7299', 'grad_norm': '0.4574', 'learning_rate': '0.0001658', 'epoch': '1.978'}
+{'loss': '0.7643', 'grad_norm': '0.4012', 'learning_rate': '0.0001658', 'epoch': '1.979'}
+{'loss': '0.6945', 'grad_norm': '0.3905', 'learning_rate': '0.0001658', 'epoch': '1.98'}
+{'loss': '0.5987', 'grad_norm': '0.3644', 'learning_rate': '0.0001657', 'epoch': '1.981'}
+{'loss': '0.7462', 'grad_norm': '0.4371', 'learning_rate': '0.0001657', 'epoch': '1.981'}
+{'loss': '0.7554', 'grad_norm': '0.444', 'learning_rate': '0.0001656', 'epoch': '1.982'}
+{'loss': '0.712', 'grad_norm': '0.4137', 'learning_rate': '0.0001656', 'epoch': '1.983'}
+{'eval_loss': '0.7582', 'eval_runtime': '690.1', 'eval_samples_per_second': '3.053', 'eval_steps_per_second': '3.053', 'epoch': '1.983'}
+{'loss': '0.626', 'grad_norm': '0.4408', 'learning_rate': '0.0001656', 'epoch': '1.984'}
+{'loss': '0.7382', 'grad_norm': '0.4698', 'learning_rate': '0.0001655', 'epoch': '1.985'}
+{'loss': '0.7037', 'grad_norm': '0.4235', 'learning_rate': '0.0001655', 'epoch': '1.986'}
+{'loss': '0.7173', 'grad_norm': '0.3863', 'learning_rate': '0.0001655', 'epoch': '1.986'}
+{'loss': '0.6885', 'grad_norm': '0.4102', 'learning_rate': '0.0001654', 'epoch': '1.987'}
+{'loss': '0.7182', 'grad_norm': '0.4487', 'learning_rate': '0.0001654', 'epoch': '1.988'}
+{'loss': '0.7738', 'grad_norm': '0.4776', 'learning_rate': '0.0001654', 'epoch': '1.989'}
+{'loss': '0.6836', 'grad_norm': '0.4028', 'learning_rate': '0.0001653', 'epoch': '1.99'}
+{'loss': '0.7448', 'grad_norm': '0.4235', 'learning_rate': '0.0001653', 'epoch': '1.991'}
+{'loss': '0.68', 'grad_norm': '0.4075', 'learning_rate': '0.0001652', 'epoch': '1.992'}
+{'loss': '0.6609', 'grad_norm': '0.3767', 'learning_rate': '0.0001652', 'epoch': '1.992'}
+{'loss': '0.6497', 'grad_norm': '0.3601', 'learning_rate': '0.0001652', 'epoch': '1.993'}
+{'loss': '0.6305', 'grad_norm': '0.3959', 'learning_rate': '0.0001651', 'epoch': '1.994'}
+{'loss': '0.7123', 'grad_norm': '0.41', 'learning_rate': '0.0001651', 'epoch': '1.995'}
+{'loss': '0.6975', 'grad_norm': '0.4195', 'learning_rate': '0.0001651', 'epoch': '1.996'}
+{'loss': '0.6844', 'grad_norm': '0.4284', 'learning_rate': '0.000165', 'epoch': '1.997'}
+{'loss': '0.7454', 'grad_norm': '0.3983', 'learning_rate': '0.000165', 'epoch': '1.997'}
+{'loss': '0.6571', 'grad_norm': '0.3876', 'learning_rate': '0.000165', 'epoch': '1.998'}
+{'loss': '0.6845', 'grad_norm': '0.38', 'learning_rate': '0.0001649', 'epoch': '1.999'}
+{'loss': '0.6115', 'grad_norm': '0.3861', 'learning_rate': '0.0001649', 'epoch': '2'}
+{'loss': '0.6193', 'grad_norm': '0.3991', 'learning_rate': '0.0001648', 'epoch': '2.001'}
+{'loss': '0.6479', 'grad_norm': '0.4138', 'learning_rate': '0.0001648', 'epoch': '2.002'}
+{'loss': '0.6163', 'grad_norm': '0.4194', 'learning_rate': '0.0001648', 'epoch': '2.003'}
+{'loss': '0.5329', 'grad_norm': '0.4044', 'learning_rate': '0.0001647', 'epoch': '2.003'}
+{'loss': '0.547', 'grad_norm': '0.4149', 'learning_rate': '0.0001647', 'epoch': '2.004'}
+{'loss': '0.5509', 'grad_norm': '0.4214', 'learning_rate': '0.0001647', 'epoch': '2.005'}
+{'loss': '0.5995', 'grad_norm': '0.451', 'learning_rate': '0.0001646', 'epoch': '2.006'}
+{'loss': '0.6348', 'grad_norm': '0.4232', 'learning_rate': '0.0001646', 'epoch': '2.007'}
+{'loss': '0.5488', 'grad_norm': '0.426', 'learning_rate': '0.0001646', 'epoch': '2.008'}
+{'loss': '0.5468', 'grad_norm': '0.5486', 'learning_rate': '0.0001645', 'epoch': '2.008'}
+{'loss': '0.6166', 'grad_norm': '0.4244', 'learning_rate': '0.0001645', 'epoch': '2.009'}
+{'loss': '0.5868', 'grad_norm': '0.4651', 'learning_rate': '0.0001644', 'epoch': '2.01'}
+{'loss': '0.5484', 'grad_norm': '0.4556', 'learning_rate': '0.0001644', 'epoch': '2.011'}
+{'loss': '0.6254', 'grad_norm': '0.4402', 'learning_rate': '0.0001644', 'epoch': '2.012'}
+{'loss': '0.6336', 'grad_norm': '0.5101', 'learning_rate': '0.0001643', 'epoch': '2.013'}
+{'loss': '0.6601', 'grad_norm': '0.5081', 'learning_rate': '0.0001643', 'epoch': '2.014'}
+{'loss': '0.6184', 'grad_norm': '0.4725', 'learning_rate': '0.0001643', 'epoch': '2.014'}
+{'loss': '0.6173', 'grad_norm': '0.4592', 'learning_rate': '0.0001642', 'epoch': '2.015'}
+{'loss': '0.6393', 'grad_norm': '0.4437', 'learning_rate': '0.0001642', 'epoch': '2.016'}
+{'loss': '0.6332', 'grad_norm': '0.4634', 'learning_rate': '0.0001641', 'epoch': '2.017'}
+{'loss': '0.5368', 'grad_norm': '0.4081', 'learning_rate': '0.0001641', 'epoch': '2.018'}
+{'loss': '0.4763', 'grad_norm': '0.3853', 'learning_rate': '0.0001641', 'epoch': '2.019'}
+{'loss': '0.651', 'grad_norm': '0.4522', 'learning_rate': '0.000164', 'epoch': '2.019'}
+{'loss': '0.5857', 'grad_norm': '0.4028', 'learning_rate': '0.000164', 'epoch': '2.02'}
+{'loss': '0.5961', 'grad_norm': '0.378', 'learning_rate': '0.000164', 'epoch': '2.021'}
+{'loss': '0.5697', 'grad_norm': '0.4297', 'learning_rate': '0.0001639', 'epoch': '2.022'}
+{'loss': '0.6611', 'grad_norm': '0.4522', 'learning_rate': '0.0001639', 'epoch': '2.023'}
+{'loss': '0.5388', 'grad_norm': '0.4118', 'learning_rate': '0.0001639', 'epoch': '2.024'}
+{'loss': '0.6615', 'grad_norm': '0.4894', 'learning_rate': '0.0001638', 'epoch': '2.024'}
+{'loss': '0.5514', 'grad_norm': '0.3742', 'learning_rate': '0.0001638', 'epoch': '2.025'}
+{'eval_loss': '0.7633', 'eval_runtime': '688.9', 'eval_samples_per_second': '3.059', 'eval_steps_per_second': '3.059', 'epoch': '2.025'}
+{'loss': '0.5733', 'grad_norm': '0.4576', 'learning_rate': '0.0001637', 'epoch': '2.026'}
+{'loss': '0.5994', 'grad_norm': '0.4009', 'learning_rate': '0.0001637', 'epoch': '2.027'}
+{'loss': '0.6229', 'grad_norm': '0.4491', 'learning_rate': '0.0001637', 'epoch': '2.028'}
+{'loss': '0.5126', 'grad_norm': '0.4153', 'learning_rate': '0.0001636', 'epoch': '2.029'}
+{'loss': '0.578', 'grad_norm': '0.4186', 'learning_rate': '0.0001636', 'epoch': '2.03'}
+{'loss': '0.5561', 'grad_norm': '0.4462', 'learning_rate': '0.0001636', 'epoch': '2.03'}
+{'loss': '0.5382', 'grad_norm': '0.4455', 'learning_rate': '0.0001635', 'epoch': '2.031'}
+{'loss': '0.6435', 'grad_norm': '0.453', 'learning_rate': '0.0001635', 'epoch': '2.032'}
+{'loss': '0.6316', 'grad_norm': '0.8912', 'learning_rate': '0.0001634', 'epoch': '2.033'}
+{'loss': '0.557', 'grad_norm': '0.4621', 'learning_rate': '0.0001634', 'epoch': '2.034'}
+{'loss': '0.5404', 'grad_norm': '0.3867', 'learning_rate': '0.0001634', 'epoch': '2.035'}
+{'loss': '0.5629', 'grad_norm': '0.3896', 'learning_rate': '0.0001633', 'epoch': '2.035'}
+{'loss': '0.5741', 'grad_norm': '0.4891', 'learning_rate': '0.0001633', 'epoch': '2.036'}
+{'loss': '0.6362', 'grad_norm': '0.4931', 'learning_rate': '0.0001633', 'epoch': '2.037'}
+{'loss': '0.5513', 'grad_norm': '0.4112', 'learning_rate': '0.0001632', 'epoch': '2.038'}
+{'loss': '0.5508', 'grad_norm': '0.482', 'learning_rate': '0.0001632', 'epoch': '2.039'}
+{'loss': '0.7608', 'grad_norm': '0.5131', 'learning_rate': '0.0001631', 'epoch': '2.04'}
+{'loss': '0.6045', 'grad_norm': '0.4455', 'learning_rate': '0.0001631', 'epoch': '2.041'}
+{'loss': '0.5879', 'grad_norm': '0.4689', 'learning_rate': '0.0001631', 'epoch': '2.041'}
+{'loss': '0.5822', 'grad_norm': '0.4547', 'learning_rate': '0.000163', 'epoch': '2.042'}
+{'loss': '0.5841', 'grad_norm': '0.467', 'learning_rate': '0.000163', 'epoch': '2.043'}
+{'loss': '0.5877', 'grad_norm': '0.4389', 'learning_rate': '0.000163', 'epoch': '2.044'}
+{'loss': '0.5628', 'grad_norm': '0.4144', 'learning_rate': '0.0001629', 'epoch': '2.045'}
+{'loss': '0.6157', 'grad_norm': '0.4742', 'learning_rate': '0.0001629', 'epoch': '2.046'}
+{'loss': '0.5762', 'grad_norm': '0.461', 'learning_rate': '0.0001628', 'epoch': '2.046'}
+{'loss': '0.6326', 'grad_norm': '0.4988', 'learning_rate': '0.0001628', 'epoch': '2.047'}
+{'loss': '0.6072', 'grad_norm': '0.4664', 'learning_rate': '0.0001628', 'epoch': '2.048'}
+{'loss': '0.6508', 'grad_norm': '0.5363', 'learning_rate': '0.0001627', 'epoch': '2.049'}
+{'loss': '0.6264', 'grad_norm': '0.4593', 'learning_rate': '0.0001627', 'epoch': '2.05'}
+{'loss': '0.6185', 'grad_norm': '0.4467', 'learning_rate': '0.0001627', 'epoch': '2.051'}
+{'loss': '0.5723', 'grad_norm': '0.4965', 'learning_rate': '0.0001626', 'epoch': '2.051'}
+{'loss': '0.5788', 'grad_norm': '0.4692', 'learning_rate': '0.0001626', 'epoch': '2.052'}
+{'loss': '0.6448', 'grad_norm': '0.4984', 'learning_rate': '0.0001625', 'epoch': '2.053'}
+{'loss': '0.6241', 'grad_norm': '0.4802', 'learning_rate': '0.0001625', 'epoch': '2.054'}
+{'loss': '0.4956', 'grad_norm': '0.3652', 'learning_rate': '0.0001625', 'epoch': '2.055'}
+{'loss': '0.6184', 'grad_norm': '0.5041', 'learning_rate': '0.0001624', 'epoch': '2.056'}
+{'loss': '0.5593', 'grad_norm': '0.4312', 'learning_rate': '0.0001624', 'epoch': '2.057'}
+{'loss': '0.6024', 'grad_norm': '0.5174', 'learning_rate': '0.0001624', 'epoch': '2.057'}
+{'loss': '0.5789', 'grad_norm': '0.4102', 'learning_rate': '0.0001623', 'epoch': '2.058'}
+{'loss': '0.6216', 'grad_norm': '0.5168', 'learning_rate': '0.0001623', 'epoch': '2.059'}
+{'loss': '0.5767', 'grad_norm': '0.5255', 'learning_rate': '0.0001622', 'epoch': '2.06'}
+{'loss': '0.5495', 'grad_norm': '0.3826', 'learning_rate': '0.0001622', 'epoch': '2.061'}
+{'loss': '0.5396', 'grad_norm': '0.481', 'learning_rate': '0.0001622', 'epoch': '2.062'}
+{'loss': '0.5737', 'grad_norm': '0.4237', 'learning_rate': '0.0001621', 'epoch': '2.062'}
+{'loss': '0.6151', 'grad_norm': '0.4641', 'learning_rate': '0.0001621', 'epoch': '2.063'}
+{'loss': '0.5868', 'grad_norm': '0.5054', 'learning_rate': '0.0001621', 'epoch': '2.064'}
+{'loss': '0.5435', 'grad_norm': '0.4553', 'learning_rate': '0.000162', 'epoch': '2.065'}
+{'loss': '0.6006', 'grad_norm': '0.494', 'learning_rate': '0.000162', 'epoch': '2.066'}
+{'loss': '0.6239', 'grad_norm': '0.4846', 'learning_rate': '0.0001619', 'epoch': '2.067'}
+{'loss': '0.5165', 'grad_norm': '0.4208', 'learning_rate': '0.0001619', 'epoch': '2.068'}
+{'eval_loss': '0.7677', 'eval_runtime': '676', 'eval_samples_per_second': '3.117', 'eval_steps_per_second': '3.117', 'epoch': '2.068'}
+{'loss': '0.6363', 'grad_norm': '0.4921', 'learning_rate': '0.0001619', 'epoch': '2.068'}
+{'loss': '0.5472', 'grad_norm': '0.462', 'learning_rate': '0.0001618', 'epoch': '2.069'}
+{'loss': '0.5588', 'grad_norm': '0.4513', 'learning_rate': '0.0001618', 'epoch': '2.07'}
+{'loss': '0.644', 'grad_norm': '0.4328', 'learning_rate': '0.0001618', 'epoch': '2.071'}
+{'loss': '0.5987', 'grad_norm': '0.444', 'learning_rate': '0.0001617', 'epoch': '2.072'}
+{'loss': '0.5382', 'grad_norm': '0.376', 'learning_rate': '0.0001617', 'epoch': '2.073'}
+{'loss': '0.574', 'grad_norm': '0.4258', 'learning_rate': '0.0001616', 'epoch': '2.073'}
+{'loss': '0.581', 'grad_norm': '0.4608', 'learning_rate': '0.0001616', 'epoch': '2.074'}
+{'loss': '0.6064', 'grad_norm': '0.4114', 'learning_rate': '0.0001616', 'epoch': '2.075'}
+{'loss': '0.5996', 'grad_norm': '0.4416', 'learning_rate': '0.0001615', 'epoch': '2.076'}
+{'loss': '0.6442', 'grad_norm': '0.554', 'learning_rate': '0.0001615', 'epoch': '2.077'}
+{'loss': '0.6176', 'grad_norm': '0.5103', 'learning_rate': '0.0001615', 'epoch': '2.078'}
+{'loss': '0.6109', 'grad_norm': '0.4423', 'learning_rate': '0.0001614', 'epoch': '2.078'}
+{'loss': '0.6194', 'grad_norm': '0.4499', 'learning_rate': '0.0001614', 'epoch': '2.079'}
+{'loss': '0.5284', 'grad_norm': '0.3636', 'learning_rate': '0.0001613', 'epoch': '2.08'}
+{'loss': '0.5338', 'grad_norm': '0.4603', 'learning_rate': '0.0001613', 'epoch': '2.081'}
+{'loss': '0.6192', 'grad_norm': '0.4317', 'learning_rate': '0.0001613', 'epoch': '2.082'}
+{'loss': '0.5953', 'grad_norm': '0.4334', 'learning_rate': '0.0001612', 'epoch': '2.083'}
+{'loss': '0.6766', 'grad_norm': '0.5603', 'learning_rate': '0.0001612', 'epoch': '2.084'}
+{'loss': '0.5476', 'grad_norm': '0.4264', 'learning_rate': '0.0001612', 'epoch': '2.084'}
+{'loss': '0.5703', 'grad_norm': '0.4163', 'learning_rate': '0.0001611', 'epoch': '2.085'}
+{'loss': '0.5692', 'grad_norm': '0.3922', 'learning_rate': '0.0001611', 'epoch': '2.086'}
+{'loss': '0.6015', 'grad_norm': '0.5189', 'learning_rate': '0.000161', 'epoch': '2.087'}
+{'loss': '0.6085', 'grad_norm': '0.5606', 'learning_rate': '0.000161', 'epoch': '2.088'}
+{'loss': '0.6078', 'grad_norm': '0.4685', 'learning_rate': '0.000161', 'epoch': '2.089'}
+{'loss': '0.5664', 'grad_norm': '0.4391', 'learning_rate': '0.0001609', 'epoch': '2.089'}
+{'loss': '0.5794', 'grad_norm': '0.4041', 'learning_rate': '0.0001609', 'epoch': '2.09'}
+{'loss': '0.6122', 'grad_norm': '0.4766', 'learning_rate': '0.0001608', 'epoch': '2.091'}
+{'loss': '0.6086', 'grad_norm': '0.4576', 'learning_rate': '0.0001608', 'epoch': '2.092'}
+{'loss': '0.5972', 'grad_norm': '0.4287', 'learning_rate': '0.0001608', 'epoch': '2.093'}
+{'loss': '0.6196', 'grad_norm': '0.5314', 'learning_rate': '0.0001607', 'epoch': '2.094'}
+{'loss': '0.6142', 'grad_norm': '0.5222', 'learning_rate': '0.0001607', 'epoch': '2.095'}
+{'loss': '0.6355', 'grad_norm': '0.4687', 'learning_rate': '0.0001607', 'epoch': '2.095'}
+{'loss': '0.5759', 'grad_norm': '0.4452', 'learning_rate': '0.0001606', 'epoch': '2.096'}
+{'loss': '0.5388', 'grad_norm': '0.4409', 'learning_rate': '0.0001606', 'epoch': '2.097'}
+{'loss': '0.5712', 'grad_norm': '0.4952', 'learning_rate': '0.0001605', 'epoch': '2.098'}
+{'loss': '0.6321', 'grad_norm': '0.489', 'learning_rate': '0.0001605', 'epoch': '2.099'}
+{'loss': '0.6141', 'grad_norm': '0.5183', 'learning_rate': '0.0001605', 'epoch': '2.1'}
+{'loss': '0.6079', 'grad_norm': '0.5345', 'learning_rate': '0.0001604', 'epoch': '2.1'}
+{'loss': '0.5635', 'grad_norm': '0.4366', 'learning_rate': '0.0001604', 'epoch': '2.101'}
+{'loss': '0.5708', 'grad_norm': '0.4405', 'learning_rate': '0.0001604', 'epoch': '2.102'}
+{'loss': '0.5635', 'grad_norm': '0.4343', 'learning_rate': '0.0001603', 'epoch': '2.103'}
+{'loss': '0.6325', 'grad_norm': '0.5538', 'learning_rate': '0.0001603', 'epoch': '2.104'}
+{'loss': '0.6509', 'grad_norm': '0.5426', 'learning_rate': '0.0001602', 'epoch': '2.105'}
+{'loss': '0.6154', 'grad_norm': '0.4604', 'learning_rate': '0.0001602', 'epoch': '2.105'}
+{'loss': '0.6845', 'grad_norm': '0.4866', 'learning_rate': '0.0001602', 'epoch': '2.106'}
+{'loss': '0.5982', 'grad_norm': '0.4754', 'learning_rate': '0.0001601', 'epoch': '2.107'}
+{'loss': '0.6041', 'grad_norm': '0.4135', 'learning_rate': '0.0001601', 'epoch': '2.108'}
+{'loss': '0.6491', 'grad_norm': '0.5325', 'learning_rate': '0.00016', 'epoch': '2.109'}
+{'loss': '0.6178', 'grad_norm': '0.4005', 'learning_rate': '0.00016', 'epoch': '2.11'}
+{'eval_loss': '0.7635', 'eval_runtime': '669.5', 'eval_samples_per_second': '3.147', 'eval_steps_per_second': '3.147', 'epoch': '2.11'}
+{'loss': '0.612', 'grad_norm': '0.3708', 'learning_rate': '0.00016', 'epoch': '2.111'}
+{'loss': '0.5535', 'grad_norm': '0.4098', 'learning_rate': '0.0001599', 'epoch': '2.111'}
+{'loss': '0.6166', 'grad_norm': '0.5794', 'learning_rate': '0.0001599', 'epoch': '2.112'}
+{'loss': '0.5814', 'grad_norm': '0.5013', 'learning_rate': '0.0001599', 'epoch': '2.113'}
+{'loss': '0.6293', 'grad_norm': '0.4189', 'learning_rate': '0.0001598', 'epoch': '2.114'}
+{'loss': '0.6219', 'grad_norm': '0.544', 'learning_rate': '0.0001598', 'epoch': '2.115'}
+{'loss': '0.6048', 'grad_norm': '0.4918', 'learning_rate': '0.0001597', 'epoch': '2.116'}
+{'loss': '0.5887', 'grad_norm': '0.5027', 'learning_rate': '0.0001597', 'epoch': '2.116'}
+{'loss': '0.619', 'grad_norm': '0.429', 'learning_rate': '0.0001597', 'epoch': '2.117'}
+{'loss': '0.6218', 'grad_norm': '0.4429', 'learning_rate': '0.0001596', 'epoch': '2.118'}
+{'loss': '0.6325', 'grad_norm': '0.4629', 'learning_rate': '0.0001596', 'epoch': '2.119'}
+{'loss': '0.6053', 'grad_norm': '0.4349', 'learning_rate': '0.0001595', 'epoch': '2.12'}
+{'loss': '0.583', 'grad_norm': '0.4363', 'learning_rate': '0.0001595', 'epoch': '2.121'}
+{'loss': '0.6273', 'grad_norm': '0.4553', 'learning_rate': '0.0001595', 'epoch': '2.122'}
+{'loss': '0.6343', 'grad_norm': '0.5445', 'learning_rate': '0.0001594', 'epoch': '2.122'}
+{'loss': '0.6358', 'grad_norm': '0.4698', 'learning_rate': '0.0001594', 'epoch': '2.123'}
+{'loss': '0.6478', 'grad_norm': '0.5375', 'learning_rate': '0.0001594', 'epoch': '2.124'}
+{'loss': '0.5516', 'grad_norm': '0.4433', 'learning_rate': '0.0001593', 'epoch': '2.125'}
+{'loss': '0.7512', 'grad_norm': '0.5592', 'learning_rate': '0.0001593', 'epoch': '2.126'}
+{'loss': '0.615', 'grad_norm': '0.4787', 'learning_rate': '0.0001592', 'epoch': '2.127'}
+{'loss': '0.6197', 'grad_norm': '0.476', 'learning_rate': '0.0001592', 'epoch': '2.127'}
+{'loss': '0.655', 'grad_norm': '0.4995', 'learning_rate': '0.0001592', 'epoch': '2.128'}
+{'loss': '0.6073', 'grad_norm': '0.4955', 'learning_rate': '0.0001591', 'epoch': '2.129'}
+{'loss': '0.6082', 'grad_norm': '0.4816', 'learning_rate': '0.0001591', 'epoch': '2.13'}
+{'loss': '0.5966', 'grad_norm': '0.4318', 'learning_rate': '0.000159', 'epoch': '2.131'}
+{'loss': '0.5764', 'grad_norm': '0.417', 'learning_rate': '0.000159', 'epoch': '2.132'}
+{'loss': '0.6827', 'grad_norm': '0.4895', 'learning_rate': '0.000159', 'epoch': '2.132'}
+{'loss': '0.5684', 'grad_norm': '0.4227', 'learning_rate': '0.0001589', 'epoch': '2.133'}
+{'loss': '0.5257', 'grad_norm': '0.4379', 'learning_rate': '0.0001589', 'epoch': '2.134'}
+{'loss': '0.5873', 'grad_norm': '0.4932', 'learning_rate': '0.0001588', 'epoch': '2.135'}
+{'loss': '0.6437', 'grad_norm': '0.4796', 'learning_rate': '0.0001588', 'epoch': '2.136'}
+{'loss': '0.6244', 'grad_norm': '0.4963', 'learning_rate': '0.0001588', 'epoch': '2.137'}
+{'loss': '0.6575', 'grad_norm': '0.515', 'learning_rate': '0.0001587', 'epoch': '2.138'}
+{'loss': '0.6086', 'grad_norm': '0.5272', 'learning_rate': '0.0001587', 'epoch': '2.138'}
+{'loss': '0.6214', 'grad_norm': '0.4663', 'learning_rate': '0.0001587', 'epoch': '2.139'}
+{'loss': '0.5942', 'grad_norm': '0.4423', 'learning_rate': '0.0001586', 'epoch': '2.14'}
+{'loss': '0.5674', 'grad_norm': '0.4166', 'learning_rate': '0.0001586', 'epoch': '2.141'}
+{'loss': '0.6162', 'grad_norm': '0.4777', 'learning_rate': '0.0001585', 'epoch': '2.142'}
+{'loss': '0.6195', 'grad_norm': '0.4656', 'learning_rate': '0.0001585', 'epoch': '2.143'}
+{'loss': '0.6183', 'grad_norm': '0.4626', 'learning_rate': '0.0001585', 'epoch': '2.143'}
+{'loss': '0.5486', 'grad_norm': '0.387', 'learning_rate': '0.0001584', 'epoch': '2.144'}
+{'loss': '0.6416', 'grad_norm': '0.4834', 'learning_rate': '0.0001584', 'epoch': '2.145'}
+{'loss': '0.6127', 'grad_norm': '0.5227', 'learning_rate': '0.0001583', 'epoch': '2.146'}
+{'loss': '0.6021', 'grad_norm': '0.532', 'learning_rate': '0.0001583', 'epoch': '2.147'}
+{'loss': '0.6051', 'grad_norm': '0.495', 'learning_rate': '0.0001583', 'epoch': '2.148'}
+{'loss': '0.5923', 'grad_norm': '0.5532', 'learning_rate': '0.0001582', 'epoch': '2.149'}
+{'loss': '0.6607', 'grad_norm': '0.4462', 'learning_rate': '0.0001582', 'epoch': '2.149'}
+{'loss': '0.6294', 'grad_norm': '0.4643', 'learning_rate': '0.0001581', 'epoch': '2.15'}
+{'loss': '0.5308', 'grad_norm': '0.4624', 'learning_rate': '0.0001581', 'epoch': '2.151'}
+{'loss': '0.6139', 'grad_norm': '0.4882', 'learning_rate': '0.0001581', 'epoch': '2.152'}
+{'eval_loss': '0.7655', 'eval_runtime': '681.5', 'eval_samples_per_second': '3.092', 'eval_steps_per_second': '3.092', 'epoch': '2.152'}
+{'train_runtime': '7.398e+04', 'train_samples_per_second': '1.538', 'train_steps_per_second': '0.192', 'train_loss': '0.8077', 'epoch': '2.152'}
+Saved best adapter -> task2file/sft_qwen_14B_v2/best_adapter
+Final eval_loss=0.7600579857826233, ppl=2.138400213711816
+--- Merge: task2file/sft_qwen_14B_v2/best_adapter + Models/Qwen2.5-Coder-14B-CPT -> task2file/sft_qwen_14B_v2/Models/Qwen2.5-Coder-14B-CPT-SFT_v2 ---
+Loading weights: 100%|█| 579/579 [00:00<00:00, 1040.59it/s, Materializing param=model
+Writing model shards: 100%|██████████████████████████| 16/16 [00:46<00:00,  2.93s/it]
+--- Merge complete ---
diff --git a/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/files/requirements.txt b/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/files/requirements.txt
new file mode 100644
index 0000000000000000000000000000000000000000..79a4241d8724f018c9bdfcd7c289f1f14578574b
--- /dev/null
+++ b/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/files/requirements.txt
@@ -0,0 +1,104 @@
+exceptiongroup==1.3.1
+wheel==0.45.1
+python-dateutil==2.9.0.post0
+nvidia-ml-py==13.580.82
+huggingface_hub==1.2.3
+idna==3.11
+click==8.3.1
+numpy==2.2.6
+httpx==0.28.1
+tokenizers==0.22.1
+sympy==1.13.1
+yarl==1.22.0
+async-timeout==5.0.1
+datasets==4.4.2
+platformdirs==4.5.1
+nvidia-cuda-cupti-cu12==12.1.105
+nvidia-nvtx-cu12==12.1.105
+smmap==5.0.2
+accelerate==1.12.0
+requests==2.32.5
+aiohttp==3.13.2
+bitsandbytes==0.49.0
+nvidia-cublas-cu12==12.1.3.1
+mpmath==1.3.0
+torchaudio==2.5.1+cu121
+nvidia-cuda-runtime-cu12==12.1.105
+typing-inspection==0.4.2
+GitPython==3.1.45
+xxhash==3.6.0
+nvidia-cusolver-cu12==11.4.5.107
+pydantic_core==2.41.5
+six==1.17.0
+torchvision==0.20.1+cu121
+typing_extensions==4.15.0
+triton==3.1.0
+charset-normalizer==3.4.4
+nvitop==1.6.1
+wandb==0.23.1
+regex==2025.11.3
+pip==25.3
+nvidia-cusparse-cu12==12.1.0.106
+pytz==2025.2
+Jinja2==3.1.6
+psutil==7.2.0
+pillow==12.0.0
+packaging==25.0
+safetensors==0.7.0
+sentry-sdk==2.48.0
+gitdb==4.0.12
+httpcore==1.0.9
+setuptools==80.9.0
+nvidia-cufft-cu12==11.0.2.54
+anyio==4.12.0
+transformers==5.0.0.dev0
+pydantic==2.12.5
+fsspec==2025.10.0
+filelock==3.20.0
+PyYAML==6.0.3
+hf-xet==1.2.0
+nvidia-cudnn-cu12==9.1.0.70
+tqdm==4.67.1
+MarkupSafe==2.1.5
+attrs==25.4.0
+nvidia-cuda-nvrtc-cu12==12.1.105
+peft==0.18.0
+aiohappyeyeballs==2.6.1
+networkx==3.4.2
+nvidia-nvjitlink-cu12==12.9.86
+certifi==2025.11.12
+pyarrow==22.0.0
+dill==0.4.0
+protobuf==6.33.2
+aiosignal==1.4.0
+frozenlist==1.8.0
+urllib3==2.6.2
+propcache==0.4.1
+tzdata==2025.3
+pandas==2.3.3
+annotated-types==0.7.0
+shellingham==1.5.4
+nvidia-nccl-cu12==2.21.5
+multidict==6.7.0
+nvidia-curand-cu12==10.3.2.106
+trl==0.26.2
+torch==2.5.1+cu121
+h11==0.16.0
+multiprocess==0.70.18
+typer-slim==0.21.0
+wheel==0.45.1
+tomli==2.0.1
+autocommand==2.2.2
+jaraco.context==5.3.0
+zipp==3.19.2
+packaging==24.2
+inflect==7.3.1
+typing_extensions==4.12.2
+platformdirs==4.2.2
+jaraco.functools==4.0.1
+jaraco.collections==5.1.0
+jaraco.text==3.12.1
+backports.tarfile==1.2.0
+more-itertools==10.3.0
+importlib_metadata==8.0.0
+typeguard==4.3.0
diff --git a/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/files/wandb-metadata.json b/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/files/wandb-metadata.json
new file mode 100644
index 0000000000000000000000000000000000000000..110c58f2dcf4f6589c4cc21ec0d9d3c37e780282
--- /dev/null
+++ b/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/files/wandb-metadata.json
@@ -0,0 +1,47 @@
+{
+  "os":  "Linux-6.12.46+-x86_64-with-glibc2.35",
+  "python":  "CPython 3.10.12",
+  "startedAt":  "2025-12-26T18:15:44.765252Z",
+  "args":  [
+    "--config",
+    "trainer-kit/SFT-14b/config_instruct.yaml"
+  ],
+  "program":  "/workspace/trainer-kit/SFT-14b/run_instruct.py",
+  "codePath":  "trainer-kit/SFT-14b/run_instruct.py",
+  "codePathLocal":  "trainer-kit/SFT-14b/run_instruct.py",
+  "email":  "shaiksirajuddin9949@gmail.com",
+  "root":  "task2file/sft_qwen_14B_v2",
+  "host":  "a100-2gpu-shell-session-757d587799-mfdvv",
+  "executable":  "/workspace/llm_finetuning_env/bin/python",
+  "cpu_count":  12,
+  "cpu_count_logical":  24,
+  "gpu":  "NVIDIA A100-SXM4-80GB",
+  "gpu_count":  2,
+  "disk":  {
+    "/":  {
+      "total":  "791251738624",
+      "used":  "392925650944"
+    }
+  },
+  "memory":  {
+    "total":  "359047892992"
+  },
+  "gpu_nvidia":  [
+    {
+      "name":  "NVIDIA A100-SXM4-80GB",
+      "memoryTotal":  "85899345920",
+      "cudaCores":  6912,
+      "architecture":  "Ampere",
+      "uuid":  "GPU-989794b0-ec3b-13bf-db9f-3fbe341497ba"
+    },
+    {
+      "name":  "NVIDIA A100-SXM4-80GB",
+      "memoryTotal":  "85899345920",
+      "cudaCores":  6912,
+      "architecture":  "Ampere",
+      "uuid":  "GPU-3790aa64-60ef-9eac-b0b1-b278ee8c0d40"
+    }
+  ],
+  "cudaVersion":  "13.0",
+  "writerId":  "ba9dsvvs7npkm5vvx6733495pp0yghz4"
+}
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/files/wandb-summary.json b/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/files/wandb-summary.json
new file mode 100644
index 0000000000000000000000000000000000000000..a618bfa6ccf3671ed937fecd63d0e59b80afb085
--- /dev/null
+++ b/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/files/wandb-summary.json
@@ -0,0 +1 @@
+{"train_steps_per_second":0.192,"train_loss":0.8076860591944526,"eval/steps_per_second":3.126,"_runtime":74806,"train/learning_rate":0.00015806910080135527,"train_runtime":73982.3221,"total_flos":3.049019256156893e+18,"eval/runtime":674.048,"train/epoch":2.151898734177215,"train/global_step":5100,"eval/loss":0.7600579857826233,"_step":2602,"_wandb":{"runtime":74806},"train/loss":0.6138747930526733,"train/grad_norm":0.48821282386779785,"eval/samples_per_second":3.126,"_timestamp":1.766847687515407e+09,"train_samples_per_second":1.538}
\ No newline at end of file
diff --git a/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/logs/debug-core.log b/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/logs/debug-core.log
new file mode 100644
index 0000000000000000000000000000000000000000..1bfbfdcc866b8126318e0b315e866c8e7c25c8c0
--- /dev/null
+++ b/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/logs/debug-core.log
@@ -0,0 +1,16 @@
+{"time":"2025-12-26T18:15:44.85463749Z","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmp8sddanmm/port-194421.txt","pid":194421,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2025-12-26T18:15:44.855355082Z","level":"INFO","msg":"server: will exit if parent process dies","ppid":194421}
+{"time":"2025-12-26T18:15:44.85534748Z","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-194421-194516-4014985148/socket","Net":"unix"}}
+{"time":"2025-12-26T18:15:45.035681883Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2025-12-26T18:15:45.043999704Z","level":"INFO","msg":"handleInformInit: received","streamId":"upub1jan","id":"1(@)"}
+{"time":"2025-12-26T18:15:45.212475024Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"upub1jan","id":"1(@)"}
+{"time":"2025-12-27T15:02:32.280521859Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"upub1jan","id":"1(@)"}
+{"time":"2025-12-27T15:02:32.281560524Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"upub1jan","id":"1(@)"}
+{"time":"2025-12-27T15:02:32.336985081Z","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"1(@)"}
+{"time":"2025-12-27T15:02:32.337047862Z","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"1(@)"}
+{"time":"2025-12-27T15:02:32.337066801Z","level":"INFO","msg":"server is shutting down"}
+{"time":"2025-12-27T15:02:32.337073486Z","level":"INFO","msg":"connection: closing","id":"1(@)"}
+{"time":"2025-12-27T15:02:32.33713325Z","level":"INFO","msg":"connection: closed successfully","id":"1(@)"}
+{"time":"2025-12-27T15:02:32.337139996Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"1(@)"}
+{"time":"2025-12-27T15:02:32.337231073Z","level":"INFO","msg":"server: listener closed","addr":{"Name":"/tmp/wandb-194421-194516-4014985148/socket","Net":"unix"}}
+{"time":"2025-12-27T15:02:32.337292822Z","level":"INFO","msg":"server is closed"}
diff --git a/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/logs/debug-internal.log b/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/logs/debug-internal.log
new file mode 100644
index 0000000000000000000000000000000000000000..15b2d6d11efb0a8c278e35049a993476ffa1baaf
--- /dev/null
+++ b/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/logs/debug-internal.log
@@ -0,0 +1,12 @@
+{"time":"2025-12-26T18:15:45.044149374Z","level":"INFO","msg":"stream: starting","core version":"0.23.1"}
+{"time":"2025-12-26T18:15:45.212148231Z","level":"INFO","msg":"stream: created new stream","id":"upub1jan"}
+{"time":"2025-12-26T18:15:45.212312297Z","level":"INFO","msg":"handler: started","stream_id":"upub1jan"}
+{"time":"2025-12-26T18:15:45.212463318Z","level":"INFO","msg":"stream: started","id":"upub1jan"}
+{"time":"2025-12-26T18:15:45.212498387Z","level":"INFO","msg":"writer: started","stream_id":"upub1jan"}
+{"time":"2025-12-26T18:15:45.212503642Z","level":"INFO","msg":"sender: started","stream_id":"upub1jan"}
+{"time":"2025-12-27T15:02:32.177112089Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
+{"time":"2025-12-27T15:02:32.275585886Z","level":"INFO","msg":"handler: operation stats","stats":{}}
+{"time":"2025-12-27T15:02:32.280819216Z","level":"INFO","msg":"stream: closing","id":"upub1jan"}
+{"time":"2025-12-27T15:02:32.280845154Z","level":"INFO","msg":"handler: closed","stream_id":"upub1jan"}
+{"time":"2025-12-27T15:02:32.280896164Z","level":"INFO","msg":"sender: closed","stream_id":"upub1jan"}
+{"time":"2025-12-27T15:02:32.280915121Z","level":"INFO","msg":"stream: closed","id":"upub1jan"}
diff --git a/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/logs/debug.log b/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/logs/debug.log
new file mode 100644
index 0000000000000000000000000000000000000000..87293499cc1e7f690ab9960d95a2b5911253be42
--- /dev/null
+++ b/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/logs/debug.log
@@ -0,0 +1,29 @@
+2025-12-26 18:15:44,766 INFO    MainThread:194421 [wandb_setup.py:_flush():80] Current SDK version is 0.23.1
+2025-12-26 18:15:44,767 INFO    MainThread:194421 [wandb_setup.py:_flush():80] Configure stats pid to 194421
+2025-12-26 18:15:44,767 INFO    MainThread:194421 [wandb_setup.py:_flush():80] Loading settings from /root/.config/wandb/settings
+2025-12-26 18:15:44,767 INFO    MainThread:194421 [wandb_setup.py:_flush():80] Loading settings from /workspace/wandb/settings
+2025-12-26 18:15:44,767 INFO    MainThread:194421 [wandb_setup.py:_flush():80] Loading settings from environment variables
+2025-12-26 18:15:44,767 INFO    MainThread:194421 [wandb_init.py:setup_run_log_directory():714] Logging user logs to task2file/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/logs/debug.log
+2025-12-26 18:15:44,767 INFO    MainThread:194421 [wandb_init.py:setup_run_log_directory():715] Logging internal logs to task2file/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/logs/debug-internal.log
+2025-12-26 18:15:44,767 INFO    MainThread:194421 [wandb_init.py:init():841] calling init triggers
+2025-12-26 18:15:44,767 INFO    MainThread:194421 [wandb_init.py:init():846] wandb.init called with sweep_config: {}
+config: {'model': {'repo_id': './Models/Qwen2.5-Coder-14B-CPT', 'revision': None, 'base_local_dir': 'base_model', 'trust_remote_code': True, 'tokenizer_use_fast': True, 'device_map': 'auto', 'torch_dtype': 'bfloat16', 'use_4bit': False, 'bnb_4bit_quant_type': 'nf4', 'bnb_4bit_use_double_quant': False, 'bnb_4bit_compute_dtype': 'bfloat16', 'attn_implementation': None}, 'data': {'train_jsonl': 'sft_dataset.jsonl', 'eval_jsonl': None, 'eval_split_ratio': 0.1, 'instruction_field': 'instruction', 'input_field': 'input', 'output_field': 'output', 'format_type': 'custom', 'system_prompt': 'You are a Hyperswitch Rust code analyzer. Identify functions/structs that need modification for a given task.\n\n## Output Format\n\n##OUTPUT\nExplain the data flow and why each component must change:\n- Flow: [Input → Processing → Output with arrows]\n- For each component: "The [ComponentName] ([path]) must [action] because [reason]—without this, [consequence]"\n- Explain coupling between components\n\n##SELECT\nmodify::crates/path/to/file.rs::impl::ComponentName\nadd::crates/another/file.rs::function::AnotherComponent\n<EOS>\n\n## Rules\n\n1. Use full paths: `remove::crates/folder/file.rs::Type::Name`\n2. Use `::` for nested items: `status::StructName::Type::Name`\n3. Always explain "must change because" and "without this"\n3. Types of components: function, struct, enum, impl, trait\n4. If there is extra information (e.g., enum variants), include that too.\n5. Start with ##OUTPUT, end with ##SELECT, terminate with <EOS>\n\n## Example\n\n##TASK\nAdd webhook subscription support\n\n##OUTPUT\nThe webhook system routes events via EventClass enum. Flow: webhook → EventClass → handler → processing. The EventClass enum (crates/common_enums/src/enums.rs::EventClass) must add Subscriptions variant because it defines event routing—without this, subscription events cannot be processed. The SubscriptionStatus impl (crates/common_enums/src/transformers.rs::SubscriptionStatus) must map to EventType because it converts status to events—without this, status changes don\'t trigger webhooks. These are coupled: EventClass routes to handlers that use SubscriptionStatus mappings.\n\n##SELECT\ncrates/common_enums/src/enums.rs::EventClass\ncrates/common_enums/src/transformers.rs::SubscriptionStatus\n<EOS>\n', 'custom_template': '##INSTRUCTION\n{instruction}<|im_end|>\n{input}<|im_end|>\n{output}<|im_end|>', 'max_length': 2048, 'shuffle': True, 'num_proc': 4}, 'peft': {'enabled': True, 'r': 32, 'lora_alpha': 64, 'lora_dropout': 0.05, 'bias': 'none', 'target_modules': 'auto'}, 'train': {'num_train_epochs': 6, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'gradient_accumulation_steps': 8, 'learning_rate': '2e-4', 'weight_decay': 0.0, 'warmup_ratio': 0.08, 'lr_scheduler_type': 'cosine', 'optim': 'adamw_torch', 'max_grad_norm': 1.0, 'gradient_checkpointing': True, 'logging_steps': 2, 'save_strategy': 'steps', 'save_steps': 500, 'save_total_limit': 20, 'evaluation_strategy': 'steps', 'eval_steps': 100, 'load_best_model_at_end': True, 'early_stopping': {'enabled': True, 'patience': 5, 'min_delta': 0.001, 'metric': 'eval_loss', 'mode': 'min'}, 'resume_from_checkpoint': 'auto'}, 'run_dir': 'task2file/sft_qwen_14B_v2', '_wandb': {}}
+2025-12-26 18:15:44,767 INFO    MainThread:194421 [wandb_init.py:init():889] starting backend
+2025-12-26 18:15:45,035 INFO    MainThread:194421 [wandb_init.py:init():892] sending inform_init request
+2025-12-26 18:15:45,040 INFO    MainThread:194421 [wandb_init.py:init():900] backend started and connected
+2025-12-26 18:15:45,044 INFO    MainThread:194421 [wandb_init.py:init():970] updated telemetry
+2025-12-26 18:15:45,045 INFO    MainThread:194421 [wandb_init.py:init():994] communicating run to backend with 90.0 second timeout
+2025-12-26 18:15:45,420 INFO    MainThread:194421 [wandb_init.py:init():1041] starting run threads in backend
+2025-12-26 18:15:45,537 INFO    MainThread:194421 [wandb_run.py:_console_start():2521] atexit reg
+2025-12-26 18:15:45,537 INFO    MainThread:194421 [wandb_run.py:_redirect():2369] redirect: wrap_raw
+2025-12-26 18:15:45,537 INFO    MainThread:194421 [wandb_run.py:_redirect():2438] Wrapping output streams.
+2025-12-26 18:15:45,537 INFO    MainThread:194421 [wandb_run.py:_redirect():2461] Redirects installed.
+2025-12-26 18:15:45,542 INFO    MainThread:194421 [wandb_init.py:init():1081] run started, returning control to user process
+2025-12-26 18:17:10,652 INFO    MainThread:194421 [wandb_run.py:_config_callback():1396] config_cb None None {'peft_config': {'default': {'task_type': 'CAUSAL_LM', 'peft_type': 'LORA', 'auto_mapping': None, 'peft_version': '0.18.0', 'base_model_name_or_path': 'Models/Qwen2.5-Coder-14B-CPT', 'revision': None, 'inference_mode': False, 'r': 32, 'target_modules': ['o_proj', 'v_proj', 'k_proj', 'q_proj'], 'exclude_modules': None, 'lora_alpha': 64, 'lora_dropout': 0.05, 'fan_in_fan_out': False, 'bias': 'none', 'use_rslora': False, 'modules_to_save': None, 'init_lora_weights': True, 'layers_to_transform': None, 'layers_pattern': None, 'rank_pattern': {}, 'alpha_pattern': {}, 'megatron_config': None, 'megatron_core': 'megatron.core', 'trainable_token_indices': None, 'loftq_config': {}, 'eva_config': None, 'corda_config': None, 'use_dora': False, 'alora_invocation_tokens': None, 'use_qalora': False, 'qalora_group_size': 16, 'layer_replication': None, 'runtime_config': {'ephemeral_gpu_offload': False}, 'lora_bias': False, 'target_parameters': None, 'arrow_config': None, 'ensure_weight_tying': False}}, 'vocab_size': 152064, 'max_position_embeddings': 32768, 'hidden_size': 5120, 'intermediate_size': 13824, 'num_hidden_layers': 48, 'num_attention_heads': 40, 'use_sliding_window': False, 'sliding_window': None, 'max_window_layers': 48, 'num_key_value_heads': 8, 'hidden_act': 'silu', 'initializer_range': 0.02, 'rms_norm_eps': 1e-06, 'use_cache': False, 'attention_dropout': 0.0, 'layer_types': ['full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention', 'full_attention'], 'rope_parameters': {'rope_theta': 1000000.0, 'rope_type': 'default'}, 'return_dict': True, 'output_hidden_states': False, 'dtype': 'bfloat16', 'tie_word_embeddings': False, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'architectures': ['Qwen2ForCausalLM'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'task_specific_params': None, 'problem_type': None, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 151643, 'pad_token_id': None, 'eos_token_id': 151643, 'sep_token_id': None, 'decoder_start_token_id': None, '_name_or_path': 'Models/Qwen2.5-Coder-14B-CPT', 'transformers_version': '5.0.0.dev0', 'model_type': 'qwen2', 'output_attentions': False, 'output_dir': 'task2file/sft_qwen_14B_v2/checkpoints', 'do_train': False, 'do_eval': True, 'do_predict': False, 'eval_strategy': 'steps', 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'gradient_accumulation_steps': 8, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 0.0002, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 6.0, 'max_steps': -1, 'lr_scheduler_type': 'cosine', 'lr_scheduler_kwargs': None, 'warmup_ratio': 0.08, 'warmup_steps': 0.08, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': None, 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 2, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 500, 'save_total_limit': 20, 'enable_jit_checkpoint': False, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'use_cpu': False, 'seed': 42, 'data_seed': None, 'bf16': True, 'fp16': False, 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': -1, 'ddp_backend': None, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': 100, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'run_name': None, 'disable_tqdm': False, 'remove_unused_columns': False, 'label_names': None, 'load_best_model_at_end': True, 'metric_for_best_model': 'eval_loss', 'greater_is_better': False, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'parallelism_config': None, 'deepspeed': None, 'label_smoothing_factor': 0.0, 'optim': 'adamw_torch', 'optim_args': None, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'project': 'huggingface', 'trackio_space_id': 'trackio', 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'push_to_hub': False, 'resume_from_checkpoint': None, 'hub_model_id': None, 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_always_push': False, 'hub_revision': None, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'auto_find_batch_size': False, 'full_determinism': False, 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_num_input_tokens_seen': 'no', 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': False, 'liger_kernel_config': None, 'eval_use_gather_object': False, 'average_tokens_across_devices': True}
+2025-12-26 18:17:10,660 INFO    MainThread:194421 [wandb_config.py:__setitem__():154] [no run ID] config set model/num_parameters = 14820365312 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x7ed24f0556c0>>
+2025-12-26 18:17:10,660 INFO    MainThread:194421 [wandb_run.py:_config_callback():1396] config_cb model/num_parameters 14820365312 None
+2025-12-27 15:02:31,903 INFO    MainThread:194421 [wandb_run.py:_finish():2287] finishing run sirajuddin-shaik-007/sft-training/upub1jan
+2025-12-27 15:02:31,905 INFO    MainThread:194421 [wandb_run.py:_atexit_cleanup():2486] got exitcode: 0
+2025-12-27 15:02:31,906 INFO    MainThread:194421 [wandb_run.py:_restore():2468] restore
+2025-12-27 15:02:31,906 INFO    MainThread:194421 [wandb_run.py:_restore():2474] restore done
+2025-12-27 15:02:32,279 INFO    MainThread:194421 [wandb_run.py:_footer_sync_info():3862] logging synced files
diff --git a/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/run-upub1jan.wandb b/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/run-upub1jan.wandb
new file mode 100644
index 0000000000000000000000000000000000000000..995064203d546210fe2c964f18b6118cc738d3e8
--- /dev/null
+++ b/sft_qwen_14B_v2/wandb/run-20251226_181544-upub1jan/run-upub1jan.wandb
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba81531ad81e50fa84b5e419944d7823f811008e4c5c537862fe28d5e9fc7b63
+size 42375196