commit 4360715cbd1fea3f933edb0e0a6caa6f739ab8a3
Author: ModelHub XC <noreply@modelhub.org.cn>
Date:   Thu Jun 4 10:56:16 2026 +0800

    初始化项目，由ModelHub XC社区提供模型
    
    Model: ferrazzipietro/unsup-Llama-3.2-1B-Instruct-datav2-3ep
    Source: Original Platform

diff --git a/.gitattributes b/.gitattributes
new file mode 100644
index 0000000..52373fe
--- /dev/null
+++ b/.gitattributes
@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/README.md b/README.md
new file mode 100644
index 0000000..bc5f30d
--- /dev/null
+++ b/README.md
@@ -0,0 +1,199 @@
+---
+library_name: transformers
+tags: []
+---
+
+# 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. -->
+
+This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
+
+- **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]
\ No newline at end of file
diff --git a/config.json b/config.json
new file mode 100644
index 0000000..9802341
--- /dev/null
+++ b/config.json
@@ -0,0 +1,39 @@
+{
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 128000,
+  "eos_token_id": [
+    128001,
+    128008,
+    128009
+  ],
+  "head_dim": 64,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "max_position_embeddings": 131072,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 16,
+  "num_key_value_heads": 8,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": {
+    "factor": 32.0,
+    "high_freq_factor": 4.0,
+    "low_freq_factor": 1.0,
+    "original_max_position_embeddings": 8192,
+    "rope_type": "llama3"
+  },
+  "rope_theta": 500000.0,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.0",
+  "use_cache": true,
+  "vocab_size": 128256
+}
diff --git a/generation_config.json b/generation_config.json
new file mode 100644
index 0000000..60d1426
--- /dev/null
+++ b/generation_config.json
@@ -0,0 +1,10 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 128000,
+  "eos_token_id": [
+    128001,
+    128008,
+    128009
+  ],
+  "transformers_version": "4.51.0"
+}
diff --git a/model.safetensors b/model.safetensors
new file mode 100644
index 0000000..0d7d3dd
--- /dev/null
+++ b/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a0c82d0e1f62537eed47b0ccfb8e52394d0f7b6dd568465ed04deb3e7a976b75
+size 2996982344
diff --git a/special_tokens_map.json b/special_tokens_map.json
new file mode 100644
index 0000000..b43be96
--- /dev/null
+++ b/special_tokens_map.json
@@ -0,0 +1,17 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|eot_id|>"
+}
diff --git a/tokenizer.json b/tokenizer.json
new file mode 100644
index 0000000..1c1d8d5
--- /dev/null
+++ b/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
+size 17209920
diff --git a/tokenizer_config.json b/tokenizer_config.json
new file mode 100644
index 0000000..7739bcb
--- /dev/null
+++ b/tokenizer_config.json
@@ -0,0 +1,2065 @@
+{
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|finetune_right_pad_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|eom_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<|python_tag|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{{- bos_token }}\n{%- if custom_tools is defined %}\n    {%- set tools = custom_tools %}\n{%- endif %}\n{%- if not tools_in_user_message is defined %}\n    {%- set tools_in_user_message = true %}\n{%- endif %}\n{%- if not date_string is defined %}\n    {%- if strftime_now is defined %}\n        {%- set date_string = strftime_now(\"%d %b %Y\") %}\n    {%- else %}\n        {%- set date_string = \"26 Jul 2024\" %}\n    {%- endif %}\n{%- endif %}\n{%- if not tools is defined %}\n    {%- set tools = none %}\n{%- endif %}\n\n{#- This block extracts the system message, so we can slot it into the right place. #}\n{%- if messages[0]['role'] == 'system' %}\n    {%- set system_message = messages[0]['content']|trim %}\n    {%- set messages = messages[1:] %}\n{%- else %}\n    {%- set system_message = \"\" %}\n{%- endif %}\n\n{#- System message #}\n{{- \"<|start_header_id|>system<|end_header_id|>\\n\\n\" }}\n{%- if tools is not none %}\n    {{- \"Environment: ipython\\n\" }}\n{%- endif %}\n{{- \"Cutting Knowledge Date: December 2023\\n\" }}\n{{- \"Today Date: \" + date_string + \"\\n\\n\" }}\n{%- if tools is not none and not tools_in_user_message %}\n    {{- \"You have access to the following functions. To call a function, please respond with JSON for a function call.\" }}\n    {{- 'Respond in the format {\"name\": function name, \"parameters\": dictionary of argument name and its value}.' }}\n    {{- \"Do not use variables.\\n\\n\" }}\n    {%- for t in tools %}\n        {{- t | tojson(indent=4) }}\n        {{- \"\\n\\n\" }}\n    {%- endfor %}\n{%- endif %}\n{{- system_message }}\n{{- \"<|eot_id|>\" }}\n\n{#- Custom tools are passed in a user message with some extra guidance #}\n{%- if tools_in_user_message and not tools is none %}\n    {#- Extract the first user message so we can plug it in here #}\n    {%- if messages | length != 0 %}\n        {%- set first_user_message = messages[0]['content']|trim %}\n        {%- set messages = messages[1:] %}\n    {%- else %}\n        {{- raise_exception(\"Cannot put tools in the first user message when there's no first user message!\") }}\n{%- endif %}\n    {{- '<|start_header_id|>user<|end_header_id|>\\n\\n' -}}\n    {{- \"Given the following functions, please respond with a JSON for a function call \" }}\n    {{- \"with its proper arguments that best answers the given prompt.\\n\\n\" }}\n    {{- 'Respond in the format {\"name\": function name, \"parameters\": dictionary of argument name and its value}.' }}\n    {{- \"Do not use variables.\\n\\n\" }}\n    {%- for t in tools %}\n        {{- t | tojson(indent=4) }}\n        {{- \"\\n\\n\" }}\n    {%- endfor %}\n    {{- first_user_message + \"<|eot_id|>\"}}\n{%- endif %}\n\n{%- for message in messages %}\n    {%- if not (message.role == 'ipython' or message.role == 'tool' or 'tool_calls' in message) %}\n        {{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\\n\\n'+ message['content'] | trim + '<|eot_id|>' }}\n    {%- elif 'tool_calls' in message %}\n        {%- if not message.tool_calls|length == 1 %}\n            {{- raise_exception(\"This model only supports single tool-calls at once!\") }}\n        {%- endif %}\n        {%- set tool_call = message.tool_calls[0].function %}\n        {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' -}}\n        {{- '{\"name\": \"' + tool_call.name + '\", ' }}\n        {{- '\"parameters\": ' }}\n        {{- tool_call.arguments | tojson }}\n        {{- \"}\" }}\n        {{- \"<|eot_id|>\" }}\n    {%- elif message.role == \"tool\" or message.role == \"ipython\" %}\n        {{- \"<|start_header_id|>ipython<|end_header_id|>\\n\\n\" }}\n        {%- if message.content is mapping or message.content is iterable %}\n            {{- message.content | tojson }}\n        {%- else %}\n            {{- message.content }}\n        {%- endif %}\n        {{- \"<|eot_id|>\" }}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "extra_special_tokens": {},
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 16384,
+  "pad_token": "<|eot_id|>",
+  "tokenizer_class": "PreTrainedTokenizer"
+}
diff --git a/trainer_state.json b/trainer_state.json
new file mode 100644
index 0000000..ee40ba1
--- /dev/null
+++ b/trainer_state.json
@@ -0,0 +1,33566 @@
+{
+  "best_global_step": 12000,
+  "best_metric": 0.3009350597858429,
+  "best_model_checkpoint": null,
+  "epoch": 2.99971659791542,
+  "eval_steps": 1000,
+  "global_step": 23817,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00012595648203545676,
+      "grad_norm": 107.34553527832031,
+      "learning_rate": 0.0,
+      "loss": 4.9992,
+      "step": 1
+    },
+    {
+      "epoch": 0.0006297824101772837,
+      "grad_norm": 132.82723999023438,
+      "learning_rate": 5.037783375314861e-07,
+      "loss": 5.2462,
+      "step": 5
+    },
+    {
+      "epoch": 0.0012595648203545674,
+      "grad_norm": 61.73713684082031,
+      "learning_rate": 1.1335012594458437e-06,
+      "loss": 4.9216,
+      "step": 10
+    },
+    {
+      "epoch": 0.0018893472305318512,
+      "grad_norm": 29.1369571685791,
+      "learning_rate": 1.7632241813602012e-06,
+      "loss": 4.3103,
+      "step": 15
+    },
+    {
+      "epoch": 0.0025191296407091348,
+      "grad_norm": 13.688713073730469,
+      "learning_rate": 2.392947103274559e-06,
+      "loss": 3.7045,
+      "step": 20
+    },
+    {
+      "epoch": 0.0031489120508864186,
+      "grad_norm": 7.063476085662842,
+      "learning_rate": 3.0226700251889166e-06,
+      "loss": 3.255,
+      "step": 25
+    },
+    {
+      "epoch": 0.0037786944610637024,
+      "grad_norm": 4.568771839141846,
+      "learning_rate": 3.6523929471032744e-06,
+      "loss": 2.9188,
+      "step": 30
+    },
+    {
+      "epoch": 0.004408476871240986,
+      "grad_norm": 4.0639753341674805,
+      "learning_rate": 4.282115869017632e-06,
+      "loss": 2.6194,
+      "step": 35
+    },
+    {
+      "epoch": 0.0050382592814182696,
+      "grad_norm": 3.363410234451294,
+      "learning_rate": 4.911838790931989e-06,
+      "loss": 2.3438,
+      "step": 40
+    },
+    {
+      "epoch": 0.005668041691595554,
+      "grad_norm": 3.086527109146118,
+      "learning_rate": 5.541561712846347e-06,
+      "loss": 2.21,
+      "step": 45
+    },
+    {
+      "epoch": 0.006297824101772837,
+      "grad_norm": 2.987258195877075,
+      "learning_rate": 6.171284634760705e-06,
+      "loss": 2.063,
+      "step": 50
+    },
+    {
+      "epoch": 0.006927606511950121,
+      "grad_norm": 2.956822156906128,
+      "learning_rate": 6.801007556675062e-06,
+      "loss": 1.9592,
+      "step": 55
+    },
+    {
+      "epoch": 0.007557388922127405,
+      "grad_norm": 2.7860183715820312,
+      "learning_rate": 7.43073047858942e-06,
+      "loss": 1.8058,
+      "step": 60
+    },
+    {
+      "epoch": 0.008187171332304689,
+      "grad_norm": 2.930774211883545,
+      "learning_rate": 8.060453400503778e-06,
+      "loss": 1.7843,
+      "step": 65
+    },
+    {
+      "epoch": 0.008816953742481972,
+      "grad_norm": 2.797520637512207,
+      "learning_rate": 8.690176322418136e-06,
+      "loss": 1.6922,
+      "step": 70
+    },
+    {
+      "epoch": 0.009446736152659256,
+      "grad_norm": 2.945425271987915,
+      "learning_rate": 9.319899244332492e-06,
+      "loss": 1.6863,
+      "step": 75
+    },
+    {
+      "epoch": 0.010076518562836539,
+      "grad_norm": 3.775211811065674,
+      "learning_rate": 9.94962216624685e-06,
+      "loss": 1.6918,
+      "step": 80
+    },
+    {
+      "epoch": 0.010706300973013824,
+      "grad_norm": 4.795872688293457,
+      "learning_rate": 1.0579345088161209e-05,
+      "loss": 1.5709,
+      "step": 85
+    },
+    {
+      "epoch": 0.011336083383191108,
+      "grad_norm": 6.182225227355957,
+      "learning_rate": 1.1209068010075565e-05,
+      "loss": 1.582,
+      "step": 90
+    },
+    {
+      "epoch": 0.011965865793368391,
+      "grad_norm": 2.548555612564087,
+      "learning_rate": 1.1838790931989923e-05,
+      "loss": 1.4157,
+      "step": 95
+    },
+    {
+      "epoch": 0.012595648203545674,
+      "grad_norm": 2.5125293731689453,
+      "learning_rate": 1.246851385390428e-05,
+      "loss": 1.4848,
+      "step": 100
+    },
+    {
+      "epoch": 0.01322543061372296,
+      "grad_norm": 2.3581464290618896,
+      "learning_rate": 1.309823677581864e-05,
+      "loss": 1.4169,
+      "step": 105
+    },
+    {
+      "epoch": 0.013855213023900243,
+      "grad_norm": 2.4205543994903564,
+      "learning_rate": 1.3727959697732996e-05,
+      "loss": 1.3803,
+      "step": 110
+    },
+    {
+      "epoch": 0.014484995434077526,
+      "grad_norm": 2.251526355743408,
+      "learning_rate": 1.4357682619647355e-05,
+      "loss": 1.3928,
+      "step": 115
+    },
+    {
+      "epoch": 0.01511477784425481,
+      "grad_norm": 2.3966078758239746,
+      "learning_rate": 1.4987405541561711e-05,
+      "loss": 1.3994,
+      "step": 120
+    },
+    {
+      "epoch": 0.015744560254432095,
+      "grad_norm": 2.2904422283172607,
+      "learning_rate": 1.561712846347607e-05,
+      "loss": 1.3884,
+      "step": 125
+    },
+    {
+      "epoch": 0.016374342664609378,
+      "grad_norm": 2.238481283187866,
+      "learning_rate": 1.6246851385390428e-05,
+      "loss": 1.3419,
+      "step": 130
+    },
+    {
+      "epoch": 0.01700412507478666,
+      "grad_norm": 2.3208041191101074,
+      "learning_rate": 1.6876574307304786e-05,
+      "loss": 1.3282,
+      "step": 135
+    },
+    {
+      "epoch": 0.017633907484963945,
+      "grad_norm": 2.34198260307312,
+      "learning_rate": 1.750629722921914e-05,
+      "loss": 1.3398,
+      "step": 140
+    },
+    {
+      "epoch": 0.018263689895141228,
+      "grad_norm": 2.316502332687378,
+      "learning_rate": 1.81360201511335e-05,
+      "loss": 1.3171,
+      "step": 145
+    },
+    {
+      "epoch": 0.01889347230531851,
+      "grad_norm": 2.1474759578704834,
+      "learning_rate": 1.8765743073047857e-05,
+      "loss": 1.2896,
+      "step": 150
+    },
+    {
+      "epoch": 0.019523254715495795,
+      "grad_norm": 2.2437920570373535,
+      "learning_rate": 1.9395465994962215e-05,
+      "loss": 1.3504,
+      "step": 155
+    },
+    {
+      "epoch": 0.020153037125673078,
+      "grad_norm": 2.3530609607696533,
+      "learning_rate": 2.002518891687657e-05,
+      "loss": 1.2937,
+      "step": 160
+    },
+    {
+      "epoch": 0.020782819535850365,
+      "grad_norm": 3.199941635131836,
+      "learning_rate": 2.065491183879093e-05,
+      "loss": 1.2841,
+      "step": 165
+    },
+    {
+      "epoch": 0.02141260194602765,
+      "grad_norm": 2.3354921340942383,
+      "learning_rate": 2.128463476070529e-05,
+      "loss": 1.2573,
+      "step": 170
+    },
+    {
+      "epoch": 0.022042384356204932,
+      "grad_norm": 1.9546146392822266,
+      "learning_rate": 2.1914357682619645e-05,
+      "loss": 1.2179,
+      "step": 175
+    },
+    {
+      "epoch": 0.022672166766382215,
+      "grad_norm": 2.329948902130127,
+      "learning_rate": 2.2544080604534003e-05,
+      "loss": 1.2417,
+      "step": 180
+    },
+    {
+      "epoch": 0.0233019491765595,
+      "grad_norm": 2.110067367553711,
+      "learning_rate": 2.3173803526448358e-05,
+      "loss": 1.289,
+      "step": 185
+    },
+    {
+      "epoch": 0.023931731586736782,
+      "grad_norm": 1.8875372409820557,
+      "learning_rate": 2.380352644836272e-05,
+      "loss": 1.2183,
+      "step": 190
+    },
+    {
+      "epoch": 0.024561513996914065,
+      "grad_norm": 2.3789267539978027,
+      "learning_rate": 2.4433249370277077e-05,
+      "loss": 1.3136,
+      "step": 195
+    },
+    {
+      "epoch": 0.02519129640709135,
+      "grad_norm": 2.4072821140289307,
+      "learning_rate": 2.5062972292191432e-05,
+      "loss": 1.2241,
+      "step": 200
+    },
+    {
+      "epoch": 0.025821078817268632,
+      "grad_norm": 1.936230182647705,
+      "learning_rate": 2.569269521410579e-05,
+      "loss": 1.2715,
+      "step": 205
+    },
+    {
+      "epoch": 0.02645086122744592,
+      "grad_norm": 1.8356448411941528,
+      "learning_rate": 2.632241813602015e-05,
+      "loss": 1.1728,
+      "step": 210
+    },
+    {
+      "epoch": 0.027080643637623202,
+      "grad_norm": 1.7642450332641602,
+      "learning_rate": 2.6952141057934507e-05,
+      "loss": 1.2022,
+      "step": 215
+    },
+    {
+      "epoch": 0.027710426047800486,
+      "grad_norm": 1.8820688724517822,
+      "learning_rate": 2.7581863979848865e-05,
+      "loss": 1.1563,
+      "step": 220
+    },
+    {
+      "epoch": 0.02834020845797777,
+      "grad_norm": 1.8875645399093628,
+      "learning_rate": 2.821158690176322e-05,
+      "loss": 1.1473,
+      "step": 225
+    },
+    {
+      "epoch": 0.028969990868155052,
+      "grad_norm": 1.950432538986206,
+      "learning_rate": 2.884130982367758e-05,
+      "loss": 1.1478,
+      "step": 230
+    },
+    {
+      "epoch": 0.029599773278332336,
+      "grad_norm": 1.8610332012176514,
+      "learning_rate": 2.9471032745591936e-05,
+      "loss": 1.2026,
+      "step": 235
+    },
+    {
+      "epoch": 0.03022955568850962,
+      "grad_norm": 1.865122675895691,
+      "learning_rate": 3.0100755667506295e-05,
+      "loss": 1.1748,
+      "step": 240
+    },
+    {
+      "epoch": 0.030859338098686902,
+      "grad_norm": 1.7126635313034058,
+      "learning_rate": 3.0730478589420656e-05,
+      "loss": 1.0982,
+      "step": 245
+    },
+    {
+      "epoch": 0.03148912050886419,
+      "grad_norm": 1.809869408607483,
+      "learning_rate": 3.136020151133501e-05,
+      "loss": 1.1032,
+      "step": 250
+    },
+    {
+      "epoch": 0.03211890291904147,
+      "grad_norm": 1.9480305910110474,
+      "learning_rate": 3.1989924433249366e-05,
+      "loss": 1.1195,
+      "step": 255
+    },
+    {
+      "epoch": 0.032748685329218756,
+      "grad_norm": 1.762191653251648,
+      "learning_rate": 3.2619647355163724e-05,
+      "loss": 1.1413,
+      "step": 260
+    },
+    {
+      "epoch": 0.03337846773939604,
+      "grad_norm": 1.5951004028320312,
+      "learning_rate": 3.324937027707808e-05,
+      "loss": 1.1514,
+      "step": 265
+    },
+    {
+      "epoch": 0.03400825014957332,
+      "grad_norm": 1.9166375398635864,
+      "learning_rate": 3.387909319899244e-05,
+      "loss": 1.1129,
+      "step": 270
+    },
+    {
+      "epoch": 0.034638032559750606,
+      "grad_norm": 1.8051133155822754,
+      "learning_rate": 3.45088161209068e-05,
+      "loss": 1.1827,
+      "step": 275
+    },
+    {
+      "epoch": 0.03526781496992789,
+      "grad_norm": 1.8178284168243408,
+      "learning_rate": 3.513853904282116e-05,
+      "loss": 1.1567,
+      "step": 280
+    },
+    {
+      "epoch": 0.03589759738010517,
+      "grad_norm": 1.5286771059036255,
+      "learning_rate": 3.5768261964735515e-05,
+      "loss": 1.011,
+      "step": 285
+    },
+    {
+      "epoch": 0.036527379790282456,
+      "grad_norm": 1.6900701522827148,
+      "learning_rate": 3.639798488664987e-05,
+      "loss": 1.059,
+      "step": 290
+    },
+    {
+      "epoch": 0.03715716220045974,
+      "grad_norm": 1.5855501890182495,
+      "learning_rate": 3.702770780856423e-05,
+      "loss": 1.0943,
+      "step": 295
+    },
+    {
+      "epoch": 0.03778694461063702,
+      "grad_norm": 1.650087833404541,
+      "learning_rate": 3.765743073047858e-05,
+      "loss": 1.0843,
+      "step": 300
+    },
+    {
+      "epoch": 0.038416727020814306,
+      "grad_norm": 1.772127389907837,
+      "learning_rate": 3.828715365239294e-05,
+      "loss": 1.1161,
+      "step": 305
+    },
+    {
+      "epoch": 0.03904650943099159,
+      "grad_norm": 1.8052891492843628,
+      "learning_rate": 3.8916876574307306e-05,
+      "loss": 1.1493,
+      "step": 310
+    },
+    {
+      "epoch": 0.03967629184116887,
+      "grad_norm": 1.816603183746338,
+      "learning_rate": 3.954659949622166e-05,
+      "loss": 1.1113,
+      "step": 315
+    },
+    {
+      "epoch": 0.040306074251346156,
+      "grad_norm": 1.7114448547363281,
+      "learning_rate": 4.0176322418136016e-05,
+      "loss": 1.1124,
+      "step": 320
+    },
+    {
+      "epoch": 0.04093585666152345,
+      "grad_norm": 1.693428874015808,
+      "learning_rate": 4.080604534005038e-05,
+      "loss": 1.0994,
+      "step": 325
+    },
+    {
+      "epoch": 0.04156563907170073,
+      "grad_norm": 1.6820402145385742,
+      "learning_rate": 4.143576826196473e-05,
+      "loss": 1.0989,
+      "step": 330
+    },
+    {
+      "epoch": 0.042195421481878014,
+      "grad_norm": 1.5841305255889893,
+      "learning_rate": 4.206549118387909e-05,
+      "loss": 1.123,
+      "step": 335
+    },
+    {
+      "epoch": 0.0428252038920553,
+      "grad_norm": 1.714936375617981,
+      "learning_rate": 4.269521410579344e-05,
+      "loss": 1.0054,
+      "step": 340
+    },
+    {
+      "epoch": 0.04345498630223258,
+      "grad_norm": 1.6568523645401,
+      "learning_rate": 4.332493702770781e-05,
+      "loss": 1.0397,
+      "step": 345
+    },
+    {
+      "epoch": 0.044084768712409864,
+      "grad_norm": 1.9612834453582764,
+      "learning_rate": 4.3954659949622165e-05,
+      "loss": 1.0481,
+      "step": 350
+    },
+    {
+      "epoch": 0.04471455112258715,
+      "grad_norm": 1.5859555006027222,
+      "learning_rate": 4.4584382871536516e-05,
+      "loss": 1.0754,
+      "step": 355
+    },
+    {
+      "epoch": 0.04534433353276443,
+      "grad_norm": 1.6280044317245483,
+      "learning_rate": 4.521410579345088e-05,
+      "loss": 1.0696,
+      "step": 360
+    },
+    {
+      "epoch": 0.045974115942941714,
+      "grad_norm": 1.5000571012496948,
+      "learning_rate": 4.584382871536523e-05,
+      "loss": 1.0508,
+      "step": 365
+    },
+    {
+      "epoch": 0.046603898353119,
+      "grad_norm": 1.4447442293167114,
+      "learning_rate": 4.647355163727959e-05,
+      "loss": 1.0924,
+      "step": 370
+    },
+    {
+      "epoch": 0.04723368076329628,
+      "grad_norm": 1.6207387447357178,
+      "learning_rate": 4.7103274559193956e-05,
+      "loss": 1.0477,
+      "step": 375
+    },
+    {
+      "epoch": 0.047863463173473564,
+      "grad_norm": 1.5458993911743164,
+      "learning_rate": 4.773299748110831e-05,
+      "loss": 1.0905,
+      "step": 380
+    },
+    {
+      "epoch": 0.04849324558365085,
+      "grad_norm": 1.5125916004180908,
+      "learning_rate": 4.8362720403022666e-05,
+      "loss": 1.0858,
+      "step": 385
+    },
+    {
+      "epoch": 0.04912302799382813,
+      "grad_norm": 1.5438019037246704,
+      "learning_rate": 4.899244332493702e-05,
+      "loss": 1.1133,
+      "step": 390
+    },
+    {
+      "epoch": 0.049752810404005414,
+      "grad_norm": 1.5938135385513306,
+      "learning_rate": 4.962216624685138e-05,
+      "loss": 1.0814,
+      "step": 395
+    },
+    {
+      "epoch": 0.0503825928141827,
+      "grad_norm": 1.4425631761550903,
+      "learning_rate": 5.025188916876574e-05,
+      "loss": 1.0668,
+      "step": 400
+    },
+    {
+      "epoch": 0.05101237522435998,
+      "grad_norm": 1.505650520324707,
+      "learning_rate": 5.088161209068009e-05,
+      "loss": 1.1201,
+      "step": 405
+    },
+    {
+      "epoch": 0.051642157634537264,
+      "grad_norm": 1.421877145767212,
+      "learning_rate": 5.151133501259446e-05,
+      "loss": 1.0199,
+      "step": 410
+    },
+    {
+      "epoch": 0.052271940044714554,
+      "grad_norm": 1.6243743896484375,
+      "learning_rate": 5.2141057934508815e-05,
+      "loss": 1.0791,
+      "step": 415
+    },
+    {
+      "epoch": 0.05290172245489184,
+      "grad_norm": 1.5055480003356934,
+      "learning_rate": 5.2770780856423166e-05,
+      "loss": 1.0553,
+      "step": 420
+    },
+    {
+      "epoch": 0.05353150486506912,
+      "grad_norm": 1.5624281167984009,
+      "learning_rate": 5.340050377833753e-05,
+      "loss": 1.0778,
+      "step": 425
+    },
+    {
+      "epoch": 0.054161287275246404,
+      "grad_norm": 1.4252930879592896,
+      "learning_rate": 5.403022670025188e-05,
+      "loss": 1.0354,
+      "step": 430
+    },
+    {
+      "epoch": 0.05479106968542369,
+      "grad_norm": 1.4582849740982056,
+      "learning_rate": 5.465994962216624e-05,
+      "loss": 0.9825,
+      "step": 435
+    },
+    {
+      "epoch": 0.05542085209560097,
+      "grad_norm": 1.4336227178573608,
+      "learning_rate": 5.52896725440806e-05,
+      "loss": 1.0753,
+      "step": 440
+    },
+    {
+      "epoch": 0.056050634505778255,
+      "grad_norm": 1.4182472229003906,
+      "learning_rate": 5.591939546599496e-05,
+      "loss": 1.0005,
+      "step": 445
+    },
+    {
+      "epoch": 0.05668041691595554,
+      "grad_norm": 1.4060423374176025,
+      "learning_rate": 5.6549118387909316e-05,
+      "loss": 1.0311,
+      "step": 450
+    },
+    {
+      "epoch": 0.05731019932613282,
+      "grad_norm": 1.5005807876586914,
+      "learning_rate": 5.717884130982367e-05,
+      "loss": 1.0815,
+      "step": 455
+    },
+    {
+      "epoch": 0.057939981736310105,
+      "grad_norm": 1.3563963174819946,
+      "learning_rate": 5.780856423173803e-05,
+      "loss": 1.0099,
+      "step": 460
+    },
+    {
+      "epoch": 0.05856976414648739,
+      "grad_norm": 1.4664981365203857,
+      "learning_rate": 5.843828715365239e-05,
+      "loss": 1.0221,
+      "step": 465
+    },
+    {
+      "epoch": 0.05919954655666467,
+      "grad_norm": 1.3635119199752808,
+      "learning_rate": 5.906801007556674e-05,
+      "loss": 1.0864,
+      "step": 470
+    },
+    {
+      "epoch": 0.059829328966841955,
+      "grad_norm": 1.3736735582351685,
+      "learning_rate": 5.9697732997481107e-05,
+      "loss": 1.0949,
+      "step": 475
+    },
+    {
+      "epoch": 0.06045911137701924,
+      "grad_norm": 1.3973110914230347,
+      "learning_rate": 6.0327455919395465e-05,
+      "loss": 1.0936,
+      "step": 480
+    },
+    {
+      "epoch": 0.06108889378719652,
+      "grad_norm": 1.4158260822296143,
+      "learning_rate": 6.0957178841309816e-05,
+      "loss": 1.0389,
+      "step": 485
+    },
+    {
+      "epoch": 0.061718676197373805,
+      "grad_norm": 1.3810491561889648,
+      "learning_rate": 6.158690176322417e-05,
+      "loss": 1.0027,
+      "step": 490
+    },
+    {
+      "epoch": 0.06234845860755109,
+      "grad_norm": 1.3191838264465332,
+      "learning_rate": 6.221662468513854e-05,
+      "loss": 1.0626,
+      "step": 495
+    },
+    {
+      "epoch": 0.06297824101772838,
+      "grad_norm": 1.2959917783737183,
+      "learning_rate": 6.28463476070529e-05,
+      "loss": 1.064,
+      "step": 500
+    },
+    {
+      "epoch": 0.06360802342790566,
+      "grad_norm": 1.321399450302124,
+      "learning_rate": 6.347607052896724e-05,
+      "loss": 1.0407,
+      "step": 505
+    },
+    {
+      "epoch": 0.06423780583808295,
+      "grad_norm": 1.464593768119812,
+      "learning_rate": 6.410579345088161e-05,
+      "loss": 1.0961,
+      "step": 510
+    },
+    {
+      "epoch": 0.06486758824826022,
+      "grad_norm": 1.1570991277694702,
+      "learning_rate": 6.473551637279596e-05,
+      "loss": 1.0373,
+      "step": 515
+    },
+    {
+      "epoch": 0.06549737065843751,
+      "grad_norm": 1.2346997261047363,
+      "learning_rate": 6.536523929471032e-05,
+      "loss": 1.0016,
+      "step": 520
+    },
+    {
+      "epoch": 0.06612715306861479,
+      "grad_norm": 1.2645131349563599,
+      "learning_rate": 6.599496221662469e-05,
+      "loss": 1.0175,
+      "step": 525
+    },
+    {
+      "epoch": 0.06675693547879208,
+      "grad_norm": 1.7016206979751587,
+      "learning_rate": 6.662468513853903e-05,
+      "loss": 1.0517,
+      "step": 530
+    },
+    {
+      "epoch": 0.06738671788896936,
+      "grad_norm": 1.2666237354278564,
+      "learning_rate": 6.725440806045339e-05,
+      "loss": 0.9876,
+      "step": 535
+    },
+    {
+      "epoch": 0.06801650029914665,
+      "grad_norm": 1.365417242050171,
+      "learning_rate": 6.788413098236775e-05,
+      "loss": 1.0262,
+      "step": 540
+    },
+    {
+      "epoch": 0.06864628270932392,
+      "grad_norm": 1.1083537340164185,
+      "learning_rate": 6.851385390428211e-05,
+      "loss": 0.979,
+      "step": 545
+    },
+    {
+      "epoch": 0.06927606511950121,
+      "grad_norm": 1.2279447317123413,
+      "learning_rate": 6.914357682619647e-05,
+      "loss": 1.0566,
+      "step": 550
+    },
+    {
+      "epoch": 0.0699058475296785,
+      "grad_norm": 1.2323870658874512,
+      "learning_rate": 6.977329974811082e-05,
+      "loss": 1.0047,
+      "step": 555
+    },
+    {
+      "epoch": 0.07053562993985578,
+      "grad_norm": 1.2166273593902588,
+      "learning_rate": 7.040302267002518e-05,
+      "loss": 0.989,
+      "step": 560
+    },
+    {
+      "epoch": 0.07116541235003307,
+      "grad_norm": 1.3858731985092163,
+      "learning_rate": 7.103274559193954e-05,
+      "loss": 1.0389,
+      "step": 565
+    },
+    {
+      "epoch": 0.07179519476021035,
+      "grad_norm": 1.1922657489776611,
+      "learning_rate": 7.16624685138539e-05,
+      "loss": 1.0214,
+      "step": 570
+    },
+    {
+      "epoch": 0.07242497717038764,
+      "grad_norm": 1.2922885417938232,
+      "learning_rate": 7.229219143576826e-05,
+      "loss": 1.0647,
+      "step": 575
+    },
+    {
+      "epoch": 0.07305475958056491,
+      "grad_norm": 1.174737811088562,
+      "learning_rate": 7.292191435768262e-05,
+      "loss": 1.039,
+      "step": 580
+    },
+    {
+      "epoch": 0.0736845419907422,
+      "grad_norm": 1.4469586610794067,
+      "learning_rate": 7.355163727959697e-05,
+      "loss": 0.9553,
+      "step": 585
+    },
+    {
+      "epoch": 0.07431432440091948,
+      "grad_norm": 1.2497671842575073,
+      "learning_rate": 7.418136020151133e-05,
+      "loss": 1.0676,
+      "step": 590
+    },
+    {
+      "epoch": 0.07494410681109677,
+      "grad_norm": 1.0431822538375854,
+      "learning_rate": 7.481108312342569e-05,
+      "loss": 0.9664,
+      "step": 595
+    },
+    {
+      "epoch": 0.07557388922127405,
+      "grad_norm": 1.1643774509429932,
+      "learning_rate": 7.544080604534005e-05,
+      "loss": 1.0354,
+      "step": 600
+    },
+    {
+      "epoch": 0.07620367163145134,
+      "grad_norm": 1.277813196182251,
+      "learning_rate": 7.60705289672544e-05,
+      "loss": 1.0078,
+      "step": 605
+    },
+    {
+      "epoch": 0.07683345404162861,
+      "grad_norm": 1.17416250705719,
+      "learning_rate": 7.670025188916876e-05,
+      "loss": 0.9802,
+      "step": 610
+    },
+    {
+      "epoch": 0.0774632364518059,
+      "grad_norm": 1.3622347116470337,
+      "learning_rate": 7.732997481108312e-05,
+      "loss": 0.9323,
+      "step": 615
+    },
+    {
+      "epoch": 0.07809301886198318,
+      "grad_norm": 1.1814721822738647,
+      "learning_rate": 7.795969773299747e-05,
+      "loss": 0.989,
+      "step": 620
+    },
+    {
+      "epoch": 0.07872280127216047,
+      "grad_norm": 1.1661953926086426,
+      "learning_rate": 7.858942065491183e-05,
+      "loss": 0.9873,
+      "step": 625
+    },
+    {
+      "epoch": 0.07935258368233775,
+      "grad_norm": 1.1950370073318481,
+      "learning_rate": 7.921914357682618e-05,
+      "loss": 0.9997,
+      "step": 630
+    },
+    {
+      "epoch": 0.07998236609251504,
+      "grad_norm": 1.2635341882705688,
+      "learning_rate": 7.984886649874056e-05,
+      "loss": 0.9939,
+      "step": 635
+    },
+    {
+      "epoch": 0.08061214850269231,
+      "grad_norm": 1.094709038734436,
+      "learning_rate": 8.047858942065491e-05,
+      "loss": 1.035,
+      "step": 640
+    },
+    {
+      "epoch": 0.0812419309128696,
+      "grad_norm": 1.0304499864578247,
+      "learning_rate": 8.110831234256926e-05,
+      "loss": 1.0385,
+      "step": 645
+    },
+    {
+      "epoch": 0.0818717133230469,
+      "grad_norm": 1.203204870223999,
+      "learning_rate": 8.173803526448362e-05,
+      "loss": 1.0159,
+      "step": 650
+    },
+    {
+      "epoch": 0.08250149573322417,
+      "grad_norm": 1.317717432975769,
+      "learning_rate": 8.236775818639797e-05,
+      "loss": 0.9972,
+      "step": 655
+    },
+    {
+      "epoch": 0.08313127814340146,
+      "grad_norm": 1.1352335214614868,
+      "learning_rate": 8.299748110831233e-05,
+      "loss": 1.0027,
+      "step": 660
+    },
+    {
+      "epoch": 0.08376106055357874,
+      "grad_norm": 1.1126285791397095,
+      "learning_rate": 8.362720403022669e-05,
+      "loss": 0.9824,
+      "step": 665
+    },
+    {
+      "epoch": 0.08439084296375603,
+      "grad_norm": 1.1858975887298584,
+      "learning_rate": 8.425692695214106e-05,
+      "loss": 1.0084,
+      "step": 670
+    },
+    {
+      "epoch": 0.0850206253739333,
+      "grad_norm": 1.1154075860977173,
+      "learning_rate": 8.488664987405541e-05,
+      "loss": 0.968,
+      "step": 675
+    },
+    {
+      "epoch": 0.0856504077841106,
+      "grad_norm": 1.1235395669937134,
+      "learning_rate": 8.551637279596977e-05,
+      "loss": 0.9828,
+      "step": 680
+    },
+    {
+      "epoch": 0.08628019019428787,
+      "grad_norm": 1.251595139503479,
+      "learning_rate": 8.614609571788412e-05,
+      "loss": 0.9945,
+      "step": 685
+    },
+    {
+      "epoch": 0.08690997260446516,
+      "grad_norm": 1.2071354389190674,
+      "learning_rate": 8.677581863979848e-05,
+      "loss": 1.0263,
+      "step": 690
+    },
+    {
+      "epoch": 0.08753975501464244,
+      "grad_norm": 1.1954138278961182,
+      "learning_rate": 8.740554156171283e-05,
+      "loss": 0.9987,
+      "step": 695
+    },
+    {
+      "epoch": 0.08816953742481973,
+      "grad_norm": 1.088759422302246,
+      "learning_rate": 8.803526448362719e-05,
+      "loss": 1.0323,
+      "step": 700
+    },
+    {
+      "epoch": 0.088799319834997,
+      "grad_norm": 1.0866519212722778,
+      "learning_rate": 8.866498740554156e-05,
+      "loss": 1.0009,
+      "step": 705
+    },
+    {
+      "epoch": 0.0894291022451743,
+      "grad_norm": 1.1450897455215454,
+      "learning_rate": 8.929471032745592e-05,
+      "loss": 1.0493,
+      "step": 710
+    },
+    {
+      "epoch": 0.09005888465535157,
+      "grad_norm": 1.0619937181472778,
+      "learning_rate": 8.992443324937027e-05,
+      "loss": 1.0007,
+      "step": 715
+    },
+    {
+      "epoch": 0.09068866706552886,
+      "grad_norm": 1.1548367738723755,
+      "learning_rate": 9.055415617128463e-05,
+      "loss": 1.006,
+      "step": 720
+    },
+    {
+      "epoch": 0.09131844947570614,
+      "grad_norm": 1.0847381353378296,
+      "learning_rate": 9.118387909319898e-05,
+      "loss": 0.9703,
+      "step": 725
+    },
+    {
+      "epoch": 0.09194823188588343,
+      "grad_norm": 1.1227048635482788,
+      "learning_rate": 9.181360201511333e-05,
+      "loss": 0.957,
+      "step": 730
+    },
+    {
+      "epoch": 0.09257801429606072,
+      "grad_norm": 1.0912779569625854,
+      "learning_rate": 9.24433249370277e-05,
+      "loss": 0.9866,
+      "step": 735
+    },
+    {
+      "epoch": 0.093207796706238,
+      "grad_norm": 1.0490583181381226,
+      "learning_rate": 9.307304785894206e-05,
+      "loss": 1.0605,
+      "step": 740
+    },
+    {
+      "epoch": 0.09383757911641528,
+      "grad_norm": 1.0930982828140259,
+      "learning_rate": 9.370277078085642e-05,
+      "loss": 1.0061,
+      "step": 745
+    },
+    {
+      "epoch": 0.09446736152659256,
+      "grad_norm": 1.1814302206039429,
+      "learning_rate": 9.433249370277077e-05,
+      "loss": 1.0187,
+      "step": 750
+    },
+    {
+      "epoch": 0.09509714393676985,
+      "grad_norm": 1.113162636756897,
+      "learning_rate": 9.496221662468513e-05,
+      "loss": 0.9929,
+      "step": 755
+    },
+    {
+      "epoch": 0.09572692634694713,
+      "grad_norm": 1.0789388418197632,
+      "learning_rate": 9.559193954659948e-05,
+      "loss": 0.9699,
+      "step": 760
+    },
+    {
+      "epoch": 0.09635670875712442,
+      "grad_norm": 1.0445059537887573,
+      "learning_rate": 9.622166246851384e-05,
+      "loss": 0.929,
+      "step": 765
+    },
+    {
+      "epoch": 0.0969864911673017,
+      "grad_norm": 1.0267964601516724,
+      "learning_rate": 9.685138539042821e-05,
+      "loss": 0.9581,
+      "step": 770
+    },
+    {
+      "epoch": 0.09761627357747898,
+      "grad_norm": 1.0283193588256836,
+      "learning_rate": 9.748110831234256e-05,
+      "loss": 1.0331,
+      "step": 775
+    },
+    {
+      "epoch": 0.09824605598765626,
+      "grad_norm": 1.0210477113723755,
+      "learning_rate": 9.811083123425692e-05,
+      "loss": 0.96,
+      "step": 780
+    },
+    {
+      "epoch": 0.09887583839783355,
+      "grad_norm": 1.0163402557373047,
+      "learning_rate": 9.874055415617127e-05,
+      "loss": 0.9923,
+      "step": 785
+    },
+    {
+      "epoch": 0.09950562080801083,
+      "grad_norm": 1.0012452602386475,
+      "learning_rate": 9.937027707808563e-05,
+      "loss": 0.9722,
+      "step": 790
+    },
+    {
+      "epoch": 0.10013540321818812,
+      "grad_norm": 0.9810453653335571,
+      "learning_rate": 9.999999999999999e-05,
+      "loss": 0.991,
+      "step": 795
+    },
+    {
+      "epoch": 0.1007651856283654,
+      "grad_norm": 1.1151692867279053,
+      "learning_rate": 0.00010062972292191434,
+      "loss": 0.9783,
+      "step": 800
+    },
+    {
+      "epoch": 0.10139496803854268,
+      "grad_norm": 1.1338117122650146,
+      "learning_rate": 0.00010125944584382871,
+      "loss": 0.9979,
+      "step": 805
+    },
+    {
+      "epoch": 0.10202475044871996,
+      "grad_norm": 0.9878106117248535,
+      "learning_rate": 0.00010188916876574307,
+      "loss": 0.9068,
+      "step": 810
+    },
+    {
+      "epoch": 0.10265453285889725,
+      "grad_norm": 1.0334627628326416,
+      "learning_rate": 0.00010251889168765742,
+      "loss": 0.9389,
+      "step": 815
+    },
+    {
+      "epoch": 0.10328431526907453,
+      "grad_norm": 0.9542704224586487,
+      "learning_rate": 0.00010314861460957178,
+      "loss": 0.9699,
+      "step": 820
+    },
+    {
+      "epoch": 0.10391409767925182,
+      "grad_norm": 1.003753423690796,
+      "learning_rate": 0.00010377833753148613,
+      "loss": 0.9309,
+      "step": 825
+    },
+    {
+      "epoch": 0.10454388008942911,
+      "grad_norm": 0.9803423285484314,
+      "learning_rate": 0.00010440806045340049,
+      "loss": 0.9711,
+      "step": 830
+    },
+    {
+      "epoch": 0.10517366249960639,
+      "grad_norm": 0.9765311479568481,
+      "learning_rate": 0.00010503778337531486,
+      "loss": 1.0237,
+      "step": 835
+    },
+    {
+      "epoch": 0.10580344490978368,
+      "grad_norm": 1.035510540008545,
+      "learning_rate": 0.00010566750629722922,
+      "loss": 0.9737,
+      "step": 840
+    },
+    {
+      "epoch": 0.10643322731996095,
+      "grad_norm": 3.4597954750061035,
+      "learning_rate": 0.00010629722921914357,
+      "loss": 1.038,
+      "step": 845
+    },
+    {
+      "epoch": 0.10706300973013824,
+      "grad_norm": 1.0254745483398438,
+      "learning_rate": 0.00010692695214105793,
+      "loss": 1.0044,
+      "step": 850
+    },
+    {
+      "epoch": 0.10769279214031552,
+      "grad_norm": 4.8941521644592285,
+      "learning_rate": 0.00010755667506297228,
+      "loss": 1.1038,
+      "step": 855
+    },
+    {
+      "epoch": 0.10832257455049281,
+      "grad_norm": 1.6676890850067139,
+      "learning_rate": 0.00010818639798488663,
+      "loss": 1.0043,
+      "step": 860
+    },
+    {
+      "epoch": 0.10895235696067009,
+      "grad_norm": 5.738070964813232,
+      "learning_rate": 0.00010881612090680099,
+      "loss": 1.0502,
+      "step": 865
+    },
+    {
+      "epoch": 0.10958213937084738,
+      "grad_norm": 1.1913108825683594,
+      "learning_rate": 0.00010944584382871536,
+      "loss": 1.0071,
+      "step": 870
+    },
+    {
+      "epoch": 0.11021192178102465,
+      "grad_norm": 1.0302019119262695,
+      "learning_rate": 0.00011007556675062972,
+      "loss": 0.9732,
+      "step": 875
+    },
+    {
+      "epoch": 0.11084170419120194,
+      "grad_norm": 0.92161625623703,
+      "learning_rate": 0.00011070528967254407,
+      "loss": 0.9414,
+      "step": 880
+    },
+    {
+      "epoch": 0.11147148660137922,
+      "grad_norm": 0.954598605632782,
+      "learning_rate": 0.00011133501259445843,
+      "loss": 0.9772,
+      "step": 885
+    },
+    {
+      "epoch": 0.11210126901155651,
+      "grad_norm": 0.9241647720336914,
+      "learning_rate": 0.00011196473551637278,
+      "loss": 0.9498,
+      "step": 890
+    },
+    {
+      "epoch": 0.11273105142173379,
+      "grad_norm": 0.9744060039520264,
+      "learning_rate": 0.00011259445843828714,
+      "loss": 0.9501,
+      "step": 895
+    },
+    {
+      "epoch": 0.11336083383191108,
+      "grad_norm": 1.0800458192825317,
+      "learning_rate": 0.0001132241813602015,
+      "loss": 0.9823,
+      "step": 900
+    },
+    {
+      "epoch": 0.11399061624208835,
+      "grad_norm": 1.0275344848632812,
+      "learning_rate": 0.00011385390428211587,
+      "loss": 1.0426,
+      "step": 905
+    },
+    {
+      "epoch": 0.11462039865226564,
+      "grad_norm": 1.0069867372512817,
+      "learning_rate": 0.00011448362720403022,
+      "loss": 0.9933,
+      "step": 910
+    },
+    {
+      "epoch": 0.11525018106244292,
+      "grad_norm": 1.0309741497039795,
+      "learning_rate": 0.00011511335012594457,
+      "loss": 0.9792,
+      "step": 915
+    },
+    {
+      "epoch": 0.11587996347262021,
+      "grad_norm": 0.9738866090774536,
+      "learning_rate": 0.00011574307304785893,
+      "loss": 1.0193,
+      "step": 920
+    },
+    {
+      "epoch": 0.1165097458827975,
+      "grad_norm": 0.9231003522872925,
+      "learning_rate": 0.00011637279596977329,
+      "loss": 0.9741,
+      "step": 925
+    },
+    {
+      "epoch": 0.11713952829297478,
+      "grad_norm": 1.1318124532699585,
+      "learning_rate": 0.00011700251889168764,
+      "loss": 0.9644,
+      "step": 930
+    },
+    {
+      "epoch": 0.11776931070315207,
+      "grad_norm": 1.033288598060608,
+      "learning_rate": 0.00011763224181360201,
+      "loss": 0.9216,
+      "step": 935
+    },
+    {
+      "epoch": 0.11839909311332934,
+      "grad_norm": 1.003190517425537,
+      "learning_rate": 0.00011826196473551637,
+      "loss": 0.9521,
+      "step": 940
+    },
+    {
+      "epoch": 0.11902887552350663,
+      "grad_norm": 1.0145738124847412,
+      "learning_rate": 0.00011889168765743072,
+      "loss": 0.9676,
+      "step": 945
+    },
+    {
+      "epoch": 0.11965865793368391,
+      "grad_norm": 1.1370879411697388,
+      "learning_rate": 0.00011952141057934508,
+      "loss": 0.9987,
+      "step": 950
+    },
+    {
+      "epoch": 0.1202884403438612,
+      "grad_norm": 0.9657129645347595,
+      "learning_rate": 0.00012015113350125943,
+      "loss": 0.9622,
+      "step": 955
+    },
+    {
+      "epoch": 0.12091822275403848,
+      "grad_norm": 0.9489335417747498,
+      "learning_rate": 0.00012078085642317378,
+      "loss": 0.9402,
+      "step": 960
+    },
+    {
+      "epoch": 0.12154800516421577,
+      "grad_norm": 1.0598636865615845,
+      "learning_rate": 0.00012141057934508814,
+      "loss": 1.0047,
+      "step": 965
+    },
+    {
+      "epoch": 0.12217778757439304,
+      "grad_norm": 0.9747732281684875,
+      "learning_rate": 0.00012204030226700251,
+      "loss": 1.009,
+      "step": 970
+    },
+    {
+      "epoch": 0.12280756998457033,
+      "grad_norm": 0.9424954056739807,
+      "learning_rate": 0.00012267002518891686,
+      "loss": 0.9603,
+      "step": 975
+    },
+    {
+      "epoch": 0.12343735239474761,
+      "grad_norm": 1.0061867237091064,
+      "learning_rate": 0.00012329974811083123,
+      "loss": 0.9494,
+      "step": 980
+    },
+    {
+      "epoch": 0.1240671348049249,
+      "grad_norm": 0.924182116985321,
+      "learning_rate": 0.00012392947103274558,
+      "loss": 0.9804,
+      "step": 985
+    },
+    {
+      "epoch": 0.12469691721510218,
+      "grad_norm": 0.983267605304718,
+      "learning_rate": 0.00012455919395465995,
+      "loss": 0.9814,
+      "step": 990
+    },
+    {
+      "epoch": 0.12532669962527945,
+      "grad_norm": 0.896524965763092,
+      "learning_rate": 0.0001251889168765743,
+      "loss": 0.9533,
+      "step": 995
+    },
+    {
+      "epoch": 0.12595648203545676,
+      "grad_norm": 0.8669747710227966,
+      "learning_rate": 0.00012581863979848864,
+      "loss": 0.9544,
+      "step": 1000
+    },
+    {
+      "epoch": 0.12595648203545676,
+      "eval_loss": 0.383962482213974,
+      "eval_runtime": 6.2938,
+      "eval_samples_per_second": 158.887,
+      "eval_steps_per_second": 10.01,
+      "step": 1000
+    },
+    {
+      "epoch": 0.12658626444563403,
+      "grad_norm": 0.9055171012878418,
+      "learning_rate": 0.000126448362720403,
+      "loss": 0.9353,
+      "step": 1005
+    },
+    {
+      "epoch": 0.1272160468558113,
+      "grad_norm": 0.9889428019523621,
+      "learning_rate": 0.00012707808564231738,
+      "loss": 0.9508,
+      "step": 1010
+    },
+    {
+      "epoch": 0.12784582926598861,
+      "grad_norm": 0.8966602683067322,
+      "learning_rate": 0.00012770780856423173,
+      "loss": 0.995,
+      "step": 1015
+    },
+    {
+      "epoch": 0.1284756116761659,
+      "grad_norm": 0.9995138645172119,
+      "learning_rate": 0.0001283375314861461,
+      "loss": 0.9624,
+      "step": 1020
+    },
+    {
+      "epoch": 0.12910539408634317,
+      "grad_norm": 0.8536145687103271,
+      "learning_rate": 0.00012896725440806044,
+      "loss": 0.9549,
+      "step": 1025
+    },
+    {
+      "epoch": 0.12973517649652044,
+      "grad_norm": 0.8860256671905518,
+      "learning_rate": 0.00012959697732997479,
+      "loss": 1.0021,
+      "step": 1030
+    },
+    {
+      "epoch": 0.13036495890669775,
+      "grad_norm": 0.8574298620223999,
+      "learning_rate": 0.00013022670025188916,
+      "loss": 0.9798,
+      "step": 1035
+    },
+    {
+      "epoch": 0.13099474131687502,
+      "grad_norm": 1.1180200576782227,
+      "learning_rate": 0.00013085642317380353,
+      "loss": 0.9225,
+      "step": 1040
+    },
+    {
+      "epoch": 0.1316245237270523,
+      "grad_norm": 0.9391751289367676,
+      "learning_rate": 0.00013148614609571787,
+      "loss": 0.9467,
+      "step": 1045
+    },
+    {
+      "epoch": 0.13225430613722958,
+      "grad_norm": 0.8861620426177979,
+      "learning_rate": 0.00013211586901763222,
+      "loss": 0.9413,
+      "step": 1050
+    },
+    {
+      "epoch": 0.13288408854740688,
+      "grad_norm": 0.8499036431312561,
+      "learning_rate": 0.0001327455919395466,
+      "loss": 0.9644,
+      "step": 1055
+    },
+    {
+      "epoch": 0.13351387095758416,
+      "grad_norm": 0.9816482067108154,
+      "learning_rate": 0.00013337531486146094,
+      "loss": 0.9552,
+      "step": 1060
+    },
+    {
+      "epoch": 0.13414365336776143,
+      "grad_norm": 0.9725036025047302,
+      "learning_rate": 0.0001340050377833753,
+      "loss": 0.9461,
+      "step": 1065
+    },
+    {
+      "epoch": 0.1347734357779387,
+      "grad_norm": 0.9366094470024109,
+      "learning_rate": 0.00013463476070528968,
+      "loss": 0.9305,
+      "step": 1070
+    },
+    {
+      "epoch": 0.13540321818811601,
+      "grad_norm": 0.9212390780448914,
+      "learning_rate": 0.00013526448362720402,
+      "loss": 0.9551,
+      "step": 1075
+    },
+    {
+      "epoch": 0.1360330005982933,
+      "grad_norm": 0.8980582356452942,
+      "learning_rate": 0.00013589420654911837,
+      "loss": 0.9491,
+      "step": 1080
+    },
+    {
+      "epoch": 0.13666278300847057,
+      "grad_norm": 0.9107893109321594,
+      "learning_rate": 0.00013652392947103274,
+      "loss": 0.9366,
+      "step": 1085
+    },
+    {
+      "epoch": 0.13729256541864784,
+      "grad_norm": 0.8583124876022339,
+      "learning_rate": 0.00013715365239294708,
+      "loss": 0.9628,
+      "step": 1090
+    },
+    {
+      "epoch": 0.13792234782882515,
+      "grad_norm": 0.877052903175354,
+      "learning_rate": 0.00013778337531486146,
+      "loss": 0.9675,
+      "step": 1095
+    },
+    {
+      "epoch": 0.13855213023900242,
+      "grad_norm": 0.8020456433296204,
+      "learning_rate": 0.0001384130982367758,
+      "loss": 0.9015,
+      "step": 1100
+    },
+    {
+      "epoch": 0.1391819126491797,
+      "grad_norm": 0.8703967928886414,
+      "learning_rate": 0.00013904282115869017,
+      "loss": 0.9658,
+      "step": 1105
+    },
+    {
+      "epoch": 0.139811695059357,
+      "grad_norm": 0.7955961227416992,
+      "learning_rate": 0.00013967254408060452,
+      "loss": 0.9084,
+      "step": 1110
+    },
+    {
+      "epoch": 0.14044147746953428,
+      "grad_norm": 0.893059492111206,
+      "learning_rate": 0.0001403022670025189,
+      "loss": 0.9591,
+      "step": 1115
+    },
+    {
+      "epoch": 0.14107125987971156,
+      "grad_norm": 0.8481057286262512,
+      "learning_rate": 0.00014093198992443323,
+      "loss": 0.9588,
+      "step": 1120
+    },
+    {
+      "epoch": 0.14170104228988883,
+      "grad_norm": 0.8342163562774658,
+      "learning_rate": 0.00014156171284634758,
+      "loss": 0.947,
+      "step": 1125
+    },
+    {
+      "epoch": 0.14233082470006614,
+      "grad_norm": 0.790868878364563,
+      "learning_rate": 0.00014219143576826195,
+      "loss": 0.9366,
+      "step": 1130
+    },
+    {
+      "epoch": 0.14296060711024342,
+      "grad_norm": 0.8430061340332031,
+      "learning_rate": 0.0001428211586901763,
+      "loss": 0.9014,
+      "step": 1135
+    },
+    {
+      "epoch": 0.1435903895204207,
+      "grad_norm": 0.9150258302688599,
+      "learning_rate": 0.00014345088161209067,
+      "loss": 0.9546,
+      "step": 1140
+    },
+    {
+      "epoch": 0.14422017193059797,
+      "grad_norm": 0.8204888105392456,
+      "learning_rate": 0.00014408060453400504,
+      "loss": 0.9159,
+      "step": 1145
+    },
+    {
+      "epoch": 0.14484995434077527,
+      "grad_norm": 0.7595349550247192,
+      "learning_rate": 0.00014471032745591938,
+      "loss": 0.9179,
+      "step": 1150
+    },
+    {
+      "epoch": 0.14547973675095255,
+      "grad_norm": 0.8642888069152832,
+      "learning_rate": 0.00014534005037783373,
+      "loss": 1.0338,
+      "step": 1155
+    },
+    {
+      "epoch": 0.14610951916112982,
+      "grad_norm": 0.9633650183677673,
+      "learning_rate": 0.0001459697732997481,
+      "loss": 0.9638,
+      "step": 1160
+    },
+    {
+      "epoch": 0.1467393015713071,
+      "grad_norm": 0.8363626599311829,
+      "learning_rate": 0.00014659949622166244,
+      "loss": 0.8828,
+      "step": 1165
+    },
+    {
+      "epoch": 0.1473690839814844,
+      "grad_norm": 0.8199290633201599,
+      "learning_rate": 0.00014722921914357682,
+      "loss": 0.9577,
+      "step": 1170
+    },
+    {
+      "epoch": 0.14799886639166168,
+      "grad_norm": 0.7671203017234802,
+      "learning_rate": 0.0001478589420654912,
+      "loss": 0.9381,
+      "step": 1175
+    },
+    {
+      "epoch": 0.14862864880183896,
+      "grad_norm": 0.8354636430740356,
+      "learning_rate": 0.00014848866498740553,
+      "loss": 1.0019,
+      "step": 1180
+    },
+    {
+      "epoch": 0.14925843121201623,
+      "grad_norm": 0.911165714263916,
+      "learning_rate": 0.00014911838790931988,
+      "loss": 0.8985,
+      "step": 1185
+    },
+    {
+      "epoch": 0.14988821362219354,
+      "grad_norm": 0.8125472664833069,
+      "learning_rate": 0.00014974811083123425,
+      "loss": 0.9628,
+      "step": 1190
+    },
+    {
+      "epoch": 0.15051799603237082,
+      "grad_norm": 0.8937430381774902,
+      "learning_rate": 0.00015037783375314862,
+      "loss": 0.9843,
+      "step": 1195
+    },
+    {
+      "epoch": 0.1511477784425481,
+      "grad_norm": 0.9609346985816956,
+      "learning_rate": 0.00015100755667506297,
+      "loss": 0.9552,
+      "step": 1200
+    },
+    {
+      "epoch": 0.1517775608527254,
+      "grad_norm": 0.7975132465362549,
+      "learning_rate": 0.0001516372795969773,
+      "loss": 0.9799,
+      "step": 1205
+    },
+    {
+      "epoch": 0.15240734326290267,
+      "grad_norm": 0.8690225481987,
+      "learning_rate": 0.00015226700251889168,
+      "loss": 0.9604,
+      "step": 1210
+    },
+    {
+      "epoch": 0.15303712567307995,
+      "grad_norm": 0.7486653923988342,
+      "learning_rate": 0.00015289672544080603,
+      "loss": 0.9022,
+      "step": 1215
+    },
+    {
+      "epoch": 0.15366690808325723,
+      "grad_norm": 0.8420302271842957,
+      "learning_rate": 0.0001535264483627204,
+      "loss": 0.8791,
+      "step": 1220
+    },
+    {
+      "epoch": 0.15429669049343453,
+      "grad_norm": 0.8187466263771057,
+      "learning_rate": 0.00015415617128463474,
+      "loss": 0.9332,
+      "step": 1225
+    },
+    {
+      "epoch": 0.1549264729036118,
+      "grad_norm": 0.8711130619049072,
+      "learning_rate": 0.0001547858942065491,
+      "loss": 0.8924,
+      "step": 1230
+    },
+    {
+      "epoch": 0.15555625531378908,
+      "grad_norm": 0.8086002469062805,
+      "learning_rate": 0.00015541561712846346,
+      "loss": 0.9491,
+      "step": 1235
+    },
+    {
+      "epoch": 0.15618603772396636,
+      "grad_norm": 0.8274957537651062,
+      "learning_rate": 0.0001560453400503778,
+      "loss": 0.9392,
+      "step": 1240
+    },
+    {
+      "epoch": 0.15681582013414366,
+      "grad_norm": 0.831676721572876,
+      "learning_rate": 0.00015667506297229218,
+      "loss": 1.0327,
+      "step": 1245
+    },
+    {
+      "epoch": 0.15744560254432094,
+      "grad_norm": 0.8806201219558716,
+      "learning_rate": 0.00015730478589420652,
+      "loss": 0.8607,
+      "step": 1250
+    },
+    {
+      "epoch": 0.15807538495449822,
+      "grad_norm": 0.905436635017395,
+      "learning_rate": 0.00015793450881612092,
+      "loss": 0.9301,
+      "step": 1255
+    },
+    {
+      "epoch": 0.1587051673646755,
+      "grad_norm": 0.8631262183189392,
+      "learning_rate": 0.00015856423173803526,
+      "loss": 0.9443,
+      "step": 1260
+    },
+    {
+      "epoch": 0.1593349497748528,
+      "grad_norm": 0.7483521699905396,
+      "learning_rate": 0.0001591939546599496,
+      "loss": 0.901,
+      "step": 1265
+    },
+    {
+      "epoch": 0.15996473218503007,
+      "grad_norm": 0.8273198008537292,
+      "learning_rate": 0.00015982367758186398,
+      "loss": 0.9608,
+      "step": 1270
+    },
+    {
+      "epoch": 0.16059451459520735,
+      "grad_norm": 0.7562909722328186,
+      "learning_rate": 0.00016045340050377832,
+      "loss": 0.9219,
+      "step": 1275
+    },
+    {
+      "epoch": 0.16122429700538463,
+      "grad_norm": 0.8585835099220276,
+      "learning_rate": 0.0001610831234256927,
+      "loss": 0.942,
+      "step": 1280
+    },
+    {
+      "epoch": 0.16185407941556193,
+      "grad_norm": 0.8192921876907349,
+      "learning_rate": 0.00016171284634760704,
+      "loss": 0.9531,
+      "step": 1285
+    },
+    {
+      "epoch": 0.1624838618257392,
+      "grad_norm": 0.8301946520805359,
+      "learning_rate": 0.00016234256926952139,
+      "loss": 0.8972,
+      "step": 1290
+    },
+    {
+      "epoch": 0.16311364423591648,
+      "grad_norm": 0.8291681408882141,
+      "learning_rate": 0.00016297229219143576,
+      "loss": 0.9653,
+      "step": 1295
+    },
+    {
+      "epoch": 0.1637434266460938,
+      "grad_norm": 0.8672564625740051,
+      "learning_rate": 0.0001636020151133501,
+      "loss": 0.9498,
+      "step": 1300
+    },
+    {
+      "epoch": 0.16437320905627106,
+      "grad_norm": 0.7432397603988647,
+      "learning_rate": 0.00016423173803526445,
+      "loss": 0.8782,
+      "step": 1305
+    },
+    {
+      "epoch": 0.16500299146644834,
+      "grad_norm": 0.7710584402084351,
+      "learning_rate": 0.00016486146095717882,
+      "loss": 0.8872,
+      "step": 1310
+    },
+    {
+      "epoch": 0.16563277387662562,
+      "grad_norm": 0.7810630798339844,
+      "learning_rate": 0.00016549118387909316,
+      "loss": 0.9357,
+      "step": 1315
+    },
+    {
+      "epoch": 0.16626255628680292,
+      "grad_norm": 0.7368482947349548,
+      "learning_rate": 0.00016612090680100756,
+      "loss": 0.8935,
+      "step": 1320
+    },
+    {
+      "epoch": 0.1668923386969802,
+      "grad_norm": 0.7725487947463989,
+      "learning_rate": 0.0001667506297229219,
+      "loss": 0.9241,
+      "step": 1325
+    },
+    {
+      "epoch": 0.16752212110715747,
+      "grad_norm": 0.7551338076591492,
+      "learning_rate": 0.00016738035264483628,
+      "loss": 0.8878,
+      "step": 1330
+    },
+    {
+      "epoch": 0.16815190351733475,
+      "grad_norm": 0.8027164340019226,
+      "learning_rate": 0.00016801007556675062,
+      "loss": 0.9149,
+      "step": 1335
+    },
+    {
+      "epoch": 0.16878168592751205,
+      "grad_norm": 0.7476945519447327,
+      "learning_rate": 0.00016863979848866497,
+      "loss": 0.9241,
+      "step": 1340
+    },
+    {
+      "epoch": 0.16941146833768933,
+      "grad_norm": 0.7967312335968018,
+      "learning_rate": 0.00016926952141057934,
+      "loss": 0.932,
+      "step": 1345
+    },
+    {
+      "epoch": 0.1700412507478666,
+      "grad_norm": 0.809727668762207,
+      "learning_rate": 0.00016989924433249368,
+      "loss": 0.922,
+      "step": 1350
+    },
+    {
+      "epoch": 0.17067103315804388,
+      "grad_norm": 0.7631811499595642,
+      "learning_rate": 0.00017052896725440806,
+      "loss": 0.94,
+      "step": 1355
+    },
+    {
+      "epoch": 0.1713008155682212,
+      "grad_norm": 0.6545524001121521,
+      "learning_rate": 0.0001711586901763224,
+      "loss": 0.8898,
+      "step": 1360
+    },
+    {
+      "epoch": 0.17193059797839846,
+      "grad_norm": 0.8232229351997375,
+      "learning_rate": 0.00017178841309823675,
+      "loss": 0.9235,
+      "step": 1365
+    },
+    {
+      "epoch": 0.17256038038857574,
+      "grad_norm": 0.8617391586303711,
+      "learning_rate": 0.00017241813602015112,
+      "loss": 0.9491,
+      "step": 1370
+    },
+    {
+      "epoch": 0.17319016279875304,
+      "grad_norm": 0.7971004247665405,
+      "learning_rate": 0.00017304785894206546,
+      "loss": 0.8749,
+      "step": 1375
+    },
+    {
+      "epoch": 0.17381994520893032,
+      "grad_norm": 0.7876558899879456,
+      "learning_rate": 0.0001736775818639798,
+      "loss": 0.9954,
+      "step": 1380
+    },
+    {
+      "epoch": 0.1744497276191076,
+      "grad_norm": 0.8051108121871948,
+      "learning_rate": 0.00017430730478589418,
+      "loss": 0.897,
+      "step": 1385
+    },
+    {
+      "epoch": 0.17507951002928487,
+      "grad_norm": 0.8449770212173462,
+      "learning_rate": 0.00017493702770780855,
+      "loss": 0.8881,
+      "step": 1390
+    },
+    {
+      "epoch": 0.17570929243946218,
+      "grad_norm": 0.8217072486877441,
+      "learning_rate": 0.00017556675062972292,
+      "loss": 0.9027,
+      "step": 1395
+    },
+    {
+      "epoch": 0.17633907484963945,
+      "grad_norm": 0.806914210319519,
+      "learning_rate": 0.00017619647355163727,
+      "loss": 0.9222,
+      "step": 1400
+    },
+    {
+      "epoch": 0.17696885725981673,
+      "grad_norm": 0.8344951868057251,
+      "learning_rate": 0.00017682619647355164,
+      "loss": 0.9462,
+      "step": 1405
+    },
+    {
+      "epoch": 0.177598639669994,
+      "grad_norm": 0.7249205112457275,
+      "learning_rate": 0.00017745591939546598,
+      "loss": 0.917,
+      "step": 1410
+    },
+    {
+      "epoch": 0.1782284220801713,
+      "grad_norm": 0.8052341341972351,
+      "learning_rate": 0.00017808564231738033,
+      "loss": 0.9168,
+      "step": 1415
+    },
+    {
+      "epoch": 0.1788582044903486,
+      "grad_norm": 0.7675748467445374,
+      "learning_rate": 0.0001787153652392947,
+      "loss": 0.9186,
+      "step": 1420
+    },
+    {
+      "epoch": 0.17948798690052586,
+      "grad_norm": 0.7672801613807678,
+      "learning_rate": 0.00017934508816120904,
+      "loss": 0.8637,
+      "step": 1425
+    },
+    {
+      "epoch": 0.18011776931070314,
+      "grad_norm": 0.7517289519309998,
+      "learning_rate": 0.00017997481108312342,
+      "loss": 0.9053,
+      "step": 1430
+    },
+    {
+      "epoch": 0.18074755172088044,
+      "grad_norm": 0.7253280878067017,
+      "learning_rate": 0.00018060453400503776,
+      "loss": 0.9047,
+      "step": 1435
+    },
+    {
+      "epoch": 0.18137733413105772,
+      "grad_norm": 0.7113356590270996,
+      "learning_rate": 0.0001812342569269521,
+      "loss": 0.9288,
+      "step": 1440
+    },
+    {
+      "epoch": 0.182007116541235,
+      "grad_norm": 0.6626010537147522,
+      "learning_rate": 0.00018186397984886648,
+      "loss": 0.8942,
+      "step": 1445
+    },
+    {
+      "epoch": 0.18263689895141227,
+      "grad_norm": 0.7033849358558655,
+      "learning_rate": 0.00018249370277078082,
+      "loss": 0.9086,
+      "step": 1450
+    },
+    {
+      "epoch": 0.18326668136158958,
+      "grad_norm": 0.701263427734375,
+      "learning_rate": 0.00018312342569269522,
+      "loss": 0.9429,
+      "step": 1455
+    },
+    {
+      "epoch": 0.18389646377176685,
+      "grad_norm": 0.7362795472145081,
+      "learning_rate": 0.00018375314861460957,
+      "loss": 0.8955,
+      "step": 1460
+    },
+    {
+      "epoch": 0.18452624618194413,
+      "grad_norm": 0.7902641296386719,
+      "learning_rate": 0.00018438287153652394,
+      "loss": 0.8535,
+      "step": 1465
+    },
+    {
+      "epoch": 0.18515602859212144,
+      "grad_norm": 0.6855788230895996,
+      "learning_rate": 0.00018501259445843828,
+      "loss": 0.9166,
+      "step": 1470
+    },
+    {
+      "epoch": 0.1857858110022987,
+      "grad_norm": 0.6782147884368896,
+      "learning_rate": 0.00018564231738035263,
+      "loss": 0.8755,
+      "step": 1475
+    },
+    {
+      "epoch": 0.186415593412476,
+      "grad_norm": 0.6875694990158081,
+      "learning_rate": 0.000186272040302267,
+      "loss": 0.8712,
+      "step": 1480
+    },
+    {
+      "epoch": 0.18704537582265326,
+      "grad_norm": 0.7253673672676086,
+      "learning_rate": 0.00018690176322418134,
+      "loss": 0.8933,
+      "step": 1485
+    },
+    {
+      "epoch": 0.18767515823283057,
+      "grad_norm": 0.8096954822540283,
+      "learning_rate": 0.0001875314861460957,
+      "loss": 0.9359,
+      "step": 1490
+    },
+    {
+      "epoch": 0.18830494064300785,
+      "grad_norm": 0.7597787380218506,
+      "learning_rate": 0.00018816120906801006,
+      "loss": 0.9341,
+      "step": 1495
+    },
+    {
+      "epoch": 0.18893472305318512,
+      "grad_norm": 0.7736676931381226,
+      "learning_rate": 0.0001887909319899244,
+      "loss": 0.9281,
+      "step": 1500
+    },
+    {
+      "epoch": 0.1895645054633624,
+      "grad_norm": 0.6343753337860107,
+      "learning_rate": 0.00018942065491183878,
+      "loss": 0.9177,
+      "step": 1505
+    },
+    {
+      "epoch": 0.1901942878735397,
+      "grad_norm": 0.728712260723114,
+      "learning_rate": 0.00019005037783375312,
+      "loss": 0.9371,
+      "step": 1510
+    },
+    {
+      "epoch": 0.19082407028371698,
+      "grad_norm": 0.7092194557189941,
+      "learning_rate": 0.00019068010075566746,
+      "loss": 0.8902,
+      "step": 1515
+    },
+    {
+      "epoch": 0.19145385269389426,
+      "grad_norm": 0.7485836744308472,
+      "learning_rate": 0.00019130982367758186,
+      "loss": 0.8931,
+      "step": 1520
+    },
+    {
+      "epoch": 0.19208363510407153,
+      "grad_norm": 0.7485086917877197,
+      "learning_rate": 0.0001919395465994962,
+      "loss": 0.9368,
+      "step": 1525
+    },
+    {
+      "epoch": 0.19271341751424884,
+      "grad_norm": 0.7100546360015869,
+      "learning_rate": 0.00019256926952141058,
+      "loss": 0.8803,
+      "step": 1530
+    },
+    {
+      "epoch": 0.1933431999244261,
+      "grad_norm": 0.7371817827224731,
+      "learning_rate": 0.00019319899244332492,
+      "loss": 0.8963,
+      "step": 1535
+    },
+    {
+      "epoch": 0.1939729823346034,
+      "grad_norm": 0.6849647164344788,
+      "learning_rate": 0.0001938287153652393,
+      "loss": 0.9137,
+      "step": 1540
+    },
+    {
+      "epoch": 0.19460276474478067,
+      "grad_norm": 0.7108625173568726,
+      "learning_rate": 0.00019445843828715364,
+      "loss": 0.9078,
+      "step": 1545
+    },
+    {
+      "epoch": 0.19523254715495797,
+      "grad_norm": 0.7581806182861328,
+      "learning_rate": 0.00019508816120906799,
+      "loss": 0.9002,
+      "step": 1550
+    },
+    {
+      "epoch": 0.19586232956513525,
+      "grad_norm": 0.7299503087997437,
+      "learning_rate": 0.00019571788413098236,
+      "loss": 0.8897,
+      "step": 1555
+    },
+    {
+      "epoch": 0.19649211197531252,
+      "grad_norm": 0.7815247774124146,
+      "learning_rate": 0.0001963476070528967,
+      "loss": 0.8454,
+      "step": 1560
+    },
+    {
+      "epoch": 0.19712189438548983,
+      "grad_norm": 0.7475869655609131,
+      "learning_rate": 0.00019697732997481105,
+      "loss": 0.9482,
+      "step": 1565
+    },
+    {
+      "epoch": 0.1977516767956671,
+      "grad_norm": 0.7469599843025208,
+      "learning_rate": 0.00019760705289672542,
+      "loss": 0.9048,
+      "step": 1570
+    },
+    {
+      "epoch": 0.19838145920584438,
+      "grad_norm": 0.6186767220497131,
+      "learning_rate": 0.00019823677581863976,
+      "loss": 0.8698,
+      "step": 1575
+    },
+    {
+      "epoch": 0.19901124161602166,
+      "grad_norm": 0.843999445438385,
+      "learning_rate": 0.00019886649874055413,
+      "loss": 0.9567,
+      "step": 1580
+    },
+    {
+      "epoch": 0.19964102402619896,
+      "grad_norm": 0.749344527721405,
+      "learning_rate": 0.00019949622166246848,
+      "loss": 0.9234,
+      "step": 1585
+    },
+    {
+      "epoch": 0.20027080643637624,
+      "grad_norm": 0.6822441220283508,
+      "learning_rate": 0.00020012594458438288,
+      "loss": 0.8915,
+      "step": 1590
+    },
+    {
+      "epoch": 0.2009005888465535,
+      "grad_norm": 0.7193272113800049,
+      "learning_rate": 0.00020075566750629722,
+      "loss": 0.8922,
+      "step": 1595
+    },
+    {
+      "epoch": 0.2015303712567308,
+      "grad_norm": 0.7202250361442566,
+      "learning_rate": 0.00020138539042821157,
+      "loss": 0.9026,
+      "step": 1600
+    },
+    {
+      "epoch": 0.2021601536669081,
+      "grad_norm": 0.6946163773536682,
+      "learning_rate": 0.00020201511335012594,
+      "loss": 0.9181,
+      "step": 1605
+    },
+    {
+      "epoch": 0.20278993607708537,
+      "grad_norm": 0.7185525894165039,
+      "learning_rate": 0.00020264483627204028,
+      "loss": 0.8809,
+      "step": 1610
+    },
+    {
+      "epoch": 0.20341971848726265,
+      "grad_norm": 0.6290002465248108,
+      "learning_rate": 0.00020327455919395466,
+      "loss": 0.9033,
+      "step": 1615
+    },
+    {
+      "epoch": 0.20404950089743992,
+      "grad_norm": 0.6773431897163391,
+      "learning_rate": 0.000203904282115869,
+      "loss": 0.838,
+      "step": 1620
+    },
+    {
+      "epoch": 0.20467928330761723,
+      "grad_norm": 0.7076095342636108,
+      "learning_rate": 0.00020453400503778335,
+      "loss": 0.9158,
+      "step": 1625
+    },
+    {
+      "epoch": 0.2053090657177945,
+      "grad_norm": 0.7354462146759033,
+      "learning_rate": 0.00020516372795969772,
+      "loss": 0.8336,
+      "step": 1630
+    },
+    {
+      "epoch": 0.20593884812797178,
+      "grad_norm": 0.6885705590248108,
+      "learning_rate": 0.00020579345088161206,
+      "loss": 0.8971,
+      "step": 1635
+    },
+    {
+      "epoch": 0.20656863053814906,
+      "grad_norm": 0.697887659072876,
+      "learning_rate": 0.00020642317380352643,
+      "loss": 0.851,
+      "step": 1640
+    },
+    {
+      "epoch": 0.20719841294832636,
+      "grad_norm": 0.7369652986526489,
+      "learning_rate": 0.00020705289672544078,
+      "loss": 0.8567,
+      "step": 1645
+    },
+    {
+      "epoch": 0.20782819535850364,
+      "grad_norm": 0.7226613759994507,
+      "learning_rate": 0.00020768261964735512,
+      "loss": 0.9038,
+      "step": 1650
+    },
+    {
+      "epoch": 0.2084579777686809,
+      "grad_norm": 0.6973157525062561,
+      "learning_rate": 0.00020831234256926952,
+      "loss": 0.8443,
+      "step": 1655
+    },
+    {
+      "epoch": 0.20908776017885822,
+      "grad_norm": 0.7276191115379333,
+      "learning_rate": 0.00020894206549118387,
+      "loss": 0.8985,
+      "step": 1660
+    },
+    {
+      "epoch": 0.2097175425890355,
+      "grad_norm": 0.694542646408081,
+      "learning_rate": 0.00020957178841309824,
+      "loss": 0.8914,
+      "step": 1665
+    },
+    {
+      "epoch": 0.21034732499921277,
+      "grad_norm": 0.8255221843719482,
+      "learning_rate": 0.00021020151133501258,
+      "loss": 0.9072,
+      "step": 1670
+    },
+    {
+      "epoch": 0.21097710740939005,
+      "grad_norm": 0.637487530708313,
+      "learning_rate": 0.00021083123425692693,
+      "loss": 0.8637,
+      "step": 1675
+    },
+    {
+      "epoch": 0.21160688981956735,
+      "grad_norm": 0.6839597821235657,
+      "learning_rate": 0.0002114609571788413,
+      "loss": 0.8736,
+      "step": 1680
+    },
+    {
+      "epoch": 0.21223667222974463,
+      "grad_norm": 0.6435440182685852,
+      "learning_rate": 0.00021209068010075564,
+      "loss": 0.8725,
+      "step": 1685
+    },
+    {
+      "epoch": 0.2128664546399219,
+      "grad_norm": 0.7100492715835571,
+      "learning_rate": 0.00021272040302267002,
+      "loss": 0.9169,
+      "step": 1690
+    },
+    {
+      "epoch": 0.21349623705009918,
+      "grad_norm": 0.6926056742668152,
+      "learning_rate": 0.00021335012594458436,
+      "loss": 0.8549,
+      "step": 1695
+    },
+    {
+      "epoch": 0.21412601946027648,
+      "grad_norm": 0.8507684469223022,
+      "learning_rate": 0.0002139798488664987,
+      "loss": 0.9011,
+      "step": 1700
+    },
+    {
+      "epoch": 0.21475580187045376,
+      "grad_norm": 0.7276325821876526,
+      "learning_rate": 0.00021460957178841308,
+      "loss": 0.8607,
+      "step": 1705
+    },
+    {
+      "epoch": 0.21538558428063104,
+      "grad_norm": 0.6535823941230774,
+      "learning_rate": 0.00021523929471032742,
+      "loss": 0.8558,
+      "step": 1710
+    },
+    {
+      "epoch": 0.2160153666908083,
+      "grad_norm": 0.6517070531845093,
+      "learning_rate": 0.0002158690176322418,
+      "loss": 0.8703,
+      "step": 1715
+    },
+    {
+      "epoch": 0.21664514910098562,
+      "grad_norm": 0.7442309260368347,
+      "learning_rate": 0.00021649874055415614,
+      "loss": 0.8961,
+      "step": 1720
+    },
+    {
+      "epoch": 0.2172749315111629,
+      "grad_norm": 0.7261196374893188,
+      "learning_rate": 0.00021712846347607054,
+      "loss": 0.8902,
+      "step": 1725
+    },
+    {
+      "epoch": 0.21790471392134017,
+      "grad_norm": 0.7019686698913574,
+      "learning_rate": 0.00021775818639798488,
+      "loss": 0.8929,
+      "step": 1730
+    },
+    {
+      "epoch": 0.21853449633151745,
+      "grad_norm": 0.7852956056594849,
+      "learning_rate": 0.00021838790931989923,
+      "loss": 0.8766,
+      "step": 1735
+    },
+    {
+      "epoch": 0.21916427874169475,
+      "grad_norm": 0.7370544672012329,
+      "learning_rate": 0.0002190176322418136,
+      "loss": 0.855,
+      "step": 1740
+    },
+    {
+      "epoch": 0.21979406115187203,
+      "grad_norm": 0.6246267557144165,
+      "learning_rate": 0.00021964735516372794,
+      "loss": 0.9127,
+      "step": 1745
+    },
+    {
+      "epoch": 0.2204238435620493,
+      "grad_norm": 0.6939797401428223,
+      "learning_rate": 0.0002202770780856423,
+      "loss": 0.8878,
+      "step": 1750
+    },
+    {
+      "epoch": 0.2210536259722266,
+      "grad_norm": 0.6594600081443787,
+      "learning_rate": 0.00022090680100755666,
+      "loss": 0.9105,
+      "step": 1755
+    },
+    {
+      "epoch": 0.22168340838240388,
+      "grad_norm": 0.6578107476234436,
+      "learning_rate": 0.000221536523929471,
+      "loss": 0.9016,
+      "step": 1760
+    },
+    {
+      "epoch": 0.22231319079258116,
+      "grad_norm": 0.6889748573303223,
+      "learning_rate": 0.00022216624685138538,
+      "loss": 0.9091,
+      "step": 1765
+    },
+    {
+      "epoch": 0.22294297320275844,
+      "grad_norm": 0.6207224130630493,
+      "learning_rate": 0.00022279596977329972,
+      "loss": 0.9058,
+      "step": 1770
+    },
+    {
+      "epoch": 0.22357275561293574,
+      "grad_norm": 0.6724773645401001,
+      "learning_rate": 0.00022342569269521406,
+      "loss": 0.9144,
+      "step": 1775
+    },
+    {
+      "epoch": 0.22420253802311302,
+      "grad_norm": 0.702472448348999,
+      "learning_rate": 0.00022405541561712844,
+      "loss": 0.9,
+      "step": 1780
+    },
+    {
+      "epoch": 0.2248323204332903,
+      "grad_norm": 0.6482950448989868,
+      "learning_rate": 0.00022468513853904278,
+      "loss": 0.88,
+      "step": 1785
+    },
+    {
+      "epoch": 0.22546210284346757,
+      "grad_norm": 0.7253268957138062,
+      "learning_rate": 0.00022531486146095718,
+      "loss": 0.9147,
+      "step": 1790
+    },
+    {
+      "epoch": 0.22609188525364488,
+      "grad_norm": 0.7196680307388306,
+      "learning_rate": 0.00022594458438287152,
+      "loss": 0.8687,
+      "step": 1795
+    },
+    {
+      "epoch": 0.22672166766382215,
+      "grad_norm": 0.6720924973487854,
+      "learning_rate": 0.0002265743073047859,
+      "loss": 0.9173,
+      "step": 1800
+    },
+    {
+      "epoch": 0.22735145007399943,
+      "grad_norm": 0.6656882762908936,
+      "learning_rate": 0.00022720403022670024,
+      "loss": 0.8237,
+      "step": 1805
+    },
+    {
+      "epoch": 0.2279812324841767,
+      "grad_norm": 0.6303510665893555,
+      "learning_rate": 0.00022783375314861459,
+      "loss": 0.891,
+      "step": 1810
+    },
+    {
+      "epoch": 0.228611014894354,
+      "grad_norm": 0.6595205068588257,
+      "learning_rate": 0.00022846347607052896,
+      "loss": 0.8745,
+      "step": 1815
+    },
+    {
+      "epoch": 0.22924079730453129,
+      "grad_norm": 0.6373685002326965,
+      "learning_rate": 0.0002290931989924433,
+      "loss": 0.895,
+      "step": 1820
+    },
+    {
+      "epoch": 0.22987057971470856,
+      "grad_norm": 0.6187670230865479,
+      "learning_rate": 0.00022972292191435767,
+      "loss": 0.8954,
+      "step": 1825
+    },
+    {
+      "epoch": 0.23050036212488584,
+      "grad_norm": 0.6348496079444885,
+      "learning_rate": 0.00023035264483627202,
+      "loss": 0.8462,
+      "step": 1830
+    },
+    {
+      "epoch": 0.23113014453506314,
+      "grad_norm": 0.6880120038986206,
+      "learning_rate": 0.00023098236775818636,
+      "loss": 0.883,
+      "step": 1835
+    },
+    {
+      "epoch": 0.23175992694524042,
+      "grad_norm": 0.7668615579605103,
+      "learning_rate": 0.00023161209068010073,
+      "loss": 0.9134,
+      "step": 1840
+    },
+    {
+      "epoch": 0.2323897093554177,
+      "grad_norm": 0.6664952635765076,
+      "learning_rate": 0.00023224181360201508,
+      "loss": 0.9276,
+      "step": 1845
+    },
+    {
+      "epoch": 0.233019491765595,
+      "grad_norm": 0.754509449005127,
+      "learning_rate": 0.00023287153652392942,
+      "loss": 0.858,
+      "step": 1850
+    },
+    {
+      "epoch": 0.23364927417577228,
+      "grad_norm": 0.6345789432525635,
+      "learning_rate": 0.00023350125944584382,
+      "loss": 0.9048,
+      "step": 1855
+    },
+    {
+      "epoch": 0.23427905658594955,
+      "grad_norm": 0.6877152323722839,
+      "learning_rate": 0.00023413098236775817,
+      "loss": 0.9023,
+      "step": 1860
+    },
+    {
+      "epoch": 0.23490883899612683,
+      "grad_norm": 0.6173678636550903,
+      "learning_rate": 0.00023476070528967254,
+      "loss": 0.8951,
+      "step": 1865
+    },
+    {
+      "epoch": 0.23553862140630413,
+      "grad_norm": 0.6912857294082642,
+      "learning_rate": 0.00023539042821158688,
+      "loss": 0.8412,
+      "step": 1870
+    },
+    {
+      "epoch": 0.2361684038164814,
+      "grad_norm": 0.6385686993598938,
+      "learning_rate": 0.00023602015113350126,
+      "loss": 0.8954,
+      "step": 1875
+    },
+    {
+      "epoch": 0.23679818622665869,
+      "grad_norm": 0.6755088567733765,
+      "learning_rate": 0.0002366498740554156,
+      "loss": 0.8964,
+      "step": 1880
+    },
+    {
+      "epoch": 0.23742796863683596,
+      "grad_norm": 0.6391545534133911,
+      "learning_rate": 0.00023727959697732995,
+      "loss": 0.9294,
+      "step": 1885
+    },
+    {
+      "epoch": 0.23805775104701327,
+      "grad_norm": 0.7155817747116089,
+      "learning_rate": 0.00023790931989924432,
+      "loss": 0.8967,
+      "step": 1890
+    },
+    {
+      "epoch": 0.23868753345719054,
+      "grad_norm": 0.681224524974823,
+      "learning_rate": 0.00023853904282115866,
+      "loss": 0.8997,
+      "step": 1895
+    },
+    {
+      "epoch": 0.23931731586736782,
+      "grad_norm": 0.6473144888877869,
+      "learning_rate": 0.00023916876574307303,
+      "loss": 0.9172,
+      "step": 1900
+    },
+    {
+      "epoch": 0.2399470982775451,
+      "grad_norm": 0.6562004685401917,
+      "learning_rate": 0.00023979848866498738,
+      "loss": 0.8488,
+      "step": 1905
+    },
+    {
+      "epoch": 0.2405768806877224,
+      "grad_norm": 0.6842007637023926,
+      "learning_rate": 0.00024042821158690172,
+      "loss": 0.9183,
+      "step": 1910
+    },
+    {
+      "epoch": 0.24120666309789968,
+      "grad_norm": 0.5957079529762268,
+      "learning_rate": 0.0002410579345088161,
+      "loss": 0.8293,
+      "step": 1915
+    },
+    {
+      "epoch": 0.24183644550807695,
+      "grad_norm": 0.6745590567588806,
+      "learning_rate": 0.00024168765743073044,
+      "loss": 0.8219,
+      "step": 1920
+    },
+    {
+      "epoch": 0.24246622791825426,
+      "grad_norm": 0.6895525455474854,
+      "learning_rate": 0.00024231738035264484,
+      "loss": 0.9034,
+      "step": 1925
+    },
+    {
+      "epoch": 0.24309601032843153,
+      "grad_norm": 0.7394620776176453,
+      "learning_rate": 0.00024294710327455918,
+      "loss": 0.8702,
+      "step": 1930
+    },
+    {
+      "epoch": 0.2437257927386088,
+      "grad_norm": 0.7846884727478027,
+      "learning_rate": 0.00024357682619647353,
+      "loss": 0.9143,
+      "step": 1935
+    },
+    {
+      "epoch": 0.24435557514878609,
+      "grad_norm": 0.594127893447876,
+      "learning_rate": 0.00024420654911838787,
+      "loss": 0.8838,
+      "step": 1940
+    },
+    {
+      "epoch": 0.2449853575589634,
+      "grad_norm": 0.6737518906593323,
+      "learning_rate": 0.00024483627204030224,
+      "loss": 0.8657,
+      "step": 1945
+    },
+    {
+      "epoch": 0.24561513996914067,
+      "grad_norm": 0.6851866245269775,
+      "learning_rate": 0.0002454659949622166,
+      "loss": 0.9133,
+      "step": 1950
+    },
+    {
+      "epoch": 0.24624492237931794,
+      "grad_norm": 0.6238758563995361,
+      "learning_rate": 0.000246095717884131,
+      "loss": 0.8816,
+      "step": 1955
+    },
+    {
+      "epoch": 0.24687470478949522,
+      "grad_norm": 0.6002854704856873,
+      "learning_rate": 0.0002467254408060453,
+      "loss": 0.8888,
+      "step": 1960
+    },
+    {
+      "epoch": 0.24750448719967252,
+      "grad_norm": 0.6201847791671753,
+      "learning_rate": 0.0002473551637279597,
+      "loss": 0.8299,
+      "step": 1965
+    },
+    {
+      "epoch": 0.2481342696098498,
+      "grad_norm": 0.6619172692298889,
+      "learning_rate": 0.00024798488664987405,
+      "loss": 0.9297,
+      "step": 1970
+    },
+    {
+      "epoch": 0.24876405202002708,
+      "grad_norm": 0.6359203457832336,
+      "learning_rate": 0.00024861460957178837,
+      "loss": 0.8811,
+      "step": 1975
+    },
+    {
+      "epoch": 0.24939383443020435,
+      "grad_norm": 0.6441104412078857,
+      "learning_rate": 0.00024924433249370274,
+      "loss": 0.8704,
+      "step": 1980
+    },
+    {
+      "epoch": 0.25002361684038166,
+      "grad_norm": 0.7083386778831482,
+      "learning_rate": 0.0002498740554156171,
+      "loss": 0.8877,
+      "step": 1985
+    },
+    {
+      "epoch": 0.2506533992505589,
+      "grad_norm": 0.642206072807312,
+      "learning_rate": 0.0002505037783375315,
+      "loss": 0.8661,
+      "step": 1990
+    },
+    {
+      "epoch": 0.2512831816607362,
+      "grad_norm": 0.6782190203666687,
+      "learning_rate": 0.00025113350125944585,
+      "loss": 0.901,
+      "step": 1995
+    },
+    {
+      "epoch": 0.2519129640709135,
+      "grad_norm": 0.6277428269386292,
+      "learning_rate": 0.00025176322418136017,
+      "loss": 0.8212,
+      "step": 2000
+    },
+    {
+      "epoch": 0.2519129640709135,
+      "eval_loss": 0.35735848546028137,
+      "eval_runtime": 6.2326,
+      "eval_samples_per_second": 160.447,
+      "eval_steps_per_second": 10.108,
+      "step": 2000
+    },
+    {
+      "epoch": 0.25254274648109076,
+      "grad_norm": 0.5980456471443176,
+      "learning_rate": 0.00025239294710327454,
+      "loss": 0.8808,
+      "step": 2005
+    },
+    {
+      "epoch": 0.25317252889126807,
+      "grad_norm": 0.6398759484291077,
+      "learning_rate": 0.0002530226700251889,
+      "loss": 0.8817,
+      "step": 2010
+    },
+    {
+      "epoch": 0.25380231130144537,
+      "grad_norm": 0.5681187510490417,
+      "learning_rate": 0.00025365239294710323,
+      "loss": 0.8672,
+      "step": 2015
+    },
+    {
+      "epoch": 0.2544320937116226,
+      "grad_norm": 0.6202912926673889,
+      "learning_rate": 0.0002542821158690176,
+      "loss": 0.8627,
+      "step": 2020
+    },
+    {
+      "epoch": 0.2550618761217999,
+      "grad_norm": 0.5921783447265625,
+      "learning_rate": 0.000254911838790932,
+      "loss": 0.8214,
+      "step": 2025
+    },
+    {
+      "epoch": 0.25569165853197723,
+      "grad_norm": 0.629782497882843,
+      "learning_rate": 0.00025554156171284635,
+      "loss": 0.8995,
+      "step": 2030
+    },
+    {
+      "epoch": 0.2563214409421545,
+      "grad_norm": 0.6545585989952087,
+      "learning_rate": 0.00025617128463476066,
+      "loss": 0.8422,
+      "step": 2035
+    },
+    {
+      "epoch": 0.2569512233523318,
+      "grad_norm": 0.6024030447006226,
+      "learning_rate": 0.00025680100755667504,
+      "loss": 0.8341,
+      "step": 2040
+    },
+    {
+      "epoch": 0.25758100576250903,
+      "grad_norm": 0.6795976161956787,
+      "learning_rate": 0.0002574307304785894,
+      "loss": 0.852,
+      "step": 2045
+    },
+    {
+      "epoch": 0.25821078817268633,
+      "grad_norm": 0.6465495228767395,
+      "learning_rate": 0.0002580604534005037,
+      "loss": 0.8514,
+      "step": 2050
+    },
+    {
+      "epoch": 0.25884057058286364,
+      "grad_norm": 0.6498434543609619,
+      "learning_rate": 0.0002586901763224181,
+      "loss": 0.8906,
+      "step": 2055
+    },
+    {
+      "epoch": 0.2594703529930409,
+      "grad_norm": 0.7072421908378601,
+      "learning_rate": 0.00025931989924433247,
+      "loss": 0.9061,
+      "step": 2060
+    },
+    {
+      "epoch": 0.2601001354032182,
+      "grad_norm": 0.5902896523475647,
+      "learning_rate": 0.00025994962216624684,
+      "loss": 0.8327,
+      "step": 2065
+    },
+    {
+      "epoch": 0.2607299178133955,
+      "grad_norm": 0.6410335302352905,
+      "learning_rate": 0.0002605793450881612,
+      "loss": 0.9002,
+      "step": 2070
+    },
+    {
+      "epoch": 0.26135970022357274,
+      "grad_norm": 1.628951072692871,
+      "learning_rate": 0.00026120906801007553,
+      "loss": 0.8944,
+      "step": 2075
+    },
+    {
+      "epoch": 0.26198948263375005,
+      "grad_norm": 0.6544843316078186,
+      "learning_rate": 0.0002618387909319899,
+      "loss": 0.8656,
+      "step": 2080
+    },
+    {
+      "epoch": 0.2626192650439273,
+      "grad_norm": 15.444189071655273,
+      "learning_rate": 0.0002624685138539043,
+      "loss": 0.9639,
+      "step": 2085
+    },
+    {
+      "epoch": 0.2632490474541046,
+      "grad_norm": 8.399425506591797,
+      "learning_rate": 0.0002630982367758186,
+      "loss": 1.1367,
+      "step": 2090
+    },
+    {
+      "epoch": 0.2638788298642819,
+      "grad_norm": 24.009044647216797,
+      "learning_rate": 0.00026372795969773296,
+      "loss": 1.1429,
+      "step": 2095
+    },
+    {
+      "epoch": 0.26450861227445915,
+      "grad_norm": 4.226770877838135,
+      "learning_rate": 0.00026435768261964733,
+      "loss": 0.974,
+      "step": 2100
+    },
+    {
+      "epoch": 0.26513839468463646,
+      "grad_norm": 1.0910799503326416,
+      "learning_rate": 0.0002649874055415617,
+      "loss": 1.0182,
+      "step": 2105
+    },
+    {
+      "epoch": 0.26576817709481376,
+      "grad_norm": 2.8835411071777344,
+      "learning_rate": 0.000265617128463476,
+      "loss": 1.0283,
+      "step": 2110
+    },
+    {
+      "epoch": 0.266397959504991,
+      "grad_norm": 2.8626575469970703,
+      "learning_rate": 0.0002662468513853904,
+      "loss": 0.9273,
+      "step": 2115
+    },
+    {
+      "epoch": 0.2670277419151683,
+      "grad_norm": 1.4587650299072266,
+      "learning_rate": 0.00026687657430730477,
+      "loss": 0.9578,
+      "step": 2120
+    },
+    {
+      "epoch": 0.2676575243253456,
+      "grad_norm": 0.7692992091178894,
+      "learning_rate": 0.00026750629722921914,
+      "loss": 0.8701,
+      "step": 2125
+    },
+    {
+      "epoch": 0.26828730673552287,
+      "grad_norm": 0.8609071373939514,
+      "learning_rate": 0.0002681360201511335,
+      "loss": 0.8718,
+      "step": 2130
+    },
+    {
+      "epoch": 0.26891708914570017,
+      "grad_norm": 0.7419576048851013,
+      "learning_rate": 0.00026876574307304783,
+      "loss": 0.8732,
+      "step": 2135
+    },
+    {
+      "epoch": 0.2695468715558774,
+      "grad_norm": 0.8134281635284424,
+      "learning_rate": 0.0002693954659949622,
+      "loss": 0.9112,
+      "step": 2140
+    },
+    {
+      "epoch": 0.2701766539660547,
+      "grad_norm": 0.7559547424316406,
+      "learning_rate": 0.00027002518891687657,
+      "loss": 0.8804,
+      "step": 2145
+    },
+    {
+      "epoch": 0.27080643637623203,
+      "grad_norm": 0.7497460842132568,
+      "learning_rate": 0.0002706549118387909,
+      "loss": 0.8439,
+      "step": 2150
+    },
+    {
+      "epoch": 0.2714362187864093,
+      "grad_norm": 0.775444746017456,
+      "learning_rate": 0.00027128463476070526,
+      "loss": 0.888,
+      "step": 2155
+    },
+    {
+      "epoch": 0.2720660011965866,
+      "grad_norm": 0.7074035406112671,
+      "learning_rate": 0.00027191435768261963,
+      "loss": 0.8628,
+      "step": 2160
+    },
+    {
+      "epoch": 0.2726957836067639,
+      "grad_norm": 0.730311393737793,
+      "learning_rate": 0.00027254408060453395,
+      "loss": 0.8908,
+      "step": 2165
+    },
+    {
+      "epoch": 0.27332556601694113,
+      "grad_norm": 0.7610625624656677,
+      "learning_rate": 0.0002731738035264483,
+      "loss": 0.8954,
+      "step": 2170
+    },
+    {
+      "epoch": 0.27395534842711844,
+      "grad_norm": 0.6473423838615417,
+      "learning_rate": 0.0002738035264483627,
+      "loss": 0.8488,
+      "step": 2175
+    },
+    {
+      "epoch": 0.2745851308372957,
+      "grad_norm": 0.7084975838661194,
+      "learning_rate": 0.00027443324937027707,
+      "loss": 0.8631,
+      "step": 2180
+    },
+    {
+      "epoch": 0.275214913247473,
+      "grad_norm": 0.6844817996025085,
+      "learning_rate": 0.0002750629722921914,
+      "loss": 0.9021,
+      "step": 2185
+    },
+    {
+      "epoch": 0.2758446956576503,
+      "grad_norm": 0.641327440738678,
+      "learning_rate": 0.0002756926952141058,
+      "loss": 0.9002,
+      "step": 2190
+    },
+    {
+      "epoch": 0.27647447806782754,
+      "grad_norm": 0.7175489664077759,
+      "learning_rate": 0.0002763224181360201,
+      "loss": 0.8794,
+      "step": 2195
+    },
+    {
+      "epoch": 0.27710426047800485,
+      "grad_norm": 0.6306767463684082,
+      "learning_rate": 0.0002769521410579345,
+      "loss": 0.8732,
+      "step": 2200
+    },
+    {
+      "epoch": 0.27773404288818215,
+      "grad_norm": 0.6501113176345825,
+      "learning_rate": 0.00027758186397984887,
+      "loss": 0.8725,
+      "step": 2205
+    },
+    {
+      "epoch": 0.2783638252983594,
+      "grad_norm": 0.5996410250663757,
+      "learning_rate": 0.0002782115869017632,
+      "loss": 0.8828,
+      "step": 2210
+    },
+    {
+      "epoch": 0.2789936077085367,
+      "grad_norm": 0.6551349759101868,
+      "learning_rate": 0.00027884130982367756,
+      "loss": 0.8725,
+      "step": 2215
+    },
+    {
+      "epoch": 0.279623390118714,
+      "grad_norm": 0.6475560069084167,
+      "learning_rate": 0.00027947103274559193,
+      "loss": 0.9333,
+      "step": 2220
+    },
+    {
+      "epoch": 0.28025317252889126,
+      "grad_norm": 0.6957899928092957,
+      "learning_rate": 0.00028010075566750625,
+      "loss": 0.8933,
+      "step": 2225
+    },
+    {
+      "epoch": 0.28088295493906856,
+      "grad_norm": 0.6194736361503601,
+      "learning_rate": 0.0002807304785894206,
+      "loss": 0.9268,
+      "step": 2230
+    },
+    {
+      "epoch": 0.2815127373492458,
+      "grad_norm": 0.6293075084686279,
+      "learning_rate": 0.000281360201511335,
+      "loss": 0.8985,
+      "step": 2235
+    },
+    {
+      "epoch": 0.2821425197594231,
+      "grad_norm": 0.6805360317230225,
+      "learning_rate": 0.0002819899244332493,
+      "loss": 0.854,
+      "step": 2240
+    },
+    {
+      "epoch": 0.2827723021696004,
+      "grad_norm": 0.6671084761619568,
+      "learning_rate": 0.0002826196473551637,
+      "loss": 0.8774,
+      "step": 2245
+    },
+    {
+      "epoch": 0.28340208457977767,
+      "grad_norm": 0.5680047273635864,
+      "learning_rate": 0.00028324937027707805,
+      "loss": 0.8273,
+      "step": 2250
+    },
+    {
+      "epoch": 0.284031866989955,
+      "grad_norm": 0.5691477060317993,
+      "learning_rate": 0.0002838790931989924,
+      "loss": 0.8633,
+      "step": 2255
+    },
+    {
+      "epoch": 0.2846616494001323,
+      "grad_norm": 0.6509323120117188,
+      "learning_rate": 0.0002845088161209068,
+      "loss": 0.8991,
+      "step": 2260
+    },
+    {
+      "epoch": 0.2852914318103095,
+      "grad_norm": 0.714750349521637,
+      "learning_rate": 0.00028513853904282117,
+      "loss": 0.8863,
+      "step": 2265
+    },
+    {
+      "epoch": 0.28592121422048683,
+      "grad_norm": 0.6934742331504822,
+      "learning_rate": 0.0002857682619647355,
+      "loss": 0.8699,
+      "step": 2270
+    },
+    {
+      "epoch": 0.2865509966306641,
+      "grad_norm": 0.6048073172569275,
+      "learning_rate": 0.00028639798488664986,
+      "loss": 0.8983,
+      "step": 2275
+    },
+    {
+      "epoch": 0.2871807790408414,
+      "grad_norm": 0.6630669236183167,
+      "learning_rate": 0.00028702770780856423,
+      "loss": 0.9142,
+      "step": 2280
+    },
+    {
+      "epoch": 0.2878105614510187,
+      "grad_norm": 0.6518734693527222,
+      "learning_rate": 0.00028765743073047855,
+      "loss": 0.8734,
+      "step": 2285
+    },
+    {
+      "epoch": 0.28844034386119594,
+      "grad_norm": 0.5939868688583374,
+      "learning_rate": 0.0002882871536523929,
+      "loss": 0.8873,
+      "step": 2290
+    },
+    {
+      "epoch": 0.28907012627137324,
+      "grad_norm": 0.6081305742263794,
+      "learning_rate": 0.0002889168765743073,
+      "loss": 0.8735,
+      "step": 2295
+    },
+    {
+      "epoch": 0.28969990868155054,
+      "grad_norm": 0.5869495272636414,
+      "learning_rate": 0.0002895465994962216,
+      "loss": 0.8694,
+      "step": 2300
+    },
+    {
+      "epoch": 0.2903296910917278,
+      "grad_norm": 0.6381964683532715,
+      "learning_rate": 0.000290176322418136,
+      "loss": 0.8638,
+      "step": 2305
+    },
+    {
+      "epoch": 0.2909594735019051,
+      "grad_norm": 0.5546308755874634,
+      "learning_rate": 0.00029080604534005035,
+      "loss": 0.8748,
+      "step": 2310
+    },
+    {
+      "epoch": 0.2915892559120824,
+      "grad_norm": 0.7318828701972961,
+      "learning_rate": 0.0002914357682619647,
+      "loss": 0.8594,
+      "step": 2315
+    },
+    {
+      "epoch": 0.29221903832225965,
+      "grad_norm": 0.5685531497001648,
+      "learning_rate": 0.00029206549118387904,
+      "loss": 0.8815,
+      "step": 2320
+    },
+    {
+      "epoch": 0.29284882073243695,
+      "grad_norm": 0.6351069808006287,
+      "learning_rate": 0.00029269521410579347,
+      "loss": 0.8351,
+      "step": 2325
+    },
+    {
+      "epoch": 0.2934786031426142,
+      "grad_norm": 0.5828582048416138,
+      "learning_rate": 0.0002933249370277078,
+      "loss": 0.8678,
+      "step": 2330
+    },
+    {
+      "epoch": 0.2941083855527915,
+      "grad_norm": 0.5991604924201965,
+      "learning_rate": 0.00029395465994962216,
+      "loss": 0.8939,
+      "step": 2335
+    },
+    {
+      "epoch": 0.2947381679629688,
+      "grad_norm": 0.5732405781745911,
+      "learning_rate": 0.00029458438287153653,
+      "loss": 0.8594,
+      "step": 2340
+    },
+    {
+      "epoch": 0.29536795037314606,
+      "grad_norm": 0.5813714265823364,
+      "learning_rate": 0.00029521410579345085,
+      "loss": 0.8412,
+      "step": 2345
+    },
+    {
+      "epoch": 0.29599773278332336,
+      "grad_norm": 0.5281296968460083,
+      "learning_rate": 0.0002958438287153652,
+      "loss": 0.9049,
+      "step": 2350
+    },
+    {
+      "epoch": 0.29662751519350067,
+      "grad_norm": 0.6491068005561829,
+      "learning_rate": 0.0002964735516372796,
+      "loss": 0.8955,
+      "step": 2355
+    },
+    {
+      "epoch": 0.2972572976036779,
+      "grad_norm": 0.6236696839332581,
+      "learning_rate": 0.0002971032745591939,
+      "loss": 0.8792,
+      "step": 2360
+    },
+    {
+      "epoch": 0.2978870800138552,
+      "grad_norm": 0.605625331401825,
+      "learning_rate": 0.0002977329974811083,
+      "loss": 0.8448,
+      "step": 2365
+    },
+    {
+      "epoch": 0.29851686242403247,
+      "grad_norm": 0.6011054515838623,
+      "learning_rate": 0.00029836272040302265,
+      "loss": 0.911,
+      "step": 2370
+    },
+    {
+      "epoch": 0.2991466448342098,
+      "grad_norm": 0.5662422180175781,
+      "learning_rate": 0.00029899244332493697,
+      "loss": 0.872,
+      "step": 2375
+    },
+    {
+      "epoch": 0.2997764272443871,
+      "grad_norm": 0.8375005125999451,
+      "learning_rate": 0.00029962216624685134,
+      "loss": 0.7924,
+      "step": 2380
+    },
+    {
+      "epoch": 0.3004062096545643,
+      "grad_norm": 0.5916186571121216,
+      "learning_rate": 0.0002999999935557256,
+      "loss": 0.9007,
+      "step": 2385
+    },
+    {
+      "epoch": 0.30103599206474163,
+      "grad_norm": 0.6436251997947693,
+      "learning_rate": 0.00029999992105764553,
+      "loss": 0.8247,
+      "step": 2390
+    },
+    {
+      "epoch": 0.30166577447491894,
+      "grad_norm": 0.6368377208709717,
+      "learning_rate": 0.0002999997680061815,
+      "loss": 0.891,
+      "step": 2395
+    },
+    {
+      "epoch": 0.3022955568850962,
+      "grad_norm": 0.5848705172538757,
+      "learning_rate": 0.0002999995344014156,
+      "loss": 0.8335,
+      "step": 2400
+    },
+    {
+      "epoch": 0.3029253392952735,
+      "grad_norm": 0.5829634070396423,
+      "learning_rate": 0.0002999992202434735,
+      "loss": 0.8705,
+      "step": 2405
+    },
+    {
+      "epoch": 0.3035551217054508,
+      "grad_norm": 0.6242154240608215,
+      "learning_rate": 0.0002999988255325237,
+      "loss": 0.8819,
+      "step": 2410
+    },
+    {
+      "epoch": 0.30418490411562804,
+      "grad_norm": 0.5757481455802917,
+      "learning_rate": 0.0002999983502687783,
+      "loss": 0.8748,
+      "step": 2415
+    },
+    {
+      "epoch": 0.30481468652580535,
+      "grad_norm": 0.5024969577789307,
+      "learning_rate": 0.00029999779445249243,
+      "loss": 0.8534,
+      "step": 2420
+    },
+    {
+      "epoch": 0.3054444689359826,
+      "grad_norm": 0.5515364408493042,
+      "learning_rate": 0.00029999715808396463,
+      "loss": 0.8535,
+      "step": 2425
+    },
+    {
+      "epoch": 0.3060742513461599,
+      "grad_norm": 0.5151112079620361,
+      "learning_rate": 0.00029999644116353666,
+      "loss": 0.8686,
+      "step": 2430
+    },
+    {
+      "epoch": 0.3067040337563372,
+      "grad_norm": 0.5231375098228455,
+      "learning_rate": 0.0002999956436915935,
+      "loss": 0.8465,
+      "step": 2435
+    },
+    {
+      "epoch": 0.30733381616651445,
+      "grad_norm": 0.5415048003196716,
+      "learning_rate": 0.0002999947656685634,
+      "loss": 0.853,
+      "step": 2440
+    },
+    {
+      "epoch": 0.30796359857669175,
+      "grad_norm": 0.5642004609107971,
+      "learning_rate": 0.00029999380709491794,
+      "loss": 0.8827,
+      "step": 2445
+    },
+    {
+      "epoch": 0.30859338098686906,
+      "grad_norm": 0.6197057366371155,
+      "learning_rate": 0.0002999927679711718,
+      "loss": 0.9072,
+      "step": 2450
+    },
+    {
+      "epoch": 0.3092231633970463,
+      "grad_norm": 0.5865146517753601,
+      "learning_rate": 0.0002999916482978831,
+      "loss": 0.837,
+      "step": 2455
+    },
+    {
+      "epoch": 0.3098529458072236,
+      "grad_norm": 0.5961802005767822,
+      "learning_rate": 0.0002999904480756531,
+      "loss": 0.8657,
+      "step": 2460
+    },
+    {
+      "epoch": 0.31048272821740086,
+      "grad_norm": 0.5736685395240784,
+      "learning_rate": 0.0002999891673051263,
+      "loss": 0.872,
+      "step": 2465
+    },
+    {
+      "epoch": 0.31111251062757816,
+      "grad_norm": 0.5412915945053101,
+      "learning_rate": 0.0002999878059869905,
+      "loss": 0.8327,
+      "step": 2470
+    },
+    {
+      "epoch": 0.31174229303775547,
+      "grad_norm": 0.5011366605758667,
+      "learning_rate": 0.0002999863641219769,
+      "loss": 0.8418,
+      "step": 2475
+    },
+    {
+      "epoch": 0.3123720754479327,
+      "grad_norm": 0.5566514134407043,
+      "learning_rate": 0.0002999848417108597,
+      "loss": 0.8768,
+      "step": 2480
+    },
+    {
+      "epoch": 0.31300185785811,
+      "grad_norm": 0.5639830231666565,
+      "learning_rate": 0.0002999832387544564,
+      "loss": 0.8178,
+      "step": 2485
+    },
+    {
+      "epoch": 0.3136316402682873,
+      "grad_norm": 0.5784679055213928,
+      "learning_rate": 0.000299981555253628,
+      "loss": 0.8698,
+      "step": 2490
+    },
+    {
+      "epoch": 0.3142614226784646,
+      "grad_norm": 0.5428637266159058,
+      "learning_rate": 0.00029997979120927846,
+      "loss": 0.8671,
+      "step": 2495
+    },
+    {
+      "epoch": 0.3148912050886419,
+      "grad_norm": 0.5629287362098694,
+      "learning_rate": 0.00029997794662235515,
+      "loss": 0.873,
+      "step": 2500
+    },
+    {
+      "epoch": 0.3155209874988192,
+      "grad_norm": 0.5561172366142273,
+      "learning_rate": 0.00029997602149384856,
+      "loss": 0.8664,
+      "step": 2505
+    },
+    {
+      "epoch": 0.31615076990899643,
+      "grad_norm": 0.5451831221580505,
+      "learning_rate": 0.0002999740158247927,
+      "loss": 0.8349,
+      "step": 2510
+    },
+    {
+      "epoch": 0.31678055231917374,
+      "grad_norm": 0.5645403861999512,
+      "learning_rate": 0.00029997192961626456,
+      "loss": 0.8924,
+      "step": 2515
+    },
+    {
+      "epoch": 0.317410334729351,
+      "grad_norm": 0.5120379328727722,
+      "learning_rate": 0.00029996976286938444,
+      "loss": 0.8606,
+      "step": 2520
+    },
+    {
+      "epoch": 0.3180401171395283,
+      "grad_norm": 0.45988166332244873,
+      "learning_rate": 0.0002999675155853161,
+      "loss": 0.8285,
+      "step": 2525
+    },
+    {
+      "epoch": 0.3186698995497056,
+      "grad_norm": 0.5446504950523376,
+      "learning_rate": 0.00029996518776526614,
+      "loss": 0.8913,
+      "step": 2530
+    },
+    {
+      "epoch": 0.31929968195988284,
+      "grad_norm": 0.648369550704956,
+      "learning_rate": 0.00029996277941048485,
+      "loss": 0.8753,
+      "step": 2535
+    },
+    {
+      "epoch": 0.31992946437006015,
+      "grad_norm": 0.6404165029525757,
+      "learning_rate": 0.0002999602905222655,
+      "loss": 0.8747,
+      "step": 2540
+    },
+    {
+      "epoch": 0.32055924678023745,
+      "grad_norm": 0.46791502833366394,
+      "learning_rate": 0.0002999577211019447,
+      "loss": 0.8132,
+      "step": 2545
+    },
+    {
+      "epoch": 0.3211890291904147,
+      "grad_norm": 0.5365081429481506,
+      "learning_rate": 0.00029995507115090225,
+      "loss": 0.8363,
+      "step": 2550
+    },
+    {
+      "epoch": 0.321818811600592,
+      "grad_norm": 0.5029319524765015,
+      "learning_rate": 0.00029995234067056124,
+      "loss": 0.8297,
+      "step": 2555
+    },
+    {
+      "epoch": 0.32244859401076925,
+      "grad_norm": 0.509843647480011,
+      "learning_rate": 0.00029994952966238804,
+      "loss": 0.828,
+      "step": 2560
+    },
+    {
+      "epoch": 0.32307837642094656,
+      "grad_norm": 0.479045569896698,
+      "learning_rate": 0.0002999466381278922,
+      "loss": 0.8689,
+      "step": 2565
+    },
+    {
+      "epoch": 0.32370815883112386,
+      "grad_norm": 0.5639600157737732,
+      "learning_rate": 0.0002999436660686265,
+      "loss": 0.8521,
+      "step": 2570
+    },
+    {
+      "epoch": 0.3243379412413011,
+      "grad_norm": 0.5077898502349854,
+      "learning_rate": 0.00029994061348618715,
+      "loss": 0.835,
+      "step": 2575
+    },
+    {
+      "epoch": 0.3249677236514784,
+      "grad_norm": 0.45198580622673035,
+      "learning_rate": 0.00029993748038221324,
+      "loss": 0.8394,
+      "step": 2580
+    },
+    {
+      "epoch": 0.3255975060616557,
+      "grad_norm": 0.5617688894271851,
+      "learning_rate": 0.0002999342667583875,
+      "loss": 0.8285,
+      "step": 2585
+    },
+    {
+      "epoch": 0.32622728847183297,
+      "grad_norm": 0.5159285664558411,
+      "learning_rate": 0.0002999309726164356,
+      "loss": 0.7654,
+      "step": 2590
+    },
+    {
+      "epoch": 0.32685707088201027,
+      "grad_norm": 0.526965320110321,
+      "learning_rate": 0.00029992759795812666,
+      "loss": 0.8392,
+      "step": 2595
+    },
+    {
+      "epoch": 0.3274868532921876,
+      "grad_norm": 0.4861494302749634,
+      "learning_rate": 0.0002999241427852729,
+      "loss": 0.8177,
+      "step": 2600
+    },
+    {
+      "epoch": 0.3281166357023648,
+      "grad_norm": 0.5498744249343872,
+      "learning_rate": 0.0002999206070997298,
+      "loss": 0.8006,
+      "step": 2605
+    },
+    {
+      "epoch": 0.3287464181125421,
+      "grad_norm": 0.526978075504303,
+      "learning_rate": 0.0002999169909033962,
+      "loss": 0.8261,
+      "step": 2610
+    },
+    {
+      "epoch": 0.3293762005227194,
+      "grad_norm": 0.5078813433647156,
+      "learning_rate": 0.0002999132941982139,
+      "loss": 0.8396,
+      "step": 2615
+    },
+    {
+      "epoch": 0.3300059829328967,
+      "grad_norm": 0.5390729308128357,
+      "learning_rate": 0.00029990951698616834,
+      "loss": 0.8695,
+      "step": 2620
+    },
+    {
+      "epoch": 0.330635765343074,
+      "grad_norm": 0.520889401435852,
+      "learning_rate": 0.00029990565926928787,
+      "loss": 0.8489,
+      "step": 2625
+    },
+    {
+      "epoch": 0.33126554775325123,
+      "grad_norm": 0.6547030210494995,
+      "learning_rate": 0.00029990172104964413,
+      "loss": 0.8821,
+      "step": 2630
+    },
+    {
+      "epoch": 0.33189533016342854,
+      "grad_norm": 0.5034601092338562,
+      "learning_rate": 0.00029989770232935204,
+      "loss": 0.8202,
+      "step": 2635
+    },
+    {
+      "epoch": 0.33252511257360584,
+      "grad_norm": 0.5204071402549744,
+      "learning_rate": 0.0002998936031105698,
+      "loss": 0.852,
+      "step": 2640
+    },
+    {
+      "epoch": 0.3331548949837831,
+      "grad_norm": 0.499221533536911,
+      "learning_rate": 0.0002998894233954988,
+      "loss": 0.8338,
+      "step": 2645
+    },
+    {
+      "epoch": 0.3337846773939604,
+      "grad_norm": 0.5096358060836792,
+      "learning_rate": 0.0002998851631863835,
+      "loss": 0.8149,
+      "step": 2650
+    },
+    {
+      "epoch": 0.3344144598041377,
+      "grad_norm": 0.4654362201690674,
+      "learning_rate": 0.0002998808224855119,
+      "loss": 0.8461,
+      "step": 2655
+    },
+    {
+      "epoch": 0.33504424221431495,
+      "grad_norm": 0.7029035091400146,
+      "learning_rate": 0.00029987640129521497,
+      "loss": 0.8137,
+      "step": 2660
+    },
+    {
+      "epoch": 0.33567402462449225,
+      "grad_norm": 0.5634217262268066,
+      "learning_rate": 0.000299871899617867,
+      "loss": 0.8434,
+      "step": 2665
+    },
+    {
+      "epoch": 0.3363038070346695,
+      "grad_norm": 0.5168646574020386,
+      "learning_rate": 0.0002998673174558855,
+      "loss": 0.8554,
+      "step": 2670
+    },
+    {
+      "epoch": 0.3369335894448468,
+      "grad_norm": 0.4693644344806671,
+      "learning_rate": 0.00029986265481173123,
+      "loss": 0.8246,
+      "step": 2675
+    },
+    {
+      "epoch": 0.3375633718550241,
+      "grad_norm": 0.44928330183029175,
+      "learning_rate": 0.00029985791168790805,
+      "loss": 0.8554,
+      "step": 2680
+    },
+    {
+      "epoch": 0.33819315426520136,
+      "grad_norm": 0.5288846492767334,
+      "learning_rate": 0.0002998530880869632,
+      "loss": 0.8319,
+      "step": 2685
+    },
+    {
+      "epoch": 0.33882293667537866,
+      "grad_norm": 0.4755760431289673,
+      "learning_rate": 0.00029984818401148706,
+      "loss": 0.874,
+      "step": 2690
+    },
+    {
+      "epoch": 0.33945271908555597,
+      "grad_norm": 0.541684091091156,
+      "learning_rate": 0.0002998431994641132,
+      "loss": 0.8526,
+      "step": 2695
+    },
+    {
+      "epoch": 0.3400825014957332,
+      "grad_norm": 0.5160995125770569,
+      "learning_rate": 0.0002998381344475184,
+      "loss": 0.8749,
+      "step": 2700
+    },
+    {
+      "epoch": 0.3407122839059105,
+      "grad_norm": 0.5409444570541382,
+      "learning_rate": 0.00029983298896442276,
+      "loss": 0.8118,
+      "step": 2705
+    },
+    {
+      "epoch": 0.34134206631608777,
+      "grad_norm": 0.5148081183433533,
+      "learning_rate": 0.00029982776301758956,
+      "loss": 0.8685,
+      "step": 2710
+    },
+    {
+      "epoch": 0.34197184872626507,
+      "grad_norm": 0.5689860582351685,
+      "learning_rate": 0.0002998224566098251,
+      "loss": 0.8476,
+      "step": 2715
+    },
+    {
+      "epoch": 0.3426016311364424,
+      "grad_norm": 0.520268440246582,
+      "learning_rate": 0.00029981706974397917,
+      "loss": 0.8128,
+      "step": 2720
+    },
+    {
+      "epoch": 0.3432314135466196,
+      "grad_norm": 0.49906817078590393,
+      "learning_rate": 0.00029981160242294457,
+      "loss": 0.836,
+      "step": 2725
+    },
+    {
+      "epoch": 0.34386119595679693,
+      "grad_norm": 0.47317516803741455,
+      "learning_rate": 0.0002998060546496575,
+      "loss": 0.8251,
+      "step": 2730
+    },
+    {
+      "epoch": 0.34449097836697423,
+      "grad_norm": 0.49573519825935364,
+      "learning_rate": 0.0002998004264270971,
+      "loss": 0.832,
+      "step": 2735
+    },
+    {
+      "epoch": 0.3451207607771515,
+      "grad_norm": 0.43803608417510986,
+      "learning_rate": 0.0002997947177582859,
+      "loss": 0.7875,
+      "step": 2740
+    },
+    {
+      "epoch": 0.3457505431873288,
+      "grad_norm": 0.5324883460998535,
+      "learning_rate": 0.0002997889286462896,
+      "loss": 0.824,
+      "step": 2745
+    },
+    {
+      "epoch": 0.3463803255975061,
+      "grad_norm": 0.5902321934700012,
+      "learning_rate": 0.00029978305909421707,
+      "loss": 0.8265,
+      "step": 2750
+    },
+    {
+      "epoch": 0.34701010800768334,
+      "grad_norm": 0.5052042007446289,
+      "learning_rate": 0.0002997771091052204,
+      "loss": 0.7715,
+      "step": 2755
+    },
+    {
+      "epoch": 0.34763989041786064,
+      "grad_norm": 0.4439961314201355,
+      "learning_rate": 0.0002997710786824949,
+      "loss": 0.8387,
+      "step": 2760
+    },
+    {
+      "epoch": 0.3482696728280379,
+      "grad_norm": 0.5099385976791382,
+      "learning_rate": 0.0002997649678292789,
+      "loss": 0.8424,
+      "step": 2765
+    },
+    {
+      "epoch": 0.3488994552382152,
+      "grad_norm": 0.4415825605392456,
+      "learning_rate": 0.00029975877654885426,
+      "loss": 0.8066,
+      "step": 2770
+    },
+    {
+      "epoch": 0.3495292376483925,
+      "grad_norm": 0.5088052153587341,
+      "learning_rate": 0.0002997525048445458,
+      "loss": 0.8172,
+      "step": 2775
+    },
+    {
+      "epoch": 0.35015902005856975,
+      "grad_norm": 0.5503986477851868,
+      "learning_rate": 0.00029974615271972146,
+      "loss": 0.873,
+      "step": 2780
+    },
+    {
+      "epoch": 0.35078880246874705,
+      "grad_norm": 0.4609704613685608,
+      "learning_rate": 0.0002997397201777926,
+      "loss": 0.8041,
+      "step": 2785
+    },
+    {
+      "epoch": 0.35141858487892436,
+      "grad_norm": 0.49494004249572754,
+      "learning_rate": 0.00029973320722221356,
+      "loss": 0.916,
+      "step": 2790
+    },
+    {
+      "epoch": 0.3520483672891016,
+      "grad_norm": 0.4820273518562317,
+      "learning_rate": 0.00029972661385648197,
+      "loss": 0.8597,
+      "step": 2795
+    },
+    {
+      "epoch": 0.3526781496992789,
+      "grad_norm": 0.467641681432724,
+      "learning_rate": 0.0002997199400841386,
+      "loss": 0.7944,
+      "step": 2800
+    },
+    {
+      "epoch": 0.35330793210945616,
+      "grad_norm": 0.49666082859039307,
+      "learning_rate": 0.00029971318590876745,
+      "loss": 0.8204,
+      "step": 2805
+    },
+    {
+      "epoch": 0.35393771451963346,
+      "grad_norm": 0.4578961133956909,
+      "learning_rate": 0.00029970635133399565,
+      "loss": 0.8426,
+      "step": 2810
+    },
+    {
+      "epoch": 0.35456749692981077,
+      "grad_norm": 0.48815369606018066,
+      "learning_rate": 0.00029969943636349363,
+      "loss": 0.8277,
+      "step": 2815
+    },
+    {
+      "epoch": 0.355197279339988,
+      "grad_norm": 0.4686887562274933,
+      "learning_rate": 0.0002996924410009747,
+      "loss": 0.7936,
+      "step": 2820
+    },
+    {
+      "epoch": 0.3558270617501653,
+      "grad_norm": 0.4245474636554718,
+      "learning_rate": 0.0002996853652501956,
+      "loss": 0.8085,
+      "step": 2825
+    },
+    {
+      "epoch": 0.3564568441603426,
+      "grad_norm": 0.5085129737854004,
+      "learning_rate": 0.0002996782091149562,
+      "loss": 0.8584,
+      "step": 2830
+    },
+    {
+      "epoch": 0.35708662657051987,
+      "grad_norm": 0.4415908455848694,
+      "learning_rate": 0.0002996709725990995,
+      "loss": 0.8234,
+      "step": 2835
+    },
+    {
+      "epoch": 0.3577164089806972,
+      "grad_norm": 0.44018128514289856,
+      "learning_rate": 0.00029966365570651164,
+      "loss": 0.8566,
+      "step": 2840
+    },
+    {
+      "epoch": 0.3583461913908745,
+      "grad_norm": 0.4675704836845398,
+      "learning_rate": 0.000299656258441122,
+      "loss": 0.8164,
+      "step": 2845
+    },
+    {
+      "epoch": 0.35897597380105173,
+      "grad_norm": 0.47553756833076477,
+      "learning_rate": 0.0002996487808069031,
+      "loss": 0.8177,
+      "step": 2850
+    },
+    {
+      "epoch": 0.35960575621122903,
+      "grad_norm": 0.5298905372619629,
+      "learning_rate": 0.00029964122280787053,
+      "loss": 0.8537,
+      "step": 2855
+    },
+    {
+      "epoch": 0.3602355386214063,
+      "grad_norm": 0.484838604927063,
+      "learning_rate": 0.0002996335844480832,
+      "loss": 0.8495,
+      "step": 2860
+    },
+    {
+      "epoch": 0.3608653210315836,
+      "grad_norm": 0.4366026818752289,
+      "learning_rate": 0.000299625865731643,
+      "loss": 0.8073,
+      "step": 2865
+    },
+    {
+      "epoch": 0.3614951034417609,
+      "grad_norm": 0.4988342225551605,
+      "learning_rate": 0.00029961806666269503,
+      "loss": 0.8127,
+      "step": 2870
+    },
+    {
+      "epoch": 0.36212488585193814,
+      "grad_norm": 0.5028805732727051,
+      "learning_rate": 0.00029961018724542767,
+      "loss": 0.8711,
+      "step": 2875
+    },
+    {
+      "epoch": 0.36275466826211544,
+      "grad_norm": 0.5009424686431885,
+      "learning_rate": 0.00029960222748407226,
+      "loss": 0.8015,
+      "step": 2880
+    },
+    {
+      "epoch": 0.36338445067229275,
+      "grad_norm": 0.4522798955440521,
+      "learning_rate": 0.00029959418738290344,
+      "loss": 0.8261,
+      "step": 2885
+    },
+    {
+      "epoch": 0.36401423308247,
+      "grad_norm": 0.49349725246429443,
+      "learning_rate": 0.00029958606694623893,
+      "loss": 0.8006,
+      "step": 2890
+    },
+    {
+      "epoch": 0.3646440154926473,
+      "grad_norm": 0.46870625019073486,
+      "learning_rate": 0.00029957786617843956,
+      "loss": 0.8285,
+      "step": 2895
+    },
+    {
+      "epoch": 0.36527379790282455,
+      "grad_norm": 0.5234463810920715,
+      "learning_rate": 0.0002995695850839093,
+      "loss": 0.8497,
+      "step": 2900
+    },
+    {
+      "epoch": 0.36590358031300185,
+      "grad_norm": 0.487884521484375,
+      "learning_rate": 0.0002995612236670953,
+      "loss": 0.8033,
+      "step": 2905
+    },
+    {
+      "epoch": 0.36653336272317916,
+      "grad_norm": 0.4760074317455292,
+      "learning_rate": 0.0002995527819324879,
+      "loss": 0.8412,
+      "step": 2910
+    },
+    {
+      "epoch": 0.3671631451333564,
+      "grad_norm": 0.4630395472049713,
+      "learning_rate": 0.0002995442598846205,
+      "loss": 0.8244,
+      "step": 2915
+    },
+    {
+      "epoch": 0.3677929275435337,
+      "grad_norm": 0.4981043040752411,
+      "learning_rate": 0.0002995356575280695,
+      "loss": 0.798,
+      "step": 2920
+    },
+    {
+      "epoch": 0.368422709953711,
+      "grad_norm": 0.4630597233772278,
+      "learning_rate": 0.00029952697486745466,
+      "loss": 0.8032,
+      "step": 2925
+    },
+    {
+      "epoch": 0.36905249236388826,
+      "grad_norm": 0.4962010979652405,
+      "learning_rate": 0.00029951821190743884,
+      "loss": 0.8183,
+      "step": 2930
+    },
+    {
+      "epoch": 0.36968227477406557,
+      "grad_norm": 0.47193852066993713,
+      "learning_rate": 0.00029950936865272775,
+      "loss": 0.841,
+      "step": 2935
+    },
+    {
+      "epoch": 0.37031205718424287,
+      "grad_norm": 0.4802277982234955,
+      "learning_rate": 0.0002995004451080706,
+      "loss": 0.7433,
+      "step": 2940
+    },
+    {
+      "epoch": 0.3709418395944201,
+      "grad_norm": 0.43486830592155457,
+      "learning_rate": 0.00029949144127825947,
+      "loss": 0.8051,
+      "step": 2945
+    },
+    {
+      "epoch": 0.3715716220045974,
+      "grad_norm": 0.5078021287918091,
+      "learning_rate": 0.0002994823571681296,
+      "loss": 0.8662,
+      "step": 2950
+    },
+    {
+      "epoch": 0.3722014044147747,
+      "grad_norm": 0.44738146662712097,
+      "learning_rate": 0.0002994731927825594,
+      "loss": 0.7997,
+      "step": 2955
+    },
+    {
+      "epoch": 0.372831186824952,
+      "grad_norm": 0.44323739409446716,
+      "learning_rate": 0.0002994639481264704,
+      "loss": 0.8481,
+      "step": 2960
+    },
+    {
+      "epoch": 0.3734609692351293,
+      "grad_norm": 0.525050938129425,
+      "learning_rate": 0.0002994546232048271,
+      "loss": 0.8375,
+      "step": 2965
+    },
+    {
+      "epoch": 0.37409075164530653,
+      "grad_norm": 0.4817000925540924,
+      "learning_rate": 0.00029944521802263723,
+      "loss": 0.8,
+      "step": 2970
+    },
+    {
+      "epoch": 0.37472053405548383,
+      "grad_norm": 0.5709639191627502,
+      "learning_rate": 0.00029943573258495165,
+      "loss": 0.8104,
+      "step": 2975
+    },
+    {
+      "epoch": 0.37535031646566114,
+      "grad_norm": 0.518618643283844,
+      "learning_rate": 0.00029942616689686416,
+      "loss": 0.7948,
+      "step": 2980
+    },
+    {
+      "epoch": 0.3759800988758384,
+      "grad_norm": 0.4226647615432739,
+      "learning_rate": 0.00029941652096351174,
+      "loss": 0.7599,
+      "step": 2985
+    },
+    {
+      "epoch": 0.3766098812860157,
+      "grad_norm": 0.4751405119895935,
+      "learning_rate": 0.0002994067947900746,
+      "loss": 0.8015,
+      "step": 2990
+    },
+    {
+      "epoch": 0.37723966369619294,
+      "grad_norm": 0.4653600752353668,
+      "learning_rate": 0.0002993969883817758,
+      "loss": 0.7758,
+      "step": 2995
+    },
+    {
+      "epoch": 0.37786944610637024,
+      "grad_norm": 0.512941837310791,
+      "learning_rate": 0.00029938710174388163,
+      "loss": 0.8188,
+      "step": 3000
+    },
+    {
+      "epoch": 0.37786944610637024,
+      "eval_loss": 0.3438470661640167,
+      "eval_runtime": 6.225,
+      "eval_samples_per_second": 160.642,
+      "eval_steps_per_second": 10.12,
+      "step": 3000
+    },
+    {
+      "epoch": 0.37849922851654755,
+      "grad_norm": 0.43524855375289917,
+      "learning_rate": 0.0002993771348817015,
+      "loss": 0.803,
+      "step": 3005
+    },
+    {
+      "epoch": 0.3791290109267248,
+      "grad_norm": 0.4569668173789978,
+      "learning_rate": 0.0002993670878005878,
+      "loss": 0.8777,
+      "step": 3010
+    },
+    {
+      "epoch": 0.3797587933369021,
+      "grad_norm": 0.4643417000770569,
+      "learning_rate": 0.00029935696050593604,
+      "loss": 0.7621,
+      "step": 3015
+    },
+    {
+      "epoch": 0.3803885757470794,
+      "grad_norm": 0.4604712128639221,
+      "learning_rate": 0.00029934675300318485,
+      "loss": 0.8216,
+      "step": 3020
+    },
+    {
+      "epoch": 0.38101835815725665,
+      "grad_norm": 0.4307630956172943,
+      "learning_rate": 0.0002993364652978158,
+      "loss": 0.8163,
+      "step": 3025
+    },
+    {
+      "epoch": 0.38164814056743396,
+      "grad_norm": 0.44455698132514954,
+      "learning_rate": 0.00029932609739535365,
+      "loss": 0.818,
+      "step": 3030
+    },
+    {
+      "epoch": 0.38227792297761126,
+      "grad_norm": 0.43203669786453247,
+      "learning_rate": 0.0002993156493013663,
+      "loss": 0.8168,
+      "step": 3035
+    },
+    {
+      "epoch": 0.3829077053877885,
+      "grad_norm": 0.42328670620918274,
+      "learning_rate": 0.00029930512102146453,
+      "loss": 0.8025,
+      "step": 3040
+    },
+    {
+      "epoch": 0.3835374877979658,
+      "grad_norm": 0.43900108337402344,
+      "learning_rate": 0.0002992945125613023,
+      "loss": 0.7595,
+      "step": 3045
+    },
+    {
+      "epoch": 0.38416727020814306,
+      "grad_norm": 0.46638986468315125,
+      "learning_rate": 0.00029928382392657656,
+      "loss": 0.8208,
+      "step": 3050
+    },
+    {
+      "epoch": 0.38479705261832037,
+      "grad_norm": 0.4279174208641052,
+      "learning_rate": 0.00029927305512302736,
+      "loss": 0.8151,
+      "step": 3055
+    },
+    {
+      "epoch": 0.38542683502849767,
+      "grad_norm": 0.4648323357105255,
+      "learning_rate": 0.0002992622061564378,
+      "loss": 0.7666,
+      "step": 3060
+    },
+    {
+      "epoch": 0.3860566174386749,
+      "grad_norm": 0.45052894949913025,
+      "learning_rate": 0.000299251277032634,
+      "loss": 0.7995,
+      "step": 3065
+    },
+    {
+      "epoch": 0.3866863998488522,
+      "grad_norm": 0.46262305974960327,
+      "learning_rate": 0.0002992402677574852,
+      "loss": 0.8175,
+      "step": 3070
+    },
+    {
+      "epoch": 0.38731618225902953,
+      "grad_norm": 0.4934038519859314,
+      "learning_rate": 0.00029922917833690365,
+      "loss": 0.821,
+      "step": 3075
+    },
+    {
+      "epoch": 0.3879459646692068,
+      "grad_norm": 0.46494096517562866,
+      "learning_rate": 0.0002992180087768445,
+      "loss": 0.8081,
+      "step": 3080
+    },
+    {
+      "epoch": 0.3885757470793841,
+      "grad_norm": 0.9760459661483765,
+      "learning_rate": 0.0002992067590833062,
+      "loss": 0.7673,
+      "step": 3085
+    },
+    {
+      "epoch": 0.38920552948956133,
+      "grad_norm": 0.7070348262786865,
+      "learning_rate": 0.00029919542926233,
+      "loss": 0.8017,
+      "step": 3090
+    },
+    {
+      "epoch": 0.38983531189973863,
+      "grad_norm": 0.6773821711540222,
+      "learning_rate": 0.00029918401932000027,
+      "loss": 0.7946,
+      "step": 3095
+    },
+    {
+      "epoch": 0.39046509430991594,
+      "grad_norm": 0.4955935478210449,
+      "learning_rate": 0.0002991725292624445,
+      "loss": 0.8431,
+      "step": 3100
+    },
+    {
+      "epoch": 0.3910948767200932,
+      "grad_norm": 0.7728490829467773,
+      "learning_rate": 0.000299160959095833,
+      "loss": 0.8024,
+      "step": 3105
+    },
+    {
+      "epoch": 0.3917246591302705,
+      "grad_norm": 0.6880044341087341,
+      "learning_rate": 0.00029914930882637926,
+      "loss": 0.788,
+      "step": 3110
+    },
+    {
+      "epoch": 0.3923544415404478,
+      "grad_norm": 0.4916500747203827,
+      "learning_rate": 0.0002991375784603398,
+      "loss": 0.7878,
+      "step": 3115
+    },
+    {
+      "epoch": 0.39298422395062504,
+      "grad_norm": 0.5188093781471252,
+      "learning_rate": 0.00029912576800401403,
+      "loss": 0.8404,
+      "step": 3120
+    },
+    {
+      "epoch": 0.39361400636080235,
+      "grad_norm": 0.5084378123283386,
+      "learning_rate": 0.0002991138774637444,
+      "loss": 0.8277,
+      "step": 3125
+    },
+    {
+      "epoch": 0.39424378877097965,
+      "grad_norm": 0.4139776825904846,
+      "learning_rate": 0.0002991019068459165,
+      "loss": 0.7672,
+      "step": 3130
+    },
+    {
+      "epoch": 0.3948735711811569,
+      "grad_norm": 0.4756976366043091,
+      "learning_rate": 0.0002990898561569588,
+      "loss": 0.7936,
+      "step": 3135
+    },
+    {
+      "epoch": 0.3955033535913342,
+      "grad_norm": 0.46053630113601685,
+      "learning_rate": 0.0002990777254033427,
+      "loss": 0.8102,
+      "step": 3140
+    },
+    {
+      "epoch": 0.39613313600151145,
+      "grad_norm": 0.48546189069747925,
+      "learning_rate": 0.00029906551459158283,
+      "loss": 0.8184,
+      "step": 3145
+    },
+    {
+      "epoch": 0.39676291841168876,
+      "grad_norm": 0.477192223072052,
+      "learning_rate": 0.0002990532237282366,
+      "loss": 0.828,
+      "step": 3150
+    },
+    {
+      "epoch": 0.39739270082186606,
+      "grad_norm": 0.48900333046913147,
+      "learning_rate": 0.00029904085281990447,
+      "loss": 0.8183,
+      "step": 3155
+    },
+    {
+      "epoch": 0.3980224832320433,
+      "grad_norm": 0.5019087791442871,
+      "learning_rate": 0.0002990284018732299,
+      "loss": 0.8002,
+      "step": 3160
+    },
+    {
+      "epoch": 0.3986522656422206,
+      "grad_norm": 0.5127068758010864,
+      "learning_rate": 0.0002990158708948994,
+      "loss": 0.8088,
+      "step": 3165
+    },
+    {
+      "epoch": 0.3992820480523979,
+      "grad_norm": 0.44172775745391846,
+      "learning_rate": 0.00029900325989164233,
+      "loss": 0.8013,
+      "step": 3170
+    },
+    {
+      "epoch": 0.39991183046257517,
+      "grad_norm": 0.5318475961685181,
+      "learning_rate": 0.0002989905688702311,
+      "loss": 0.8239,
+      "step": 3175
+    },
+    {
+      "epoch": 0.4005416128727525,
+      "grad_norm": 0.4257467985153198,
+      "learning_rate": 0.0002989777978374811,
+      "loss": 0.7714,
+      "step": 3180
+    },
+    {
+      "epoch": 0.4011713952829297,
+      "grad_norm": 0.42196664214134216,
+      "learning_rate": 0.0002989649468002506,
+      "loss": 0.7987,
+      "step": 3185
+    },
+    {
+      "epoch": 0.401801177693107,
+      "grad_norm": 0.47977736592292786,
+      "learning_rate": 0.000298952015765441,
+      "loss": 0.7699,
+      "step": 3190
+    },
+    {
+      "epoch": 0.40243096010328433,
+      "grad_norm": 0.4841693639755249,
+      "learning_rate": 0.0002989390047399965,
+      "loss": 0.7916,
+      "step": 3195
+    },
+    {
+      "epoch": 0.4030607425134616,
+      "grad_norm": 0.5104061961174011,
+      "learning_rate": 0.0002989259137309043,
+      "loss": 0.8244,
+      "step": 3200
+    },
+    {
+      "epoch": 0.4036905249236389,
+      "grad_norm": 0.46594473719596863,
+      "learning_rate": 0.00029891274274519464,
+      "loss": 0.786,
+      "step": 3205
+    },
+    {
+      "epoch": 0.4043203073338162,
+      "grad_norm": 0.4309998154640198,
+      "learning_rate": 0.0002988994917899405,
+      "loss": 0.8266,
+      "step": 3210
+    },
+    {
+      "epoch": 0.40495008974399344,
+      "grad_norm": 0.4976588785648346,
+      "learning_rate": 0.00029888616087225817,
+      "loss": 0.7911,
+      "step": 3215
+    },
+    {
+      "epoch": 0.40557987215417074,
+      "grad_norm": 0.47657066583633423,
+      "learning_rate": 0.00029887274999930647,
+      "loss": 0.7926,
+      "step": 3220
+    },
+    {
+      "epoch": 0.40620965456434804,
+      "grad_norm": 0.42497026920318604,
+      "learning_rate": 0.0002988592591782874,
+      "loss": 0.7838,
+      "step": 3225
+    },
+    {
+      "epoch": 0.4068394369745253,
+      "grad_norm": 0.4974801540374756,
+      "learning_rate": 0.00029884568841644587,
+      "loss": 0.7854,
+      "step": 3230
+    },
+    {
+      "epoch": 0.4074692193847026,
+      "grad_norm": 0.43505486845970154,
+      "learning_rate": 0.00029883203772106966,
+      "loss": 0.8689,
+      "step": 3235
+    },
+    {
+      "epoch": 0.40809900179487985,
+      "grad_norm": 0.5216085314750671,
+      "learning_rate": 0.0002988183070994895,
+      "loss": 0.8445,
+      "step": 3240
+    },
+    {
+      "epoch": 0.40872878420505715,
+      "grad_norm": 0.5993830561637878,
+      "learning_rate": 0.0002988044965590791,
+      "loss": 0.7944,
+      "step": 3245
+    },
+    {
+      "epoch": 0.40935856661523445,
+      "grad_norm": 0.7245651483535767,
+      "learning_rate": 0.00029879060610725494,
+      "loss": 0.8175,
+      "step": 3250
+    },
+    {
+      "epoch": 0.4099883490254117,
+      "grad_norm": 0.4758714735507965,
+      "learning_rate": 0.00029877663575147653,
+      "loss": 0.7862,
+      "step": 3255
+    },
+    {
+      "epoch": 0.410618131435589,
+      "grad_norm": 0.5264742970466614,
+      "learning_rate": 0.0002987625854992464,
+      "loss": 0.7625,
+      "step": 3260
+    },
+    {
+      "epoch": 0.4112479138457663,
+      "grad_norm": 0.46857550740242004,
+      "learning_rate": 0.0002987484553581097,
+      "loss": 0.7878,
+      "step": 3265
+    },
+    {
+      "epoch": 0.41187769625594356,
+      "grad_norm": 0.4588899314403534,
+      "learning_rate": 0.0002987342453356547,
+      "loss": 0.8435,
+      "step": 3270
+    },
+    {
+      "epoch": 0.41250747866612086,
+      "grad_norm": 0.47005462646484375,
+      "learning_rate": 0.0002987199554395125,
+      "loss": 0.8343,
+      "step": 3275
+    },
+    {
+      "epoch": 0.4131372610762981,
+      "grad_norm": 0.4855548143386841,
+      "learning_rate": 0.00029870558567735716,
+      "loss": 0.7944,
+      "step": 3280
+    },
+    {
+      "epoch": 0.4137670434864754,
+      "grad_norm": 0.4832567572593689,
+      "learning_rate": 0.00029869113605690545,
+      "loss": 0.7999,
+      "step": 3285
+    },
+    {
+      "epoch": 0.4143968258966527,
+      "grad_norm": 0.4483296573162079,
+      "learning_rate": 0.00029867660658591724,
+      "loss": 0.8074,
+      "step": 3290
+    },
+    {
+      "epoch": 0.41502660830682997,
+      "grad_norm": 0.5084306001663208,
+      "learning_rate": 0.00029866199727219514,
+      "loss": 0.8173,
+      "step": 3295
+    },
+    {
+      "epoch": 0.4156563907170073,
+      "grad_norm": 0.43247321248054504,
+      "learning_rate": 0.00029864730812358473,
+      "loss": 0.7904,
+      "step": 3300
+    },
+    {
+      "epoch": 0.4162861731271846,
+      "grad_norm": 0.4278540313243866,
+      "learning_rate": 0.0002986325391479744,
+      "loss": 0.79,
+      "step": 3305
+    },
+    {
+      "epoch": 0.4169159555373618,
+      "grad_norm": 0.4396720230579376,
+      "learning_rate": 0.00029861769035329546,
+      "loss": 0.7737,
+      "step": 3310
+    },
+    {
+      "epoch": 0.41754573794753913,
+      "grad_norm": 0.4305702745914459,
+      "learning_rate": 0.0002986027617475219,
+      "loss": 0.8133,
+      "step": 3315
+    },
+    {
+      "epoch": 0.41817552035771643,
+      "grad_norm": 0.4455117881298065,
+      "learning_rate": 0.0002985877533386709,
+      "loss": 0.7932,
+      "step": 3320
+    },
+    {
+      "epoch": 0.4188053027678937,
+      "grad_norm": 0.45051881670951843,
+      "learning_rate": 0.00029857266513480226,
+      "loss": 0.8162,
+      "step": 3325
+    },
+    {
+      "epoch": 0.419435085178071,
+      "grad_norm": 0.47537773847579956,
+      "learning_rate": 0.0002985574971440187,
+      "loss": 0.7931,
+      "step": 3330
+    },
+    {
+      "epoch": 0.42006486758824824,
+      "grad_norm": 0.46828627586364746,
+      "learning_rate": 0.0002985422493744657,
+      "loss": 0.8399,
+      "step": 3335
+    },
+    {
+      "epoch": 0.42069464999842554,
+      "grad_norm": 0.4528372585773468,
+      "learning_rate": 0.00029852692183433176,
+      "loss": 0.7821,
+      "step": 3340
+    },
+    {
+      "epoch": 0.42132443240860284,
+      "grad_norm": 0.4476306736469269,
+      "learning_rate": 0.00029851151453184807,
+      "loss": 0.7986,
+      "step": 3345
+    },
+    {
+      "epoch": 0.4219542148187801,
+      "grad_norm": 0.4092450439929962,
+      "learning_rate": 0.00029849602747528874,
+      "loss": 0.7827,
+      "step": 3350
+    },
+    {
+      "epoch": 0.4225839972289574,
+      "grad_norm": 0.4776279330253601,
+      "learning_rate": 0.00029848046067297064,
+      "loss": 0.8269,
+      "step": 3355
+    },
+    {
+      "epoch": 0.4232137796391347,
+      "grad_norm": 0.45867466926574707,
+      "learning_rate": 0.00029846481413325346,
+      "loss": 0.8094,
+      "step": 3360
+    },
+    {
+      "epoch": 0.42384356204931195,
+      "grad_norm": 0.4665123522281647,
+      "learning_rate": 0.00029844908786453986,
+      "loss": 0.7288,
+      "step": 3365
+    },
+    {
+      "epoch": 0.42447334445948925,
+      "grad_norm": 0.4820057451725006,
+      "learning_rate": 0.0002984332818752751,
+      "loss": 0.8012,
+      "step": 3370
+    },
+    {
+      "epoch": 0.4251031268696665,
+      "grad_norm": 0.46213796734809875,
+      "learning_rate": 0.00029841739617394737,
+      "loss": 0.7693,
+      "step": 3375
+    },
+    {
+      "epoch": 0.4257329092798438,
+      "grad_norm": 0.4329429864883423,
+      "learning_rate": 0.0002984014307690878,
+      "loss": 0.7215,
+      "step": 3380
+    },
+    {
+      "epoch": 0.4263626916900211,
+      "grad_norm": 0.437621146440506,
+      "learning_rate": 0.00029838538566926993,
+      "loss": 0.7839,
+      "step": 3385
+    },
+    {
+      "epoch": 0.42699247410019836,
+      "grad_norm": 0.4661789536476135,
+      "learning_rate": 0.0002983692608831105,
+      "loss": 0.7827,
+      "step": 3390
+    },
+    {
+      "epoch": 0.42762225651037566,
+      "grad_norm": 0.4203425645828247,
+      "learning_rate": 0.0002983530564192689,
+      "loss": 0.8096,
+      "step": 3395
+    },
+    {
+      "epoch": 0.42825203892055297,
+      "grad_norm": 0.4614803194999695,
+      "learning_rate": 0.00029833677228644726,
+      "loss": 0.8189,
+      "step": 3400
+    },
+    {
+      "epoch": 0.4288818213307302,
+      "grad_norm": 0.4247860908508301,
+      "learning_rate": 0.0002983204084933905,
+      "loss": 0.8123,
+      "step": 3405
+    },
+    {
+      "epoch": 0.4295116037409075,
+      "grad_norm": 0.4418291449546814,
+      "learning_rate": 0.0002983039650488864,
+      "loss": 0.8036,
+      "step": 3410
+    },
+    {
+      "epoch": 0.4301413861510848,
+      "grad_norm": 0.46780282258987427,
+      "learning_rate": 0.00029828744196176547,
+      "loss": 0.8122,
+      "step": 3415
+    },
+    {
+      "epoch": 0.4307711685612621,
+      "grad_norm": 0.44973024725914,
+      "learning_rate": 0.0002982708392409009,
+      "loss": 0.7813,
+      "step": 3420
+    },
+    {
+      "epoch": 0.4314009509714394,
+      "grad_norm": 0.3922254741191864,
+      "learning_rate": 0.00029825415689520887,
+      "loss": 0.7809,
+      "step": 3425
+    },
+    {
+      "epoch": 0.4320307333816166,
+      "grad_norm": 0.391084223985672,
+      "learning_rate": 0.00029823739493364804,
+      "loss": 0.7757,
+      "step": 3430
+    },
+    {
+      "epoch": 0.43266051579179393,
+      "grad_norm": 0.4502919316291809,
+      "learning_rate": 0.00029822055336522005,
+      "loss": 0.7688,
+      "step": 3435
+    },
+    {
+      "epoch": 0.43329029820197124,
+      "grad_norm": 0.475436270236969,
+      "learning_rate": 0.0002982036321989692,
+      "loss": 0.7667,
+      "step": 3440
+    },
+    {
+      "epoch": 0.4339200806121485,
+      "grad_norm": 0.42362409830093384,
+      "learning_rate": 0.00029818663144398253,
+      "loss": 0.8098,
+      "step": 3445
+    },
+    {
+      "epoch": 0.4345498630223258,
+      "grad_norm": 0.45329517126083374,
+      "learning_rate": 0.0002981695511093898,
+      "loss": 0.7706,
+      "step": 3450
+    },
+    {
+      "epoch": 0.4351796454325031,
+      "grad_norm": 0.4297536313533783,
+      "learning_rate": 0.00029815239120436365,
+      "loss": 0.808,
+      "step": 3455
+    },
+    {
+      "epoch": 0.43580942784268034,
+      "grad_norm": 0.4800092577934265,
+      "learning_rate": 0.0002981351517381192,
+      "loss": 0.7973,
+      "step": 3460
+    },
+    {
+      "epoch": 0.43643921025285765,
+      "grad_norm": 0.5014777779579163,
+      "learning_rate": 0.00029811783271991454,
+      "loss": 0.8098,
+      "step": 3465
+    },
+    {
+      "epoch": 0.4370689926630349,
+      "grad_norm": 0.4412321448326111,
+      "learning_rate": 0.00029810043415905027,
+      "loss": 0.7669,
+      "step": 3470
+    },
+    {
+      "epoch": 0.4376987750732122,
+      "grad_norm": 0.4491146206855774,
+      "learning_rate": 0.00029808295606486993,
+      "loss": 0.7599,
+      "step": 3475
+    },
+    {
+      "epoch": 0.4383285574833895,
+      "grad_norm": 0.42482897639274597,
+      "learning_rate": 0.0002980653984467596,
+      "loss": 0.7501,
+      "step": 3480
+    },
+    {
+      "epoch": 0.43895833989356675,
+      "grad_norm": 0.4166581332683563,
+      "learning_rate": 0.0002980477613141482,
+      "loss": 0.7807,
+      "step": 3485
+    },
+    {
+      "epoch": 0.43958812230374406,
+      "grad_norm": 0.48076120018959045,
+      "learning_rate": 0.0002980300446765071,
+      "loss": 0.821,
+      "step": 3490
+    },
+    {
+      "epoch": 0.44021790471392136,
+      "grad_norm": 0.4148639142513275,
+      "learning_rate": 0.00029801224854335073,
+      "loss": 0.781,
+      "step": 3495
+    },
+    {
+      "epoch": 0.4408476871240986,
+      "grad_norm": 0.41731131076812744,
+      "learning_rate": 0.00029799437292423586,
+      "loss": 0.7784,
+      "step": 3500
+    },
+    {
+      "epoch": 0.4414774695342759,
+      "grad_norm": 0.4514264762401581,
+      "learning_rate": 0.00029797641782876224,
+      "loss": 0.8066,
+      "step": 3505
+    },
+    {
+      "epoch": 0.4421072519444532,
+      "grad_norm": 0.44717252254486084,
+      "learning_rate": 0.00029795838326657204,
+      "loss": 0.7761,
+      "step": 3510
+    },
+    {
+      "epoch": 0.44273703435463047,
+      "grad_norm": 0.42850586771965027,
+      "learning_rate": 0.00029794026924735034,
+      "loss": 0.783,
+      "step": 3515
+    },
+    {
+      "epoch": 0.44336681676480777,
+      "grad_norm": 0.7937319278717041,
+      "learning_rate": 0.00029792207578082476,
+      "loss": 0.7894,
+      "step": 3520
+    },
+    {
+      "epoch": 0.443996599174985,
+      "grad_norm": 0.4401470124721527,
+      "learning_rate": 0.0002979038028767656,
+      "loss": 0.8046,
+      "step": 3525
+    },
+    {
+      "epoch": 0.4446263815851623,
+      "grad_norm": 0.45515474677085876,
+      "learning_rate": 0.00029788545054498577,
+      "loss": 0.8095,
+      "step": 3530
+    },
+    {
+      "epoch": 0.4452561639953396,
+      "grad_norm": 0.4676735997200012,
+      "learning_rate": 0.00029786701879534093,
+      "loss": 0.7969,
+      "step": 3535
+    },
+    {
+      "epoch": 0.4458859464055169,
+      "grad_norm": 0.42322975397109985,
+      "learning_rate": 0.0002978485076377294,
+      "loss": 0.8336,
+      "step": 3540
+    },
+    {
+      "epoch": 0.4465157288156942,
+      "grad_norm": 0.4256497025489807,
+      "learning_rate": 0.000297829917082092,
+      "loss": 0.7773,
+      "step": 3545
+    },
+    {
+      "epoch": 0.4471455112258715,
+      "grad_norm": 0.40527772903442383,
+      "learning_rate": 0.00029781124713841237,
+      "loss": 0.8058,
+      "step": 3550
+    },
+    {
+      "epoch": 0.44777529363604873,
+      "grad_norm": 0.4047418534755707,
+      "learning_rate": 0.0002977924978167166,
+      "loss": 0.7769,
+      "step": 3555
+    },
+    {
+      "epoch": 0.44840507604622604,
+      "grad_norm": 0.4016299545764923,
+      "learning_rate": 0.00029777366912707366,
+      "loss": 0.7531,
+      "step": 3560
+    },
+    {
+      "epoch": 0.4490348584564033,
+      "grad_norm": 0.4176371395587921,
+      "learning_rate": 0.00029775476107959486,
+      "loss": 0.7865,
+      "step": 3565
+    },
+    {
+      "epoch": 0.4496646408665806,
+      "grad_norm": 0.44107553362846375,
+      "learning_rate": 0.00029773577368443426,
+      "loss": 0.7735,
+      "step": 3570
+    },
+    {
+      "epoch": 0.4502944232767579,
+      "grad_norm": 0.43897122144699097,
+      "learning_rate": 0.00029771670695178857,
+      "loss": 0.7715,
+      "step": 3575
+    },
+    {
+      "epoch": 0.45092420568693514,
+      "grad_norm": 0.4024025499820709,
+      "learning_rate": 0.000297697560891897,
+      "loss": 0.7442,
+      "step": 3580
+    },
+    {
+      "epoch": 0.45155398809711245,
+      "grad_norm": 0.4896734356880188,
+      "learning_rate": 0.0002976783355150415,
+      "loss": 0.829,
+      "step": 3585
+    },
+    {
+      "epoch": 0.45218377050728975,
+      "grad_norm": 0.47621142864227295,
+      "learning_rate": 0.0002976590308315465,
+      "loss": 0.7915,
+      "step": 3590
+    },
+    {
+      "epoch": 0.452813552917467,
+      "grad_norm": 0.3869519829750061,
+      "learning_rate": 0.00029763964685177905,
+      "loss": 0.7696,
+      "step": 3595
+    },
+    {
+      "epoch": 0.4534433353276443,
+      "grad_norm": 0.4245865046977997,
+      "learning_rate": 0.0002976201835861488,
+      "loss": 0.7281,
+      "step": 3600
+    },
+    {
+      "epoch": 0.4540731177378216,
+      "grad_norm": 0.40193241834640503,
+      "learning_rate": 0.0002976006410451079,
+      "loss": 0.7435,
+      "step": 3605
+    },
+    {
+      "epoch": 0.45470290014799886,
+      "grad_norm": 0.4142551124095917,
+      "learning_rate": 0.00029758101923915123,
+      "loss": 0.7627,
+      "step": 3610
+    },
+    {
+      "epoch": 0.45533268255817616,
+      "grad_norm": 0.4214671552181244,
+      "learning_rate": 0.0002975613181788162,
+      "loss": 0.8084,
+      "step": 3615
+    },
+    {
+      "epoch": 0.4559624649683534,
+      "grad_norm": 0.46768540143966675,
+      "learning_rate": 0.0002975415378746826,
+      "loss": 0.7502,
+      "step": 3620
+    },
+    {
+      "epoch": 0.4565922473785307,
+      "grad_norm": 0.43812718987464905,
+      "learning_rate": 0.00029752167833737295,
+      "loss": 0.7555,
+      "step": 3625
+    },
+    {
+      "epoch": 0.457222029788708,
+      "grad_norm": 0.44065701961517334,
+      "learning_rate": 0.00029750173957755223,
+      "loss": 0.7824,
+      "step": 3630
+    },
+    {
+      "epoch": 0.45785181219888527,
+      "grad_norm": 0.46632614731788635,
+      "learning_rate": 0.00029748172160592816,
+      "loss": 0.7787,
+      "step": 3635
+    },
+    {
+      "epoch": 0.45848159460906257,
+      "grad_norm": 0.3616549074649811,
+      "learning_rate": 0.00029746162443325066,
+      "loss": 0.766,
+      "step": 3640
+    },
+    {
+      "epoch": 0.4591113770192399,
+      "grad_norm": 0.39852583408355713,
+      "learning_rate": 0.00029744144807031253,
+      "loss": 0.7318,
+      "step": 3645
+    },
+    {
+      "epoch": 0.4597411594294171,
+      "grad_norm": 0.4082608222961426,
+      "learning_rate": 0.0002974211925279488,
+      "loss": 0.7726,
+      "step": 3650
+    },
+    {
+      "epoch": 0.4603709418395944,
+      "grad_norm": 0.4750503599643707,
+      "learning_rate": 0.00029740085781703726,
+      "loss": 0.7953,
+      "step": 3655
+    },
+    {
+      "epoch": 0.4610007242497717,
+      "grad_norm": 0.439531534910202,
+      "learning_rate": 0.0002973804439484981,
+      "loss": 0.788,
+      "step": 3660
+    },
+    {
+      "epoch": 0.461630506659949,
+      "grad_norm": 0.41563692688941956,
+      "learning_rate": 0.000297359950933294,
+      "loss": 0.7935,
+      "step": 3665
+    },
+    {
+      "epoch": 0.4622602890701263,
+      "grad_norm": 0.4645535945892334,
+      "learning_rate": 0.00029733937878243015,
+      "loss": 0.7716,
+      "step": 3670
+    },
+    {
+      "epoch": 0.46289007148030353,
+      "grad_norm": 0.4334595501422882,
+      "learning_rate": 0.0002973187275069544,
+      "loss": 0.7455,
+      "step": 3675
+    },
+    {
+      "epoch": 0.46351985389048084,
+      "grad_norm": 0.4452027678489685,
+      "learning_rate": 0.0002972979971179568,
+      "loss": 0.7533,
+      "step": 3680
+    },
+    {
+      "epoch": 0.46414963630065814,
+      "grad_norm": 0.4289001226425171,
+      "learning_rate": 0.0002972771876265701,
+      "loss": 0.8066,
+      "step": 3685
+    },
+    {
+      "epoch": 0.4647794187108354,
+      "grad_norm": 0.44446882605552673,
+      "learning_rate": 0.0002972562990439694,
+      "loss": 0.8017,
+      "step": 3690
+    },
+    {
+      "epoch": 0.4654092011210127,
+      "grad_norm": 0.4466266930103302,
+      "learning_rate": 0.00029723533138137256,
+      "loss": 0.7686,
+      "step": 3695
+    },
+    {
+      "epoch": 0.46603898353119,
+      "grad_norm": 0.44262874126434326,
+      "learning_rate": 0.0002972142846500395,
+      "loss": 0.7835,
+      "step": 3700
+    },
+    {
+      "epoch": 0.46666876594136725,
+      "grad_norm": 0.40932217240333557,
+      "learning_rate": 0.0002971931588612729,
+      "loss": 0.7844,
+      "step": 3705
+    },
+    {
+      "epoch": 0.46729854835154455,
+      "grad_norm": 0.38220685720443726,
+      "learning_rate": 0.0002971719540264177,
+      "loss": 0.7682,
+      "step": 3710
+    },
+    {
+      "epoch": 0.4679283307617218,
+      "grad_norm": 0.4890231788158417,
+      "learning_rate": 0.0002971506701568614,
+      "loss": 0.7883,
+      "step": 3715
+    },
+    {
+      "epoch": 0.4685581131718991,
+      "grad_norm": 0.44211700558662415,
+      "learning_rate": 0.00029712930726403397,
+      "loss": 0.7287,
+      "step": 3720
+    },
+    {
+      "epoch": 0.4691878955820764,
+      "grad_norm": 0.4585074782371521,
+      "learning_rate": 0.0002971078653594078,
+      "loss": 0.7452,
+      "step": 3725
+    },
+    {
+      "epoch": 0.46981767799225366,
+      "grad_norm": 0.3818283975124359,
+      "learning_rate": 0.00029708634445449754,
+      "loss": 0.751,
+      "step": 3730
+    },
+    {
+      "epoch": 0.47044746040243096,
+      "grad_norm": 0.4351007640361786,
+      "learning_rate": 0.00029706474456086054,
+      "loss": 0.7665,
+      "step": 3735
+    },
+    {
+      "epoch": 0.47107724281260827,
+      "grad_norm": 0.4167363941669464,
+      "learning_rate": 0.0002970430656900964,
+      "loss": 0.7421,
+      "step": 3740
+    },
+    {
+      "epoch": 0.4717070252227855,
+      "grad_norm": 0.40461474657058716,
+      "learning_rate": 0.0002970213078538472,
+      "loss": 0.7496,
+      "step": 3745
+    },
+    {
+      "epoch": 0.4723368076329628,
+      "grad_norm": 0.38994473218917847,
+      "learning_rate": 0.00029699947106379734,
+      "loss": 0.773,
+      "step": 3750
+    },
+    {
+      "epoch": 0.4729665900431401,
+      "grad_norm": 0.42335331439971924,
+      "learning_rate": 0.0002969775553316737,
+      "loss": 0.7496,
+      "step": 3755
+    },
+    {
+      "epoch": 0.47359637245331737,
+      "grad_norm": 0.39755743741989136,
+      "learning_rate": 0.0002969555606692455,
+      "loss": 0.7794,
+      "step": 3760
+    },
+    {
+      "epoch": 0.4742261548634947,
+      "grad_norm": 0.4671246409416199,
+      "learning_rate": 0.0002969334870883244,
+      "loss": 0.8289,
+      "step": 3765
+    },
+    {
+      "epoch": 0.4748559372736719,
+      "grad_norm": 0.4498395621776581,
+      "learning_rate": 0.00029691133460076443,
+      "loss": 0.7856,
+      "step": 3770
+    },
+    {
+      "epoch": 0.47548571968384923,
+      "grad_norm": 0.4068240225315094,
+      "learning_rate": 0.00029688910321846193,
+      "loss": 0.7572,
+      "step": 3775
+    },
+    {
+      "epoch": 0.47611550209402653,
+      "grad_norm": 0.4673171043395996,
+      "learning_rate": 0.0002968667929533557,
+      "loss": 0.7972,
+      "step": 3780
+    },
+    {
+      "epoch": 0.4767452845042038,
+      "grad_norm": 0.4210684597492218,
+      "learning_rate": 0.00029684440381742697,
+      "loss": 0.7566,
+      "step": 3785
+    },
+    {
+      "epoch": 0.4773750669143811,
+      "grad_norm": 0.4167214632034302,
+      "learning_rate": 0.000296821935822699,
+      "loss": 0.7383,
+      "step": 3790
+    },
+    {
+      "epoch": 0.4780048493245584,
+      "grad_norm": 0.3826481103897095,
+      "learning_rate": 0.0002967993889812378,
+      "loss": 0.7749,
+      "step": 3795
+    },
+    {
+      "epoch": 0.47863463173473564,
+      "grad_norm": 0.416202187538147,
+      "learning_rate": 0.0002967767633051514,
+      "loss": 0.7755,
+      "step": 3800
+    },
+    {
+      "epoch": 0.47926441414491294,
+      "grad_norm": 0.38089507818222046,
+      "learning_rate": 0.0002967540588065904,
+      "loss": 0.7813,
+      "step": 3805
+    },
+    {
+      "epoch": 0.4798941965550902,
+      "grad_norm": 0.5257447957992554,
+      "learning_rate": 0.0002967312754977476,
+      "loss": 0.7408,
+      "step": 3810
+    },
+    {
+      "epoch": 0.4805239789652675,
+      "grad_norm": 0.4472402334213257,
+      "learning_rate": 0.00029670841339085813,
+      "loss": 0.7946,
+      "step": 3815
+    },
+    {
+      "epoch": 0.4811537613754448,
+      "grad_norm": 0.39956340193748474,
+      "learning_rate": 0.00029668547249819957,
+      "loss": 0.7469,
+      "step": 3820
+    },
+    {
+      "epoch": 0.48178354378562205,
+      "grad_norm": 0.41508588194847107,
+      "learning_rate": 0.00029666245283209154,
+      "loss": 0.7328,
+      "step": 3825
+    },
+    {
+      "epoch": 0.48241332619579935,
+      "grad_norm": 0.3888874053955078,
+      "learning_rate": 0.00029663935440489624,
+      "loss": 0.7529,
+      "step": 3830
+    },
+    {
+      "epoch": 0.48304310860597666,
+      "grad_norm": 0.39619430899620056,
+      "learning_rate": 0.00029661617722901806,
+      "loss": 0.7406,
+      "step": 3835
+    },
+    {
+      "epoch": 0.4836728910161539,
+      "grad_norm": 0.38242536783218384,
+      "learning_rate": 0.0002965929213169036,
+      "loss": 0.7355,
+      "step": 3840
+    },
+    {
+      "epoch": 0.4843026734263312,
+      "grad_norm": 0.37065988779067993,
+      "learning_rate": 0.0002965695866810419,
+      "loss": 0.7087,
+      "step": 3845
+    },
+    {
+      "epoch": 0.4849324558365085,
+      "grad_norm": 0.4015233814716339,
+      "learning_rate": 0.0002965461733339641,
+      "loss": 0.7531,
+      "step": 3850
+    },
+    {
+      "epoch": 0.48556223824668576,
+      "grad_norm": 0.4394242465496063,
+      "learning_rate": 0.0002965226812882438,
+      "loss": 0.7619,
+      "step": 3855
+    },
+    {
+      "epoch": 0.48619202065686307,
+      "grad_norm": 0.39809074997901917,
+      "learning_rate": 0.00029649911055649666,
+      "loss": 0.7702,
+      "step": 3860
+    },
+    {
+      "epoch": 0.4868218030670403,
+      "grad_norm": 0.5184118747711182,
+      "learning_rate": 0.0002964754611513808,
+      "loss": 0.7931,
+      "step": 3865
+    },
+    {
+      "epoch": 0.4874515854772176,
+      "grad_norm": 0.5002544522285461,
+      "learning_rate": 0.00029645173308559644,
+      "loss": 0.7989,
+      "step": 3870
+    },
+    {
+      "epoch": 0.4880813678873949,
+      "grad_norm": 0.551458477973938,
+      "learning_rate": 0.0002964279263718861,
+      "loss": 0.7345,
+      "step": 3875
+    },
+    {
+      "epoch": 0.48871115029757217,
+      "grad_norm": 0.499612420797348,
+      "learning_rate": 0.0002964040410230345,
+      "loss": 0.7885,
+      "step": 3880
+    },
+    {
+      "epoch": 0.4893409327077495,
+      "grad_norm": 0.5279458165168762,
+      "learning_rate": 0.0002963800770518687,
+      "loss": 0.795,
+      "step": 3885
+    },
+    {
+      "epoch": 0.4899707151179268,
+      "grad_norm": 0.47077476978302,
+      "learning_rate": 0.0002963560344712578,
+      "loss": 0.7716,
+      "step": 3890
+    },
+    {
+      "epoch": 0.49060049752810403,
+      "grad_norm": 0.8377729058265686,
+      "learning_rate": 0.0002963319132941133,
+      "loss": 0.7625,
+      "step": 3895
+    },
+    {
+      "epoch": 0.49123027993828133,
+      "grad_norm": 0.43871793150901794,
+      "learning_rate": 0.0002963077135333888,
+      "loss": 0.734,
+      "step": 3900
+    },
+    {
+      "epoch": 0.4918600623484586,
+      "grad_norm": 0.44589656591415405,
+      "learning_rate": 0.00029628343520208004,
+      "loss": 0.7735,
+      "step": 3905
+    },
+    {
+      "epoch": 0.4924898447586359,
+      "grad_norm": 0.7113938927650452,
+      "learning_rate": 0.00029625907831322515,
+      "loss": 0.7611,
+      "step": 3910
+    },
+    {
+      "epoch": 0.4931196271688132,
+      "grad_norm": 0.3830680847167969,
+      "learning_rate": 0.0002962346428799043,
+      "loss": 0.7399,
+      "step": 3915
+    },
+    {
+      "epoch": 0.49374940957899044,
+      "grad_norm": 0.4787169396877289,
+      "learning_rate": 0.00029621012891523985,
+      "loss": 0.7572,
+      "step": 3920
+    },
+    {
+      "epoch": 0.49437919198916774,
+      "grad_norm": 0.428469717502594,
+      "learning_rate": 0.0002961855364323964,
+      "loss": 0.7548,
+      "step": 3925
+    },
+    {
+      "epoch": 0.49500897439934505,
+      "grad_norm": 0.3982272148132324,
+      "learning_rate": 0.00029616086544458065,
+      "loss": 0.7846,
+      "step": 3930
+    },
+    {
+      "epoch": 0.4956387568095223,
+      "grad_norm": 0.45961281657218933,
+      "learning_rate": 0.00029613611596504146,
+      "loss": 0.8041,
+      "step": 3935
+    },
+    {
+      "epoch": 0.4962685392196996,
+      "grad_norm": 0.400146484375,
+      "learning_rate": 0.00029611128800706996,
+      "loss": 0.7395,
+      "step": 3940
+    },
+    {
+      "epoch": 0.4968983216298769,
+      "grad_norm": 0.3984740674495697,
+      "learning_rate": 0.00029608638158399925,
+      "loss": 0.7569,
+      "step": 3945
+    },
+    {
+      "epoch": 0.49752810404005415,
+      "grad_norm": 0.4343632161617279,
+      "learning_rate": 0.0002960613967092046,
+      "loss": 0.7958,
+      "step": 3950
+    },
+    {
+      "epoch": 0.49815788645023146,
+      "grad_norm": 0.4569413363933563,
+      "learning_rate": 0.0002960363333961036,
+      "loss": 0.7673,
+      "step": 3955
+    },
+    {
+      "epoch": 0.4987876688604087,
+      "grad_norm": 0.4818612039089203,
+      "learning_rate": 0.0002960111916581557,
+      "loss": 0.7621,
+      "step": 3960
+    },
+    {
+      "epoch": 0.499417451270586,
+      "grad_norm": 0.4524887800216675,
+      "learning_rate": 0.0002959859715088626,
+      "loss": 0.7793,
+      "step": 3965
+    },
+    {
+      "epoch": 0.5000472336807633,
+      "grad_norm": 0.4414517283439636,
+      "learning_rate": 0.0002959606729617682,
+      "loss": 0.7557,
+      "step": 3970
+    },
+    {
+      "epoch": 0.5006770160909406,
+      "grad_norm": 0.3883852958679199,
+      "learning_rate": 0.0002959352960304583,
+      "loss": 0.7146,
+      "step": 3975
+    },
+    {
+      "epoch": 0.5013067985011178,
+      "grad_norm": 0.34819212555885315,
+      "learning_rate": 0.00029590984072856084,
+      "loss": 0.7271,
+      "step": 3980
+    },
+    {
+      "epoch": 0.5019365809112951,
+      "grad_norm": 0.3943585753440857,
+      "learning_rate": 0.0002958843070697461,
+      "loss": 0.817,
+      "step": 3985
+    },
+    {
+      "epoch": 0.5025663633214724,
+      "grad_norm": 0.3881372809410095,
+      "learning_rate": 0.000295858695067726,
+      "loss": 0.6997,
+      "step": 3990
+    },
+    {
+      "epoch": 0.5031961457316497,
+      "grad_norm": 0.4077765941619873,
+      "learning_rate": 0.00029583300473625497,
+      "loss": 0.789,
+      "step": 3995
+    },
+    {
+      "epoch": 0.503825928141827,
+      "grad_norm": 0.4519467353820801,
+      "learning_rate": 0.0002958072360891292,
+      "loss": 0.7081,
+      "step": 4000
+    },
+    {
+      "epoch": 0.503825928141827,
+      "eval_loss": 0.3237769305706024,
+      "eval_runtime": 6.2328,
+      "eval_samples_per_second": 160.442,
+      "eval_steps_per_second": 10.108,
+      "step": 4000
+    },
+    {
+      "epoch": 0.5044557105520043,
+      "grad_norm": 0.4161011278629303,
+      "learning_rate": 0.00029578138914018704,
+      "loss": 0.7426,
+      "step": 4005
+    },
+    {
+      "epoch": 0.5050854929621815,
+      "grad_norm": 0.4170863926410675,
+      "learning_rate": 0.0002957554639033089,
+      "loss": 0.7614,
+      "step": 4010
+    },
+    {
+      "epoch": 0.5057152753723588,
+      "grad_norm": 0.41827666759490967,
+      "learning_rate": 0.0002957294603924172,
+      "loss": 0.7339,
+      "step": 4015
+    },
+    {
+      "epoch": 0.5063450577825361,
+      "grad_norm": 0.4575699269771576,
+      "learning_rate": 0.0002957033786214766,
+      "loss": 0.7506,
+      "step": 4020
+    },
+    {
+      "epoch": 0.5069748401927134,
+      "grad_norm": 0.39499175548553467,
+      "learning_rate": 0.00029567721860449333,
+      "loss": 0.7227,
+      "step": 4025
+    },
+    {
+      "epoch": 0.5076046226028907,
+      "grad_norm": 0.4190625548362732,
+      "learning_rate": 0.00029565098035551606,
+      "loss": 0.7375,
+      "step": 4030
+    },
+    {
+      "epoch": 0.5082344050130679,
+      "grad_norm": 0.41470181941986084,
+      "learning_rate": 0.00029562466388863534,
+      "loss": 0.7953,
+      "step": 4035
+    },
+    {
+      "epoch": 0.5088641874232452,
+      "grad_norm": 0.5376180410385132,
+      "learning_rate": 0.00029559826921798373,
+      "loss": 0.7927,
+      "step": 4040
+    },
+    {
+      "epoch": 0.5094939698334225,
+      "grad_norm": 0.41073331236839294,
+      "learning_rate": 0.0002955717963577357,
+      "loss": 0.7175,
+      "step": 4045
+    },
+    {
+      "epoch": 0.5101237522435998,
+      "grad_norm": 0.3894195854663849,
+      "learning_rate": 0.0002955452453221078,
+      "loss": 0.743,
+      "step": 4050
+    },
+    {
+      "epoch": 0.5107535346537772,
+      "grad_norm": 0.37404918670654297,
+      "learning_rate": 0.00029551861612535856,
+      "loss": 0.6833,
+      "step": 4055
+    },
+    {
+      "epoch": 0.5113833170639545,
+      "grad_norm": 0.4107655882835388,
+      "learning_rate": 0.0002954919087817885,
+      "loss": 0.7588,
+      "step": 4060
+    },
+    {
+      "epoch": 0.5120130994741316,
+      "grad_norm": 0.4086790382862091,
+      "learning_rate": 0.00029546512330574004,
+      "loss": 0.7328,
+      "step": 4065
+    },
+    {
+      "epoch": 0.512642881884309,
+      "grad_norm": 0.4246830344200134,
+      "learning_rate": 0.0002954382597115976,
+      "loss": 0.7171,
+      "step": 4070
+    },
+    {
+      "epoch": 0.5132726642944863,
+      "grad_norm": 0.4021676182746887,
+      "learning_rate": 0.00029541131801378743,
+      "loss": 0.8009,
+      "step": 4075
+    },
+    {
+      "epoch": 0.5139024467046636,
+      "grad_norm": 0.42611315846443176,
+      "learning_rate": 0.00029538429822677806,
+      "loss": 0.7338,
+      "step": 4080
+    },
+    {
+      "epoch": 0.5145322291148409,
+      "grad_norm": 0.40971845388412476,
+      "learning_rate": 0.0002953572003650795,
+      "loss": 0.7883,
+      "step": 4085
+    },
+    {
+      "epoch": 0.5151620115250181,
+      "grad_norm": 0.4226076304912567,
+      "learning_rate": 0.0002953300244432441,
+      "loss": 0.7696,
+      "step": 4090
+    },
+    {
+      "epoch": 0.5157917939351954,
+      "grad_norm": 0.4504645764827728,
+      "learning_rate": 0.0002953027704758659,
+      "loss": 0.7123,
+      "step": 4095
+    },
+    {
+      "epoch": 0.5164215763453727,
+      "grad_norm": 0.4032662510871887,
+      "learning_rate": 0.00029527543847758086,
+      "loss": 0.6786,
+      "step": 4100
+    },
+    {
+      "epoch": 0.51705135875555,
+      "grad_norm": 0.4030795097351074,
+      "learning_rate": 0.00029524802846306694,
+      "loss": 0.7335,
+      "step": 4105
+    },
+    {
+      "epoch": 0.5176811411657273,
+      "grad_norm": 0.4887290596961975,
+      "learning_rate": 0.0002952205404470439,
+      "loss": 0.7238,
+      "step": 4110
+    },
+    {
+      "epoch": 0.5183109235759045,
+      "grad_norm": 0.4061615467071533,
+      "learning_rate": 0.00029519297444427343,
+      "loss": 0.7733,
+      "step": 4115
+    },
+    {
+      "epoch": 0.5189407059860818,
+      "grad_norm": 0.4060840308666229,
+      "learning_rate": 0.00029516533046955917,
+      "loss": 0.7268,
+      "step": 4120
+    },
+    {
+      "epoch": 0.5195704883962591,
+      "grad_norm": 0.4293743371963501,
+      "learning_rate": 0.0002951376085377465,
+      "loss": 0.7234,
+      "step": 4125
+    },
+    {
+      "epoch": 0.5202002708064364,
+      "grad_norm": 0.410264790058136,
+      "learning_rate": 0.00029510980866372273,
+      "loss": 0.774,
+      "step": 4130
+    },
+    {
+      "epoch": 0.5208300532166137,
+      "grad_norm": 0.3944232761859894,
+      "learning_rate": 0.0002950819308624171,
+      "loss": 0.7517,
+      "step": 4135
+    },
+    {
+      "epoch": 0.521459835626791,
+      "grad_norm": 0.4245687425136566,
+      "learning_rate": 0.0002950539751488005,
+      "loss": 0.7612,
+      "step": 4140
+    },
+    {
+      "epoch": 0.5220896180369682,
+      "grad_norm": 0.3795585632324219,
+      "learning_rate": 0.00029502594153788593,
+      "loss": 0.7778,
+      "step": 4145
+    },
+    {
+      "epoch": 0.5227194004471455,
+      "grad_norm": 0.42911338806152344,
+      "learning_rate": 0.000294997830044728,
+      "loss": 0.8033,
+      "step": 4150
+    },
+    {
+      "epoch": 0.5233491828573228,
+      "grad_norm": 0.4150259494781494,
+      "learning_rate": 0.0002949696406844232,
+      "loss": 0.7326,
+      "step": 4155
+    },
+    {
+      "epoch": 0.5239789652675001,
+      "grad_norm": 0.3846616744995117,
+      "learning_rate": 0.0002949413734721099,
+      "loss": 0.7085,
+      "step": 4160
+    },
+    {
+      "epoch": 0.5246087476776774,
+      "grad_norm": 0.3434165418148041,
+      "learning_rate": 0.00029491302842296824,
+      "loss": 0.711,
+      "step": 4165
+    },
+    {
+      "epoch": 0.5252385300878546,
+      "grad_norm": 0.33985382318496704,
+      "learning_rate": 0.0002948846055522202,
+      "loss": 0.7493,
+      "step": 4170
+    },
+    {
+      "epoch": 0.5258683124980319,
+      "grad_norm": 0.3809979259967804,
+      "learning_rate": 0.0002948561048751294,
+      "loss": 0.7224,
+      "step": 4175
+    },
+    {
+      "epoch": 0.5264980949082092,
+      "grad_norm": 0.45042338967323303,
+      "learning_rate": 0.00029482752640700143,
+      "loss": 0.7554,
+      "step": 4180
+    },
+    {
+      "epoch": 0.5271278773183865,
+      "grad_norm": 0.4068913757801056,
+      "learning_rate": 0.00029479887016318357,
+      "loss": 0.7267,
+      "step": 4185
+    },
+    {
+      "epoch": 0.5277576597285638,
+      "grad_norm": 0.41964098811149597,
+      "learning_rate": 0.0002947701361590649,
+      "loss": 0.7255,
+      "step": 4190
+    },
+    {
+      "epoch": 0.5283874421387411,
+      "grad_norm": 0.3956906795501709,
+      "learning_rate": 0.0002947413244100762,
+      "loss": 0.7272,
+      "step": 4195
+    },
+    {
+      "epoch": 0.5290172245489183,
+      "grad_norm": 0.41254857182502747,
+      "learning_rate": 0.0002947124349316901,
+      "loss": 0.7155,
+      "step": 4200
+    },
+    {
+      "epoch": 0.5296470069590956,
+      "grad_norm": 0.4162386655807495,
+      "learning_rate": 0.0002946834677394208,
+      "loss": 0.7729,
+      "step": 4205
+    },
+    {
+      "epoch": 0.5302767893692729,
+      "grad_norm": 0.4521070420742035,
+      "learning_rate": 0.00029465442284882436,
+      "loss": 0.7328,
+      "step": 4210
+    },
+    {
+      "epoch": 0.5309065717794502,
+      "grad_norm": 0.3702057898044586,
+      "learning_rate": 0.00029462530027549866,
+      "loss": 0.7592,
+      "step": 4215
+    },
+    {
+      "epoch": 0.5315363541896275,
+      "grad_norm": 0.4132764935493469,
+      "learning_rate": 0.00029459610003508313,
+      "loss": 0.7238,
+      "step": 4220
+    },
+    {
+      "epoch": 0.5321661365998047,
+      "grad_norm": 0.3817763328552246,
+      "learning_rate": 0.0002945668221432589,
+      "loss": 0.7524,
+      "step": 4225
+    },
+    {
+      "epoch": 0.532795919009982,
+      "grad_norm": 0.41137659549713135,
+      "learning_rate": 0.000294537466615749,
+      "loss": 0.7405,
+      "step": 4230
+    },
+    {
+      "epoch": 0.5334257014201593,
+      "grad_norm": 0.446150541305542,
+      "learning_rate": 0.00029450803346831787,
+      "loss": 0.7481,
+      "step": 4235
+    },
+    {
+      "epoch": 0.5340554838303366,
+      "grad_norm": 0.37535202503204346,
+      "learning_rate": 0.0002944785227167719,
+      "loss": 0.7505,
+      "step": 4240
+    },
+    {
+      "epoch": 0.5346852662405139,
+      "grad_norm": 0.4109747111797333,
+      "learning_rate": 0.000294448934376959,
+      "loss": 0.7406,
+      "step": 4245
+    },
+    {
+      "epoch": 0.5353150486506912,
+      "grad_norm": 0.4233269989490509,
+      "learning_rate": 0.00029441926846476873,
+      "loss": 0.7823,
+      "step": 4250
+    },
+    {
+      "epoch": 0.5359448310608684,
+      "grad_norm": 0.40127456188201904,
+      "learning_rate": 0.00029438952499613244,
+      "loss": 0.7486,
+      "step": 4255
+    },
+    {
+      "epoch": 0.5365746134710457,
+      "grad_norm": 0.40279653668403625,
+      "learning_rate": 0.000294359703987023,
+      "loss": 0.7157,
+      "step": 4260
+    },
+    {
+      "epoch": 0.537204395881223,
+      "grad_norm": 0.34208250045776367,
+      "learning_rate": 0.000294329805453455,
+      "loss": 0.7158,
+      "step": 4265
+    },
+    {
+      "epoch": 0.5378341782914003,
+      "grad_norm": 0.41574689745903015,
+      "learning_rate": 0.0002942998294114846,
+      "loss": 0.7668,
+      "step": 4270
+    },
+    {
+      "epoch": 0.5384639607015776,
+      "grad_norm": 0.401426762342453,
+      "learning_rate": 0.0002942697758772097,
+      "loss": 0.734,
+      "step": 4275
+    },
+    {
+      "epoch": 0.5390937431117548,
+      "grad_norm": 0.4085477292537689,
+      "learning_rate": 0.00029423964486676964,
+      "loss": 0.7448,
+      "step": 4280
+    },
+    {
+      "epoch": 0.5397235255219321,
+      "grad_norm": 0.43037959933280945,
+      "learning_rate": 0.0002942094363963456,
+      "loss": 0.7618,
+      "step": 4285
+    },
+    {
+      "epoch": 0.5403533079321095,
+      "grad_norm": 0.34685570001602173,
+      "learning_rate": 0.00029417915048216003,
+      "loss": 0.7314,
+      "step": 4290
+    },
+    {
+      "epoch": 0.5409830903422868,
+      "grad_norm": 0.3967381417751312,
+      "learning_rate": 0.00029414878714047725,
+      "loss": 0.7465,
+      "step": 4295
+    },
+    {
+      "epoch": 0.5416128727524641,
+      "grad_norm": 0.36378154158592224,
+      "learning_rate": 0.0002941183463876031,
+      "loss": 0.7372,
+      "step": 4300
+    },
+    {
+      "epoch": 0.5422426551626414,
+      "grad_norm": 0.3804253339767456,
+      "learning_rate": 0.00029408782823988494,
+      "loss": 0.7488,
+      "step": 4305
+    },
+    {
+      "epoch": 0.5428724375728186,
+      "grad_norm": 0.3679543137550354,
+      "learning_rate": 0.00029405723271371166,
+      "loss": 0.7253,
+      "step": 4310
+    },
+    {
+      "epoch": 0.5435022199829959,
+      "grad_norm": 0.35688257217407227,
+      "learning_rate": 0.0002940265598255138,
+      "loss": 0.7523,
+      "step": 4315
+    },
+    {
+      "epoch": 0.5441320023931732,
+      "grad_norm": 0.40890881419181824,
+      "learning_rate": 0.00029399580959176344,
+      "loss": 0.756,
+      "step": 4320
+    },
+    {
+      "epoch": 0.5447617848033505,
+      "grad_norm": 0.478547602891922,
+      "learning_rate": 0.00029396498202897406,
+      "loss": 0.7249,
+      "step": 4325
+    },
+    {
+      "epoch": 0.5453915672135278,
+      "grad_norm": 0.40117356181144714,
+      "learning_rate": 0.0002939340771537009,
+      "loss": 0.7466,
+      "step": 4330
+    },
+    {
+      "epoch": 0.546021349623705,
+      "grad_norm": 0.42868953943252563,
+      "learning_rate": 0.0002939030949825404,
+      "loss": 0.7894,
+      "step": 4335
+    },
+    {
+      "epoch": 0.5466511320338823,
+      "grad_norm": 0.41796940565109253,
+      "learning_rate": 0.0002938720355321309,
+      "loss": 0.7446,
+      "step": 4340
+    },
+    {
+      "epoch": 0.5472809144440596,
+      "grad_norm": 0.427336186170578,
+      "learning_rate": 0.0002938408988191519,
+      "loss": 0.7824,
+      "step": 4345
+    },
+    {
+      "epoch": 0.5479106968542369,
+      "grad_norm": 0.38179048895835876,
+      "learning_rate": 0.00029380968486032456,
+      "loss": 0.7427,
+      "step": 4350
+    },
+    {
+      "epoch": 0.5485404792644142,
+      "grad_norm": 0.39974477887153625,
+      "learning_rate": 0.0002937783936724115,
+      "loss": 0.7347,
+      "step": 4355
+    },
+    {
+      "epoch": 0.5491702616745914,
+      "grad_norm": 0.3805896043777466,
+      "learning_rate": 0.00029374702527221674,
+      "loss": 0.7547,
+      "step": 4360
+    },
+    {
+      "epoch": 0.5498000440847687,
+      "grad_norm": 0.43362486362457275,
+      "learning_rate": 0.0002937155796765859,
+      "loss": 0.7651,
+      "step": 4365
+    },
+    {
+      "epoch": 0.550429826494946,
+      "grad_norm": 0.38877996802330017,
+      "learning_rate": 0.000293684056902406,
+      "loss": 0.7054,
+      "step": 4370
+    },
+    {
+      "epoch": 0.5510596089051233,
+      "grad_norm": 0.393184095621109,
+      "learning_rate": 0.00029365245696660544,
+      "loss": 0.7453,
+      "step": 4375
+    },
+    {
+      "epoch": 0.5516893913153006,
+      "grad_norm": 0.3892836570739746,
+      "learning_rate": 0.0002936207798861541,
+      "loss": 0.7036,
+      "step": 4380
+    },
+    {
+      "epoch": 0.5523191737254779,
+      "grad_norm": 0.3737259805202484,
+      "learning_rate": 0.0002935890256780633,
+      "loss": 0.7403,
+      "step": 4385
+    },
+    {
+      "epoch": 0.5529489561356551,
+      "grad_norm": 0.36731937527656555,
+      "learning_rate": 0.00029355719435938585,
+      "loss": 0.7098,
+      "step": 4390
+    },
+    {
+      "epoch": 0.5535787385458324,
+      "grad_norm": 0.40238016843795776,
+      "learning_rate": 0.00029352528594721577,
+      "loss": 0.7625,
+      "step": 4395
+    },
+    {
+      "epoch": 0.5542085209560097,
+      "grad_norm": 0.3878697454929352,
+      "learning_rate": 0.0002934933004586887,
+      "loss": 0.7486,
+      "step": 4400
+    },
+    {
+      "epoch": 0.554838303366187,
+      "grad_norm": 0.36463412642478943,
+      "learning_rate": 0.00029346123791098157,
+      "loss": 0.7489,
+      "step": 4405
+    },
+    {
+      "epoch": 0.5554680857763643,
+      "grad_norm": 0.3860667049884796,
+      "learning_rate": 0.0002934290983213126,
+      "loss": 0.7503,
+      "step": 4410
+    },
+    {
+      "epoch": 0.5560978681865415,
+      "grad_norm": 0.40702390670776367,
+      "learning_rate": 0.0002933968817069417,
+      "loss": 0.6892,
+      "step": 4415
+    },
+    {
+      "epoch": 0.5567276505967188,
+      "grad_norm": 0.4769366979598999,
+      "learning_rate": 0.0002933645880851697,
+      "loss": 0.7285,
+      "step": 4420
+    },
+    {
+      "epoch": 0.5573574330068961,
+      "grad_norm": 0.37400034070014954,
+      "learning_rate": 0.00029333221747333913,
+      "loss": 0.7055,
+      "step": 4425
+    },
+    {
+      "epoch": 0.5579872154170734,
+      "grad_norm": 0.4280668795108795,
+      "learning_rate": 0.00029329976988883374,
+      "loss": 0.7629,
+      "step": 4430
+    },
+    {
+      "epoch": 0.5586169978272507,
+      "grad_norm": 0.3710954189300537,
+      "learning_rate": 0.00029326724534907856,
+      "loss": 0.696,
+      "step": 4435
+    },
+    {
+      "epoch": 0.559246780237428,
+      "grad_norm": 0.4311872720718384,
+      "learning_rate": 0.0002932346438715401,
+      "loss": 0.726,
+      "step": 4440
+    },
+    {
+      "epoch": 0.5598765626476052,
+      "grad_norm": 0.3708207309246063,
+      "learning_rate": 0.000293201965473726,
+      "loss": 0.7308,
+      "step": 4445
+    },
+    {
+      "epoch": 0.5605063450577825,
+      "grad_norm": 0.36177051067352295,
+      "learning_rate": 0.00029316921017318536,
+      "loss": 0.7403,
+      "step": 4450
+    },
+    {
+      "epoch": 0.5611361274679598,
+      "grad_norm": 0.4313011169433594,
+      "learning_rate": 0.0002931363779875086,
+      "loss": 0.7053,
+      "step": 4455
+    },
+    {
+      "epoch": 0.5617659098781371,
+      "grad_norm": 0.36055561900138855,
+      "learning_rate": 0.0002931034689343272,
+      "loss": 0.7544,
+      "step": 4460
+    },
+    {
+      "epoch": 0.5623956922883144,
+      "grad_norm": 0.37126588821411133,
+      "learning_rate": 0.0002930704830313142,
+      "loss": 0.7444,
+      "step": 4465
+    },
+    {
+      "epoch": 0.5630254746984916,
+      "grad_norm": 0.35056549310684204,
+      "learning_rate": 0.00029303742029618377,
+      "loss": 0.7251,
+      "step": 4470
+    },
+    {
+      "epoch": 0.5636552571086689,
+      "grad_norm": 0.3944834768772125,
+      "learning_rate": 0.0002930042807466913,
+      "loss": 0.771,
+      "step": 4475
+    },
+    {
+      "epoch": 0.5642850395188462,
+      "grad_norm": 0.39250391721725464,
+      "learning_rate": 0.0002929710644006334,
+      "loss": 0.7177,
+      "step": 4480
+    },
+    {
+      "epoch": 0.5649148219290235,
+      "grad_norm": 0.41848230361938477,
+      "learning_rate": 0.00029293777127584826,
+      "loss": 0.7362,
+      "step": 4485
+    },
+    {
+      "epoch": 0.5655446043392008,
+      "grad_norm": 0.3222586214542389,
+      "learning_rate": 0.00029290440139021477,
+      "loss": 0.6746,
+      "step": 4490
+    },
+    {
+      "epoch": 0.5661743867493781,
+      "grad_norm": 0.4275425672531128,
+      "learning_rate": 0.00029287095476165356,
+      "loss": 0.7641,
+      "step": 4495
+    },
+    {
+      "epoch": 0.5668041691595553,
+      "grad_norm": 0.37914222478866577,
+      "learning_rate": 0.0002928374314081261,
+      "loss": 0.7367,
+      "step": 4500
+    },
+    {
+      "epoch": 0.5674339515697326,
+      "grad_norm": 0.36903491616249084,
+      "learning_rate": 0.00029280383134763516,
+      "loss": 0.726,
+      "step": 4505
+    },
+    {
+      "epoch": 0.56806373397991,
+      "grad_norm": 0.45876601338386536,
+      "learning_rate": 0.0002927701545982249,
+      "loss": 0.7285,
+      "step": 4510
+    },
+    {
+      "epoch": 0.5686935163900873,
+      "grad_norm": 0.3885752856731415,
+      "learning_rate": 0.0002927364011779803,
+      "loss": 0.7111,
+      "step": 4515
+    },
+    {
+      "epoch": 0.5693232988002646,
+      "grad_norm": 0.3738529086112976,
+      "learning_rate": 0.00029270257110502784,
+      "loss": 0.7381,
+      "step": 4520
+    },
+    {
+      "epoch": 0.5699530812104417,
+      "grad_norm": 0.38678133487701416,
+      "learning_rate": 0.0002926686643975351,
+      "loss": 0.7069,
+      "step": 4525
+    },
+    {
+      "epoch": 0.570582863620619,
+      "grad_norm": 0.38699817657470703,
+      "learning_rate": 0.0002926346810737106,
+      "loss": 0.7456,
+      "step": 4530
+    },
+    {
+      "epoch": 0.5712126460307964,
+      "grad_norm": 0.39948272705078125,
+      "learning_rate": 0.0002926006211518043,
+      "loss": 0.7018,
+      "step": 4535
+    },
+    {
+      "epoch": 0.5718424284409737,
+      "grad_norm": 0.36441704630851746,
+      "learning_rate": 0.00029256648465010706,
+      "loss": 0.7155,
+      "step": 4540
+    },
+    {
+      "epoch": 0.572472210851151,
+      "grad_norm": 0.38412773609161377,
+      "learning_rate": 0.00029253227158695103,
+      "loss": 0.7131,
+      "step": 4545
+    },
+    {
+      "epoch": 0.5731019932613282,
+      "grad_norm": 0.3713320791721344,
+      "learning_rate": 0.0002924979819807094,
+      "loss": 0.7109,
+      "step": 4550
+    },
+    {
+      "epoch": 0.5737317756715055,
+      "grad_norm": 0.41460588574409485,
+      "learning_rate": 0.00029246361584979637,
+      "loss": 0.7218,
+      "step": 4555
+    },
+    {
+      "epoch": 0.5743615580816828,
+      "grad_norm": 0.37706735730171204,
+      "learning_rate": 0.0002924291732126675,
+      "loss": 0.7364,
+      "step": 4560
+    },
+    {
+      "epoch": 0.5749913404918601,
+      "grad_norm": 0.3931211829185486,
+      "learning_rate": 0.00029239465408781914,
+      "loss": 0.793,
+      "step": 4565
+    },
+    {
+      "epoch": 0.5756211229020374,
+      "grad_norm": 0.4280949831008911,
+      "learning_rate": 0.0002923600584937889,
+      "loss": 0.7577,
+      "step": 4570
+    },
+    {
+      "epoch": 0.5762509053122147,
+      "grad_norm": 0.408357173204422,
+      "learning_rate": 0.0002923253864491554,
+      "loss": 0.6866,
+      "step": 4575
+    },
+    {
+      "epoch": 0.5768806877223919,
+      "grad_norm": 0.3654685318470001,
+      "learning_rate": 0.0002922906379725383,
+      "loss": 0.7409,
+      "step": 4580
+    },
+    {
+      "epoch": 0.5775104701325692,
+      "grad_norm": 0.3723433017730713,
+      "learning_rate": 0.0002922558130825984,
+      "loss": 0.7106,
+      "step": 4585
+    },
+    {
+      "epoch": 0.5781402525427465,
+      "grad_norm": 0.40489017963409424,
+      "learning_rate": 0.00029222091179803735,
+      "loss": 0.7311,
+      "step": 4590
+    },
+    {
+      "epoch": 0.5787700349529238,
+      "grad_norm": 0.40270909667015076,
+      "learning_rate": 0.000292185934137598,
+      "loss": 0.7393,
+      "step": 4595
+    },
+    {
+      "epoch": 0.5793998173631011,
+      "grad_norm": 0.4228857159614563,
+      "learning_rate": 0.0002921508801200642,
+      "loss": 0.7253,
+      "step": 4600
+    },
+    {
+      "epoch": 0.5800295997732783,
+      "grad_norm": 0.39830881357192993,
+      "learning_rate": 0.0002921157497642607,
+      "loss": 0.7413,
+      "step": 4605
+    },
+    {
+      "epoch": 0.5806593821834556,
+      "grad_norm": 0.40520498156547546,
+      "learning_rate": 0.00029208054308905323,
+      "loss": 0.6902,
+      "step": 4610
+    },
+    {
+      "epoch": 0.5812891645936329,
+      "grad_norm": 0.3546881377696991,
+      "learning_rate": 0.0002920452601133487,
+      "loss": 0.7104,
+      "step": 4615
+    },
+    {
+      "epoch": 0.5819189470038102,
+      "grad_norm": 0.40294864773750305,
+      "learning_rate": 0.0002920099008560949,
+      "loss": 0.7258,
+      "step": 4620
+    },
+    {
+      "epoch": 0.5825487294139875,
+      "grad_norm": 0.36979302763938904,
+      "learning_rate": 0.0002919744653362804,
+      "loss": 0.708,
+      "step": 4625
+    },
+    {
+      "epoch": 0.5831785118241648,
+      "grad_norm": 0.42616382241249084,
+      "learning_rate": 0.000291938953572935,
+      "loss": 0.7044,
+      "step": 4630
+    },
+    {
+      "epoch": 0.583808294234342,
+      "grad_norm": 0.3644506335258484,
+      "learning_rate": 0.0002919033655851293,
+      "loss": 0.7277,
+      "step": 4635
+    },
+    {
+      "epoch": 0.5844380766445193,
+      "grad_norm": 0.34578534960746765,
+      "learning_rate": 0.0002918677013919749,
+      "loss": 0.7233,
+      "step": 4640
+    },
+    {
+      "epoch": 0.5850678590546966,
+      "grad_norm": 0.3914281725883484,
+      "learning_rate": 0.00029183196101262423,
+      "loss": 0.6829,
+      "step": 4645
+    },
+    {
+      "epoch": 0.5856976414648739,
+      "grad_norm": 0.35399550199508667,
+      "learning_rate": 0.0002917961444662707,
+      "loss": 0.7371,
+      "step": 4650
+    },
+    {
+      "epoch": 0.5863274238750512,
+      "grad_norm": 0.3999468684196472,
+      "learning_rate": 0.0002917602517721486,
+      "loss": 0.7228,
+      "step": 4655
+    },
+    {
+      "epoch": 0.5869572062852284,
+      "grad_norm": 0.4196580946445465,
+      "learning_rate": 0.0002917242829495332,
+      "loss": 0.7013,
+      "step": 4660
+    },
+    {
+      "epoch": 0.5875869886954057,
+      "grad_norm": 0.38301941752433777,
+      "learning_rate": 0.0002916882380177405,
+      "loss": 0.7409,
+      "step": 4665
+    },
+    {
+      "epoch": 0.588216771105583,
+      "grad_norm": 0.3997241258621216,
+      "learning_rate": 0.0002916521169961275,
+      "loss": 0.7216,
+      "step": 4670
+    },
+    {
+      "epoch": 0.5888465535157603,
+      "grad_norm": 0.3389524817466736,
+      "learning_rate": 0.00029161591990409203,
+      "loss": 0.7109,
+      "step": 4675
+    },
+    {
+      "epoch": 0.5894763359259376,
+      "grad_norm": 0.38282495737075806,
+      "learning_rate": 0.0002915796467610727,
+      "loss": 0.7608,
+      "step": 4680
+    },
+    {
+      "epoch": 0.5901061183361149,
+      "grad_norm": 0.40406349301338196,
+      "learning_rate": 0.000291543297586549,
+      "loss": 0.7062,
+      "step": 4685
+    },
+    {
+      "epoch": 0.5907359007462921,
+      "grad_norm": 0.37658101320266724,
+      "learning_rate": 0.0002915068724000413,
+      "loss": 0.7305,
+      "step": 4690
+    },
+    {
+      "epoch": 0.5913656831564694,
+      "grad_norm": 0.397401362657547,
+      "learning_rate": 0.0002914703712211108,
+      "loss": 0.7276,
+      "step": 4695
+    },
+    {
+      "epoch": 0.5919954655666467,
+      "grad_norm": 0.4348791539669037,
+      "learning_rate": 0.0002914337940693594,
+      "loss": 0.7572,
+      "step": 4700
+    },
+    {
+      "epoch": 0.592625247976824,
+      "grad_norm": 0.372659295797348,
+      "learning_rate": 0.0002913971409644299,
+      "loss": 0.7436,
+      "step": 4705
+    },
+    {
+      "epoch": 0.5932550303870013,
+      "grad_norm": 0.3933033049106598,
+      "learning_rate": 0.0002913604119260059,
+      "loss": 0.7229,
+      "step": 4710
+    },
+    {
+      "epoch": 0.5938848127971785,
+      "grad_norm": 0.35579994320869446,
+      "learning_rate": 0.0002913236069738116,
+      "loss": 0.7055,
+      "step": 4715
+    },
+    {
+      "epoch": 0.5945145952073558,
+      "grad_norm": 0.40102267265319824,
+      "learning_rate": 0.0002912867261276122,
+      "loss": 0.7167,
+      "step": 4720
+    },
+    {
+      "epoch": 0.5951443776175331,
+      "grad_norm": 0.3881862163543701,
+      "learning_rate": 0.0002912497694072136,
+      "loss": 0.7395,
+      "step": 4725
+    },
+    {
+      "epoch": 0.5957741600277104,
+      "grad_norm": 0.43878352642059326,
+      "learning_rate": 0.00029121273683246234,
+      "loss": 0.7251,
+      "step": 4730
+    },
+    {
+      "epoch": 0.5964039424378877,
+      "grad_norm": 0.3500851094722748,
+      "learning_rate": 0.0002911756284232457,
+      "loss": 0.6989,
+      "step": 4735
+    },
+    {
+      "epoch": 0.5970337248480649,
+      "grad_norm": 0.3887772262096405,
+      "learning_rate": 0.00029113844419949184,
+      "loss": 0.7324,
+      "step": 4740
+    },
+    {
+      "epoch": 0.5976635072582422,
+      "grad_norm": 0.376321405172348,
+      "learning_rate": 0.0002911011841811695,
+      "loss": 0.7239,
+      "step": 4745
+    },
+    {
+      "epoch": 0.5982932896684195,
+      "grad_norm": 0.3770900070667267,
+      "learning_rate": 0.00029106384838828816,
+      "loss": 0.6973,
+      "step": 4750
+    },
+    {
+      "epoch": 0.5989230720785969,
+      "grad_norm": 0.366205096244812,
+      "learning_rate": 0.000291026436840898,
+      "loss": 0.749,
+      "step": 4755
+    },
+    {
+      "epoch": 0.5995528544887742,
+      "grad_norm": 0.40675458312034607,
+      "learning_rate": 0.00029098894955908983,
+      "loss": 0.7155,
+      "step": 4760
+    },
+    {
+      "epoch": 0.6001826368989515,
+      "grad_norm": 0.4125664532184601,
+      "learning_rate": 0.0002909513865629953,
+      "loss": 0.7455,
+      "step": 4765
+    },
+    {
+      "epoch": 0.6008124193091287,
+      "grad_norm": 0.39872869849205017,
+      "learning_rate": 0.0002909137478727864,
+      "loss": 0.7194,
+      "step": 4770
+    },
+    {
+      "epoch": 0.601442201719306,
+      "grad_norm": 0.3824906051158905,
+      "learning_rate": 0.00029087603350867616,
+      "loss": 0.742,
+      "step": 4775
+    },
+    {
+      "epoch": 0.6020719841294833,
+      "grad_norm": 0.37914398312568665,
+      "learning_rate": 0.00029083824349091794,
+      "loss": 0.692,
+      "step": 4780
+    },
+    {
+      "epoch": 0.6027017665396606,
+      "grad_norm": 0.35589495301246643,
+      "learning_rate": 0.0002908003778398059,
+      "loss": 0.6706,
+      "step": 4785
+    },
+    {
+      "epoch": 0.6033315489498379,
+      "grad_norm": 0.34000104665756226,
+      "learning_rate": 0.0002907624365756748,
+      "loss": 0.7506,
+      "step": 4790
+    },
+    {
+      "epoch": 0.6039613313600151,
+      "grad_norm": 0.34795689582824707,
+      "learning_rate": 0.0002907244197188998,
+      "loss": 0.7097,
+      "step": 4795
+    },
+    {
+      "epoch": 0.6045911137701924,
+      "grad_norm": 0.38767385482788086,
+      "learning_rate": 0.00029068632728989697,
+      "loss": 0.6986,
+      "step": 4800
+    },
+    {
+      "epoch": 0.6052208961803697,
+      "grad_norm": 0.40651988983154297,
+      "learning_rate": 0.00029064815930912276,
+      "loss": 0.7159,
+      "step": 4805
+    },
+    {
+      "epoch": 0.605850678590547,
+      "grad_norm": 0.37715932726860046,
+      "learning_rate": 0.00029060991579707424,
+      "loss": 0.7189,
+      "step": 4810
+    },
+    {
+      "epoch": 0.6064804610007243,
+      "grad_norm": 0.3925745487213135,
+      "learning_rate": 0.0002905715967742891,
+      "loss": 0.6956,
+      "step": 4815
+    },
+    {
+      "epoch": 0.6071102434109016,
+      "grad_norm": 0.33669450879096985,
+      "learning_rate": 0.0002905332022613455,
+      "loss": 0.6806,
+      "step": 4820
+    },
+    {
+      "epoch": 0.6077400258210788,
+      "grad_norm": 0.38812607526779175,
+      "learning_rate": 0.00029049473227886214,
+      "loss": 0.6997,
+      "step": 4825
+    },
+    {
+      "epoch": 0.6083698082312561,
+      "grad_norm": 0.3890033960342407,
+      "learning_rate": 0.00029045618684749833,
+      "loss": 0.7306,
+      "step": 4830
+    },
+    {
+      "epoch": 0.6089995906414334,
+      "grad_norm": 0.4020345211029053,
+      "learning_rate": 0.00029041756598795383,
+      "loss": 0.7357,
+      "step": 4835
+    },
+    {
+      "epoch": 0.6096293730516107,
+      "grad_norm": 0.39244237542152405,
+      "learning_rate": 0.0002903788697209689,
+      "loss": 0.6956,
+      "step": 4840
+    },
+    {
+      "epoch": 0.610259155461788,
+      "grad_norm": 0.35866880416870117,
+      "learning_rate": 0.0002903400980673243,
+      "loss": 0.7219,
+      "step": 4845
+    },
+    {
+      "epoch": 0.6108889378719652,
+      "grad_norm": 0.3912501931190491,
+      "learning_rate": 0.0002903012510478414,
+      "loss": 0.7194,
+      "step": 4850
+    },
+    {
+      "epoch": 0.6115187202821425,
+      "grad_norm": 0.3933585584163666,
+      "learning_rate": 0.00029026232868338184,
+      "loss": 0.7136,
+      "step": 4855
+    },
+    {
+      "epoch": 0.6121485026923198,
+      "grad_norm": 0.37603482604026794,
+      "learning_rate": 0.0002902233309948479,
+      "loss": 0.7208,
+      "step": 4860
+    },
+    {
+      "epoch": 0.6127782851024971,
+      "grad_norm": 0.36919742822647095,
+      "learning_rate": 0.00029018425800318205,
+      "loss": 0.7499,
+      "step": 4865
+    },
+    {
+      "epoch": 0.6134080675126744,
+      "grad_norm": 0.38067975640296936,
+      "learning_rate": 0.0002901451097293676,
+      "loss": 0.7468,
+      "step": 4870
+    },
+    {
+      "epoch": 0.6140378499228517,
+      "grad_norm": 0.33954986929893494,
+      "learning_rate": 0.00029010588619442793,
+      "loss": 0.6894,
+      "step": 4875
+    },
+    {
+      "epoch": 0.6146676323330289,
+      "grad_norm": 0.38103437423706055,
+      "learning_rate": 0.000290066587419427,
+      "loss": 0.6661,
+      "step": 4880
+    },
+    {
+      "epoch": 0.6152974147432062,
+      "grad_norm": 0.3855966031551361,
+      "learning_rate": 0.00029002721342546924,
+      "loss": 0.7138,
+      "step": 4885
+    },
+    {
+      "epoch": 0.6159271971533835,
+      "grad_norm": 0.5084123611450195,
+      "learning_rate": 0.00028998776423369923,
+      "loss": 0.7005,
+      "step": 4890
+    },
+    {
+      "epoch": 0.6165569795635608,
+      "grad_norm": 0.36192139983177185,
+      "learning_rate": 0.0002899482398653022,
+      "loss": 0.7386,
+      "step": 4895
+    },
+    {
+      "epoch": 0.6171867619737381,
+      "grad_norm": 0.37423619627952576,
+      "learning_rate": 0.0002899086403415037,
+      "loss": 0.7172,
+      "step": 4900
+    },
+    {
+      "epoch": 0.6178165443839153,
+      "grad_norm": 0.3741579055786133,
+      "learning_rate": 0.00028986896568356933,
+      "loss": 0.7519,
+      "step": 4905
+    },
+    {
+      "epoch": 0.6184463267940926,
+      "grad_norm": 0.4323353171348572,
+      "learning_rate": 0.0002898292159128055,
+      "loss": 0.7325,
+      "step": 4910
+    },
+    {
+      "epoch": 0.6190761092042699,
+      "grad_norm": 0.3273026645183563,
+      "learning_rate": 0.00028978939105055873,
+      "loss": 0.7211,
+      "step": 4915
+    },
+    {
+      "epoch": 0.6197058916144472,
+      "grad_norm": 0.38831016421318054,
+      "learning_rate": 0.0002897494911182158,
+      "loss": 0.6435,
+      "step": 4920
+    },
+    {
+      "epoch": 0.6203356740246245,
+      "grad_norm": 0.36923748254776,
+      "learning_rate": 0.00028970951613720397,
+      "loss": 0.7184,
+      "step": 4925
+    },
+    {
+      "epoch": 0.6209654564348017,
+      "grad_norm": 0.3658188283443451,
+      "learning_rate": 0.0002896694661289906,
+      "loss": 0.7171,
+      "step": 4930
+    },
+    {
+      "epoch": 0.621595238844979,
+      "grad_norm": 0.3589092493057251,
+      "learning_rate": 0.00028962934111508357,
+      "loss": 0.7173,
+      "step": 4935
+    },
+    {
+      "epoch": 0.6222250212551563,
+      "grad_norm": 0.41886886954307556,
+      "learning_rate": 0.00028958914111703086,
+      "loss": 0.7412,
+      "step": 4940
+    },
+    {
+      "epoch": 0.6228548036653336,
+      "grad_norm": 0.34496763348579407,
+      "learning_rate": 0.0002895488661564208,
+      "loss": 0.6608,
+      "step": 4945
+    },
+    {
+      "epoch": 0.6234845860755109,
+      "grad_norm": 0.3527592122554779,
+      "learning_rate": 0.000289508516254882,
+      "loss": 0.7179,
+      "step": 4950
+    },
+    {
+      "epoch": 0.6241143684856882,
+      "grad_norm": 0.3406129479408264,
+      "learning_rate": 0.0002894680914340833,
+      "loss": 0.6862,
+      "step": 4955
+    },
+    {
+      "epoch": 0.6247441508958654,
+      "grad_norm": 0.33078086376190186,
+      "learning_rate": 0.00028942759171573374,
+      "loss": 0.6804,
+      "step": 4960
+    },
+    {
+      "epoch": 0.6253739333060427,
+      "grad_norm": 0.3582599461078644,
+      "learning_rate": 0.00028938701712158247,
+      "loss": 0.6681,
+      "step": 4965
+    },
+    {
+      "epoch": 0.62600371571622,
+      "grad_norm": 0.3656567633152008,
+      "learning_rate": 0.0002893463676734191,
+      "loss": 0.6714,
+      "step": 4970
+    },
+    {
+      "epoch": 0.6266334981263973,
+      "grad_norm": 0.35537272691726685,
+      "learning_rate": 0.00028930564339307337,
+      "loss": 0.6917,
+      "step": 4975
+    },
+    {
+      "epoch": 0.6272632805365747,
+      "grad_norm": 0.35100945830345154,
+      "learning_rate": 0.0002892648443024149,
+      "loss": 0.7217,
+      "step": 4980
+    },
+    {
+      "epoch": 0.6278930629467518,
+      "grad_norm": 0.34070494771003723,
+      "learning_rate": 0.000289223970423354,
+      "loss": 0.7237,
+      "step": 4985
+    },
+    {
+      "epoch": 0.6285228453569291,
+      "grad_norm": 0.3810268044471741,
+      "learning_rate": 0.00028918302177784075,
+      "loss": 0.7513,
+      "step": 4990
+    },
+    {
+      "epoch": 0.6291526277671065,
+      "grad_norm": 0.3511486053466797,
+      "learning_rate": 0.0002891419983878655,
+      "loss": 0.7112,
+      "step": 4995
+    },
+    {
+      "epoch": 0.6297824101772838,
+      "grad_norm": 0.30101874470710754,
+      "learning_rate": 0.0002891009002754588,
+      "loss": 0.6666,
+      "step": 5000
+    },
+    {
+      "epoch": 0.6297824101772838,
+      "eval_loss": 0.31327521800994873,
+      "eval_runtime": 6.2403,
+      "eval_samples_per_second": 160.248,
+      "eval_steps_per_second": 10.096,
+      "step": 5000
+    },
+    {
+      "epoch": 0.6304121925874611,
+      "grad_norm": 0.3446876108646393,
+      "learning_rate": 0.00028905972746269125,
+      "loss": 0.6651,
+      "step": 5005
+    },
+    {
+      "epoch": 0.6310419749976384,
+      "grad_norm": 0.3606228232383728,
+      "learning_rate": 0.0002890184799716736,
+      "loss": 0.7387,
+      "step": 5010
+    },
+    {
+      "epoch": 0.6316717574078156,
+      "grad_norm": 0.37057119607925415,
+      "learning_rate": 0.0002889771578245567,
+      "loss": 0.7044,
+      "step": 5015
+    },
+    {
+      "epoch": 0.6323015398179929,
+      "grad_norm": 0.36304429173469543,
+      "learning_rate": 0.0002889357610435314,
+      "loss": 0.7391,
+      "step": 5020
+    },
+    {
+      "epoch": 0.6329313222281702,
+      "grad_norm": 0.38329148292541504,
+      "learning_rate": 0.00028889428965082886,
+      "loss": 0.7045,
+      "step": 5025
+    },
+    {
+      "epoch": 0.6335611046383475,
+      "grad_norm": 0.3362608850002289,
+      "learning_rate": 0.00028885274366872006,
+      "loss": 0.6865,
+      "step": 5030
+    },
+    {
+      "epoch": 0.6341908870485248,
+      "grad_norm": 0.4079527258872986,
+      "learning_rate": 0.00028881112311951625,
+      "loss": 0.6892,
+      "step": 5035
+    },
+    {
+      "epoch": 0.634820669458702,
+      "grad_norm": 0.35261860489845276,
+      "learning_rate": 0.00028876942802556847,
+      "loss": 0.7189,
+      "step": 5040
+    },
+    {
+      "epoch": 0.6354504518688793,
+      "grad_norm": 0.40486040711402893,
+      "learning_rate": 0.00028872765840926804,
+      "loss": 0.7385,
+      "step": 5045
+    },
+    {
+      "epoch": 0.6360802342790566,
+      "grad_norm": 0.32852765917778015,
+      "learning_rate": 0.0002886858142930462,
+      "loss": 0.6267,
+      "step": 5050
+    },
+    {
+      "epoch": 0.6367100166892339,
+      "grad_norm": 0.31455445289611816,
+      "learning_rate": 0.0002886438956993741,
+      "loss": 0.6813,
+      "step": 5055
+    },
+    {
+      "epoch": 0.6373397990994112,
+      "grad_norm": 0.3047012686729431,
+      "learning_rate": 0.00028860190265076304,
+      "loss": 0.6862,
+      "step": 5060
+    },
+    {
+      "epoch": 0.6379695815095885,
+      "grad_norm": 0.34203359484672546,
+      "learning_rate": 0.0002885598351697643,
+      "loss": 0.6996,
+      "step": 5065
+    },
+    {
+      "epoch": 0.6385993639197657,
+      "grad_norm": 0.4077922999858856,
+      "learning_rate": 0.0002885176932789691,
+      "loss": 0.7018,
+      "step": 5070
+    },
+    {
+      "epoch": 0.639229146329943,
+      "grad_norm": 0.3590135872364044,
+      "learning_rate": 0.00028847547700100836,
+      "loss": 0.6741,
+      "step": 5075
+    },
+    {
+      "epoch": 0.6398589287401203,
+      "grad_norm": 0.33030763268470764,
+      "learning_rate": 0.0002884331863585535,
+      "loss": 0.6775,
+      "step": 5080
+    },
+    {
+      "epoch": 0.6404887111502976,
+      "grad_norm": 0.3921838104724884,
+      "learning_rate": 0.0002883908213743153,
+      "loss": 0.7359,
+      "step": 5085
+    },
+    {
+      "epoch": 0.6411184935604749,
+      "grad_norm": 0.35765379667282104,
+      "learning_rate": 0.0002883483820710449,
+      "loss": 0.6953,
+      "step": 5090
+    },
+    {
+      "epoch": 0.6417482759706521,
+      "grad_norm": 0.3486902415752411,
+      "learning_rate": 0.0002883058684715331,
+      "loss": 0.6848,
+      "step": 5095
+    },
+    {
+      "epoch": 0.6423780583808294,
+      "grad_norm": 0.35446256399154663,
+      "learning_rate": 0.0002882632805986108,
+      "loss": 0.7031,
+      "step": 5100
+    },
+    {
+      "epoch": 0.6430078407910067,
+      "grad_norm": 0.3666916489601135,
+      "learning_rate": 0.00028822061847514843,
+      "loss": 0.7135,
+      "step": 5105
+    },
+    {
+      "epoch": 0.643637623201184,
+      "grad_norm": 0.38766369223594666,
+      "learning_rate": 0.00028817788212405666,
+      "loss": 0.6623,
+      "step": 5110
+    },
+    {
+      "epoch": 0.6442674056113613,
+      "grad_norm": 0.3532891273498535,
+      "learning_rate": 0.0002881350715682859,
+      "loss": 0.699,
+      "step": 5115
+    },
+    {
+      "epoch": 0.6448971880215385,
+      "grad_norm": 0.36512479186058044,
+      "learning_rate": 0.0002880921868308263,
+      "loss": 0.6859,
+      "step": 5120
+    },
+    {
+      "epoch": 0.6455269704317158,
+      "grad_norm": 0.34285515546798706,
+      "learning_rate": 0.0002880492279347081,
+      "loss": 0.7254,
+      "step": 5125
+    },
+    {
+      "epoch": 0.6461567528418931,
+      "grad_norm": 0.3731713891029358,
+      "learning_rate": 0.00028800619490300107,
+      "loss": 0.6995,
+      "step": 5130
+    },
+    {
+      "epoch": 0.6467865352520704,
+      "grad_norm": 0.37182632088661194,
+      "learning_rate": 0.000287963087758815,
+      "loss": 0.7262,
+      "step": 5135
+    },
+    {
+      "epoch": 0.6474163176622477,
+      "grad_norm": 0.371231347322464,
+      "learning_rate": 0.0002879199065252994,
+      "loss": 0.7051,
+      "step": 5140
+    },
+    {
+      "epoch": 0.648046100072425,
+      "grad_norm": 0.35507723689079285,
+      "learning_rate": 0.00028787665122564357,
+      "loss": 0.6799,
+      "step": 5145
+    },
+    {
+      "epoch": 0.6486758824826022,
+      "grad_norm": 0.4001401662826538,
+      "learning_rate": 0.0002878333218830766,
+      "loss": 0.7718,
+      "step": 5150
+    },
+    {
+      "epoch": 0.6493056648927795,
+      "grad_norm": 0.36585733294487,
+      "learning_rate": 0.0002877899185208673,
+      "loss": 0.6652,
+      "step": 5155
+    },
+    {
+      "epoch": 0.6499354473029568,
+      "grad_norm": 0.3719576895236969,
+      "learning_rate": 0.00028774644116232436,
+      "loss": 0.7232,
+      "step": 5160
+    },
+    {
+      "epoch": 0.6505652297131341,
+      "grad_norm": 0.40236014127731323,
+      "learning_rate": 0.000287702889830796,
+      "loss": 0.6697,
+      "step": 5165
+    },
+    {
+      "epoch": 0.6511950121233114,
+      "grad_norm": 0.4343264400959015,
+      "learning_rate": 0.00028765926454967037,
+      "loss": 0.6877,
+      "step": 5170
+    },
+    {
+      "epoch": 0.6518247945334886,
+      "grad_norm": 0.3576568067073822,
+      "learning_rate": 0.00028761556534237514,
+      "loss": 0.7239,
+      "step": 5175
+    },
+    {
+      "epoch": 0.6524545769436659,
+      "grad_norm": 0.33383145928382874,
+      "learning_rate": 0.00028757179223237793,
+      "loss": 0.6822,
+      "step": 5180
+    },
+    {
+      "epoch": 0.6530843593538432,
+      "grad_norm": 0.353253573179245,
+      "learning_rate": 0.0002875279452431858,
+      "loss": 0.6925,
+      "step": 5185
+    },
+    {
+      "epoch": 0.6537141417640205,
+      "grad_norm": 0.3755667209625244,
+      "learning_rate": 0.0002874840243983455,
+      "loss": 0.6872,
+      "step": 5190
+    },
+    {
+      "epoch": 0.6543439241741978,
+      "grad_norm": 0.3973848521709442,
+      "learning_rate": 0.00028744002972144376,
+      "loss": 0.7251,
+      "step": 5195
+    },
+    {
+      "epoch": 0.6549737065843751,
+      "grad_norm": 0.3476422131061554,
+      "learning_rate": 0.0002873959612361066,
+      "loss": 0.6964,
+      "step": 5200
+    },
+    {
+      "epoch": 0.6556034889945523,
+      "grad_norm": 0.42737796902656555,
+      "learning_rate": 0.0002873518189659997,
+      "loss": 0.7106,
+      "step": 5205
+    },
+    {
+      "epoch": 0.6562332714047296,
+      "grad_norm": 0.3009507358074188,
+      "learning_rate": 0.00028730760293482863,
+      "loss": 0.6614,
+      "step": 5210
+    },
+    {
+      "epoch": 0.656863053814907,
+      "grad_norm": 0.38053247332572937,
+      "learning_rate": 0.00028726331316633835,
+      "loss": 0.6963,
+      "step": 5215
+    },
+    {
+      "epoch": 0.6574928362250843,
+      "grad_norm": 0.4153291583061218,
+      "learning_rate": 0.00028721894968431345,
+      "loss": 0.7471,
+      "step": 5220
+    },
+    {
+      "epoch": 0.6581226186352616,
+      "grad_norm": 0.36470016837120056,
+      "learning_rate": 0.0002871745125125782,
+      "loss": 0.6558,
+      "step": 5225
+    },
+    {
+      "epoch": 0.6587524010454388,
+      "grad_norm": 0.3935704827308655,
+      "learning_rate": 0.00028713000167499627,
+      "loss": 0.7025,
+      "step": 5230
+    },
+    {
+      "epoch": 0.659382183455616,
+      "grad_norm": 0.36777618527412415,
+      "learning_rate": 0.0002870854171954711,
+      "loss": 0.7386,
+      "step": 5235
+    },
+    {
+      "epoch": 0.6600119658657934,
+      "grad_norm": 0.36549127101898193,
+      "learning_rate": 0.0002870407590979455,
+      "loss": 0.703,
+      "step": 5240
+    },
+    {
+      "epoch": 0.6606417482759707,
+      "grad_norm": 0.37523144483566284,
+      "learning_rate": 0.00028699602740640194,
+      "loss": 0.6708,
+      "step": 5245
+    },
+    {
+      "epoch": 0.661271530686148,
+      "grad_norm": 0.3451475203037262,
+      "learning_rate": 0.00028695122214486237,
+      "loss": 0.6776,
+      "step": 5250
+    },
+    {
+      "epoch": 0.6619013130963253,
+      "grad_norm": 0.35215169191360474,
+      "learning_rate": 0.00028690634333738816,
+      "loss": 0.6983,
+      "step": 5255
+    },
+    {
+      "epoch": 0.6625310955065025,
+      "grad_norm": 0.37627631425857544,
+      "learning_rate": 0.00028686139100808037,
+      "loss": 0.6844,
+      "step": 5260
+    },
+    {
+      "epoch": 0.6631608779166798,
+      "grad_norm": 0.34171178936958313,
+      "learning_rate": 0.0002868163651810793,
+      "loss": 0.7068,
+      "step": 5265
+    },
+    {
+      "epoch": 0.6637906603268571,
+      "grad_norm": 0.3566179573535919,
+      "learning_rate": 0.0002867712658805649,
+      "loss": 0.6618,
+      "step": 5270
+    },
+    {
+      "epoch": 0.6644204427370344,
+      "grad_norm": 0.3453030586242676,
+      "learning_rate": 0.00028672609313075664,
+      "loss": 0.7046,
+      "step": 5275
+    },
+    {
+      "epoch": 0.6650502251472117,
+      "grad_norm": 0.40633949637413025,
+      "learning_rate": 0.00028668084695591316,
+      "loss": 0.6931,
+      "step": 5280
+    },
+    {
+      "epoch": 0.6656800075573889,
+      "grad_norm": 0.3927484154701233,
+      "learning_rate": 0.00028663552738033275,
+      "loss": 0.7051,
+      "step": 5285
+    },
+    {
+      "epoch": 0.6663097899675662,
+      "grad_norm": 0.35829389095306396,
+      "learning_rate": 0.000286590134428353,
+      "loss": 0.7051,
+      "step": 5290
+    },
+    {
+      "epoch": 0.6669395723777435,
+      "grad_norm": 0.4202066957950592,
+      "learning_rate": 0.00028654466812435105,
+      "loss": 0.7179,
+      "step": 5295
+    },
+    {
+      "epoch": 0.6675693547879208,
+      "grad_norm": 0.37852293252944946,
+      "learning_rate": 0.0002864991284927433,
+      "loss": 0.7107,
+      "step": 5300
+    },
+    {
+      "epoch": 0.6681991371980981,
+      "grad_norm": 0.3831678330898285,
+      "learning_rate": 0.0002864535155579856,
+      "loss": 0.659,
+      "step": 5305
+    },
+    {
+      "epoch": 0.6688289196082754,
+      "grad_norm": 0.3563750684261322,
+      "learning_rate": 0.0002864078293445731,
+      "loss": 0.7111,
+      "step": 5310
+    },
+    {
+      "epoch": 0.6694587020184526,
+      "grad_norm": 0.3460354804992676,
+      "learning_rate": 0.0002863620698770403,
+      "loss": 0.6822,
+      "step": 5315
+    },
+    {
+      "epoch": 0.6700884844286299,
+      "grad_norm": 0.36469632387161255,
+      "learning_rate": 0.0002863162371799612,
+      "loss": 0.6298,
+      "step": 5320
+    },
+    {
+      "epoch": 0.6707182668388072,
+      "grad_norm": 0.3730217218399048,
+      "learning_rate": 0.00028627033127794896,
+      "loss": 0.7137,
+      "step": 5325
+    },
+    {
+      "epoch": 0.6713480492489845,
+      "grad_norm": 0.347002774477005,
+      "learning_rate": 0.00028622435219565606,
+      "loss": 0.6873,
+      "step": 5330
+    },
+    {
+      "epoch": 0.6719778316591618,
+      "grad_norm": 0.35723358392715454,
+      "learning_rate": 0.00028617829995777433,
+      "loss": 0.7055,
+      "step": 5335
+    },
+    {
+      "epoch": 0.672607614069339,
+      "grad_norm": 0.3175225257873535,
+      "learning_rate": 0.0002861321745890349,
+      "loss": 0.6702,
+      "step": 5340
+    },
+    {
+      "epoch": 0.6732373964795163,
+      "grad_norm": 0.3599521517753601,
+      "learning_rate": 0.00028608597611420807,
+      "loss": 0.6646,
+      "step": 5345
+    },
+    {
+      "epoch": 0.6738671788896936,
+      "grad_norm": 0.4381812810897827,
+      "learning_rate": 0.00028603970455810357,
+      "loss": 0.7122,
+      "step": 5350
+    },
+    {
+      "epoch": 0.6744969612998709,
+      "grad_norm": 0.3400894105434418,
+      "learning_rate": 0.00028599335994557027,
+      "loss": 0.705,
+      "step": 5355
+    },
+    {
+      "epoch": 0.6751267437100482,
+      "grad_norm": 0.3332962989807129,
+      "learning_rate": 0.00028594694230149625,
+      "loss": 0.6497,
+      "step": 5360
+    },
+    {
+      "epoch": 0.6757565261202254,
+      "grad_norm": 0.386343389749527,
+      "learning_rate": 0.00028590045165080883,
+      "loss": 0.6344,
+      "step": 5365
+    },
+    {
+      "epoch": 0.6763863085304027,
+      "grad_norm": 0.4404468834400177,
+      "learning_rate": 0.0002858538880184746,
+      "loss": 0.7115,
+      "step": 5370
+    },
+    {
+      "epoch": 0.67701609094058,
+      "grad_norm": 0.35227730870246887,
+      "learning_rate": 0.00028580725142949925,
+      "loss": 0.702,
+      "step": 5375
+    },
+    {
+      "epoch": 0.6776458733507573,
+      "grad_norm": 0.38216719031333923,
+      "learning_rate": 0.00028576054190892775,
+      "loss": 0.6845,
+      "step": 5380
+    },
+    {
+      "epoch": 0.6782756557609346,
+      "grad_norm": 0.3602873682975769,
+      "learning_rate": 0.0002857137594818441,
+      "loss": 0.7156,
+      "step": 5385
+    },
+    {
+      "epoch": 0.6789054381711119,
+      "grad_norm": 0.38896870613098145,
+      "learning_rate": 0.00028566690417337166,
+      "loss": 0.7029,
+      "step": 5390
+    },
+    {
+      "epoch": 0.6795352205812891,
+      "grad_norm": 0.3434313237667084,
+      "learning_rate": 0.0002856199760086726,
+      "loss": 0.687,
+      "step": 5395
+    },
+    {
+      "epoch": 0.6801650029914664,
+      "grad_norm": 0.381331205368042,
+      "learning_rate": 0.0002855729750129487,
+      "loss": 0.6597,
+      "step": 5400
+    },
+    {
+      "epoch": 0.6807947854016437,
+      "grad_norm": 0.35004013776779175,
+      "learning_rate": 0.0002855259012114403,
+      "loss": 0.6604,
+      "step": 5405
+    },
+    {
+      "epoch": 0.681424567811821,
+      "grad_norm": 0.3601452112197876,
+      "learning_rate": 0.0002854787546294272,
+      "loss": 0.6949,
+      "step": 5410
+    },
+    {
+      "epoch": 0.6820543502219983,
+      "grad_norm": 0.3827126920223236,
+      "learning_rate": 0.0002854315352922282,
+      "loss": 0.7121,
+      "step": 5415
+    },
+    {
+      "epoch": 0.6826841326321755,
+      "grad_norm": 0.35859569907188416,
+      "learning_rate": 0.0002853842432252012,
+      "loss": 0.6662,
+      "step": 5420
+    },
+    {
+      "epoch": 0.6833139150423528,
+      "grad_norm": 0.36607855558395386,
+      "learning_rate": 0.00028533687845374304,
+      "loss": 0.6716,
+      "step": 5425
+    },
+    {
+      "epoch": 0.6839436974525301,
+      "grad_norm": 0.3658086061477661,
+      "learning_rate": 0.00028528944100328975,
+      "loss": 0.6718,
+      "step": 5430
+    },
+    {
+      "epoch": 0.6845734798627074,
+      "grad_norm": 0.3442821800708771,
+      "learning_rate": 0.00028524193089931633,
+      "loss": 0.6474,
+      "step": 5435
+    },
+    {
+      "epoch": 0.6852032622728847,
+      "grad_norm": 0.38460132479667664,
+      "learning_rate": 0.0002851943481673367,
+      "loss": 0.6973,
+      "step": 5440
+    },
+    {
+      "epoch": 0.685833044683062,
+      "grad_norm": 0.3717944622039795,
+      "learning_rate": 0.000285146692832904,
+      "loss": 0.6962,
+      "step": 5445
+    },
+    {
+      "epoch": 0.6864628270932392,
+      "grad_norm": 0.42136862874031067,
+      "learning_rate": 0.00028509896492161013,
+      "loss": 0.6783,
+      "step": 5450
+    },
+    {
+      "epoch": 0.6870926095034166,
+      "grad_norm": 0.37208443880081177,
+      "learning_rate": 0.0002850511644590862,
+      "loss": 0.6915,
+      "step": 5455
+    },
+    {
+      "epoch": 0.6877223919135939,
+      "grad_norm": 0.3807058036327362,
+      "learning_rate": 0.000285003291471002,
+      "loss": 0.7269,
+      "step": 5460
+    },
+    {
+      "epoch": 0.6883521743237712,
+      "grad_norm": 0.38431763648986816,
+      "learning_rate": 0.00028495534598306645,
+      "loss": 0.6589,
+      "step": 5465
+    },
+    {
+      "epoch": 0.6889819567339485,
+      "grad_norm": 0.372773140668869,
+      "learning_rate": 0.0002849073280210274,
+      "loss": 0.6922,
+      "step": 5470
+    },
+    {
+      "epoch": 0.6896117391441257,
+      "grad_norm": 0.3280029892921448,
+      "learning_rate": 0.00028485923761067164,
+      "loss": 0.6887,
+      "step": 5475
+    },
+    {
+      "epoch": 0.690241521554303,
+      "grad_norm": 0.3463418483734131,
+      "learning_rate": 0.0002848110747778247,
+      "loss": 0.6565,
+      "step": 5480
+    },
+    {
+      "epoch": 0.6908713039644803,
+      "grad_norm": 0.3423214256763458,
+      "learning_rate": 0.00028476283954835123,
+      "loss": 0.6412,
+      "step": 5485
+    },
+    {
+      "epoch": 0.6915010863746576,
+      "grad_norm": 0.3461606204509735,
+      "learning_rate": 0.0002847145319481546,
+      "loss": 0.6803,
+      "step": 5490
+    },
+    {
+      "epoch": 0.6921308687848349,
+      "grad_norm": 0.38746729493141174,
+      "learning_rate": 0.0002846661520031772,
+      "loss": 0.6424,
+      "step": 5495
+    },
+    {
+      "epoch": 0.6927606511950122,
+      "grad_norm": 0.32353097200393677,
+      "learning_rate": 0.00028461769973939997,
+      "loss": 0.6761,
+      "step": 5500
+    },
+    {
+      "epoch": 0.6933904336051894,
+      "grad_norm": 0.3790241777896881,
+      "learning_rate": 0.00028456917518284304,
+      "loss": 0.6683,
+      "step": 5505
+    },
+    {
+      "epoch": 0.6940202160153667,
+      "grad_norm": 0.3713475465774536,
+      "learning_rate": 0.0002845205783595651,
+      "loss": 0.6663,
+      "step": 5510
+    },
+    {
+      "epoch": 0.694649998425544,
+      "grad_norm": 0.3859196901321411,
+      "learning_rate": 0.00028447190929566384,
+      "loss": 0.6717,
+      "step": 5515
+    },
+    {
+      "epoch": 0.6952797808357213,
+      "grad_norm": 0.34451383352279663,
+      "learning_rate": 0.0002844231680172756,
+      "loss": 0.6368,
+      "step": 5520
+    },
+    {
+      "epoch": 0.6959095632458986,
+      "grad_norm": 0.3519328534603119,
+      "learning_rate": 0.00028437435455057564,
+      "loss": 0.6882,
+      "step": 5525
+    },
+    {
+      "epoch": 0.6965393456560758,
+      "grad_norm": 0.382755309343338,
+      "learning_rate": 0.0002843254689217778,
+      "loss": 0.6415,
+      "step": 5530
+    },
+    {
+      "epoch": 0.6971691280662531,
+      "grad_norm": 0.35310298204421997,
+      "learning_rate": 0.0002842765111571349,
+      "loss": 0.6744,
+      "step": 5535
+    },
+    {
+      "epoch": 0.6977989104764304,
+      "grad_norm": 0.3392702341079712,
+      "learning_rate": 0.0002842274812829382,
+      "loss": 0.6705,
+      "step": 5540
+    },
+    {
+      "epoch": 0.6984286928866077,
+      "grad_norm": 0.36502036452293396,
+      "learning_rate": 0.00028417837932551805,
+      "loss": 0.6777,
+      "step": 5545
+    },
+    {
+      "epoch": 0.699058475296785,
+      "grad_norm": 0.36270782351493835,
+      "learning_rate": 0.0002841292053112432,
+      "loss": 0.6988,
+      "step": 5550
+    },
+    {
+      "epoch": 0.6996882577069622,
+      "grad_norm": 0.3752531111240387,
+      "learning_rate": 0.0002840799592665213,
+      "loss": 0.6745,
+      "step": 5555
+    },
+    {
+      "epoch": 0.7003180401171395,
+      "grad_norm": 0.32373905181884766,
+      "learning_rate": 0.00028403064121779853,
+      "loss": 0.664,
+      "step": 5560
+    },
+    {
+      "epoch": 0.7009478225273168,
+      "grad_norm": 0.4017639756202698,
+      "learning_rate": 0.0002839812511915599,
+      "loss": 0.6793,
+      "step": 5565
+    },
+    {
+      "epoch": 0.7015776049374941,
+      "grad_norm": 0.33867186307907104,
+      "learning_rate": 0.00028393178921432883,
+      "loss": 0.6811,
+      "step": 5570
+    },
+    {
+      "epoch": 0.7022073873476714,
+      "grad_norm": 0.3769174814224243,
+      "learning_rate": 0.0002838822553126677,
+      "loss": 0.7118,
+      "step": 5575
+    },
+    {
+      "epoch": 0.7028371697578487,
+      "grad_norm": 0.36820533871650696,
+      "learning_rate": 0.00028383264951317727,
+      "loss": 0.6581,
+      "step": 5580
+    },
+    {
+      "epoch": 0.7034669521680259,
+      "grad_norm": 0.37128061056137085,
+      "learning_rate": 0.00028378297184249694,
+      "loss": 0.6722,
+      "step": 5585
+    },
+    {
+      "epoch": 0.7040967345782032,
+      "grad_norm": 0.39225873351097107,
+      "learning_rate": 0.00028373322232730483,
+      "loss": 0.6846,
+      "step": 5590
+    },
+    {
+      "epoch": 0.7047265169883805,
+      "grad_norm": 0.3394504189491272,
+      "learning_rate": 0.0002836834009943175,
+      "loss": 0.6815,
+      "step": 5595
+    },
+    {
+      "epoch": 0.7053562993985578,
+      "grad_norm": 0.37265124917030334,
+      "learning_rate": 0.0002836335078702903,
+      "loss": 0.6614,
+      "step": 5600
+    },
+    {
+      "epoch": 0.7059860818087351,
+      "grad_norm": 0.33066150546073914,
+      "learning_rate": 0.00028358354298201673,
+      "loss": 0.6701,
+      "step": 5605
+    },
+    {
+      "epoch": 0.7066158642189123,
+      "grad_norm": 0.35536128282546997,
+      "learning_rate": 0.0002835335063563293,
+      "loss": 0.6149,
+      "step": 5610
+    },
+    {
+      "epoch": 0.7072456466290896,
+      "grad_norm": 0.35491225123405457,
+      "learning_rate": 0.0002834833980200987,
+      "loss": 0.6773,
+      "step": 5615
+    },
+    {
+      "epoch": 0.7078754290392669,
+      "grad_norm": 0.37837696075439453,
+      "learning_rate": 0.0002834332180002343,
+      "loss": 0.6899,
+      "step": 5620
+    },
+    {
+      "epoch": 0.7085052114494442,
+      "grad_norm": 0.3391937017440796,
+      "learning_rate": 0.0002833829663236838,
+      "loss": 0.7041,
+      "step": 5625
+    },
+    {
+      "epoch": 0.7091349938596215,
+      "grad_norm": 0.3482423424720764,
+      "learning_rate": 0.00028333264301743375,
+      "loss": 0.6597,
+      "step": 5630
+    },
+    {
+      "epoch": 0.7097647762697988,
+      "grad_norm": 0.4188586175441742,
+      "learning_rate": 0.00028328224810850866,
+      "loss": 0.6916,
+      "step": 5635
+    },
+    {
+      "epoch": 0.710394558679976,
+      "grad_norm": 0.32832324504852295,
+      "learning_rate": 0.0002832317816239718,
+      "loss": 0.6791,
+      "step": 5640
+    },
+    {
+      "epoch": 0.7110243410901533,
+      "grad_norm": 0.343058705329895,
+      "learning_rate": 0.00028318124359092496,
+      "loss": 0.6423,
+      "step": 5645
+    },
+    {
+      "epoch": 0.7116541235003306,
+      "grad_norm": 0.37011584639549255,
+      "learning_rate": 0.0002831306340365081,
+      "loss": 0.6783,
+      "step": 5650
+    },
+    {
+      "epoch": 0.7122839059105079,
+      "grad_norm": 0.38297170400619507,
+      "learning_rate": 0.00028307995298789974,
+      "loss": 0.6751,
+      "step": 5655
+    },
+    {
+      "epoch": 0.7129136883206852,
+      "grad_norm": 0.38705122470855713,
+      "learning_rate": 0.00028302920047231677,
+      "loss": 0.6844,
+      "step": 5660
+    },
+    {
+      "epoch": 0.7135434707308624,
+      "grad_norm": 0.3647492527961731,
+      "learning_rate": 0.0002829783765170144,
+      "loss": 0.6811,
+      "step": 5665
+    },
+    {
+      "epoch": 0.7141732531410397,
+      "grad_norm": 0.3796983063220978,
+      "learning_rate": 0.0002829274811492863,
+      "loss": 0.6766,
+      "step": 5670
+    },
+    {
+      "epoch": 0.714803035551217,
+      "grad_norm": 0.36972787976264954,
+      "learning_rate": 0.00028287651439646444,
+      "loss": 0.6701,
+      "step": 5675
+    },
+    {
+      "epoch": 0.7154328179613944,
+      "grad_norm": 0.37298983335494995,
+      "learning_rate": 0.0002828254762859192,
+      "loss": 0.6439,
+      "step": 5680
+    },
+    {
+      "epoch": 0.7160626003715717,
+      "grad_norm": 0.3464621603488922,
+      "learning_rate": 0.0002827743668450591,
+      "loss": 0.6626,
+      "step": 5685
+    },
+    {
+      "epoch": 0.716692382781749,
+      "grad_norm": 0.34213629364967346,
+      "learning_rate": 0.00028272318610133104,
+      "loss": 0.6987,
+      "step": 5690
+    },
+    {
+      "epoch": 0.7173221651919262,
+      "grad_norm": 0.38596463203430176,
+      "learning_rate": 0.0002826719340822204,
+      "loss": 0.6846,
+      "step": 5695
+    },
+    {
+      "epoch": 0.7179519476021035,
+      "grad_norm": 0.3410765826702118,
+      "learning_rate": 0.0002826206108152506,
+      "loss": 0.6769,
+      "step": 5700
+    },
+    {
+      "epoch": 0.7185817300122808,
+      "grad_norm": 0.3370499610900879,
+      "learning_rate": 0.0002825692163279834,
+      "loss": 0.6563,
+      "step": 5705
+    },
+    {
+      "epoch": 0.7192115124224581,
+      "grad_norm": 0.3973693549633026,
+      "learning_rate": 0.0002825177506480189,
+      "loss": 0.6587,
+      "step": 5710
+    },
+    {
+      "epoch": 0.7198412948326354,
+      "grad_norm": 0.3341182470321655,
+      "learning_rate": 0.0002824662138029952,
+      "loss": 0.6489,
+      "step": 5715
+    },
+    {
+      "epoch": 0.7204710772428126,
+      "grad_norm": 0.3598056733608246,
+      "learning_rate": 0.00028241460582058883,
+      "loss": 0.6623,
+      "step": 5720
+    },
+    {
+      "epoch": 0.7211008596529899,
+      "grad_norm": 0.34275728464126587,
+      "learning_rate": 0.00028236292672851443,
+      "loss": 0.6987,
+      "step": 5725
+    },
+    {
+      "epoch": 0.7217306420631672,
+      "grad_norm": 0.3606712222099304,
+      "learning_rate": 0.000282311176554525,
+      "loss": 0.6947,
+      "step": 5730
+    },
+    {
+      "epoch": 0.7223604244733445,
+      "grad_norm": 0.32409214973449707,
+      "learning_rate": 0.0002822593553264114,
+      "loss": 0.6468,
+      "step": 5735
+    },
+    {
+      "epoch": 0.7229902068835218,
+      "grad_norm": 0.3465891182422638,
+      "learning_rate": 0.00028220746307200287,
+      "loss": 0.647,
+      "step": 5740
+    },
+    {
+      "epoch": 0.723619989293699,
+      "grad_norm": 0.3540678918361664,
+      "learning_rate": 0.0002821554998191667,
+      "loss": 0.6964,
+      "step": 5745
+    },
+    {
+      "epoch": 0.7242497717038763,
+      "grad_norm": 0.35845157504081726,
+      "learning_rate": 0.0002821034655958084,
+      "loss": 0.6599,
+      "step": 5750
+    },
+    {
+      "epoch": 0.7248795541140536,
+      "grad_norm": 0.3469247817993164,
+      "learning_rate": 0.00028205136042987156,
+      "loss": 0.6518,
+      "step": 5755
+    },
+    {
+      "epoch": 0.7255093365242309,
+      "grad_norm": 0.3693814277648926,
+      "learning_rate": 0.0002819991843493377,
+      "loss": 0.6339,
+      "step": 5760
+    },
+    {
+      "epoch": 0.7261391189344082,
+      "grad_norm": 0.35166436433792114,
+      "learning_rate": 0.0002819469373822268,
+      "loss": 0.6593,
+      "step": 5765
+    },
+    {
+      "epoch": 0.7267689013445855,
+      "grad_norm": 0.376717746257782,
+      "learning_rate": 0.00028189461955659644,
+      "loss": 0.6583,
+      "step": 5770
+    },
+    {
+      "epoch": 0.7273986837547627,
+      "grad_norm": 0.36365002393722534,
+      "learning_rate": 0.0002818422309005426,
+      "loss": 0.707,
+      "step": 5775
+    },
+    {
+      "epoch": 0.72802846616494,
+      "grad_norm": 0.3356451392173767,
+      "learning_rate": 0.00028178977144219914,
+      "loss": 0.6439,
+      "step": 5780
+    },
+    {
+      "epoch": 0.7286582485751173,
+      "grad_norm": 0.33520832657814026,
+      "learning_rate": 0.00028173724120973806,
+      "loss": 0.6276,
+      "step": 5785
+    },
+    {
+      "epoch": 0.7292880309852946,
+      "grad_norm": 0.3459213376045227,
+      "learning_rate": 0.00028168464023136926,
+      "loss": 0.648,
+      "step": 5790
+    },
+    {
+      "epoch": 0.7299178133954719,
+      "grad_norm": 0.3563973903656006,
+      "learning_rate": 0.0002816319685353406,
+      "loss": 0.6579,
+      "step": 5795
+    },
+    {
+      "epoch": 0.7305475958056491,
+      "grad_norm": 0.3637474775314331,
+      "learning_rate": 0.0002815792261499381,
+      "loss": 0.6828,
+      "step": 5800
+    },
+    {
+      "epoch": 0.7311773782158264,
+      "grad_norm": 0.38304394483566284,
+      "learning_rate": 0.00028152641310348554,
+      "loss": 0.6348,
+      "step": 5805
+    },
+    {
+      "epoch": 0.7318071606260037,
+      "grad_norm": 0.33336034417152405,
+      "learning_rate": 0.0002814735294243448,
+      "loss": 0.6337,
+      "step": 5810
+    },
+    {
+      "epoch": 0.732436943036181,
+      "grad_norm": 0.34154805541038513,
+      "learning_rate": 0.0002814205751409156,
+      "loss": 0.6885,
+      "step": 5815
+    },
+    {
+      "epoch": 0.7330667254463583,
+      "grad_norm": 0.3780697286128998,
+      "learning_rate": 0.00028136755028163556,
+      "loss": 0.6558,
+      "step": 5820
+    },
+    {
+      "epoch": 0.7336965078565356,
+      "grad_norm": 0.3496229946613312,
+      "learning_rate": 0.0002813144548749802,
+      "loss": 0.7058,
+      "step": 5825
+    },
+    {
+      "epoch": 0.7343262902667128,
+      "grad_norm": 0.36560389399528503,
+      "learning_rate": 0.0002812612889494631,
+      "loss": 0.6991,
+      "step": 5830
+    },
+    {
+      "epoch": 0.7349560726768901,
+      "grad_norm": 0.3215349316596985,
+      "learning_rate": 0.00028120805253363545,
+      "loss": 0.612,
+      "step": 5835
+    },
+    {
+      "epoch": 0.7355858550870674,
+      "grad_norm": 0.36016130447387695,
+      "learning_rate": 0.00028115474565608656,
+      "loss": 0.6905,
+      "step": 5840
+    },
+    {
+      "epoch": 0.7362156374972447,
+      "grad_norm": 0.3493592441082001,
+      "learning_rate": 0.00028110136834544336,
+      "loss": 0.6922,
+      "step": 5845
+    },
+    {
+      "epoch": 0.736845419907422,
+      "grad_norm": 0.34350746870040894,
+      "learning_rate": 0.00028104792063037064,
+      "loss": 0.6238,
+      "step": 5850
+    },
+    {
+      "epoch": 0.7374752023175992,
+      "grad_norm": 0.3633589446544647,
+      "learning_rate": 0.0002809944025395711,
+      "loss": 0.6775,
+      "step": 5855
+    },
+    {
+      "epoch": 0.7381049847277765,
+      "grad_norm": 0.3892457187175751,
+      "learning_rate": 0.00028094081410178515,
+      "loss": 0.6756,
+      "step": 5860
+    },
+    {
+      "epoch": 0.7387347671379538,
+      "grad_norm": 0.33569657802581787,
+      "learning_rate": 0.00028088715534579104,
+      "loss": 0.63,
+      "step": 5865
+    },
+    {
+      "epoch": 0.7393645495481311,
+      "grad_norm": 0.36327067017555237,
+      "learning_rate": 0.0002808334263004047,
+      "loss": 0.6653,
+      "step": 5870
+    },
+    {
+      "epoch": 0.7399943319583084,
+      "grad_norm": 0.32698652148246765,
+      "learning_rate": 0.00028077962699448,
+      "loss": 0.655,
+      "step": 5875
+    },
+    {
+      "epoch": 0.7406241143684857,
+      "grad_norm": 0.35473042726516724,
+      "learning_rate": 0.0002807257574569082,
+      "loss": 0.6341,
+      "step": 5880
+    },
+    {
+      "epoch": 0.7412538967786629,
+      "grad_norm": 0.33008939027786255,
+      "learning_rate": 0.0002806718177166185,
+      "loss": 0.6614,
+      "step": 5885
+    },
+    {
+      "epoch": 0.7418836791888402,
+      "grad_norm": 0.3434574007987976,
+      "learning_rate": 0.0002806178078025779,
+      "loss": 0.6313,
+      "step": 5890
+    },
+    {
+      "epoch": 0.7425134615990175,
+      "grad_norm": 0.30766573548316956,
+      "learning_rate": 0.00028056372774379085,
+      "loss": 0.6296,
+      "step": 5895
+    },
+    {
+      "epoch": 0.7431432440091948,
+      "grad_norm": 0.3676775097846985,
+      "learning_rate": 0.00028050957756929965,
+      "loss": 0.628,
+      "step": 5900
+    },
+    {
+      "epoch": 0.7437730264193722,
+      "grad_norm": 0.3424786925315857,
+      "learning_rate": 0.0002804553573081841,
+      "loss": 0.6141,
+      "step": 5905
+    },
+    {
+      "epoch": 0.7444028088295493,
+      "grad_norm": 0.391250878572464,
+      "learning_rate": 0.0002804010669895618,
+      "loss": 0.6615,
+      "step": 5910
+    },
+    {
+      "epoch": 0.7450325912397266,
+      "grad_norm": 0.34186193346977234,
+      "learning_rate": 0.0002803467066425878,
+      "loss": 0.6389,
+      "step": 5915
+    },
+    {
+      "epoch": 0.745662373649904,
+      "grad_norm": 0.37509649991989136,
+      "learning_rate": 0.0002802922762964549,
+      "loss": 0.6397,
+      "step": 5920
+    },
+    {
+      "epoch": 0.7462921560600813,
+      "grad_norm": 0.3327299654483795,
+      "learning_rate": 0.00028023777598039346,
+      "loss": 0.6241,
+      "step": 5925
+    },
+    {
+      "epoch": 0.7469219384702586,
+      "grad_norm": 0.37098389863967896,
+      "learning_rate": 0.0002801832057236714,
+      "loss": 0.7004,
+      "step": 5930
+    },
+    {
+      "epoch": 0.7475517208804358,
+      "grad_norm": 0.36630627512931824,
+      "learning_rate": 0.00028012856555559415,
+      "loss": 0.6201,
+      "step": 5935
+    },
+    {
+      "epoch": 0.7481815032906131,
+      "grad_norm": 0.3580261170864105,
+      "learning_rate": 0.00028007385550550475,
+      "loss": 0.6969,
+      "step": 5940
+    },
+    {
+      "epoch": 0.7488112857007904,
+      "grad_norm": 0.3491668105125427,
+      "learning_rate": 0.0002800190756027837,
+      "loss": 0.6457,
+      "step": 5945
+    },
+    {
+      "epoch": 0.7494410681109677,
+      "grad_norm": 0.2999480664730072,
+      "learning_rate": 0.0002799642258768491,
+      "loss": 0.6398,
+      "step": 5950
+    },
+    {
+      "epoch": 0.750070850521145,
+      "grad_norm": 0.33795973658561707,
+      "learning_rate": 0.00027990930635715655,
+      "loss": 0.6672,
+      "step": 5955
+    },
+    {
+      "epoch": 0.7507006329313223,
+      "grad_norm": 0.39881202578544617,
+      "learning_rate": 0.00027985431707319903,
+      "loss": 0.6796,
+      "step": 5960
+    },
+    {
+      "epoch": 0.7513304153414995,
+      "grad_norm": 0.4092641770839691,
+      "learning_rate": 0.0002797992580545071,
+      "loss": 0.6488,
+      "step": 5965
+    },
+    {
+      "epoch": 0.7519601977516768,
+      "grad_norm": 0.33037346601486206,
+      "learning_rate": 0.0002797441293306486,
+      "loss": 0.667,
+      "step": 5970
+    },
+    {
+      "epoch": 0.7525899801618541,
+      "grad_norm": 0.35514095425605774,
+      "learning_rate": 0.00027968893093122896,
+      "loss": 0.6984,
+      "step": 5975
+    },
+    {
+      "epoch": 0.7532197625720314,
+      "grad_norm": 0.4268254339694977,
+      "learning_rate": 0.0002796336628858911,
+      "loss": 0.6762,
+      "step": 5980
+    },
+    {
+      "epoch": 0.7538495449822087,
+      "grad_norm": 0.33386656641960144,
+      "learning_rate": 0.00027957832522431503,
+      "loss": 0.6438,
+      "step": 5985
+    },
+    {
+      "epoch": 0.7544793273923859,
+      "grad_norm": 0.374845415353775,
+      "learning_rate": 0.00027952291797621846,
+      "loss": 0.6422,
+      "step": 5990
+    },
+    {
+      "epoch": 0.7551091098025632,
+      "grad_norm": 0.32742077112197876,
+      "learning_rate": 0.0002794674411713563,
+      "loss": 0.6685,
+      "step": 5995
+    },
+    {
+      "epoch": 0.7557388922127405,
+      "grad_norm": 0.3118845820426941,
+      "learning_rate": 0.0002794118948395209,
+      "loss": 0.6273,
+      "step": 6000
+    },
+    {
+      "epoch": 0.7557388922127405,
+      "eval_loss": 0.3097546696662903,
+      "eval_runtime": 6.2567,
+      "eval_samples_per_second": 159.828,
+      "eval_steps_per_second": 10.069,
+      "step": 6000
+    },
+    {
+      "epoch": 0.7563686746229178,
+      "grad_norm": 0.3407754898071289,
+      "learning_rate": 0.00027935627901054197,
+      "loss": 0.6712,
+      "step": 6005
+    },
+    {
+      "epoch": 0.7569984570330951,
+      "grad_norm": 0.34817007184028625,
+      "learning_rate": 0.0002793005937142863,
+      "loss": 0.6492,
+      "step": 6010
+    },
+    {
+      "epoch": 0.7576282394432724,
+      "grad_norm": 0.36492645740509033,
+      "learning_rate": 0.00027924483898065833,
+      "loss": 0.6467,
+      "step": 6015
+    },
+    {
+      "epoch": 0.7582580218534496,
+      "grad_norm": 0.33556580543518066,
+      "learning_rate": 0.0002791890148395995,
+      "loss": 0.6486,
+      "step": 6020
+    },
+    {
+      "epoch": 0.7588878042636269,
+      "grad_norm": 0.36699965596199036,
+      "learning_rate": 0.00027913312132108874,
+      "loss": 0.6909,
+      "step": 6025
+    },
+    {
+      "epoch": 0.7595175866738042,
+      "grad_norm": 0.32526010274887085,
+      "learning_rate": 0.0002790771584551421,
+      "loss": 0.6234,
+      "step": 6030
+    },
+    {
+      "epoch": 0.7601473690839815,
+      "grad_norm": 0.38366591930389404,
+      "learning_rate": 0.00027902112627181295,
+      "loss": 0.6195,
+      "step": 6035
+    },
+    {
+      "epoch": 0.7607771514941588,
+      "grad_norm": 0.33587443828582764,
+      "learning_rate": 0.0002789650248011918,
+      "loss": 0.6546,
+      "step": 6040
+    },
+    {
+      "epoch": 0.761406933904336,
+      "grad_norm": 0.36170026659965515,
+      "learning_rate": 0.00027890885407340653,
+      "loss": 0.6294,
+      "step": 6045
+    },
+    {
+      "epoch": 0.7620367163145133,
+      "grad_norm": 0.34692490100860596,
+      "learning_rate": 0.000278852614118622,
+      "loss": 0.6468,
+      "step": 6050
+    },
+    {
+      "epoch": 0.7626664987246906,
+      "grad_norm": 0.346608966588974,
+      "learning_rate": 0.0002787963049670404,
+      "loss": 0.6714,
+      "step": 6055
+    },
+    {
+      "epoch": 0.7632962811348679,
+      "grad_norm": 0.3632940948009491,
+      "learning_rate": 0.00027873992664890097,
+      "loss": 0.6772,
+      "step": 6060
+    },
+    {
+      "epoch": 0.7639260635450452,
+      "grad_norm": 0.38135001063346863,
+      "learning_rate": 0.00027868347919448027,
+      "loss": 0.658,
+      "step": 6065
+    },
+    {
+      "epoch": 0.7645558459552225,
+      "grad_norm": 0.3518752455711365,
+      "learning_rate": 0.00027862696263409177,
+      "loss": 0.6445,
+      "step": 6070
+    },
+    {
+      "epoch": 0.7651856283653997,
+      "grad_norm": 0.33004361391067505,
+      "learning_rate": 0.00027857037699808613,
+      "loss": 0.6553,
+      "step": 6075
+    },
+    {
+      "epoch": 0.765815410775577,
+      "grad_norm": 0.36370858550071716,
+      "learning_rate": 0.0002785137223168512,
+      "loss": 0.6632,
+      "step": 6080
+    },
+    {
+      "epoch": 0.7664451931857543,
+      "grad_norm": 0.3472859561443329,
+      "learning_rate": 0.0002784569986208119,
+      "loss": 0.626,
+      "step": 6085
+    },
+    {
+      "epoch": 0.7670749755959316,
+      "grad_norm": 0.3560635447502136,
+      "learning_rate": 0.00027840020594043,
+      "loss": 0.6628,
+      "step": 6090
+    },
+    {
+      "epoch": 0.7677047580061089,
+      "grad_norm": 0.3515082895755768,
+      "learning_rate": 0.00027834334430620455,
+      "loss": 0.7061,
+      "step": 6095
+    },
+    {
+      "epoch": 0.7683345404162861,
+      "grad_norm": 0.3222733736038208,
+      "learning_rate": 0.00027828641374867154,
+      "loss": 0.617,
+      "step": 6100
+    },
+    {
+      "epoch": 0.7689643228264634,
+      "grad_norm": 0.3362828493118286,
+      "learning_rate": 0.00027822941429840397,
+      "loss": 0.6825,
+      "step": 6105
+    },
+    {
+      "epoch": 0.7695941052366407,
+      "grad_norm": 0.34228187799453735,
+      "learning_rate": 0.0002781723459860119,
+      "loss": 0.6306,
+      "step": 6110
+    },
+    {
+      "epoch": 0.770223887646818,
+      "grad_norm": 0.3672444820404053,
+      "learning_rate": 0.0002781152088421422,
+      "loss": 0.6601,
+      "step": 6115
+    },
+    {
+      "epoch": 0.7708536700569953,
+      "grad_norm": 0.3703080415725708,
+      "learning_rate": 0.00027805800289747894,
+      "loss": 0.6385,
+      "step": 6120
+    },
+    {
+      "epoch": 0.7714834524671725,
+      "grad_norm": 0.34456151723861694,
+      "learning_rate": 0.0002780007281827429,
+      "loss": 0.6635,
+      "step": 6125
+    },
+    {
+      "epoch": 0.7721132348773498,
+      "grad_norm": 0.3449029326438904,
+      "learning_rate": 0.00027794338472869205,
+      "loss": 0.6258,
+      "step": 6130
+    },
+    {
+      "epoch": 0.7727430172875271,
+      "grad_norm": 0.3441922068595886,
+      "learning_rate": 0.0002778859725661211,
+      "loss": 0.627,
+      "step": 6135
+    },
+    {
+      "epoch": 0.7733727996977044,
+      "grad_norm": 0.3855600357055664,
+      "learning_rate": 0.00027782849172586156,
+      "loss": 0.6205,
+      "step": 6140
+    },
+    {
+      "epoch": 0.7740025821078818,
+      "grad_norm": 0.3838488757610321,
+      "learning_rate": 0.0002777709422387821,
+      "loss": 0.6463,
+      "step": 6145
+    },
+    {
+      "epoch": 0.7746323645180591,
+      "grad_norm": 0.3128564953804016,
+      "learning_rate": 0.00027771332413578805,
+      "loss": 0.6639,
+      "step": 6150
+    },
+    {
+      "epoch": 0.7752621469282363,
+      "grad_norm": 0.32142025232315063,
+      "learning_rate": 0.00027765563744782166,
+      "loss": 0.6187,
+      "step": 6155
+    },
+    {
+      "epoch": 0.7758919293384136,
+      "grad_norm": 0.34378373622894287,
+      "learning_rate": 0.000277597882205862,
+      "loss": 0.659,
+      "step": 6160
+    },
+    {
+      "epoch": 0.7765217117485909,
+      "grad_norm": 0.35872867703437805,
+      "learning_rate": 0.0002775400584409249,
+      "loss": 0.6245,
+      "step": 6165
+    },
+    {
+      "epoch": 0.7771514941587682,
+      "grad_norm": 0.32217180728912354,
+      "learning_rate": 0.00027748216618406316,
+      "loss": 0.6216,
+      "step": 6170
+    },
+    {
+      "epoch": 0.7777812765689455,
+      "grad_norm": 0.3139524757862091,
+      "learning_rate": 0.00027742420546636616,
+      "loss": 0.6831,
+      "step": 6175
+    },
+    {
+      "epoch": 0.7784110589791227,
+      "grad_norm": 0.3159128427505493,
+      "learning_rate": 0.00027736617631896017,
+      "loss": 0.6417,
+      "step": 6180
+    },
+    {
+      "epoch": 0.7790408413893,
+      "grad_norm": 0.36738142371177673,
+      "learning_rate": 0.0002773080787730081,
+      "loss": 0.6592,
+      "step": 6185
+    },
+    {
+      "epoch": 0.7796706237994773,
+      "grad_norm": 0.31971079111099243,
+      "learning_rate": 0.0002772499128597097,
+      "loss": 0.6296,
+      "step": 6190
+    },
+    {
+      "epoch": 0.7803004062096546,
+      "grad_norm": 0.3699764609336853,
+      "learning_rate": 0.00027719167861030145,
+      "loss": 0.6161,
+      "step": 6195
+    },
+    {
+      "epoch": 0.7809301886198319,
+      "grad_norm": 0.3316752016544342,
+      "learning_rate": 0.0002771333760560564,
+      "loss": 0.6698,
+      "step": 6200
+    },
+    {
+      "epoch": 0.7815599710300092,
+      "grad_norm": 0.34318891167640686,
+      "learning_rate": 0.00027707500522828433,
+      "loss": 0.6312,
+      "step": 6205
+    },
+    {
+      "epoch": 0.7821897534401864,
+      "grad_norm": 0.3325194716453552,
+      "learning_rate": 0.00027701656615833185,
+      "loss": 0.6515,
+      "step": 6210
+    },
+    {
+      "epoch": 0.7828195358503637,
+      "grad_norm": 0.3374411463737488,
+      "learning_rate": 0.0002769580588775819,
+      "loss": 0.6811,
+      "step": 6215
+    },
+    {
+      "epoch": 0.783449318260541,
+      "grad_norm": 0.3507198989391327,
+      "learning_rate": 0.00027689948341745433,
+      "loss": 0.6177,
+      "step": 6220
+    },
+    {
+      "epoch": 0.7840791006707183,
+      "grad_norm": 0.3619876205921173,
+      "learning_rate": 0.00027684083980940543,
+      "loss": 0.6812,
+      "step": 6225
+    },
+    {
+      "epoch": 0.7847088830808956,
+      "grad_norm": 0.3660729229450226,
+      "learning_rate": 0.00027678212808492824,
+      "loss": 0.6888,
+      "step": 6230
+    },
+    {
+      "epoch": 0.7853386654910728,
+      "grad_norm": 0.37557917833328247,
+      "learning_rate": 0.00027672334827555226,
+      "loss": 0.6516,
+      "step": 6235
+    },
+    {
+      "epoch": 0.7859684479012501,
+      "grad_norm": 0.37117084860801697,
+      "learning_rate": 0.00027666450041284363,
+      "loss": 0.6503,
+      "step": 6240
+    },
+    {
+      "epoch": 0.7865982303114274,
+      "grad_norm": 0.3434617519378662,
+      "learning_rate": 0.00027660558452840487,
+      "loss": 0.6582,
+      "step": 6245
+    },
+    {
+      "epoch": 0.7872280127216047,
+      "grad_norm": 0.3878399431705475,
+      "learning_rate": 0.0002765466006538753,
+      "loss": 0.6309,
+      "step": 6250
+    },
+    {
+      "epoch": 0.787857795131782,
+      "grad_norm": 0.3379189968109131,
+      "learning_rate": 0.0002764875488209305,
+      "loss": 0.6802,
+      "step": 6255
+    },
+    {
+      "epoch": 0.7884875775419593,
+      "grad_norm": 0.3534158170223236,
+      "learning_rate": 0.0002764284290612827,
+      "loss": 0.6248,
+      "step": 6260
+    },
+    {
+      "epoch": 0.7891173599521365,
+      "grad_norm": 0.3273150324821472,
+      "learning_rate": 0.0002763692414066806,
+      "loss": 0.617,
+      "step": 6265
+    },
+    {
+      "epoch": 0.7897471423623138,
+      "grad_norm": 0.4256115257740021,
+      "learning_rate": 0.0002763099858889093,
+      "loss": 0.6452,
+      "step": 6270
+    },
+    {
+      "epoch": 0.7903769247724911,
+      "grad_norm": 0.34881314635276794,
+      "learning_rate": 0.0002762506625397903,
+      "loss": 0.6545,
+      "step": 6275
+    },
+    {
+      "epoch": 0.7910067071826684,
+      "grad_norm": 0.3283347487449646,
+      "learning_rate": 0.0002761912713911817,
+      "loss": 0.6819,
+      "step": 6280
+    },
+    {
+      "epoch": 0.7916364895928457,
+      "grad_norm": 0.33939605951309204,
+      "learning_rate": 0.0002761318124749778,
+      "loss": 0.6188,
+      "step": 6285
+    },
+    {
+      "epoch": 0.7922662720030229,
+      "grad_norm": 0.3786788582801819,
+      "learning_rate": 0.00027607228582310947,
+      "loss": 0.6583,
+      "step": 6290
+    },
+    {
+      "epoch": 0.7928960544132002,
+      "grad_norm": 0.34528714418411255,
+      "learning_rate": 0.0002760126914675439,
+      "loss": 0.6594,
+      "step": 6295
+    },
+    {
+      "epoch": 0.7935258368233775,
+      "grad_norm": 0.3494967818260193,
+      "learning_rate": 0.00027595302944028447,
+      "loss": 0.6241,
+      "step": 6300
+    },
+    {
+      "epoch": 0.7941556192335548,
+      "grad_norm": 0.350005179643631,
+      "learning_rate": 0.00027589329977337126,
+      "loss": 0.6724,
+      "step": 6305
+    },
+    {
+      "epoch": 0.7947854016437321,
+      "grad_norm": 0.3381168246269226,
+      "learning_rate": 0.0002758335024988803,
+      "loss": 0.6062,
+      "step": 6310
+    },
+    {
+      "epoch": 0.7954151840539094,
+      "grad_norm": 0.32583653926849365,
+      "learning_rate": 0.0002757736376489242,
+      "loss": 0.6602,
+      "step": 6315
+    },
+    {
+      "epoch": 0.7960449664640866,
+      "grad_norm": 0.33687326312065125,
+      "learning_rate": 0.0002757137052556517,
+      "loss": 0.6391,
+      "step": 6320
+    },
+    {
+      "epoch": 0.7966747488742639,
+      "grad_norm": 0.35395026206970215,
+      "learning_rate": 0.00027565370535124784,
+      "loss": 0.6445,
+      "step": 6325
+    },
+    {
+      "epoch": 0.7973045312844412,
+      "grad_norm": 0.3484829068183899,
+      "learning_rate": 0.000275593637967934,
+      "loss": 0.6242,
+      "step": 6330
+    },
+    {
+      "epoch": 0.7979343136946185,
+      "grad_norm": 0.32783517241477966,
+      "learning_rate": 0.0002755335031379677,
+      "loss": 0.6481,
+      "step": 6335
+    },
+    {
+      "epoch": 0.7985640961047958,
+      "grad_norm": 0.3683319389820099,
+      "learning_rate": 0.0002754733008936427,
+      "loss": 0.6506,
+      "step": 6340
+    },
+    {
+      "epoch": 0.799193878514973,
+      "grad_norm": 0.360219269990921,
+      "learning_rate": 0.00027541303126728907,
+      "loss": 0.6377,
+      "step": 6345
+    },
+    {
+      "epoch": 0.7998236609251503,
+      "grad_norm": 0.3323548436164856,
+      "learning_rate": 0.00027535269429127283,
+      "loss": 0.6278,
+      "step": 6350
+    },
+    {
+      "epoch": 0.8004534433353276,
+      "grad_norm": 0.33823835849761963,
+      "learning_rate": 0.0002752922899979965,
+      "loss": 0.5999,
+      "step": 6355
+    },
+    {
+      "epoch": 0.801083225745505,
+      "grad_norm": 0.35394924879074097,
+      "learning_rate": 0.0002752318184198984,
+      "loss": 0.6873,
+      "step": 6360
+    },
+    {
+      "epoch": 0.8017130081556822,
+      "grad_norm": 0.35529881715774536,
+      "learning_rate": 0.00027517127958945315,
+      "loss": 0.6183,
+      "step": 6365
+    },
+    {
+      "epoch": 0.8023427905658594,
+      "grad_norm": 0.35854044556617737,
+      "learning_rate": 0.00027511067353917166,
+      "loss": 0.6394,
+      "step": 6370
+    },
+    {
+      "epoch": 0.8029725729760367,
+      "grad_norm": 0.32757097482681274,
+      "learning_rate": 0.0002750500003016006,
+      "loss": 0.6383,
+      "step": 6375
+    },
+    {
+      "epoch": 0.803602355386214,
+      "grad_norm": 0.3267909586429596,
+      "learning_rate": 0.0002749892599093229,
+      "loss": 0.5951,
+      "step": 6380
+    },
+    {
+      "epoch": 0.8042321377963914,
+      "grad_norm": 0.31262004375457764,
+      "learning_rate": 0.0002749284523949576,
+      "loss": 0.6497,
+      "step": 6385
+    },
+    {
+      "epoch": 0.8048619202065687,
+      "grad_norm": 0.34036824107170105,
+      "learning_rate": 0.00027486757779115973,
+      "loss": 0.6295,
+      "step": 6390
+    },
+    {
+      "epoch": 0.805491702616746,
+      "grad_norm": 0.3461470901966095,
+      "learning_rate": 0.0002748066361306203,
+      "loss": 0.6537,
+      "step": 6395
+    },
+    {
+      "epoch": 0.8061214850269232,
+      "grad_norm": 0.35146886110305786,
+      "learning_rate": 0.00027474562744606636,
+      "loss": 0.6217,
+      "step": 6400
+    },
+    {
+      "epoch": 0.8067512674371005,
+      "grad_norm": 0.37654054164886475,
+      "learning_rate": 0.000274684551770261,
+      "loss": 0.6417,
+      "step": 6405
+    },
+    {
+      "epoch": 0.8073810498472778,
+      "grad_norm": 0.36115625500679016,
+      "learning_rate": 0.0002746234091360032,
+      "loss": 0.6638,
+      "step": 6410
+    },
+    {
+      "epoch": 0.8080108322574551,
+      "grad_norm": 0.3503740727901459,
+      "learning_rate": 0.00027456219957612804,
+      "loss": 0.6652,
+      "step": 6415
+    },
+    {
+      "epoch": 0.8086406146676324,
+      "grad_norm": 0.3303118646144867,
+      "learning_rate": 0.0002745009231235064,
+      "loss": 0.614,
+      "step": 6420
+    },
+    {
+      "epoch": 0.8092703970778096,
+      "grad_norm": 0.35880813002586365,
+      "learning_rate": 0.00027443957981104517,
+      "loss": 0.6449,
+      "step": 6425
+    },
+    {
+      "epoch": 0.8099001794879869,
+      "grad_norm": 0.3664454221725464,
+      "learning_rate": 0.000274378169671687,
+      "loss": 0.6448,
+      "step": 6430
+    },
+    {
+      "epoch": 0.8105299618981642,
+      "grad_norm": 0.38473254442214966,
+      "learning_rate": 0.00027431669273841067,
+      "loss": 0.6576,
+      "step": 6435
+    },
+    {
+      "epoch": 0.8111597443083415,
+      "grad_norm": 0.3694675862789154,
+      "learning_rate": 0.0002742551490442307,
+      "loss": 0.6365,
+      "step": 6440
+    },
+    {
+      "epoch": 0.8117895267185188,
+      "grad_norm": 0.32066047191619873,
+      "learning_rate": 0.0002741935386221973,
+      "loss": 0.6563,
+      "step": 6445
+    },
+    {
+      "epoch": 0.8124193091286961,
+      "grad_norm": 0.3764455020427704,
+      "learning_rate": 0.0002741318615053968,
+      "loss": 0.61,
+      "step": 6450
+    },
+    {
+      "epoch": 0.8130490915388733,
+      "grad_norm": 0.3913812041282654,
+      "learning_rate": 0.00027407011772695124,
+      "loss": 0.6606,
+      "step": 6455
+    },
+    {
+      "epoch": 0.8136788739490506,
+      "grad_norm": 0.2876626253128052,
+      "learning_rate": 0.0002740083073200184,
+      "loss": 0.6123,
+      "step": 6460
+    },
+    {
+      "epoch": 0.8143086563592279,
+      "grad_norm": 0.37668120861053467,
+      "learning_rate": 0.0002739464303177919,
+      "loss": 0.6323,
+      "step": 6465
+    },
+    {
+      "epoch": 0.8149384387694052,
+      "grad_norm": 0.3343159854412079,
+      "learning_rate": 0.000273884486753501,
+      "loss": 0.6051,
+      "step": 6470
+    },
+    {
+      "epoch": 0.8155682211795825,
+      "grad_norm": 0.3852281868457794,
+      "learning_rate": 0.00027382247666041097,
+      "loss": 0.6614,
+      "step": 6475
+    },
+    {
+      "epoch": 0.8161980035897597,
+      "grad_norm": 0.36491283774375916,
+      "learning_rate": 0.0002737604000718225,
+      "loss": 0.6383,
+      "step": 6480
+    },
+    {
+      "epoch": 0.816827785999937,
+      "grad_norm": 0.32019633054733276,
+      "learning_rate": 0.00027369825702107224,
+      "loss": 0.623,
+      "step": 6485
+    },
+    {
+      "epoch": 0.8174575684101143,
+      "grad_norm": 0.3173837661743164,
+      "learning_rate": 0.0002736360475415324,
+      "loss": 0.599,
+      "step": 6490
+    },
+    {
+      "epoch": 0.8180873508202916,
+      "grad_norm": 0.31505605578422546,
+      "learning_rate": 0.00027357377166661086,
+      "loss": 0.6341,
+      "step": 6495
+    },
+    {
+      "epoch": 0.8187171332304689,
+      "grad_norm": 0.3370759189128876,
+      "learning_rate": 0.00027351142942975124,
+      "loss": 0.6296,
+      "step": 6500
+    },
+    {
+      "epoch": 0.8193469156406462,
+      "grad_norm": 0.3554564416408539,
+      "learning_rate": 0.0002734490208644327,
+      "loss": 0.6587,
+      "step": 6505
+    },
+    {
+      "epoch": 0.8199766980508234,
+      "grad_norm": 0.3487757444381714,
+      "learning_rate": 0.0002733865460041701,
+      "loss": 0.6292,
+      "step": 6510
+    },
+    {
+      "epoch": 0.8206064804610007,
+      "grad_norm": 0.3280607759952545,
+      "learning_rate": 0.0002733240048825138,
+      "loss": 0.5964,
+      "step": 6515
+    },
+    {
+      "epoch": 0.821236262871178,
+      "grad_norm": 0.35416868329048157,
+      "learning_rate": 0.0002732613975330499,
+      "loss": 0.6089,
+      "step": 6520
+    },
+    {
+      "epoch": 0.8218660452813553,
+      "grad_norm": 0.3558996915817261,
+      "learning_rate": 0.00027319872398939995,
+      "loss": 0.5791,
+      "step": 6525
+    },
+    {
+      "epoch": 0.8224958276915326,
+      "grad_norm": 0.35394206643104553,
+      "learning_rate": 0.000273135984285221,
+      "loss": 0.6183,
+      "step": 6530
+    },
+    {
+      "epoch": 0.8231256101017098,
+      "grad_norm": 0.33172932267189026,
+      "learning_rate": 0.0002730731784542058,
+      "loss": 0.605,
+      "step": 6535
+    },
+    {
+      "epoch": 0.8237553925118871,
+      "grad_norm": 0.3498142957687378,
+      "learning_rate": 0.00027301030653008253,
+      "loss": 0.6199,
+      "step": 6540
+    },
+    {
+      "epoch": 0.8243851749220644,
+      "grad_norm": 0.3364173471927643,
+      "learning_rate": 0.0002729473685466148,
+      "loss": 0.6352,
+      "step": 6545
+    },
+    {
+      "epoch": 0.8250149573322417,
+      "grad_norm": 0.38148370385169983,
+      "learning_rate": 0.00027288436453760164,
+      "loss": 0.6216,
+      "step": 6550
+    },
+    {
+      "epoch": 0.825644739742419,
+      "grad_norm": 0.33975306153297424,
+      "learning_rate": 0.0002728212945368778,
+      "loss": 0.6155,
+      "step": 6555
+    },
+    {
+      "epoch": 0.8262745221525962,
+      "grad_norm": 0.3361944854259491,
+      "learning_rate": 0.0002727581585783133,
+      "loss": 0.6084,
+      "step": 6560
+    },
+    {
+      "epoch": 0.8269043045627735,
+      "grad_norm": 0.3503773808479309,
+      "learning_rate": 0.00027269495669581353,
+      "loss": 0.6355,
+      "step": 6565
+    },
+    {
+      "epoch": 0.8275340869729508,
+      "grad_norm": 0.35406753420829773,
+      "learning_rate": 0.00027263168892331934,
+      "loss": 0.624,
+      "step": 6570
+    },
+    {
+      "epoch": 0.8281638693831281,
+      "grad_norm": 0.3337428569793701,
+      "learning_rate": 0.00027256835529480697,
+      "loss": 0.6451,
+      "step": 6575
+    },
+    {
+      "epoch": 0.8287936517933054,
+      "grad_norm": 0.3431616425514221,
+      "learning_rate": 0.00027250495584428807,
+      "loss": 0.5969,
+      "step": 6580
+    },
+    {
+      "epoch": 0.8294234342034827,
+      "grad_norm": 0.4032285511493683,
+      "learning_rate": 0.0002724414906058096,
+      "loss": 0.5954,
+      "step": 6585
+    },
+    {
+      "epoch": 0.8300532166136599,
+      "grad_norm": 0.3352124094963074,
+      "learning_rate": 0.00027237795961345383,
+      "loss": 0.6077,
+      "step": 6590
+    },
+    {
+      "epoch": 0.8306829990238372,
+      "grad_norm": 0.3181077837944031,
+      "learning_rate": 0.0002723143629013383,
+      "loss": 0.6107,
+      "step": 6595
+    },
+    {
+      "epoch": 0.8313127814340145,
+      "grad_norm": 0.32390958070755005,
+      "learning_rate": 0.000272250700503616,
+      "loss": 0.6345,
+      "step": 6600
+    },
+    {
+      "epoch": 0.8319425638441919,
+      "grad_norm": 0.3199234902858734,
+      "learning_rate": 0.0002721869724544749,
+      "loss": 0.6268,
+      "step": 6605
+    },
+    {
+      "epoch": 0.8325723462543692,
+      "grad_norm": 0.38811957836151123,
+      "learning_rate": 0.00027212317878813863,
+      "loss": 0.643,
+      "step": 6610
+    },
+    {
+      "epoch": 0.8332021286645463,
+      "grad_norm": 0.38010820746421814,
+      "learning_rate": 0.00027205931953886575,
+      "loss": 0.6055,
+      "step": 6615
+    },
+    {
+      "epoch": 0.8338319110747237,
+      "grad_norm": 0.3288145065307617,
+      "learning_rate": 0.00027199539474095013,
+      "loss": 0.6311,
+      "step": 6620
+    },
+    {
+      "epoch": 0.834461693484901,
+      "grad_norm": 0.33361807465553284,
+      "learning_rate": 0.0002719314044287209,
+      "loss": 0.6083,
+      "step": 6625
+    },
+    {
+      "epoch": 0.8350914758950783,
+      "grad_norm": 0.350864440202713,
+      "learning_rate": 0.0002718673486365423,
+      "loss": 0.5969,
+      "step": 6630
+    },
+    {
+      "epoch": 0.8357212583052556,
+      "grad_norm": 0.35795754194259644,
+      "learning_rate": 0.0002718032273988137,
+      "loss": 0.6623,
+      "step": 6635
+    },
+    {
+      "epoch": 0.8363510407154329,
+      "grad_norm": 0.3748815357685089,
+      "learning_rate": 0.0002717390407499697,
+      "loss": 0.6301,
+      "step": 6640
+    },
+    {
+      "epoch": 0.8369808231256101,
+      "grad_norm": 0.3146851360797882,
+      "learning_rate": 0.00027167478872448,
+      "loss": 0.62,
+      "step": 6645
+    },
+    {
+      "epoch": 0.8376106055357874,
+      "grad_norm": 0.3758367598056793,
+      "learning_rate": 0.0002716104713568495,
+      "loss": 0.6202,
+      "step": 6650
+    },
+    {
+      "epoch": 0.8382403879459647,
+      "grad_norm": 0.4035817086696625,
+      "learning_rate": 0.0002715460886816179,
+      "loss": 0.606,
+      "step": 6655
+    },
+    {
+      "epoch": 0.838870170356142,
+      "grad_norm": 0.3586306869983673,
+      "learning_rate": 0.00027148164073336026,
+      "loss": 0.6523,
+      "step": 6660
+    },
+    {
+      "epoch": 0.8394999527663193,
+      "grad_norm": 0.33375057578086853,
+      "learning_rate": 0.0002714171275466866,
+      "loss": 0.6193,
+      "step": 6665
+    },
+    {
+      "epoch": 0.8401297351764965,
+      "grad_norm": 0.30258163809776306,
+      "learning_rate": 0.0002713525491562421,
+      "loss": 0.6225,
+      "step": 6670
+    },
+    {
+      "epoch": 0.8407595175866738,
+      "grad_norm": 0.33032524585723877,
+      "learning_rate": 0.00027128790559670667,
+      "loss": 0.628,
+      "step": 6675
+    },
+    {
+      "epoch": 0.8413892999968511,
+      "grad_norm": 0.36689457297325134,
+      "learning_rate": 0.00027122319690279535,
+      "loss": 0.6341,
+      "step": 6680
+    },
+    {
+      "epoch": 0.8420190824070284,
+      "grad_norm": 0.35744035243988037,
+      "learning_rate": 0.00027115842310925837,
+      "loss": 0.5945,
+      "step": 6685
+    },
+    {
+      "epoch": 0.8426488648172057,
+      "grad_norm": 0.3377218246459961,
+      "learning_rate": 0.0002710935842508806,
+      "loss": 0.6216,
+      "step": 6690
+    },
+    {
+      "epoch": 0.843278647227383,
+      "grad_norm": 0.3244309723377228,
+      "learning_rate": 0.000271028680362482,
+      "loss": 0.6045,
+      "step": 6695
+    },
+    {
+      "epoch": 0.8439084296375602,
+      "grad_norm": 0.34593185782432556,
+      "learning_rate": 0.00027096371147891744,
+      "loss": 0.6277,
+      "step": 6700
+    },
+    {
+      "epoch": 0.8445382120477375,
+      "grad_norm": 0.3151993751525879,
+      "learning_rate": 0.0002708986776350767,
+      "loss": 0.5929,
+      "step": 6705
+    },
+    {
+      "epoch": 0.8451679944579148,
+      "grad_norm": 0.38307860493659973,
+      "learning_rate": 0.0002708335788658845,
+      "loss": 0.5934,
+      "step": 6710
+    },
+    {
+      "epoch": 0.8457977768680921,
+      "grad_norm": 0.3155449330806732,
+      "learning_rate": 0.0002707684152063003,
+      "loss": 0.5838,
+      "step": 6715
+    },
+    {
+      "epoch": 0.8464275592782694,
+      "grad_norm": 0.3827744424343109,
+      "learning_rate": 0.00027070318669131845,
+      "loss": 0.5976,
+      "step": 6720
+    },
+    {
+      "epoch": 0.8470573416884466,
+      "grad_norm": 0.35382625460624695,
+      "learning_rate": 0.00027063789335596825,
+      "loss": 0.5997,
+      "step": 6725
+    },
+    {
+      "epoch": 0.8476871240986239,
+      "grad_norm": 0.36884164810180664,
+      "learning_rate": 0.00027057253523531365,
+      "loss": 0.6373,
+      "step": 6730
+    },
+    {
+      "epoch": 0.8483169065088012,
+      "grad_norm": 0.35557276010513306,
+      "learning_rate": 0.0002705071123644534,
+      "loss": 0.6717,
+      "step": 6735
+    },
+    {
+      "epoch": 0.8489466889189785,
+      "grad_norm": 0.3088480234146118,
+      "learning_rate": 0.00027044162477852124,
+      "loss": 0.6011,
+      "step": 6740
+    },
+    {
+      "epoch": 0.8495764713291558,
+      "grad_norm": 0.33665964007377625,
+      "learning_rate": 0.0002703760725126853,
+      "loss": 0.6039,
+      "step": 6745
+    },
+    {
+      "epoch": 0.850206253739333,
+      "grad_norm": 0.3297666311264038,
+      "learning_rate": 0.0002703104556021488,
+      "loss": 0.6226,
+      "step": 6750
+    },
+    {
+      "epoch": 0.8508360361495103,
+      "grad_norm": 0.33897268772125244,
+      "learning_rate": 0.00027024477408214945,
+      "loss": 0.5564,
+      "step": 6755
+    },
+    {
+      "epoch": 0.8514658185596876,
+      "grad_norm": 0.3549680709838867,
+      "learning_rate": 0.0002701790279879597,
+      "loss": 0.5989,
+      "step": 6760
+    },
+    {
+      "epoch": 0.8520956009698649,
+      "grad_norm": 0.31162139773368835,
+      "learning_rate": 0.0002701132173548868,
+      "loss": 0.6363,
+      "step": 6765
+    },
+    {
+      "epoch": 0.8527253833800422,
+      "grad_norm": 0.35543885827064514,
+      "learning_rate": 0.0002700473422182724,
+      "loss": 0.6228,
+      "step": 6770
+    },
+    {
+      "epoch": 0.8533551657902195,
+      "grad_norm": 0.3361263871192932,
+      "learning_rate": 0.0002699814026134932,
+      "loss": 0.5957,
+      "step": 6775
+    },
+    {
+      "epoch": 0.8539849482003967,
+      "grad_norm": 0.2764013707637787,
+      "learning_rate": 0.00026991539857596,
+      "loss": 0.5982,
+      "step": 6780
+    },
+    {
+      "epoch": 0.854614730610574,
+      "grad_norm": 0.3229328691959381,
+      "learning_rate": 0.0002698493301411187,
+      "loss": 0.6562,
+      "step": 6785
+    },
+    {
+      "epoch": 0.8552445130207513,
+      "grad_norm": 0.3163946270942688,
+      "learning_rate": 0.00026978319734444943,
+      "loss": 0.6125,
+      "step": 6790
+    },
+    {
+      "epoch": 0.8558742954309286,
+      "grad_norm": 0.38809090852737427,
+      "learning_rate": 0.0002697170002214671,
+      "loss": 0.6308,
+      "step": 6795
+    },
+    {
+      "epoch": 0.8565040778411059,
+      "grad_norm": 0.3416973650455475,
+      "learning_rate": 0.0002696507388077209,
+      "loss": 0.6565,
+      "step": 6800
+    },
+    {
+      "epoch": 0.8571338602512831,
+      "grad_norm": 0.3220008909702301,
+      "learning_rate": 0.00026958441313879494,
+      "loss": 0.6211,
+      "step": 6805
+    },
+    {
+      "epoch": 0.8577636426614604,
+      "grad_norm": 0.34507647156715393,
+      "learning_rate": 0.00026951802325030755,
+      "loss": 0.6384,
+      "step": 6810
+    },
+    {
+      "epoch": 0.8583934250716377,
+      "grad_norm": 0.3345770239830017,
+      "learning_rate": 0.00026945156917791154,
+      "loss": 0.6566,
+      "step": 6815
+    },
+    {
+      "epoch": 0.859023207481815,
+      "grad_norm": 0.32488980889320374,
+      "learning_rate": 0.0002693850509572943,
+      "loss": 0.626,
+      "step": 6820
+    },
+    {
+      "epoch": 0.8596529898919923,
+      "grad_norm": 0.3537434935569763,
+      "learning_rate": 0.00026931846862417766,
+      "loss": 0.6539,
+      "step": 6825
+    },
+    {
+      "epoch": 0.8602827723021697,
+      "grad_norm": 0.3165736794471741,
+      "learning_rate": 0.0002692518222143179,
+      "loss": 0.6468,
+      "step": 6830
+    },
+    {
+      "epoch": 0.8609125547123468,
+      "grad_norm": 0.34746891260147095,
+      "learning_rate": 0.0002691851117635056,
+      "loss": 0.6498,
+      "step": 6835
+    },
+    {
+      "epoch": 0.8615423371225241,
+      "grad_norm": 0.3370078206062317,
+      "learning_rate": 0.00026911833730756577,
+      "loss": 0.5951,
+      "step": 6840
+    },
+    {
+      "epoch": 0.8621721195327015,
+      "grad_norm": 0.3180099427700043,
+      "learning_rate": 0.00026905149888235787,
+      "loss": 0.609,
+      "step": 6845
+    },
+    {
+      "epoch": 0.8628019019428788,
+      "grad_norm": 0.34123897552490234,
+      "learning_rate": 0.0002689845965237757,
+      "loss": 0.6228,
+      "step": 6850
+    },
+    {
+      "epoch": 0.8634316843530561,
+      "grad_norm": 0.3529733717441559,
+      "learning_rate": 0.00026891763026774725,
+      "loss": 0.6101,
+      "step": 6855
+    },
+    {
+      "epoch": 0.8640614667632333,
+      "grad_norm": 0.3116464614868164,
+      "learning_rate": 0.00026885060015023496,
+      "loss": 0.5734,
+      "step": 6860
+    },
+    {
+      "epoch": 0.8646912491734106,
+      "grad_norm": 0.3331621587276459,
+      "learning_rate": 0.00026878350620723556,
+      "loss": 0.6004,
+      "step": 6865
+    },
+    {
+      "epoch": 0.8653210315835879,
+      "grad_norm": 0.3215835690498352,
+      "learning_rate": 0.00026871634847478007,
+      "loss": 0.6105,
+      "step": 6870
+    },
+    {
+      "epoch": 0.8659508139937652,
+      "grad_norm": 0.3454177677631378,
+      "learning_rate": 0.0002686491269889336,
+      "loss": 0.6203,
+      "step": 6875
+    },
+    {
+      "epoch": 0.8665805964039425,
+      "grad_norm": 0.3336181640625,
+      "learning_rate": 0.0002685818417857958,
+      "loss": 0.6179,
+      "step": 6880
+    },
+    {
+      "epoch": 0.8672103788141198,
+      "grad_norm": 0.3452587127685547,
+      "learning_rate": 0.00026851449290150024,
+      "loss": 0.5918,
+      "step": 6885
+    },
+    {
+      "epoch": 0.867840161224297,
+      "grad_norm": 0.37552833557128906,
+      "learning_rate": 0.0002684470803722148,
+      "loss": 0.6284,
+      "step": 6890
+    },
+    {
+      "epoch": 0.8684699436344743,
+      "grad_norm": 0.33525559306144714,
+      "learning_rate": 0.0002683796042341416,
+      "loss": 0.6465,
+      "step": 6895
+    },
+    {
+      "epoch": 0.8690997260446516,
+      "grad_norm": 0.3272569477558136,
+      "learning_rate": 0.00026831206452351683,
+      "loss": 0.636,
+      "step": 6900
+    },
+    {
+      "epoch": 0.8697295084548289,
+      "grad_norm": 0.35215091705322266,
+      "learning_rate": 0.0002682444612766109,
+      "loss": 0.6415,
+      "step": 6905
+    },
+    {
+      "epoch": 0.8703592908650062,
+      "grad_norm": 0.33025211095809937,
+      "learning_rate": 0.0002681767945297282,
+      "loss": 0.6677,
+      "step": 6910
+    },
+    {
+      "epoch": 0.8709890732751834,
+      "grad_norm": 0.34073176980018616,
+      "learning_rate": 0.0002681090643192075,
+      "loss": 0.6386,
+      "step": 6915
+    },
+    {
+      "epoch": 0.8716188556853607,
+      "grad_norm": 0.4070134162902832,
+      "learning_rate": 0.0002680412706814213,
+      "loss": 0.6365,
+      "step": 6920
+    },
+    {
+      "epoch": 0.872248638095538,
+      "grad_norm": 0.33693283796310425,
+      "learning_rate": 0.00026797341365277644,
+      "loss": 0.6465,
+      "step": 6925
+    },
+    {
+      "epoch": 0.8728784205057153,
+      "grad_norm": 0.3678983747959137,
+      "learning_rate": 0.0002679054932697136,
+      "loss": 0.594,
+      "step": 6930
+    },
+    {
+      "epoch": 0.8735082029158926,
+      "grad_norm": 0.31632333993911743,
+      "learning_rate": 0.00026783750956870764,
+      "loss": 0.6128,
+      "step": 6935
+    },
+    {
+      "epoch": 0.8741379853260698,
+      "grad_norm": 0.3184865713119507,
+      "learning_rate": 0.0002677694625862674,
+      "loss": 0.5955,
+      "step": 6940
+    },
+    {
+      "epoch": 0.8747677677362471,
+      "grad_norm": 0.33729860186576843,
+      "learning_rate": 0.00026770135235893556,
+      "loss": 0.609,
+      "step": 6945
+    },
+    {
+      "epoch": 0.8753975501464244,
+      "grad_norm": 0.3195466995239258,
+      "learning_rate": 0.0002676331789232889,
+      "loss": 0.6399,
+      "step": 6950
+    },
+    {
+      "epoch": 0.8760273325566017,
+      "grad_norm": 0.35504212975502014,
+      "learning_rate": 0.0002675649423159382,
+      "loss": 0.6162,
+      "step": 6955
+    },
+    {
+      "epoch": 0.876657114966779,
+      "grad_norm": 0.3598940372467041,
+      "learning_rate": 0.000267496642573528,
+      "loss": 0.6117,
+      "step": 6960
+    },
+    {
+      "epoch": 0.8772868973769563,
+      "grad_norm": 0.32016637921333313,
+      "learning_rate": 0.0002674282797327368,
+      "loss": 0.6129,
+      "step": 6965
+    },
+    {
+      "epoch": 0.8779166797871335,
+      "grad_norm": 0.36968451738357544,
+      "learning_rate": 0.00026735985383027704,
+      "loss": 0.619,
+      "step": 6970
+    },
+    {
+      "epoch": 0.8785464621973108,
+      "grad_norm": 0.3299955427646637,
+      "learning_rate": 0.000267291364902895,
+      "loss": 0.5894,
+      "step": 6975
+    },
+    {
+      "epoch": 0.8791762446074881,
+      "grad_norm": 0.34892305731773376,
+      "learning_rate": 0.0002672228129873708,
+      "loss": 0.6152,
+      "step": 6980
+    },
+    {
+      "epoch": 0.8798060270176654,
+      "grad_norm": 0.379016637802124,
+      "learning_rate": 0.00026715419812051833,
+      "loss": 0.6633,
+      "step": 6985
+    },
+    {
+      "epoch": 0.8804358094278427,
+      "grad_norm": 0.3378797173500061,
+      "learning_rate": 0.00026708552033918544,
+      "loss": 0.5911,
+      "step": 6990
+    },
+    {
+      "epoch": 0.8810655918380199,
+      "grad_norm": 0.3348138928413391,
+      "learning_rate": 0.0002670167796802536,
+      "loss": 0.5841,
+      "step": 6995
+    },
+    {
+      "epoch": 0.8816953742481972,
+      "grad_norm": 0.36374861001968384,
+      "learning_rate": 0.0002669479761806381,
+      "loss": 0.5973,
+      "step": 7000
+    },
+    {
+      "epoch": 0.8816953742481972,
+      "eval_loss": 0.3066178560256958,
+      "eval_runtime": 6.2494,
+      "eval_samples_per_second": 160.014,
+      "eval_steps_per_second": 10.081,
+      "step": 7000
+    },
+    {
+      "epoch": 0.8823251566583745,
+      "grad_norm": 0.31616318225860596,
+      "learning_rate": 0.000266879109877288,
+      "loss": 0.6302,
+      "step": 7005
+    },
+    {
+      "epoch": 0.8829549390685518,
+      "grad_norm": 0.37413114309310913,
+      "learning_rate": 0.00026681018080718615,
+      "loss": 0.6141,
+      "step": 7010
+    },
+    {
+      "epoch": 0.8835847214787291,
+      "grad_norm": 0.3616124987602234,
+      "learning_rate": 0.0002667411890073489,
+      "loss": 0.6081,
+      "step": 7015
+    },
+    {
+      "epoch": 0.8842145038889064,
+      "grad_norm": 0.3536156713962555,
+      "learning_rate": 0.00026667213451482655,
+      "loss": 0.6101,
+      "step": 7020
+    },
+    {
+      "epoch": 0.8848442862990836,
+      "grad_norm": 0.2826579809188843,
+      "learning_rate": 0.00026660301736670293,
+      "loss": 0.5803,
+      "step": 7025
+    },
+    {
+      "epoch": 0.8854740687092609,
+      "grad_norm": 0.3352709710597992,
+      "learning_rate": 0.00026653383760009546,
+      "loss": 0.5994,
+      "step": 7030
+    },
+    {
+      "epoch": 0.8861038511194382,
+      "grad_norm": 0.320122092962265,
+      "learning_rate": 0.00026646459525215524,
+      "loss": 0.6159,
+      "step": 7035
+    },
+    {
+      "epoch": 0.8867336335296155,
+      "grad_norm": 0.3512963652610779,
+      "learning_rate": 0.0002663952903600671,
+      "loss": 0.6034,
+      "step": 7040
+    },
+    {
+      "epoch": 0.8873634159397928,
+      "grad_norm": 0.358071506023407,
+      "learning_rate": 0.00026632592296104926,
+      "loss": 0.6155,
+      "step": 7045
+    },
+    {
+      "epoch": 0.88799319834997,
+      "grad_norm": 0.342318058013916,
+      "learning_rate": 0.0002662564930923536,
+      "loss": 0.5997,
+      "step": 7050
+    },
+    {
+      "epoch": 0.8886229807601473,
+      "grad_norm": 0.291960746049881,
+      "learning_rate": 0.0002661870007912656,
+      "loss": 0.5721,
+      "step": 7055
+    },
+    {
+      "epoch": 0.8892527631703246,
+      "grad_norm": 0.3608805239200592,
+      "learning_rate": 0.0002661174460951042,
+      "loss": 0.6248,
+      "step": 7060
+    },
+    {
+      "epoch": 0.889882545580502,
+      "grad_norm": 0.329289972782135,
+      "learning_rate": 0.0002660478290412218,
+      "loss": 0.6163,
+      "step": 7065
+    },
+    {
+      "epoch": 0.8905123279906793,
+      "grad_norm": 0.352383553981781,
+      "learning_rate": 0.0002659781496670044,
+      "loss": 0.6252,
+      "step": 7070
+    },
+    {
+      "epoch": 0.8911421104008566,
+      "grad_norm": 0.3424574136734009,
+      "learning_rate": 0.0002659084080098714,
+      "loss": 0.5562,
+      "step": 7075
+    },
+    {
+      "epoch": 0.8917718928110338,
+      "grad_norm": 0.32095563411712646,
+      "learning_rate": 0.0002658386041072757,
+      "loss": 0.6232,
+      "step": 7080
+    },
+    {
+      "epoch": 0.892401675221211,
+      "grad_norm": 0.3307218849658966,
+      "learning_rate": 0.00026576873799670356,
+      "loss": 0.5958,
+      "step": 7085
+    },
+    {
+      "epoch": 0.8930314576313884,
+      "grad_norm": 0.31858259439468384,
+      "learning_rate": 0.00026569880971567464,
+      "loss": 0.6128,
+      "step": 7090
+    },
+    {
+      "epoch": 0.8936612400415657,
+      "grad_norm": 0.3014832139015198,
+      "learning_rate": 0.00026562881930174213,
+      "loss": 0.5886,
+      "step": 7095
+    },
+    {
+      "epoch": 0.894291022451743,
+      "grad_norm": 0.35925576090812683,
+      "learning_rate": 0.00026555876679249234,
+      "loss": 0.6032,
+      "step": 7100
+    },
+    {
+      "epoch": 0.8949208048619202,
+      "grad_norm": 0.337100625038147,
+      "learning_rate": 0.0002654886522255452,
+      "loss": 0.6217,
+      "step": 7105
+    },
+    {
+      "epoch": 0.8955505872720975,
+      "grad_norm": 0.34906861186027527,
+      "learning_rate": 0.00026541847563855373,
+      "loss": 0.5999,
+      "step": 7110
+    },
+    {
+      "epoch": 0.8961803696822748,
+      "grad_norm": 0.2829444110393524,
+      "learning_rate": 0.00026534823706920443,
+      "loss": 0.5747,
+      "step": 7115
+    },
+    {
+      "epoch": 0.8968101520924521,
+      "grad_norm": 0.3298097550868988,
+      "learning_rate": 0.00026527793655521697,
+      "loss": 0.5959,
+      "step": 7120
+    },
+    {
+      "epoch": 0.8974399345026294,
+      "grad_norm": 0.3762158453464508,
+      "learning_rate": 0.0002652075741343444,
+      "loss": 0.6325,
+      "step": 7125
+    },
+    {
+      "epoch": 0.8980697169128066,
+      "grad_norm": 0.3318065106868744,
+      "learning_rate": 0.00026513714984437284,
+      "loss": 0.6015,
+      "step": 7130
+    },
+    {
+      "epoch": 0.8986994993229839,
+      "grad_norm": 0.3132246434688568,
+      "learning_rate": 0.0002650666637231218,
+      "loss": 0.6317,
+      "step": 7135
+    },
+    {
+      "epoch": 0.8993292817331612,
+      "grad_norm": 0.3308473527431488,
+      "learning_rate": 0.00026499611580844403,
+      "loss": 0.6364,
+      "step": 7140
+    },
+    {
+      "epoch": 0.8999590641433385,
+      "grad_norm": 0.31450155377388,
+      "learning_rate": 0.0002649255061382252,
+      "loss": 0.6186,
+      "step": 7145
+    },
+    {
+      "epoch": 0.9005888465535158,
+      "grad_norm": 0.3408615291118622,
+      "learning_rate": 0.00026485483475038445,
+      "loss": 0.5954,
+      "step": 7150
+    },
+    {
+      "epoch": 0.9012186289636931,
+      "grad_norm": 0.34355321526527405,
+      "learning_rate": 0.0002647841016828738,
+      "loss": 0.6143,
+      "step": 7155
+    },
+    {
+      "epoch": 0.9018484113738703,
+      "grad_norm": 0.35341107845306396,
+      "learning_rate": 0.00026471330697367865,
+      "loss": 0.5887,
+      "step": 7160
+    },
+    {
+      "epoch": 0.9024781937840476,
+      "grad_norm": 0.3439336121082306,
+      "learning_rate": 0.0002646424506608173,
+      "loss": 0.6152,
+      "step": 7165
+    },
+    {
+      "epoch": 0.9031079761942249,
+      "grad_norm": 0.32301509380340576,
+      "learning_rate": 0.00026457153278234126,
+      "loss": 0.6191,
+      "step": 7170
+    },
+    {
+      "epoch": 0.9037377586044022,
+      "grad_norm": 0.3085480034351349,
+      "learning_rate": 0.000264500553376335,
+      "loss": 0.5993,
+      "step": 7175
+    },
+    {
+      "epoch": 0.9043675410145795,
+      "grad_norm": 0.3285475969314575,
+      "learning_rate": 0.0002644295124809161,
+      "loss": 0.5832,
+      "step": 7180
+    },
+    {
+      "epoch": 0.9049973234247567,
+      "grad_norm": 0.3160327076911926,
+      "learning_rate": 0.0002643584101342352,
+      "loss": 0.6258,
+      "step": 7185
+    },
+    {
+      "epoch": 0.905627105834934,
+      "grad_norm": 0.30449238419532776,
+      "learning_rate": 0.0002642872463744759,
+      "loss": 0.62,
+      "step": 7190
+    },
+    {
+      "epoch": 0.9062568882451113,
+      "grad_norm": 0.31154754757881165,
+      "learning_rate": 0.00026421602123985455,
+      "loss": 0.5888,
+      "step": 7195
+    },
+    {
+      "epoch": 0.9068866706552886,
+      "grad_norm": 0.32224607467651367,
+      "learning_rate": 0.0002641447347686209,
+      "loss": 0.5971,
+      "step": 7200
+    },
+    {
+      "epoch": 0.9075164530654659,
+      "grad_norm": 0.33809399604797363,
+      "learning_rate": 0.0002640733869990573,
+      "loss": 0.5942,
+      "step": 7205
+    },
+    {
+      "epoch": 0.9081462354756432,
+      "grad_norm": 0.337990403175354,
+      "learning_rate": 0.0002640019779694792,
+      "loss": 0.5996,
+      "step": 7210
+    },
+    {
+      "epoch": 0.9087760178858204,
+      "grad_norm": 0.33843520283699036,
+      "learning_rate": 0.0002639305077182348,
+      "loss": 0.6009,
+      "step": 7215
+    },
+    {
+      "epoch": 0.9094058002959977,
+      "grad_norm": 0.31854307651519775,
+      "learning_rate": 0.00026385897628370536,
+      "loss": 0.5929,
+      "step": 7220
+    },
+    {
+      "epoch": 0.910035582706175,
+      "grad_norm": 0.31263160705566406,
+      "learning_rate": 0.0002637873837043049,
+      "loss": 0.5861,
+      "step": 7225
+    },
+    {
+      "epoch": 0.9106653651163523,
+      "grad_norm": 0.3141006827354431,
+      "learning_rate": 0.00026371573001848005,
+      "loss": 0.6204,
+      "step": 7230
+    },
+    {
+      "epoch": 0.9112951475265296,
+      "grad_norm": 0.3565130829811096,
+      "learning_rate": 0.00026364401526471077,
+      "loss": 0.6051,
+      "step": 7235
+    },
+    {
+      "epoch": 0.9119249299367068,
+      "grad_norm": 0.3886755108833313,
+      "learning_rate": 0.0002635722394815094,
+      "loss": 0.6162,
+      "step": 7240
+    },
+    {
+      "epoch": 0.9125547123468841,
+      "grad_norm": 0.32173478603363037,
+      "learning_rate": 0.0002635004027074211,
+      "loss": 0.5908,
+      "step": 7245
+    },
+    {
+      "epoch": 0.9131844947570614,
+      "grad_norm": 0.3483346998691559,
+      "learning_rate": 0.0002634285049810239,
+      "loss": 0.5934,
+      "step": 7250
+    },
+    {
+      "epoch": 0.9138142771672387,
+      "grad_norm": 0.31829094886779785,
+      "learning_rate": 0.00026335654634092857,
+      "loss": 0.6205,
+      "step": 7255
+    },
+    {
+      "epoch": 0.914444059577416,
+      "grad_norm": 0.2864934206008911,
+      "learning_rate": 0.0002632845268257785,
+      "loss": 0.5486,
+      "step": 7260
+    },
+    {
+      "epoch": 0.9150738419875933,
+      "grad_norm": 0.34583529829978943,
+      "learning_rate": 0.0002632124464742499,
+      "loss": 0.5994,
+      "step": 7265
+    },
+    {
+      "epoch": 0.9157036243977705,
+      "grad_norm": 0.3405662775039673,
+      "learning_rate": 0.00026314030532505146,
+      "loss": 0.5941,
+      "step": 7270
+    },
+    {
+      "epoch": 0.9163334068079478,
+      "grad_norm": 0.319985568523407,
+      "learning_rate": 0.00026306810341692464,
+      "loss": 0.5949,
+      "step": 7275
+    },
+    {
+      "epoch": 0.9169631892181251,
+      "grad_norm": 0.3206420838832855,
+      "learning_rate": 0.00026299584078864354,
+      "loss": 0.5895,
+      "step": 7280
+    },
+    {
+      "epoch": 0.9175929716283024,
+      "grad_norm": 0.33022215962409973,
+      "learning_rate": 0.00026292351747901486,
+      "loss": 0.6018,
+      "step": 7285
+    },
+    {
+      "epoch": 0.9182227540384797,
+      "grad_norm": 0.3440692722797394,
+      "learning_rate": 0.00026285113352687785,
+      "loss": 0.5818,
+      "step": 7290
+    },
+    {
+      "epoch": 0.9188525364486569,
+      "grad_norm": 0.3580811619758606,
+      "learning_rate": 0.0002627786889711043,
+      "loss": 0.6024,
+      "step": 7295
+    },
+    {
+      "epoch": 0.9194823188588342,
+      "grad_norm": 0.3101358413696289,
+      "learning_rate": 0.0002627061838505987,
+      "loss": 0.6241,
+      "step": 7300
+    },
+    {
+      "epoch": 0.9201121012690116,
+      "grad_norm": 0.3681425452232361,
+      "learning_rate": 0.00026263361820429783,
+      "loss": 0.5759,
+      "step": 7305
+    },
+    {
+      "epoch": 0.9207418836791889,
+      "grad_norm": 0.3331769108772278,
+      "learning_rate": 0.0002625609920711712,
+      "loss": 0.5696,
+      "step": 7310
+    },
+    {
+      "epoch": 0.9213716660893662,
+      "grad_norm": 0.34252071380615234,
+      "learning_rate": 0.00026248830549022064,
+      "loss": 0.6171,
+      "step": 7315
+    },
+    {
+      "epoch": 0.9220014484995434,
+      "grad_norm": 0.31009170413017273,
+      "learning_rate": 0.00026241555850048056,
+      "loss": 0.5758,
+      "step": 7320
+    },
+    {
+      "epoch": 0.9226312309097207,
+      "grad_norm": 0.33126717805862427,
+      "learning_rate": 0.00026234275114101765,
+      "loss": 0.557,
+      "step": 7325
+    },
+    {
+      "epoch": 0.923261013319898,
+      "grad_norm": 0.35423141717910767,
+      "learning_rate": 0.00026226988345093126,
+      "loss": 0.6239,
+      "step": 7330
+    },
+    {
+      "epoch": 0.9238907957300753,
+      "grad_norm": 0.31321558356285095,
+      "learning_rate": 0.0002621969554693529,
+      "loss": 0.5796,
+      "step": 7335
+    },
+    {
+      "epoch": 0.9245205781402526,
+      "grad_norm": 0.38709312677383423,
+      "learning_rate": 0.00026212396723544664,
+      "loss": 0.5831,
+      "step": 7340
+    },
+    {
+      "epoch": 0.9251503605504299,
+      "grad_norm": 0.3205506205558777,
+      "learning_rate": 0.0002620509187884088,
+      "loss": 0.5577,
+      "step": 7345
+    },
+    {
+      "epoch": 0.9257801429606071,
+      "grad_norm": 0.3263196647167206,
+      "learning_rate": 0.00026197781016746804,
+      "loss": 0.5729,
+      "step": 7350
+    },
+    {
+      "epoch": 0.9264099253707844,
+      "grad_norm": 0.3553536534309387,
+      "learning_rate": 0.0002619046414118854,
+      "loss": 0.5968,
+      "step": 7355
+    },
+    {
+      "epoch": 0.9270397077809617,
+      "grad_norm": 0.4170524477958679,
+      "learning_rate": 0.0002618314125609541,
+      "loss": 0.5731,
+      "step": 7360
+    },
+    {
+      "epoch": 0.927669490191139,
+      "grad_norm": 0.3739701807498932,
+      "learning_rate": 0.00026175812365399976,
+      "loss": 0.5785,
+      "step": 7365
+    },
+    {
+      "epoch": 0.9282992726013163,
+      "grad_norm": 0.32139813899993896,
+      "learning_rate": 0.0002616847747303802,
+      "loss": 0.5909,
+      "step": 7370
+    },
+    {
+      "epoch": 0.9289290550114935,
+      "grad_norm": 0.3099890947341919,
+      "learning_rate": 0.00026161136582948544,
+      "loss": 0.5579,
+      "step": 7375
+    },
+    {
+      "epoch": 0.9295588374216708,
+      "grad_norm": 0.349729984998703,
+      "learning_rate": 0.0002615378969907378,
+      "loss": 0.5762,
+      "step": 7380
+    },
+    {
+      "epoch": 0.9301886198318481,
+      "grad_norm": 0.3257734775543213,
+      "learning_rate": 0.00026146436825359167,
+      "loss": 0.6216,
+      "step": 7385
+    },
+    {
+      "epoch": 0.9308184022420254,
+      "grad_norm": 0.3399578332901001,
+      "learning_rate": 0.0002613907796575337,
+      "loss": 0.5694,
+      "step": 7390
+    },
+    {
+      "epoch": 0.9314481846522027,
+      "grad_norm": 0.3863985240459442,
+      "learning_rate": 0.0002613171312420826,
+      "loss": 0.6416,
+      "step": 7395
+    },
+    {
+      "epoch": 0.93207796706238,
+      "grad_norm": 0.3288150429725647,
+      "learning_rate": 0.0002612434230467892,
+      "loss": 0.5839,
+      "step": 7400
+    },
+    {
+      "epoch": 0.9327077494725572,
+      "grad_norm": 0.37783902883529663,
+      "learning_rate": 0.00026116965511123664,
+      "loss": 0.5919,
+      "step": 7405
+    },
+    {
+      "epoch": 0.9333375318827345,
+      "grad_norm": 0.36346110701560974,
+      "learning_rate": 0.00026109582747503986,
+      "loss": 0.5796,
+      "step": 7410
+    },
+    {
+      "epoch": 0.9339673142929118,
+      "grad_norm": 0.3194875419139862,
+      "learning_rate": 0.00026102194017784606,
+      "loss": 0.5808,
+      "step": 7415
+    },
+    {
+      "epoch": 0.9345970967030891,
+      "grad_norm": 0.286823570728302,
+      "learning_rate": 0.00026094799325933435,
+      "loss": 0.5605,
+      "step": 7420
+    },
+    {
+      "epoch": 0.9352268791132664,
+      "grad_norm": 0.3147251307964325,
+      "learning_rate": 0.0002608739867592159,
+      "loss": 0.572,
+      "step": 7425
+    },
+    {
+      "epoch": 0.9358566615234436,
+      "grad_norm": 0.34172821044921875,
+      "learning_rate": 0.000260799920717234,
+      "loss": 0.5763,
+      "step": 7430
+    },
+    {
+      "epoch": 0.9364864439336209,
+      "grad_norm": 0.32804232835769653,
+      "learning_rate": 0.0002607257951731637,
+      "loss": 0.5925,
+      "step": 7435
+    },
+    {
+      "epoch": 0.9371162263437982,
+      "grad_norm": 0.2969893515110016,
+      "learning_rate": 0.0002606516101668122,
+      "loss": 0.5754,
+      "step": 7440
+    },
+    {
+      "epoch": 0.9377460087539755,
+      "grad_norm": 0.3364142179489136,
+      "learning_rate": 0.00026057736573801844,
+      "loss": 0.6248,
+      "step": 7445
+    },
+    {
+      "epoch": 0.9383757911641528,
+      "grad_norm": 0.3493711054325104,
+      "learning_rate": 0.0002605030619266534,
+      "loss": 0.5828,
+      "step": 7450
+    },
+    {
+      "epoch": 0.9390055735743301,
+      "grad_norm": 0.3338306248188019,
+      "learning_rate": 0.00026042869877262,
+      "loss": 0.5947,
+      "step": 7455
+    },
+    {
+      "epoch": 0.9396353559845073,
+      "grad_norm": 0.30441364645957947,
+      "learning_rate": 0.0002603542763158529,
+      "loss": 0.5743,
+      "step": 7460
+    },
+    {
+      "epoch": 0.9402651383946846,
+      "grad_norm": 0.31838342547416687,
+      "learning_rate": 0.0002602797945963186,
+      "loss": 0.5493,
+      "step": 7465
+    },
+    {
+      "epoch": 0.9408949208048619,
+      "grad_norm": 0.3308780789375305,
+      "learning_rate": 0.0002602052536540156,
+      "loss": 0.5984,
+      "step": 7470
+    },
+    {
+      "epoch": 0.9415247032150392,
+      "grad_norm": 0.30487555265426636,
+      "learning_rate": 0.00026013065352897407,
+      "loss": 0.5687,
+      "step": 7475
+    },
+    {
+      "epoch": 0.9421544856252165,
+      "grad_norm": 0.33297523856163025,
+      "learning_rate": 0.0002600559942612559,
+      "loss": 0.5728,
+      "step": 7480
+    },
+    {
+      "epoch": 0.9427842680353937,
+      "grad_norm": 0.3194848299026489,
+      "learning_rate": 0.00025998127589095483,
+      "loss": 0.5939,
+      "step": 7485
+    },
+    {
+      "epoch": 0.943414050445571,
+      "grad_norm": 0.3401489555835724,
+      "learning_rate": 0.0002599064984581964,
+      "loss": 0.5282,
+      "step": 7490
+    },
+    {
+      "epoch": 0.9440438328557483,
+      "grad_norm": 0.3722991943359375,
+      "learning_rate": 0.0002598316620031378,
+      "loss": 0.6044,
+      "step": 7495
+    },
+    {
+      "epoch": 0.9446736152659256,
+      "grad_norm": 0.3582395613193512,
+      "learning_rate": 0.0002597567665659678,
+      "loss": 0.574,
+      "step": 7500
+    },
+    {
+      "epoch": 0.9453033976761029,
+      "grad_norm": 0.30922654271125793,
+      "learning_rate": 0.0002596818121869071,
+      "loss": 0.6086,
+      "step": 7505
+    },
+    {
+      "epoch": 0.9459331800862802,
+      "grad_norm": 0.34381213784217834,
+      "learning_rate": 0.00025960679890620785,
+      "loss": 0.6032,
+      "step": 7510
+    },
+    {
+      "epoch": 0.9465629624964574,
+      "grad_norm": 0.3153468072414398,
+      "learning_rate": 0.0002595317267641539,
+      "loss": 0.5758,
+      "step": 7515
+    },
+    {
+      "epoch": 0.9471927449066347,
+      "grad_norm": 0.30763527750968933,
+      "learning_rate": 0.0002594565958010607,
+      "loss": 0.6036,
+      "step": 7520
+    },
+    {
+      "epoch": 0.947822527316812,
+      "grad_norm": 0.33897343277931213,
+      "learning_rate": 0.00025938140605727536,
+      "loss": 0.5879,
+      "step": 7525
+    },
+    {
+      "epoch": 0.9484523097269894,
+      "grad_norm": 0.2996034324169159,
+      "learning_rate": 0.00025930615757317635,
+      "loss": 0.6095,
+      "step": 7530
+    },
+    {
+      "epoch": 0.9490820921371667,
+      "grad_norm": 0.37265533208847046,
+      "learning_rate": 0.00025923085038917395,
+      "loss": 0.5718,
+      "step": 7535
+    },
+    {
+      "epoch": 0.9497118745473438,
+      "grad_norm": 0.32904815673828125,
+      "learning_rate": 0.00025915548454570977,
+      "loss": 0.5689,
+      "step": 7540
+    },
+    {
+      "epoch": 0.9503416569575212,
+      "grad_norm": 0.3493824005126953,
+      "learning_rate": 0.000259080060083257,
+      "loss": 0.594,
+      "step": 7545
+    },
+    {
+      "epoch": 0.9509714393676985,
+      "grad_norm": 0.33561789989471436,
+      "learning_rate": 0.0002590045770423204,
+      "loss": 0.5604,
+      "step": 7550
+    },
+    {
+      "epoch": 0.9516012217778758,
+      "grad_norm": 0.3272433876991272,
+      "learning_rate": 0.00025892903546343587,
+      "loss": 0.5819,
+      "step": 7555
+    },
+    {
+      "epoch": 0.9522310041880531,
+      "grad_norm": 0.34539222717285156,
+      "learning_rate": 0.00025885343538717116,
+      "loss": 0.591,
+      "step": 7560
+    },
+    {
+      "epoch": 0.9528607865982303,
+      "grad_norm": 0.3331897258758545,
+      "learning_rate": 0.0002587777768541252,
+      "loss": 0.5885,
+      "step": 7565
+    },
+    {
+      "epoch": 0.9534905690084076,
+      "grad_norm": 0.3285147547721863,
+      "learning_rate": 0.00025870205990492827,
+      "loss": 0.5561,
+      "step": 7570
+    },
+    {
+      "epoch": 0.9541203514185849,
+      "grad_norm": 0.3221907317638397,
+      "learning_rate": 0.0002586262845802422,
+      "loss": 0.5837,
+      "step": 7575
+    },
+    {
+      "epoch": 0.9547501338287622,
+      "grad_norm": 0.4986007511615753,
+      "learning_rate": 0.00025855045092076,
+      "loss": 0.5645,
+      "step": 7580
+    },
+    {
+      "epoch": 0.9553799162389395,
+      "grad_norm": 0.33891043066978455,
+      "learning_rate": 0.00025847455896720615,
+      "loss": 0.5801,
+      "step": 7585
+    },
+    {
+      "epoch": 0.9560096986491168,
+      "grad_norm": 0.345480740070343,
+      "learning_rate": 0.00025839860876033626,
+      "loss": 0.5876,
+      "step": 7590
+    },
+    {
+      "epoch": 0.956639481059294,
+      "grad_norm": 0.39212220907211304,
+      "learning_rate": 0.0002583226003409374,
+      "loss": 0.5949,
+      "step": 7595
+    },
+    {
+      "epoch": 0.9572692634694713,
+      "grad_norm": 0.3195202648639679,
+      "learning_rate": 0.00025824653374982776,
+      "loss": 0.592,
+      "step": 7600
+    },
+    {
+      "epoch": 0.9578990458796486,
+      "grad_norm": 0.31688785552978516,
+      "learning_rate": 0.00025817040902785694,
+      "loss": 0.5432,
+      "step": 7605
+    },
+    {
+      "epoch": 0.9585288282898259,
+      "grad_norm": 0.3165288269519806,
+      "learning_rate": 0.00025809422621590554,
+      "loss": 0.552,
+      "step": 7610
+    },
+    {
+      "epoch": 0.9591586107000032,
+      "grad_norm": 0.33528926968574524,
+      "learning_rate": 0.0002580179853548856,
+      "loss": 0.5745,
+      "step": 7615
+    },
+    {
+      "epoch": 0.9597883931101804,
+      "grad_norm": 0.34123846888542175,
+      "learning_rate": 0.0002579416864857401,
+      "loss": 0.6019,
+      "step": 7620
+    },
+    {
+      "epoch": 0.9604181755203577,
+      "grad_norm": 0.3223724663257599,
+      "learning_rate": 0.0002578653296494433,
+      "loss": 0.5725,
+      "step": 7625
+    },
+    {
+      "epoch": 0.961047957930535,
+      "grad_norm": 0.349751740694046,
+      "learning_rate": 0.0002577889148870006,
+      "loss": 0.5739,
+      "step": 7630
+    },
+    {
+      "epoch": 0.9616777403407123,
+      "grad_norm": 0.3111324608325958,
+      "learning_rate": 0.0002577124422394484,
+      "loss": 0.5555,
+      "step": 7635
+    },
+    {
+      "epoch": 0.9623075227508896,
+      "grad_norm": 0.364615797996521,
+      "learning_rate": 0.00025763591174785433,
+      "loss": 0.5789,
+      "step": 7640
+    },
+    {
+      "epoch": 0.9629373051610669,
+      "grad_norm": 0.31817707419395447,
+      "learning_rate": 0.000257559323453317,
+      "loss": 0.5799,
+      "step": 7645
+    },
+    {
+      "epoch": 0.9635670875712441,
+      "grad_norm": 0.33710840344429016,
+      "learning_rate": 0.000257482677396966,
+      "loss": 0.6,
+      "step": 7650
+    },
+    {
+      "epoch": 0.9641968699814214,
+      "grad_norm": 0.3512105345726013,
+      "learning_rate": 0.00025740597361996215,
+      "loss": 0.5772,
+      "step": 7655
+    },
+    {
+      "epoch": 0.9648266523915987,
+      "grad_norm": 0.32505640387535095,
+      "learning_rate": 0.00025732921216349705,
+      "loss": 0.5872,
+      "step": 7660
+    },
+    {
+      "epoch": 0.965456434801776,
+      "grad_norm": 0.32156363129615784,
+      "learning_rate": 0.0002572523930687933,
+      "loss": 0.5842,
+      "step": 7665
+    },
+    {
+      "epoch": 0.9660862172119533,
+      "grad_norm": 0.313147634267807,
+      "learning_rate": 0.0002571755163771046,
+      "loss": 0.5697,
+      "step": 7670
+    },
+    {
+      "epoch": 0.9667159996221305,
+      "grad_norm": 0.3494894504547119,
+      "learning_rate": 0.00025709858212971545,
+      "loss": 0.5651,
+      "step": 7675
+    },
+    {
+      "epoch": 0.9673457820323078,
+      "grad_norm": 0.317107230424881,
+      "learning_rate": 0.00025702159036794135,
+      "loss": 0.5563,
+      "step": 7680
+    },
+    {
+      "epoch": 0.9679755644424851,
+      "grad_norm": 0.3228907585144043,
+      "learning_rate": 0.00025694454113312854,
+      "loss": 0.5642,
+      "step": 7685
+    },
+    {
+      "epoch": 0.9686053468526624,
+      "grad_norm": 0.33400991559028625,
+      "learning_rate": 0.00025686743446665426,
+      "loss": 0.5738,
+      "step": 7690
+    },
+    {
+      "epoch": 0.9692351292628397,
+      "grad_norm": 0.35151737928390503,
+      "learning_rate": 0.0002567902704099266,
+      "loss": 0.562,
+      "step": 7695
+    },
+    {
+      "epoch": 0.969864911673017,
+      "grad_norm": 0.33582988381385803,
+      "learning_rate": 0.00025671304900438437,
+      "loss": 0.5724,
+      "step": 7700
+    },
+    {
+      "epoch": 0.9704946940831942,
+      "grad_norm": 0.4050043523311615,
+      "learning_rate": 0.00025663577029149727,
+      "loss": 0.6038,
+      "step": 7705
+    },
+    {
+      "epoch": 0.9711244764933715,
+      "grad_norm": 0.3320407271385193,
+      "learning_rate": 0.00025655843431276565,
+      "loss": 0.5725,
+      "step": 7710
+    },
+    {
+      "epoch": 0.9717542589035488,
+      "grad_norm": 0.33253729343414307,
+      "learning_rate": 0.00025648104110972074,
+      "loss": 0.559,
+      "step": 7715
+    },
+    {
+      "epoch": 0.9723840413137261,
+      "grad_norm": 0.3316608667373657,
+      "learning_rate": 0.0002564035907239245,
+      "loss": 0.5813,
+      "step": 7720
+    },
+    {
+      "epoch": 0.9730138237239034,
+      "grad_norm": 0.35272932052612305,
+      "learning_rate": 0.0002563260831969695,
+      "loss": 0.5544,
+      "step": 7725
+    },
+    {
+      "epoch": 0.9736436061340806,
+      "grad_norm": 0.2942962348461151,
+      "learning_rate": 0.00025624851857047914,
+      "loss": 0.5741,
+      "step": 7730
+    },
+    {
+      "epoch": 0.9742733885442579,
+      "grad_norm": 0.30799049139022827,
+      "learning_rate": 0.0002561708968861073,
+      "loss": 0.5604,
+      "step": 7735
+    },
+    {
+      "epoch": 0.9749031709544352,
+      "grad_norm": 0.2929095923900604,
+      "learning_rate": 0.00025609321818553864,
+      "loss": 0.5399,
+      "step": 7740
+    },
+    {
+      "epoch": 0.9755329533646125,
+      "grad_norm": 0.3074556291103363,
+      "learning_rate": 0.00025601548251048833,
+      "loss": 0.5714,
+      "step": 7745
+    },
+    {
+      "epoch": 0.9761627357747898,
+      "grad_norm": 0.3233494162559509,
+      "learning_rate": 0.0002559376899027024,
+      "loss": 0.5559,
+      "step": 7750
+    },
+    {
+      "epoch": 0.976792518184967,
+      "grad_norm": 0.3106531500816345,
+      "learning_rate": 0.000255859840403957,
+      "loss": 0.5462,
+      "step": 7755
+    },
+    {
+      "epoch": 0.9774223005951443,
+      "grad_norm": 0.35069772601127625,
+      "learning_rate": 0.00025578193405605923,
+      "loss": 0.5635,
+      "step": 7760
+    },
+    {
+      "epoch": 0.9780520830053216,
+      "grad_norm": 0.310811311006546,
+      "learning_rate": 0.00025570397090084656,
+      "loss": 0.5658,
+      "step": 7765
+    },
+    {
+      "epoch": 0.978681865415499,
+      "grad_norm": 0.36216944456100464,
+      "learning_rate": 0.000255625950980187,
+      "loss": 0.5785,
+      "step": 7770
+    },
+    {
+      "epoch": 0.9793116478256763,
+      "grad_norm": 0.30353617668151855,
+      "learning_rate": 0.000255547874335979,
+      "loss": 0.5347,
+      "step": 7775
+    },
+    {
+      "epoch": 0.9799414302358536,
+      "grad_norm": 0.3112618029117584,
+      "learning_rate": 0.00025546974101015154,
+      "loss": 0.5559,
+      "step": 7780
+    },
+    {
+      "epoch": 0.9805712126460308,
+      "grad_norm": 0.3782903552055359,
+      "learning_rate": 0.00025539155104466394,
+      "loss": 0.5717,
+      "step": 7785
+    },
+    {
+      "epoch": 0.9812009950562081,
+      "grad_norm": 0.3308548331260681,
+      "learning_rate": 0.000255313304481506,
+      "loss": 0.5511,
+      "step": 7790
+    },
+    {
+      "epoch": 0.9818307774663854,
+      "grad_norm": 0.2971625328063965,
+      "learning_rate": 0.000255235001362698,
+      "loss": 0.5411,
+      "step": 7795
+    },
+    {
+      "epoch": 0.9824605598765627,
+      "grad_norm": 0.3594948351383209,
+      "learning_rate": 0.0002551566417302904,
+      "loss": 0.5817,
+      "step": 7800
+    },
+    {
+      "epoch": 0.98309034228674,
+      "grad_norm": 0.3537582755088806,
+      "learning_rate": 0.0002550782256263642,
+      "loss": 0.5631,
+      "step": 7805
+    },
+    {
+      "epoch": 0.9837201246969172,
+      "grad_norm": 0.3132795989513397,
+      "learning_rate": 0.0002549997530930306,
+      "loss": 0.5725,
+      "step": 7810
+    },
+    {
+      "epoch": 0.9843499071070945,
+      "grad_norm": 0.3250652551651001,
+      "learning_rate": 0.00025492122417243113,
+      "loss": 0.5786,
+      "step": 7815
+    },
+    {
+      "epoch": 0.9849796895172718,
+      "grad_norm": 0.3318973183631897,
+      "learning_rate": 0.0002548426389067376,
+      "loss": 0.5399,
+      "step": 7820
+    },
+    {
+      "epoch": 0.9856094719274491,
+      "grad_norm": 0.3335192799568176,
+      "learning_rate": 0.00025476399733815214,
+      "loss": 0.5693,
+      "step": 7825
+    },
+    {
+      "epoch": 0.9862392543376264,
+      "grad_norm": 0.31399449706077576,
+      "learning_rate": 0.00025468529950890703,
+      "loss": 0.5821,
+      "step": 7830
+    },
+    {
+      "epoch": 0.9868690367478037,
+      "grad_norm": 0.33886855840682983,
+      "learning_rate": 0.00025460654546126485,
+      "loss": 0.556,
+      "step": 7835
+    },
+    {
+      "epoch": 0.9874988191579809,
+      "grad_norm": 0.3620472848415375,
+      "learning_rate": 0.0002545277352375183,
+      "loss": 0.6104,
+      "step": 7840
+    },
+    {
+      "epoch": 0.9881286015681582,
+      "grad_norm": 0.31123921275138855,
+      "learning_rate": 0.0002544488688799902,
+      "loss": 0.5802,
+      "step": 7845
+    },
+    {
+      "epoch": 0.9887583839783355,
+      "grad_norm": 0.33104339241981506,
+      "learning_rate": 0.0002543699464310337,
+      "loss": 0.5882,
+      "step": 7850
+    },
+    {
+      "epoch": 0.9893881663885128,
+      "grad_norm": 0.3223660886287689,
+      "learning_rate": 0.00025429096793303186,
+      "loss": 0.5649,
+      "step": 7855
+    },
+    {
+      "epoch": 0.9900179487986901,
+      "grad_norm": 0.3436056077480316,
+      "learning_rate": 0.000254211933428398,
+      "loss": 0.5546,
+      "step": 7860
+    },
+    {
+      "epoch": 0.9906477312088673,
+      "grad_norm": 0.29697200655937195,
+      "learning_rate": 0.00025413284295957547,
+      "loss": 0.5434,
+      "step": 7865
+    },
+    {
+      "epoch": 0.9912775136190446,
+      "grad_norm": 0.32985180616378784,
+      "learning_rate": 0.0002540536965690376,
+      "loss": 0.5737,
+      "step": 7870
+    },
+    {
+      "epoch": 0.9919072960292219,
+      "grad_norm": 0.31599709391593933,
+      "learning_rate": 0.0002539744942992878,
+      "loss": 0.5452,
+      "step": 7875
+    },
+    {
+      "epoch": 0.9925370784393992,
+      "grad_norm": 0.30331170558929443,
+      "learning_rate": 0.00025389523619285956,
+      "loss": 0.5593,
+      "step": 7880
+    },
+    {
+      "epoch": 0.9931668608495765,
+      "grad_norm": 0.3150465786457062,
+      "learning_rate": 0.0002538159222923163,
+      "loss": 0.5518,
+      "step": 7885
+    },
+    {
+      "epoch": 0.9937966432597538,
+      "grad_norm": 0.3179359436035156,
+      "learning_rate": 0.00025373655264025134,
+      "loss": 0.5546,
+      "step": 7890
+    },
+    {
+      "epoch": 0.994426425669931,
+      "grad_norm": 0.3226470947265625,
+      "learning_rate": 0.000253657127279288,
+      "loss": 0.58,
+      "step": 7895
+    },
+    {
+      "epoch": 0.9950562080801083,
+      "grad_norm": 0.3453287184238434,
+      "learning_rate": 0.0002535776462520795,
+      "loss": 0.5681,
+      "step": 7900
+    },
+    {
+      "epoch": 0.9956859904902856,
+      "grad_norm": 0.3329002261161804,
+      "learning_rate": 0.0002534981096013091,
+      "loss": 0.548,
+      "step": 7905
+    },
+    {
+      "epoch": 0.9963157729004629,
+      "grad_norm": 0.32592061161994934,
+      "learning_rate": 0.00025341851736968956,
+      "loss": 0.5244,
+      "step": 7910
+    },
+    {
+      "epoch": 0.9969455553106402,
+      "grad_norm": 0.32833319902420044,
+      "learning_rate": 0.00025333886959996396,
+      "loss": 0.5558,
+      "step": 7915
+    },
+    {
+      "epoch": 0.9975753377208174,
+      "grad_norm": 0.3146878182888031,
+      "learning_rate": 0.00025325916633490487,
+      "loss": 0.595,
+      "step": 7920
+    },
+    {
+      "epoch": 0.9982051201309947,
+      "grad_norm": 0.3828830122947693,
+      "learning_rate": 0.00025317940761731476,
+      "loss": 0.5675,
+      "step": 7925
+    },
+    {
+      "epoch": 0.998834902541172,
+      "grad_norm": 0.3208398222923279,
+      "learning_rate": 0.0002530995934900259,
+      "loss": 0.5439,
+      "step": 7930
+    },
+    {
+      "epoch": 0.9994646849513493,
+      "grad_norm": 0.3446502983570099,
+      "learning_rate": 0.00025301972399590023,
+      "loss": 0.5276,
+      "step": 7935
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.31275373697280884,
+      "learning_rate": 0.0002529397991778297,
+      "loss": 0.543,
+      "step": 7940
+    },
+    {
+      "epoch": 1.0006297824101773,
+      "grad_norm": 0.3219754099845886,
+      "learning_rate": 0.0002528598190787355,
+      "loss": 0.4901,
+      "step": 7945
+    },
+    {
+      "epoch": 1.0012595648203546,
+      "grad_norm": 0.33292412757873535,
+      "learning_rate": 0.0002527797837415689,
+      "loss": 0.4794,
+      "step": 7950
+    },
+    {
+      "epoch": 1.001889347230532,
+      "grad_norm": 0.36561062932014465,
+      "learning_rate": 0.00025269969320931065,
+      "loss": 0.4948,
+      "step": 7955
+    },
+    {
+      "epoch": 1.0025191296407092,
+      "grad_norm": 0.2977091372013092,
+      "learning_rate": 0.0002526195475249713,
+      "loss": 0.5172,
+      "step": 7960
+    },
+    {
+      "epoch": 1.0031489120508865,
+      "grad_norm": 0.3075500428676605,
+      "learning_rate": 0.00025253934673159084,
+      "loss": 0.4755,
+      "step": 7965
+    },
+    {
+      "epoch": 1.0037786944610636,
+      "grad_norm": 0.30956047773361206,
+      "learning_rate": 0.00025245909087223895,
+      "loss": 0.4783,
+      "step": 7970
+    },
+    {
+      "epoch": 1.004408476871241,
+      "grad_norm": 0.34965232014656067,
+      "learning_rate": 0.00025237877999001484,
+      "loss": 0.4876,
+      "step": 7975
+    },
+    {
+      "epoch": 1.0050382592814182,
+      "grad_norm": 0.3290039896965027,
+      "learning_rate": 0.00025229841412804726,
+      "loss": 0.501,
+      "step": 7980
+    },
+    {
+      "epoch": 1.0056680416915955,
+      "grad_norm": 0.3144761323928833,
+      "learning_rate": 0.00025221799332949456,
+      "loss": 0.4923,
+      "step": 7985
+    },
+    {
+      "epoch": 1.0062978241017728,
+      "grad_norm": 0.3586188554763794,
+      "learning_rate": 0.0002521375176375446,
+      "loss": 0.487,
+      "step": 7990
+    },
+    {
+      "epoch": 1.0069276065119501,
+      "grad_norm": 0.3210572302341461,
+      "learning_rate": 0.0002520569870954146,
+      "loss": 0.4916,
+      "step": 7995
+    },
+    {
+      "epoch": 1.0075573889221274,
+      "grad_norm": 0.3171830177307129,
+      "learning_rate": 0.0002519764017463512,
+      "loss": 0.4834,
+      "step": 8000
+    },
+    {
+      "epoch": 1.0075573889221274,
+      "eval_loss": 0.30723655223846436,
+      "eval_runtime": 6.2539,
+      "eval_samples_per_second": 159.899,
+      "eval_steps_per_second": 10.074,
+      "step": 8000
+    },
+    {
+      "epoch": 1.0081871713323047,
+      "grad_norm": 0.3511858880519867,
+      "learning_rate": 0.00025189576163363076,
+      "loss": 0.4937,
+      "step": 8005
+    },
+    {
+      "epoch": 1.008816953742482,
+      "grad_norm": 0.3305964171886444,
+      "learning_rate": 0.00025181506680055875,
+      "loss": 0.4665,
+      "step": 8010
+    },
+    {
+      "epoch": 1.0094467361526593,
+      "grad_norm": 0.3735099732875824,
+      "learning_rate": 0.00025173431729047014,
+      "loss": 0.5116,
+      "step": 8015
+    },
+    {
+      "epoch": 1.0100765185628366,
+      "grad_norm": 0.34169599413871765,
+      "learning_rate": 0.0002516535131467293,
+      "loss": 0.475,
+      "step": 8020
+    },
+    {
+      "epoch": 1.0107063009730137,
+      "grad_norm": 0.3473950922489166,
+      "learning_rate": 0.00025157265441272993,
+      "loss": 0.4812,
+      "step": 8025
+    },
+    {
+      "epoch": 1.011336083383191,
+      "grad_norm": 0.31877681612968445,
+      "learning_rate": 0.00025149174113189496,
+      "loss": 0.4906,
+      "step": 8030
+    },
+    {
+      "epoch": 1.0119658657933683,
+      "grad_norm": 0.364511638879776,
+      "learning_rate": 0.0002514107733476766,
+      "loss": 0.4926,
+      "step": 8035
+    },
+    {
+      "epoch": 1.0125956482035456,
+      "grad_norm": 0.3073696792125702,
+      "learning_rate": 0.00025132975110355664,
+      "loss": 0.4994,
+      "step": 8040
+    },
+    {
+      "epoch": 1.013225430613723,
+      "grad_norm": 0.3270637094974518,
+      "learning_rate": 0.0002512486744430456,
+      "loss": 0.468,
+      "step": 8045
+    },
+    {
+      "epoch": 1.0138552130239002,
+      "grad_norm": 0.3626968264579773,
+      "learning_rate": 0.0002511675434096837,
+      "loss": 0.5139,
+      "step": 8050
+    },
+    {
+      "epoch": 1.0144849954340776,
+      "grad_norm": 0.30527931451797485,
+      "learning_rate": 0.00025108635804704,
+      "loss": 0.4922,
+      "step": 8055
+    },
+    {
+      "epoch": 1.0151147778442549,
+      "grad_norm": 0.3518252968788147,
+      "learning_rate": 0.000251005118398713,
+      "loss": 0.5297,
+      "step": 8060
+    },
+    {
+      "epoch": 1.0157445602544322,
+      "grad_norm": 0.3298850655555725,
+      "learning_rate": 0.0002509238245083302,
+      "loss": 0.5292,
+      "step": 8065
+    },
+    {
+      "epoch": 1.0163743426646095,
+      "grad_norm": 0.3175168037414551,
+      "learning_rate": 0.0002508424764195484,
+      "loss": 0.4907,
+      "step": 8070
+    },
+    {
+      "epoch": 1.0170041250747868,
+      "grad_norm": 0.33489352464675903,
+      "learning_rate": 0.0002507610741760531,
+      "loss": 0.4869,
+      "step": 8075
+    },
+    {
+      "epoch": 1.0176339074849639,
+      "grad_norm": 0.2922315299510956,
+      "learning_rate": 0.0002506796178215595,
+      "loss": 0.474,
+      "step": 8080
+    },
+    {
+      "epoch": 1.0182636898951412,
+      "grad_norm": 0.32073619961738586,
+      "learning_rate": 0.00025059810739981125,
+      "loss": 0.4951,
+      "step": 8085
+    },
+    {
+      "epoch": 1.0188934723053185,
+      "grad_norm": 0.2875652611255646,
+      "learning_rate": 0.0002505165429545815,
+      "loss": 0.5104,
+      "step": 8090
+    },
+    {
+      "epoch": 1.0195232547154958,
+      "grad_norm": 0.33247148990631104,
+      "learning_rate": 0.0002504349245296721,
+      "loss": 0.489,
+      "step": 8095
+    },
+    {
+      "epoch": 1.020153037125673,
+      "grad_norm": 0.29777953028678894,
+      "learning_rate": 0.0002503532521689141,
+      "loss": 0.5172,
+      "step": 8100
+    },
+    {
+      "epoch": 1.0207828195358504,
+      "grad_norm": 0.3418375253677368,
+      "learning_rate": 0.0002502715259161673,
+      "loss": 0.4464,
+      "step": 8105
+    },
+    {
+      "epoch": 1.0214126019460277,
+      "grad_norm": 0.39162155985832214,
+      "learning_rate": 0.0002501897458153207,
+      "loss": 0.4953,
+      "step": 8110
+    },
+    {
+      "epoch": 1.022042384356205,
+      "grad_norm": 0.32206737995147705,
+      "learning_rate": 0.000250107911910292,
+      "loss": 0.4732,
+      "step": 8115
+    },
+    {
+      "epoch": 1.0226721667663823,
+      "grad_norm": 0.37178757786750793,
+      "learning_rate": 0.0002500260242450279,
+      "loss": 0.504,
+      "step": 8120
+    },
+    {
+      "epoch": 1.0233019491765596,
+      "grad_norm": 0.33448055386543274,
+      "learning_rate": 0.0002499440828635039,
+      "loss": 0.4774,
+      "step": 8125
+    },
+    {
+      "epoch": 1.0239317315867367,
+      "grad_norm": 0.344594806432724,
+      "learning_rate": 0.00024986208780972455,
+      "loss": 0.4948,
+      "step": 8130
+    },
+    {
+      "epoch": 1.024561513996914,
+      "grad_norm": 0.3440978527069092,
+      "learning_rate": 0.00024978003912772283,
+      "loss": 0.4979,
+      "step": 8135
+    },
+    {
+      "epoch": 1.0251912964070913,
+      "grad_norm": 0.2915257513523102,
+      "learning_rate": 0.000249697936861561,
+      "loss": 0.4875,
+      "step": 8140
+    },
+    {
+      "epoch": 1.0258210788172686,
+      "grad_norm": 0.271371990442276,
+      "learning_rate": 0.0002496157810553296,
+      "loss": 0.4929,
+      "step": 8145
+    },
+    {
+      "epoch": 1.0264508612274459,
+      "grad_norm": 0.3228522539138794,
+      "learning_rate": 0.0002495335717531484,
+      "loss": 0.4706,
+      "step": 8150
+    },
+    {
+      "epoch": 1.0270806436376232,
+      "grad_norm": 0.3222556412220001,
+      "learning_rate": 0.00024945130899916554,
+      "loss": 0.487,
+      "step": 8155
+    },
+    {
+      "epoch": 1.0277104260478005,
+      "grad_norm": 0.32311001420021057,
+      "learning_rate": 0.00024936899283755807,
+      "loss": 0.5144,
+      "step": 8160
+    },
+    {
+      "epoch": 1.0283402084579778,
+      "grad_norm": 0.2946212589740753,
+      "learning_rate": 0.0002492866233125316,
+      "loss": 0.4867,
+      "step": 8165
+    },
+    {
+      "epoch": 1.028969990868155,
+      "grad_norm": 0.32464465498924255,
+      "learning_rate": 0.0002492042004683205,
+      "loss": 0.4729,
+      "step": 8170
+    },
+    {
+      "epoch": 1.0295997732783324,
+      "grad_norm": 0.3378526270389557,
+      "learning_rate": 0.0002491217243491876,
+      "loss": 0.4843,
+      "step": 8175
+    },
+    {
+      "epoch": 1.0302295556885097,
+      "grad_norm": 0.35685908794403076,
+      "learning_rate": 0.0002490391949994246,
+      "loss": 0.4941,
+      "step": 8180
+    },
+    {
+      "epoch": 1.0308593380986868,
+      "grad_norm": 0.30618053674697876,
+      "learning_rate": 0.0002489566124633516,
+      "loss": 0.4985,
+      "step": 8185
+    },
+    {
+      "epoch": 1.031489120508864,
+      "grad_norm": 0.34786808490753174,
+      "learning_rate": 0.0002488739767853173,
+      "loss": 0.4914,
+      "step": 8190
+    },
+    {
+      "epoch": 1.0321189029190414,
+      "grad_norm": 0.35167476534843445,
+      "learning_rate": 0.00024879128800969893,
+      "loss": 0.493,
+      "step": 8195
+    },
+    {
+      "epoch": 1.0327486853292187,
+      "grad_norm": 0.3278263509273529,
+      "learning_rate": 0.00024870854618090225,
+      "loss": 0.4676,
+      "step": 8200
+    },
+    {
+      "epoch": 1.033378467739396,
+      "grad_norm": 0.36896881461143494,
+      "learning_rate": 0.00024862575134336154,
+      "loss": 0.4995,
+      "step": 8205
+    },
+    {
+      "epoch": 1.0340082501495733,
+      "grad_norm": 0.3700760304927826,
+      "learning_rate": 0.00024854290354153953,
+      "loss": 0.5189,
+      "step": 8210
+    },
+    {
+      "epoch": 1.0346380325597506,
+      "grad_norm": 0.3370974063873291,
+      "learning_rate": 0.00024846000281992733,
+      "loss": 0.5044,
+      "step": 8215
+    },
+    {
+      "epoch": 1.035267814969928,
+      "grad_norm": 0.3200768232345581,
+      "learning_rate": 0.00024837704922304457,
+      "loss": 0.4779,
+      "step": 8220
+    },
+    {
+      "epoch": 1.0358975973801052,
+      "grad_norm": 0.2786978781223297,
+      "learning_rate": 0.0002482940427954392,
+      "loss": 0.4677,
+      "step": 8225
+    },
+    {
+      "epoch": 1.0365273797902825,
+      "grad_norm": 0.3220120668411255,
+      "learning_rate": 0.00024821098358168757,
+      "loss": 0.4503,
+      "step": 8230
+    },
+    {
+      "epoch": 1.0371571622004598,
+      "grad_norm": 0.3315715491771698,
+      "learning_rate": 0.00024812787162639444,
+      "loss": 0.4715,
+      "step": 8235
+    },
+    {
+      "epoch": 1.037786944610637,
+      "grad_norm": 0.3595867455005646,
+      "learning_rate": 0.00024804470697419273,
+      "loss": 0.4712,
+      "step": 8240
+    },
+    {
+      "epoch": 1.0384167270208142,
+      "grad_norm": 0.29993361234664917,
+      "learning_rate": 0.00024796148966974376,
+      "loss": 0.47,
+      "step": 8245
+    },
+    {
+      "epoch": 1.0390465094309915,
+      "grad_norm": 0.39950379729270935,
+      "learning_rate": 0.00024787821975773717,
+      "loss": 0.5233,
+      "step": 8250
+    },
+    {
+      "epoch": 1.0396762918411688,
+      "grad_norm": 0.312003493309021,
+      "learning_rate": 0.0002477948972828908,
+      "loss": 0.4836,
+      "step": 8255
+    },
+    {
+      "epoch": 1.0403060742513461,
+      "grad_norm": 0.29678481817245483,
+      "learning_rate": 0.0002477115222899507,
+      "loss": 0.4928,
+      "step": 8260
+    },
+    {
+      "epoch": 1.0409358566615234,
+      "grad_norm": 0.35694456100463867,
+      "learning_rate": 0.0002476280948236912,
+      "loss": 0.4925,
+      "step": 8265
+    },
+    {
+      "epoch": 1.0415656390717007,
+      "grad_norm": 0.3164297640323639,
+      "learning_rate": 0.00024754461492891474,
+      "loss": 0.4828,
+      "step": 8270
+    },
+    {
+      "epoch": 1.042195421481878,
+      "grad_norm": 0.37906938791275024,
+      "learning_rate": 0.00024746108265045184,
+      "loss": 0.4989,
+      "step": 8275
+    },
+    {
+      "epoch": 1.0428252038920554,
+      "grad_norm": 0.3458475172519684,
+      "learning_rate": 0.0002473774980331614,
+      "loss": 0.5072,
+      "step": 8280
+    },
+    {
+      "epoch": 1.0434549863022327,
+      "grad_norm": 0.36052700877189636,
+      "learning_rate": 0.0002472938611219301,
+      "loss": 0.4872,
+      "step": 8285
+    },
+    {
+      "epoch": 1.04408476871241,
+      "grad_norm": 0.4497036337852478,
+      "learning_rate": 0.00024721017196167297,
+      "loss": 0.4921,
+      "step": 8290
+    },
+    {
+      "epoch": 1.044714551122587,
+      "grad_norm": 0.357461154460907,
+      "learning_rate": 0.000247126430597333,
+      "loss": 0.5035,
+      "step": 8295
+    },
+    {
+      "epoch": 1.0453443335327643,
+      "grad_norm": 0.3499346375465393,
+      "learning_rate": 0.00024704263707388117,
+      "loss": 0.5242,
+      "step": 8300
+    },
+    {
+      "epoch": 1.0459741159429417,
+      "grad_norm": 0.2994784414768219,
+      "learning_rate": 0.0002469587914363166,
+      "loss": 0.4575,
+      "step": 8305
+    },
+    {
+      "epoch": 1.046603898353119,
+      "grad_norm": 0.3699876666069031,
+      "learning_rate": 0.0002468748937296662,
+      "loss": 0.4804,
+      "step": 8310
+    },
+    {
+      "epoch": 1.0472336807632963,
+      "grad_norm": 0.37695133686065674,
+      "learning_rate": 0.000246790943998985,
+      "loss": 0.4914,
+      "step": 8315
+    },
+    {
+      "epoch": 1.0478634631734736,
+      "grad_norm": 0.30732589960098267,
+      "learning_rate": 0.0002467069422893559,
+      "loss": 0.458,
+      "step": 8320
+    },
+    {
+      "epoch": 1.0484932455836509,
+      "grad_norm": 0.3094361424446106,
+      "learning_rate": 0.0002466228886458899,
+      "loss": 0.4584,
+      "step": 8325
+    },
+    {
+      "epoch": 1.0491230279938282,
+      "grad_norm": 0.3499257564544678,
+      "learning_rate": 0.0002465387831137255,
+      "loss": 0.4717,
+      "step": 8330
+    },
+    {
+      "epoch": 1.0497528104040055,
+      "grad_norm": 0.32755059003829956,
+      "learning_rate": 0.0002464546257380294,
+      "loss": 0.49,
+      "step": 8335
+    },
+    {
+      "epoch": 1.0503825928141828,
+      "grad_norm": 0.3201046884059906,
+      "learning_rate": 0.000246370416563996,
+      "loss": 0.4833,
+      "step": 8340
+    },
+    {
+      "epoch": 1.05101237522436,
+      "grad_norm": 0.2581581771373749,
+      "learning_rate": 0.0002462861556368476,
+      "loss": 0.465,
+      "step": 8345
+    },
+    {
+      "epoch": 1.0516421576345372,
+      "grad_norm": 0.3480297923088074,
+      "learning_rate": 0.00024620184300183423,
+      "loss": 0.4756,
+      "step": 8350
+    },
+    {
+      "epoch": 1.0522719400447145,
+      "grad_norm": 0.36630478501319885,
+      "learning_rate": 0.00024611747870423366,
+      "loss": 0.5051,
+      "step": 8355
+    },
+    {
+      "epoch": 1.0529017224548918,
+      "grad_norm": 0.3450157940387726,
+      "learning_rate": 0.0002460330627893515,
+      "loss": 0.4996,
+      "step": 8360
+    },
+    {
+      "epoch": 1.053531504865069,
+      "grad_norm": 0.30790945887565613,
+      "learning_rate": 0.000245948595302521,
+      "loss": 0.4826,
+      "step": 8365
+    },
+    {
+      "epoch": 1.0541612872752464,
+      "grad_norm": 0.39590683579444885,
+      "learning_rate": 0.00024586407628910306,
+      "loss": 0.4963,
+      "step": 8370
+    },
+    {
+      "epoch": 1.0547910696854237,
+      "grad_norm": 0.3294634521007538,
+      "learning_rate": 0.00024577950579448643,
+      "loss": 0.4868,
+      "step": 8375
+    },
+    {
+      "epoch": 1.055420852095601,
+      "grad_norm": 0.33493947982788086,
+      "learning_rate": 0.00024569488386408736,
+      "loss": 0.4773,
+      "step": 8380
+    },
+    {
+      "epoch": 1.0560506345057783,
+      "grad_norm": 0.32626229524612427,
+      "learning_rate": 0.00024561021054334974,
+      "loss": 0.4898,
+      "step": 8385
+    },
+    {
+      "epoch": 1.0566804169159556,
+      "grad_norm": 0.3181340992450714,
+      "learning_rate": 0.00024552548587774507,
+      "loss": 0.4757,
+      "step": 8390
+    },
+    {
+      "epoch": 1.057310199326133,
+      "grad_norm": 0.3592873215675354,
+      "learning_rate": 0.0002454407099127725,
+      "loss": 0.5034,
+      "step": 8395
+    },
+    {
+      "epoch": 1.0579399817363102,
+      "grad_norm": 0.3184007406234741,
+      "learning_rate": 0.00024535588269395856,
+      "loss": 0.4929,
+      "step": 8400
+    },
+    {
+      "epoch": 1.0585697641464873,
+      "grad_norm": 0.3555738627910614,
+      "learning_rate": 0.00024527100426685746,
+      "loss": 0.503,
+      "step": 8405
+    },
+    {
+      "epoch": 1.0591995465566646,
+      "grad_norm": 0.29637908935546875,
+      "learning_rate": 0.0002451860746770509,
+      "loss": 0.4716,
+      "step": 8410
+    },
+    {
+      "epoch": 1.059829328966842,
+      "grad_norm": 0.3031441569328308,
+      "learning_rate": 0.0002451010939701479,
+      "loss": 0.4757,
+      "step": 8415
+    },
+    {
+      "epoch": 1.0604591113770192,
+      "grad_norm": 0.28256094455718994,
+      "learning_rate": 0.0002450160621917851,
+      "loss": 0.4558,
+      "step": 8420
+    },
+    {
+      "epoch": 1.0610888937871965,
+      "grad_norm": 0.3192931115627289,
+      "learning_rate": 0.0002449309793876266,
+      "loss": 0.499,
+      "step": 8425
+    },
+    {
+      "epoch": 1.0617186761973738,
+      "grad_norm": 0.2788430154323578,
+      "learning_rate": 0.00024484584560336363,
+      "loss": 0.4616,
+      "step": 8430
+    },
+    {
+      "epoch": 1.0623484586075511,
+      "grad_norm": 0.35733649134635925,
+      "learning_rate": 0.00024476066088471507,
+      "loss": 0.4926,
+      "step": 8435
+    },
+    {
+      "epoch": 1.0629782410177284,
+      "grad_norm": 0.3398718535900116,
+      "learning_rate": 0.00024467542527742707,
+      "loss": 0.4944,
+      "step": 8440
+    },
+    {
+      "epoch": 1.0636080234279057,
+      "grad_norm": 0.3338175117969513,
+      "learning_rate": 0.000244590138827273,
+      "loss": 0.5181,
+      "step": 8445
+    },
+    {
+      "epoch": 1.064237805838083,
+      "grad_norm": 0.33433952927589417,
+      "learning_rate": 0.00024450480158005384,
+      "loss": 0.4837,
+      "step": 8450
+    },
+    {
+      "epoch": 1.0648675882482603,
+      "grad_norm": 0.3656097948551178,
+      "learning_rate": 0.0002444194135815974,
+      "loss": 0.4639,
+      "step": 8455
+    },
+    {
+      "epoch": 1.0654973706584374,
+      "grad_norm": 0.31470635533332825,
+      "learning_rate": 0.0002443339748777592,
+      "loss": 0.4718,
+      "step": 8460
+    },
+    {
+      "epoch": 1.0661271530686147,
+      "grad_norm": 0.29020166397094727,
+      "learning_rate": 0.00024424848551442166,
+      "loss": 0.4712,
+      "step": 8465
+    },
+    {
+      "epoch": 1.066756935478792,
+      "grad_norm": 0.34259042143821716,
+      "learning_rate": 0.00024416294553749446,
+      "loss": 0.5252,
+      "step": 8470
+    },
+    {
+      "epoch": 1.0673867178889693,
+      "grad_norm": 0.33828607201576233,
+      "learning_rate": 0.0002440773549929146,
+      "loss": 0.4663,
+      "step": 8475
+    },
+    {
+      "epoch": 1.0680165002991466,
+      "grad_norm": 0.35722973942756653,
+      "learning_rate": 0.00024399171392664622,
+      "loss": 0.4868,
+      "step": 8480
+    },
+    {
+      "epoch": 1.068646282709324,
+      "grad_norm": 0.3226557970046997,
+      "learning_rate": 0.00024390602238468043,
+      "loss": 0.4785,
+      "step": 8485
+    },
+    {
+      "epoch": 1.0692760651195012,
+      "grad_norm": 0.3097434639930725,
+      "learning_rate": 0.0002438202804130356,
+      "loss": 0.4677,
+      "step": 8490
+    },
+    {
+      "epoch": 1.0699058475296785,
+      "grad_norm": 0.3146856725215912,
+      "learning_rate": 0.00024373448805775709,
+      "loss": 0.4802,
+      "step": 8495
+    },
+    {
+      "epoch": 1.0705356299398558,
+      "grad_norm": 0.3576582372188568,
+      "learning_rate": 0.00024364864536491739,
+      "loss": 0.5113,
+      "step": 8500
+    },
+    {
+      "epoch": 1.0711654123500332,
+      "grad_norm": 0.33004313707351685,
+      "learning_rate": 0.0002435627523806159,
+      "loss": 0.4625,
+      "step": 8505
+    },
+    {
+      "epoch": 1.0717951947602105,
+      "grad_norm": 0.3689037263393402,
+      "learning_rate": 0.00024347680915097928,
+      "loss": 0.4923,
+      "step": 8510
+    },
+    {
+      "epoch": 1.0724249771703875,
+      "grad_norm": 0.28334125876426697,
+      "learning_rate": 0.00024339081572216084,
+      "loss": 0.4818,
+      "step": 8515
+    },
+    {
+      "epoch": 1.0730547595805648,
+      "grad_norm": 0.3461993336677551,
+      "learning_rate": 0.00024330477214034113,
+      "loss": 0.4807,
+      "step": 8520
+    },
+    {
+      "epoch": 1.0736845419907421,
+      "grad_norm": 0.32148951292037964,
+      "learning_rate": 0.00024321867845172743,
+      "loss": 0.4829,
+      "step": 8525
+    },
+    {
+      "epoch": 1.0743143244009195,
+      "grad_norm": 0.31461793184280396,
+      "learning_rate": 0.0002431325347025541,
+      "loss": 0.5045,
+      "step": 8530
+    },
+    {
+      "epoch": 1.0749441068110968,
+      "grad_norm": 0.30194273591041565,
+      "learning_rate": 0.00024304634093908224,
+      "loss": 0.4747,
+      "step": 8535
+    },
+    {
+      "epoch": 1.075573889221274,
+      "grad_norm": 0.27379968762397766,
+      "learning_rate": 0.0002429600972075999,
+      "loss": 0.4382,
+      "step": 8540
+    },
+    {
+      "epoch": 1.0762036716314514,
+      "grad_norm": 0.3732368052005768,
+      "learning_rate": 0.0002428738035544219,
+      "loss": 0.4704,
+      "step": 8545
+    },
+    {
+      "epoch": 1.0768334540416287,
+      "grad_norm": 0.3252260088920593,
+      "learning_rate": 0.00024278746002588997,
+      "loss": 0.4929,
+      "step": 8550
+    },
+    {
+      "epoch": 1.077463236451806,
+      "grad_norm": 0.31606802344322205,
+      "learning_rate": 0.00024270106666837246,
+      "loss": 0.4698,
+      "step": 8555
+    },
+    {
+      "epoch": 1.0780930188619833,
+      "grad_norm": 0.320529967546463,
+      "learning_rate": 0.00024261462352826468,
+      "loss": 0.4531,
+      "step": 8560
+    },
+    {
+      "epoch": 1.0787228012721606,
+      "grad_norm": 0.36827871203422546,
+      "learning_rate": 0.00024252813065198852,
+      "loss": 0.4948,
+      "step": 8565
+    },
+    {
+      "epoch": 1.0793525836823377,
+      "grad_norm": 0.3132867217063904,
+      "learning_rate": 0.00024244158808599264,
+      "loss": 0.4836,
+      "step": 8570
+    },
+    {
+      "epoch": 1.079982366092515,
+      "grad_norm": 0.32383888959884644,
+      "learning_rate": 0.00024235499587675236,
+      "loss": 0.4749,
+      "step": 8575
+    },
+    {
+      "epoch": 1.0806121485026923,
+      "grad_norm": 0.32294297218322754,
+      "learning_rate": 0.0002422683540707697,
+      "loss": 0.4616,
+      "step": 8580
+    },
+    {
+      "epoch": 1.0812419309128696,
+      "grad_norm": 0.3049245774745941,
+      "learning_rate": 0.00024218166271457322,
+      "loss": 0.4871,
+      "step": 8585
+    },
+    {
+      "epoch": 1.0818717133230469,
+      "grad_norm": 0.3330252170562744,
+      "learning_rate": 0.00024209492185471826,
+      "loss": 0.4908,
+      "step": 8590
+    },
+    {
+      "epoch": 1.0825014957332242,
+      "grad_norm": 0.35933157801628113,
+      "learning_rate": 0.00024200813153778654,
+      "loss": 0.4936,
+      "step": 8595
+    },
+    {
+      "epoch": 1.0831312781434015,
+      "grad_norm": 0.345434308052063,
+      "learning_rate": 0.00024192129181038654,
+      "loss": 0.4637,
+      "step": 8600
+    },
+    {
+      "epoch": 1.0837610605535788,
+      "grad_norm": 0.3012515604496002,
+      "learning_rate": 0.0002418344027191531,
+      "loss": 0.4719,
+      "step": 8605
+    },
+    {
+      "epoch": 1.084390842963756,
+      "grad_norm": 0.3081362247467041,
+      "learning_rate": 0.0002417474643107477,
+      "loss": 0.4852,
+      "step": 8610
+    },
+    {
+      "epoch": 1.0850206253739334,
+      "grad_norm": 0.367389053106308,
+      "learning_rate": 0.00024166047663185826,
+      "loss": 0.5046,
+      "step": 8615
+    },
+    {
+      "epoch": 1.0856504077841107,
+      "grad_norm": 0.3392958641052246,
+      "learning_rate": 0.0002415734397291991,
+      "loss": 0.5087,
+      "step": 8620
+    },
+    {
+      "epoch": 1.0862801901942878,
+      "grad_norm": 0.2843685746192932,
+      "learning_rate": 0.000241486353649511,
+      "loss": 0.4722,
+      "step": 8625
+    },
+    {
+      "epoch": 1.086909972604465,
+      "grad_norm": 0.29619672894477844,
+      "learning_rate": 0.00024139921843956128,
+      "loss": 0.4645,
+      "step": 8630
+    },
+    {
+      "epoch": 1.0875397550146424,
+      "grad_norm": 0.30029621720314026,
+      "learning_rate": 0.00024131203414614347,
+      "loss": 0.4434,
+      "step": 8635
+    },
+    {
+      "epoch": 1.0881695374248197,
+      "grad_norm": 0.3630850911140442,
+      "learning_rate": 0.00024122480081607755,
+      "loss": 0.4772,
+      "step": 8640
+    },
+    {
+      "epoch": 1.088799319834997,
+      "grad_norm": 0.32482001185417175,
+      "learning_rate": 0.00024113751849620974,
+      "loss": 0.4441,
+      "step": 8645
+    },
+    {
+      "epoch": 1.0894291022451743,
+      "grad_norm": 0.3149590492248535,
+      "learning_rate": 0.00024105018723341275,
+      "loss": 0.468,
+      "step": 8650
+    },
+    {
+      "epoch": 1.0900588846553516,
+      "grad_norm": 0.34652113914489746,
+      "learning_rate": 0.0002409628070745854,
+      "loss": 0.4706,
+      "step": 8655
+    },
+    {
+      "epoch": 1.090688667065529,
+      "grad_norm": 0.31633374094963074,
+      "learning_rate": 0.00024087537806665279,
+      "loss": 0.4693,
+      "step": 8660
+    },
+    {
+      "epoch": 1.0913184494757062,
+      "grad_norm": 0.31668806076049805,
+      "learning_rate": 0.00024078790025656638,
+      "loss": 0.4619,
+      "step": 8665
+    },
+    {
+      "epoch": 1.0919482318858835,
+      "grad_norm": 0.3093356490135193,
+      "learning_rate": 0.00024070037369130375,
+      "loss": 0.485,
+      "step": 8670
+    },
+    {
+      "epoch": 1.0925780142960608,
+      "grad_norm": 0.31765609979629517,
+      "learning_rate": 0.0002406127984178686,
+      "loss": 0.4696,
+      "step": 8675
+    },
+    {
+      "epoch": 1.093207796706238,
+      "grad_norm": 0.35910454392433167,
+      "learning_rate": 0.00024052517448329086,
+      "loss": 0.4781,
+      "step": 8680
+    },
+    {
+      "epoch": 1.0938375791164152,
+      "grad_norm": 0.37290528416633606,
+      "learning_rate": 0.00024043750193462665,
+      "loss": 0.4824,
+      "step": 8685
+    },
+    {
+      "epoch": 1.0944673615265925,
+      "grad_norm": 0.3106020390987396,
+      "learning_rate": 0.00024034978081895807,
+      "loss": 0.4607,
+      "step": 8690
+    },
+    {
+      "epoch": 1.0950971439367698,
+      "grad_norm": 0.3306252062320709,
+      "learning_rate": 0.0002402620111833934,
+      "loss": 0.4725,
+      "step": 8695
+    },
+    {
+      "epoch": 1.0957269263469471,
+      "grad_norm": 0.2956124544143677,
+      "learning_rate": 0.00024017419307506687,
+      "loss": 0.4784,
+      "step": 8700
+    },
+    {
+      "epoch": 1.0963567087571244,
+      "grad_norm": 0.3285719156265259,
+      "learning_rate": 0.00024008632654113894,
+      "loss": 0.4856,
+      "step": 8705
+    },
+    {
+      "epoch": 1.0969864911673017,
+      "grad_norm": 0.3430241644382477,
+      "learning_rate": 0.00023999841162879583,
+      "loss": 0.5017,
+      "step": 8710
+    },
+    {
+      "epoch": 1.097616273577479,
+      "grad_norm": 0.33543142676353455,
+      "learning_rate": 0.00023991044838524985,
+      "loss": 0.516,
+      "step": 8715
+    },
+    {
+      "epoch": 1.0982460559876563,
+      "grad_norm": 0.28755661845207214,
+      "learning_rate": 0.0002398224368577394,
+      "loss": 0.4645,
+      "step": 8720
+    },
+    {
+      "epoch": 1.0988758383978336,
+      "grad_norm": 0.34112608432769775,
+      "learning_rate": 0.00023973437709352851,
+      "loss": 0.5134,
+      "step": 8725
+    },
+    {
+      "epoch": 1.0995056208080107,
+      "grad_norm": 0.3198321759700775,
+      "learning_rate": 0.00023964626913990743,
+      "loss": 0.4939,
+      "step": 8730
+    },
+    {
+      "epoch": 1.100135403218188,
+      "grad_norm": 0.2985571026802063,
+      "learning_rate": 0.00023955811304419205,
+      "loss": 0.4817,
+      "step": 8735
+    },
+    {
+      "epoch": 1.1007651856283653,
+      "grad_norm": 0.32038047909736633,
+      "learning_rate": 0.0002394699088537243,
+      "loss": 0.524,
+      "step": 8740
+    },
+    {
+      "epoch": 1.1013949680385426,
+      "grad_norm": 0.3562256395816803,
+      "learning_rate": 0.00023938165661587175,
+      "loss": 0.4779,
+      "step": 8745
+    },
+    {
+      "epoch": 1.10202475044872,
+      "grad_norm": 0.3481481969356537,
+      "learning_rate": 0.00023929335637802788,
+      "loss": 0.4861,
+      "step": 8750
+    },
+    {
+      "epoch": 1.1026545328588973,
+      "grad_norm": 0.3087615966796875,
+      "learning_rate": 0.00023920500818761198,
+      "loss": 0.473,
+      "step": 8755
+    },
+    {
+      "epoch": 1.1032843152690746,
+      "grad_norm": 0.34575629234313965,
+      "learning_rate": 0.00023911661209206903,
+      "loss": 0.4709,
+      "step": 8760
+    },
+    {
+      "epoch": 1.1039140976792519,
+      "grad_norm": 0.3505946099758148,
+      "learning_rate": 0.0002390281681388697,
+      "loss": 0.4766,
+      "step": 8765
+    },
+    {
+      "epoch": 1.1045438800894292,
+      "grad_norm": 0.40102317929267883,
+      "learning_rate": 0.0002389396763755105,
+      "loss": 0.5048,
+      "step": 8770
+    },
+    {
+      "epoch": 1.1051736624996065,
+      "grad_norm": 0.3319726884365082,
+      "learning_rate": 0.0002388511368495135,
+      "loss": 0.4768,
+      "step": 8775
+    },
+    {
+      "epoch": 1.1058034449097838,
+      "grad_norm": 0.3191297948360443,
+      "learning_rate": 0.00023876254960842645,
+      "loss": 0.5009,
+      "step": 8780
+    },
+    {
+      "epoch": 1.1064332273199609,
+      "grad_norm": 0.3122735619544983,
+      "learning_rate": 0.00023867391469982268,
+      "loss": 0.4777,
+      "step": 8785
+    },
+    {
+      "epoch": 1.1070630097301382,
+      "grad_norm": 0.33340710401535034,
+      "learning_rate": 0.0002385852321713012,
+      "loss": 0.459,
+      "step": 8790
+    },
+    {
+      "epoch": 1.1076927921403155,
+      "grad_norm": 0.32803764939308167,
+      "learning_rate": 0.00023849650207048655,
+      "loss": 0.4784,
+      "step": 8795
+    },
+    {
+      "epoch": 1.1083225745504928,
+      "grad_norm": 0.35463786125183105,
+      "learning_rate": 0.00023840772444502878,
+      "loss": 0.4739,
+      "step": 8800
+    },
+    {
+      "epoch": 1.10895235696067,
+      "grad_norm": 0.3237099349498749,
+      "learning_rate": 0.00023831889934260357,
+      "loss": 0.4652,
+      "step": 8805
+    },
+    {
+      "epoch": 1.1095821393708474,
+      "grad_norm": 0.34681713581085205,
+      "learning_rate": 0.000238230026810912,
+      "loss": 0.4872,
+      "step": 8810
+    },
+    {
+      "epoch": 1.1102119217810247,
+      "grad_norm": 0.3360891342163086,
+      "learning_rate": 0.00023814110689768066,
+      "loss": 0.496,
+      "step": 8815
+    },
+    {
+      "epoch": 1.110841704191202,
+      "grad_norm": 0.32971322536468506,
+      "learning_rate": 0.0002380521396506615,
+      "loss": 0.4468,
+      "step": 8820
+    },
+    {
+      "epoch": 1.1114714866013793,
+      "grad_norm": 0.3112764060497284,
+      "learning_rate": 0.00023796312511763205,
+      "loss": 0.4985,
+      "step": 8825
+    },
+    {
+      "epoch": 1.1121012690115566,
+      "grad_norm": 0.30539095401763916,
+      "learning_rate": 0.0002378740633463951,
+      "loss": 0.4835,
+      "step": 8830
+    },
+    {
+      "epoch": 1.1127310514217337,
+      "grad_norm": 0.274139940738678,
+      "learning_rate": 0.00023778495438477894,
+      "loss": 0.5014,
+      "step": 8835
+    },
+    {
+      "epoch": 1.113360833831911,
+      "grad_norm": 0.2877870500087738,
+      "learning_rate": 0.000237695798280637,
+      "loss": 0.4842,
+      "step": 8840
+    },
+    {
+      "epoch": 1.1139906162420883,
+      "grad_norm": 0.262893944978714,
+      "learning_rate": 0.00023760659508184823,
+      "loss": 0.4754,
+      "step": 8845
+    },
+    {
+      "epoch": 1.1146203986522656,
+      "grad_norm": 0.3255792260169983,
+      "learning_rate": 0.00023751734483631672,
+      "loss": 0.489,
+      "step": 8850
+    },
+    {
+      "epoch": 1.115250181062443,
+      "grad_norm": 0.3453415632247925,
+      "learning_rate": 0.00023742804759197195,
+      "loss": 0.4624,
+      "step": 8855
+    },
+    {
+      "epoch": 1.1158799634726202,
+      "grad_norm": 0.3276025354862213,
+      "learning_rate": 0.00023733870339676856,
+      "loss": 0.4629,
+      "step": 8860
+    },
+    {
+      "epoch": 1.1165097458827975,
+      "grad_norm": 0.32096150517463684,
+      "learning_rate": 0.0002372493122986864,
+      "loss": 0.4482,
+      "step": 8865
+    },
+    {
+      "epoch": 1.1171395282929748,
+      "grad_norm": 0.33016180992126465,
+      "learning_rate": 0.00023715987434573055,
+      "loss": 0.493,
+      "step": 8870
+    },
+    {
+      "epoch": 1.117769310703152,
+      "grad_norm": 0.2946653366088867,
+      "learning_rate": 0.00023707038958593126,
+      "loss": 0.4365,
+      "step": 8875
+    },
+    {
+      "epoch": 1.1183990931133294,
+      "grad_norm": 0.37148308753967285,
+      "learning_rate": 0.00023698085806734385,
+      "loss": 0.4974,
+      "step": 8880
+    },
+    {
+      "epoch": 1.1190288755235067,
+      "grad_norm": 0.3068748116493225,
+      "learning_rate": 0.00023689127983804882,
+      "loss": 0.4886,
+      "step": 8885
+    },
+    {
+      "epoch": 1.1196586579336838,
+      "grad_norm": 0.3096564710140228,
+      "learning_rate": 0.00023680165494615167,
+      "loss": 0.4592,
+      "step": 8890
+    },
+    {
+      "epoch": 1.120288440343861,
+      "grad_norm": 0.3341507613658905,
+      "learning_rate": 0.00023671198343978308,
+      "loss": 0.4258,
+      "step": 8895
+    },
+    {
+      "epoch": 1.1209182227540384,
+      "grad_norm": 0.30653128027915955,
+      "learning_rate": 0.00023662226536709868,
+      "loss": 0.486,
+      "step": 8900
+    },
+    {
+      "epoch": 1.1215480051642157,
+      "grad_norm": 0.30991849303245544,
+      "learning_rate": 0.00023653250077627908,
+      "loss": 0.4879,
+      "step": 8905
+    },
+    {
+      "epoch": 1.122177787574393,
+      "grad_norm": 0.3082162141799927,
+      "learning_rate": 0.00023644268971552998,
+      "loss": 0.4538,
+      "step": 8910
+    },
+    {
+      "epoch": 1.1228075699845703,
+      "grad_norm": 0.30248114466667175,
+      "learning_rate": 0.00023635283223308193,
+      "loss": 0.4501,
+      "step": 8915
+    },
+    {
+      "epoch": 1.1234373523947476,
+      "grad_norm": 0.34090158343315125,
+      "learning_rate": 0.00023626292837719047,
+      "loss": 0.4825,
+      "step": 8920
+    },
+    {
+      "epoch": 1.124067134804925,
+      "grad_norm": 0.28670960664749146,
+      "learning_rate": 0.00023617297819613598,
+      "loss": 0.4422,
+      "step": 8925
+    },
+    {
+      "epoch": 1.1246969172151022,
+      "grad_norm": 0.37079116702079773,
+      "learning_rate": 0.0002360829817382239,
+      "loss": 0.4725,
+      "step": 8930
+    },
+    {
+      "epoch": 1.1253266996252795,
+      "grad_norm": 0.35876086354255676,
+      "learning_rate": 0.00023599293905178417,
+      "loss": 0.4672,
+      "step": 8935
+    },
+    {
+      "epoch": 1.1259564820354568,
+      "grad_norm": 0.28581666946411133,
+      "learning_rate": 0.00023590285018517196,
+      "loss": 0.4597,
+      "step": 8940
+    },
+    {
+      "epoch": 1.126586264445634,
+      "grad_norm": 0.34076693654060364,
+      "learning_rate": 0.00023581271518676694,
+      "loss": 0.4894,
+      "step": 8945
+    },
+    {
+      "epoch": 1.1272160468558112,
+      "grad_norm": 0.29919254779815674,
+      "learning_rate": 0.0002357225341049737,
+      "loss": 0.4538,
+      "step": 8950
+    },
+    {
+      "epoch": 1.1278458292659885,
+      "grad_norm": 0.2799806594848633,
+      "learning_rate": 0.00023563230698822154,
+      "loss": 0.4814,
+      "step": 8955
+    },
+    {
+      "epoch": 1.1284756116761658,
+      "grad_norm": 0.3249780833721161,
+      "learning_rate": 0.00023554203388496446,
+      "loss": 0.4825,
+      "step": 8960
+    },
+    {
+      "epoch": 1.1291053940863431,
+      "grad_norm": 0.3509981036186218,
+      "learning_rate": 0.0002354517148436812,
+      "loss": 0.4468,
+      "step": 8965
+    },
+    {
+      "epoch": 1.1297351764965204,
+      "grad_norm": 0.33016157150268555,
+      "learning_rate": 0.0002353613499128752,
+      "loss": 0.449,
+      "step": 8970
+    },
+    {
+      "epoch": 1.1303649589066977,
+      "grad_norm": 0.2889571487903595,
+      "learning_rate": 0.00023527093914107436,
+      "loss": 0.4584,
+      "step": 8975
+    },
+    {
+      "epoch": 1.130994741316875,
+      "grad_norm": 0.31957536935806274,
+      "learning_rate": 0.00023518048257683145,
+      "loss": 0.4807,
+      "step": 8980
+    },
+    {
+      "epoch": 1.1316245237270524,
+      "grad_norm": 0.31418105959892273,
+      "learning_rate": 0.00023508998026872365,
+      "loss": 0.4755,
+      "step": 8985
+    },
+    {
+      "epoch": 1.1322543061372297,
+      "grad_norm": 0.3458874523639679,
+      "learning_rate": 0.00023499943226535278,
+      "loss": 0.4906,
+      "step": 8990
+    },
+    {
+      "epoch": 1.132884088547407,
+      "grad_norm": 0.3091862201690674,
+      "learning_rate": 0.0002349088386153452,
+      "loss": 0.4786,
+      "step": 8995
+    },
+    {
+      "epoch": 1.133513870957584,
+      "grad_norm": 0.2758231461048126,
+      "learning_rate": 0.00023481819936735178,
+      "loss": 0.4189,
+      "step": 9000
+    },
+    {
+      "epoch": 1.133513870957584,
+      "eval_loss": 0.3038506805896759,
+      "eval_runtime": 6.258,
+      "eval_samples_per_second": 159.795,
+      "eval_steps_per_second": 10.067,
+      "step": 9000
+    },
+    {
+      "epoch": 1.1341436533677614,
+      "grad_norm": 0.3153883218765259,
+      "learning_rate": 0.00023472751457004782,
+      "loss": 0.4802,
+      "step": 9005
+    },
+    {
+      "epoch": 1.1347734357779387,
+      "grad_norm": 0.3110881745815277,
+      "learning_rate": 0.00023463678427213317,
+      "loss": 0.4488,
+      "step": 9010
+    },
+    {
+      "epoch": 1.135403218188116,
+      "grad_norm": 0.30957111716270447,
+      "learning_rate": 0.00023454600852233206,
+      "loss": 0.476,
+      "step": 9015
+    },
+    {
+      "epoch": 1.1360330005982933,
+      "grad_norm": 0.3130200207233429,
+      "learning_rate": 0.00023445518736939312,
+      "loss": 0.4396,
+      "step": 9020
+    },
+    {
+      "epoch": 1.1366627830084706,
+      "grad_norm": 0.31500178575515747,
+      "learning_rate": 0.0002343643208620894,
+      "loss": 0.4644,
+      "step": 9025
+    },
+    {
+      "epoch": 1.1372925654186479,
+      "grad_norm": 0.3096972703933716,
+      "learning_rate": 0.00023427340904921834,
+      "loss": 0.4775,
+      "step": 9030
+    },
+    {
+      "epoch": 1.1379223478288252,
+      "grad_norm": 0.3503490388393402,
+      "learning_rate": 0.00023418245197960155,
+      "loss": 0.4617,
+      "step": 9035
+    },
+    {
+      "epoch": 1.1385521302390025,
+      "grad_norm": 0.31281721591949463,
+      "learning_rate": 0.00023409144970208516,
+      "loss": 0.4703,
+      "step": 9040
+    },
+    {
+      "epoch": 1.1391819126491798,
+      "grad_norm": 0.3011356592178345,
+      "learning_rate": 0.0002340004022655394,
+      "loss": 0.4472,
+      "step": 9045
+    },
+    {
+      "epoch": 1.139811695059357,
+      "grad_norm": 0.3240005075931549,
+      "learning_rate": 0.00023390930971885888,
+      "loss": 0.4726,
+      "step": 9050
+    },
+    {
+      "epoch": 1.1404414774695342,
+      "grad_norm": 0.35690784454345703,
+      "learning_rate": 0.0002338181721109623,
+      "loss": 0.4601,
+      "step": 9055
+    },
+    {
+      "epoch": 1.1410712598797115,
+      "grad_norm": 0.30888888239860535,
+      "learning_rate": 0.0002337269894907927,
+      "loss": 0.45,
+      "step": 9060
+    },
+    {
+      "epoch": 1.1417010422898888,
+      "grad_norm": 0.3118223249912262,
+      "learning_rate": 0.00023363576190731726,
+      "loss": 0.4456,
+      "step": 9065
+    },
+    {
+      "epoch": 1.142330824700066,
+      "grad_norm": 0.3156544864177704,
+      "learning_rate": 0.0002335444894095272,
+      "loss": 0.4744,
+      "step": 9070
+    },
+    {
+      "epoch": 1.1429606071102434,
+      "grad_norm": 0.33679795265197754,
+      "learning_rate": 0.00023345317204643797,
+      "loss": 0.4662,
+      "step": 9075
+    },
+    {
+      "epoch": 1.1435903895204207,
+      "grad_norm": 0.32647955417633057,
+      "learning_rate": 0.00023336180986708904,
+      "loss": 0.4573,
+      "step": 9080
+    },
+    {
+      "epoch": 1.144220171930598,
+      "grad_norm": 0.3759111762046814,
+      "learning_rate": 0.00023327040292054412,
+      "loss": 0.4439,
+      "step": 9085
+    },
+    {
+      "epoch": 1.1448499543407753,
+      "grad_norm": 0.31271886825561523,
+      "learning_rate": 0.00023317895125589066,
+      "loss": 0.4778,
+      "step": 9090
+    },
+    {
+      "epoch": 1.1454797367509526,
+      "grad_norm": 0.2915593385696411,
+      "learning_rate": 0.0002330874549222404,
+      "loss": 0.4646,
+      "step": 9095
+    },
+    {
+      "epoch": 1.14610951916113,
+      "grad_norm": 0.3337639570236206,
+      "learning_rate": 0.00023299591396872893,
+      "loss": 0.4597,
+      "step": 9100
+    },
+    {
+      "epoch": 1.1467393015713072,
+      "grad_norm": 0.3345816433429718,
+      "learning_rate": 0.0002329043284445158,
+      "loss": 0.519,
+      "step": 9105
+    },
+    {
+      "epoch": 1.1473690839814843,
+      "grad_norm": 0.31568819284439087,
+      "learning_rate": 0.0002328126983987846,
+      "loss": 0.455,
+      "step": 9110
+    },
+    {
+      "epoch": 1.1479988663916616,
+      "grad_norm": 0.3630363643169403,
+      "learning_rate": 0.00023272102388074265,
+      "loss": 0.4544,
+      "step": 9115
+    },
+    {
+      "epoch": 1.148628648801839,
+      "grad_norm": 0.30382248759269714,
+      "learning_rate": 0.00023262930493962142,
+      "loss": 0.485,
+      "step": 9120
+    },
+    {
+      "epoch": 1.1492584312120162,
+      "grad_norm": 0.30339518189430237,
+      "learning_rate": 0.0002325375416246759,
+      "loss": 0.474,
+      "step": 9125
+    },
+    {
+      "epoch": 1.1498882136221935,
+      "grad_norm": 0.33041009306907654,
+      "learning_rate": 0.00023244573398518523,
+      "loss": 0.447,
+      "step": 9130
+    },
+    {
+      "epoch": 1.1505179960323708,
+      "grad_norm": 0.35708925127983093,
+      "learning_rate": 0.00023235388207045214,
+      "loss": 0.4801,
+      "step": 9135
+    },
+    {
+      "epoch": 1.1511477784425481,
+      "grad_norm": 0.3497597575187683,
+      "learning_rate": 0.00023226198592980318,
+      "loss": 0.4753,
+      "step": 9140
+    },
+    {
+      "epoch": 1.1517775608527254,
+      "grad_norm": 0.31747546792030334,
+      "learning_rate": 0.00023217004561258876,
+      "loss": 0.4642,
+      "step": 9145
+    },
+    {
+      "epoch": 1.1524073432629027,
+      "grad_norm": 0.31225451827049255,
+      "learning_rate": 0.00023207806116818283,
+      "loss": 0.501,
+      "step": 9150
+    },
+    {
+      "epoch": 1.15303712567308,
+      "grad_norm": 0.31150931119918823,
+      "learning_rate": 0.00023198603264598327,
+      "loss": 0.447,
+      "step": 9155
+    },
+    {
+      "epoch": 1.1536669080832573,
+      "grad_norm": 0.29207199811935425,
+      "learning_rate": 0.00023189396009541135,
+      "loss": 0.448,
+      "step": 9160
+    },
+    {
+      "epoch": 1.1542966904934344,
+      "grad_norm": 0.33640962839126587,
+      "learning_rate": 0.00023180184356591223,
+      "loss": 0.4725,
+      "step": 9165
+    },
+    {
+      "epoch": 1.1549264729036117,
+      "grad_norm": 0.292582631111145,
+      "learning_rate": 0.00023170968310695457,
+      "loss": 0.4603,
+      "step": 9170
+    },
+    {
+      "epoch": 1.155556255313789,
+      "grad_norm": 0.3217863142490387,
+      "learning_rate": 0.00023161747876803066,
+      "loss": 0.4386,
+      "step": 9175
+    },
+    {
+      "epoch": 1.1561860377239663,
+      "grad_norm": 0.32607826590538025,
+      "learning_rate": 0.00023152523059865622,
+      "loss": 0.4747,
+      "step": 9180
+    },
+    {
+      "epoch": 1.1568158201341436,
+      "grad_norm": 0.35956209897994995,
+      "learning_rate": 0.00023143293864837078,
+      "loss": 0.4563,
+      "step": 9185
+    },
+    {
+      "epoch": 1.157445602544321,
+      "grad_norm": 0.3542852997779846,
+      "learning_rate": 0.00023134060296673716,
+      "loss": 0.4907,
+      "step": 9190
+    },
+    {
+      "epoch": 1.1580753849544982,
+      "grad_norm": 0.3324996829032898,
+      "learning_rate": 0.0002312482236033417,
+      "loss": 0.4539,
+      "step": 9195
+    },
+    {
+      "epoch": 1.1587051673646755,
+      "grad_norm": 0.3436378836631775,
+      "learning_rate": 0.00023115580060779429,
+      "loss": 0.5107,
+      "step": 9200
+    },
+    {
+      "epoch": 1.1593349497748529,
+      "grad_norm": 0.2886941730976105,
+      "learning_rate": 0.00023106333402972813,
+      "loss": 0.4547,
+      "step": 9205
+    },
+    {
+      "epoch": 1.1599647321850302,
+      "grad_norm": 0.30411913990974426,
+      "learning_rate": 0.00023097082391879993,
+      "loss": 0.4517,
+      "step": 9210
+    },
+    {
+      "epoch": 1.1605945145952075,
+      "grad_norm": 0.3265014886856079,
+      "learning_rate": 0.00023087827032468975,
+      "loss": 0.4589,
+      "step": 9215
+    },
+    {
+      "epoch": 1.1612242970053845,
+      "grad_norm": 0.2876526713371277,
+      "learning_rate": 0.00023078567329710091,
+      "loss": 0.466,
+      "step": 9220
+    },
+    {
+      "epoch": 1.1618540794155618,
+      "grad_norm": 0.29947248101234436,
+      "learning_rate": 0.0002306930328857602,
+      "loss": 0.4459,
+      "step": 9225
+    },
+    {
+      "epoch": 1.1624838618257392,
+      "grad_norm": 0.33246028423309326,
+      "learning_rate": 0.00023060034914041753,
+      "loss": 0.4826,
+      "step": 9230
+    },
+    {
+      "epoch": 1.1631136442359165,
+      "grad_norm": 0.29653674364089966,
+      "learning_rate": 0.0002305076221108463,
+      "loss": 0.4394,
+      "step": 9235
+    },
+    {
+      "epoch": 1.1637434266460938,
+      "grad_norm": 0.30506858229637146,
+      "learning_rate": 0.00023041485184684308,
+      "loss": 0.4645,
+      "step": 9240
+    },
+    {
+      "epoch": 1.164373209056271,
+      "grad_norm": 0.2603437304496765,
+      "learning_rate": 0.00023032203839822748,
+      "loss": 0.4536,
+      "step": 9245
+    },
+    {
+      "epoch": 1.1650029914664484,
+      "grad_norm": 0.3310236632823944,
+      "learning_rate": 0.00023022918181484254,
+      "loss": 0.4653,
+      "step": 9250
+    },
+    {
+      "epoch": 1.1656327738766257,
+      "grad_norm": 0.3645521104335785,
+      "learning_rate": 0.0002301362821465543,
+      "loss": 0.4404,
+      "step": 9255
+    },
+    {
+      "epoch": 1.166262556286803,
+      "grad_norm": 0.33431464433670044,
+      "learning_rate": 0.00023004333944325208,
+      "loss": 0.4389,
+      "step": 9260
+    },
+    {
+      "epoch": 1.1668923386969803,
+      "grad_norm": 0.31086647510528564,
+      "learning_rate": 0.00022995035375484817,
+      "loss": 0.493,
+      "step": 9265
+    },
+    {
+      "epoch": 1.1675221211071576,
+      "grad_norm": 0.34322085976600647,
+      "learning_rate": 0.00022985732513127805,
+      "loss": 0.4839,
+      "step": 9270
+    },
+    {
+      "epoch": 1.1681519035173347,
+      "grad_norm": 0.3111884593963623,
+      "learning_rate": 0.0002297642536225002,
+      "loss": 0.4473,
+      "step": 9275
+    },
+    {
+      "epoch": 1.168781685927512,
+      "grad_norm": 0.3494400084018707,
+      "learning_rate": 0.00022967113927849613,
+      "loss": 0.469,
+      "step": 9280
+    },
+    {
+      "epoch": 1.1694114683376893,
+      "grad_norm": 0.27351829409599304,
+      "learning_rate": 0.00022957798214927037,
+      "loss": 0.4617,
+      "step": 9285
+    },
+    {
+      "epoch": 1.1700412507478666,
+      "grad_norm": 0.3605945408344269,
+      "learning_rate": 0.00022948478228485046,
+      "loss": 0.444,
+      "step": 9290
+    },
+    {
+      "epoch": 1.1706710331580439,
+      "grad_norm": 0.31383225321769714,
+      "learning_rate": 0.0002293915397352869,
+      "loss": 0.4716,
+      "step": 9295
+    },
+    {
+      "epoch": 1.1713008155682212,
+      "grad_norm": 0.3261600732803345,
+      "learning_rate": 0.00022929825455065292,
+      "loss": 0.4646,
+      "step": 9300
+    },
+    {
+      "epoch": 1.1719305979783985,
+      "grad_norm": 0.29624396562576294,
+      "learning_rate": 0.00022920492678104492,
+      "loss": 0.4636,
+      "step": 9305
+    },
+    {
+      "epoch": 1.1725603803885758,
+      "grad_norm": 0.39078545570373535,
+      "learning_rate": 0.00022911155647658201,
+      "loss": 0.4933,
+      "step": 9310
+    },
+    {
+      "epoch": 1.173190162798753,
+      "grad_norm": 0.2990373373031616,
+      "learning_rate": 0.00022901814368740615,
+      "loss": 0.4726,
+      "step": 9315
+    },
+    {
+      "epoch": 1.1738199452089304,
+      "grad_norm": 0.28325891494750977,
+      "learning_rate": 0.00022892468846368217,
+      "loss": 0.4428,
+      "step": 9320
+    },
+    {
+      "epoch": 1.1744497276191077,
+      "grad_norm": 0.3357643187046051,
+      "learning_rate": 0.0002288311908555977,
+      "loss": 0.4618,
+      "step": 9325
+    },
+    {
+      "epoch": 1.1750795100292848,
+      "grad_norm": 0.31550613045692444,
+      "learning_rate": 0.00022873765091336302,
+      "loss": 0.4607,
+      "step": 9330
+    },
+    {
+      "epoch": 1.175709292439462,
+      "grad_norm": 0.30639806389808655,
+      "learning_rate": 0.00022864406868721118,
+      "loss": 0.458,
+      "step": 9335
+    },
+    {
+      "epoch": 1.1763390748496394,
+      "grad_norm": 0.3836449086666107,
+      "learning_rate": 0.0002285504442273981,
+      "loss": 0.4788,
+      "step": 9340
+    },
+    {
+      "epoch": 1.1769688572598167,
+      "grad_norm": 0.2955804467201233,
+      "learning_rate": 0.00022845677758420217,
+      "loss": 0.4636,
+      "step": 9345
+    },
+    {
+      "epoch": 1.177598639669994,
+      "grad_norm": 0.3264003098011017,
+      "learning_rate": 0.0002283630688079245,
+      "loss": 0.4769,
+      "step": 9350
+    },
+    {
+      "epoch": 1.1782284220801713,
+      "grad_norm": 0.34578555822372437,
+      "learning_rate": 0.00022826931794888894,
+      "loss": 0.4784,
+      "step": 9355
+    },
+    {
+      "epoch": 1.1788582044903486,
+      "grad_norm": 0.37039560079574585,
+      "learning_rate": 0.00022817552505744178,
+      "loss": 0.5042,
+      "step": 9360
+    },
+    {
+      "epoch": 1.179487986900526,
+      "grad_norm": 0.319118857383728,
+      "learning_rate": 0.00022808169018395192,
+      "loss": 0.4607,
+      "step": 9365
+    },
+    {
+      "epoch": 1.1801177693107032,
+      "grad_norm": 0.32380104064941406,
+      "learning_rate": 0.00022798781337881086,
+      "loss": 0.4606,
+      "step": 9370
+    },
+    {
+      "epoch": 1.1807475517208805,
+      "grad_norm": 0.3038274943828583,
+      "learning_rate": 0.00022789389469243256,
+      "loss": 0.448,
+      "step": 9375
+    },
+    {
+      "epoch": 1.1813773341310578,
+      "grad_norm": 0.3078247308731079,
+      "learning_rate": 0.00022779993417525356,
+      "loss": 0.4683,
+      "step": 9380
+    },
+    {
+      "epoch": 1.182007116541235,
+      "grad_norm": 0.2909676432609558,
+      "learning_rate": 0.00022770593187773275,
+      "loss": 0.4778,
+      "step": 9385
+    },
+    {
+      "epoch": 1.1826368989514122,
+      "grad_norm": 0.3095955550670624,
+      "learning_rate": 0.00022761188785035155,
+      "loss": 0.4523,
+      "step": 9390
+    },
+    {
+      "epoch": 1.1832666813615895,
+      "grad_norm": 0.2969966530799866,
+      "learning_rate": 0.0002275178021436137,
+      "loss": 0.4735,
+      "step": 9395
+    },
+    {
+      "epoch": 1.1838964637717668,
+      "grad_norm": 0.2896679937839508,
+      "learning_rate": 0.00022742367480804544,
+      "loss": 0.45,
+      "step": 9400
+    },
+    {
+      "epoch": 1.1845262461819441,
+      "grad_norm": 0.31511151790618896,
+      "learning_rate": 0.0002273295058941952,
+      "loss": 0.4614,
+      "step": 9405
+    },
+    {
+      "epoch": 1.1851560285921214,
+      "grad_norm": 0.3440285623073578,
+      "learning_rate": 0.00022723529545263399,
+      "loss": 0.4593,
+      "step": 9410
+    },
+    {
+      "epoch": 1.1857858110022987,
+      "grad_norm": 0.29399538040161133,
+      "learning_rate": 0.00022714104353395483,
+      "loss": 0.4519,
+      "step": 9415
+    },
+    {
+      "epoch": 1.186415593412476,
+      "grad_norm": 0.3958999812602997,
+      "learning_rate": 0.00022704675018877322,
+      "loss": 0.4838,
+      "step": 9420
+    },
+    {
+      "epoch": 1.1870453758226533,
+      "grad_norm": 0.2960554361343384,
+      "learning_rate": 0.0002269524154677268,
+      "loss": 0.459,
+      "step": 9425
+    },
+    {
+      "epoch": 1.1876751582328307,
+      "grad_norm": 0.32369253039360046,
+      "learning_rate": 0.00022685803942147555,
+      "loss": 0.4542,
+      "step": 9430
+    },
+    {
+      "epoch": 1.188304940643008,
+      "grad_norm": 0.320547491312027,
+      "learning_rate": 0.00022676362210070144,
+      "loss": 0.4853,
+      "step": 9435
+    },
+    {
+      "epoch": 1.188934723053185,
+      "grad_norm": 0.2984744906425476,
+      "learning_rate": 0.00022666916355610885,
+      "loss": 0.4201,
+      "step": 9440
+    },
+    {
+      "epoch": 1.1895645054633623,
+      "grad_norm": 0.34194597601890564,
+      "learning_rate": 0.00022657466383842407,
+      "loss": 0.4705,
+      "step": 9445
+    },
+    {
+      "epoch": 1.1901942878735396,
+      "grad_norm": 0.29718858003616333,
+      "learning_rate": 0.0002264801229983957,
+      "loss": 0.4403,
+      "step": 9450
+    },
+    {
+      "epoch": 1.190824070283717,
+      "grad_norm": 0.29723846912384033,
+      "learning_rate": 0.0002263855410867943,
+      "loss": 0.4841,
+      "step": 9455
+    },
+    {
+      "epoch": 1.1914538526938943,
+      "grad_norm": 0.31662440299987793,
+      "learning_rate": 0.00022629091815441245,
+      "loss": 0.456,
+      "step": 9460
+    },
+    {
+      "epoch": 1.1920836351040716,
+      "grad_norm": 0.3458605408668518,
+      "learning_rate": 0.0002261962542520649,
+      "loss": 0.4504,
+      "step": 9465
+    },
+    {
+      "epoch": 1.1927134175142489,
+      "grad_norm": 0.31829431653022766,
+      "learning_rate": 0.00022610154943058833,
+      "loss": 0.4821,
+      "step": 9470
+    },
+    {
+      "epoch": 1.1933431999244262,
+      "grad_norm": 0.3380287289619446,
+      "learning_rate": 0.00022600680374084138,
+      "loss": 0.4963,
+      "step": 9475
+    },
+    {
+      "epoch": 1.1939729823346035,
+      "grad_norm": 0.3048580288887024,
+      "learning_rate": 0.00022591201723370458,
+      "loss": 0.4443,
+      "step": 9480
+    },
+    {
+      "epoch": 1.1946027647447806,
+      "grad_norm": 0.34586548805236816,
+      "learning_rate": 0.0002258171899600806,
+      "loss": 0.473,
+      "step": 9485
+    },
+    {
+      "epoch": 1.195232547154958,
+      "grad_norm": 0.2828037440776825,
+      "learning_rate": 0.0002257223219708937,
+      "loss": 0.4539,
+      "step": 9490
+    },
+    {
+      "epoch": 1.1958623295651352,
+      "grad_norm": 0.31300345063209534,
+      "learning_rate": 0.00022562741331709024,
+      "loss": 0.4353,
+      "step": 9495
+    },
+    {
+      "epoch": 1.1964921119753125,
+      "grad_norm": 0.311260849237442,
+      "learning_rate": 0.0002255324640496383,
+      "loss": 0.4553,
+      "step": 9500
+    },
+    {
+      "epoch": 1.1971218943854898,
+      "grad_norm": 0.2941080331802368,
+      "learning_rate": 0.0002254374742195279,
+      "loss": 0.4464,
+      "step": 9505
+    },
+    {
+      "epoch": 1.197751676795667,
+      "grad_norm": 0.26669132709503174,
+      "learning_rate": 0.00022534244387777057,
+      "loss": 0.4368,
+      "step": 9510
+    },
+    {
+      "epoch": 1.1983814592058444,
+      "grad_norm": 0.2933709919452667,
+      "learning_rate": 0.00022524737307539995,
+      "loss": 0.4526,
+      "step": 9515
+    },
+    {
+      "epoch": 1.1990112416160217,
+      "grad_norm": 0.338360458612442,
+      "learning_rate": 0.0002251522618634711,
+      "loss": 0.4625,
+      "step": 9520
+    },
+    {
+      "epoch": 1.199641024026199,
+      "grad_norm": 0.31670835614204407,
+      "learning_rate": 0.00022505711029306098,
+      "loss": 0.4553,
+      "step": 9525
+    },
+    {
+      "epoch": 1.2002708064363763,
+      "grad_norm": 0.3221518099308014,
+      "learning_rate": 0.00022496191841526813,
+      "loss": 0.475,
+      "step": 9530
+    },
+    {
+      "epoch": 1.2009005888465536,
+      "grad_norm": 0.32984668016433716,
+      "learning_rate": 0.00022486668628121282,
+      "loss": 0.4526,
+      "step": 9535
+    },
+    {
+      "epoch": 1.2015303712567307,
+      "grad_norm": 0.2793140113353729,
+      "learning_rate": 0.00022477141394203678,
+      "loss": 0.4374,
+      "step": 9540
+    },
+    {
+      "epoch": 1.2021601536669082,
+      "grad_norm": 0.3125605881214142,
+      "learning_rate": 0.00022467610144890357,
+      "loss": 0.4569,
+      "step": 9545
+    },
+    {
+      "epoch": 1.2027899360770853,
+      "grad_norm": 0.2892754375934601,
+      "learning_rate": 0.00022458074885299808,
+      "loss": 0.4747,
+      "step": 9550
+    },
+    {
+      "epoch": 1.2034197184872626,
+      "grad_norm": 0.3224146068096161,
+      "learning_rate": 0.00022448535620552684,
+      "loss": 0.4372,
+      "step": 9555
+    },
+    {
+      "epoch": 1.20404950089744,
+      "grad_norm": 0.33973759412765503,
+      "learning_rate": 0.00022438992355771787,
+      "loss": 0.4368,
+      "step": 9560
+    },
+    {
+      "epoch": 1.2046792833076172,
+      "grad_norm": 0.37665504217147827,
+      "learning_rate": 0.00022429445096082073,
+      "loss": 0.4747,
+      "step": 9565
+    },
+    {
+      "epoch": 1.2053090657177945,
+      "grad_norm": 0.2834467589855194,
+      "learning_rate": 0.00022419893846610634,
+      "loss": 0.4841,
+      "step": 9570
+    },
+    {
+      "epoch": 1.2059388481279718,
+      "grad_norm": 0.3729229271411896,
+      "learning_rate": 0.00022410338612486715,
+      "loss": 0.475,
+      "step": 9575
+    },
+    {
+      "epoch": 1.2065686305381491,
+      "grad_norm": 0.30668923258781433,
+      "learning_rate": 0.00022400779398841684,
+      "loss": 0.4271,
+      "step": 9580
+    },
+    {
+      "epoch": 1.2071984129483264,
+      "grad_norm": 0.33016908168792725,
+      "learning_rate": 0.00022391216210809072,
+      "loss": 0.4553,
+      "step": 9585
+    },
+    {
+      "epoch": 1.2078281953585037,
+      "grad_norm": 0.30926114320755005,
+      "learning_rate": 0.00022381649053524518,
+      "loss": 0.4512,
+      "step": 9590
+    },
+    {
+      "epoch": 1.2084579777686808,
+      "grad_norm": 0.3481772840023041,
+      "learning_rate": 0.00022372077932125809,
+      "loss": 0.4707,
+      "step": 9595
+    },
+    {
+      "epoch": 1.2090877601788583,
+      "grad_norm": 0.2549537420272827,
+      "learning_rate": 0.0002236250285175285,
+      "loss": 0.4686,
+      "step": 9600
+    },
+    {
+      "epoch": 1.2097175425890354,
+      "grad_norm": 0.3111298978328705,
+      "learning_rate": 0.00022352923817547688,
+      "loss": 0.4535,
+      "step": 9605
+    },
+    {
+      "epoch": 1.2103473249992127,
+      "grad_norm": 0.29062095284461975,
+      "learning_rate": 0.00022343340834654472,
+      "loss": 0.4612,
+      "step": 9610
+    },
+    {
+      "epoch": 1.21097710740939,
+      "grad_norm": 0.3373335897922516,
+      "learning_rate": 0.0002233375390821949,
+      "loss": 0.4233,
+      "step": 9615
+    },
+    {
+      "epoch": 1.2116068898195673,
+      "grad_norm": 0.308648943901062,
+      "learning_rate": 0.0002232416304339114,
+      "loss": 0.4535,
+      "step": 9620
+    },
+    {
+      "epoch": 1.2122366722297446,
+      "grad_norm": 0.32941722869873047,
+      "learning_rate": 0.00022314568245319935,
+      "loss": 0.4564,
+      "step": 9625
+    },
+    {
+      "epoch": 1.212866454639922,
+      "grad_norm": 0.33229124546051025,
+      "learning_rate": 0.00022304969519158495,
+      "loss": 0.458,
+      "step": 9630
+    },
+    {
+      "epoch": 1.2134962370500992,
+      "grad_norm": 0.29093366861343384,
+      "learning_rate": 0.00022295366870061565,
+      "loss": 0.4315,
+      "step": 9635
+    },
+    {
+      "epoch": 1.2141260194602765,
+      "grad_norm": 0.3482106328010559,
+      "learning_rate": 0.00022285760303185982,
+      "loss": 0.4311,
+      "step": 9640
+    },
+    {
+      "epoch": 1.2147558018704538,
+      "grad_norm": 0.29717814922332764,
+      "learning_rate": 0.0002227614982369069,
+      "loss": 0.4261,
+      "step": 9645
+    },
+    {
+      "epoch": 1.215385584280631,
+      "grad_norm": 0.3359118700027466,
+      "learning_rate": 0.00022266535436736738,
+      "loss": 0.4698,
+      "step": 9650
+    },
+    {
+      "epoch": 1.2160153666908082,
+      "grad_norm": 0.3095514476299286,
+      "learning_rate": 0.0002225691714748727,
+      "loss": 0.4463,
+      "step": 9655
+    },
+    {
+      "epoch": 1.2166451491009855,
+      "grad_norm": 0.29095733165740967,
+      "learning_rate": 0.0002224729496110753,
+      "loss": 0.4662,
+      "step": 9660
+    },
+    {
+      "epoch": 1.2172749315111628,
+      "grad_norm": 0.34425532817840576,
+      "learning_rate": 0.00022237668882764847,
+      "loss": 0.4579,
+      "step": 9665
+    },
+    {
+      "epoch": 1.2179047139213401,
+      "grad_norm": 0.32856446504592896,
+      "learning_rate": 0.0002222803891762865,
+      "loss": 0.4648,
+      "step": 9670
+    },
+    {
+      "epoch": 1.2185344963315174,
+      "grad_norm": 0.35708895325660706,
+      "learning_rate": 0.00022218405070870451,
+      "loss": 0.4579,
+      "step": 9675
+    },
+    {
+      "epoch": 1.2191642787416948,
+      "grad_norm": 0.26759231090545654,
+      "learning_rate": 0.0002220876734766384,
+      "loss": 0.4321,
+      "step": 9680
+    },
+    {
+      "epoch": 1.219794061151872,
+      "grad_norm": 0.27995094656944275,
+      "learning_rate": 0.00022199125753184497,
+      "loss": 0.4552,
+      "step": 9685
+    },
+    {
+      "epoch": 1.2204238435620494,
+      "grad_norm": 0.3591984510421753,
+      "learning_rate": 0.00022189480292610187,
+      "loss": 0.4685,
+      "step": 9690
+    },
+    {
+      "epoch": 1.2210536259722267,
+      "grad_norm": 0.2892036736011505,
+      "learning_rate": 0.00022179830971120722,
+      "loss": 0.4609,
+      "step": 9695
+    },
+    {
+      "epoch": 1.221683408382404,
+      "grad_norm": 0.3287111520767212,
+      "learning_rate": 0.00022170177793898028,
+      "loss": 0.479,
+      "step": 9700
+    },
+    {
+      "epoch": 1.222313190792581,
+      "grad_norm": 0.3088148832321167,
+      "learning_rate": 0.00022160520766126074,
+      "loss": 0.4597,
+      "step": 9705
+    },
+    {
+      "epoch": 1.2229429732027584,
+      "grad_norm": 0.3263307511806488,
+      "learning_rate": 0.0002215085989299091,
+      "loss": 0.4801,
+      "step": 9710
+    },
+    {
+      "epoch": 1.2235727556129357,
+      "grad_norm": 0.283078134059906,
+      "learning_rate": 0.0002214119517968063,
+      "loss": 0.4476,
+      "step": 9715
+    },
+    {
+      "epoch": 1.224202538023113,
+      "grad_norm": 0.3226225674152374,
+      "learning_rate": 0.00022131526631385422,
+      "loss": 0.4644,
+      "step": 9720
+    },
+    {
+      "epoch": 1.2248323204332903,
+      "grad_norm": 0.32242435216903687,
+      "learning_rate": 0.00022121854253297514,
+      "loss": 0.4477,
+      "step": 9725
+    },
+    {
+      "epoch": 1.2254621028434676,
+      "grad_norm": 0.3373146057128906,
+      "learning_rate": 0.0002211217805061119,
+      "loss": 0.4541,
+      "step": 9730
+    },
+    {
+      "epoch": 1.2260918852536449,
+      "grad_norm": 0.28866246342658997,
+      "learning_rate": 0.00022102498028522786,
+      "loss": 0.4388,
+      "step": 9735
+    },
+    {
+      "epoch": 1.2267216676638222,
+      "grad_norm": 0.308704674243927,
+      "learning_rate": 0.00022092814192230711,
+      "loss": 0.425,
+      "step": 9740
+    },
+    {
+      "epoch": 1.2273514500739995,
+      "grad_norm": 0.3144040107727051,
+      "learning_rate": 0.00022083126546935394,
+      "loss": 0.4532,
+      "step": 9745
+    },
+    {
+      "epoch": 1.2279812324841768,
+      "grad_norm": 0.29848021268844604,
+      "learning_rate": 0.00022073435097839329,
+      "loss": 0.457,
+      "step": 9750
+    },
+    {
+      "epoch": 1.228611014894354,
+      "grad_norm": 0.35102754831314087,
+      "learning_rate": 0.00022063739850147036,
+      "loss": 0.4258,
+      "step": 9755
+    },
+    {
+      "epoch": 1.2292407973045312,
+      "grad_norm": 0.32105547189712524,
+      "learning_rate": 0.000220540408090651,
+      "loss": 0.4226,
+      "step": 9760
+    },
+    {
+      "epoch": 1.2298705797147085,
+      "grad_norm": 0.3647817075252533,
+      "learning_rate": 0.0002204433797980211,
+      "loss": 0.4556,
+      "step": 9765
+    },
+    {
+      "epoch": 1.2305003621248858,
+      "grad_norm": 0.3260333836078644,
+      "learning_rate": 0.00022034631367568718,
+      "loss": 0.4834,
+      "step": 9770
+    },
+    {
+      "epoch": 1.231130144535063,
+      "grad_norm": 0.30218422412872314,
+      "learning_rate": 0.00022024920977577596,
+      "loss": 0.4327,
+      "step": 9775
+    },
+    {
+      "epoch": 1.2317599269452404,
+      "grad_norm": 0.3666177988052368,
+      "learning_rate": 0.0002201520681504344,
+      "loss": 0.4361,
+      "step": 9780
+    },
+    {
+      "epoch": 1.2323897093554177,
+      "grad_norm": 0.3113807737827301,
+      "learning_rate": 0.00022005488885182975,
+      "loss": 0.4554,
+      "step": 9785
+    },
+    {
+      "epoch": 1.233019491765595,
+      "grad_norm": 0.31085875630378723,
+      "learning_rate": 0.00021995767193214963,
+      "loss": 0.4391,
+      "step": 9790
+    },
+    {
+      "epoch": 1.2336492741757723,
+      "grad_norm": 0.304509699344635,
+      "learning_rate": 0.0002198604174436017,
+      "loss": 0.4754,
+      "step": 9795
+    },
+    {
+      "epoch": 1.2342790565859496,
+      "grad_norm": 0.2930733263492584,
+      "learning_rate": 0.0002197631254384138,
+      "loss": 0.4194,
+      "step": 9800
+    },
+    {
+      "epoch": 1.234908838996127,
+      "grad_norm": 0.30277615785598755,
+      "learning_rate": 0.00021966579596883394,
+      "loss": 0.4506,
+      "step": 9805
+    },
+    {
+      "epoch": 1.2355386214063042,
+      "grad_norm": 0.2824211120605469,
+      "learning_rate": 0.00021956842908713037,
+      "loss": 0.4398,
+      "step": 9810
+    },
+    {
+      "epoch": 1.2361684038164813,
+      "grad_norm": 0.31834569573402405,
+      "learning_rate": 0.00021947102484559121,
+      "loss": 0.4756,
+      "step": 9815
+    },
+    {
+      "epoch": 1.2367981862266586,
+      "grad_norm": 0.355283260345459,
+      "learning_rate": 0.00021937358329652488,
+      "loss": 0.456,
+      "step": 9820
+    },
+    {
+      "epoch": 1.237427968636836,
+      "grad_norm": 0.2955317497253418,
+      "learning_rate": 0.00021927610449225962,
+      "loss": 0.4462,
+      "step": 9825
+    },
+    {
+      "epoch": 1.2380577510470132,
+      "grad_norm": 0.2653120756149292,
+      "learning_rate": 0.00021917858848514383,
+      "loss": 0.4197,
+      "step": 9830
+    },
+    {
+      "epoch": 1.2386875334571905,
+      "grad_norm": 0.3773416578769684,
+      "learning_rate": 0.0002190810353275458,
+      "loss": 0.4263,
+      "step": 9835
+    },
+    {
+      "epoch": 1.2393173158673678,
+      "grad_norm": 0.28635114431381226,
+      "learning_rate": 0.00021898344507185384,
+      "loss": 0.4705,
+      "step": 9840
+    },
+    {
+      "epoch": 1.2399470982775451,
+      "grad_norm": 0.3044835031032562,
+      "learning_rate": 0.00021888581777047608,
+      "loss": 0.4671,
+      "step": 9845
+    },
+    {
+      "epoch": 1.2405768806877224,
+      "grad_norm": 0.293748676776886,
+      "learning_rate": 0.0002187881534758407,
+      "loss": 0.436,
+      "step": 9850
+    },
+    {
+      "epoch": 1.2412066630978997,
+      "grad_norm": 0.3891184628009796,
+      "learning_rate": 0.00021869045224039564,
+      "loss": 0.456,
+      "step": 9855
+    },
+    {
+      "epoch": 1.241836445508077,
+      "grad_norm": 0.3140691816806793,
+      "learning_rate": 0.0002185927141166086,
+      "loss": 0.4402,
+      "step": 9860
+    },
+    {
+      "epoch": 1.2424662279182543,
+      "grad_norm": 0.33889827132225037,
+      "learning_rate": 0.00021849493915696738,
+      "loss": 0.4363,
+      "step": 9865
+    },
+    {
+      "epoch": 1.2430960103284314,
+      "grad_norm": 0.3084375858306885,
+      "learning_rate": 0.0002183971274139791,
+      "loss": 0.4295,
+      "step": 9870
+    },
+    {
+      "epoch": 1.2437257927386087,
+      "grad_norm": 0.3091178834438324,
+      "learning_rate": 0.00021829927894017115,
+      "loss": 0.4263,
+      "step": 9875
+    },
+    {
+      "epoch": 1.244355575148786,
+      "grad_norm": 0.3208729922771454,
+      "learning_rate": 0.00021820139378809025,
+      "loss": 0.4233,
+      "step": 9880
+    },
+    {
+      "epoch": 1.2449853575589633,
+      "grad_norm": 0.30196666717529297,
+      "learning_rate": 0.000218103472010303,
+      "loss": 0.4265,
+      "step": 9885
+    },
+    {
+      "epoch": 1.2456151399691406,
+      "grad_norm": 0.3044353127479553,
+      "learning_rate": 0.0002180055136593956,
+      "loss": 0.48,
+      "step": 9890
+    },
+    {
+      "epoch": 1.246244922379318,
+      "grad_norm": 0.31633850932121277,
+      "learning_rate": 0.000217907518787974,
+      "loss": 0.4708,
+      "step": 9895
+    },
+    {
+      "epoch": 1.2468747047894952,
+      "grad_norm": 0.29174062609672546,
+      "learning_rate": 0.0002178094874486636,
+      "loss": 0.4135,
+      "step": 9900
+    },
+    {
+      "epoch": 1.2475044871996726,
+      "grad_norm": 0.33092647790908813,
+      "learning_rate": 0.00021771141969410956,
+      "loss": 0.4541,
+      "step": 9905
+    },
+    {
+      "epoch": 1.2481342696098499,
+      "grad_norm": 0.30151379108428955,
+      "learning_rate": 0.00021761331557697635,
+      "loss": 0.4397,
+      "step": 9910
+    },
+    {
+      "epoch": 1.2487640520200272,
+      "grad_norm": 0.31203630566596985,
+      "learning_rate": 0.00021751517514994836,
+      "loss": 0.454,
+      "step": 9915
+    },
+    {
+      "epoch": 1.2493938344302045,
+      "grad_norm": 0.30847153067588806,
+      "learning_rate": 0.00021741699846572902,
+      "loss": 0.4309,
+      "step": 9920
+    },
+    {
+      "epoch": 1.2500236168403815,
+      "grad_norm": 0.2937026619911194,
+      "learning_rate": 0.00021731878557704158,
+      "loss": 0.4206,
+      "step": 9925
+    },
+    {
+      "epoch": 1.2506533992505589,
+      "grad_norm": 0.2875721752643585,
+      "learning_rate": 0.0002172205365366285,
+      "loss": 0.4385,
+      "step": 9930
+    },
+    {
+      "epoch": 1.2512831816607362,
+      "grad_norm": 0.2834903299808502,
+      "learning_rate": 0.00021712225139725188,
+      "loss": 0.423,
+      "step": 9935
+    },
+    {
+      "epoch": 1.2519129640709135,
+      "grad_norm": 0.3069617748260498,
+      "learning_rate": 0.000217023930211693,
+      "loss": 0.4536,
+      "step": 9940
+    },
+    {
+      "epoch": 1.2525427464810908,
+      "grad_norm": 0.32263246178627014,
+      "learning_rate": 0.0002169255730327526,
+      "loss": 0.4281,
+      "step": 9945
+    },
+    {
+      "epoch": 1.253172528891268,
+      "grad_norm": 0.2980237603187561,
+      "learning_rate": 0.00021682717991325075,
+      "loss": 0.4163,
+      "step": 9950
+    },
+    {
+      "epoch": 1.2538023113014454,
+      "grad_norm": 0.3552669584751129,
+      "learning_rate": 0.0002167287509060268,
+      "loss": 0.4378,
+      "step": 9955
+    },
+    {
+      "epoch": 1.2544320937116227,
+      "grad_norm": 0.3207598924636841,
+      "learning_rate": 0.00021663028606393932,
+      "loss": 0.4411,
+      "step": 9960
+    },
+    {
+      "epoch": 1.2550618761218,
+      "grad_norm": 0.3187711238861084,
+      "learning_rate": 0.0002165317854398663,
+      "loss": 0.4384,
+      "step": 9965
+    },
+    {
+      "epoch": 1.2556916585319773,
+      "grad_norm": 0.3156946897506714,
+      "learning_rate": 0.00021643324908670472,
+      "loss": 0.4227,
+      "step": 9970
+    },
+    {
+      "epoch": 1.2563214409421546,
+      "grad_norm": 0.3305997848510742,
+      "learning_rate": 0.00021633467705737085,
+      "loss": 0.4521,
+      "step": 9975
+    },
+    {
+      "epoch": 1.2569512233523317,
+      "grad_norm": 0.2964983880519867,
+      "learning_rate": 0.00021623606940480015,
+      "loss": 0.4373,
+      "step": 9980
+    },
+    {
+      "epoch": 1.257581005762509,
+      "grad_norm": 0.29807519912719727,
+      "learning_rate": 0.00021613742618194727,
+      "loss": 0.4591,
+      "step": 9985
+    },
+    {
+      "epoch": 1.2582107881726863,
+      "grad_norm": 0.29127413034439087,
+      "learning_rate": 0.00021603874744178576,
+      "loss": 0.43,
+      "step": 9990
+    },
+    {
+      "epoch": 1.2588405705828636,
+      "grad_norm": 0.339418888092041,
+      "learning_rate": 0.00021594003323730836,
+      "loss": 0.4407,
+      "step": 9995
+    },
+    {
+      "epoch": 1.2594703529930409,
+      "grad_norm": 0.3419913053512573,
+      "learning_rate": 0.0002158412836215269,
+      "loss": 0.4678,
+      "step": 10000
+    },
+    {
+      "epoch": 1.2594703529930409,
+      "eval_loss": 0.30844178795814514,
+      "eval_runtime": 6.157,
+      "eval_samples_per_second": 162.416,
+      "eval_steps_per_second": 10.232,
+      "step": 10000
+    },
+    {
+      "epoch": 1.2601001354032182,
+      "grad_norm": 0.3139461576938629,
+      "learning_rate": 0.00021574249864747216,
+      "loss": 0.4491,
+      "step": 10005
+    },
+    {
+      "epoch": 1.2607299178133955,
+      "grad_norm": 0.319892555475235,
+      "learning_rate": 0.00021564367836819393,
+      "loss": 0.4648,
+      "step": 10010
+    },
+    {
+      "epoch": 1.2613597002235728,
+      "grad_norm": 0.30732426047325134,
+      "learning_rate": 0.00021554482283676093,
+      "loss": 0.4113,
+      "step": 10015
+    },
+    {
+      "epoch": 1.26198948263375,
+      "grad_norm": 0.3234427571296692,
+      "learning_rate": 0.00021544593210626092,
+      "loss": 0.4461,
+      "step": 10020
+    },
+    {
+      "epoch": 1.2626192650439272,
+      "grad_norm": 0.3298225998878479,
+      "learning_rate": 0.00021534700622980038,
+      "loss": 0.4487,
+      "step": 10025
+    },
+    {
+      "epoch": 1.2632490474541047,
+      "grad_norm": 0.3394641578197479,
+      "learning_rate": 0.0002152480452605048,
+      "loss": 0.4653,
+      "step": 10030
+    },
+    {
+      "epoch": 1.2638788298642818,
+      "grad_norm": 0.29091107845306396,
+      "learning_rate": 0.00021514904925151854,
+      "loss": 0.4639,
+      "step": 10035
+    },
+    {
+      "epoch": 1.264508612274459,
+      "grad_norm": 0.27975961565971375,
+      "learning_rate": 0.00021505001825600461,
+      "loss": 0.4094,
+      "step": 10040
+    },
+    {
+      "epoch": 1.2651383946846364,
+      "grad_norm": 0.2882293164730072,
+      "learning_rate": 0.00021495095232714503,
+      "loss": 0.4212,
+      "step": 10045
+    },
+    {
+      "epoch": 1.2657681770948137,
+      "grad_norm": 0.31701260805130005,
+      "learning_rate": 0.0002148518515181404,
+      "loss": 0.4427,
+      "step": 10050
+    },
+    {
+      "epoch": 1.266397959504991,
+      "grad_norm": 0.33051052689552307,
+      "learning_rate": 0.00021475271588221014,
+      "loss": 0.4331,
+      "step": 10055
+    },
+    {
+      "epoch": 1.2670277419151683,
+      "grad_norm": 0.32075920701026917,
+      "learning_rate": 0.00021465354547259234,
+      "loss": 0.4486,
+      "step": 10060
+    },
+    {
+      "epoch": 1.2676575243253456,
+      "grad_norm": 0.3044838309288025,
+      "learning_rate": 0.00021455434034254375,
+      "loss": 0.4141,
+      "step": 10065
+    },
+    {
+      "epoch": 1.268287306735523,
+      "grad_norm": 0.31618407368659973,
+      "learning_rate": 0.00021445510054533983,
+      "loss": 0.446,
+      "step": 10070
+    },
+    {
+      "epoch": 1.2689170891457002,
+      "grad_norm": 0.3025960624217987,
+      "learning_rate": 0.0002143558261342746,
+      "loss": 0.4233,
+      "step": 10075
+    },
+    {
+      "epoch": 1.2695468715558773,
+      "grad_norm": 0.2974034249782562,
+      "learning_rate": 0.0002142565171626607,
+      "loss": 0.4078,
+      "step": 10080
+    },
+    {
+      "epoch": 1.2701766539660548,
+      "grad_norm": 0.34097397327423096,
+      "learning_rate": 0.0002141571736838293,
+      "loss": 0.4555,
+      "step": 10085
+    },
+    {
+      "epoch": 1.270806436376232,
+      "grad_norm": 0.30995890498161316,
+      "learning_rate": 0.0002140577957511302,
+      "loss": 0.4388,
+      "step": 10090
+    },
+    {
+      "epoch": 1.2714362187864092,
+      "grad_norm": 0.24191588163375854,
+      "learning_rate": 0.00021395838341793145,
+      "loss": 0.4114,
+      "step": 10095
+    },
+    {
+      "epoch": 1.2720660011965865,
+      "grad_norm": 0.31779953837394714,
+      "learning_rate": 0.00021385893673761986,
+      "loss": 0.4169,
+      "step": 10100
+    },
+    {
+      "epoch": 1.2726957836067638,
+      "grad_norm": 0.31599584221839905,
+      "learning_rate": 0.0002137594557636006,
+      "loss": 0.4081,
+      "step": 10105
+    },
+    {
+      "epoch": 1.2733255660169411,
+      "grad_norm": 0.31904011964797974,
+      "learning_rate": 0.00021365994054929713,
+      "loss": 0.4406,
+      "step": 10110
+    },
+    {
+      "epoch": 1.2739553484271184,
+      "grad_norm": 0.2923012375831604,
+      "learning_rate": 0.00021356039114815145,
+      "loss": 0.4335,
+      "step": 10115
+    },
+    {
+      "epoch": 1.2745851308372957,
+      "grad_norm": 0.27983418107032776,
+      "learning_rate": 0.00021346080761362385,
+      "loss": 0.4039,
+      "step": 10120
+    },
+    {
+      "epoch": 1.275214913247473,
+      "grad_norm": 0.29870182275772095,
+      "learning_rate": 0.000213361189999193,
+      "loss": 0.4311,
+      "step": 10125
+    },
+    {
+      "epoch": 1.2758446956576504,
+      "grad_norm": 0.3060225546360016,
+      "learning_rate": 0.00021326153835835574,
+      "loss": 0.4722,
+      "step": 10130
+    },
+    {
+      "epoch": 1.2764744780678274,
+      "grad_norm": 0.38860756158828735,
+      "learning_rate": 0.00021316185274462734,
+      "loss": 0.4276,
+      "step": 10135
+    },
+    {
+      "epoch": 1.277104260478005,
+      "grad_norm": 0.32171720266342163,
+      "learning_rate": 0.0002130621332115413,
+      "loss": 0.4334,
+      "step": 10140
+    },
+    {
+      "epoch": 1.277734042888182,
+      "grad_norm": 0.2947072684764862,
+      "learning_rate": 0.00021296237981264916,
+      "loss": 0.411,
+      "step": 10145
+    },
+    {
+      "epoch": 1.2783638252983593,
+      "grad_norm": 0.2904439866542816,
+      "learning_rate": 0.00021286259260152088,
+      "loss": 0.4222,
+      "step": 10150
+    },
+    {
+      "epoch": 1.2789936077085367,
+      "grad_norm": 0.2517947554588318,
+      "learning_rate": 0.00021276277163174444,
+      "loss": 0.4336,
+      "step": 10155
+    },
+    {
+      "epoch": 1.279623390118714,
+      "grad_norm": 0.295692503452301,
+      "learning_rate": 0.00021266291695692602,
+      "loss": 0.4617,
+      "step": 10160
+    },
+    {
+      "epoch": 1.2802531725288913,
+      "grad_norm": 0.3214627802371979,
+      "learning_rate": 0.00021256302863068976,
+      "loss": 0.4327,
+      "step": 10165
+    },
+    {
+      "epoch": 1.2808829549390686,
+      "grad_norm": 0.3030719459056854,
+      "learning_rate": 0.00021246310670667808,
+      "loss": 0.4289,
+      "step": 10170
+    },
+    {
+      "epoch": 1.2815127373492459,
+      "grad_norm": 0.32924139499664307,
+      "learning_rate": 0.00021236315123855128,
+      "loss": 0.4391,
+      "step": 10175
+    },
+    {
+      "epoch": 1.2821425197594232,
+      "grad_norm": 0.2978973984718323,
+      "learning_rate": 0.00021226316227998773,
+      "loss": 0.4356,
+      "step": 10180
+    },
+    {
+      "epoch": 1.2827723021696005,
+      "grad_norm": 0.289858341217041,
+      "learning_rate": 0.00021216313988468375,
+      "loss": 0.4302,
+      "step": 10185
+    },
+    {
+      "epoch": 1.2834020845797776,
+      "grad_norm": 0.28235578536987305,
+      "learning_rate": 0.00021206308410635376,
+      "loss": 0.4581,
+      "step": 10190
+    },
+    {
+      "epoch": 1.284031866989955,
+      "grad_norm": 0.28610706329345703,
+      "learning_rate": 0.0002119629949987299,
+      "loss": 0.4233,
+      "step": 10195
+    },
+    {
+      "epoch": 1.2846616494001322,
+      "grad_norm": 0.347464382648468,
+      "learning_rate": 0.00021186287261556238,
+      "loss": 0.4191,
+      "step": 10200
+    },
+    {
+      "epoch": 1.2852914318103095,
+      "grad_norm": 0.3228091299533844,
+      "learning_rate": 0.00021176271701061914,
+      "loss": 0.4162,
+      "step": 10205
+    },
+    {
+      "epoch": 1.2859212142204868,
+      "grad_norm": 0.34487780928611755,
+      "learning_rate": 0.00021166252823768606,
+      "loss": 0.4383,
+      "step": 10210
+    },
+    {
+      "epoch": 1.286550996630664,
+      "grad_norm": 0.34411466121673584,
+      "learning_rate": 0.00021156230635056676,
+      "loss": 0.4532,
+      "step": 10215
+    },
+    {
+      "epoch": 1.2871807790408414,
+      "grad_norm": 0.38219863176345825,
+      "learning_rate": 0.00021146205140308273,
+      "loss": 0.4656,
+      "step": 10220
+    },
+    {
+      "epoch": 1.2878105614510187,
+      "grad_norm": 0.3240879774093628,
+      "learning_rate": 0.00021136176344907322,
+      "loss": 0.4174,
+      "step": 10225
+    },
+    {
+      "epoch": 1.288440343861196,
+      "grad_norm": 0.34157487750053406,
+      "learning_rate": 0.00021126144254239503,
+      "loss": 0.4297,
+      "step": 10230
+    },
+    {
+      "epoch": 1.2890701262713733,
+      "grad_norm": 0.2788861095905304,
+      "learning_rate": 0.00021116108873692286,
+      "loss": 0.429,
+      "step": 10235
+    },
+    {
+      "epoch": 1.2896999086815506,
+      "grad_norm": 0.28119325637817383,
+      "learning_rate": 0.00021106070208654895,
+      "loss": 0.4145,
+      "step": 10240
+    },
+    {
+      "epoch": 1.2903296910917277,
+      "grad_norm": 0.32004043459892273,
+      "learning_rate": 0.00021096028264518325,
+      "loss": 0.4361,
+      "step": 10245
+    },
+    {
+      "epoch": 1.2909594735019052,
+      "grad_norm": 0.3054758310317993,
+      "learning_rate": 0.0002108598304667533,
+      "loss": 0.4331,
+      "step": 10250
+    },
+    {
+      "epoch": 1.2915892559120823,
+      "grad_norm": 0.3827783167362213,
+      "learning_rate": 0.0002107593456052042,
+      "loss": 0.4246,
+      "step": 10255
+    },
+    {
+      "epoch": 1.2922190383222596,
+      "grad_norm": 0.3008691370487213,
+      "learning_rate": 0.00021065882811449862,
+      "loss": 0.4448,
+      "step": 10260
+    },
+    {
+      "epoch": 1.292848820732437,
+      "grad_norm": 0.3227977752685547,
+      "learning_rate": 0.00021055827804861675,
+      "loss": 0.4308,
+      "step": 10265
+    },
+    {
+      "epoch": 1.2934786031426142,
+      "grad_norm": 0.32592520117759705,
+      "learning_rate": 0.00021045769546155623,
+      "loss": 0.4472,
+      "step": 10270
+    },
+    {
+      "epoch": 1.2941083855527915,
+      "grad_norm": 0.30866268277168274,
+      "learning_rate": 0.00021035708040733231,
+      "loss": 0.4193,
+      "step": 10275
+    },
+    {
+      "epoch": 1.2947381679629688,
+      "grad_norm": 0.36590054631233215,
+      "learning_rate": 0.0002102564329399775,
+      "loss": 0.4554,
+      "step": 10280
+    },
+    {
+      "epoch": 1.2953679503731461,
+      "grad_norm": 0.34002235531806946,
+      "learning_rate": 0.00021015575311354175,
+      "loss": 0.465,
+      "step": 10285
+    },
+    {
+      "epoch": 1.2959977327833234,
+      "grad_norm": 0.26847660541534424,
+      "learning_rate": 0.00021005504098209248,
+      "loss": 0.4226,
+      "step": 10290
+    },
+    {
+      "epoch": 1.2966275151935007,
+      "grad_norm": 0.2904103398323059,
+      "learning_rate": 0.00020995429659971445,
+      "loss": 0.4135,
+      "step": 10295
+    },
+    {
+      "epoch": 1.2972572976036778,
+      "grad_norm": 0.2799352705478668,
+      "learning_rate": 0.00020985352002050962,
+      "loss": 0.4241,
+      "step": 10300
+    },
+    {
+      "epoch": 1.2978870800138553,
+      "grad_norm": 0.3527425229549408,
+      "learning_rate": 0.00020975271129859734,
+      "loss": 0.4397,
+      "step": 10305
+    },
+    {
+      "epoch": 1.2985168624240324,
+      "grad_norm": 0.30795904994010925,
+      "learning_rate": 0.00020965187048811417,
+      "loss": 0.4248,
+      "step": 10310
+    },
+    {
+      "epoch": 1.2991466448342097,
+      "grad_norm": 0.31814008951187134,
+      "learning_rate": 0.00020955099764321402,
+      "loss": 0.4501,
+      "step": 10315
+    },
+    {
+      "epoch": 1.299776427244387,
+      "grad_norm": 0.29917100071907043,
+      "learning_rate": 0.0002094500928180678,
+      "loss": 0.4511,
+      "step": 10320
+    },
+    {
+      "epoch": 1.3004062096545643,
+      "grad_norm": 0.32853367924690247,
+      "learning_rate": 0.00020934915606686373,
+      "loss": 0.4055,
+      "step": 10325
+    },
+    {
+      "epoch": 1.3010359920647416,
+      "grad_norm": 0.420550137758255,
+      "learning_rate": 0.00020924818744380723,
+      "loss": 0.4417,
+      "step": 10330
+    },
+    {
+      "epoch": 1.301665774474919,
+      "grad_norm": 0.3183051347732544,
+      "learning_rate": 0.0002091471870031207,
+      "loss": 0.4256,
+      "step": 10335
+    },
+    {
+      "epoch": 1.3022955568850962,
+      "grad_norm": 0.30520761013031006,
+      "learning_rate": 0.00020904615479904362,
+      "loss": 0.4213,
+      "step": 10340
+    },
+    {
+      "epoch": 1.3029253392952735,
+      "grad_norm": 0.3484478294849396,
+      "learning_rate": 0.0002089450908858327,
+      "loss": 0.4202,
+      "step": 10345
+    },
+    {
+      "epoch": 1.3035551217054508,
+      "grad_norm": 0.3063777983188629,
+      "learning_rate": 0.00020884399531776154,
+      "loss": 0.4121,
+      "step": 10350
+    },
+    {
+      "epoch": 1.304184904115628,
+      "grad_norm": 0.35436901450157166,
+      "learning_rate": 0.00020874286814912072,
+      "loss": 0.4351,
+      "step": 10355
+    },
+    {
+      "epoch": 1.3048146865258055,
+      "grad_norm": 0.3233969211578369,
+      "learning_rate": 0.00020864170943421786,
+      "loss": 0.4326,
+      "step": 10360
+    },
+    {
+      "epoch": 1.3054444689359825,
+      "grad_norm": 0.34073448181152344,
+      "learning_rate": 0.0002085405192273776,
+      "loss": 0.4454,
+      "step": 10365
+    },
+    {
+      "epoch": 1.3060742513461598,
+      "grad_norm": 0.28455135226249695,
+      "learning_rate": 0.00020843929758294121,
+      "loss": 0.4511,
+      "step": 10370
+    },
+    {
+      "epoch": 1.3067040337563371,
+      "grad_norm": 0.31585589051246643,
+      "learning_rate": 0.0002083380445552672,
+      "loss": 0.4258,
+      "step": 10375
+    },
+    {
+      "epoch": 1.3073338161665145,
+      "grad_norm": 0.31528952717781067,
+      "learning_rate": 0.00020823676019873064,
+      "loss": 0.424,
+      "step": 10380
+    },
+    {
+      "epoch": 1.3079635985766918,
+      "grad_norm": 0.3014485836029053,
+      "learning_rate": 0.00020813544456772362,
+      "loss": 0.4429,
+      "step": 10385
+    },
+    {
+      "epoch": 1.308593380986869,
+      "grad_norm": 0.2870473861694336,
+      "learning_rate": 0.00020803409771665484,
+      "loss": 0.439,
+      "step": 10390
+    },
+    {
+      "epoch": 1.3092231633970464,
+      "grad_norm": 0.2971458435058594,
+      "learning_rate": 0.00020793271969994997,
+      "loss": 0.4233,
+      "step": 10395
+    },
+    {
+      "epoch": 1.3098529458072237,
+      "grad_norm": 0.2853131890296936,
+      "learning_rate": 0.00020783131057205135,
+      "loss": 0.4164,
+      "step": 10400
+    },
+    {
+      "epoch": 1.310482728217401,
+      "grad_norm": 0.29392004013061523,
+      "learning_rate": 0.00020772987038741793,
+      "loss": 0.4234,
+      "step": 10405
+    },
+    {
+      "epoch": 1.311112510627578,
+      "grad_norm": 0.2874060273170471,
+      "learning_rate": 0.00020762839920052543,
+      "loss": 0.4413,
+      "step": 10410
+    },
+    {
+      "epoch": 1.3117422930377556,
+      "grad_norm": 0.2806376516819,
+      "learning_rate": 0.00020752689706586615,
+      "loss": 0.4223,
+      "step": 10415
+    },
+    {
+      "epoch": 1.3123720754479327,
+      "grad_norm": 0.28510767221450806,
+      "learning_rate": 0.00020742536403794908,
+      "loss": 0.4183,
+      "step": 10420
+    },
+    {
+      "epoch": 1.31300185785811,
+      "grad_norm": 0.3087919056415558,
+      "learning_rate": 0.00020732380017129983,
+      "loss": 0.4241,
+      "step": 10425
+    },
+    {
+      "epoch": 1.3136316402682873,
+      "grad_norm": 0.2965323328971863,
+      "learning_rate": 0.00020722220552046048,
+      "loss": 0.4225,
+      "step": 10430
+    },
+    {
+      "epoch": 1.3142614226784646,
+      "grad_norm": 0.2907772660255432,
+      "learning_rate": 0.00020712058013998963,
+      "loss": 0.4176,
+      "step": 10435
+    },
+    {
+      "epoch": 1.3148912050886419,
+      "grad_norm": 0.3242434859275818,
+      "learning_rate": 0.0002070189240844625,
+      "loss": 0.4377,
+      "step": 10440
+    },
+    {
+      "epoch": 1.3155209874988192,
+      "grad_norm": 0.28129857778549194,
+      "learning_rate": 0.00020691723740847066,
+      "loss": 0.425,
+      "step": 10445
+    },
+    {
+      "epoch": 1.3161507699089965,
+      "grad_norm": 0.3053089380264282,
+      "learning_rate": 0.00020681552016662224,
+      "loss": 0.4066,
+      "step": 10450
+    },
+    {
+      "epoch": 1.3167805523191738,
+      "grad_norm": 0.27167361974716187,
+      "learning_rate": 0.00020671377241354168,
+      "loss": 0.4458,
+      "step": 10455
+    },
+    {
+      "epoch": 1.317410334729351,
+      "grad_norm": 0.29331174492836,
+      "learning_rate": 0.00020661199420386986,
+      "loss": 0.427,
+      "step": 10460
+    },
+    {
+      "epoch": 1.3180401171395282,
+      "grad_norm": 0.329908162355423,
+      "learning_rate": 0.00020651018559226394,
+      "loss": 0.4292,
+      "step": 10465
+    },
+    {
+      "epoch": 1.3186698995497057,
+      "grad_norm": 0.32669904828071594,
+      "learning_rate": 0.0002064083466333976,
+      "loss": 0.4118,
+      "step": 10470
+    },
+    {
+      "epoch": 1.3192996819598828,
+      "grad_norm": 0.35706159472465515,
+      "learning_rate": 0.00020630647738196058,
+      "loss": 0.4433,
+      "step": 10475
+    },
+    {
+      "epoch": 1.31992946437006,
+      "grad_norm": 0.3119877278804779,
+      "learning_rate": 0.00020620457789265905,
+      "loss": 0.4206,
+      "step": 10480
+    },
+    {
+      "epoch": 1.3205592467802374,
+      "grad_norm": 0.34798958897590637,
+      "learning_rate": 0.00020610264822021532,
+      "loss": 0.39,
+      "step": 10485
+    },
+    {
+      "epoch": 1.3211890291904147,
+      "grad_norm": 0.36972302198410034,
+      "learning_rate": 0.000206000688419368,
+      "loss": 0.4402,
+      "step": 10490
+    },
+    {
+      "epoch": 1.321818811600592,
+      "grad_norm": 0.27949050068855286,
+      "learning_rate": 0.00020589869854487175,
+      "loss": 0.4221,
+      "step": 10495
+    },
+    {
+      "epoch": 1.3224485940107693,
+      "grad_norm": 0.30757853388786316,
+      "learning_rate": 0.00020579667865149758,
+      "loss": 0.4402,
+      "step": 10500
+    },
+    {
+      "epoch": 1.3230783764209466,
+      "grad_norm": 0.3018808364868164,
+      "learning_rate": 0.0002056946287940324,
+      "loss": 0.4088,
+      "step": 10505
+    },
+    {
+      "epoch": 1.323708158831124,
+      "grad_norm": 0.2630440592765808,
+      "learning_rate": 0.00020559254902727942,
+      "loss": 0.4062,
+      "step": 10510
+    },
+    {
+      "epoch": 1.3243379412413012,
+      "grad_norm": 0.3145885169506073,
+      "learning_rate": 0.00020549043940605767,
+      "loss": 0.4301,
+      "step": 10515
+    },
+    {
+      "epoch": 1.3249677236514783,
+      "grad_norm": 0.3040730655193329,
+      "learning_rate": 0.0002053882999852025,
+      "loss": 0.4267,
+      "step": 10520
+    },
+    {
+      "epoch": 1.3255975060616558,
+      "grad_norm": 0.2861897945404053,
+      "learning_rate": 0.00020528613081956498,
+      "loss": 0.4115,
+      "step": 10525
+    },
+    {
+      "epoch": 1.326227288471833,
+      "grad_norm": 0.2938830256462097,
+      "learning_rate": 0.00020518393196401234,
+      "loss": 0.4315,
+      "step": 10530
+    },
+    {
+      "epoch": 1.3268570708820102,
+      "grad_norm": 0.24550281465053558,
+      "learning_rate": 0.0002050817034734277,
+      "loss": 0.4181,
+      "step": 10535
+    },
+    {
+      "epoch": 1.3274868532921875,
+      "grad_norm": 0.30074000358581543,
+      "learning_rate": 0.00020497944540271017,
+      "loss": 0.4016,
+      "step": 10540
+    },
+    {
+      "epoch": 1.3281166357023648,
+      "grad_norm": 0.34675145149230957,
+      "learning_rate": 0.0002048771578067745,
+      "loss": 0.4157,
+      "step": 10545
+    },
+    {
+      "epoch": 1.3287464181125421,
+      "grad_norm": 0.3144848644733429,
+      "learning_rate": 0.00020477484074055157,
+      "loss": 0.4024,
+      "step": 10550
+    },
+    {
+      "epoch": 1.3293762005227194,
+      "grad_norm": 0.32153722643852234,
+      "learning_rate": 0.00020467249425898805,
+      "loss": 0.4114,
+      "step": 10555
+    },
+    {
+      "epoch": 1.3300059829328967,
+      "grad_norm": 0.301707923412323,
+      "learning_rate": 0.0002045701184170462,
+      "loss": 0.423,
+      "step": 10560
+    },
+    {
+      "epoch": 1.330635765343074,
+      "grad_norm": 0.25224459171295166,
+      "learning_rate": 0.00020446771326970424,
+      "loss": 0.4037,
+      "step": 10565
+    },
+    {
+      "epoch": 1.3312655477532513,
+      "grad_norm": 0.3072243928909302,
+      "learning_rate": 0.00020436527887195607,
+      "loss": 0.4279,
+      "step": 10570
+    },
+    {
+      "epoch": 1.3318953301634284,
+      "grad_norm": 0.36949509382247925,
+      "learning_rate": 0.00020426281527881137,
+      "loss": 0.4259,
+      "step": 10575
+    },
+    {
+      "epoch": 1.332525112573606,
+      "grad_norm": 0.30465519428253174,
+      "learning_rate": 0.00020416032254529535,
+      "loss": 0.457,
+      "step": 10580
+    },
+    {
+      "epoch": 1.333154894983783,
+      "grad_norm": 0.2719140350818634,
+      "learning_rate": 0.00020405780072644896,
+      "loss": 0.3927,
+      "step": 10585
+    },
+    {
+      "epoch": 1.3337846773939603,
+      "grad_norm": 0.33556681871414185,
+      "learning_rate": 0.00020395524987732876,
+      "loss": 0.4341,
+      "step": 10590
+    },
+    {
+      "epoch": 1.3344144598041376,
+      "grad_norm": 0.3145639896392822,
+      "learning_rate": 0.0002038526700530069,
+      "loss": 0.4176,
+      "step": 10595
+    },
+    {
+      "epoch": 1.335044242214315,
+      "grad_norm": 0.31328147649765015,
+      "learning_rate": 0.00020375006130857111,
+      "loss": 0.4332,
+      "step": 10600
+    },
+    {
+      "epoch": 1.3356740246244923,
+      "grad_norm": 0.3016543388366699,
+      "learning_rate": 0.00020364742369912464,
+      "loss": 0.4173,
+      "step": 10605
+    },
+    {
+      "epoch": 1.3363038070346696,
+      "grad_norm": 0.31259703636169434,
+      "learning_rate": 0.0002035447572797862,
+      "loss": 0.4091,
+      "step": 10610
+    },
+    {
+      "epoch": 1.3369335894448469,
+      "grad_norm": 0.34624606370925903,
+      "learning_rate": 0.00020344206210569,
+      "loss": 0.4408,
+      "step": 10615
+    },
+    {
+      "epoch": 1.3375633718550242,
+      "grad_norm": 0.3144773542881012,
+      "learning_rate": 0.00020333933823198566,
+      "loss": 0.3863,
+      "step": 10620
+    },
+    {
+      "epoch": 1.3381931542652015,
+      "grad_norm": 0.3231208026409149,
+      "learning_rate": 0.00020323658571383833,
+      "loss": 0.4151,
+      "step": 10625
+    },
+    {
+      "epoch": 1.3388229366753785,
+      "grad_norm": 0.3022227883338928,
+      "learning_rate": 0.00020313380460642842,
+      "loss": 0.4108,
+      "step": 10630
+    },
+    {
+      "epoch": 1.339452719085556,
+      "grad_norm": 0.2899850606918335,
+      "learning_rate": 0.00020303099496495172,
+      "loss": 0.412,
+      "step": 10635
+    },
+    {
+      "epoch": 1.3400825014957332,
+      "grad_norm": 0.31005537509918213,
+      "learning_rate": 0.00020292815684461936,
+      "loss": 0.4114,
+      "step": 10640
+    },
+    {
+      "epoch": 1.3407122839059105,
+      "grad_norm": 0.29457420110702515,
+      "learning_rate": 0.00020282529030065784,
+      "loss": 0.4292,
+      "step": 10645
+    },
+    {
+      "epoch": 1.3413420663160878,
+      "grad_norm": 0.31712374091148376,
+      "learning_rate": 0.00020272239538830867,
+      "loss": 0.4029,
+      "step": 10650
+    },
+    {
+      "epoch": 1.341971848726265,
+      "grad_norm": 0.3228032886981964,
+      "learning_rate": 0.00020261947216282896,
+      "loss": 0.414,
+      "step": 10655
+    },
+    {
+      "epoch": 1.3426016311364424,
+      "grad_norm": 0.305351197719574,
+      "learning_rate": 0.00020251652067949068,
+      "loss": 0.4233,
+      "step": 10660
+    },
+    {
+      "epoch": 1.3432314135466197,
+      "grad_norm": 0.30317017436027527,
+      "learning_rate": 0.00020241354099358123,
+      "loss": 0.3816,
+      "step": 10665
+    },
+    {
+      "epoch": 1.343861195956797,
+      "grad_norm": 0.3036525845527649,
+      "learning_rate": 0.00020231053316040293,
+      "loss": 0.4115,
+      "step": 10670
+    },
+    {
+      "epoch": 1.3444909783669743,
+      "grad_norm": 0.33367687463760376,
+      "learning_rate": 0.00020220749723527353,
+      "loss": 0.449,
+      "step": 10675
+    },
+    {
+      "epoch": 1.3451207607771516,
+      "grad_norm": 0.28938767313957214,
+      "learning_rate": 0.00020210443327352553,
+      "loss": 0.3919,
+      "step": 10680
+    },
+    {
+      "epoch": 1.3457505431873287,
+      "grad_norm": 0.2946431338787079,
+      "learning_rate": 0.00020200134133050666,
+      "loss": 0.4043,
+      "step": 10685
+    },
+    {
+      "epoch": 1.3463803255975062,
+      "grad_norm": 0.31588709354400635,
+      "learning_rate": 0.00020189822146157962,
+      "loss": 0.4136,
+      "step": 10690
+    },
+    {
+      "epoch": 1.3470101080076833,
+      "grad_norm": 0.2830824851989746,
+      "learning_rate": 0.00020179507372212224,
+      "loss": 0.4164,
+      "step": 10695
+    },
+    {
+      "epoch": 1.3476398904178606,
+      "grad_norm": 0.31364426016807556,
+      "learning_rate": 0.0002016918981675271,
+      "loss": 0.4197,
+      "step": 10700
+    },
+    {
+      "epoch": 1.348269672828038,
+      "grad_norm": 0.32086437940597534,
+      "learning_rate": 0.00020158869485320194,
+      "loss": 0.4346,
+      "step": 10705
+    },
+    {
+      "epoch": 1.3488994552382152,
+      "grad_norm": 0.30549678206443787,
+      "learning_rate": 0.0002014854638345692,
+      "loss": 0.4134,
+      "step": 10710
+    },
+    {
+      "epoch": 1.3495292376483925,
+      "grad_norm": 0.2996455132961273,
+      "learning_rate": 0.00020138220516706634,
+      "loss": 0.3846,
+      "step": 10715
+    },
+    {
+      "epoch": 1.3501590200585698,
+      "grad_norm": 0.3013511002063751,
+      "learning_rate": 0.00020127891890614556,
+      "loss": 0.3994,
+      "step": 10720
+    },
+    {
+      "epoch": 1.350788802468747,
+      "grad_norm": 0.28055283427238464,
+      "learning_rate": 0.00020117560510727402,
+      "loss": 0.4163,
+      "step": 10725
+    },
+    {
+      "epoch": 1.3514185848789244,
+      "grad_norm": 0.3024522364139557,
+      "learning_rate": 0.00020107226382593357,
+      "loss": 0.4042,
+      "step": 10730
+    },
+    {
+      "epoch": 1.3520483672891017,
+      "grad_norm": 0.28080272674560547,
+      "learning_rate": 0.00020096889511762083,
+      "loss": 0.4176,
+      "step": 10735
+    },
+    {
+      "epoch": 1.3526781496992788,
+      "grad_norm": 0.3069353997707367,
+      "learning_rate": 0.00020086549903784715,
+      "loss": 0.4189,
+      "step": 10740
+    },
+    {
+      "epoch": 1.353307932109456,
+      "grad_norm": 0.2898117005825043,
+      "learning_rate": 0.00020076207564213866,
+      "loss": 0.4342,
+      "step": 10745
+    },
+    {
+      "epoch": 1.3539377145196334,
+      "grad_norm": 0.3365933299064636,
+      "learning_rate": 0.00020065862498603592,
+      "loss": 0.3944,
+      "step": 10750
+    },
+    {
+      "epoch": 1.3545674969298107,
+      "grad_norm": 0.29901427030563354,
+      "learning_rate": 0.00020055514712509446,
+      "loss": 0.4059,
+      "step": 10755
+    },
+    {
+      "epoch": 1.355197279339988,
+      "grad_norm": 0.2927230894565582,
+      "learning_rate": 0.00020045164211488417,
+      "loss": 0.4137,
+      "step": 10760
+    },
+    {
+      "epoch": 1.3558270617501653,
+      "grad_norm": 0.35867777466773987,
+      "learning_rate": 0.00020034811001098964,
+      "loss": 0.4108,
+      "step": 10765
+    },
+    {
+      "epoch": 1.3564568441603426,
+      "grad_norm": 0.2955409586429596,
+      "learning_rate": 0.00020024455086900994,
+      "loss": 0.4328,
+      "step": 10770
+    },
+    {
+      "epoch": 1.35708662657052,
+      "grad_norm": 0.29247814416885376,
+      "learning_rate": 0.00020014096474455873,
+      "loss": 0.4014,
+      "step": 10775
+    },
+    {
+      "epoch": 1.3577164089806972,
+      "grad_norm": 0.30858153104782104,
+      "learning_rate": 0.00020003735169326413,
+      "loss": 0.4112,
+      "step": 10780
+    },
+    {
+      "epoch": 1.3583461913908745,
+      "grad_norm": 0.4134693145751953,
+      "learning_rate": 0.0001999337117707687,
+      "loss": 0.4062,
+      "step": 10785
+    },
+    {
+      "epoch": 1.3589759738010518,
+      "grad_norm": 0.3120553195476532,
+      "learning_rate": 0.0001998300450327294,
+      "loss": 0.4049,
+      "step": 10790
+    },
+    {
+      "epoch": 1.359605756211229,
+      "grad_norm": 0.3146657645702362,
+      "learning_rate": 0.00019972635153481767,
+      "loss": 0.4029,
+      "step": 10795
+    },
+    {
+      "epoch": 1.3602355386214062,
+      "grad_norm": 0.2997225821018219,
+      "learning_rate": 0.00019962263133271933,
+      "loss": 0.3792,
+      "step": 10800
+    },
+    {
+      "epoch": 1.3608653210315835,
+      "grad_norm": 0.32136911153793335,
+      "learning_rate": 0.0001995188844821345,
+      "loss": 0.3987,
+      "step": 10805
+    },
+    {
+      "epoch": 1.3614951034417608,
+      "grad_norm": 0.30875489115715027,
+      "learning_rate": 0.0001994151110387775,
+      "loss": 0.4211,
+      "step": 10810
+    },
+    {
+      "epoch": 1.3621248858519381,
+      "grad_norm": 0.30939677357673645,
+      "learning_rate": 0.00019931131105837714,
+      "loss": 0.451,
+      "step": 10815
+    },
+    {
+      "epoch": 1.3627546682621154,
+      "grad_norm": 0.27874892950057983,
+      "learning_rate": 0.0001992074845966764,
+      "loss": 0.4102,
+      "step": 10820
+    },
+    {
+      "epoch": 1.3633844506722927,
+      "grad_norm": 0.28371527791023254,
+      "learning_rate": 0.00019910363170943233,
+      "loss": 0.4153,
+      "step": 10825
+    },
+    {
+      "epoch": 1.36401423308247,
+      "grad_norm": 0.2852970063686371,
+      "learning_rate": 0.00019899975245241643,
+      "loss": 0.409,
+      "step": 10830
+    },
+    {
+      "epoch": 1.3646440154926474,
+      "grad_norm": 0.300521582365036,
+      "learning_rate": 0.00019889584688141418,
+      "loss": 0.4032,
+      "step": 10835
+    },
+    {
+      "epoch": 1.3652737979028244,
+      "grad_norm": 0.30631181597709656,
+      "learning_rate": 0.00019879191505222526,
+      "loss": 0.4299,
+      "step": 10840
+    },
+    {
+      "epoch": 1.365903580313002,
+      "grad_norm": 0.3514620363712311,
+      "learning_rate": 0.00019868795702066342,
+      "loss": 0.4051,
+      "step": 10845
+    },
+    {
+      "epoch": 1.366533362723179,
+      "grad_norm": 0.27533403038978577,
+      "learning_rate": 0.00019858397284255657,
+      "loss": 0.4108,
+      "step": 10850
+    },
+    {
+      "epoch": 1.3671631451333564,
+      "grad_norm": 0.3143390119075775,
+      "learning_rate": 0.00019847996257374645,
+      "loss": 0.426,
+      "step": 10855
+    },
+    {
+      "epoch": 1.3677929275435337,
+      "grad_norm": 0.3388061821460724,
+      "learning_rate": 0.00019837592627008904,
+      "loss": 0.4163,
+      "step": 10860
+    },
+    {
+      "epoch": 1.368422709953711,
+      "grad_norm": 0.34078383445739746,
+      "learning_rate": 0.00019827186398745417,
+      "loss": 0.4015,
+      "step": 10865
+    },
+    {
+      "epoch": 1.3690524923638883,
+      "grad_norm": 0.33532068133354187,
+      "learning_rate": 0.00019816777578172582,
+      "loss": 0.4436,
+      "step": 10870
+    },
+    {
+      "epoch": 1.3696822747740656,
+      "grad_norm": 0.3230116069316864,
+      "learning_rate": 0.0001980636617088015,
+      "loss": 0.4239,
+      "step": 10875
+    },
+    {
+      "epoch": 1.3703120571842429,
+      "grad_norm": 0.31974872946739197,
+      "learning_rate": 0.00019795952182459297,
+      "loss": 0.4313,
+      "step": 10880
+    },
+    {
+      "epoch": 1.3709418395944202,
+      "grad_norm": 0.2825758159160614,
+      "learning_rate": 0.0001978553561850257,
+      "loss": 0.4045,
+      "step": 10885
+    },
+    {
+      "epoch": 1.3715716220045975,
+      "grad_norm": 0.2678980529308319,
+      "learning_rate": 0.00019775116484603908,
+      "loss": 0.3899,
+      "step": 10890
+    },
+    {
+      "epoch": 1.3722014044147746,
+      "grad_norm": 0.3492506146430969,
+      "learning_rate": 0.00019764694786358612,
+      "loss": 0.3807,
+      "step": 10895
+    },
+    {
+      "epoch": 1.372831186824952,
+      "grad_norm": 0.30808547139167786,
+      "learning_rate": 0.00019754270529363384,
+      "loss": 0.4163,
+      "step": 10900
+    },
+    {
+      "epoch": 1.3734609692351292,
+      "grad_norm": 0.30980342626571655,
+      "learning_rate": 0.0001974384371921628,
+      "loss": 0.3843,
+      "step": 10905
+    },
+    {
+      "epoch": 1.3740907516453065,
+      "grad_norm": 0.2915787100791931,
+      "learning_rate": 0.00019733414361516736,
+      "loss": 0.4208,
+      "step": 10910
+    },
+    {
+      "epoch": 1.3747205340554838,
+      "grad_norm": 0.30979228019714355,
+      "learning_rate": 0.00019722982461865555,
+      "loss": 0.4188,
+      "step": 10915
+    },
+    {
+      "epoch": 1.375350316465661,
+      "grad_norm": 0.28953999280929565,
+      "learning_rate": 0.00019712548025864918,
+      "loss": 0.3934,
+      "step": 10920
+    },
+    {
+      "epoch": 1.3759800988758384,
+      "grad_norm": 0.31495416164398193,
+      "learning_rate": 0.00019702111059118334,
+      "loss": 0.4117,
+      "step": 10925
+    },
+    {
+      "epoch": 1.3766098812860157,
+      "grad_norm": 0.38459569215774536,
+      "learning_rate": 0.00019691671567230714,
+      "loss": 0.4229,
+      "step": 10930
+    },
+    {
+      "epoch": 1.377239663696193,
+      "grad_norm": 0.31138870120048523,
+      "learning_rate": 0.00019681229555808285,
+      "loss": 0.4284,
+      "step": 10935
+    },
+    {
+      "epoch": 1.3778694461063703,
+      "grad_norm": 0.2761414051055908,
+      "learning_rate": 0.0001967078503045866,
+      "loss": 0.3838,
+      "step": 10940
+    },
+    {
+      "epoch": 1.3784992285165476,
+      "grad_norm": 0.31627506017684937,
+      "learning_rate": 0.00019660337996790772,
+      "loss": 0.4008,
+      "step": 10945
+    },
+    {
+      "epoch": 1.3791290109267247,
+      "grad_norm": 0.29025107622146606,
+      "learning_rate": 0.00019649888460414937,
+      "loss": 0.409,
+      "step": 10950
+    },
+    {
+      "epoch": 1.3797587933369022,
+      "grad_norm": 0.3379102349281311,
+      "learning_rate": 0.0001963943642694278,
+      "loss": 0.4213,
+      "step": 10955
+    },
+    {
+      "epoch": 1.3803885757470793,
+      "grad_norm": 0.3209204375743866,
+      "learning_rate": 0.00019628981901987285,
+      "loss": 0.3834,
+      "step": 10960
+    },
+    {
+      "epoch": 1.3810183581572566,
+      "grad_norm": 0.31717419624328613,
+      "learning_rate": 0.0001961852489116277,
+      "loss": 0.4499,
+      "step": 10965
+    },
+    {
+      "epoch": 1.381648140567434,
+      "grad_norm": 0.27936458587646484,
+      "learning_rate": 0.00019608065400084898,
+      "loss": 0.3987,
+      "step": 10970
+    },
+    {
+      "epoch": 1.3822779229776112,
+      "grad_norm": 0.28877684473991394,
+      "learning_rate": 0.00019597603434370637,
+      "loss": 0.4252,
+      "step": 10975
+    },
+    {
+      "epoch": 1.3829077053877885,
+      "grad_norm": 0.3423072397708893,
+      "learning_rate": 0.00019587138999638316,
+      "loss": 0.421,
+      "step": 10980
+    },
+    {
+      "epoch": 1.3835374877979658,
+      "grad_norm": 0.26486262679100037,
+      "learning_rate": 0.00019576672101507568,
+      "loss": 0.4104,
+      "step": 10985
+    },
+    {
+      "epoch": 1.3841672702081431,
+      "grad_norm": 0.2929472029209137,
+      "learning_rate": 0.00019566202745599365,
+      "loss": 0.4127,
+      "step": 10990
+    },
+    {
+      "epoch": 1.3847970526183204,
+      "grad_norm": 0.2696884870529175,
+      "learning_rate": 0.00019555730937535976,
+      "loss": 0.4067,
+      "step": 10995
+    },
+    {
+      "epoch": 1.3854268350284977,
+      "grad_norm": 0.32420167326927185,
+      "learning_rate": 0.0001954525668294102,
+      "loss": 0.4136,
+      "step": 11000
+    },
+    {
+      "epoch": 1.3854268350284977,
+      "eval_loss": 0.3039778470993042,
+      "eval_runtime": 6.1549,
+      "eval_samples_per_second": 162.472,
+      "eval_steps_per_second": 10.236,
+      "step": 11000
+    },
+    {
+      "epoch": 1.3860566174386748,
+      "grad_norm": 0.3149106502532959,
+      "learning_rate": 0.00019534779987439395,
+      "loss": 0.3954,
+      "step": 11005
+    },
+    {
+      "epoch": 1.3866863998488523,
+      "grad_norm": 0.332868367433548,
+      "learning_rate": 0.0001952430085665733,
+      "loss": 0.4178,
+      "step": 11010
+    },
+    {
+      "epoch": 1.3873161822590294,
+      "grad_norm": 0.285671591758728,
+      "learning_rate": 0.00019513819296222362,
+      "loss": 0.3788,
+      "step": 11015
+    },
+    {
+      "epoch": 1.3879459646692067,
+      "grad_norm": 0.3317325711250305,
+      "learning_rate": 0.0001950333531176332,
+      "loss": 0.4091,
+      "step": 11020
+    },
+    {
+      "epoch": 1.388575747079384,
+      "grad_norm": 0.27808326482772827,
+      "learning_rate": 0.00019492848908910356,
+      "loss": 0.4104,
+      "step": 11025
+    },
+    {
+      "epoch": 1.3892055294895613,
+      "grad_norm": 0.29725268483161926,
+      "learning_rate": 0.00019482360093294897,
+      "loss": 0.3981,
+      "step": 11030
+    },
+    {
+      "epoch": 1.3898353118997386,
+      "grad_norm": 0.24770186841487885,
+      "learning_rate": 0.0001947186887054968,
+      "loss": 0.4052,
+      "step": 11035
+    },
+    {
+      "epoch": 1.390465094309916,
+      "grad_norm": 0.31627580523490906,
+      "learning_rate": 0.00019461375246308734,
+      "loss": 0.4051,
+      "step": 11040
+    },
+    {
+      "epoch": 1.3910948767200932,
+      "grad_norm": 0.2721163332462311,
+      "learning_rate": 0.00019450879226207368,
+      "loss": 0.3962,
+      "step": 11045
+    },
+    {
+      "epoch": 1.3917246591302705,
+      "grad_norm": 0.31926798820495605,
+      "learning_rate": 0.00019440380815882187,
+      "loss": 0.3964,
+      "step": 11050
+    },
+    {
+      "epoch": 1.3923544415404479,
+      "grad_norm": 0.3047574460506439,
+      "learning_rate": 0.0001942988002097108,
+      "loss": 0.3818,
+      "step": 11055
+    },
+    {
+      "epoch": 1.392984223950625,
+      "grad_norm": 0.35394978523254395,
+      "learning_rate": 0.00019419376847113216,
+      "loss": 0.4398,
+      "step": 11060
+    },
+    {
+      "epoch": 1.3936140063608025,
+      "grad_norm": 0.2855307459831238,
+      "learning_rate": 0.00019408871299949037,
+      "loss": 0.4089,
+      "step": 11065
+    },
+    {
+      "epoch": 1.3942437887709795,
+      "grad_norm": 0.3066868484020233,
+      "learning_rate": 0.00019398363385120254,
+      "loss": 0.3987,
+      "step": 11070
+    },
+    {
+      "epoch": 1.3948735711811568,
+      "grad_norm": 0.312775194644928,
+      "learning_rate": 0.0001938785310826987,
+      "loss": 0.3794,
+      "step": 11075
+    },
+    {
+      "epoch": 1.3955033535913342,
+      "grad_norm": 0.3235652446746826,
+      "learning_rate": 0.00019377340475042136,
+      "loss": 0.3852,
+      "step": 11080
+    },
+    {
+      "epoch": 1.3961331360015115,
+      "grad_norm": 0.33732032775878906,
+      "learning_rate": 0.00019366825491082574,
+      "loss": 0.4003,
+      "step": 11085
+    },
+    {
+      "epoch": 1.3967629184116888,
+      "grad_norm": 0.33549800515174866,
+      "learning_rate": 0.00019356308162037976,
+      "loss": 0.3699,
+      "step": 11090
+    },
+    {
+      "epoch": 1.397392700821866,
+      "grad_norm": 0.3360839784145355,
+      "learning_rate": 0.00019345788493556394,
+      "loss": 0.394,
+      "step": 11095
+    },
+    {
+      "epoch": 1.3980224832320434,
+      "grad_norm": 0.3089699447154999,
+      "learning_rate": 0.00019335266491287112,
+      "loss": 0.4016,
+      "step": 11100
+    },
+    {
+      "epoch": 1.3986522656422207,
+      "grad_norm": 0.30863386392593384,
+      "learning_rate": 0.00019324742160880702,
+      "loss": 0.3973,
+      "step": 11105
+    },
+    {
+      "epoch": 1.399282048052398,
+      "grad_norm": 0.30803561210632324,
+      "learning_rate": 0.00019314215507988965,
+      "loss": 0.4119,
+      "step": 11110
+    },
+    {
+      "epoch": 1.399911830462575,
+      "grad_norm": 0.2869633138179779,
+      "learning_rate": 0.0001930368653826495,
+      "loss": 0.4098,
+      "step": 11115
+    },
+    {
+      "epoch": 1.4005416128727526,
+      "grad_norm": 0.25851666927337646,
+      "learning_rate": 0.00019293155257362957,
+      "loss": 0.4034,
+      "step": 11120
+    },
+    {
+      "epoch": 1.4011713952829297,
+      "grad_norm": 0.32763540744781494,
+      "learning_rate": 0.00019282621670938527,
+      "loss": 0.4121,
+      "step": 11125
+    },
+    {
+      "epoch": 1.401801177693107,
+      "grad_norm": 0.3531438410282135,
+      "learning_rate": 0.00019272085784648432,
+      "loss": 0.4021,
+      "step": 11130
+    },
+    {
+      "epoch": 1.4024309601032843,
+      "grad_norm": 0.27890294790267944,
+      "learning_rate": 0.00019261547604150687,
+      "loss": 0.3872,
+      "step": 11135
+    },
+    {
+      "epoch": 1.4030607425134616,
+      "grad_norm": 0.26616647839546204,
+      "learning_rate": 0.00019251007135104534,
+      "loss": 0.4293,
+      "step": 11140
+    },
+    {
+      "epoch": 1.4036905249236389,
+      "grad_norm": 0.3214140236377716,
+      "learning_rate": 0.0001924046438317045,
+      "loss": 0.3974,
+      "step": 11145
+    },
+    {
+      "epoch": 1.4043203073338162,
+      "grad_norm": 0.31075042486190796,
+      "learning_rate": 0.00019229919354010126,
+      "loss": 0.3978,
+      "step": 11150
+    },
+    {
+      "epoch": 1.4049500897439935,
+      "grad_norm": 0.31546491384506226,
+      "learning_rate": 0.00019219372053286485,
+      "loss": 0.3937,
+      "step": 11155
+    },
+    {
+      "epoch": 1.4055798721541708,
+      "grad_norm": 0.33116820454597473,
+      "learning_rate": 0.00019208822486663677,
+      "loss": 0.3779,
+      "step": 11160
+    },
+    {
+      "epoch": 1.406209654564348,
+      "grad_norm": 0.30159297585487366,
+      "learning_rate": 0.0001919827065980705,
+      "loss": 0.3822,
+      "step": 11165
+    },
+    {
+      "epoch": 1.4068394369745252,
+      "grad_norm": 0.29656147956848145,
+      "learning_rate": 0.00019187716578383178,
+      "loss": 0.4047,
+      "step": 11170
+    },
+    {
+      "epoch": 1.4074692193847027,
+      "grad_norm": 0.3193992078304291,
+      "learning_rate": 0.0001917716024805985,
+      "loss": 0.4088,
+      "step": 11175
+    },
+    {
+      "epoch": 1.4080990017948798,
+      "grad_norm": 0.29688236117362976,
+      "learning_rate": 0.0001916660167450605,
+      "loss": 0.3693,
+      "step": 11180
+    },
+    {
+      "epoch": 1.408728784205057,
+      "grad_norm": 0.33146485686302185,
+      "learning_rate": 0.00019156040863391977,
+      "loss": 0.3865,
+      "step": 11185
+    },
+    {
+      "epoch": 1.4093585666152344,
+      "grad_norm": 0.3015727698802948,
+      "learning_rate": 0.00019145477820389027,
+      "loss": 0.3857,
+      "step": 11190
+    },
+    {
+      "epoch": 1.4099883490254117,
+      "grad_norm": 0.27797931432724,
+      "learning_rate": 0.00019134912551169796,
+      "loss": 0.4148,
+      "step": 11195
+    },
+    {
+      "epoch": 1.410618131435589,
+      "grad_norm": 0.30010297894477844,
+      "learning_rate": 0.00019124345061408067,
+      "loss": 0.4076,
+      "step": 11200
+    },
+    {
+      "epoch": 1.4112479138457663,
+      "grad_norm": 0.29101455211639404,
+      "learning_rate": 0.00019113775356778833,
+      "loss": 0.3802,
+      "step": 11205
+    },
+    {
+      "epoch": 1.4118776962559436,
+      "grad_norm": 0.29706794023513794,
+      "learning_rate": 0.00019103203442958266,
+      "loss": 0.3867,
+      "step": 11210
+    },
+    {
+      "epoch": 1.412507478666121,
+      "grad_norm": 0.2546458840370178,
+      "learning_rate": 0.00019092629325623723,
+      "loss": 0.3964,
+      "step": 11215
+    },
+    {
+      "epoch": 1.4131372610762982,
+      "grad_norm": 0.3409089148044586,
+      "learning_rate": 0.0001908205301045375,
+      "loss": 0.4171,
+      "step": 11220
+    },
+    {
+      "epoch": 1.4137670434864753,
+      "grad_norm": 0.27688878774642944,
+      "learning_rate": 0.00019071474503128057,
+      "loss": 0.405,
+      "step": 11225
+    },
+    {
+      "epoch": 1.4143968258966528,
+      "grad_norm": 0.30704399943351746,
+      "learning_rate": 0.00019060893809327563,
+      "loss": 0.4024,
+      "step": 11230
+    },
+    {
+      "epoch": 1.41502660830683,
+      "grad_norm": 0.2823016941547394,
+      "learning_rate": 0.00019050310934734326,
+      "loss": 0.3908,
+      "step": 11235
+    },
+    {
+      "epoch": 1.4156563907170072,
+      "grad_norm": 0.3309246897697449,
+      "learning_rate": 0.000190397258850316,
+      "loss": 0.4049,
+      "step": 11240
+    },
+    {
+      "epoch": 1.4162861731271845,
+      "grad_norm": 0.2959790527820587,
+      "learning_rate": 0.00019029138665903794,
+      "loss": 0.4031,
+      "step": 11245
+    },
+    {
+      "epoch": 1.4169159555373618,
+      "grad_norm": 0.29836803674697876,
+      "learning_rate": 0.00019018549283036497,
+      "loss": 0.4103,
+      "step": 11250
+    },
+    {
+      "epoch": 1.4175457379475391,
+      "grad_norm": 0.3187415301799774,
+      "learning_rate": 0.00019007957742116433,
+      "loss": 0.4055,
+      "step": 11255
+    },
+    {
+      "epoch": 1.4181755203577164,
+      "grad_norm": 0.3521386981010437,
+      "learning_rate": 0.00018997364048831515,
+      "loss": 0.3839,
+      "step": 11260
+    },
+    {
+      "epoch": 1.4188053027678937,
+      "grad_norm": 0.3985449969768524,
+      "learning_rate": 0.00018986768208870792,
+      "loss": 0.4058,
+      "step": 11265
+    },
+    {
+      "epoch": 1.419435085178071,
+      "grad_norm": 0.30885374546051025,
+      "learning_rate": 0.00018976170227924473,
+      "loss": 0.394,
+      "step": 11270
+    },
+    {
+      "epoch": 1.4200648675882483,
+      "grad_norm": 0.2981209456920624,
+      "learning_rate": 0.00018965570111683917,
+      "loss": 0.3917,
+      "step": 11275
+    },
+    {
+      "epoch": 1.4206946499984254,
+      "grad_norm": 0.2993827164173126,
+      "learning_rate": 0.00018954967865841629,
+      "loss": 0.4016,
+      "step": 11280
+    },
+    {
+      "epoch": 1.421324432408603,
+      "grad_norm": 0.283632755279541,
+      "learning_rate": 0.00018944363496091254,
+      "loss": 0.3873,
+      "step": 11285
+    },
+    {
+      "epoch": 1.42195421481878,
+      "grad_norm": 0.2871907353401184,
+      "learning_rate": 0.0001893375700812758,
+      "loss": 0.4136,
+      "step": 11290
+    },
+    {
+      "epoch": 1.4225839972289573,
+      "grad_norm": 0.3341853618621826,
+      "learning_rate": 0.00018923148407646537,
+      "loss": 0.409,
+      "step": 11295
+    },
+    {
+      "epoch": 1.4232137796391346,
+      "grad_norm": 0.32463696599006653,
+      "learning_rate": 0.00018912537700345192,
+      "loss": 0.3912,
+      "step": 11300
+    },
+    {
+      "epoch": 1.423843562049312,
+      "grad_norm": 0.33242395520210266,
+      "learning_rate": 0.00018901924891921726,
+      "loss": 0.4158,
+      "step": 11305
+    },
+    {
+      "epoch": 1.4244733444594893,
+      "grad_norm": 0.301289439201355,
+      "learning_rate": 0.00018891309988075463,
+      "loss": 0.4012,
+      "step": 11310
+    },
+    {
+      "epoch": 1.4251031268696666,
+      "grad_norm": 0.28636494278907776,
+      "learning_rate": 0.00018880692994506845,
+      "loss": 0.3817,
+      "step": 11315
+    },
+    {
+      "epoch": 1.4257329092798439,
+      "grad_norm": 0.2837861478328705,
+      "learning_rate": 0.00018870073916917455,
+      "loss": 0.4116,
+      "step": 11320
+    },
+    {
+      "epoch": 1.4263626916900212,
+      "grad_norm": 0.31169527769088745,
+      "learning_rate": 0.0001885945276100996,
+      "loss": 0.3967,
+      "step": 11325
+    },
+    {
+      "epoch": 1.4269924741001985,
+      "grad_norm": 0.31035301089286804,
+      "learning_rate": 0.00018848829532488177,
+      "loss": 0.407,
+      "step": 11330
+    },
+    {
+      "epoch": 1.4276222565103756,
+      "grad_norm": 0.3047008812427521,
+      "learning_rate": 0.00018838204237057023,
+      "loss": 0.3939,
+      "step": 11335
+    },
+    {
+      "epoch": 1.428252038920553,
+      "grad_norm": 0.2646077871322632,
+      "learning_rate": 0.00018827576880422515,
+      "loss": 0.3881,
+      "step": 11340
+    },
+    {
+      "epoch": 1.4288818213307302,
+      "grad_norm": 0.31041520833969116,
+      "learning_rate": 0.00018816947468291788,
+      "loss": 0.3822,
+      "step": 11345
+    },
+    {
+      "epoch": 1.4295116037409075,
+      "grad_norm": 0.2699204385280609,
+      "learning_rate": 0.00018806316006373086,
+      "loss": 0.3895,
+      "step": 11350
+    },
+    {
+      "epoch": 1.4301413861510848,
+      "grad_norm": 0.285363107919693,
+      "learning_rate": 0.00018795682500375742,
+      "loss": 0.4027,
+      "step": 11355
+    },
+    {
+      "epoch": 1.430771168561262,
+      "grad_norm": 0.27154308557510376,
+      "learning_rate": 0.00018785046956010194,
+      "loss": 0.3815,
+      "step": 11360
+    },
+    {
+      "epoch": 1.4314009509714394,
+      "grad_norm": 0.29652640223503113,
+      "learning_rate": 0.00018774409378987972,
+      "loss": 0.4003,
+      "step": 11365
+    },
+    {
+      "epoch": 1.4320307333816167,
+      "grad_norm": 0.2921524941921234,
+      "learning_rate": 0.00018763769775021695,
+      "loss": 0.3828,
+      "step": 11370
+    },
+    {
+      "epoch": 1.432660515791794,
+      "grad_norm": 0.26934945583343506,
+      "learning_rate": 0.00018753128149825074,
+      "loss": 0.3999,
+      "step": 11375
+    },
+    {
+      "epoch": 1.4332902982019713,
+      "grad_norm": 0.29320502281188965,
+      "learning_rate": 0.00018742484509112907,
+      "loss": 0.4034,
+      "step": 11380
+    },
+    {
+      "epoch": 1.4339200806121486,
+      "grad_norm": 0.2842418849468231,
+      "learning_rate": 0.00018731838858601074,
+      "loss": 0.3877,
+      "step": 11385
+    },
+    {
+      "epoch": 1.4345498630223257,
+      "grad_norm": 0.31208139657974243,
+      "learning_rate": 0.00018721191204006525,
+      "loss": 0.3731,
+      "step": 11390
+    },
+    {
+      "epoch": 1.4351796454325032,
+      "grad_norm": 0.2809062600135803,
+      "learning_rate": 0.00018710541551047303,
+      "loss": 0.3939,
+      "step": 11395
+    },
+    {
+      "epoch": 1.4358094278426803,
+      "grad_norm": 0.308969646692276,
+      "learning_rate": 0.00018699889905442508,
+      "loss": 0.3874,
+      "step": 11400
+    },
+    {
+      "epoch": 1.4364392102528576,
+      "grad_norm": 0.3051275610923767,
+      "learning_rate": 0.00018689236272912316,
+      "loss": 0.3676,
+      "step": 11405
+    },
+    {
+      "epoch": 1.437068992663035,
+      "grad_norm": 0.31084486842155457,
+      "learning_rate": 0.0001867858065917798,
+      "loss": 0.3954,
+      "step": 11410
+    },
+    {
+      "epoch": 1.4376987750732122,
+      "grad_norm": 0.28356167674064636,
+      "learning_rate": 0.000186679230699618,
+      "loss": 0.3701,
+      "step": 11415
+    },
+    {
+      "epoch": 1.4383285574833895,
+      "grad_norm": 0.3026244044303894,
+      "learning_rate": 0.0001865726351098715,
+      "loss": 0.3797,
+      "step": 11420
+    },
+    {
+      "epoch": 1.4389583398935668,
+      "grad_norm": 0.2909928560256958,
+      "learning_rate": 0.00018646601987978452,
+      "loss": 0.4022,
+      "step": 11425
+    },
+    {
+      "epoch": 1.439588122303744,
+      "grad_norm": 0.3085511326789856,
+      "learning_rate": 0.00018635938506661183,
+      "loss": 0.4099,
+      "step": 11430
+    },
+    {
+      "epoch": 1.4402179047139214,
+      "grad_norm": 0.28047701716423035,
+      "learning_rate": 0.0001862527307276189,
+      "loss": 0.3789,
+      "step": 11435
+    },
+    {
+      "epoch": 1.4408476871240987,
+      "grad_norm": 0.2697209119796753,
+      "learning_rate": 0.00018614605692008146,
+      "loss": 0.3864,
+      "step": 11440
+    },
+    {
+      "epoch": 1.4414774695342758,
+      "grad_norm": 0.40744665265083313,
+      "learning_rate": 0.0001860393637012858,
+      "loss": 0.4085,
+      "step": 11445
+    },
+    {
+      "epoch": 1.4421072519444533,
+      "grad_norm": 0.25875118374824524,
+      "learning_rate": 0.00018593265112852854,
+      "loss": 0.4033,
+      "step": 11450
+    },
+    {
+      "epoch": 1.4427370343546304,
+      "grad_norm": 0.2960642874240875,
+      "learning_rate": 0.00018582591925911694,
+      "loss": 0.4214,
+      "step": 11455
+    },
+    {
+      "epoch": 1.4433668167648077,
+      "grad_norm": 0.2711925506591797,
+      "learning_rate": 0.00018571916815036824,
+      "loss": 0.3537,
+      "step": 11460
+    },
+    {
+      "epoch": 1.443996599174985,
+      "grad_norm": 0.28002485632896423,
+      "learning_rate": 0.0001856123978596104,
+      "loss": 0.3787,
+      "step": 11465
+    },
+    {
+      "epoch": 1.4446263815851623,
+      "grad_norm": 0.3143458366394043,
+      "learning_rate": 0.00018550560844418138,
+      "loss": 0.3553,
+      "step": 11470
+    },
+    {
+      "epoch": 1.4452561639953396,
+      "grad_norm": 0.3184334337711334,
+      "learning_rate": 0.00018539879996142962,
+      "loss": 0.385,
+      "step": 11475
+    },
+    {
+      "epoch": 1.445885946405517,
+      "grad_norm": 0.3327188789844513,
+      "learning_rate": 0.00018529197246871368,
+      "loss": 0.4074,
+      "step": 11480
+    },
+    {
+      "epoch": 1.4465157288156942,
+      "grad_norm": 0.317942351102829,
+      "learning_rate": 0.0001851851260234024,
+      "loss": 0.3995,
+      "step": 11485
+    },
+    {
+      "epoch": 1.4471455112258715,
+      "grad_norm": 0.2567351758480072,
+      "learning_rate": 0.00018507826068287473,
+      "loss": 0.3661,
+      "step": 11490
+    },
+    {
+      "epoch": 1.4477752936360488,
+      "grad_norm": 0.29439592361450195,
+      "learning_rate": 0.0001849713765045198,
+      "loss": 0.3759,
+      "step": 11495
+    },
+    {
+      "epoch": 1.448405076046226,
+      "grad_norm": 0.3125048279762268,
+      "learning_rate": 0.0001848644735457368,
+      "loss": 0.4107,
+      "step": 11500
+    },
+    {
+      "epoch": 1.4490348584564032,
+      "grad_norm": 0.2855313718318939,
+      "learning_rate": 0.00018475755186393516,
+      "loss": 0.4061,
+      "step": 11505
+    },
+    {
+      "epoch": 1.4496646408665805,
+      "grad_norm": 0.3040854036808014,
+      "learning_rate": 0.00018465061151653423,
+      "loss": 0.3902,
+      "step": 11510
+    },
+    {
+      "epoch": 1.4502944232767578,
+      "grad_norm": 0.28425633907318115,
+      "learning_rate": 0.0001845436525609634,
+      "loss": 0.3861,
+      "step": 11515
+    },
+    {
+      "epoch": 1.4509242056869351,
+      "grad_norm": 0.31335607171058655,
+      "learning_rate": 0.00018443667505466205,
+      "loss": 0.3949,
+      "step": 11520
+    },
+    {
+      "epoch": 1.4515539880971124,
+      "grad_norm": 0.2725260555744171,
+      "learning_rate": 0.00018432967905507967,
+      "loss": 0.3979,
+      "step": 11525
+    },
+    {
+      "epoch": 1.4521837705072898,
+      "grad_norm": 0.2674049437046051,
+      "learning_rate": 0.00018422266461967537,
+      "loss": 0.3747,
+      "step": 11530
+    },
+    {
+      "epoch": 1.452813552917467,
+      "grad_norm": 0.3076520562171936,
+      "learning_rate": 0.0001841156318059185,
+      "loss": 0.385,
+      "step": 11535
+    },
+    {
+      "epoch": 1.4534433353276444,
+      "grad_norm": 0.23340527713298798,
+      "learning_rate": 0.00018400858067128806,
+      "loss": 0.3736,
+      "step": 11540
+    },
+    {
+      "epoch": 1.4540731177378217,
+      "grad_norm": 0.29402169585227966,
+      "learning_rate": 0.00018390151127327295,
+      "loss": 0.3994,
+      "step": 11545
+    },
+    {
+      "epoch": 1.454702900147999,
+      "grad_norm": 0.32409217953681946,
+      "learning_rate": 0.00018379442366937187,
+      "loss": 0.3979,
+      "step": 11550
+    },
+    {
+      "epoch": 1.455332682558176,
+      "grad_norm": 0.28875911235809326,
+      "learning_rate": 0.00018368731791709337,
+      "loss": 0.365,
+      "step": 11555
+    },
+    {
+      "epoch": 1.4559624649683534,
+      "grad_norm": 0.26838234066963196,
+      "learning_rate": 0.0001835801940739556,
+      "loss": 0.3912,
+      "step": 11560
+    },
+    {
+      "epoch": 1.4565922473785307,
+      "grad_norm": 0.31797516345977783,
+      "learning_rate": 0.00018347305219748665,
+      "loss": 0.3622,
+      "step": 11565
+    },
+    {
+      "epoch": 1.457222029788708,
+      "grad_norm": 0.31115812063217163,
+      "learning_rate": 0.00018336589234522398,
+      "loss": 0.4283,
+      "step": 11570
+    },
+    {
+      "epoch": 1.4578518121988853,
+      "grad_norm": 0.2730168402194977,
+      "learning_rate": 0.00018325871457471496,
+      "loss": 0.3864,
+      "step": 11575
+    },
+    {
+      "epoch": 1.4584815946090626,
+      "grad_norm": 0.28333088755607605,
+      "learning_rate": 0.00018315151894351657,
+      "loss": 0.3451,
+      "step": 11580
+    },
+    {
+      "epoch": 1.4591113770192399,
+      "grad_norm": 0.3169468343257904,
+      "learning_rate": 0.00018304430550919522,
+      "loss": 0.3719,
+      "step": 11585
+    },
+    {
+      "epoch": 1.4597411594294172,
+      "grad_norm": 0.3411467969417572,
+      "learning_rate": 0.000182937074329327,
+      "loss": 0.4073,
+      "step": 11590
+    },
+    {
+      "epoch": 1.4603709418395945,
+      "grad_norm": 0.3131183385848999,
+      "learning_rate": 0.0001828298254614975,
+      "loss": 0.4117,
+      "step": 11595
+    },
+    {
+      "epoch": 1.4610007242497716,
+      "grad_norm": 0.25929832458496094,
+      "learning_rate": 0.0001827225589633018,
+      "loss": 0.3834,
+      "step": 11600
+    },
+    {
+      "epoch": 1.461630506659949,
+      "grad_norm": 0.32609832286834717,
+      "learning_rate": 0.00018261527489234444,
+      "loss": 0.3972,
+      "step": 11605
+    },
+    {
+      "epoch": 1.4622602890701262,
+      "grad_norm": 0.3089287579059601,
+      "learning_rate": 0.00018250797330623953,
+      "loss": 0.3727,
+      "step": 11610
+    },
+    {
+      "epoch": 1.4628900714803035,
+      "grad_norm": 0.2891997992992401,
+      "learning_rate": 0.00018240065426261033,
+      "loss": 0.3891,
+      "step": 11615
+    },
+    {
+      "epoch": 1.4635198538904808,
+      "grad_norm": 0.3119528293609619,
+      "learning_rate": 0.00018229331781908971,
+      "loss": 0.388,
+      "step": 11620
+    },
+    {
+      "epoch": 1.464149636300658,
+      "grad_norm": 0.3314844071865082,
+      "learning_rate": 0.00018218596403331977,
+      "loss": 0.3803,
+      "step": 11625
+    },
+    {
+      "epoch": 1.4647794187108354,
+      "grad_norm": 0.27267536520957947,
+      "learning_rate": 0.00018207859296295197,
+      "loss": 0.3665,
+      "step": 11630
+    },
+    {
+      "epoch": 1.4654092011210127,
+      "grad_norm": 0.30490440130233765,
+      "learning_rate": 0.00018197120466564693,
+      "loss": 0.4051,
+      "step": 11635
+    },
+    {
+      "epoch": 1.46603898353119,
+      "grad_norm": 0.3182273209095001,
+      "learning_rate": 0.00018186379919907472,
+      "loss": 0.38,
+      "step": 11640
+    },
+    {
+      "epoch": 1.4666687659413673,
+      "grad_norm": 0.3026832044124603,
+      "learning_rate": 0.00018175637662091448,
+      "loss": 0.3371,
+      "step": 11645
+    },
+    {
+      "epoch": 1.4672985483515446,
+      "grad_norm": 0.3287534713745117,
+      "learning_rate": 0.0001816489369888546,
+      "loss": 0.4234,
+      "step": 11650
+    },
+    {
+      "epoch": 1.4679283307617217,
+      "grad_norm": 0.28076720237731934,
+      "learning_rate": 0.00018154148036059263,
+      "loss": 0.3825,
+      "step": 11655
+    },
+    {
+      "epoch": 1.4685581131718992,
+      "grad_norm": 0.304766446352005,
+      "learning_rate": 0.0001814340067938352,
+      "loss": 0.3905,
+      "step": 11660
+    },
+    {
+      "epoch": 1.4691878955820763,
+      "grad_norm": 0.30473533272743225,
+      "learning_rate": 0.00018132651634629812,
+      "loss": 0.409,
+      "step": 11665
+    },
+    {
+      "epoch": 1.4698176779922536,
+      "grad_norm": 0.32186418771743774,
+      "learning_rate": 0.00018121900907570618,
+      "loss": 0.3741,
+      "step": 11670
+    },
+    {
+      "epoch": 1.470447460402431,
+      "grad_norm": 0.33314061164855957,
+      "learning_rate": 0.00018111148503979326,
+      "loss": 0.3981,
+      "step": 11675
+    },
+    {
+      "epoch": 1.4710772428126082,
+      "grad_norm": 0.3202495872974396,
+      "learning_rate": 0.00018100394429630223,
+      "loss": 0.4014,
+      "step": 11680
+    },
+    {
+      "epoch": 1.4717070252227855,
+      "grad_norm": 0.2801063656806946,
+      "learning_rate": 0.00018089638690298488,
+      "loss": 0.3827,
+      "step": 11685
+    },
+    {
+      "epoch": 1.4723368076329628,
+      "grad_norm": 0.3252180516719818,
+      "learning_rate": 0.000180788812917602,
+      "loss": 0.4207,
+      "step": 11690
+    },
+    {
+      "epoch": 1.4729665900431401,
+      "grad_norm": 0.279823899269104,
+      "learning_rate": 0.0001806812223979233,
+      "loss": 0.4092,
+      "step": 11695
+    },
+    {
+      "epoch": 1.4735963724533174,
+      "grad_norm": 0.29136526584625244,
+      "learning_rate": 0.00018057361540172733,
+      "loss": 0.3939,
+      "step": 11700
+    },
+    {
+      "epoch": 1.4742261548634947,
+      "grad_norm": 0.2708832621574402,
+      "learning_rate": 0.00018046599198680153,
+      "loss": 0.3645,
+      "step": 11705
+    },
+    {
+      "epoch": 1.4748559372736718,
+      "grad_norm": 0.34708496928215027,
+      "learning_rate": 0.00018035835221094214,
+      "loss": 0.3814,
+      "step": 11710
+    },
+    {
+      "epoch": 1.4754857196838493,
+      "grad_norm": 0.3081948161125183,
+      "learning_rate": 0.00018025069613195413,
+      "loss": 0.3738,
+      "step": 11715
+    },
+    {
+      "epoch": 1.4761155020940264,
+      "grad_norm": 0.26891911029815674,
+      "learning_rate": 0.0001801430238076513,
+      "loss": 0.3724,
+      "step": 11720
+    },
+    {
+      "epoch": 1.4767452845042037,
+      "grad_norm": 0.3266797363758087,
+      "learning_rate": 0.00018003533529585612,
+      "loss": 0.3749,
+      "step": 11725
+    },
+    {
+      "epoch": 1.477375066914381,
+      "grad_norm": 0.25788089632987976,
+      "learning_rate": 0.00017992763065439982,
+      "loss": 0.3661,
+      "step": 11730
+    },
+    {
+      "epoch": 1.4780048493245583,
+      "grad_norm": 0.301270067691803,
+      "learning_rate": 0.00017981990994112227,
+      "loss": 0.3832,
+      "step": 11735
+    },
+    {
+      "epoch": 1.4786346317347356,
+      "grad_norm": 0.2785583734512329,
+      "learning_rate": 0.0001797121732138719,
+      "loss": 0.357,
+      "step": 11740
+    },
+    {
+      "epoch": 1.479264414144913,
+      "grad_norm": 0.3153518736362457,
+      "learning_rate": 0.00017960442053050583,
+      "loss": 0.3964,
+      "step": 11745
+    },
+    {
+      "epoch": 1.4798941965550902,
+      "grad_norm": 0.2862750291824341,
+      "learning_rate": 0.00017949665194888972,
+      "loss": 0.3781,
+      "step": 11750
+    },
+    {
+      "epoch": 1.4805239789652676,
+      "grad_norm": 0.31263992190361023,
+      "learning_rate": 0.00017938886752689765,
+      "loss": 0.3822,
+      "step": 11755
+    },
+    {
+      "epoch": 1.4811537613754449,
+      "grad_norm": 0.31964340806007385,
+      "learning_rate": 0.00017928106732241248,
+      "loss": 0.3757,
+      "step": 11760
+    },
+    {
+      "epoch": 1.481783543785622,
+      "grad_norm": 0.29111340641975403,
+      "learning_rate": 0.0001791732513933253,
+      "loss": 0.362,
+      "step": 11765
+    },
+    {
+      "epoch": 1.4824133261957995,
+      "grad_norm": 0.32248637080192566,
+      "learning_rate": 0.00017906541979753572,
+      "loss": 0.3978,
+      "step": 11770
+    },
+    {
+      "epoch": 1.4830431086059765,
+      "grad_norm": 0.2964222729206085,
+      "learning_rate": 0.0001789575725929518,
+      "loss": 0.3853,
+      "step": 11775
+    },
+    {
+      "epoch": 1.4836728910161538,
+      "grad_norm": 0.32823482155799866,
+      "learning_rate": 0.0001788497098374899,
+      "loss": 0.3828,
+      "step": 11780
+    },
+    {
+      "epoch": 1.4843026734263312,
+      "grad_norm": 0.30054226517677307,
+      "learning_rate": 0.0001787418315890748,
+      "loss": 0.38,
+      "step": 11785
+    },
+    {
+      "epoch": 1.4849324558365085,
+      "grad_norm": 0.30829596519470215,
+      "learning_rate": 0.0001786339379056397,
+      "loss": 0.3645,
+      "step": 11790
+    },
+    {
+      "epoch": 1.4855622382466858,
+      "grad_norm": 0.3095497786998749,
+      "learning_rate": 0.00017852602884512584,
+      "loss": 0.3727,
+      "step": 11795
+    },
+    {
+      "epoch": 1.486192020656863,
+      "grad_norm": 0.29647621512413025,
+      "learning_rate": 0.00017841810446548283,
+      "loss": 0.3764,
+      "step": 11800
+    },
+    {
+      "epoch": 1.4868218030670404,
+      "grad_norm": 0.3227784037590027,
+      "learning_rate": 0.00017831016482466864,
+      "loss": 0.3797,
+      "step": 11805
+    },
+    {
+      "epoch": 1.4874515854772177,
+      "grad_norm": 0.32365646958351135,
+      "learning_rate": 0.00017820220998064927,
+      "loss": 0.3766,
+      "step": 11810
+    },
+    {
+      "epoch": 1.488081367887395,
+      "grad_norm": 0.36090198159217834,
+      "learning_rate": 0.0001780942399913989,
+      "loss": 0.4015,
+      "step": 11815
+    },
+    {
+      "epoch": 1.488711150297572,
+      "grad_norm": 0.28814610838890076,
+      "learning_rate": 0.00017798625491489994,
+      "loss": 0.3616,
+      "step": 11820
+    },
+    {
+      "epoch": 1.4893409327077496,
+      "grad_norm": 0.2654825747013092,
+      "learning_rate": 0.00017787825480914283,
+      "loss": 0.3462,
+      "step": 11825
+    },
+    {
+      "epoch": 1.4899707151179267,
+      "grad_norm": 0.2913071811199188,
+      "learning_rate": 0.000177770239732126,
+      "loss": 0.3707,
+      "step": 11830
+    },
+    {
+      "epoch": 1.490600497528104,
+      "grad_norm": 0.33099865913391113,
+      "learning_rate": 0.0001776622097418562,
+      "loss": 0.3644,
+      "step": 11835
+    },
+    {
+      "epoch": 1.4912302799382813,
+      "grad_norm": 0.2980974018573761,
+      "learning_rate": 0.0001775541648963478,
+      "loss": 0.3839,
+      "step": 11840
+    },
+    {
+      "epoch": 1.4918600623484586,
+      "grad_norm": 0.2673074007034302,
+      "learning_rate": 0.00017744610525362352,
+      "loss": 0.3736,
+      "step": 11845
+    },
+    {
+      "epoch": 1.4924898447586359,
+      "grad_norm": 0.26277023553848267,
+      "learning_rate": 0.00017733803087171372,
+      "loss": 0.3463,
+      "step": 11850
+    },
+    {
+      "epoch": 1.4931196271688132,
+      "grad_norm": 0.27924680709838867,
+      "learning_rate": 0.00017722994180865696,
+      "loss": 0.4095,
+      "step": 11855
+    },
+    {
+      "epoch": 1.4937494095789905,
+      "grad_norm": 0.2761695086956024,
+      "learning_rate": 0.00017712183812249938,
+      "loss": 0.3748,
+      "step": 11860
+    },
+    {
+      "epoch": 1.4943791919891678,
+      "grad_norm": 0.312854528427124,
+      "learning_rate": 0.00017701371987129523,
+      "loss": 0.3748,
+      "step": 11865
+    },
+    {
+      "epoch": 1.495008974399345,
+      "grad_norm": 0.3033592998981476,
+      "learning_rate": 0.00017690558711310644,
+      "loss": 0.3728,
+      "step": 11870
+    },
+    {
+      "epoch": 1.4956387568095222,
+      "grad_norm": 0.2711508572101593,
+      "learning_rate": 0.00017679743990600281,
+      "loss": 0.3748,
+      "step": 11875
+    },
+    {
+      "epoch": 1.4962685392196997,
+      "grad_norm": 0.28003159165382385,
+      "learning_rate": 0.00017668927830806177,
+      "loss": 0.3658,
+      "step": 11880
+    },
+    {
+      "epoch": 1.4968983216298768,
+      "grad_norm": 0.2750314772129059,
+      "learning_rate": 0.0001765811023773687,
+      "loss": 0.3705,
+      "step": 11885
+    },
+    {
+      "epoch": 1.497528104040054,
+      "grad_norm": 0.31037452816963196,
+      "learning_rate": 0.00017647291217201644,
+      "loss": 0.3718,
+      "step": 11890
+    },
+    {
+      "epoch": 1.4981578864502314,
+      "grad_norm": 0.33681520819664,
+      "learning_rate": 0.00017636470775010563,
+      "loss": 0.37,
+      "step": 11895
+    },
+    {
+      "epoch": 1.4987876688604087,
+      "grad_norm": 0.2735719084739685,
+      "learning_rate": 0.00017625648916974452,
+      "loss": 0.3898,
+      "step": 11900
+    },
+    {
+      "epoch": 1.499417451270586,
+      "grad_norm": 0.2873845398426056,
+      "learning_rate": 0.00017614825648904902,
+      "loss": 0.387,
+      "step": 11905
+    },
+    {
+      "epoch": 1.5000472336807633,
+      "grad_norm": 0.2826070189476013,
+      "learning_rate": 0.00017604000976614243,
+      "loss": 0.3656,
+      "step": 11910
+    },
+    {
+      "epoch": 1.5006770160909406,
+      "grad_norm": 0.2709527015686035,
+      "learning_rate": 0.00017593174905915581,
+      "loss": 0.3583,
+      "step": 11915
+    },
+    {
+      "epoch": 1.5013067985011177,
+      "grad_norm": 0.3088144063949585,
+      "learning_rate": 0.00017582347442622755,
+      "loss": 0.3715,
+      "step": 11920
+    },
+    {
+      "epoch": 1.5019365809112952,
+      "grad_norm": 0.27996301651000977,
+      "learning_rate": 0.0001757151859255038,
+      "loss": 0.3636,
+      "step": 11925
+    },
+    {
+      "epoch": 1.5025663633214723,
+      "grad_norm": 0.3117114007472992,
+      "learning_rate": 0.00017560688361513766,
+      "loss": 0.351,
+      "step": 11930
+    },
+    {
+      "epoch": 1.5031961457316498,
+      "grad_norm": 0.32614433765411377,
+      "learning_rate": 0.00017549856755329012,
+      "loss": 0.3711,
+      "step": 11935
+    },
+    {
+      "epoch": 1.503825928141827,
+      "grad_norm": 0.23831017315387726,
+      "learning_rate": 0.0001753902377981294,
+      "loss": 0.3645,
+      "step": 11940
+    },
+    {
+      "epoch": 1.5044557105520044,
+      "grad_norm": 0.27338019013404846,
+      "learning_rate": 0.000175281894407831,
+      "loss": 0.3606,
+      "step": 11945
+    },
+    {
+      "epoch": 1.5050854929621815,
+      "grad_norm": 0.2813990116119385,
+      "learning_rate": 0.0001751735374405778,
+      "loss": 0.3637,
+      "step": 11950
+    },
+    {
+      "epoch": 1.5057152753723588,
+      "grad_norm": 0.2607782781124115,
+      "learning_rate": 0.00017506516695455992,
+      "loss": 0.3493,
+      "step": 11955
+    },
+    {
+      "epoch": 1.5063450577825361,
+      "grad_norm": 0.2825680077075958,
+      "learning_rate": 0.0001749567830079749,
+      "loss": 0.3474,
+      "step": 11960
+    },
+    {
+      "epoch": 1.5069748401927134,
+      "grad_norm": 0.2957023084163666,
+      "learning_rate": 0.00017484838565902735,
+      "loss": 0.3852,
+      "step": 11965
+    },
+    {
+      "epoch": 1.5076046226028907,
+      "grad_norm": 0.31363338232040405,
+      "learning_rate": 0.00017473997496592904,
+      "loss": 0.3944,
+      "step": 11970
+    },
+    {
+      "epoch": 1.5082344050130678,
+      "grad_norm": 0.271010160446167,
+      "learning_rate": 0.00017463155098689908,
+      "loss": 0.3667,
+      "step": 11975
+    },
+    {
+      "epoch": 1.5088641874232454,
+      "grad_norm": 0.28360188007354736,
+      "learning_rate": 0.00017452311378016362,
+      "loss": 0.3564,
+      "step": 11980
+    },
+    {
+      "epoch": 1.5094939698334224,
+      "grad_norm": 0.28345590829849243,
+      "learning_rate": 0.00017441466340395583,
+      "loss": 0.358,
+      "step": 11985
+    },
+    {
+      "epoch": 1.5101237522436,
+      "grad_norm": 0.23574601113796234,
+      "learning_rate": 0.00017430619991651614,
+      "loss": 0.3588,
+      "step": 11990
+    },
+    {
+      "epoch": 1.510753534653777,
+      "grad_norm": 0.32633906602859497,
+      "learning_rate": 0.0001741977233760919,
+      "loss": 0.3786,
+      "step": 11995
+    },
+    {
+      "epoch": 1.5113833170639546,
+      "grad_norm": 0.31216609477996826,
+      "learning_rate": 0.00017408923384093746,
+      "loss": 0.3949,
+      "step": 12000
+    },
+    {
+      "epoch": 1.5113833170639546,
+      "eval_loss": 0.3009350597858429,
+      "eval_runtime": 6.1573,
+      "eval_samples_per_second": 162.409,
+      "eval_steps_per_second": 10.232,
+      "step": 12000
+    },
+    {
+      "epoch": 1.5120130994741316,
+      "grad_norm": 0.2735341191291809,
+      "learning_rate": 0.00017398073136931416,
+      "loss": 0.3667,
+      "step": 12005
+    },
+    {
+      "epoch": 1.512642881884309,
+      "grad_norm": 0.3168368637561798,
+      "learning_rate": 0.0001738722160194904,
+      "loss": 0.3693,
+      "step": 12010
+    },
+    {
+      "epoch": 1.5132726642944863,
+      "grad_norm": 0.27563655376434326,
+      "learning_rate": 0.0001737636878497413,
+      "loss": 0.3721,
+      "step": 12015
+    },
+    {
+      "epoch": 1.5139024467046636,
+      "grad_norm": 0.27887552976608276,
+      "learning_rate": 0.00017365514691834898,
+      "loss": 0.402,
+      "step": 12020
+    },
+    {
+      "epoch": 1.5145322291148409,
+      "grad_norm": 0.30676189064979553,
+      "learning_rate": 0.0001735465932836024,
+      "loss": 0.3875,
+      "step": 12025
+    },
+    {
+      "epoch": 1.515162011525018,
+      "grad_norm": 0.30623871088027954,
+      "learning_rate": 0.00017343802700379746,
+      "loss": 0.3644,
+      "step": 12030
+    },
+    {
+      "epoch": 1.5157917939351955,
+      "grad_norm": 0.2534305453300476,
+      "learning_rate": 0.00017332944813723658,
+      "loss": 0.3753,
+      "step": 12035
+    },
+    {
+      "epoch": 1.5164215763453726,
+      "grad_norm": 0.29374125599861145,
+      "learning_rate": 0.00017322085674222916,
+      "loss": 0.3964,
+      "step": 12040
+    },
+    {
+      "epoch": 1.51705135875555,
+      "grad_norm": 0.2833009362220764,
+      "learning_rate": 0.00017311225287709126,
+      "loss": 0.3778,
+      "step": 12045
+    },
+    {
+      "epoch": 1.5176811411657272,
+      "grad_norm": 0.273299902677536,
+      "learning_rate": 0.0001730036366001456,
+      "loss": 0.3661,
+      "step": 12050
+    },
+    {
+      "epoch": 1.5183109235759045,
+      "grad_norm": 0.32840752601623535,
+      "learning_rate": 0.00017289500796972165,
+      "loss": 0.3564,
+      "step": 12055
+    },
+    {
+      "epoch": 1.5189407059860818,
+      "grad_norm": 0.289202481508255,
+      "learning_rate": 0.00017278636704415545,
+      "loss": 0.3885,
+      "step": 12060
+    },
+    {
+      "epoch": 1.519570488396259,
+      "grad_norm": 0.28327277302742004,
+      "learning_rate": 0.0001726777138817896,
+      "loss": 0.376,
+      "step": 12065
+    },
+    {
+      "epoch": 1.5202002708064364,
+      "grad_norm": 0.2617267370223999,
+      "learning_rate": 0.00017256904854097343,
+      "loss": 0.353,
+      "step": 12070
+    },
+    {
+      "epoch": 1.5208300532166137,
+      "grad_norm": 0.2693130671977997,
+      "learning_rate": 0.00017246037108006266,
+      "loss": 0.386,
+      "step": 12075
+    },
+    {
+      "epoch": 1.521459835626791,
+      "grad_norm": 0.260217547416687,
+      "learning_rate": 0.00017235168155741956,
+      "loss": 0.3773,
+      "step": 12080
+    },
+    {
+      "epoch": 1.522089618036968,
+      "grad_norm": 0.2806963622570038,
+      "learning_rate": 0.0001722429800314129,
+      "loss": 0.3703,
+      "step": 12085
+    },
+    {
+      "epoch": 1.5227194004471456,
+      "grad_norm": 0.2797011435031891,
+      "learning_rate": 0.00017213426656041787,
+      "loss": 0.3523,
+      "step": 12090
+    },
+    {
+      "epoch": 1.5233491828573227,
+      "grad_norm": 0.3413710296154022,
+      "learning_rate": 0.00017202554120281612,
+      "loss": 0.3825,
+      "step": 12095
+    },
+    {
+      "epoch": 1.5239789652675002,
+      "grad_norm": 0.2759542167186737,
+      "learning_rate": 0.0001719168040169956,
+      "loss": 0.346,
+      "step": 12100
+    },
+    {
+      "epoch": 1.5246087476776773,
+      "grad_norm": 0.28816723823547363,
+      "learning_rate": 0.00017180805506135068,
+      "loss": 0.3772,
+      "step": 12105
+    },
+    {
+      "epoch": 1.5252385300878546,
+      "grad_norm": 0.2563376724720001,
+      "learning_rate": 0.00017169929439428207,
+      "loss": 0.3661,
+      "step": 12110
+    },
+    {
+      "epoch": 1.525868312498032,
+      "grad_norm": 0.29572755098342896,
+      "learning_rate": 0.0001715905220741967,
+      "loss": 0.3428,
+      "step": 12115
+    },
+    {
+      "epoch": 1.5264980949082092,
+      "grad_norm": 0.28491732478141785,
+      "learning_rate": 0.0001714817381595078,
+      "loss": 0.3778,
+      "step": 12120
+    },
+    {
+      "epoch": 1.5271278773183865,
+      "grad_norm": 0.28429144620895386,
+      "learning_rate": 0.0001713729427086348,
+      "loss": 0.351,
+      "step": 12125
+    },
+    {
+      "epoch": 1.5277576597285638,
+      "grad_norm": 0.3044835925102234,
+      "learning_rate": 0.00017126413578000342,
+      "loss": 0.3651,
+      "step": 12130
+    },
+    {
+      "epoch": 1.5283874421387411,
+      "grad_norm": 0.30945730209350586,
+      "learning_rate": 0.0001711553174320453,
+      "loss": 0.3731,
+      "step": 12135
+    },
+    {
+      "epoch": 1.5290172245489182,
+      "grad_norm": 0.26389655470848083,
+      "learning_rate": 0.00017104648772319853,
+      "loss": 0.3527,
+      "step": 12140
+    },
+    {
+      "epoch": 1.5296470069590957,
+      "grad_norm": 0.3144720196723938,
+      "learning_rate": 0.0001709376467119071,
+      "loss": 0.3776,
+      "step": 12145
+    },
+    {
+      "epoch": 1.5302767893692728,
+      "grad_norm": 0.2860710918903351,
+      "learning_rate": 0.00017082879445662113,
+      "loss": 0.3575,
+      "step": 12150
+    },
+    {
+      "epoch": 1.5309065717794503,
+      "grad_norm": 0.2869095802307129,
+      "learning_rate": 0.00017071993101579674,
+      "loss": 0.3322,
+      "step": 12155
+    },
+    {
+      "epoch": 1.5315363541896274,
+      "grad_norm": 0.2524400055408478,
+      "learning_rate": 0.00017061105644789612,
+      "loss": 0.3743,
+      "step": 12160
+    },
+    {
+      "epoch": 1.5321661365998047,
+      "grad_norm": 0.2670304477214813,
+      "learning_rate": 0.00017050217081138736,
+      "loss": 0.3735,
+      "step": 12165
+    },
+    {
+      "epoch": 1.532795919009982,
+      "grad_norm": 0.2701478898525238,
+      "learning_rate": 0.00017039327416474456,
+      "loss": 0.3467,
+      "step": 12170
+    },
+    {
+      "epoch": 1.5334257014201593,
+      "grad_norm": 0.2941682040691376,
+      "learning_rate": 0.0001702843665664477,
+      "loss": 0.3895,
+      "step": 12175
+    },
+    {
+      "epoch": 1.5340554838303366,
+      "grad_norm": 0.28004932403564453,
+      "learning_rate": 0.00017017544807498264,
+      "loss": 0.3666,
+      "step": 12180
+    },
+    {
+      "epoch": 1.534685266240514,
+      "grad_norm": 0.29110807180404663,
+      "learning_rate": 0.00017006651874884116,
+      "loss": 0.3628,
+      "step": 12185
+    },
+    {
+      "epoch": 1.5353150486506912,
+      "grad_norm": 0.2467578798532486,
+      "learning_rate": 0.00016995757864652066,
+      "loss": 0.35,
+      "step": 12190
+    },
+    {
+      "epoch": 1.5359448310608683,
+      "grad_norm": 0.3148331046104431,
+      "learning_rate": 0.00016984862782652463,
+      "loss": 0.3535,
+      "step": 12195
+    },
+    {
+      "epoch": 1.5365746134710458,
+      "grad_norm": 0.28578343987464905,
+      "learning_rate": 0.00016973966634736202,
+      "loss": 0.3477,
+      "step": 12200
+    },
+    {
+      "epoch": 1.537204395881223,
+      "grad_norm": 0.24588525295257568,
+      "learning_rate": 0.0001696306942675477,
+      "loss": 0.3459,
+      "step": 12205
+    },
+    {
+      "epoch": 1.5378341782914005,
+      "grad_norm": 0.2754054665565491,
+      "learning_rate": 0.00016952171164560213,
+      "loss": 0.3555,
+      "step": 12210
+    },
+    {
+      "epoch": 1.5384639607015775,
+      "grad_norm": 0.28237447142601013,
+      "learning_rate": 0.00016941271854005148,
+      "loss": 0.3446,
+      "step": 12215
+    },
+    {
+      "epoch": 1.5390937431117548,
+      "grad_norm": 0.27689647674560547,
+      "learning_rate": 0.00016930371500942755,
+      "loss": 0.3651,
+      "step": 12220
+    },
+    {
+      "epoch": 1.5397235255219321,
+      "grad_norm": 0.29644525051116943,
+      "learning_rate": 0.0001691947011122677,
+      "loss": 0.3482,
+      "step": 12225
+    },
+    {
+      "epoch": 1.5403533079321095,
+      "grad_norm": 0.3168468475341797,
+      "learning_rate": 0.0001690856769071149,
+      "loss": 0.3859,
+      "step": 12230
+    },
+    {
+      "epoch": 1.5409830903422868,
+      "grad_norm": 0.282879501581192,
+      "learning_rate": 0.0001689766424525177,
+      "loss": 0.3742,
+      "step": 12235
+    },
+    {
+      "epoch": 1.541612872752464,
+      "grad_norm": 0.2539578676223755,
+      "learning_rate": 0.00016886759780702996,
+      "loss": 0.3467,
+      "step": 12240
+    },
+    {
+      "epoch": 1.5422426551626414,
+      "grad_norm": 0.3353635370731354,
+      "learning_rate": 0.00016875854302921122,
+      "loss": 0.3831,
+      "step": 12245
+    },
+    {
+      "epoch": 1.5428724375728184,
+      "grad_norm": 0.2890516519546509,
+      "learning_rate": 0.0001686494781776264,
+      "loss": 0.3672,
+      "step": 12250
+    },
+    {
+      "epoch": 1.543502219982996,
+      "grad_norm": 0.3136516213417053,
+      "learning_rate": 0.00016854040331084583,
+      "loss": 0.37,
+      "step": 12255
+    },
+    {
+      "epoch": 1.544132002393173,
+      "grad_norm": 0.29757821559906006,
+      "learning_rate": 0.0001684313184874451,
+      "loss": 0.3681,
+      "step": 12260
+    },
+    {
+      "epoch": 1.5447617848033506,
+      "grad_norm": 0.3504684269428253,
+      "learning_rate": 0.0001683222237660054,
+      "loss": 0.3868,
+      "step": 12265
+    },
+    {
+      "epoch": 1.5453915672135277,
+      "grad_norm": 0.25241127610206604,
+      "learning_rate": 0.00016821311920511297,
+      "loss": 0.3425,
+      "step": 12270
+    },
+    {
+      "epoch": 1.546021349623705,
+      "grad_norm": 0.27286654710769653,
+      "learning_rate": 0.00016810400486335953,
+      "loss": 0.3604,
+      "step": 12275
+    },
+    {
+      "epoch": 1.5466511320338823,
+      "grad_norm": 0.3104652166366577,
+      "learning_rate": 0.0001679948807993419,
+      "loss": 0.3487,
+      "step": 12280
+    },
+    {
+      "epoch": 1.5472809144440596,
+      "grad_norm": 0.2972196042537689,
+      "learning_rate": 0.00016788574707166226,
+      "loss": 0.3555,
+      "step": 12285
+    },
+    {
+      "epoch": 1.5479106968542369,
+      "grad_norm": 0.29232388734817505,
+      "learning_rate": 0.00016777660373892787,
+      "loss": 0.3654,
+      "step": 12290
+    },
+    {
+      "epoch": 1.5485404792644142,
+      "grad_norm": 0.29798245429992676,
+      "learning_rate": 0.00016766745085975126,
+      "loss": 0.3575,
+      "step": 12295
+    },
+    {
+      "epoch": 1.5491702616745915,
+      "grad_norm": 0.2721775472164154,
+      "learning_rate": 0.0001675582884927499,
+      "loss": 0.3409,
+      "step": 12300
+    },
+    {
+      "epoch": 1.5498000440847686,
+      "grad_norm": 0.3131150007247925,
+      "learning_rate": 0.00016744911669654662,
+      "loss": 0.3695,
+      "step": 12305
+    },
+    {
+      "epoch": 1.550429826494946,
+      "grad_norm": 0.29543280601501465,
+      "learning_rate": 0.00016733993552976901,
+      "loss": 0.3572,
+      "step": 12310
+    },
+    {
+      "epoch": 1.5510596089051232,
+      "grad_norm": 0.3287052512168884,
+      "learning_rate": 0.00016723074505105,
+      "loss": 0.3681,
+      "step": 12315
+    },
+    {
+      "epoch": 1.5516893913153007,
+      "grad_norm": 0.2833183705806732,
+      "learning_rate": 0.0001671215453190273,
+      "loss": 0.3709,
+      "step": 12320
+    },
+    {
+      "epoch": 1.5523191737254778,
+      "grad_norm": 0.2558510899543762,
+      "learning_rate": 0.00016701233639234363,
+      "loss": 0.3404,
+      "step": 12325
+    },
+    {
+      "epoch": 1.552948956135655,
+      "grad_norm": 0.2524779438972473,
+      "learning_rate": 0.0001669031183296467,
+      "loss": 0.3492,
+      "step": 12330
+    },
+    {
+      "epoch": 1.5535787385458324,
+      "grad_norm": 0.2844880521297455,
+      "learning_rate": 0.00016679389118958918,
+      "loss": 0.3538,
+      "step": 12335
+    },
+    {
+      "epoch": 1.5542085209560097,
+      "grad_norm": 0.28060173988342285,
+      "learning_rate": 0.0001666846550308285,
+      "loss": 0.3615,
+      "step": 12340
+    },
+    {
+      "epoch": 1.554838303366187,
+      "grad_norm": 0.2490835040807724,
+      "learning_rate": 0.00016657540991202687,
+      "loss": 0.3655,
+      "step": 12345
+    },
+    {
+      "epoch": 1.5554680857763643,
+      "grad_norm": 0.27524054050445557,
+      "learning_rate": 0.00016646615589185153,
+      "loss": 0.3412,
+      "step": 12350
+    },
+    {
+      "epoch": 1.5560978681865416,
+      "grad_norm": 0.31142935156822205,
+      "learning_rate": 0.00016635689302897435,
+      "loss": 0.347,
+      "step": 12355
+    },
+    {
+      "epoch": 1.5567276505967187,
+      "grad_norm": 0.28053995966911316,
+      "learning_rate": 0.00016624762138207197,
+      "loss": 0.3838,
+      "step": 12360
+    },
+    {
+      "epoch": 1.5573574330068962,
+      "grad_norm": 0.2476169764995575,
+      "learning_rate": 0.0001661383410098258,
+      "loss": 0.3636,
+      "step": 12365
+    },
+    {
+      "epoch": 1.5579872154170733,
+      "grad_norm": 0.4054109752178192,
+      "learning_rate": 0.00016602905197092183,
+      "loss": 0.3657,
+      "step": 12370
+    },
+    {
+      "epoch": 1.5586169978272508,
+      "grad_norm": 0.2735072672367096,
+      "learning_rate": 0.00016591975432405084,
+      "loss": 0.3593,
+      "step": 12375
+    },
+    {
+      "epoch": 1.559246780237428,
+      "grad_norm": 0.2994532883167267,
+      "learning_rate": 0.00016581044812790817,
+      "loss": 0.3641,
+      "step": 12380
+    },
+    {
+      "epoch": 1.5598765626476052,
+      "grad_norm": 0.263090044260025,
+      "learning_rate": 0.0001657011334411936,
+      "loss": 0.3711,
+      "step": 12385
+    },
+    {
+      "epoch": 1.5605063450577825,
+      "grad_norm": 0.25073063373565674,
+      "learning_rate": 0.0001655918103226118,
+      "loss": 0.3554,
+      "step": 12390
+    },
+    {
+      "epoch": 1.5611361274679598,
+      "grad_norm": 0.2575080096721649,
+      "learning_rate": 0.00016548247883087168,
+      "loss": 0.3744,
+      "step": 12395
+    },
+    {
+      "epoch": 1.5617659098781371,
+      "grad_norm": 0.2630578577518463,
+      "learning_rate": 0.00016537313902468677,
+      "loss": 0.3501,
+      "step": 12400
+    },
+    {
+      "epoch": 1.5623956922883144,
+      "grad_norm": 0.3097805678844452,
+      "learning_rate": 0.00016526379096277503,
+      "loss": 0.3586,
+      "step": 12405
+    },
+    {
+      "epoch": 1.5630254746984917,
+      "grad_norm": 0.3104281723499298,
+      "learning_rate": 0.0001651544347038589,
+      "loss": 0.3643,
+      "step": 12410
+    },
+    {
+      "epoch": 1.5636552571086688,
+      "grad_norm": 0.3604758381843567,
+      "learning_rate": 0.0001650450703066652,
+      "loss": 0.3645,
+      "step": 12415
+    },
+    {
+      "epoch": 1.5642850395188463,
+      "grad_norm": 0.30638590455055237,
+      "learning_rate": 0.000164935697829925,
+      "loss": 0.3572,
+      "step": 12420
+    },
+    {
+      "epoch": 1.5649148219290234,
+      "grad_norm": 0.27940669655799866,
+      "learning_rate": 0.00016482631733237397,
+      "loss": 0.3636,
+      "step": 12425
+    },
+    {
+      "epoch": 1.565544604339201,
+      "grad_norm": 0.28857216238975525,
+      "learning_rate": 0.00016471692887275185,
+      "loss": 0.3601,
+      "step": 12430
+    },
+    {
+      "epoch": 1.566174386749378,
+      "grad_norm": 0.2992657721042633,
+      "learning_rate": 0.0001646075325098027,
+      "loss": 0.3621,
+      "step": 12435
+    },
+    {
+      "epoch": 1.5668041691595553,
+      "grad_norm": 0.28050917387008667,
+      "learning_rate": 0.00016449812830227498,
+      "loss": 0.3623,
+      "step": 12440
+    },
+    {
+      "epoch": 1.5674339515697326,
+      "grad_norm": 0.269634485244751,
+      "learning_rate": 0.0001643887163089212,
+      "loss": 0.3375,
+      "step": 12445
+    },
+    {
+      "epoch": 1.56806373397991,
+      "grad_norm": 0.2825991213321686,
+      "learning_rate": 0.00016427929658849807,
+      "loss": 0.3523,
+      "step": 12450
+    },
+    {
+      "epoch": 1.5686935163900873,
+      "grad_norm": 0.3219839334487915,
+      "learning_rate": 0.00016416986919976645,
+      "loss": 0.3588,
+      "step": 12455
+    },
+    {
+      "epoch": 1.5693232988002646,
+      "grad_norm": 0.2681691646575928,
+      "learning_rate": 0.00016406043420149146,
+      "loss": 0.3466,
+      "step": 12460
+    },
+    {
+      "epoch": 1.5699530812104419,
+      "grad_norm": 0.2719057500362396,
+      "learning_rate": 0.0001639509916524421,
+      "loss": 0.3599,
+      "step": 12465
+    },
+    {
+      "epoch": 1.570582863620619,
+      "grad_norm": 0.24405649304389954,
+      "learning_rate": 0.00016384154161139158,
+      "loss": 0.3402,
+      "step": 12470
+    },
+    {
+      "epoch": 1.5712126460307965,
+      "grad_norm": 0.306537002325058,
+      "learning_rate": 0.00016373208413711696,
+      "loss": 0.3283,
+      "step": 12475
+    },
+    {
+      "epoch": 1.5718424284409735,
+      "grad_norm": 0.28490665555000305,
+      "learning_rate": 0.0001636226192883996,
+      "loss": 0.3529,
+      "step": 12480
+    },
+    {
+      "epoch": 1.572472210851151,
+      "grad_norm": 0.2510652542114258,
+      "learning_rate": 0.00016351314712402442,
+      "loss": 0.3228,
+      "step": 12485
+    },
+    {
+      "epoch": 1.5731019932613282,
+      "grad_norm": 0.2670060694217682,
+      "learning_rate": 0.0001634036677027806,
+      "loss": 0.3592,
+      "step": 12490
+    },
+    {
+      "epoch": 1.5737317756715055,
+      "grad_norm": 0.29240545630455017,
+      "learning_rate": 0.00016329418108346105,
+      "loss": 0.3717,
+      "step": 12495
+    },
+    {
+      "epoch": 1.5743615580816828,
+      "grad_norm": 0.29088887572288513,
+      "learning_rate": 0.00016318468732486255,
+      "loss": 0.3679,
+      "step": 12500
+    },
+    {
+      "epoch": 1.57499134049186,
+      "grad_norm": 0.25105100870132446,
+      "learning_rate": 0.0001630751864857858,
+      "loss": 0.3464,
+      "step": 12505
+    },
+    {
+      "epoch": 1.5756211229020374,
+      "grad_norm": 0.26624953746795654,
+      "learning_rate": 0.00016296567862503526,
+      "loss": 0.3552,
+      "step": 12510
+    },
+    {
+      "epoch": 1.5762509053122147,
+      "grad_norm": 0.28500837087631226,
+      "learning_rate": 0.00016285616380141914,
+      "loss": 0.3591,
+      "step": 12515
+    },
+    {
+      "epoch": 1.576880687722392,
+      "grad_norm": 0.2937677502632141,
+      "learning_rate": 0.00016274664207374936,
+      "loss": 0.3664,
+      "step": 12520
+    },
+    {
+      "epoch": 1.577510470132569,
+      "grad_norm": 0.28588148951530457,
+      "learning_rate": 0.00016263711350084165,
+      "loss": 0.3767,
+      "step": 12525
+    },
+    {
+      "epoch": 1.5781402525427466,
+      "grad_norm": 0.31547772884368896,
+      "learning_rate": 0.0001625275781415153,
+      "loss": 0.3521,
+      "step": 12530
+    },
+    {
+      "epoch": 1.5787700349529237,
+      "grad_norm": 0.29322996735572815,
+      "learning_rate": 0.00016241803605459334,
+      "loss": 0.3777,
+      "step": 12535
+    },
+    {
+      "epoch": 1.5793998173631012,
+      "grad_norm": 0.29141756892204285,
+      "learning_rate": 0.00016230848729890238,
+      "loss": 0.3367,
+      "step": 12540
+    },
+    {
+      "epoch": 1.5800295997732783,
+      "grad_norm": 0.2583523094654083,
+      "learning_rate": 0.00016219893193327258,
+      "loss": 0.3473,
+      "step": 12545
+    },
+    {
+      "epoch": 1.5806593821834556,
+      "grad_norm": 0.26929906010627747,
+      "learning_rate": 0.00016208937001653765,
+      "loss": 0.3622,
+      "step": 12550
+    },
+    {
+      "epoch": 1.581289164593633,
+      "grad_norm": 0.2727062702178955,
+      "learning_rate": 0.0001619798016075349,
+      "loss": 0.3607,
+      "step": 12555
+    },
+    {
+      "epoch": 1.5819189470038102,
+      "grad_norm": 0.35366252064704895,
+      "learning_rate": 0.000161870226765105,
+      "loss": 0.3453,
+      "step": 12560
+    },
+    {
+      "epoch": 1.5825487294139875,
+      "grad_norm": 0.30689889192581177,
+      "learning_rate": 0.00016176064554809225,
+      "loss": 0.3672,
+      "step": 12565
+    },
+    {
+      "epoch": 1.5831785118241648,
+      "grad_norm": 0.2855357825756073,
+      "learning_rate": 0.00016165105801534414,
+      "loss": 0.3715,
+      "step": 12570
+    },
+    {
+      "epoch": 1.583808294234342,
+      "grad_norm": 0.33706697821617126,
+      "learning_rate": 0.00016154146422571176,
+      "loss": 0.3645,
+      "step": 12575
+    },
+    {
+      "epoch": 1.5844380766445192,
+      "grad_norm": 0.24410569667816162,
+      "learning_rate": 0.00016143186423804944,
+      "loss": 0.3576,
+      "step": 12580
+    },
+    {
+      "epoch": 1.5850678590546967,
+      "grad_norm": 0.33356451988220215,
+      "learning_rate": 0.00016132225811121492,
+      "loss": 0.3774,
+      "step": 12585
+    },
+    {
+      "epoch": 1.5856976414648738,
+      "grad_norm": 0.2804293632507324,
+      "learning_rate": 0.00016121264590406912,
+      "loss": 0.3656,
+      "step": 12590
+    },
+    {
+      "epoch": 1.5863274238750513,
+      "grad_norm": 0.29394668340682983,
+      "learning_rate": 0.0001611030276754764,
+      "loss": 0.3468,
+      "step": 12595
+    },
+    {
+      "epoch": 1.5869572062852284,
+      "grad_norm": 0.2657965421676636,
+      "learning_rate": 0.0001609934034843042,
+      "loss": 0.3518,
+      "step": 12600
+    },
+    {
+      "epoch": 1.5875869886954057,
+      "grad_norm": 0.25583842396736145,
+      "learning_rate": 0.00016088377338942318,
+      "loss": 0.361,
+      "step": 12605
+    },
+    {
+      "epoch": 1.588216771105583,
+      "grad_norm": 0.27986687421798706,
+      "learning_rate": 0.00016077413744970722,
+      "loss": 0.3771,
+      "step": 12610
+    },
+    {
+      "epoch": 1.5888465535157603,
+      "grad_norm": 0.3200220763683319,
+      "learning_rate": 0.0001606644957240334,
+      "loss": 0.3666,
+      "step": 12615
+    },
+    {
+      "epoch": 1.5894763359259376,
+      "grad_norm": 0.29622554779052734,
+      "learning_rate": 0.00016055484827128173,
+      "loss": 0.3469,
+      "step": 12620
+    },
+    {
+      "epoch": 1.590106118336115,
+      "grad_norm": 0.3073137700557709,
+      "learning_rate": 0.00016044519515033545,
+      "loss": 0.3382,
+      "step": 12625
+    },
+    {
+      "epoch": 1.5907359007462922,
+      "grad_norm": 0.31342241168022156,
+      "learning_rate": 0.00016033553642008077,
+      "loss": 0.357,
+      "step": 12630
+    },
+    {
+      "epoch": 1.5913656831564693,
+      "grad_norm": 0.2913351058959961,
+      "learning_rate": 0.00016022587213940698,
+      "loss": 0.3487,
+      "step": 12635
+    },
+    {
+      "epoch": 1.5919954655666468,
+      "grad_norm": 0.2823300361633301,
+      "learning_rate": 0.00016011620236720621,
+      "loss": 0.3367,
+      "step": 12640
+    },
+    {
+      "epoch": 1.592625247976824,
+      "grad_norm": 0.3134678304195404,
+      "learning_rate": 0.00016000652716237373,
+      "loss": 0.3393,
+      "step": 12645
+    },
+    {
+      "epoch": 1.5932550303870014,
+      "grad_norm": 0.3235761821269989,
+      "learning_rate": 0.0001598968465838076,
+      "loss": 0.3752,
+      "step": 12650
+    },
+    {
+      "epoch": 1.5938848127971785,
+      "grad_norm": 0.26606664061546326,
+      "learning_rate": 0.00015978716069040875,
+      "loss": 0.3413,
+      "step": 12655
+    },
+    {
+      "epoch": 1.5945145952073558,
+      "grad_norm": 0.30575528740882874,
+      "learning_rate": 0.0001596774695410811,
+      "loss": 0.3715,
+      "step": 12660
+    },
+    {
+      "epoch": 1.5951443776175331,
+      "grad_norm": 0.3017826974391937,
+      "learning_rate": 0.0001595677731947312,
+      "loss": 0.3586,
+      "step": 12665
+    },
+    {
+      "epoch": 1.5957741600277104,
+      "grad_norm": 0.3203217089176178,
+      "learning_rate": 0.00015945807171026855,
+      "loss": 0.3753,
+      "step": 12670
+    },
+    {
+      "epoch": 1.5964039424378877,
+      "grad_norm": 0.2615835666656494,
+      "learning_rate": 0.00015934836514660536,
+      "loss": 0.3641,
+      "step": 12675
+    },
+    {
+      "epoch": 1.5970337248480648,
+      "grad_norm": 0.2814265191555023,
+      "learning_rate": 0.00015923865356265652,
+      "loss": 0.3467,
+      "step": 12680
+    },
+    {
+      "epoch": 1.5976635072582424,
+      "grad_norm": 0.28240394592285156,
+      "learning_rate": 0.00015912893701733975,
+      "loss": 0.3405,
+      "step": 12685
+    },
+    {
+      "epoch": 1.5982932896684194,
+      "grad_norm": 0.2675967514514923,
+      "learning_rate": 0.0001590192155695752,
+      "loss": 0.3341,
+      "step": 12690
+    },
+    {
+      "epoch": 1.598923072078597,
+      "grad_norm": 0.33408063650131226,
+      "learning_rate": 0.00015890948927828593,
+      "loss": 0.3431,
+      "step": 12695
+    },
+    {
+      "epoch": 1.599552854488774,
+      "grad_norm": 0.2793383300304413,
+      "learning_rate": 0.00015879975820239737,
+      "loss": 0.3334,
+      "step": 12700
+    },
+    {
+      "epoch": 1.6001826368989516,
+      "grad_norm": 0.29299893975257874,
+      "learning_rate": 0.00015869002240083765,
+      "loss": 0.3479,
+      "step": 12705
+    },
+    {
+      "epoch": 1.6008124193091287,
+      "grad_norm": 0.2782432436943054,
+      "learning_rate": 0.0001585802819325374,
+      "loss": 0.3491,
+      "step": 12710
+    },
+    {
+      "epoch": 1.601442201719306,
+      "grad_norm": 0.2812289297580719,
+      "learning_rate": 0.00015847053685642977,
+      "loss": 0.3406,
+      "step": 12715
+    },
+    {
+      "epoch": 1.6020719841294833,
+      "grad_norm": 0.2464970499277115,
+      "learning_rate": 0.00015836078723145032,
+      "loss": 0.3539,
+      "step": 12720
+    },
+    {
+      "epoch": 1.6027017665396606,
+      "grad_norm": 0.3087675869464874,
+      "learning_rate": 0.0001582510331165372,
+      "loss": 0.356,
+      "step": 12725
+    },
+    {
+      "epoch": 1.6033315489498379,
+      "grad_norm": 0.2726532816886902,
+      "learning_rate": 0.0001581412745706308,
+      "loss": 0.3443,
+      "step": 12730
+    },
+    {
+      "epoch": 1.603961331360015,
+      "grad_norm": 0.28401410579681396,
+      "learning_rate": 0.00015803151165267397,
+      "loss": 0.3359,
+      "step": 12735
+    },
+    {
+      "epoch": 1.6045911137701925,
+      "grad_norm": 0.2700473666191101,
+      "learning_rate": 0.00015792174442161194,
+      "loss": 0.3523,
+      "step": 12740
+    },
+    {
+      "epoch": 1.6052208961803696,
+      "grad_norm": 0.32183146476745605,
+      "learning_rate": 0.00015781197293639223,
+      "loss": 0.3765,
+      "step": 12745
+    },
+    {
+      "epoch": 1.605850678590547,
+      "grad_norm": 0.2805304229259491,
+      "learning_rate": 0.0001577021972559646,
+      "loss": 0.3546,
+      "step": 12750
+    },
+    {
+      "epoch": 1.6064804610007242,
+      "grad_norm": 0.30137768387794495,
+      "learning_rate": 0.00015759241743928108,
+      "loss": 0.3721,
+      "step": 12755
+    },
+    {
+      "epoch": 1.6071102434109017,
+      "grad_norm": 0.25476494431495667,
+      "learning_rate": 0.00015748263354529597,
+      "loss": 0.3281,
+      "step": 12760
+    },
+    {
+      "epoch": 1.6077400258210788,
+      "grad_norm": 0.3311167061328888,
+      "learning_rate": 0.0001573728456329657,
+      "loss": 0.3875,
+      "step": 12765
+    },
+    {
+      "epoch": 1.608369808231256,
+      "grad_norm": 0.27148258686065674,
+      "learning_rate": 0.00015726305376124897,
+      "loss": 0.3547,
+      "step": 12770
+    },
+    {
+      "epoch": 1.6089995906414334,
+      "grad_norm": 0.25366437435150146,
+      "learning_rate": 0.00015715325798910644,
+      "loss": 0.3423,
+      "step": 12775
+    },
+    {
+      "epoch": 1.6096293730516107,
+      "grad_norm": 0.2699160873889923,
+      "learning_rate": 0.000157043458375501,
+      "loss": 0.3347,
+      "step": 12780
+    },
+    {
+      "epoch": 1.610259155461788,
+      "grad_norm": 0.2672334611415863,
+      "learning_rate": 0.00015693365497939743,
+      "loss": 0.3354,
+      "step": 12785
+    },
+    {
+      "epoch": 1.610888937871965,
+      "grad_norm": 0.3269018828868866,
+      "learning_rate": 0.00015682384785976284,
+      "loss": 0.3427,
+      "step": 12790
+    },
+    {
+      "epoch": 1.6115187202821426,
+      "grad_norm": 0.2637929320335388,
+      "learning_rate": 0.00015671403707556605,
+      "loss": 0.3501,
+      "step": 12795
+    },
+    {
+      "epoch": 1.6121485026923197,
+      "grad_norm": 0.2606005072593689,
+      "learning_rate": 0.00015660422268577801,
+      "loss": 0.3387,
+      "step": 12800
+    },
+    {
+      "epoch": 1.6127782851024972,
+      "grad_norm": 0.30220791697502136,
+      "learning_rate": 0.00015649440474937152,
+      "loss": 0.3489,
+      "step": 12805
+    },
+    {
+      "epoch": 1.6134080675126743,
+      "grad_norm": 0.29726284742355347,
+      "learning_rate": 0.0001563845833253213,
+      "loss": 0.3358,
+      "step": 12810
+    },
+    {
+      "epoch": 1.6140378499228518,
+      "grad_norm": 0.2928326427936554,
+      "learning_rate": 0.000156274758472604,
+      "loss": 0.3236,
+      "step": 12815
+    },
+    {
+      "epoch": 1.614667632333029,
+      "grad_norm": 0.31645599007606506,
+      "learning_rate": 0.0001561649302501981,
+      "loss": 0.3571,
+      "step": 12820
+    },
+    {
+      "epoch": 1.6152974147432062,
+      "grad_norm": 0.26705339550971985,
+      "learning_rate": 0.00015605509871708382,
+      "loss": 0.3671,
+      "step": 12825
+    },
+    {
+      "epoch": 1.6159271971533835,
+      "grad_norm": 0.2691219449043274,
+      "learning_rate": 0.00015594526393224322,
+      "loss": 0.3452,
+      "step": 12830
+    },
+    {
+      "epoch": 1.6165569795635608,
+      "grad_norm": 0.2822478413581848,
+      "learning_rate": 0.00015583542595466005,
+      "loss": 0.3273,
+      "step": 12835
+    },
+    {
+      "epoch": 1.6171867619737381,
+      "grad_norm": 0.2974461019039154,
+      "learning_rate": 0.00015572558484331994,
+      "loss": 0.3652,
+      "step": 12840
+    },
+    {
+      "epoch": 1.6178165443839152,
+      "grad_norm": 0.2611928880214691,
+      "learning_rate": 0.00015561574065720986,
+      "loss": 0.3445,
+      "step": 12845
+    },
+    {
+      "epoch": 1.6184463267940927,
+      "grad_norm": 0.2836850583553314,
+      "learning_rate": 0.00015550589345531885,
+      "loss": 0.3326,
+      "step": 12850
+    },
+    {
+      "epoch": 1.6190761092042698,
+      "grad_norm": 0.2735482156276703,
+      "learning_rate": 0.00015539604329663725,
+      "loss": 0.3532,
+      "step": 12855
+    },
+    {
+      "epoch": 1.6197058916144473,
+      "grad_norm": 0.34394770860671997,
+      "learning_rate": 0.0001552861902401572,
+      "loss": 0.3532,
+      "step": 12860
+    },
+    {
+      "epoch": 1.6203356740246244,
+      "grad_norm": 0.2786300778388977,
+      "learning_rate": 0.0001551763343448722,
+      "loss": 0.3591,
+      "step": 12865
+    },
+    {
+      "epoch": 1.6209654564348017,
+      "grad_norm": 0.26574140787124634,
+      "learning_rate": 0.00015506647566977737,
+      "loss": 0.3527,
+      "step": 12870
+    },
+    {
+      "epoch": 1.621595238844979,
+      "grad_norm": 0.23986276984214783,
+      "learning_rate": 0.00015495661427386944,
+      "loss": 0.3437,
+      "step": 12875
+    },
+    {
+      "epoch": 1.6222250212551563,
+      "grad_norm": 0.29332754015922546,
+      "learning_rate": 0.0001548467502161464,
+      "loss": 0.3323,
+      "step": 12880
+    },
+    {
+      "epoch": 1.6228548036653336,
+      "grad_norm": 0.34338971972465515,
+      "learning_rate": 0.0001547368835556078,
+      "loss": 0.3367,
+      "step": 12885
+    },
+    {
+      "epoch": 1.623484586075511,
+      "grad_norm": 0.3112575113773346,
+      "learning_rate": 0.00015462701435125451,
+      "loss": 0.3392,
+      "step": 12890
+    },
+    {
+      "epoch": 1.6241143684856882,
+      "grad_norm": 0.26299479603767395,
+      "learning_rate": 0.0001545171426620888,
+      "loss": 0.3194,
+      "step": 12895
+    },
+    {
+      "epoch": 1.6247441508958653,
+      "grad_norm": 0.27403828501701355,
+      "learning_rate": 0.00015440726854711436,
+      "loss": 0.3344,
+      "step": 12900
+    },
+    {
+      "epoch": 1.6253739333060429,
+      "grad_norm": 0.2603330910205841,
+      "learning_rate": 0.000154297392065336,
+      "loss": 0.3564,
+      "step": 12905
+    },
+    {
+      "epoch": 1.62600371571622,
+      "grad_norm": 0.2812626361846924,
+      "learning_rate": 0.00015418751327575994,
+      "loss": 0.3583,
+      "step": 12910
+    },
+    {
+      "epoch": 1.6266334981263975,
+      "grad_norm": 0.27280038595199585,
+      "learning_rate": 0.0001540776322373936,
+      "loss": 0.3568,
+      "step": 12915
+    },
+    {
+      "epoch": 1.6272632805365745,
+      "grad_norm": 0.3073441982269287,
+      "learning_rate": 0.0001539677490092456,
+      "loss": 0.3336,
+      "step": 12920
+    },
+    {
+      "epoch": 1.6278930629467518,
+      "grad_norm": 0.2868177890777588,
+      "learning_rate": 0.00015385786365032576,
+      "loss": 0.3455,
+      "step": 12925
+    },
+    {
+      "epoch": 1.6285228453569291,
+      "grad_norm": 0.2661624550819397,
+      "learning_rate": 0.000153747976219645,
+      "loss": 0.3377,
+      "step": 12930
+    },
+    {
+      "epoch": 1.6291526277671065,
+      "grad_norm": 0.3179323673248291,
+      "learning_rate": 0.0001536380867762154,
+      "loss": 0.3706,
+      "step": 12935
+    },
+    {
+      "epoch": 1.6297824101772838,
+      "grad_norm": 0.30941662192344666,
+      "learning_rate": 0.0001535281953790501,
+      "loss": 0.3514,
+      "step": 12940
+    },
+    {
+      "epoch": 1.630412192587461,
+      "grad_norm": 0.3018413782119751,
+      "learning_rate": 0.0001534183020871633,
+      "loss": 0.3516,
+      "step": 12945
+    },
+    {
+      "epoch": 1.6310419749976384,
+      "grad_norm": 0.34621462225914,
+      "learning_rate": 0.00015330840695957019,
+      "loss": 0.3522,
+      "step": 12950
+    },
+    {
+      "epoch": 1.6316717574078154,
+      "grad_norm": 0.2858521342277527,
+      "learning_rate": 0.000153198510055287,
+      "loss": 0.3378,
+      "step": 12955
+    },
+    {
+      "epoch": 1.632301539817993,
+      "grad_norm": 0.2880783975124359,
+      "learning_rate": 0.00015308861143333076,
+      "loss": 0.3615,
+      "step": 12960
+    },
+    {
+      "epoch": 1.63293132222817,
+      "grad_norm": 0.24324443936347961,
+      "learning_rate": 0.00015297871115271976,
+      "loss": 0.3346,
+      "step": 12965
+    },
+    {
+      "epoch": 1.6335611046383476,
+      "grad_norm": 0.26982635259628296,
+      "learning_rate": 0.00015286880927247273,
+      "loss": 0.3423,
+      "step": 12970
+    },
+    {
+      "epoch": 1.6341908870485247,
+      "grad_norm": 0.27813199162483215,
+      "learning_rate": 0.00015275890585160961,
+      "loss": 0.3545,
+      "step": 12975
+    },
+    {
+      "epoch": 1.634820669458702,
+      "grad_norm": 0.27575090527534485,
+      "learning_rate": 0.00015264900094915106,
+      "loss": 0.3357,
+      "step": 12980
+    },
+    {
+      "epoch": 1.6354504518688793,
+      "grad_norm": 0.25838521122932434,
+      "learning_rate": 0.00015253909462411847,
+      "loss": 0.3244,
+      "step": 12985
+    },
+    {
+      "epoch": 1.6360802342790566,
+      "grad_norm": 0.2889041602611542,
+      "learning_rate": 0.00015242918693553404,
+      "loss": 0.3297,
+      "step": 12990
+    },
+    {
+      "epoch": 1.6367100166892339,
+      "grad_norm": 0.3074316680431366,
+      "learning_rate": 0.0001523192779424208,
+      "loss": 0.3525,
+      "step": 12995
+    },
+    {
+      "epoch": 1.6373397990994112,
+      "grad_norm": 0.26425209641456604,
+      "learning_rate": 0.00015220936770380227,
+      "loss": 0.3493,
+      "step": 13000
+    },
+    {
+      "epoch": 1.6373397990994112,
+      "eval_loss": 0.30248695611953735,
+      "eval_runtime": 6.1682,
+      "eval_samples_per_second": 162.123,
+      "eval_steps_per_second": 10.214,
+      "step": 13000
+    },
+    {
+      "epoch": 1.6379695815095885,
+      "grad_norm": 0.2873767018318176,
+      "learning_rate": 0.00015209945627870283,
+      "loss": 0.3838,
+      "step": 13005
+    },
+    {
+      "epoch": 1.6385993639197656,
+      "grad_norm": 0.2895953059196472,
+      "learning_rate": 0.0001519895437261474,
+      "loss": 0.3509,
+      "step": 13010
+    },
+    {
+      "epoch": 1.639229146329943,
+      "grad_norm": 0.2910915017127991,
+      "learning_rate": 0.0001518796301051616,
+      "loss": 0.326,
+      "step": 13015
+    },
+    {
+      "epoch": 1.6398589287401202,
+      "grad_norm": 0.2735256552696228,
+      "learning_rate": 0.00015176971547477142,
+      "loss": 0.366,
+      "step": 13020
+    },
+    {
+      "epoch": 1.6404887111502977,
+      "grad_norm": 0.3099430501461029,
+      "learning_rate": 0.00015165979989400366,
+      "loss": 0.3226,
+      "step": 13025
+    },
+    {
+      "epoch": 1.6411184935604748,
+      "grad_norm": 0.2963193655014038,
+      "learning_rate": 0.00015154988342188543,
+      "loss": 0.3301,
+      "step": 13030
+    },
+    {
+      "epoch": 1.641748275970652,
+      "grad_norm": 0.28377631306648254,
+      "learning_rate": 0.0001514399661174444,
+      "loss": 0.3143,
+      "step": 13035
+    },
+    {
+      "epoch": 1.6423780583808294,
+      "grad_norm": 0.25847306847572327,
+      "learning_rate": 0.00015133004803970866,
+      "loss": 0.325,
+      "step": 13040
+    },
+    {
+      "epoch": 1.6430078407910067,
+      "grad_norm": 0.2919864058494568,
+      "learning_rate": 0.00015122012924770675,
+      "loss": 0.3543,
+      "step": 13045
+    },
+    {
+      "epoch": 1.643637623201184,
+      "grad_norm": 0.31185677647590637,
+      "learning_rate": 0.00015111020980046756,
+      "loss": 0.3546,
+      "step": 13050
+    },
+    {
+      "epoch": 1.6442674056113613,
+      "grad_norm": 0.27933362126350403,
+      "learning_rate": 0.00015100028975702036,
+      "loss": 0.3344,
+      "step": 13055
+    },
+    {
+      "epoch": 1.6448971880215386,
+      "grad_norm": 0.2898799777030945,
+      "learning_rate": 0.00015089036917639468,
+      "loss": 0.3473,
+      "step": 13060
+    },
+    {
+      "epoch": 1.6455269704317157,
+      "grad_norm": 0.31464672088623047,
+      "learning_rate": 0.00015078044811762047,
+      "loss": 0.3418,
+      "step": 13065
+    },
+    {
+      "epoch": 1.6461567528418932,
+      "grad_norm": 0.2676648199558258,
+      "learning_rate": 0.00015067052663972775,
+      "loss": 0.3331,
+      "step": 13070
+    },
+    {
+      "epoch": 1.6467865352520703,
+      "grad_norm": 0.30420759320259094,
+      "learning_rate": 0.0001505606048017469,
+      "loss": 0.3544,
+      "step": 13075
+    },
+    {
+      "epoch": 1.6474163176622478,
+      "grad_norm": 0.3160271942615509,
+      "learning_rate": 0.00015045068266270848,
+      "loss": 0.3526,
+      "step": 13080
+    },
+    {
+      "epoch": 1.648046100072425,
+      "grad_norm": 0.31276562809944153,
+      "learning_rate": 0.0001503407602816432,
+      "loss": 0.3213,
+      "step": 13085
+    },
+    {
+      "epoch": 1.6486758824826022,
+      "grad_norm": 0.316756933927536,
+      "learning_rate": 0.00015023083771758183,
+      "loss": 0.3446,
+      "step": 13090
+    },
+    {
+      "epoch": 1.6493056648927795,
+      "grad_norm": 0.23935994505882263,
+      "learning_rate": 0.00015012091502955533,
+      "loss": 0.3416,
+      "step": 13095
+    },
+    {
+      "epoch": 1.6499354473029568,
+      "grad_norm": 0.2719472348690033,
+      "learning_rate": 0.00015001099227659475,
+      "loss": 0.3567,
+      "step": 13100
+    },
+    {
+      "epoch": 1.6505652297131341,
+      "grad_norm": 0.3108009696006775,
+      "learning_rate": 0.00014990106951773098,
+      "loss": 0.3524,
+      "step": 13105
+    },
+    {
+      "epoch": 1.6511950121233114,
+      "grad_norm": 0.3002628982067108,
+      "learning_rate": 0.00014979114681199524,
+      "loss": 0.3314,
+      "step": 13110
+    },
+    {
+      "epoch": 1.6518247945334887,
+      "grad_norm": 0.32287389039993286,
+      "learning_rate": 0.0001496812242184184,
+      "loss": 0.3376,
+      "step": 13115
+    },
+    {
+      "epoch": 1.6524545769436658,
+      "grad_norm": 0.27193522453308105,
+      "learning_rate": 0.0001495713017960314,
+      "loss": 0.3443,
+      "step": 13120
+    },
+    {
+      "epoch": 1.6530843593538433,
+      "grad_norm": 0.30429700016975403,
+      "learning_rate": 0.00014946137960386512,
+      "loss": 0.3345,
+      "step": 13125
+    },
+    {
+      "epoch": 1.6537141417640204,
+      "grad_norm": 0.2757263481616974,
+      "learning_rate": 0.00014935145770095034,
+      "loss": 0.3405,
+      "step": 13130
+    },
+    {
+      "epoch": 1.654343924174198,
+      "grad_norm": 0.274728000164032,
+      "learning_rate": 0.00014924153614631754,
+      "loss": 0.3199,
+      "step": 13135
+    },
+    {
+      "epoch": 1.654973706584375,
+      "grad_norm": 0.2992052137851715,
+      "learning_rate": 0.0001491316149989972,
+      "loss": 0.3641,
+      "step": 13140
+    },
+    {
+      "epoch": 1.6556034889945523,
+      "grad_norm": 0.28687140345573425,
+      "learning_rate": 0.00014902169431801947,
+      "loss": 0.3586,
+      "step": 13145
+    },
+    {
+      "epoch": 1.6562332714047296,
+      "grad_norm": 0.31748563051223755,
+      "learning_rate": 0.00014891177416241416,
+      "loss": 0.3318,
+      "step": 13150
+    },
+    {
+      "epoch": 1.656863053814907,
+      "grad_norm": 0.2876995801925659,
+      "learning_rate": 0.00014880185459121103,
+      "loss": 0.3446,
+      "step": 13155
+    },
+    {
+      "epoch": 1.6574928362250843,
+      "grad_norm": 0.2874261736869812,
+      "learning_rate": 0.00014869193566343934,
+      "loss": 0.3058,
+      "step": 13160
+    },
+    {
+      "epoch": 1.6581226186352616,
+      "grad_norm": 0.2720824182033539,
+      "learning_rate": 0.00014858201743812806,
+      "loss": 0.3332,
+      "step": 13165
+    },
+    {
+      "epoch": 1.6587524010454389,
+      "grad_norm": 0.27765411138534546,
+      "learning_rate": 0.00014847209997430582,
+      "loss": 0.3428,
+      "step": 13170
+    },
+    {
+      "epoch": 1.659382183455616,
+      "grad_norm": 0.28871631622314453,
+      "learning_rate": 0.0001483621833310008,
+      "loss": 0.3325,
+      "step": 13175
+    },
+    {
+      "epoch": 1.6600119658657935,
+      "grad_norm": 0.2875865697860718,
+      "learning_rate": 0.00014825226756724077,
+      "loss": 0.3527,
+      "step": 13180
+    },
+    {
+      "epoch": 1.6606417482759706,
+      "grad_norm": 0.2774711549282074,
+      "learning_rate": 0.00014814235274205297,
+      "loss": 0.335,
+      "step": 13185
+    },
+    {
+      "epoch": 1.661271530686148,
+      "grad_norm": 0.2727283537387848,
+      "learning_rate": 0.00014803243891446416,
+      "loss": 0.3393,
+      "step": 13190
+    },
+    {
+      "epoch": 1.6619013130963252,
+      "grad_norm": 0.27977532148361206,
+      "learning_rate": 0.00014792252614350055,
+      "loss": 0.3566,
+      "step": 13195
+    },
+    {
+      "epoch": 1.6625310955065025,
+      "grad_norm": 0.29823413491249084,
+      "learning_rate": 0.0001478126144881879,
+      "loss": 0.3287,
+      "step": 13200
+    },
+    {
+      "epoch": 1.6631608779166798,
+      "grad_norm": 0.2849923372268677,
+      "learning_rate": 0.00014770270400755125,
+      "loss": 0.3166,
+      "step": 13205
+    },
+    {
+      "epoch": 1.663790660326857,
+      "grad_norm": 0.259219229221344,
+      "learning_rate": 0.00014759279476061503,
+      "loss": 0.336,
+      "step": 13210
+    },
+    {
+      "epoch": 1.6644204427370344,
+      "grad_norm": 0.2877882719039917,
+      "learning_rate": 0.00014748288680640302,
+      "loss": 0.3506,
+      "step": 13215
+    },
+    {
+      "epoch": 1.6650502251472117,
+      "grad_norm": 0.2952651381492615,
+      "learning_rate": 0.00014737298020393828,
+      "loss": 0.3562,
+      "step": 13220
+    },
+    {
+      "epoch": 1.665680007557389,
+      "grad_norm": 0.25878390669822693,
+      "learning_rate": 0.00014726307501224312,
+      "loss": 0.3289,
+      "step": 13225
+    },
+    {
+      "epoch": 1.666309789967566,
+      "grad_norm": 0.29914605617523193,
+      "learning_rate": 0.00014715317129033924,
+      "loss": 0.3321,
+      "step": 13230
+    },
+    {
+      "epoch": 1.6669395723777436,
+      "grad_norm": 0.27533242106437683,
+      "learning_rate": 0.00014704326909724738,
+      "loss": 0.3234,
+      "step": 13235
+    },
+    {
+      "epoch": 1.6675693547879207,
+      "grad_norm": 0.2584016025066376,
+      "learning_rate": 0.0001469333684919876,
+      "loss": 0.3181,
+      "step": 13240
+    },
+    {
+      "epoch": 1.6681991371980982,
+      "grad_norm": 0.262953519821167,
+      "learning_rate": 0.00014682346953357898,
+      "loss": 0.3127,
+      "step": 13245
+    },
+    {
+      "epoch": 1.6688289196082753,
+      "grad_norm": 0.3399054706096649,
+      "learning_rate": 0.00014671357228103978,
+      "loss": 0.3529,
+      "step": 13250
+    },
+    {
+      "epoch": 1.6694587020184526,
+      "grad_norm": 0.26437637209892273,
+      "learning_rate": 0.00014660367679338732,
+      "loss": 0.318,
+      "step": 13255
+    },
+    {
+      "epoch": 1.67008848442863,
+      "grad_norm": 0.28796815872192383,
+      "learning_rate": 0.000146493783129638,
+      "loss": 0.3226,
+      "step": 13260
+    },
+    {
+      "epoch": 1.6707182668388072,
+      "grad_norm": 0.3208424150943756,
+      "learning_rate": 0.00014638389134880722,
+      "loss": 0.3661,
+      "step": 13265
+    },
+    {
+      "epoch": 1.6713480492489845,
+      "grad_norm": 0.2934640347957611,
+      "learning_rate": 0.00014627400150990941,
+      "loss": 0.3414,
+      "step": 13270
+    },
+    {
+      "epoch": 1.6719778316591618,
+      "grad_norm": 0.28860223293304443,
+      "learning_rate": 0.0001461641136719579,
+      "loss": 0.3386,
+      "step": 13275
+    },
+    {
+      "epoch": 1.672607614069339,
+      "grad_norm": 0.2960747182369232,
+      "learning_rate": 0.00014605422789396494,
+      "loss": 0.3466,
+      "step": 13280
+    },
+    {
+      "epoch": 1.6732373964795162,
+      "grad_norm": 0.25040510296821594,
+      "learning_rate": 0.00014594434423494178,
+      "loss": 0.3366,
+      "step": 13285
+    },
+    {
+      "epoch": 1.6738671788896937,
+      "grad_norm": 0.2958894371986389,
+      "learning_rate": 0.0001458344627538984,
+      "loss": 0.3614,
+      "step": 13290
+    },
+    {
+      "epoch": 1.6744969612998708,
+      "grad_norm": 0.26937004923820496,
+      "learning_rate": 0.00014572458350984362,
+      "loss": 0.3499,
+      "step": 13295
+    },
+    {
+      "epoch": 1.6751267437100483,
+      "grad_norm": 0.267607182264328,
+      "learning_rate": 0.00014561470656178517,
+      "loss": 0.3268,
+      "step": 13300
+    },
+    {
+      "epoch": 1.6757565261202254,
+      "grad_norm": 0.30197760462760925,
+      "learning_rate": 0.0001455048319687295,
+      "loss": 0.3212,
+      "step": 13305
+    },
+    {
+      "epoch": 1.6763863085304027,
+      "grad_norm": 0.29999008774757385,
+      "learning_rate": 0.0001453949597896817,
+      "loss": 0.3492,
+      "step": 13310
+    },
+    {
+      "epoch": 1.67701609094058,
+      "grad_norm": 0.30626264214515686,
+      "learning_rate": 0.00014528509008364572,
+      "loss": 0.3541,
+      "step": 13315
+    },
+    {
+      "epoch": 1.6776458733507573,
+      "grad_norm": 0.2915571630001068,
+      "learning_rate": 0.0001451752229096241,
+      "loss": 0.3231,
+      "step": 13320
+    },
+    {
+      "epoch": 1.6782756557609346,
+      "grad_norm": 0.2660951018333435,
+      "learning_rate": 0.0001450653583266179,
+      "loss": 0.321,
+      "step": 13325
+    },
+    {
+      "epoch": 1.678905438171112,
+      "grad_norm": 0.2831597924232483,
+      "learning_rate": 0.00014495549639362707,
+      "loss": 0.3243,
+      "step": 13330
+    },
+    {
+      "epoch": 1.6795352205812892,
+      "grad_norm": 0.2856467664241791,
+      "learning_rate": 0.0001448456371696499,
+      "loss": 0.3134,
+      "step": 13335
+    },
+    {
+      "epoch": 1.6801650029914663,
+      "grad_norm": 0.31137335300445557,
+      "learning_rate": 0.00014473578071368324,
+      "loss": 0.3266,
+      "step": 13340
+    },
+    {
+      "epoch": 1.6807947854016438,
+      "grad_norm": 0.3102738857269287,
+      "learning_rate": 0.0001446259270847226,
+      "loss": 0.3368,
+      "step": 13345
+    },
+    {
+      "epoch": 1.681424567811821,
+      "grad_norm": 0.2788311839103699,
+      "learning_rate": 0.00014451607634176196,
+      "loss": 0.345,
+      "step": 13350
+    },
+    {
+      "epoch": 1.6820543502219985,
+      "grad_norm": 0.26762083172798157,
+      "learning_rate": 0.0001444062285437935,
+      "loss": 0.3112,
+      "step": 13355
+    },
+    {
+      "epoch": 1.6826841326321755,
+      "grad_norm": 0.30155837535858154,
+      "learning_rate": 0.00014429638374980814,
+      "loss": 0.3353,
+      "step": 13360
+    },
+    {
+      "epoch": 1.6833139150423528,
+      "grad_norm": 0.3196204602718353,
+      "learning_rate": 0.00014418654201879498,
+      "loss": 0.3738,
+      "step": 13365
+    },
+    {
+      "epoch": 1.6839436974525301,
+      "grad_norm": 0.29560673236846924,
+      "learning_rate": 0.0001440767034097415,
+      "loss": 0.3458,
+      "step": 13370
+    },
+    {
+      "epoch": 1.6845734798627074,
+      "grad_norm": 0.30189448595046997,
+      "learning_rate": 0.00014396686798163365,
+      "loss": 0.3577,
+      "step": 13375
+    },
+    {
+      "epoch": 1.6852032622728847,
+      "grad_norm": 0.29545098543167114,
+      "learning_rate": 0.00014385703579345544,
+      "loss": 0.3299,
+      "step": 13380
+    },
+    {
+      "epoch": 1.685833044683062,
+      "grad_norm": 0.3403629660606384,
+      "learning_rate": 0.00014374720690418942,
+      "loss": 0.3349,
+      "step": 13385
+    },
+    {
+      "epoch": 1.6864628270932394,
+      "grad_norm": 0.2561693489551544,
+      "learning_rate": 0.0001436373813728161,
+      "loss": 0.321,
+      "step": 13390
+    },
+    {
+      "epoch": 1.6870926095034164,
+      "grad_norm": 0.2968713641166687,
+      "learning_rate": 0.00014352755925831428,
+      "loss": 0.3314,
+      "step": 13395
+    },
+    {
+      "epoch": 1.687722391913594,
+      "grad_norm": 0.25213027000427246,
+      "learning_rate": 0.00014341774061966096,
+      "loss": 0.3245,
+      "step": 13400
+    },
+    {
+      "epoch": 1.688352174323771,
+      "grad_norm": 0.26504096388816833,
+      "learning_rate": 0.00014330792551583133,
+      "loss": 0.324,
+      "step": 13405
+    },
+    {
+      "epoch": 1.6889819567339486,
+      "grad_norm": 0.31459683179855347,
+      "learning_rate": 0.00014319811400579854,
+      "loss": 0.33,
+      "step": 13410
+    },
+    {
+      "epoch": 1.6896117391441257,
+      "grad_norm": 0.31566324830055237,
+      "learning_rate": 0.00014308830614853392,
+      "loss": 0.3097,
+      "step": 13415
+    },
+    {
+      "epoch": 1.690241521554303,
+      "grad_norm": 0.3083827793598175,
+      "learning_rate": 0.00014297850200300683,
+      "loss": 0.3345,
+      "step": 13420
+    },
+    {
+      "epoch": 1.6908713039644803,
+      "grad_norm": 0.29203763604164124,
+      "learning_rate": 0.0001428687016281845,
+      "loss": 0.3459,
+      "step": 13425
+    },
+    {
+      "epoch": 1.6915010863746576,
+      "grad_norm": 0.28596800565719604,
+      "learning_rate": 0.00014275890508303225,
+      "loss": 0.3188,
+      "step": 13430
+    },
+    {
+      "epoch": 1.6921308687848349,
+      "grad_norm": 0.3753102421760559,
+      "learning_rate": 0.00014264911242651342,
+      "loss": 0.3457,
+      "step": 13435
+    },
+    {
+      "epoch": 1.6927606511950122,
+      "grad_norm": 0.28502312302589417,
+      "learning_rate": 0.0001425393237175891,
+      "loss": 0.3295,
+      "step": 13440
+    },
+    {
+      "epoch": 1.6933904336051895,
+      "grad_norm": 0.3175462782382965,
+      "learning_rate": 0.00014242953901521838,
+      "loss": 0.3094,
+      "step": 13445
+    },
+    {
+      "epoch": 1.6940202160153666,
+      "grad_norm": 0.25370490550994873,
+      "learning_rate": 0.00014231975837835815,
+      "loss": 0.3446,
+      "step": 13450
+    },
+    {
+      "epoch": 1.694649998425544,
+      "grad_norm": 0.2589857876300812,
+      "learning_rate": 0.00014220998186596315,
+      "loss": 0.3258,
+      "step": 13455
+    },
+    {
+      "epoch": 1.6952797808357212,
+      "grad_norm": 0.31022030115127563,
+      "learning_rate": 0.00014210020953698573,
+      "loss": 0.344,
+      "step": 13460
+    },
+    {
+      "epoch": 1.6959095632458987,
+      "grad_norm": 0.3099876046180725,
+      "learning_rate": 0.0001419904414503763,
+      "loss": 0.3425,
+      "step": 13465
+    },
+    {
+      "epoch": 1.6965393456560758,
+      "grad_norm": 0.27715328335762024,
+      "learning_rate": 0.00014188067766508273,
+      "loss": 0.3309,
+      "step": 13470
+    },
+    {
+      "epoch": 1.697169128066253,
+      "grad_norm": 0.2700579762458801,
+      "learning_rate": 0.00014177091824005075,
+      "loss": 0.3191,
+      "step": 13475
+    },
+    {
+      "epoch": 1.6977989104764304,
+      "grad_norm": 0.2773703336715698,
+      "learning_rate": 0.00014166116323422365,
+      "loss": 0.3321,
+      "step": 13480
+    },
+    {
+      "epoch": 1.6984286928866077,
+      "grad_norm": 0.2699192464351654,
+      "learning_rate": 0.00014155141270654232,
+      "loss": 0.3318,
+      "step": 13485
+    },
+    {
+      "epoch": 1.699058475296785,
+      "grad_norm": 0.26127228140830994,
+      "learning_rate": 0.00014144166671594544,
+      "loss": 0.2982,
+      "step": 13490
+    },
+    {
+      "epoch": 1.699688257706962,
+      "grad_norm": 0.37218350172042847,
+      "learning_rate": 0.000141331925321369,
+      "loss": 0.3335,
+      "step": 13495
+    },
+    {
+      "epoch": 1.7003180401171396,
+      "grad_norm": 0.26352524757385254,
+      "learning_rate": 0.0001412221885817466,
+      "loss": 0.3246,
+      "step": 13500
+    },
+    {
+      "epoch": 1.7009478225273167,
+      "grad_norm": 0.27649009227752686,
+      "learning_rate": 0.00014111245655600948,
+      "loss": 0.3117,
+      "step": 13505
+    },
+    {
+      "epoch": 1.7015776049374942,
+      "grad_norm": 0.26316478848457336,
+      "learning_rate": 0.00014100272930308623,
+      "loss": 0.3268,
+      "step": 13510
+    },
+    {
+      "epoch": 1.7022073873476713,
+      "grad_norm": 0.26319512724876404,
+      "learning_rate": 0.0001408930068819028,
+      "loss": 0.3083,
+      "step": 13515
+    },
+    {
+      "epoch": 1.7028371697578488,
+      "grad_norm": 0.26792389154434204,
+      "learning_rate": 0.00014078328935138276,
+      "loss": 0.3317,
+      "step": 13520
+    },
+    {
+      "epoch": 1.703466952168026,
+      "grad_norm": 0.2627207338809967,
+      "learning_rate": 0.0001406735767704469,
+      "loss": 0.3225,
+      "step": 13525
+    },
+    {
+      "epoch": 1.7040967345782032,
+      "grad_norm": 0.30815207958221436,
+      "learning_rate": 0.00014056386919801325,
+      "loss": 0.3201,
+      "step": 13530
+    },
+    {
+      "epoch": 1.7047265169883805,
+      "grad_norm": 0.296520471572876,
+      "learning_rate": 0.00014045416669299747,
+      "loss": 0.3189,
+      "step": 13535
+    },
+    {
+      "epoch": 1.7053562993985578,
+      "grad_norm": 0.2739796042442322,
+      "learning_rate": 0.0001403444693143122,
+      "loss": 0.3023,
+      "step": 13540
+    },
+    {
+      "epoch": 1.7059860818087351,
+      "grad_norm": 0.311927855014801,
+      "learning_rate": 0.00014023477712086743,
+      "loss": 0.3311,
+      "step": 13545
+    },
+    {
+      "epoch": 1.7066158642189122,
+      "grad_norm": 0.2842674255371094,
+      "learning_rate": 0.0001401250901715704,
+      "loss": 0.3376,
+      "step": 13550
+    },
+    {
+      "epoch": 1.7072456466290897,
+      "grad_norm": 0.30459704995155334,
+      "learning_rate": 0.00014001540852532553,
+      "loss": 0.3276,
+      "step": 13555
+    },
+    {
+      "epoch": 1.7078754290392668,
+      "grad_norm": 0.26651817560195923,
+      "learning_rate": 0.00013990573224103442,
+      "loss": 0.3309,
+      "step": 13560
+    },
+    {
+      "epoch": 1.7085052114494443,
+      "grad_norm": 0.32419687509536743,
+      "learning_rate": 0.00013979606137759563,
+      "loss": 0.314,
+      "step": 13565
+    },
+    {
+      "epoch": 1.7091349938596214,
+      "grad_norm": 0.2715966999530792,
+      "learning_rate": 0.000139686395993905,
+      "loss": 0.3293,
+      "step": 13570
+    },
+    {
+      "epoch": 1.709764776269799,
+      "grad_norm": 0.29049497842788696,
+      "learning_rate": 0.0001395767361488552,
+      "loss": 0.3159,
+      "step": 13575
+    },
+    {
+      "epoch": 1.710394558679976,
+      "grad_norm": 0.3235701024532318,
+      "learning_rate": 0.00013946708190133627,
+      "loss": 0.3422,
+      "step": 13580
+    },
+    {
+      "epoch": 1.7110243410901533,
+      "grad_norm": 0.2732395529747009,
+      "learning_rate": 0.00013935743331023492,
+      "loss": 0.317,
+      "step": 13585
+    },
+    {
+      "epoch": 1.7116541235003306,
+      "grad_norm": 0.2833672761917114,
+      "learning_rate": 0.000139247790434435,
+      "loss": 0.3619,
+      "step": 13590
+    },
+    {
+      "epoch": 1.712283905910508,
+      "grad_norm": 0.2510261535644531,
+      "learning_rate": 0.00013913815333281728,
+      "loss": 0.3215,
+      "step": 13595
+    },
+    {
+      "epoch": 1.7129136883206852,
+      "grad_norm": 0.29638463258743286,
+      "learning_rate": 0.00013902852206425925,
+      "loss": 0.3341,
+      "step": 13600
+    },
+    {
+      "epoch": 1.7135434707308623,
+      "grad_norm": 0.26883918046951294,
+      "learning_rate": 0.0001389188966876355,
+      "loss": 0.3198,
+      "step": 13605
+    },
+    {
+      "epoch": 1.7141732531410399,
+      "grad_norm": 0.280301958322525,
+      "learning_rate": 0.00013880927726181737,
+      "loss": 0.3232,
+      "step": 13610
+    },
+    {
+      "epoch": 1.714803035551217,
+      "grad_norm": 0.25223594903945923,
+      "learning_rate": 0.00013869966384567293,
+      "loss": 0.3362,
+      "step": 13615
+    },
+    {
+      "epoch": 1.7154328179613945,
+      "grad_norm": 0.29902294278144836,
+      "learning_rate": 0.00013859005649806717,
+      "loss": 0.3169,
+      "step": 13620
+    },
+    {
+      "epoch": 1.7160626003715715,
+      "grad_norm": 0.3142664134502411,
+      "learning_rate": 0.00013848045527786168,
+      "loss": 0.3149,
+      "step": 13625
+    },
+    {
+      "epoch": 1.716692382781749,
+      "grad_norm": 0.312800794839859,
+      "learning_rate": 0.0001383708602439149,
+      "loss": 0.3327,
+      "step": 13630
+    },
+    {
+      "epoch": 1.7173221651919262,
+      "grad_norm": 0.3177478015422821,
+      "learning_rate": 0.00013826127145508176,
+      "loss": 0.3215,
+      "step": 13635
+    },
+    {
+      "epoch": 1.7179519476021035,
+      "grad_norm": 0.2900395691394806,
+      "learning_rate": 0.00013815168897021398,
+      "loss": 0.3169,
+      "step": 13640
+    },
+    {
+      "epoch": 1.7185817300122808,
+      "grad_norm": 0.2877413332462311,
+      "learning_rate": 0.00013804211284815986,
+      "loss": 0.3247,
+      "step": 13645
+    },
+    {
+      "epoch": 1.719211512422458,
+      "grad_norm": 0.25947847962379456,
+      "learning_rate": 0.00013793254314776432,
+      "loss": 0.3091,
+      "step": 13650
+    },
+    {
+      "epoch": 1.7198412948326354,
+      "grad_norm": 0.270942747592926,
+      "learning_rate": 0.00013782297992786873,
+      "loss": 0.3318,
+      "step": 13655
+    },
+    {
+      "epoch": 1.7204710772428125,
+      "grad_norm": 0.2605541944503784,
+      "learning_rate": 0.00013771342324731106,
+      "loss": 0.3247,
+      "step": 13660
+    },
+    {
+      "epoch": 1.72110085965299,
+      "grad_norm": 0.25236964225769043,
+      "learning_rate": 0.00013760387316492584,
+      "loss": 0.3111,
+      "step": 13665
+    },
+    {
+      "epoch": 1.721730642063167,
+      "grad_norm": 0.2639407217502594,
+      "learning_rate": 0.00013749432973954385,
+      "loss": 0.305,
+      "step": 13670
+    },
+    {
+      "epoch": 1.7223604244733446,
+      "grad_norm": 0.3111459016799927,
+      "learning_rate": 0.0001373847930299924,
+      "loss": 0.3367,
+      "step": 13675
+    },
+    {
+      "epoch": 1.7229902068835217,
+      "grad_norm": 0.31038767099380493,
+      "learning_rate": 0.00013727526309509531,
+      "loss": 0.3223,
+      "step": 13680
+    },
+    {
+      "epoch": 1.723619989293699,
+      "grad_norm": 0.2571181058883667,
+      "learning_rate": 0.00013716573999367259,
+      "loss": 0.3057,
+      "step": 13685
+    },
+    {
+      "epoch": 1.7242497717038763,
+      "grad_norm": 0.24940542876720428,
+      "learning_rate": 0.0001370562237845406,
+      "loss": 0.319,
+      "step": 13690
+    },
+    {
+      "epoch": 1.7248795541140536,
+      "grad_norm": 0.2301412671804428,
+      "learning_rate": 0.00013694671452651216,
+      "loss": 0.3099,
+      "step": 13695
+    },
+    {
+      "epoch": 1.7255093365242309,
+      "grad_norm": 0.27043718099594116,
+      "learning_rate": 0.00013683721227839623,
+      "loss": 0.3345,
+      "step": 13700
+    },
+    {
+      "epoch": 1.7261391189344082,
+      "grad_norm": 0.26595422625541687,
+      "learning_rate": 0.00013672771709899792,
+      "loss": 0.3162,
+      "step": 13705
+    },
+    {
+      "epoch": 1.7267689013445855,
+      "grad_norm": 0.26224714517593384,
+      "learning_rate": 0.0001366182290471187,
+      "loss": 0.322,
+      "step": 13710
+    },
+    {
+      "epoch": 1.7273986837547626,
+      "grad_norm": 0.26390886306762695,
+      "learning_rate": 0.00013650874818155618,
+      "loss": 0.2964,
+      "step": 13715
+    },
+    {
+      "epoch": 1.72802846616494,
+      "grad_norm": 0.3042176365852356,
+      "learning_rate": 0.00013639927456110402,
+      "loss": 0.3128,
+      "step": 13720
+    },
+    {
+      "epoch": 1.7286582485751172,
+      "grad_norm": 0.269771009683609,
+      "learning_rate": 0.00013628980824455212,
+      "loss": 0.2963,
+      "step": 13725
+    },
+    {
+      "epoch": 1.7292880309852947,
+      "grad_norm": 0.3462948203086853,
+      "learning_rate": 0.00013618034929068634,
+      "loss": 0.3445,
+      "step": 13730
+    },
+    {
+      "epoch": 1.7299178133954718,
+      "grad_norm": 0.270379900932312,
+      "learning_rate": 0.0001360708977582887,
+      "loss": 0.3174,
+      "step": 13735
+    },
+    {
+      "epoch": 1.730547595805649,
+      "grad_norm": 0.23746255040168762,
+      "learning_rate": 0.00013596145370613715,
+      "loss": 0.3006,
+      "step": 13740
+    },
+    {
+      "epoch": 1.7311773782158264,
+      "grad_norm": 0.30519574880599976,
+      "learning_rate": 0.00013585201719300562,
+      "loss": 0.3272,
+      "step": 13745
+    },
+    {
+      "epoch": 1.7318071606260037,
+      "grad_norm": 0.3508155941963196,
+      "learning_rate": 0.000135742588277664,
+      "loss": 0.3385,
+      "step": 13750
+    },
+    {
+      "epoch": 1.732436943036181,
+      "grad_norm": 0.2649688720703125,
+      "learning_rate": 0.00013563316701887816,
+      "loss": 0.3191,
+      "step": 13755
+    },
+    {
+      "epoch": 1.7330667254463583,
+      "grad_norm": 0.25044509768486023,
+      "learning_rate": 0.0001355237534754098,
+      "loss": 0.3114,
+      "step": 13760
+    },
+    {
+      "epoch": 1.7336965078565356,
+      "grad_norm": 0.27739325165748596,
+      "learning_rate": 0.00013541434770601653,
+      "loss": 0.3555,
+      "step": 13765
+    },
+    {
+      "epoch": 1.7343262902667127,
+      "grad_norm": 0.27952834963798523,
+      "learning_rate": 0.00013530494976945172,
+      "loss": 0.3287,
+      "step": 13770
+    },
+    {
+      "epoch": 1.7349560726768902,
+      "grad_norm": 0.29794949293136597,
+      "learning_rate": 0.00013519555972446454,
+      "loss": 0.3248,
+      "step": 13775
+    },
+    {
+      "epoch": 1.7355858550870673,
+      "grad_norm": 0.3177776634693146,
+      "learning_rate": 0.00013508617762979992,
+      "loss": 0.3311,
+      "step": 13780
+    },
+    {
+      "epoch": 1.7362156374972448,
+      "grad_norm": 0.29036352038383484,
+      "learning_rate": 0.0001349768035441986,
+      "loss": 0.3021,
+      "step": 13785
+    },
+    {
+      "epoch": 1.736845419907422,
+      "grad_norm": 0.2803820073604584,
+      "learning_rate": 0.00013486743752639694,
+      "loss": 0.3021,
+      "step": 13790
+    },
+    {
+      "epoch": 1.7374752023175992,
+      "grad_norm": 0.25854361057281494,
+      "learning_rate": 0.000134758079635127,
+      "loss": 0.3215,
+      "step": 13795
+    },
+    {
+      "epoch": 1.7381049847277765,
+      "grad_norm": 0.2606901228427887,
+      "learning_rate": 0.0001346487299291165,
+      "loss": 0.3093,
+      "step": 13800
+    },
+    {
+      "epoch": 1.7387347671379538,
+      "grad_norm": 0.25198522210121155,
+      "learning_rate": 0.00013453938846708864,
+      "loss": 0.2954,
+      "step": 13805
+    },
+    {
+      "epoch": 1.7393645495481311,
+      "grad_norm": 0.27399036288261414,
+      "learning_rate": 0.00013443005530776233,
+      "loss": 0.3212,
+      "step": 13810
+    },
+    {
+      "epoch": 1.7399943319583084,
+      "grad_norm": 0.2777753174304962,
+      "learning_rate": 0.000134320730509852,
+      "loss": 0.32,
+      "step": 13815
+    },
+    {
+      "epoch": 1.7406241143684857,
+      "grad_norm": 0.28130999207496643,
+      "learning_rate": 0.0001342114141320675,
+      "loss": 0.305,
+      "step": 13820
+    },
+    {
+      "epoch": 1.7412538967786628,
+      "grad_norm": 0.28102371096611023,
+      "learning_rate": 0.00013410210623311428,
+      "loss": 0.3066,
+      "step": 13825
+    },
+    {
+      "epoch": 1.7418836791888404,
+      "grad_norm": 0.21866032481193542,
+      "learning_rate": 0.00013399280687169312,
+      "loss": 0.3181,
+      "step": 13830
+    },
+    {
+      "epoch": 1.7425134615990174,
+      "grad_norm": 0.27159667015075684,
+      "learning_rate": 0.00013388351610650045,
+      "loss": 0.2983,
+      "step": 13835
+    },
+    {
+      "epoch": 1.743143244009195,
+      "grad_norm": 0.26473724842071533,
+      "learning_rate": 0.00013377423399622764,
+      "loss": 0.3041,
+      "step": 13840
+    },
+    {
+      "epoch": 1.743773026419372,
+      "grad_norm": 0.30044063925743103,
+      "learning_rate": 0.00013366496059956184,
+      "loss": 0.3391,
+      "step": 13845
+    },
+    {
+      "epoch": 1.7444028088295493,
+      "grad_norm": 0.3015748858451843,
+      "learning_rate": 0.00013355569597518532,
+      "loss": 0.3033,
+      "step": 13850
+    },
+    {
+      "epoch": 1.7450325912397266,
+      "grad_norm": 0.27009138464927673,
+      "learning_rate": 0.00013344644018177572,
+      "loss": 0.2973,
+      "step": 13855
+    },
+    {
+      "epoch": 1.745662373649904,
+      "grad_norm": 0.28925400972366333,
+      "learning_rate": 0.00013333719327800585,
+      "loss": 0.3137,
+      "step": 13860
+    },
+    {
+      "epoch": 1.7462921560600813,
+      "grad_norm": 0.27679139375686646,
+      "learning_rate": 0.00013322795532254379,
+      "loss": 0.3119,
+      "step": 13865
+    },
+    {
+      "epoch": 1.7469219384702586,
+      "grad_norm": 0.283965140581131,
+      "learning_rate": 0.0001331187263740529,
+      "loss": 0.3151,
+      "step": 13870
+    },
+    {
+      "epoch": 1.7475517208804359,
+      "grad_norm": 0.24927465617656708,
+      "learning_rate": 0.0001330095064911915,
+      "loss": 0.2968,
+      "step": 13875
+    },
+    {
+      "epoch": 1.748181503290613,
+      "grad_norm": 0.2976732850074768,
+      "learning_rate": 0.0001329002957326132,
+      "loss": 0.3257,
+      "step": 13880
+    },
+    {
+      "epoch": 1.7488112857007905,
+      "grad_norm": 0.27860409021377563,
+      "learning_rate": 0.00013279109415696672,
+      "loss": 0.2988,
+      "step": 13885
+    },
+    {
+      "epoch": 1.7494410681109676,
+      "grad_norm": 0.28782716393470764,
+      "learning_rate": 0.0001326819018228958,
+      "loss": 0.3098,
+      "step": 13890
+    },
+    {
+      "epoch": 1.750070850521145,
+      "grad_norm": 0.24729984998703003,
+      "learning_rate": 0.0001325727187890391,
+      "loss": 0.3123,
+      "step": 13895
+    },
+    {
+      "epoch": 1.7507006329313222,
+      "grad_norm": 0.23218853771686554,
+      "learning_rate": 0.00013246354511403058,
+      "loss": 0.3025,
+      "step": 13900
+    },
+    {
+      "epoch": 1.7513304153414995,
+      "grad_norm": 0.2634672522544861,
+      "learning_rate": 0.00013235438085649893,
+      "loss": 0.3123,
+      "step": 13905
+    },
+    {
+      "epoch": 1.7519601977516768,
+      "grad_norm": 0.3087509572505951,
+      "learning_rate": 0.00013224522607506776,
+      "loss": 0.3515,
+      "step": 13910
+    },
+    {
+      "epoch": 1.752589980161854,
+      "grad_norm": 0.28160160779953003,
+      "learning_rate": 0.00013213608082835576,
+      "loss": 0.3141,
+      "step": 13915
+    },
+    {
+      "epoch": 1.7532197625720314,
+      "grad_norm": 0.2643168866634369,
+      "learning_rate": 0.0001320269451749764,
+      "loss": 0.297,
+      "step": 13920
+    },
+    {
+      "epoch": 1.7538495449822087,
+      "grad_norm": 0.34547582268714905,
+      "learning_rate": 0.00013191781917353803,
+      "loss": 0.3194,
+      "step": 13925
+    },
+    {
+      "epoch": 1.754479327392386,
+      "grad_norm": 0.29079994559288025,
+      "learning_rate": 0.00013180870288264385,
+      "loss": 0.3334,
+      "step": 13930
+    },
+    {
+      "epoch": 1.755109109802563,
+      "grad_norm": 0.2323244959115982,
+      "learning_rate": 0.00013169959636089167,
+      "loss": 0.3106,
+      "step": 13935
+    },
+    {
+      "epoch": 1.7557388922127406,
+      "grad_norm": 0.29080161452293396,
+      "learning_rate": 0.00013159049966687437,
+      "loss": 0.2978,
+      "step": 13940
+    },
+    {
+      "epoch": 1.7563686746229177,
+      "grad_norm": 0.2688988149166107,
+      "learning_rate": 0.00013148141285917924,
+      "loss": 0.3184,
+      "step": 13945
+    },
+    {
+      "epoch": 1.7569984570330952,
+      "grad_norm": 0.25353583693504333,
+      "learning_rate": 0.0001313723359963884,
+      "loss": 0.2956,
+      "step": 13950
+    },
+    {
+      "epoch": 1.7576282394432723,
+      "grad_norm": 0.32606688141822815,
+      "learning_rate": 0.0001312632691370786,
+      "loss": 0.3136,
+      "step": 13955
+    },
+    {
+      "epoch": 1.7582580218534496,
+      "grad_norm": 0.24126961827278137,
+      "learning_rate": 0.0001311542123398213,
+      "loss": 0.304,
+      "step": 13960
+    },
+    {
+      "epoch": 1.758887804263627,
+      "grad_norm": 0.2840232253074646,
+      "learning_rate": 0.0001310451656631824,
+      "loss": 0.3126,
+      "step": 13965
+    },
+    {
+      "epoch": 1.7595175866738042,
+      "grad_norm": 0.30879929661750793,
+      "learning_rate": 0.0001309361291657226,
+      "loss": 0.3115,
+      "step": 13970
+    },
+    {
+      "epoch": 1.7601473690839815,
+      "grad_norm": 0.29478558897972107,
+      "learning_rate": 0.0001308271029059969,
+      "loss": 0.3035,
+      "step": 13975
+    },
+    {
+      "epoch": 1.7607771514941588,
+      "grad_norm": 0.29496970772743225,
+      "learning_rate": 0.00013071808694255484,
+      "loss": 0.3417,
+      "step": 13980
+    },
+    {
+      "epoch": 1.7614069339043361,
+      "grad_norm": 0.27189967036247253,
+      "learning_rate": 0.00013060908133394054,
+      "loss": 0.3146,
+      "step": 13985
+    },
+    {
+      "epoch": 1.7620367163145132,
+      "grad_norm": 0.2737963795661926,
+      "learning_rate": 0.00013050008613869256,
+      "loss": 0.3223,
+      "step": 13990
+    },
+    {
+      "epoch": 1.7626664987246907,
+      "grad_norm": 0.2881993055343628,
+      "learning_rate": 0.00013039110141534367,
+      "loss": 0.3039,
+      "step": 13995
+    },
+    {
+      "epoch": 1.7632962811348678,
+      "grad_norm": 0.29045918583869934,
+      "learning_rate": 0.00013028212722242127,
+      "loss": 0.3193,
+      "step": 14000
+    },
+    {
+      "epoch": 1.7632962811348678,
+      "eval_loss": 0.3040441870689392,
+      "eval_runtime": 6.1585,
+      "eval_samples_per_second": 162.378,
+      "eval_steps_per_second": 10.23,
+      "step": 14000
+    },
+    {
+      "epoch": 1.7639260635450453,
+      "grad_norm": 0.24037687480449677,
+      "learning_rate": 0.00013017316361844692,
+      "loss": 0.2918,
+      "step": 14005
+    },
+    {
+      "epoch": 1.7645558459552224,
+      "grad_norm": 0.25562503933906555,
+      "learning_rate": 0.0001300642106619367,
+      "loss": 0.2967,
+      "step": 14010
+    },
+    {
+      "epoch": 1.7651856283653997,
+      "grad_norm": 0.3410753905773163,
+      "learning_rate": 0.00012995526841140068,
+      "loss": 0.3158,
+      "step": 14015
+    },
+    {
+      "epoch": 1.765815410775577,
+      "grad_norm": 0.2569274306297302,
+      "learning_rate": 0.00012984633692534337,
+      "loss": 0.306,
+      "step": 14020
+    },
+    {
+      "epoch": 1.7664451931857543,
+      "grad_norm": 0.26620200276374817,
+      "learning_rate": 0.00012973741626226348,
+      "loss": 0.3122,
+      "step": 14025
+    },
+    {
+      "epoch": 1.7670749755959316,
+      "grad_norm": 0.2842133045196533,
+      "learning_rate": 0.00012962850648065393,
+      "loss": 0.3253,
+      "step": 14030
+    },
+    {
+      "epoch": 1.767704758006109,
+      "grad_norm": 0.27718397974967957,
+      "learning_rate": 0.00012951960763900173,
+      "loss": 0.3187,
+      "step": 14035
+    },
+    {
+      "epoch": 1.7683345404162862,
+      "grad_norm": 0.27699559926986694,
+      "learning_rate": 0.00012941071979578805,
+      "loss": 0.33,
+      "step": 14040
+    },
+    {
+      "epoch": 1.7689643228264633,
+      "grad_norm": 0.21499434113502502,
+      "learning_rate": 0.00012930184300948819,
+      "loss": 0.2765,
+      "step": 14045
+    },
+    {
+      "epoch": 1.7695941052366408,
+      "grad_norm": 0.29474014043807983,
+      "learning_rate": 0.00012919297733857138,
+      "loss": 0.32,
+      "step": 14050
+    },
+    {
+      "epoch": 1.770223887646818,
+      "grad_norm": 0.3570992052555084,
+      "learning_rate": 0.00012908412284150104,
+      "loss": 0.3088,
+      "step": 14055
+    },
+    {
+      "epoch": 1.7708536700569955,
+      "grad_norm": 0.2408706545829773,
+      "learning_rate": 0.00012897527957673446,
+      "loss": 0.2991,
+      "step": 14060
+    },
+    {
+      "epoch": 1.7714834524671725,
+      "grad_norm": 0.23086212575435638,
+      "learning_rate": 0.00012886644760272306,
+      "loss": 0.2959,
+      "step": 14065
+    },
+    {
+      "epoch": 1.7721132348773498,
+      "grad_norm": 0.25117409229278564,
+      "learning_rate": 0.00012875762697791199,
+      "loss": 0.2933,
+      "step": 14070
+    },
+    {
+      "epoch": 1.7727430172875271,
+      "grad_norm": 0.28731420636177063,
+      "learning_rate": 0.0001286488177607405,
+      "loss": 0.3234,
+      "step": 14075
+    },
+    {
+      "epoch": 1.7733727996977044,
+      "grad_norm": 0.23875364661216736,
+      "learning_rate": 0.0001285400200096416,
+      "loss": 0.2952,
+      "step": 14080
+    },
+    {
+      "epoch": 1.7740025821078818,
+      "grad_norm": 0.2722354829311371,
+      "learning_rate": 0.0001284312337830421,
+      "loss": 0.2997,
+      "step": 14085
+    },
+    {
+      "epoch": 1.774632364518059,
+      "grad_norm": 0.27776023745536804,
+      "learning_rate": 0.00012832245913936278,
+      "loss": 0.3256,
+      "step": 14090
+    },
+    {
+      "epoch": 1.7752621469282364,
+      "grad_norm": 0.26422828435897827,
+      "learning_rate": 0.00012821369613701808,
+      "loss": 0.2983,
+      "step": 14095
+    },
+    {
+      "epoch": 1.7758919293384134,
+      "grad_norm": 0.23418962955474854,
+      "learning_rate": 0.00012810494483441614,
+      "loss": 0.3024,
+      "step": 14100
+    },
+    {
+      "epoch": 1.776521711748591,
+      "grad_norm": 0.300912082195282,
+      "learning_rate": 0.000127996205289959,
+      "loss": 0.3001,
+      "step": 14105
+    },
+    {
+      "epoch": 1.777151494158768,
+      "grad_norm": 0.2872162461280823,
+      "learning_rate": 0.00012788747756204222,
+      "loss": 0.3074,
+      "step": 14110
+    },
+    {
+      "epoch": 1.7777812765689456,
+      "grad_norm": 0.2784421145915985,
+      "learning_rate": 0.00012777876170905515,
+      "loss": 0.2978,
+      "step": 14115
+    },
+    {
+      "epoch": 1.7784110589791227,
+      "grad_norm": 0.28062257170677185,
+      "learning_rate": 0.00012767005778938062,
+      "loss": 0.2993,
+      "step": 14120
+    },
+    {
+      "epoch": 1.7790408413893,
+      "grad_norm": 0.3496231734752655,
+      "learning_rate": 0.0001275613658613951,
+      "loss": 0.3147,
+      "step": 14125
+    },
+    {
+      "epoch": 1.7796706237994773,
+      "grad_norm": 0.2595261037349701,
+      "learning_rate": 0.00012745268598346864,
+      "loss": 0.2943,
+      "step": 14130
+    },
+    {
+      "epoch": 1.7803004062096546,
+      "grad_norm": 0.2795499563217163,
+      "learning_rate": 0.00012734401821396486,
+      "loss": 0.3123,
+      "step": 14135
+    },
+    {
+      "epoch": 1.7809301886198319,
+      "grad_norm": 0.2615763247013092,
+      "learning_rate": 0.0001272353626112408,
+      "loss": 0.3059,
+      "step": 14140
+    },
+    {
+      "epoch": 1.7815599710300092,
+      "grad_norm": 0.2783886790275574,
+      "learning_rate": 0.00012712671923364706,
+      "loss": 0.3134,
+      "step": 14145
+    },
+    {
+      "epoch": 1.7821897534401865,
+      "grad_norm": 0.2884584367275238,
+      "learning_rate": 0.0001270180881395276,
+      "loss": 0.3151,
+      "step": 14150
+    },
+    {
+      "epoch": 1.7828195358503636,
+      "grad_norm": 0.2677745521068573,
+      "learning_rate": 0.0001269094693872197,
+      "loss": 0.3146,
+      "step": 14155
+    },
+    {
+      "epoch": 1.783449318260541,
+      "grad_norm": 0.25956082344055176,
+      "learning_rate": 0.0001268008630350542,
+      "loss": 0.3118,
+      "step": 14160
+    },
+    {
+      "epoch": 1.7840791006707182,
+      "grad_norm": 0.2646723985671997,
+      "learning_rate": 0.0001266922691413552,
+      "loss": 0.2861,
+      "step": 14165
+    },
+    {
+      "epoch": 1.7847088830808957,
+      "grad_norm": 0.29946067929267883,
+      "learning_rate": 0.00012658368776444004,
+      "loss": 0.3349,
+      "step": 14170
+    },
+    {
+      "epoch": 1.7853386654910728,
+      "grad_norm": 0.24171167612075806,
+      "learning_rate": 0.00012647511896261943,
+      "loss": 0.2805,
+      "step": 14175
+    },
+    {
+      "epoch": 1.78596844790125,
+      "grad_norm": 0.26428696513175964,
+      "learning_rate": 0.0001263665627941973,
+      "loss": 0.3231,
+      "step": 14180
+    },
+    {
+      "epoch": 1.7865982303114274,
+      "grad_norm": 0.2787708044052124,
+      "learning_rate": 0.0001262580193174709,
+      "loss": 0.2961,
+      "step": 14185
+    },
+    {
+      "epoch": 1.7872280127216047,
+      "grad_norm": 0.2826111614704132,
+      "learning_rate": 0.00012614948859073036,
+      "loss": 0.3343,
+      "step": 14190
+    },
+    {
+      "epoch": 1.787857795131782,
+      "grad_norm": 0.278361052274704,
+      "learning_rate": 0.00012604097067225927,
+      "loss": 0.2919,
+      "step": 14195
+    },
+    {
+      "epoch": 1.7884875775419593,
+      "grad_norm": 0.24778404831886292,
+      "learning_rate": 0.00012593246562033419,
+      "loss": 0.316,
+      "step": 14200
+    },
+    {
+      "epoch": 1.7891173599521366,
+      "grad_norm": 0.28171002864837646,
+      "learning_rate": 0.00012582397349322484,
+      "loss": 0.3076,
+      "step": 14205
+    },
+    {
+      "epoch": 1.7897471423623137,
+      "grad_norm": 0.26361143589019775,
+      "learning_rate": 0.00012571549434919392,
+      "loss": 0.2953,
+      "step": 14210
+    },
+    {
+      "epoch": 1.7903769247724912,
+      "grad_norm": 0.27602389454841614,
+      "learning_rate": 0.0001256070282464973,
+      "loss": 0.3266,
+      "step": 14215
+    },
+    {
+      "epoch": 1.7910067071826683,
+      "grad_norm": 0.2887786328792572,
+      "learning_rate": 0.00012549857524338378,
+      "loss": 0.3166,
+      "step": 14220
+    },
+    {
+      "epoch": 1.7916364895928458,
+      "grad_norm": 0.272359162569046,
+      "learning_rate": 0.00012539013539809493,
+      "loss": 0.3053,
+      "step": 14225
+    },
+    {
+      "epoch": 1.792266272003023,
+      "grad_norm": 0.2615000903606415,
+      "learning_rate": 0.00012528170876886555,
+      "loss": 0.2974,
+      "step": 14230
+    },
+    {
+      "epoch": 1.7928960544132002,
+      "grad_norm": 0.2882770597934723,
+      "learning_rate": 0.00012517329541392316,
+      "loss": 0.301,
+      "step": 14235
+    },
+    {
+      "epoch": 1.7935258368233775,
+      "grad_norm": 0.29980406165122986,
+      "learning_rate": 0.00012506489539148823,
+      "loss": 0.3009,
+      "step": 14240
+    },
+    {
+      "epoch": 1.7941556192335548,
+      "grad_norm": 0.2714889943599701,
+      "learning_rate": 0.0001249565087597741,
+      "loss": 0.2897,
+      "step": 14245
+    },
+    {
+      "epoch": 1.7947854016437321,
+      "grad_norm": 0.3578423261642456,
+      "learning_rate": 0.00012484813557698678,
+      "loss": 0.3021,
+      "step": 14250
+    },
+    {
+      "epoch": 1.7954151840539094,
+      "grad_norm": 0.29889971017837524,
+      "learning_rate": 0.00012473977590132524,
+      "loss": 0.3039,
+      "step": 14255
+    },
+    {
+      "epoch": 1.7960449664640867,
+      "grad_norm": 0.27244943380355835,
+      "learning_rate": 0.000124631429790981,
+      "loss": 0.3068,
+      "step": 14260
+    },
+    {
+      "epoch": 1.7966747488742638,
+      "grad_norm": 0.2793833613395691,
+      "learning_rate": 0.00012452309730413843,
+      "loss": 0.3081,
+      "step": 14265
+    },
+    {
+      "epoch": 1.7973045312844413,
+      "grad_norm": 0.27198326587677,
+      "learning_rate": 0.00012441477849897461,
+      "loss": 0.2957,
+      "step": 14270
+    },
+    {
+      "epoch": 1.7979343136946184,
+      "grad_norm": 0.24795940518379211,
+      "learning_rate": 0.0001243064734336591,
+      "loss": 0.3094,
+      "step": 14275
+    },
+    {
+      "epoch": 1.798564096104796,
+      "grad_norm": 0.29008451104164124,
+      "learning_rate": 0.0001241981821663543,
+      "loss": 0.3306,
+      "step": 14280
+    },
+    {
+      "epoch": 1.799193878514973,
+      "grad_norm": 0.24478363990783691,
+      "learning_rate": 0.00012408990475521508,
+      "loss": 0.291,
+      "step": 14285
+    },
+    {
+      "epoch": 1.7998236609251503,
+      "grad_norm": 0.2566664218902588,
+      "learning_rate": 0.00012398164125838881,
+      "loss": 0.3087,
+      "step": 14290
+    },
+    {
+      "epoch": 1.8004534433353276,
+      "grad_norm": 0.24992555379867554,
+      "learning_rate": 0.00012387339173401552,
+      "loss": 0.318,
+      "step": 14295
+    },
+    {
+      "epoch": 1.801083225745505,
+      "grad_norm": 0.244164377450943,
+      "learning_rate": 0.00012376515624022767,
+      "loss": 0.3096,
+      "step": 14300
+    },
+    {
+      "epoch": 1.8017130081556822,
+      "grad_norm": 0.2495235651731491,
+      "learning_rate": 0.00012365693483515016,
+      "loss": 0.283,
+      "step": 14305
+    },
+    {
+      "epoch": 1.8023427905658593,
+      "grad_norm": 0.2685554027557373,
+      "learning_rate": 0.00012354872757690038,
+      "loss": 0.3359,
+      "step": 14310
+    },
+    {
+      "epoch": 1.8029725729760369,
+      "grad_norm": 0.23964886367321014,
+      "learning_rate": 0.0001234405345235881,
+      "loss": 0.3074,
+      "step": 14315
+    },
+    {
+      "epoch": 1.803602355386214,
+      "grad_norm": 0.24736544489860535,
+      "learning_rate": 0.00012333235573331556,
+      "loss": 0.2891,
+      "step": 14320
+    },
+    {
+      "epoch": 1.8042321377963915,
+      "grad_norm": 0.2994007170200348,
+      "learning_rate": 0.00012322419126417706,
+      "loss": 0.3109,
+      "step": 14325
+    },
+    {
+      "epoch": 1.8048619202065685,
+      "grad_norm": 0.26516586542129517,
+      "learning_rate": 0.0001231160411742595,
+      "loss": 0.2974,
+      "step": 14330
+    },
+    {
+      "epoch": 1.805491702616746,
+      "grad_norm": 0.27139636874198914,
+      "learning_rate": 0.0001230079055216419,
+      "loss": 0.3023,
+      "step": 14335
+    },
+    {
+      "epoch": 1.8061214850269232,
+      "grad_norm": 0.26109209656715393,
+      "learning_rate": 0.00012289978436439558,
+      "loss": 0.3059,
+      "step": 14340
+    },
+    {
+      "epoch": 1.8067512674371005,
+      "grad_norm": 0.29744458198547363,
+      "learning_rate": 0.0001227916777605841,
+      "loss": 0.3088,
+      "step": 14345
+    },
+    {
+      "epoch": 1.8073810498472778,
+      "grad_norm": 0.27332085371017456,
+      "learning_rate": 0.0001226835857682632,
+      "loss": 0.2888,
+      "step": 14350
+    },
+    {
+      "epoch": 1.808010832257455,
+      "grad_norm": 0.2586978077888489,
+      "learning_rate": 0.00012257550844548074,
+      "loss": 0.328,
+      "step": 14355
+    },
+    {
+      "epoch": 1.8086406146676324,
+      "grad_norm": 0.29042935371398926,
+      "learning_rate": 0.00012246744585027667,
+      "loss": 0.3113,
+      "step": 14360
+    },
+    {
+      "epoch": 1.8092703970778095,
+      "grad_norm": 0.271710067987442,
+      "learning_rate": 0.000122359398040683,
+      "loss": 0.2888,
+      "step": 14365
+    },
+    {
+      "epoch": 1.809900179487987,
+      "grad_norm": 0.2969205379486084,
+      "learning_rate": 0.00012225136507472406,
+      "loss": 0.312,
+      "step": 14370
+    },
+    {
+      "epoch": 1.810529961898164,
+      "grad_norm": 0.301145076751709,
+      "learning_rate": 0.00012214334701041586,
+      "loss": 0.2952,
+      "step": 14375
+    },
+    {
+      "epoch": 1.8111597443083416,
+      "grad_norm": 0.250630259513855,
+      "learning_rate": 0.00012203534390576666,
+      "loss": 0.3073,
+      "step": 14380
+    },
+    {
+      "epoch": 1.8117895267185187,
+      "grad_norm": 0.24282781779766083,
+      "learning_rate": 0.00012192735581877654,
+      "loss": 0.2863,
+      "step": 14385
+    },
+    {
+      "epoch": 1.8124193091286962,
+      "grad_norm": 0.2824462652206421,
+      "learning_rate": 0.00012181938280743769,
+      "loss": 0.2999,
+      "step": 14390
+    },
+    {
+      "epoch": 1.8130490915388733,
+      "grad_norm": 0.2740934491157532,
+      "learning_rate": 0.00012171142492973388,
+      "loss": 0.3131,
+      "step": 14395
+    },
+    {
+      "epoch": 1.8136788739490506,
+      "grad_norm": 0.23533669114112854,
+      "learning_rate": 0.00012160348224364109,
+      "loss": 0.2846,
+      "step": 14400
+    },
+    {
+      "epoch": 1.814308656359228,
+      "grad_norm": 0.26320409774780273,
+      "learning_rate": 0.00012149555480712697,
+      "loss": 0.2954,
+      "step": 14405
+    },
+    {
+      "epoch": 1.8149384387694052,
+      "grad_norm": 0.2816338837146759,
+      "learning_rate": 0.00012138764267815105,
+      "loss": 0.2811,
+      "step": 14410
+    },
+    {
+      "epoch": 1.8155682211795825,
+      "grad_norm": 0.23801551759243011,
+      "learning_rate": 0.00012127974591466455,
+      "loss": 0.2846,
+      "step": 14415
+    },
+    {
+      "epoch": 1.8161980035897596,
+      "grad_norm": 0.3131721317768097,
+      "learning_rate": 0.00012117186457461056,
+      "loss": 0.2969,
+      "step": 14420
+    },
+    {
+      "epoch": 1.816827785999937,
+      "grad_norm": 0.2892078757286072,
+      "learning_rate": 0.00012106399871592385,
+      "loss": 0.3,
+      "step": 14425
+    },
+    {
+      "epoch": 1.8174575684101142,
+      "grad_norm": 0.253273606300354,
+      "learning_rate": 0.00012095614839653074,
+      "loss": 0.3005,
+      "step": 14430
+    },
+    {
+      "epoch": 1.8180873508202917,
+      "grad_norm": 0.2675528824329376,
+      "learning_rate": 0.00012084831367434937,
+      "loss": 0.2947,
+      "step": 14435
+    },
+    {
+      "epoch": 1.8187171332304688,
+      "grad_norm": 0.2665347158908844,
+      "learning_rate": 0.00012074049460728945,
+      "loss": 0.3012,
+      "step": 14440
+    },
+    {
+      "epoch": 1.8193469156406463,
+      "grad_norm": 0.2987824082374573,
+      "learning_rate": 0.00012063269125325228,
+      "loss": 0.2986,
+      "step": 14445
+    },
+    {
+      "epoch": 1.8199766980508234,
+      "grad_norm": 0.2429313212633133,
+      "learning_rate": 0.00012052490367013076,
+      "loss": 0.3035,
+      "step": 14450
+    },
+    {
+      "epoch": 1.8206064804610007,
+      "grad_norm": 0.28424081206321716,
+      "learning_rate": 0.00012041713191580925,
+      "loss": 0.2948,
+      "step": 14455
+    },
+    {
+      "epoch": 1.821236262871178,
+      "grad_norm": 0.25087571144104004,
+      "learning_rate": 0.00012030937604816365,
+      "loss": 0.2949,
+      "step": 14460
+    },
+    {
+      "epoch": 1.8218660452813553,
+      "grad_norm": 0.23633217811584473,
+      "learning_rate": 0.00012020163612506127,
+      "loss": 0.2669,
+      "step": 14465
+    },
+    {
+      "epoch": 1.8224958276915326,
+      "grad_norm": 0.26396888494491577,
+      "learning_rate": 0.000120093912204361,
+      "loss": 0.2912,
+      "step": 14470
+    },
+    {
+      "epoch": 1.8231256101017097,
+      "grad_norm": 0.2898525297641754,
+      "learning_rate": 0.00011998620434391299,
+      "loss": 0.319,
+      "step": 14475
+    },
+    {
+      "epoch": 1.8237553925118872,
+      "grad_norm": 0.25507113337516785,
+      "learning_rate": 0.00011987851260155881,
+      "loss": 0.3028,
+      "step": 14480
+    },
+    {
+      "epoch": 1.8243851749220643,
+      "grad_norm": 0.2405284345149994,
+      "learning_rate": 0.00011977083703513145,
+      "loss": 0.2879,
+      "step": 14485
+    },
+    {
+      "epoch": 1.8250149573322418,
+      "grad_norm": 0.27114009857177734,
+      "learning_rate": 0.00011966317770245507,
+      "loss": 0.3094,
+      "step": 14490
+    },
+    {
+      "epoch": 1.825644739742419,
+      "grad_norm": 0.2708043158054352,
+      "learning_rate": 0.0001195555346613453,
+      "loss": 0.3062,
+      "step": 14495
+    },
+    {
+      "epoch": 1.8262745221525962,
+      "grad_norm": 0.2507513463497162,
+      "learning_rate": 0.00011944790796960878,
+      "loss": 0.2832,
+      "step": 14500
+    },
+    {
+      "epoch": 1.8269043045627735,
+      "grad_norm": 0.2864154577255249,
+      "learning_rate": 0.0001193402976850436,
+      "loss": 0.3067,
+      "step": 14505
+    },
+    {
+      "epoch": 1.8275340869729508,
+      "grad_norm": 0.26530271768569946,
+      "learning_rate": 0.00011923270386543886,
+      "loss": 0.2816,
+      "step": 14510
+    },
+    {
+      "epoch": 1.8281638693831281,
+      "grad_norm": 0.24444885551929474,
+      "learning_rate": 0.00011912512656857498,
+      "loss": 0.2993,
+      "step": 14515
+    },
+    {
+      "epoch": 1.8287936517933054,
+      "grad_norm": 0.2591851055622101,
+      "learning_rate": 0.00011901756585222334,
+      "loss": 0.2926,
+      "step": 14520
+    },
+    {
+      "epoch": 1.8294234342034827,
+      "grad_norm": 0.2942061424255371,
+      "learning_rate": 0.0001189100217741466,
+      "loss": 0.3032,
+      "step": 14525
+    },
+    {
+      "epoch": 1.8300532166136598,
+      "grad_norm": 0.28199318051338196,
+      "learning_rate": 0.00011880249439209836,
+      "loss": 0.291,
+      "step": 14530
+    },
+    {
+      "epoch": 1.8306829990238374,
+      "grad_norm": 0.2743484377861023,
+      "learning_rate": 0.00011869498376382324,
+      "loss": 0.3101,
+      "step": 14535
+    },
+    {
+      "epoch": 1.8313127814340144,
+      "grad_norm": 0.24012960493564606,
+      "learning_rate": 0.00011858748994705689,
+      "loss": 0.3,
+      "step": 14540
+    },
+    {
+      "epoch": 1.831942563844192,
+      "grad_norm": 0.2856425344944,
+      "learning_rate": 0.00011848001299952598,
+      "loss": 0.3042,
+      "step": 14545
+    },
+    {
+      "epoch": 1.832572346254369,
+      "grad_norm": 0.2720118463039398,
+      "learning_rate": 0.00011837255297894808,
+      "loss": 0.298,
+      "step": 14550
+    },
+    {
+      "epoch": 1.8332021286645463,
+      "grad_norm": 0.26973578333854675,
+      "learning_rate": 0.0001182651099430317,
+      "loss": 0.2734,
+      "step": 14555
+    },
+    {
+      "epoch": 1.8338319110747237,
+      "grad_norm": 0.35720425844192505,
+      "learning_rate": 0.00011815768394947616,
+      "loss": 0.3174,
+      "step": 14560
+    },
+    {
+      "epoch": 1.834461693484901,
+      "grad_norm": 0.2649666666984558,
+      "learning_rate": 0.00011805027505597178,
+      "loss": 0.3009,
+      "step": 14565
+    },
+    {
+      "epoch": 1.8350914758950783,
+      "grad_norm": 0.2809504270553589,
+      "learning_rate": 0.00011794288332019939,
+      "loss": 0.3075,
+      "step": 14570
+    },
+    {
+      "epoch": 1.8357212583052556,
+      "grad_norm": 0.247705340385437,
+      "learning_rate": 0.00011783550879983097,
+      "loss": 0.2929,
+      "step": 14575
+    },
+    {
+      "epoch": 1.8363510407154329,
+      "grad_norm": 0.3010486364364624,
+      "learning_rate": 0.00011772815155252901,
+      "loss": 0.2923,
+      "step": 14580
+    },
+    {
+      "epoch": 1.83698082312561,
+      "grad_norm": 0.29634296894073486,
+      "learning_rate": 0.00011762081163594686,
+      "loss": 0.2956,
+      "step": 14585
+    },
+    {
+      "epoch": 1.8376106055357875,
+      "grad_norm": 0.3235035538673401,
+      "learning_rate": 0.00011751348910772844,
+      "loss": 0.299,
+      "step": 14590
+    },
+    {
+      "epoch": 1.8382403879459646,
+      "grad_norm": 0.27069565653800964,
+      "learning_rate": 0.00011740618402550849,
+      "loss": 0.2885,
+      "step": 14595
+    },
+    {
+      "epoch": 1.838870170356142,
+      "grad_norm": 0.26986175775527954,
+      "learning_rate": 0.00011729889644691227,
+      "loss": 0.2974,
+      "step": 14600
+    },
+    {
+      "epoch": 1.8394999527663192,
+      "grad_norm": 0.24633704125881195,
+      "learning_rate": 0.00011719162642955559,
+      "loss": 0.3011,
+      "step": 14605
+    },
+    {
+      "epoch": 1.8401297351764965,
+      "grad_norm": 0.2659735381603241,
+      "learning_rate": 0.00011708437403104491,
+      "loss": 0.2802,
+      "step": 14610
+    },
+    {
+      "epoch": 1.8407595175866738,
+      "grad_norm": 0.2634638547897339,
+      "learning_rate": 0.00011697713930897728,
+      "loss": 0.2834,
+      "step": 14615
+    },
+    {
+      "epoch": 1.841389299996851,
+      "grad_norm": 0.2891436517238617,
+      "learning_rate": 0.00011686992232094012,
+      "loss": 0.2892,
+      "step": 14620
+    },
+    {
+      "epoch": 1.8420190824070284,
+      "grad_norm": 0.26533305644989014,
+      "learning_rate": 0.0001167627231245115,
+      "loss": 0.2954,
+      "step": 14625
+    },
+    {
+      "epoch": 1.8426488648172057,
+      "grad_norm": 0.26114416122436523,
+      "learning_rate": 0.00011665554177725977,
+      "loss": 0.2936,
+      "step": 14630
+    },
+    {
+      "epoch": 1.843278647227383,
+      "grad_norm": 0.24782754480838776,
+      "learning_rate": 0.00011654837833674379,
+      "loss": 0.283,
+      "step": 14635
+    },
+    {
+      "epoch": 1.84390842963756,
+      "grad_norm": 0.2653804123401642,
+      "learning_rate": 0.00011644123286051274,
+      "loss": 0.2911,
+      "step": 14640
+    },
+    {
+      "epoch": 1.8445382120477376,
+      "grad_norm": 0.2524818181991577,
+      "learning_rate": 0.00011633410540610621,
+      "loss": 0.29,
+      "step": 14645
+    },
+    {
+      "epoch": 1.8451679944579147,
+      "grad_norm": 0.2844378352165222,
+      "learning_rate": 0.00011622699603105404,
+      "loss": 0.298,
+      "step": 14650
+    },
+    {
+      "epoch": 1.8457977768680922,
+      "grad_norm": 0.2608543038368225,
+      "learning_rate": 0.0001161199047928765,
+      "loss": 0.2807,
+      "step": 14655
+    },
+    {
+      "epoch": 1.8464275592782693,
+      "grad_norm": 0.2596459984779358,
+      "learning_rate": 0.000116012831749084,
+      "loss": 0.29,
+      "step": 14660
+    },
+    {
+      "epoch": 1.8470573416884466,
+      "grad_norm": 0.2654721140861511,
+      "learning_rate": 0.00011590577695717717,
+      "loss": 0.2878,
+      "step": 14665
+    },
+    {
+      "epoch": 1.847687124098624,
+      "grad_norm": 0.283388614654541,
+      "learning_rate": 0.00011579874047464696,
+      "loss": 0.2751,
+      "step": 14670
+    },
+    {
+      "epoch": 1.8483169065088012,
+      "grad_norm": 0.24917341768741608,
+      "learning_rate": 0.00011569172235897433,
+      "loss": 0.3,
+      "step": 14675
+    },
+    {
+      "epoch": 1.8489466889189785,
+      "grad_norm": 0.2464076280593872,
+      "learning_rate": 0.00011558472266763049,
+      "loss": 0.2848,
+      "step": 14680
+    },
+    {
+      "epoch": 1.8495764713291558,
+      "grad_norm": 0.2884039282798767,
+      "learning_rate": 0.00011547774145807665,
+      "loss": 0.2698,
+      "step": 14685
+    },
+    {
+      "epoch": 1.8502062537393331,
+      "grad_norm": 0.2762083411216736,
+      "learning_rate": 0.00011537077878776425,
+      "loss": 0.3151,
+      "step": 14690
+    },
+    {
+      "epoch": 1.8508360361495102,
+      "grad_norm": 0.22906774282455444,
+      "learning_rate": 0.00011526383471413463,
+      "loss": 0.2669,
+      "step": 14695
+    },
+    {
+      "epoch": 1.8514658185596877,
+      "grad_norm": 0.28603047132492065,
+      "learning_rate": 0.00011515690929461928,
+      "loss": 0.2922,
+      "step": 14700
+    },
+    {
+      "epoch": 1.8520956009698648,
+      "grad_norm": 0.26245948672294617,
+      "learning_rate": 0.00011505000258663954,
+      "loss": 0.3095,
+      "step": 14705
+    },
+    {
+      "epoch": 1.8527253833800423,
+      "grad_norm": 0.2754320800304413,
+      "learning_rate": 0.00011494311464760673,
+      "loss": 0.2843,
+      "step": 14710
+    },
+    {
+      "epoch": 1.8533551657902194,
+      "grad_norm": 0.24283255636692047,
+      "learning_rate": 0.00011483624553492212,
+      "loss": 0.3039,
+      "step": 14715
+    },
+    {
+      "epoch": 1.8539849482003967,
+      "grad_norm": 0.299950510263443,
+      "learning_rate": 0.00011472939530597691,
+      "loss": 0.3108,
+      "step": 14720
+    },
+    {
+      "epoch": 1.854614730610574,
+      "grad_norm": 0.23872928321361542,
+      "learning_rate": 0.00011462256401815205,
+      "loss": 0.3221,
+      "step": 14725
+    },
+    {
+      "epoch": 1.8552445130207513,
+      "grad_norm": 0.32674193382263184,
+      "learning_rate": 0.00011451575172881845,
+      "loss": 0.3066,
+      "step": 14730
+    },
+    {
+      "epoch": 1.8558742954309286,
+      "grad_norm": 0.2620803415775299,
+      "learning_rate": 0.00011440895849533675,
+      "loss": 0.2855,
+      "step": 14735
+    },
+    {
+      "epoch": 1.856504077841106,
+      "grad_norm": 0.2653051018714905,
+      "learning_rate": 0.0001143021843750573,
+      "loss": 0.2827,
+      "step": 14740
+    },
+    {
+      "epoch": 1.8571338602512832,
+      "grad_norm": 0.29697105288505554,
+      "learning_rate": 0.00011419542942532023,
+      "loss": 0.2848,
+      "step": 14745
+    },
+    {
+      "epoch": 1.8577636426614603,
+      "grad_norm": 0.26711151003837585,
+      "learning_rate": 0.00011408869370345545,
+      "loss": 0.301,
+      "step": 14750
+    },
+    {
+      "epoch": 1.8583934250716379,
+      "grad_norm": 0.26371288299560547,
+      "learning_rate": 0.0001139819772667824,
+      "loss": 0.2994,
+      "step": 14755
+    },
+    {
+      "epoch": 1.859023207481815,
+      "grad_norm": 0.34920862317085266,
+      "learning_rate": 0.00011387528017261035,
+      "loss": 0.2968,
+      "step": 14760
+    },
+    {
+      "epoch": 1.8596529898919925,
+      "grad_norm": 0.2951182723045349,
+      "learning_rate": 0.000113768602478238,
+      "loss": 0.2935,
+      "step": 14765
+    },
+    {
+      "epoch": 1.8602827723021695,
+      "grad_norm": 0.22442401945590973,
+      "learning_rate": 0.00011366194424095381,
+      "loss": 0.2952,
+      "step": 14770
+    },
+    {
+      "epoch": 1.8609125547123468,
+      "grad_norm": 0.26102137565612793,
+      "learning_rate": 0.00011355530551803553,
+      "loss": 0.2823,
+      "step": 14775
+    },
+    {
+      "epoch": 1.8615423371225241,
+      "grad_norm": 0.27118000388145447,
+      "learning_rate": 0.0001134486863667507,
+      "loss": 0.286,
+      "step": 14780
+    },
+    {
+      "epoch": 1.8621721195327015,
+      "grad_norm": 0.2869999408721924,
+      "learning_rate": 0.00011334208684435617,
+      "loss": 0.2734,
+      "step": 14785
+    },
+    {
+      "epoch": 1.8628019019428788,
+      "grad_norm": 0.32944396138191223,
+      "learning_rate": 0.0001132355070080984,
+      "loss": 0.3038,
+      "step": 14790
+    },
+    {
+      "epoch": 1.863431684353056,
+      "grad_norm": 0.28535759449005127,
+      "learning_rate": 0.00011312894691521312,
+      "loss": 0.3213,
+      "step": 14795
+    },
+    {
+      "epoch": 1.8640614667632334,
+      "grad_norm": 0.23751592636108398,
+      "learning_rate": 0.00011302240662292561,
+      "loss": 0.2972,
+      "step": 14800
+    },
+    {
+      "epoch": 1.8646912491734104,
+      "grad_norm": 0.2352185994386673,
+      "learning_rate": 0.00011291588618845043,
+      "loss": 0.2772,
+      "step": 14805
+    },
+    {
+      "epoch": 1.865321031583588,
+      "grad_norm": 0.24066108465194702,
+      "learning_rate": 0.00011280938566899142,
+      "loss": 0.3053,
+      "step": 14810
+    },
+    {
+      "epoch": 1.865950813993765,
+      "grad_norm": 0.33842501044273376,
+      "learning_rate": 0.0001127029051217418,
+      "loss": 0.2992,
+      "step": 14815
+    },
+    {
+      "epoch": 1.8665805964039426,
+      "grad_norm": 0.24873322248458862,
+      "learning_rate": 0.00011259644460388412,
+      "loss": 0.2887,
+      "step": 14820
+    },
+    {
+      "epoch": 1.8672103788141197,
+      "grad_norm": 0.27127575874328613,
+      "learning_rate": 0.00011249000417259005,
+      "loss": 0.2619,
+      "step": 14825
+    },
+    {
+      "epoch": 1.867840161224297,
+      "grad_norm": 0.28289374709129333,
+      "learning_rate": 0.00011238358388502059,
+      "loss": 0.2815,
+      "step": 14830
+    },
+    {
+      "epoch": 1.8684699436344743,
+      "grad_norm": 0.29764994978904724,
+      "learning_rate": 0.00011227718379832583,
+      "loss": 0.3006,
+      "step": 14835
+    },
+    {
+      "epoch": 1.8690997260446516,
+      "grad_norm": 0.2869538366794586,
+      "learning_rate": 0.00011217080396964507,
+      "loss": 0.2707,
+      "step": 14840
+    },
+    {
+      "epoch": 1.8697295084548289,
+      "grad_norm": 0.2732262909412384,
+      "learning_rate": 0.00011206444445610663,
+      "loss": 0.2768,
+      "step": 14845
+    },
+    {
+      "epoch": 1.8703592908650062,
+      "grad_norm": 0.3032742738723755,
+      "learning_rate": 0.0001119581053148281,
+      "loss": 0.2715,
+      "step": 14850
+    },
+    {
+      "epoch": 1.8709890732751835,
+      "grad_norm": 0.26171359419822693,
+      "learning_rate": 0.00011185178660291594,
+      "loss": 0.2701,
+      "step": 14855
+    },
+    {
+      "epoch": 1.8716188556853606,
+      "grad_norm": 0.30940353870391846,
+      "learning_rate": 0.00011174548837746581,
+      "loss": 0.2843,
+      "step": 14860
+    },
+    {
+      "epoch": 1.872248638095538,
+      "grad_norm": 0.2774769067764282,
+      "learning_rate": 0.00011163921069556224,
+      "loss": 0.2951,
+      "step": 14865
+    },
+    {
+      "epoch": 1.8728784205057152,
+      "grad_norm": 0.29232633113861084,
+      "learning_rate": 0.00011153295361427876,
+      "loss": 0.2938,
+      "step": 14870
+    },
+    {
+      "epoch": 1.8735082029158927,
+      "grad_norm": 0.28283149003982544,
+      "learning_rate": 0.00011142671719067793,
+      "loss": 0.2875,
+      "step": 14875
+    },
+    {
+      "epoch": 1.8741379853260698,
+      "grad_norm": 0.24245183169841766,
+      "learning_rate": 0.00011132050148181103,
+      "loss": 0.2829,
+      "step": 14880
+    },
+    {
+      "epoch": 1.874767767736247,
+      "grad_norm": 0.27495938539505005,
+      "learning_rate": 0.00011121430654471837,
+      "loss": 0.2923,
+      "step": 14885
+    },
+    {
+      "epoch": 1.8753975501464244,
+      "grad_norm": 0.3106895089149475,
+      "learning_rate": 0.00011110813243642906,
+      "loss": 0.2855,
+      "step": 14890
+    },
+    {
+      "epoch": 1.8760273325566017,
+      "grad_norm": 0.263810396194458,
+      "learning_rate": 0.00011100197921396102,
+      "loss": 0.271,
+      "step": 14895
+    },
+    {
+      "epoch": 1.876657114966779,
+      "grad_norm": 0.23044048249721527,
+      "learning_rate": 0.00011089584693432091,
+      "loss": 0.2608,
+      "step": 14900
+    },
+    {
+      "epoch": 1.8772868973769563,
+      "grad_norm": 0.29268765449523926,
+      "learning_rate": 0.00011078973565450427,
+      "loss": 0.2835,
+      "step": 14905
+    },
+    {
+      "epoch": 1.8779166797871336,
+      "grad_norm": 0.2691350281238556,
+      "learning_rate": 0.00011068364543149527,
+      "loss": 0.291,
+      "step": 14910
+    },
+    {
+      "epoch": 1.8785464621973107,
+      "grad_norm": 0.26748213171958923,
+      "learning_rate": 0.00011057757632226672,
+      "loss": 0.2994,
+      "step": 14915
+    },
+    {
+      "epoch": 1.8791762446074882,
+      "grad_norm": 0.2624029815196991,
+      "learning_rate": 0.00011047152838378018,
+      "loss": 0.2832,
+      "step": 14920
+    },
+    {
+      "epoch": 1.8798060270176653,
+      "grad_norm": 0.2670036554336548,
+      "learning_rate": 0.00011036550167298583,
+      "loss": 0.284,
+      "step": 14925
+    },
+    {
+      "epoch": 1.8804358094278428,
+      "grad_norm": 0.2848396599292755,
+      "learning_rate": 0.0001102594962468224,
+      "loss": 0.2831,
+      "step": 14930
+    },
+    {
+      "epoch": 1.88106559183802,
+      "grad_norm": 0.2502748668193817,
+      "learning_rate": 0.0001101535121622173,
+      "loss": 0.3038,
+      "step": 14935
+    },
+    {
+      "epoch": 1.8816953742481972,
+      "grad_norm": 0.2998834252357483,
+      "learning_rate": 0.0001100475494760863,
+      "loss": 0.2847,
+      "step": 14940
+    },
+    {
+      "epoch": 1.8823251566583745,
+      "grad_norm": 0.229685977101326,
+      "learning_rate": 0.00010994160824533398,
+      "loss": 0.261,
+      "step": 14945
+    },
+    {
+      "epoch": 1.8829549390685518,
+      "grad_norm": 0.26833808422088623,
+      "learning_rate": 0.00010983568852685294,
+      "loss": 0.2923,
+      "step": 14950
+    },
+    {
+      "epoch": 1.8835847214787291,
+      "grad_norm": 0.2380465716123581,
+      "learning_rate": 0.00010972979037752465,
+      "loss": 0.2664,
+      "step": 14955
+    },
+    {
+      "epoch": 1.8842145038889064,
+      "grad_norm": 0.2505188286304474,
+      "learning_rate": 0.00010962391385421876,
+      "loss": 0.2914,
+      "step": 14960
+    },
+    {
+      "epoch": 1.8848442862990837,
+      "grad_norm": 0.33335885405540466,
+      "learning_rate": 0.00010951805901379346,
+      "loss": 0.3092,
+      "step": 14965
+    },
+    {
+      "epoch": 1.8854740687092608,
+      "grad_norm": 0.22425580024719238,
+      "learning_rate": 0.0001094122259130951,
+      "loss": 0.2583,
+      "step": 14970
+    },
+    {
+      "epoch": 1.8861038511194383,
+      "grad_norm": 0.25008514523506165,
+      "learning_rate": 0.00010930641460895863,
+      "loss": 0.2936,
+      "step": 14975
+    },
+    {
+      "epoch": 1.8867336335296154,
+      "grad_norm": 0.2543163299560547,
+      "learning_rate": 0.00010920062515820707,
+      "loss": 0.2855,
+      "step": 14980
+    },
+    {
+      "epoch": 1.887363415939793,
+      "grad_norm": 0.25144490599632263,
+      "learning_rate": 0.00010909485761765172,
+      "loss": 0.2788,
+      "step": 14985
+    },
+    {
+      "epoch": 1.88799319834997,
+      "grad_norm": 0.23470145463943481,
+      "learning_rate": 0.00010898911204409218,
+      "loss": 0.2709,
+      "step": 14990
+    },
+    {
+      "epoch": 1.8886229807601473,
+      "grad_norm": 0.27916932106018066,
+      "learning_rate": 0.00010888338849431629,
+      "loss": 0.279,
+      "step": 14995
+    },
+    {
+      "epoch": 1.8892527631703246,
+      "grad_norm": 0.24980424344539642,
+      "learning_rate": 0.00010877768702509996,
+      "loss": 0.2982,
+      "step": 15000
+    },
+    {
+      "epoch": 1.8892527631703246,
+      "eval_loss": 0.3032541871070862,
+      "eval_runtime": 6.1659,
+      "eval_samples_per_second": 162.182,
+      "eval_steps_per_second": 10.217,
+      "step": 15000
+    },
+    {
+      "epoch": 1.889882545580502,
+      "grad_norm": 0.24535268545150757,
+      "learning_rate": 0.00010867200769320732,
+      "loss": 0.2667,
+      "step": 15005
+    },
+    {
+      "epoch": 1.8905123279906793,
+      "grad_norm": 0.2690826654434204,
+      "learning_rate": 0.0001085663505553906,
+      "loss": 0.2703,
+      "step": 15010
+    },
+    {
+      "epoch": 1.8911421104008566,
+      "grad_norm": 0.2511346936225891,
+      "learning_rate": 0.00010846071566839008,
+      "loss": 0.3011,
+      "step": 15015
+    },
+    {
+      "epoch": 1.8917718928110339,
+      "grad_norm": 0.28077587485313416,
+      "learning_rate": 0.00010835510308893407,
+      "loss": 0.285,
+      "step": 15020
+    },
+    {
+      "epoch": 1.892401675221211,
+      "grad_norm": 0.309238463640213,
+      "learning_rate": 0.000108249512873739,
+      "loss": 0.2894,
+      "step": 15025
+    },
+    {
+      "epoch": 1.8930314576313885,
+      "grad_norm": 0.26940178871154785,
+      "learning_rate": 0.00010814394507950917,
+      "loss": 0.2864,
+      "step": 15030
+    },
+    {
+      "epoch": 1.8936612400415656,
+      "grad_norm": 0.27850431203842163,
+      "learning_rate": 0.00010803839976293694,
+      "loss": 0.2716,
+      "step": 15035
+    },
+    {
+      "epoch": 1.894291022451743,
+      "grad_norm": 0.24114792048931122,
+      "learning_rate": 0.00010793287698070256,
+      "loss": 0.2695,
+      "step": 15040
+    },
+    {
+      "epoch": 1.8949208048619202,
+      "grad_norm": 0.3137163817882538,
+      "learning_rate": 0.0001078273767894741,
+      "loss": 0.3063,
+      "step": 15045
+    },
+    {
+      "epoch": 1.8955505872720975,
+      "grad_norm": 0.27090078592300415,
+      "learning_rate": 0.00010772189924590773,
+      "loss": 0.2643,
+      "step": 15050
+    },
+    {
+      "epoch": 1.8961803696822748,
+      "grad_norm": 0.27956193685531616,
+      "learning_rate": 0.00010761644440664714,
+      "loss": 0.271,
+      "step": 15055
+    },
+    {
+      "epoch": 1.896810152092452,
+      "grad_norm": 0.24823328852653503,
+      "learning_rate": 0.00010751101232832401,
+      "loss": 0.2849,
+      "step": 15060
+    },
+    {
+      "epoch": 1.8974399345026294,
+      "grad_norm": 0.2675158977508545,
+      "learning_rate": 0.00010740560306755787,
+      "loss": 0.2744,
+      "step": 15065
+    },
+    {
+      "epoch": 1.8980697169128065,
+      "grad_norm": 0.2589218318462372,
+      "learning_rate": 0.0001073002166809558,
+      "loss": 0.2834,
+      "step": 15070
+    },
+    {
+      "epoch": 1.898699499322984,
+      "grad_norm": 0.277705579996109,
+      "learning_rate": 0.00010719485322511273,
+      "loss": 0.2826,
+      "step": 15075
+    },
+    {
+      "epoch": 1.899329281733161,
+      "grad_norm": 0.23539955914020538,
+      "learning_rate": 0.0001070895127566113,
+      "loss": 0.2589,
+      "step": 15080
+    },
+    {
+      "epoch": 1.8999590641433386,
+      "grad_norm": 0.3010064661502838,
+      "learning_rate": 0.00010698419533202172,
+      "loss": 0.2804,
+      "step": 15085
+    },
+    {
+      "epoch": 1.9005888465535157,
+      "grad_norm": 0.25453826785087585,
+      "learning_rate": 0.00010687890100790175,
+      "loss": 0.2863,
+      "step": 15090
+    },
+    {
+      "epoch": 1.9012186289636932,
+      "grad_norm": 0.2774878144264221,
+      "learning_rate": 0.00010677362984079699,
+      "loss": 0.2933,
+      "step": 15095
+    },
+    {
+      "epoch": 1.9018484113738703,
+      "grad_norm": 0.26002323627471924,
+      "learning_rate": 0.00010666838188724038,
+      "loss": 0.2891,
+      "step": 15100
+    },
+    {
+      "epoch": 1.9024781937840476,
+      "grad_norm": 0.25788870453834534,
+      "learning_rate": 0.00010656315720375246,
+      "loss": 0.2934,
+      "step": 15105
+    },
+    {
+      "epoch": 1.903107976194225,
+      "grad_norm": 0.24301236867904663,
+      "learning_rate": 0.00010645795584684138,
+      "loss": 0.2848,
+      "step": 15110
+    },
+    {
+      "epoch": 1.9037377586044022,
+      "grad_norm": 0.309514582157135,
+      "learning_rate": 0.00010635277787300256,
+      "loss": 0.2846,
+      "step": 15115
+    },
+    {
+      "epoch": 1.9043675410145795,
+      "grad_norm": 0.274870902299881,
+      "learning_rate": 0.00010624762333871913,
+      "loss": 0.2956,
+      "step": 15120
+    },
+    {
+      "epoch": 1.9049973234247566,
+      "grad_norm": 0.24861137568950653,
+      "learning_rate": 0.00010614249230046129,
+      "loss": 0.2777,
+      "step": 15125
+    },
+    {
+      "epoch": 1.905627105834934,
+      "grad_norm": 0.26125532388687134,
+      "learning_rate": 0.00010603738481468693,
+      "loss": 0.2794,
+      "step": 15130
+    },
+    {
+      "epoch": 1.9062568882451112,
+      "grad_norm": 0.24094760417938232,
+      "learning_rate": 0.0001059323009378411,
+      "loss": 0.2633,
+      "step": 15135
+    },
+    {
+      "epoch": 1.9068866706552887,
+      "grad_norm": 0.3418034315109253,
+      "learning_rate": 0.0001058272407263563,
+      "loss": 0.3045,
+      "step": 15140
+    },
+    {
+      "epoch": 1.9075164530654658,
+      "grad_norm": 0.2657215893268585,
+      "learning_rate": 0.00010572220423665222,
+      "loss": 0.3085,
+      "step": 15145
+    },
+    {
+      "epoch": 1.9081462354756433,
+      "grad_norm": 0.23728597164154053,
+      "learning_rate": 0.00010561719152513591,
+      "loss": 0.2788,
+      "step": 15150
+    },
+    {
+      "epoch": 1.9087760178858204,
+      "grad_norm": 0.2741139829158783,
+      "learning_rate": 0.0001055122026482016,
+      "loss": 0.2855,
+      "step": 15155
+    },
+    {
+      "epoch": 1.9094058002959977,
+      "grad_norm": 0.2415517419576645,
+      "learning_rate": 0.00010540723766223064,
+      "loss": 0.2799,
+      "step": 15160
+    },
+    {
+      "epoch": 1.910035582706175,
+      "grad_norm": 0.2724277675151825,
+      "learning_rate": 0.00010530229662359162,
+      "loss": 0.2821,
+      "step": 15165
+    },
+    {
+      "epoch": 1.9106653651163523,
+      "grad_norm": 0.28418639302253723,
+      "learning_rate": 0.00010519737958864036,
+      "loss": 0.2899,
+      "step": 15170
+    },
+    {
+      "epoch": 1.9112951475265296,
+      "grad_norm": 0.26423749327659607,
+      "learning_rate": 0.00010509248661371962,
+      "loss": 0.3033,
+      "step": 15175
+    },
+    {
+      "epoch": 1.9119249299367067,
+      "grad_norm": 0.2523916959762573,
+      "learning_rate": 0.00010498761775515941,
+      "loss": 0.2763,
+      "step": 15180
+    },
+    {
+      "epoch": 1.9125547123468842,
+      "grad_norm": 0.29665645956993103,
+      "learning_rate": 0.00010488277306927663,
+      "loss": 0.2918,
+      "step": 15185
+    },
+    {
+      "epoch": 1.9131844947570613,
+      "grad_norm": 0.2941978871822357,
+      "learning_rate": 0.00010477795261237537,
+      "loss": 0.2753,
+      "step": 15190
+    },
+    {
+      "epoch": 1.9138142771672388,
+      "grad_norm": 0.2701078951358795,
+      "learning_rate": 0.00010467315644074646,
+      "loss": 0.2925,
+      "step": 15195
+    },
+    {
+      "epoch": 1.914444059577416,
+      "grad_norm": 0.2497081160545349,
+      "learning_rate": 0.00010456838461066793,
+      "loss": 0.2669,
+      "step": 15200
+    },
+    {
+      "epoch": 1.9150738419875935,
+      "grad_norm": 0.2448865920305252,
+      "learning_rate": 0.00010446363717840462,
+      "loss": 0.2766,
+      "step": 15205
+    },
+    {
+      "epoch": 1.9157036243977705,
+      "grad_norm": 0.26188936829566956,
+      "learning_rate": 0.00010435891420020833,
+      "loss": 0.2935,
+      "step": 15210
+    },
+    {
+      "epoch": 1.9163334068079478,
+      "grad_norm": 0.3044489622116089,
+      "learning_rate": 0.00010425421573231767,
+      "loss": 0.2791,
+      "step": 15215
+    },
+    {
+      "epoch": 1.9169631892181251,
+      "grad_norm": 0.30361208319664,
+      "learning_rate": 0.00010414954183095813,
+      "loss": 0.277,
+      "step": 15220
+    },
+    {
+      "epoch": 1.9175929716283024,
+      "grad_norm": 0.31100359559059143,
+      "learning_rate": 0.00010404489255234191,
+      "loss": 0.2687,
+      "step": 15225
+    },
+    {
+      "epoch": 1.9182227540384797,
+      "grad_norm": 0.26500749588012695,
+      "learning_rate": 0.00010394026795266814,
+      "loss": 0.2804,
+      "step": 15230
+    },
+    {
+      "epoch": 1.9188525364486568,
+      "grad_norm": 0.33220374584198,
+      "learning_rate": 0.00010383566808812257,
+      "loss": 0.284,
+      "step": 15235
+    },
+    {
+      "epoch": 1.9194823188588344,
+      "grad_norm": 0.23146981000900269,
+      "learning_rate": 0.00010373109301487777,
+      "loss": 0.2949,
+      "step": 15240
+    },
+    {
+      "epoch": 1.9201121012690114,
+      "grad_norm": 0.24833330512046814,
+      "learning_rate": 0.00010362654278909292,
+      "loss": 0.2685,
+      "step": 15245
+    },
+    {
+      "epoch": 1.920741883679189,
+      "grad_norm": 0.22905099391937256,
+      "learning_rate": 0.00010352201746691381,
+      "loss": 0.248,
+      "step": 15250
+    },
+    {
+      "epoch": 1.921371666089366,
+      "grad_norm": 0.2544589936733246,
+      "learning_rate": 0.00010341751710447308,
+      "loss": 0.2763,
+      "step": 15255
+    },
+    {
+      "epoch": 1.9220014484995434,
+      "grad_norm": 0.24207763373851776,
+      "learning_rate": 0.0001033130417578897,
+      "loss": 0.2691,
+      "step": 15260
+    },
+    {
+      "epoch": 1.9226312309097207,
+      "grad_norm": 0.3025490939617157,
+      "learning_rate": 0.0001032085914832693,
+      "loss": 0.2902,
+      "step": 15265
+    },
+    {
+      "epoch": 1.923261013319898,
+      "grad_norm": 0.2563372552394867,
+      "learning_rate": 0.00010310416633670413,
+      "loss": 0.2937,
+      "step": 15270
+    },
+    {
+      "epoch": 1.9238907957300753,
+      "grad_norm": 0.22143816947937012,
+      "learning_rate": 0.00010299976637427285,
+      "loss": 0.2615,
+      "step": 15275
+    },
+    {
+      "epoch": 1.9245205781402526,
+      "grad_norm": 0.26383697986602783,
+      "learning_rate": 0.00010289539165204058,
+      "loss": 0.2834,
+      "step": 15280
+    },
+    {
+      "epoch": 1.9251503605504299,
+      "grad_norm": 0.2607567310333252,
+      "learning_rate": 0.00010279104222605903,
+      "loss": 0.2875,
+      "step": 15285
+    },
+    {
+      "epoch": 1.925780142960607,
+      "grad_norm": 0.23255427181720734,
+      "learning_rate": 0.0001026867181523662,
+      "loss": 0.2645,
+      "step": 15290
+    },
+    {
+      "epoch": 1.9264099253707845,
+      "grad_norm": 0.2203371226787567,
+      "learning_rate": 0.00010258241948698641,
+      "loss": 0.276,
+      "step": 15295
+    },
+    {
+      "epoch": 1.9270397077809616,
+      "grad_norm": 0.2557859718799591,
+      "learning_rate": 0.00010247814628593052,
+      "loss": 0.2877,
+      "step": 15300
+    },
+    {
+      "epoch": 1.927669490191139,
+      "grad_norm": 0.2551586925983429,
+      "learning_rate": 0.00010237389860519557,
+      "loss": 0.2678,
+      "step": 15305
+    },
+    {
+      "epoch": 1.9282992726013162,
+      "grad_norm": 0.2592737376689911,
+      "learning_rate": 0.00010226967650076495,
+      "loss": 0.2645,
+      "step": 15310
+    },
+    {
+      "epoch": 1.9289290550114935,
+      "grad_norm": 0.25076064467430115,
+      "learning_rate": 0.00010216548002860836,
+      "loss": 0.2595,
+      "step": 15315
+    },
+    {
+      "epoch": 1.9295588374216708,
+      "grad_norm": 0.28892189264297485,
+      "learning_rate": 0.0001020613092446816,
+      "loss": 0.2658,
+      "step": 15320
+    },
+    {
+      "epoch": 1.930188619831848,
+      "grad_norm": 0.28119730949401855,
+      "learning_rate": 0.00010195716420492692,
+      "loss": 0.2783,
+      "step": 15325
+    },
+    {
+      "epoch": 1.9308184022420254,
+      "grad_norm": 0.23143291473388672,
+      "learning_rate": 0.00010185304496527239,
+      "loss": 0.2745,
+      "step": 15330
+    },
+    {
+      "epoch": 1.9314481846522027,
+      "grad_norm": 0.23947221040725708,
+      "learning_rate": 0.00010174895158163252,
+      "loss": 0.2642,
+      "step": 15335
+    },
+    {
+      "epoch": 1.93207796706238,
+      "grad_norm": 0.27924421429634094,
+      "learning_rate": 0.00010164488410990779,
+      "loss": 0.2895,
+      "step": 15340
+    },
+    {
+      "epoch": 1.932707749472557,
+      "grad_norm": 0.2736763656139374,
+      "learning_rate": 0.00010154084260598488,
+      "loss": 0.2798,
+      "step": 15345
+    },
+    {
+      "epoch": 1.9333375318827346,
+      "grad_norm": 0.26288047432899475,
+      "learning_rate": 0.00010143682712573639,
+      "loss": 0.2799,
+      "step": 15350
+    },
+    {
+      "epoch": 1.9339673142929117,
+      "grad_norm": 0.2662082016468048,
+      "learning_rate": 0.00010133283772502105,
+      "loss": 0.2708,
+      "step": 15355
+    },
+    {
+      "epoch": 1.9345970967030892,
+      "grad_norm": 0.2595316767692566,
+      "learning_rate": 0.00010122887445968358,
+      "loss": 0.2631,
+      "step": 15360
+    },
+    {
+      "epoch": 1.9352268791132663,
+      "grad_norm": 0.22839054465293884,
+      "learning_rate": 0.00010112493738555453,
+      "loss": 0.2533,
+      "step": 15365
+    },
+    {
+      "epoch": 1.9358566615234436,
+      "grad_norm": 0.25195086002349854,
+      "learning_rate": 0.0001010210265584505,
+      "loss": 0.26,
+      "step": 15370
+    },
+    {
+      "epoch": 1.936486443933621,
+      "grad_norm": 0.2431613951921463,
+      "learning_rate": 0.00010091714203417404,
+      "loss": 0.2802,
+      "step": 15375
+    },
+    {
+      "epoch": 1.9371162263437982,
+      "grad_norm": 0.24503393471240997,
+      "learning_rate": 0.00010081328386851342,
+      "loss": 0.2968,
+      "step": 15380
+    },
+    {
+      "epoch": 1.9377460087539755,
+      "grad_norm": 0.26283174753189087,
+      "learning_rate": 0.00010070945211724298,
+      "loss": 0.2831,
+      "step": 15385
+    },
+    {
+      "epoch": 1.9383757911641528,
+      "grad_norm": 0.23644685745239258,
+      "learning_rate": 0.00010060564683612264,
+      "loss": 0.2843,
+      "step": 15390
+    },
+    {
+      "epoch": 1.9390055735743301,
+      "grad_norm": 0.271457314491272,
+      "learning_rate": 0.00010050186808089828,
+      "loss": 0.2736,
+      "step": 15395
+    },
+    {
+      "epoch": 1.9396353559845072,
+      "grad_norm": 0.2437523454427719,
+      "learning_rate": 0.00010039811590730137,
+      "loss": 0.2839,
+      "step": 15400
+    },
+    {
+      "epoch": 1.9402651383946847,
+      "grad_norm": 0.25611042976379395,
+      "learning_rate": 0.00010029439037104925,
+      "loss": 0.2671,
+      "step": 15405
+    },
+    {
+      "epoch": 1.9408949208048618,
+      "grad_norm": 0.2646775245666504,
+      "learning_rate": 0.00010019069152784486,
+      "loss": 0.3072,
+      "step": 15410
+    },
+    {
+      "epoch": 1.9415247032150393,
+      "grad_norm": 0.26959145069122314,
+      "learning_rate": 0.00010008701943337695,
+      "loss": 0.2655,
+      "step": 15415
+    },
+    {
+      "epoch": 1.9421544856252164,
+      "grad_norm": 0.28409838676452637,
+      "learning_rate": 9.998337414331971e-05,
+      "loss": 0.2643,
+      "step": 15420
+    },
+    {
+      "epoch": 1.9427842680353937,
+      "grad_norm": 0.288766086101532,
+      "learning_rate": 9.987975571333303e-05,
+      "loss": 0.2849,
+      "step": 15425
+    },
+    {
+      "epoch": 1.943414050445571,
+      "grad_norm": 0.28650057315826416,
+      "learning_rate": 9.977616419906247e-05,
+      "loss": 0.2672,
+      "step": 15430
+    },
+    {
+      "epoch": 1.9440438328557483,
+      "grad_norm": 0.28229546546936035,
+      "learning_rate": 9.967259965613893e-05,
+      "loss": 0.2649,
+      "step": 15435
+    },
+    {
+      "epoch": 1.9446736152659256,
+      "grad_norm": 0.21892526745796204,
+      "learning_rate": 9.956906214017894e-05,
+      "loss": 0.2668,
+      "step": 15440
+    },
+    {
+      "epoch": 1.945303397676103,
+      "grad_norm": 0.27021822333335876,
+      "learning_rate": 9.946555170678458e-05,
+      "loss": 0.2725,
+      "step": 15445
+    },
+    {
+      "epoch": 1.9459331800862802,
+      "grad_norm": 0.2574271857738495,
+      "learning_rate": 9.936206841154328e-05,
+      "loss": 0.2643,
+      "step": 15450
+    },
+    {
+      "epoch": 1.9465629624964573,
+      "grad_norm": 0.2907993495464325,
+      "learning_rate": 9.925861231002792e-05,
+      "loss": 0.3103,
+      "step": 15455
+    },
+    {
+      "epoch": 1.9471927449066349,
+      "grad_norm": 0.225221186876297,
+      "learning_rate": 9.915518345779681e-05,
+      "loss": 0.2804,
+      "step": 15460
+    },
+    {
+      "epoch": 1.947822527316812,
+      "grad_norm": 0.2557651400566101,
+      "learning_rate": 9.905178191039365e-05,
+      "loss": 0.2735,
+      "step": 15465
+    },
+    {
+      "epoch": 1.9484523097269895,
+      "grad_norm": 0.26498880982398987,
+      "learning_rate": 9.894840772334733e-05,
+      "loss": 0.2664,
+      "step": 15470
+    },
+    {
+      "epoch": 1.9490820921371665,
+      "grad_norm": 0.2424790859222412,
+      "learning_rate": 9.884506095217222e-05,
+      "loss": 0.2693,
+      "step": 15475
+    },
+    {
+      "epoch": 1.9497118745473438,
+      "grad_norm": 0.25802767276763916,
+      "learning_rate": 9.87417416523679e-05,
+      "loss": 0.2831,
+      "step": 15480
+    },
+    {
+      "epoch": 1.9503416569575212,
+      "grad_norm": 0.2601839005947113,
+      "learning_rate": 9.863844987941912e-05,
+      "loss": 0.2629,
+      "step": 15485
+    },
+    {
+      "epoch": 1.9509714393676985,
+      "grad_norm": 0.26015961170196533,
+      "learning_rate": 9.853518568879602e-05,
+      "loss": 0.2634,
+      "step": 15490
+    },
+    {
+      "epoch": 1.9516012217778758,
+      "grad_norm": 0.2370160073041916,
+      "learning_rate": 9.843194913595374e-05,
+      "loss": 0.2557,
+      "step": 15495
+    },
+    {
+      "epoch": 1.952231004188053,
+      "grad_norm": 0.2519363462924957,
+      "learning_rate": 9.832874027633281e-05,
+      "loss": 0.255,
+      "step": 15500
+    },
+    {
+      "epoch": 1.9528607865982304,
+      "grad_norm": 0.3419806659221649,
+      "learning_rate": 9.822555916535858e-05,
+      "loss": 0.2744,
+      "step": 15505
+    },
+    {
+      "epoch": 1.9534905690084075,
+      "grad_norm": 0.24397574365139008,
+      "learning_rate": 9.812240585844176e-05,
+      "loss": 0.2619,
+      "step": 15510
+    },
+    {
+      "epoch": 1.954120351418585,
+      "grad_norm": 0.2432924211025238,
+      "learning_rate": 9.801928041097795e-05,
+      "loss": 0.2581,
+      "step": 15515
+    },
+    {
+      "epoch": 1.954750133828762,
+      "grad_norm": 0.27478650212287903,
+      "learning_rate": 9.791618287834797e-05,
+      "loss": 0.2606,
+      "step": 15520
+    },
+    {
+      "epoch": 1.9553799162389396,
+      "grad_norm": 0.29080766439437866,
+      "learning_rate": 9.781311331591747e-05,
+      "loss": 0.2656,
+      "step": 15525
+    },
+    {
+      "epoch": 1.9560096986491167,
+      "grad_norm": 0.24801793694496155,
+      "learning_rate": 9.771007177903723e-05,
+      "loss": 0.2651,
+      "step": 15530
+    },
+    {
+      "epoch": 1.956639481059294,
+      "grad_norm": 0.22467739880084991,
+      "learning_rate": 9.76070583230429e-05,
+      "loss": 0.2663,
+      "step": 15535
+    },
+    {
+      "epoch": 1.9572692634694713,
+      "grad_norm": 0.24151213467121124,
+      "learning_rate": 9.750407300325502e-05,
+      "loss": 0.2612,
+      "step": 15540
+    },
+    {
+      "epoch": 1.9578990458796486,
+      "grad_norm": 0.262352854013443,
+      "learning_rate": 9.7401115874979e-05,
+      "loss": 0.2508,
+      "step": 15545
+    },
+    {
+      "epoch": 1.9585288282898259,
+      "grad_norm": 0.2491580843925476,
+      "learning_rate": 9.72981869935053e-05,
+      "loss": 0.2735,
+      "step": 15550
+    },
+    {
+      "epoch": 1.9591586107000032,
+      "grad_norm": 0.27000558376312256,
+      "learning_rate": 9.719528641410898e-05,
+      "loss": 0.2794,
+      "step": 15555
+    },
+    {
+      "epoch": 1.9597883931101805,
+      "grad_norm": 0.2562926113605499,
+      "learning_rate": 9.709241419205008e-05,
+      "loss": 0.2829,
+      "step": 15560
+    },
+    {
+      "epoch": 1.9604181755203576,
+      "grad_norm": 0.2559642493724823,
+      "learning_rate": 9.69895703825733e-05,
+      "loss": 0.2768,
+      "step": 15565
+    },
+    {
+      "epoch": 1.961047957930535,
+      "grad_norm": 0.23282787203788757,
+      "learning_rate": 9.688675504090811e-05,
+      "loss": 0.2648,
+      "step": 15570
+    },
+    {
+      "epoch": 1.9616777403407122,
+      "grad_norm": 0.2280416637659073,
+      "learning_rate": 9.678396822226868e-05,
+      "loss": 0.2474,
+      "step": 15575
+    },
+    {
+      "epoch": 1.9623075227508897,
+      "grad_norm": 0.2516798973083496,
+      "learning_rate": 9.668120998185392e-05,
+      "loss": 0.2855,
+      "step": 15580
+    },
+    {
+      "epoch": 1.9629373051610668,
+      "grad_norm": 0.24892964959144592,
+      "learning_rate": 9.657848037484726e-05,
+      "loss": 0.2731,
+      "step": 15585
+    },
+    {
+      "epoch": 1.963567087571244,
+      "grad_norm": 0.2524420917034149,
+      "learning_rate": 9.647577945641699e-05,
+      "loss": 0.275,
+      "step": 15590
+    },
+    {
+      "epoch": 1.9641968699814214,
+      "grad_norm": 0.2617582380771637,
+      "learning_rate": 9.637310728171577e-05,
+      "loss": 0.293,
+      "step": 15595
+    },
+    {
+      "epoch": 1.9648266523915987,
+      "grad_norm": 0.2635948061943054,
+      "learning_rate": 9.627046390588086e-05,
+      "loss": 0.2642,
+      "step": 15600
+    },
+    {
+      "epoch": 1.965456434801776,
+      "grad_norm": 0.22701425850391388,
+      "learning_rate": 9.61678493840342e-05,
+      "loss": 0.2647,
+      "step": 15605
+    },
+    {
+      "epoch": 1.9660862172119533,
+      "grad_norm": 0.2594752609729767,
+      "learning_rate": 9.606526377128207e-05,
+      "loss": 0.2846,
+      "step": 15610
+    },
+    {
+      "epoch": 1.9667159996221306,
+      "grad_norm": 0.25541216135025024,
+      "learning_rate": 9.596270712271524e-05,
+      "loss": 0.2712,
+      "step": 15615
+    },
+    {
+      "epoch": 1.9673457820323077,
+      "grad_norm": 0.26473337411880493,
+      "learning_rate": 9.586017949340909e-05,
+      "loss": 0.2515,
+      "step": 15620
+    },
+    {
+      "epoch": 1.9679755644424852,
+      "grad_norm": 0.2370501607656479,
+      "learning_rate": 9.575768093842321e-05,
+      "loss": 0.2569,
+      "step": 15625
+    },
+    {
+      "epoch": 1.9686053468526623,
+      "grad_norm": 0.25999268889427185,
+      "learning_rate": 9.565521151280168e-05,
+      "loss": 0.2846,
+      "step": 15630
+    },
+    {
+      "epoch": 1.9692351292628398,
+      "grad_norm": 0.2597227394580841,
+      "learning_rate": 9.555277127157294e-05,
+      "loss": 0.2814,
+      "step": 15635
+    },
+    {
+      "epoch": 1.969864911673017,
+      "grad_norm": 0.22267143428325653,
+      "learning_rate": 9.545036026974979e-05,
+      "loss": 0.2703,
+      "step": 15640
+    },
+    {
+      "epoch": 1.9704946940831942,
+      "grad_norm": 0.2599702477455139,
+      "learning_rate": 9.534797856232913e-05,
+      "loss": 0.2741,
+      "step": 15645
+    },
+    {
+      "epoch": 1.9711244764933715,
+      "grad_norm": 0.23703083395957947,
+      "learning_rate": 9.524562620429243e-05,
+      "loss": 0.2657,
+      "step": 15650
+    },
+    {
+      "epoch": 1.9717542589035488,
+      "grad_norm": 0.24194732308387756,
+      "learning_rate": 9.514330325060515e-05,
+      "loss": 0.2613,
+      "step": 15655
+    },
+    {
+      "epoch": 1.9723840413137261,
+      "grad_norm": 0.2648374140262604,
+      "learning_rate": 9.504100975621709e-05,
+      "loss": 0.2808,
+      "step": 15660
+    },
+    {
+      "epoch": 1.9730138237239034,
+      "grad_norm": 0.2491552084684372,
+      "learning_rate": 9.493874577606218e-05,
+      "loss": 0.2622,
+      "step": 15665
+    },
+    {
+      "epoch": 1.9736436061340807,
+      "grad_norm": 0.25322696566581726,
+      "learning_rate": 9.483651136505857e-05,
+      "loss": 0.2716,
+      "step": 15670
+    },
+    {
+      "epoch": 1.9742733885442578,
+      "grad_norm": 0.22565199434757233,
+      "learning_rate": 9.473430657810838e-05,
+      "loss": 0.2947,
+      "step": 15675
+    },
+    {
+      "epoch": 1.9749031709544353,
+      "grad_norm": 0.2594245672225952,
+      "learning_rate": 9.463213147009795e-05,
+      "loss": 0.2793,
+      "step": 15680
+    },
+    {
+      "epoch": 1.9755329533646124,
+      "grad_norm": 0.2432025521993637,
+      "learning_rate": 9.452998609589769e-05,
+      "loss": 0.2559,
+      "step": 15685
+    },
+    {
+      "epoch": 1.97616273577479,
+      "grad_norm": 0.2537454068660736,
+      "learning_rate": 9.442787051036192e-05,
+      "loss": 0.2842,
+      "step": 15690
+    },
+    {
+      "epoch": 1.976792518184967,
+      "grad_norm": 0.2597581446170807,
+      "learning_rate": 9.432578476832911e-05,
+      "loss": 0.2755,
+      "step": 15695
+    },
+    {
+      "epoch": 1.9774223005951443,
+      "grad_norm": 0.25382810831069946,
+      "learning_rate": 9.42237289246216e-05,
+      "loss": 0.2653,
+      "step": 15700
+    },
+    {
+      "epoch": 1.9780520830053216,
+      "grad_norm": 0.23822832107543945,
+      "learning_rate": 9.412170303404579e-05,
+      "loss": 0.2624,
+      "step": 15705
+    },
+    {
+      "epoch": 1.978681865415499,
+      "grad_norm": 0.2722800076007843,
+      "learning_rate": 9.40197071513918e-05,
+      "loss": 0.2712,
+      "step": 15710
+    },
+    {
+      "epoch": 1.9793116478256763,
+      "grad_norm": 0.2273283749818802,
+      "learning_rate": 9.39177413314338e-05,
+      "loss": 0.2545,
+      "step": 15715
+    },
+    {
+      "epoch": 1.9799414302358536,
+      "grad_norm": 0.24674946069717407,
+      "learning_rate": 9.381580562892972e-05,
+      "loss": 0.2606,
+      "step": 15720
+    },
+    {
+      "epoch": 1.9805712126460309,
+      "grad_norm": 0.23100855946540833,
+      "learning_rate": 9.371390009862145e-05,
+      "loss": 0.2632,
+      "step": 15725
+    },
+    {
+      "epoch": 1.981200995056208,
+      "grad_norm": 0.23489323258399963,
+      "learning_rate": 9.361202479523448e-05,
+      "loss": 0.2833,
+      "step": 15730
+    },
+    {
+      "epoch": 1.9818307774663855,
+      "grad_norm": 0.26526087522506714,
+      "learning_rate": 9.35101797734783e-05,
+      "loss": 0.2717,
+      "step": 15735
+    },
+    {
+      "epoch": 1.9824605598765626,
+      "grad_norm": 0.27042046189308167,
+      "learning_rate": 9.340836508804595e-05,
+      "loss": 0.2664,
+      "step": 15740
+    },
+    {
+      "epoch": 1.98309034228674,
+      "grad_norm": 0.28461650013923645,
+      "learning_rate": 9.330658079361422e-05,
+      "loss": 0.26,
+      "step": 15745
+    },
+    {
+      "epoch": 1.9837201246969172,
+      "grad_norm": 0.26529213786125183,
+      "learning_rate": 9.320482694484356e-05,
+      "loss": 0.2808,
+      "step": 15750
+    },
+    {
+      "epoch": 1.9843499071070945,
+      "grad_norm": 0.32026639580726624,
+      "learning_rate": 9.310310359637823e-05,
+      "loss": 0.2631,
+      "step": 15755
+    },
+    {
+      "epoch": 1.9849796895172718,
+      "grad_norm": 0.2596029043197632,
+      "learning_rate": 9.300141080284588e-05,
+      "loss": 0.2771,
+      "step": 15760
+    },
+    {
+      "epoch": 1.985609471927449,
+      "grad_norm": 0.25400853157043457,
+      "learning_rate": 9.289974861885796e-05,
+      "loss": 0.2532,
+      "step": 15765
+    },
+    {
+      "epoch": 1.9862392543376264,
+      "grad_norm": 0.29176244139671326,
+      "learning_rate": 9.279811709900934e-05,
+      "loss": 0.2719,
+      "step": 15770
+    },
+    {
+      "epoch": 1.9868690367478037,
+      "grad_norm": 0.2530720829963684,
+      "learning_rate": 9.26965162978785e-05,
+      "loss": 0.2676,
+      "step": 15775
+    },
+    {
+      "epoch": 1.987498819157981,
+      "grad_norm": 0.23311518132686615,
+      "learning_rate": 9.259494627002728e-05,
+      "loss": 0.2632,
+      "step": 15780
+    },
+    {
+      "epoch": 1.988128601568158,
+      "grad_norm": 0.2402007132768631,
+      "learning_rate": 9.249340707000123e-05,
+      "loss": 0.2419,
+      "step": 15785
+    },
+    {
+      "epoch": 1.9887583839783356,
+      "grad_norm": 0.24241755902767181,
+      "learning_rate": 9.239189875232914e-05,
+      "loss": 0.2567,
+      "step": 15790
+    },
+    {
+      "epoch": 1.9893881663885127,
+      "grad_norm": 0.31180015206336975,
+      "learning_rate": 9.229042137152337e-05,
+      "loss": 0.2864,
+      "step": 15795
+    },
+    {
+      "epoch": 1.9900179487986902,
+      "grad_norm": 0.29330986738204956,
+      "learning_rate": 9.218897498207952e-05,
+      "loss": 0.2626,
+      "step": 15800
+    },
+    {
+      "epoch": 1.9906477312088673,
+      "grad_norm": 0.24687117338180542,
+      "learning_rate": 9.208755963847663e-05,
+      "loss": 0.2721,
+      "step": 15805
+    },
+    {
+      "epoch": 1.9912775136190446,
+      "grad_norm": 0.22886700928211212,
+      "learning_rate": 9.198617539517714e-05,
+      "loss": 0.2626,
+      "step": 15810
+    },
+    {
+      "epoch": 1.991907296029222,
+      "grad_norm": 0.2331649363040924,
+      "learning_rate": 9.188482230662662e-05,
+      "loss": 0.2484,
+      "step": 15815
+    },
+    {
+      "epoch": 1.9925370784393992,
+      "grad_norm": 0.23371124267578125,
+      "learning_rate": 9.178350042725397e-05,
+      "loss": 0.291,
+      "step": 15820
+    },
+    {
+      "epoch": 1.9931668608495765,
+      "grad_norm": 0.2478175163269043,
+      "learning_rate": 9.168220981147143e-05,
+      "loss": 0.2748,
+      "step": 15825
+    },
+    {
+      "epoch": 1.9937966432597538,
+      "grad_norm": 0.25952714681625366,
+      "learning_rate": 9.158095051367433e-05,
+      "loss": 0.2568,
+      "step": 15830
+    },
+    {
+      "epoch": 1.9944264256699311,
+      "grad_norm": 0.2522846758365631,
+      "learning_rate": 9.14797225882412e-05,
+      "loss": 0.2414,
+      "step": 15835
+    },
+    {
+      "epoch": 1.9950562080801082,
+      "grad_norm": 0.24577966332435608,
+      "learning_rate": 9.137852608953384e-05,
+      "loss": 0.2573,
+      "step": 15840
+    },
+    {
+      "epoch": 1.9956859904902857,
+      "grad_norm": 0.2714809775352478,
+      "learning_rate": 9.127736107189705e-05,
+      "loss": 0.2703,
+      "step": 15845
+    },
+    {
+      "epoch": 1.9963157729004628,
+      "grad_norm": 0.25562793016433716,
+      "learning_rate": 9.117622758965866e-05,
+      "loss": 0.2601,
+      "step": 15850
+    },
+    {
+      "epoch": 1.9969455553106403,
+      "grad_norm": 0.23811009526252747,
+      "learning_rate": 9.107512569712975e-05,
+      "loss": 0.2474,
+      "step": 15855
+    },
+    {
+      "epoch": 1.9975753377208174,
+      "grad_norm": 0.24953188002109528,
+      "learning_rate": 9.097405544860437e-05,
+      "loss": 0.2582,
+      "step": 15860
+    },
+    {
+      "epoch": 1.9982051201309947,
+      "grad_norm": 0.23611120879650116,
+      "learning_rate": 9.087301689835944e-05,
+      "loss": 0.253,
+      "step": 15865
+    },
+    {
+      "epoch": 1.998834902541172,
+      "grad_norm": 0.24487170577049255,
+      "learning_rate": 9.077201010065509e-05,
+      "loss": 0.2508,
+      "step": 15870
+    },
+    {
+      "epoch": 1.9994646849513493,
+      "grad_norm": 0.2839270830154419,
+      "learning_rate": 9.06710351097342e-05,
+      "loss": 0.2748,
+      "step": 15875
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.2304636538028717,
+      "learning_rate": 9.057009197982272e-05,
+      "loss": 0.2653,
+      "step": 15880
+    },
+    {
+      "epoch": 2.000629782410177,
+      "grad_norm": 0.19537301361560822,
+      "learning_rate": 9.046918076512935e-05,
+      "loss": 0.1844,
+      "step": 15885
+    },
+    {
+      "epoch": 2.0012595648203546,
+      "grad_norm": 0.22054894268512726,
+      "learning_rate": 9.036830151984571e-05,
+      "loss": 0.196,
+      "step": 15890
+    },
+    {
+      "epoch": 2.0018893472305317,
+      "grad_norm": 0.23987694084644318,
+      "learning_rate": 9.02674542981463e-05,
+      "loss": 0.1962,
+      "step": 15895
+    },
+    {
+      "epoch": 2.002519129640709,
+      "grad_norm": 0.24562768638134003,
+      "learning_rate": 9.016663915418835e-05,
+      "loss": 0.1826,
+      "step": 15900
+    },
+    {
+      "epoch": 2.0031489120508863,
+      "grad_norm": 0.27111175656318665,
+      "learning_rate": 9.00658561421119e-05,
+      "loss": 0.2025,
+      "step": 15905
+    },
+    {
+      "epoch": 2.003778694461064,
+      "grad_norm": 0.21321839094161987,
+      "learning_rate": 8.99651053160398e-05,
+      "loss": 0.1743,
+      "step": 15910
+    },
+    {
+      "epoch": 2.004408476871241,
+      "grad_norm": 0.2295263558626175,
+      "learning_rate": 8.986438673007749e-05,
+      "loss": 0.1856,
+      "step": 15915
+    },
+    {
+      "epoch": 2.0050382592814184,
+      "grad_norm": 0.22658327221870422,
+      "learning_rate": 8.976370043831313e-05,
+      "loss": 0.1896,
+      "step": 15920
+    },
+    {
+      "epoch": 2.0056680416915955,
+      "grad_norm": 0.21595464646816254,
+      "learning_rate": 8.966304649481753e-05,
+      "loss": 0.1865,
+      "step": 15925
+    },
+    {
+      "epoch": 2.006297824101773,
+      "grad_norm": 0.23339222371578217,
+      "learning_rate": 8.956242495364426e-05,
+      "loss": 0.1866,
+      "step": 15930
+    },
+    {
+      "epoch": 2.00692760651195,
+      "grad_norm": 0.20041927695274353,
+      "learning_rate": 8.946183586882929e-05,
+      "loss": 0.1745,
+      "step": 15935
+    },
+    {
+      "epoch": 2.007557388922127,
+      "grad_norm": 0.19914592802524567,
+      "learning_rate": 8.936127929439131e-05,
+      "loss": 0.1885,
+      "step": 15940
+    },
+    {
+      "epoch": 2.0081871713323047,
+      "grad_norm": 0.20688550174236298,
+      "learning_rate": 8.926075528433149e-05,
+      "loss": 0.1932,
+      "step": 15945
+    },
+    {
+      "epoch": 2.008816953742482,
+      "grad_norm": 0.23507048189640045,
+      "learning_rate": 8.916026389263358e-05,
+      "loss": 0.1865,
+      "step": 15950
+    },
+    {
+      "epoch": 2.0094467361526593,
+      "grad_norm": 0.2366725355386734,
+      "learning_rate": 8.905980517326358e-05,
+      "loss": 0.1867,
+      "step": 15955
+    },
+    {
+      "epoch": 2.0100765185628364,
+      "grad_norm": 0.20678187906742096,
+      "learning_rate": 8.895937918017028e-05,
+      "loss": 0.1785,
+      "step": 15960
+    },
+    {
+      "epoch": 2.010706300973014,
+      "grad_norm": 0.2642296850681305,
+      "learning_rate": 8.885898596728463e-05,
+      "loss": 0.1812,
+      "step": 15965
+    },
+    {
+      "epoch": 2.011336083383191,
+      "grad_norm": 0.20598894357681274,
+      "learning_rate": 8.875862558852016e-05,
+      "loss": 0.1861,
+      "step": 15970
+    },
+    {
+      "epoch": 2.0119658657933686,
+      "grad_norm": 0.23556114733219147,
+      "learning_rate": 8.865829809777265e-05,
+      "loss": 0.1873,
+      "step": 15975
+    },
+    {
+      "epoch": 2.0125956482035456,
+      "grad_norm": 0.25772175192832947,
+      "learning_rate": 8.855800354892022e-05,
+      "loss": 0.1858,
+      "step": 15980
+    },
+    {
+      "epoch": 2.013225430613723,
+      "grad_norm": 0.21538549661636353,
+      "learning_rate": 8.845774199582344e-05,
+      "loss": 0.1738,
+      "step": 15985
+    },
+    {
+      "epoch": 2.0138552130239002,
+      "grad_norm": 0.22819840908050537,
+      "learning_rate": 8.835751349232496e-05,
+      "loss": 0.1843,
+      "step": 15990
+    },
+    {
+      "epoch": 2.0144849954340773,
+      "grad_norm": 0.23319579660892487,
+      "learning_rate": 8.825731809224976e-05,
+      "loss": 0.1878,
+      "step": 15995
+    },
+    {
+      "epoch": 2.015114777844255,
+      "grad_norm": 0.24107947945594788,
+      "learning_rate": 8.815715584940511e-05,
+      "loss": 0.1867,
+      "step": 16000
+    },
+    {
+      "epoch": 2.015114777844255,
+      "eval_loss": 0.3415575921535492,
+      "eval_runtime": 6.166,
+      "eval_samples_per_second": 162.181,
+      "eval_steps_per_second": 10.217,
+      "step": 16000
+    },
+    {
+      "epoch": 2.015744560254432,
+      "grad_norm": 0.2272019386291504,
+      "learning_rate": 8.805702681758042e-05,
+      "loss": 0.1718,
+      "step": 16005
+    },
+    {
+      "epoch": 2.0163743426646095,
+      "grad_norm": 0.22147491574287415,
+      "learning_rate": 8.795693105054723e-05,
+      "loss": 0.175,
+      "step": 16010
+    },
+    {
+      "epoch": 2.0170041250747865,
+      "grad_norm": 0.21899926662445068,
+      "learning_rate": 8.785686860205929e-05,
+      "loss": 0.1749,
+      "step": 16015
+    },
+    {
+      "epoch": 2.017633907484964,
+      "grad_norm": 0.24299047887325287,
+      "learning_rate": 8.775683952585246e-05,
+      "loss": 0.1902,
+      "step": 16020
+    },
+    {
+      "epoch": 2.018263689895141,
+      "grad_norm": 0.24278461933135986,
+      "learning_rate": 8.765684387564454e-05,
+      "loss": 0.1872,
+      "step": 16025
+    },
+    {
+      "epoch": 2.0188934723053187,
+      "grad_norm": 0.24929705262184143,
+      "learning_rate": 8.75568817051355e-05,
+      "loss": 0.1838,
+      "step": 16030
+    },
+    {
+      "epoch": 2.0195232547154958,
+      "grad_norm": 0.20675018429756165,
+      "learning_rate": 8.745695306800738e-05,
+      "loss": 0.1734,
+      "step": 16035
+    },
+    {
+      "epoch": 2.0201530371256733,
+      "grad_norm": 0.25064778327941895,
+      "learning_rate": 8.73570580179241e-05,
+      "loss": 0.1821,
+      "step": 16040
+    },
+    {
+      "epoch": 2.0207828195358504,
+      "grad_norm": 0.23618988692760468,
+      "learning_rate": 8.725719660853157e-05,
+      "loss": 0.1935,
+      "step": 16045
+    },
+    {
+      "epoch": 2.0214126019460275,
+      "grad_norm": 0.2201015204191208,
+      "learning_rate": 8.715736889345766e-05,
+      "loss": 0.1806,
+      "step": 16050
+    },
+    {
+      "epoch": 2.022042384356205,
+      "grad_norm": 0.23748455941677094,
+      "learning_rate": 8.705757492631214e-05,
+      "loss": 0.1807,
+      "step": 16055
+    },
+    {
+      "epoch": 2.022672166766382,
+      "grad_norm": 0.2563530504703522,
+      "learning_rate": 8.695781476068664e-05,
+      "loss": 0.1825,
+      "step": 16060
+    },
+    {
+      "epoch": 2.0233019491765596,
+      "grad_norm": 0.27659016847610474,
+      "learning_rate": 8.685808845015464e-05,
+      "loss": 0.1861,
+      "step": 16065
+    },
+    {
+      "epoch": 2.0239317315867367,
+      "grad_norm": 0.19301186501979828,
+      "learning_rate": 8.675839604827146e-05,
+      "loss": 0.1804,
+      "step": 16070
+    },
+    {
+      "epoch": 2.024561513996914,
+      "grad_norm": 0.245374858379364,
+      "learning_rate": 8.665873760857415e-05,
+      "loss": 0.1785,
+      "step": 16075
+    },
+    {
+      "epoch": 2.0251912964070913,
+      "grad_norm": 0.21472232043743134,
+      "learning_rate": 8.655911318458166e-05,
+      "loss": 0.1785,
+      "step": 16080
+    },
+    {
+      "epoch": 2.025821078817269,
+      "grad_norm": 0.22257132828235626,
+      "learning_rate": 8.645952282979453e-05,
+      "loss": 0.1812,
+      "step": 16085
+    },
+    {
+      "epoch": 2.026450861227446,
+      "grad_norm": 0.25223472714424133,
+      "learning_rate": 8.635996659769512e-05,
+      "loss": 0.1934,
+      "step": 16090
+    },
+    {
+      "epoch": 2.0270806436376234,
+      "grad_norm": 0.22251825034618378,
+      "learning_rate": 8.626044454174724e-05,
+      "loss": 0.1895,
+      "step": 16095
+    },
+    {
+      "epoch": 2.0277104260478005,
+      "grad_norm": 0.2073337882757187,
+      "learning_rate": 8.616095671539663e-05,
+      "loss": 0.1851,
+      "step": 16100
+    },
+    {
+      "epoch": 2.0283402084579776,
+      "grad_norm": 0.21960042417049408,
+      "learning_rate": 8.606150317207053e-05,
+      "loss": 0.1809,
+      "step": 16105
+    },
+    {
+      "epoch": 2.028969990868155,
+      "grad_norm": 0.23633064329624176,
+      "learning_rate": 8.596208396517771e-05,
+      "loss": 0.1839,
+      "step": 16110
+    },
+    {
+      "epoch": 2.029599773278332,
+      "grad_norm": 0.21128375828266144,
+      "learning_rate": 8.586269914810855e-05,
+      "loss": 0.1828,
+      "step": 16115
+    },
+    {
+      "epoch": 2.0302295556885097,
+      "grad_norm": 0.24467304348945618,
+      "learning_rate": 8.576334877423505e-05,
+      "loss": 0.1784,
+      "step": 16120
+    },
+    {
+      "epoch": 2.030859338098687,
+      "grad_norm": 0.24976873397827148,
+      "learning_rate": 8.566403289691062e-05,
+      "loss": 0.1924,
+      "step": 16125
+    },
+    {
+      "epoch": 2.0314891205088643,
+      "grad_norm": 0.258323609828949,
+      "learning_rate": 8.556475156947008e-05,
+      "loss": 0.1889,
+      "step": 16130
+    },
+    {
+      "epoch": 2.0321189029190414,
+      "grad_norm": 0.24420535564422607,
+      "learning_rate": 8.546550484522973e-05,
+      "loss": 0.197,
+      "step": 16135
+    },
+    {
+      "epoch": 2.032748685329219,
+      "grad_norm": 0.2438700944185257,
+      "learning_rate": 8.536629277748746e-05,
+      "loss": 0.1958,
+      "step": 16140
+    },
+    {
+      "epoch": 2.033378467739396,
+      "grad_norm": 0.25343936681747437,
+      "learning_rate": 8.526711541952236e-05,
+      "loss": 0.1877,
+      "step": 16145
+    },
+    {
+      "epoch": 2.0340082501495735,
+      "grad_norm": 0.24403081834316254,
+      "learning_rate": 8.516797282459493e-05,
+      "loss": 0.1774,
+      "step": 16150
+    },
+    {
+      "epoch": 2.0346380325597506,
+      "grad_norm": 0.24733777344226837,
+      "learning_rate": 8.506886504594704e-05,
+      "loss": 0.1792,
+      "step": 16155
+    },
+    {
+      "epoch": 2.0352678149699277,
+      "grad_norm": 0.22619028389453888,
+      "learning_rate": 8.496979213680177e-05,
+      "loss": 0.1807,
+      "step": 16160
+    },
+    {
+      "epoch": 2.0358975973801052,
+      "grad_norm": 0.23040007054805756,
+      "learning_rate": 8.48707541503636e-05,
+      "loss": 0.1804,
+      "step": 16165
+    },
+    {
+      "epoch": 2.0365273797902823,
+      "grad_norm": 0.21034270524978638,
+      "learning_rate": 8.477175113981813e-05,
+      "loss": 0.1787,
+      "step": 16170
+    },
+    {
+      "epoch": 2.03715716220046,
+      "grad_norm": 0.21682168543338776,
+      "learning_rate": 8.467278315833224e-05,
+      "loss": 0.1817,
+      "step": 16175
+    },
+    {
+      "epoch": 2.037786944610637,
+      "grad_norm": 0.2700116038322449,
+      "learning_rate": 8.457385025905407e-05,
+      "loss": 0.1896,
+      "step": 16180
+    },
+    {
+      "epoch": 2.0384167270208144,
+      "grad_norm": 0.214239239692688,
+      "learning_rate": 8.44749524951128e-05,
+      "loss": 0.1827,
+      "step": 16185
+    },
+    {
+      "epoch": 2.0390465094309915,
+      "grad_norm": 0.2243194878101349,
+      "learning_rate": 8.437608991961885e-05,
+      "loss": 0.1833,
+      "step": 16190
+    },
+    {
+      "epoch": 2.039676291841169,
+      "grad_norm": 0.28487569093704224,
+      "learning_rate": 8.427726258566353e-05,
+      "loss": 0.1901,
+      "step": 16195
+    },
+    {
+      "epoch": 2.040306074251346,
+      "grad_norm": 0.24857446551322937,
+      "learning_rate": 8.41784705463195e-05,
+      "loss": 0.192,
+      "step": 16200
+    },
+    {
+      "epoch": 2.0409358566615237,
+      "grad_norm": 0.22208547592163086,
+      "learning_rate": 8.407971385464032e-05,
+      "loss": 0.1907,
+      "step": 16205
+    },
+    {
+      "epoch": 2.0415656390717007,
+      "grad_norm": 0.22752498090267181,
+      "learning_rate": 8.398099256366057e-05,
+      "loss": 0.1827,
+      "step": 16210
+    },
+    {
+      "epoch": 2.042195421481878,
+      "grad_norm": 0.25674304366111755,
+      "learning_rate": 8.388230672639584e-05,
+      "loss": 0.1889,
+      "step": 16215
+    },
+    {
+      "epoch": 2.0428252038920554,
+      "grad_norm": 0.22372281551361084,
+      "learning_rate": 8.378365639584264e-05,
+      "loss": 0.1816,
+      "step": 16220
+    },
+    {
+      "epoch": 2.0434549863022324,
+      "grad_norm": 0.25298216938972473,
+      "learning_rate": 8.368504162497859e-05,
+      "loss": 0.1813,
+      "step": 16225
+    },
+    {
+      "epoch": 2.04408476871241,
+      "grad_norm": 0.21058551967144012,
+      "learning_rate": 8.358646246676197e-05,
+      "loss": 0.1855,
+      "step": 16230
+    },
+    {
+      "epoch": 2.044714551122587,
+      "grad_norm": 0.2757975459098816,
+      "learning_rate": 8.348791897413196e-05,
+      "loss": 0.1749,
+      "step": 16235
+    },
+    {
+      "epoch": 2.0453443335327646,
+      "grad_norm": 0.22646676003932953,
+      "learning_rate": 8.338941120000884e-05,
+      "loss": 0.1852,
+      "step": 16240
+    },
+    {
+      "epoch": 2.0459741159429417,
+      "grad_norm": 0.23769816756248474,
+      "learning_rate": 8.329093919729342e-05,
+      "loss": 0.1869,
+      "step": 16245
+    },
+    {
+      "epoch": 2.046603898353119,
+      "grad_norm": 0.22907455265522003,
+      "learning_rate": 8.319250301886746e-05,
+      "loss": 0.1876,
+      "step": 16250
+    },
+    {
+      "epoch": 2.0472336807632963,
+      "grad_norm": 0.22925196588039398,
+      "learning_rate": 8.309410271759342e-05,
+      "loss": 0.1885,
+      "step": 16255
+    },
+    {
+      "epoch": 2.0478634631734733,
+      "grad_norm": 0.22043700516223907,
+      "learning_rate": 8.299573834631454e-05,
+      "loss": 0.181,
+      "step": 16260
+    },
+    {
+      "epoch": 2.048493245583651,
+      "grad_norm": 0.23858542740345,
+      "learning_rate": 8.289740995785468e-05,
+      "loss": 0.1898,
+      "step": 16265
+    },
+    {
+      "epoch": 2.049123027993828,
+      "grad_norm": 0.23982049524784088,
+      "learning_rate": 8.279911760501846e-05,
+      "loss": 0.1838,
+      "step": 16270
+    },
+    {
+      "epoch": 2.0497528104040055,
+      "grad_norm": 0.21694807708263397,
+      "learning_rate": 8.270086134059113e-05,
+      "loss": 0.1795,
+      "step": 16275
+    },
+    {
+      "epoch": 2.0503825928141826,
+      "grad_norm": 0.20050913095474243,
+      "learning_rate": 8.260264121733846e-05,
+      "loss": 0.175,
+      "step": 16280
+    },
+    {
+      "epoch": 2.05101237522436,
+      "grad_norm": 0.2118636816740036,
+      "learning_rate": 8.250445728800706e-05,
+      "loss": 0.1778,
+      "step": 16285
+    },
+    {
+      "epoch": 2.051642157634537,
+      "grad_norm": 0.2250407338142395,
+      "learning_rate": 8.240630960532382e-05,
+      "loss": 0.1885,
+      "step": 16290
+    },
+    {
+      "epoch": 2.0522719400447147,
+      "grad_norm": 0.2565051019191742,
+      "learning_rate": 8.230819822199642e-05,
+      "loss": 0.1901,
+      "step": 16295
+    },
+    {
+      "epoch": 2.0529017224548918,
+      "grad_norm": 0.24367564916610718,
+      "learning_rate": 8.221012319071268e-05,
+      "loss": 0.1798,
+      "step": 16300
+    },
+    {
+      "epoch": 2.0535315048650693,
+      "grad_norm": 0.24313905835151672,
+      "learning_rate": 8.211208456414135e-05,
+      "loss": 0.1908,
+      "step": 16305
+    },
+    {
+      "epoch": 2.0541612872752464,
+      "grad_norm": 0.23950958251953125,
+      "learning_rate": 8.201408239493131e-05,
+      "loss": 0.1815,
+      "step": 16310
+    },
+    {
+      "epoch": 2.0547910696854235,
+      "grad_norm": 0.24551273882389069,
+      "learning_rate": 8.1916116735712e-05,
+      "loss": 0.1941,
+      "step": 16315
+    },
+    {
+      "epoch": 2.055420852095601,
+      "grad_norm": 0.21070988476276398,
+      "learning_rate": 8.181818763909314e-05,
+      "loss": 0.1868,
+      "step": 16320
+    },
+    {
+      "epoch": 2.056050634505778,
+      "grad_norm": 0.21926933526992798,
+      "learning_rate": 8.172029515766502e-05,
+      "loss": 0.1848,
+      "step": 16325
+    },
+    {
+      "epoch": 2.0566804169159556,
+      "grad_norm": 0.22517934441566467,
+      "learning_rate": 8.162243934399812e-05,
+      "loss": 0.1912,
+      "step": 16330
+    },
+    {
+      "epoch": 2.0573101993261327,
+      "grad_norm": 0.2571990489959717,
+      "learning_rate": 8.152462025064315e-05,
+      "loss": 0.1834,
+      "step": 16335
+    },
+    {
+      "epoch": 2.05793998173631,
+      "grad_norm": 0.22555163502693176,
+      "learning_rate": 8.14268379301312e-05,
+      "loss": 0.19,
+      "step": 16340
+    },
+    {
+      "epoch": 2.0585697641464873,
+      "grad_norm": 0.2326682209968567,
+      "learning_rate": 8.13290924349737e-05,
+      "loss": 0.1786,
+      "step": 16345
+    },
+    {
+      "epoch": 2.059199546556665,
+      "grad_norm": 0.22472088038921356,
+      "learning_rate": 8.123138381766218e-05,
+      "loss": 0.1843,
+      "step": 16350
+    },
+    {
+      "epoch": 2.059829328966842,
+      "grad_norm": 0.2206810563802719,
+      "learning_rate": 8.113371213066838e-05,
+      "loss": 0.1781,
+      "step": 16355
+    },
+    {
+      "epoch": 2.0604591113770194,
+      "grad_norm": 0.2740577757358551,
+      "learning_rate": 8.103607742644426e-05,
+      "loss": 0.1875,
+      "step": 16360
+    },
+    {
+      "epoch": 2.0610888937871965,
+      "grad_norm": 0.22217485308647156,
+      "learning_rate": 8.093847975742185e-05,
+      "loss": 0.1748,
+      "step": 16365
+    },
+    {
+      "epoch": 2.0617186761973736,
+      "grad_norm": 0.2460946887731552,
+      "learning_rate": 8.084091917601336e-05,
+      "loss": 0.1839,
+      "step": 16370
+    },
+    {
+      "epoch": 2.062348458607551,
+      "grad_norm": 0.2489384114742279,
+      "learning_rate": 8.074339573461101e-05,
+      "loss": 0.1818,
+      "step": 16375
+    },
+    {
+      "epoch": 2.062978241017728,
+      "grad_norm": 0.22755055129528046,
+      "learning_rate": 8.06459094855871e-05,
+      "loss": 0.1885,
+      "step": 16380
+    },
+    {
+      "epoch": 2.0636080234279057,
+      "grad_norm": 0.22558000683784485,
+      "learning_rate": 8.054846048129406e-05,
+      "loss": 0.1805,
+      "step": 16385
+    },
+    {
+      "epoch": 2.064237805838083,
+      "grad_norm": 0.2083364725112915,
+      "learning_rate": 8.045104877406418e-05,
+      "loss": 0.1809,
+      "step": 16390
+    },
+    {
+      "epoch": 2.0648675882482603,
+      "grad_norm": 0.23679542541503906,
+      "learning_rate": 8.035367441620976e-05,
+      "loss": 0.181,
+      "step": 16395
+    },
+    {
+      "epoch": 2.0654973706584374,
+      "grad_norm": 0.2173621654510498,
+      "learning_rate": 8.025633746002311e-05,
+      "loss": 0.1857,
+      "step": 16400
+    },
+    {
+      "epoch": 2.066127153068615,
+      "grad_norm": 0.22376009821891785,
+      "learning_rate": 8.015903795777634e-05,
+      "loss": 0.1832,
+      "step": 16405
+    },
+    {
+      "epoch": 2.066756935478792,
+      "grad_norm": 0.24444858729839325,
+      "learning_rate": 8.00617759617215e-05,
+      "loss": 0.1959,
+      "step": 16410
+    },
+    {
+      "epoch": 2.0673867178889695,
+      "grad_norm": 0.21472635865211487,
+      "learning_rate": 7.996455152409055e-05,
+      "loss": 0.17,
+      "step": 16415
+    },
+    {
+      "epoch": 2.0680165002991466,
+      "grad_norm": 0.22463464736938477,
+      "learning_rate": 7.986736469709521e-05,
+      "loss": 0.1847,
+      "step": 16420
+    },
+    {
+      "epoch": 2.0686462827093237,
+      "grad_norm": 0.2251402884721756,
+      "learning_rate": 7.977021553292696e-05,
+      "loss": 0.1822,
+      "step": 16425
+    },
+    {
+      "epoch": 2.0692760651195012,
+      "grad_norm": 0.21793001890182495,
+      "learning_rate": 7.967310408375725e-05,
+      "loss": 0.1862,
+      "step": 16430
+    },
+    {
+      "epoch": 2.0699058475296783,
+      "grad_norm": 0.2344975620508194,
+      "learning_rate": 7.957603040173714e-05,
+      "loss": 0.1791,
+      "step": 16435
+    },
+    {
+      "epoch": 2.070535629939856,
+      "grad_norm": 0.23466047644615173,
+      "learning_rate": 7.947899453899725e-05,
+      "loss": 0.1867,
+      "step": 16440
+    },
+    {
+      "epoch": 2.071165412350033,
+      "grad_norm": 0.2190965861082077,
+      "learning_rate": 7.93819965476482e-05,
+      "loss": 0.1831,
+      "step": 16445
+    },
+    {
+      "epoch": 2.0717951947602105,
+      "grad_norm": 0.22384218871593475,
+      "learning_rate": 7.928503647978012e-05,
+      "loss": 0.1745,
+      "step": 16450
+    },
+    {
+      "epoch": 2.0724249771703875,
+      "grad_norm": 0.23837679624557495,
+      "learning_rate": 7.918811438746272e-05,
+      "loss": 0.1875,
+      "step": 16455
+    },
+    {
+      "epoch": 2.073054759580565,
+      "grad_norm": 0.2510152757167816,
+      "learning_rate": 7.909123032274542e-05,
+      "loss": 0.1849,
+      "step": 16460
+    },
+    {
+      "epoch": 2.073684541990742,
+      "grad_norm": 0.2514597475528717,
+      "learning_rate": 7.899438433765711e-05,
+      "loss": 0.1882,
+      "step": 16465
+    },
+    {
+      "epoch": 2.0743143244009197,
+      "grad_norm": 0.20441210269927979,
+      "learning_rate": 7.889757648420648e-05,
+      "loss": 0.1754,
+      "step": 16470
+    },
+    {
+      "epoch": 2.0749441068110968,
+      "grad_norm": 0.25783875584602356,
+      "learning_rate": 7.880080681438134e-05,
+      "loss": 0.1859,
+      "step": 16475
+    },
+    {
+      "epoch": 2.075573889221274,
+      "grad_norm": 0.2234499454498291,
+      "learning_rate": 7.870407538014933e-05,
+      "loss": 0.1842,
+      "step": 16480
+    },
+    {
+      "epoch": 2.0762036716314514,
+      "grad_norm": 0.24572981894016266,
+      "learning_rate": 7.860738223345734e-05,
+      "loss": 0.1728,
+      "step": 16485
+    },
+    {
+      "epoch": 2.0768334540416284,
+      "grad_norm": 0.23702028393745422,
+      "learning_rate": 7.851072742623194e-05,
+      "loss": 0.1748,
+      "step": 16490
+    },
+    {
+      "epoch": 2.077463236451806,
+      "grad_norm": 0.23450568318367004,
+      "learning_rate": 7.84141110103789e-05,
+      "loss": 0.1826,
+      "step": 16495
+    },
+    {
+      "epoch": 2.078093018861983,
+      "grad_norm": 0.23022450506687164,
+      "learning_rate": 7.831753303778342e-05,
+      "loss": 0.1684,
+      "step": 16500
+    },
+    {
+      "epoch": 2.0787228012721606,
+      "grad_norm": 0.22727181017398834,
+      "learning_rate": 7.822099356031014e-05,
+      "loss": 0.1751,
+      "step": 16505
+    },
+    {
+      "epoch": 2.0793525836823377,
+      "grad_norm": 0.20935000479221344,
+      "learning_rate": 7.812449262980289e-05,
+      "loss": 0.1748,
+      "step": 16510
+    },
+    {
+      "epoch": 2.079982366092515,
+      "grad_norm": 0.2445985972881317,
+      "learning_rate": 7.802803029808492e-05,
+      "loss": 0.1869,
+      "step": 16515
+    },
+    {
+      "epoch": 2.0806121485026923,
+      "grad_norm": 0.21021974086761475,
+      "learning_rate": 7.793160661695867e-05,
+      "loss": 0.1778,
+      "step": 16520
+    },
+    {
+      "epoch": 2.08124193091287,
+      "grad_norm": 0.20149335265159607,
+      "learning_rate": 7.783522163820587e-05,
+      "loss": 0.1685,
+      "step": 16525
+    },
+    {
+      "epoch": 2.081871713323047,
+      "grad_norm": 0.2342994064092636,
+      "learning_rate": 7.773887541358749e-05,
+      "loss": 0.1714,
+      "step": 16530
+    },
+    {
+      "epoch": 2.082501495733224,
+      "grad_norm": 0.2518448829650879,
+      "learning_rate": 7.764256799484364e-05,
+      "loss": 0.1899,
+      "step": 16535
+    },
+    {
+      "epoch": 2.0831312781434015,
+      "grad_norm": 0.22891752421855927,
+      "learning_rate": 7.754629943369365e-05,
+      "loss": 0.1724,
+      "step": 16540
+    },
+    {
+      "epoch": 2.0837610605535786,
+      "grad_norm": 0.2348988950252533,
+      "learning_rate": 7.74500697818358e-05,
+      "loss": 0.1772,
+      "step": 16545
+    },
+    {
+      "epoch": 2.084390842963756,
+      "grad_norm": 0.21126072108745575,
+      "learning_rate": 7.735387909094772e-05,
+      "loss": 0.182,
+      "step": 16550
+    },
+    {
+      "epoch": 2.085020625373933,
+      "grad_norm": 0.2134072482585907,
+      "learning_rate": 7.725772741268598e-05,
+      "loss": 0.1861,
+      "step": 16555
+    },
+    {
+      "epoch": 2.0856504077841107,
+      "grad_norm": 0.22559498250484467,
+      "learning_rate": 7.716161479868623e-05,
+      "loss": 0.1745,
+      "step": 16560
+    },
+    {
+      "epoch": 2.086280190194288,
+      "grad_norm": 0.2076030671596527,
+      "learning_rate": 7.706554130056315e-05,
+      "loss": 0.1811,
+      "step": 16565
+    },
+    {
+      "epoch": 2.0869099726044653,
+      "grad_norm": 0.24279461801052094,
+      "learning_rate": 7.696950696991032e-05,
+      "loss": 0.1829,
+      "step": 16570
+    },
+    {
+      "epoch": 2.0875397550146424,
+      "grad_norm": 0.21790249645709991,
+      "learning_rate": 7.687351185830058e-05,
+      "loss": 0.1835,
+      "step": 16575
+    },
+    {
+      "epoch": 2.08816953742482,
+      "grad_norm": 0.2210235744714737,
+      "learning_rate": 7.677755601728527e-05,
+      "loss": 0.1678,
+      "step": 16580
+    },
+    {
+      "epoch": 2.088799319834997,
+      "grad_norm": 0.21354030072689056,
+      "learning_rate": 7.668163949839492e-05,
+      "loss": 0.1863,
+      "step": 16585
+    },
+    {
+      "epoch": 2.089429102245174,
+      "grad_norm": 0.264240026473999,
+      "learning_rate": 7.658576235313896e-05,
+      "loss": 0.1879,
+      "step": 16590
+    },
+    {
+      "epoch": 2.0900588846553516,
+      "grad_norm": 0.2348974198102951,
+      "learning_rate": 7.648992463300561e-05,
+      "loss": 0.1796,
+      "step": 16595
+    },
+    {
+      "epoch": 2.0906886670655287,
+      "grad_norm": 0.23128418624401093,
+      "learning_rate": 7.639412638946186e-05,
+      "loss": 0.1793,
+      "step": 16600
+    },
+    {
+      "epoch": 2.091318449475706,
+      "grad_norm": 0.2405007928609848,
+      "learning_rate": 7.629836767395359e-05,
+      "loss": 0.1856,
+      "step": 16605
+    },
+    {
+      "epoch": 2.0919482318858833,
+      "grad_norm": 0.23123788833618164,
+      "learning_rate": 7.620264853790539e-05,
+      "loss": 0.1752,
+      "step": 16610
+    },
+    {
+      "epoch": 2.092578014296061,
+      "grad_norm": 0.22082751989364624,
+      "learning_rate": 7.610696903272062e-05,
+      "loss": 0.1731,
+      "step": 16615
+    },
+    {
+      "epoch": 2.093207796706238,
+      "grad_norm": 0.23356421291828156,
+      "learning_rate": 7.601132920978139e-05,
+      "loss": 0.1839,
+      "step": 16620
+    },
+    {
+      "epoch": 2.0938375791164154,
+      "grad_norm": 0.2418486326932907,
+      "learning_rate": 7.591572912044846e-05,
+      "loss": 0.1883,
+      "step": 16625
+    },
+    {
+      "epoch": 2.0944673615265925,
+      "grad_norm": 0.2357870191335678,
+      "learning_rate": 7.58201688160612e-05,
+      "loss": 0.176,
+      "step": 16630
+    },
+    {
+      "epoch": 2.09509714393677,
+      "grad_norm": 0.27169832587242126,
+      "learning_rate": 7.572464834793778e-05,
+      "loss": 0.1824,
+      "step": 16635
+    },
+    {
+      "epoch": 2.095726926346947,
+      "grad_norm": 0.23245801031589508,
+      "learning_rate": 7.562916776737488e-05,
+      "loss": 0.1937,
+      "step": 16640
+    },
+    {
+      "epoch": 2.096356708757124,
+      "grad_norm": 0.2312193661928177,
+      "learning_rate": 7.55337271256476e-05,
+      "loss": 0.1873,
+      "step": 16645
+    },
+    {
+      "epoch": 2.0969864911673017,
+      "grad_norm": 0.2394751012325287,
+      "learning_rate": 7.543832647400989e-05,
+      "loss": 0.1748,
+      "step": 16650
+    },
+    {
+      "epoch": 2.097616273577479,
+      "grad_norm": 0.2679862976074219,
+      "learning_rate": 7.534296586369402e-05,
+      "loss": 0.1868,
+      "step": 16655
+    },
+    {
+      "epoch": 2.0982460559876563,
+      "grad_norm": 0.2397966831922531,
+      "learning_rate": 7.524764534591086e-05,
+      "loss": 0.1768,
+      "step": 16660
+    },
+    {
+      "epoch": 2.0988758383978334,
+      "grad_norm": 0.22550681233406067,
+      "learning_rate": 7.515236497184965e-05,
+      "loss": 0.1764,
+      "step": 16665
+    },
+    {
+      "epoch": 2.099505620808011,
+      "grad_norm": 0.23124639689922333,
+      "learning_rate": 7.505712479267809e-05,
+      "loss": 0.1828,
+      "step": 16670
+    },
+    {
+      "epoch": 2.100135403218188,
+      "grad_norm": 0.2034096121788025,
+      "learning_rate": 7.496192485954254e-05,
+      "loss": 0.179,
+      "step": 16675
+    },
+    {
+      "epoch": 2.1007651856283656,
+      "grad_norm": 0.2237498164176941,
+      "learning_rate": 7.486676522356732e-05,
+      "loss": 0.1867,
+      "step": 16680
+    },
+    {
+      "epoch": 2.1013949680385426,
+      "grad_norm": 0.22583693265914917,
+      "learning_rate": 7.477164593585537e-05,
+      "loss": 0.1882,
+      "step": 16685
+    },
+    {
+      "epoch": 2.10202475044872,
+      "grad_norm": 0.20145735144615173,
+      "learning_rate": 7.467656704748792e-05,
+      "loss": 0.1749,
+      "step": 16690
+    },
+    {
+      "epoch": 2.1026545328588973,
+      "grad_norm": 0.204311341047287,
+      "learning_rate": 7.458152860952458e-05,
+      "loss": 0.1803,
+      "step": 16695
+    },
+    {
+      "epoch": 2.1032843152690743,
+      "grad_norm": 0.23768644034862518,
+      "learning_rate": 7.448653067300313e-05,
+      "loss": 0.1915,
+      "step": 16700
+    },
+    {
+      "epoch": 2.103914097679252,
+      "grad_norm": 0.21348991990089417,
+      "learning_rate": 7.439157328893961e-05,
+      "loss": 0.1778,
+      "step": 16705
+    },
+    {
+      "epoch": 2.104543880089429,
+      "grad_norm": 0.22427400946617126,
+      "learning_rate": 7.429665650832831e-05,
+      "loss": 0.1712,
+      "step": 16710
+    },
+    {
+      "epoch": 2.1051736624996065,
+      "grad_norm": 0.22512148320674896,
+      "learning_rate": 7.420178038214172e-05,
+      "loss": 0.1889,
+      "step": 16715
+    },
+    {
+      "epoch": 2.1058034449097836,
+      "grad_norm": 0.22715777158737183,
+      "learning_rate": 7.410694496133048e-05,
+      "loss": 0.1737,
+      "step": 16720
+    },
+    {
+      "epoch": 2.106433227319961,
+      "grad_norm": 0.2505483627319336,
+      "learning_rate": 7.401215029682339e-05,
+      "loss": 0.1809,
+      "step": 16725
+    },
+    {
+      "epoch": 2.107063009730138,
+      "grad_norm": 0.2218826860189438,
+      "learning_rate": 7.391739643952725e-05,
+      "loss": 0.1766,
+      "step": 16730
+    },
+    {
+      "epoch": 2.1076927921403157,
+      "grad_norm": 0.2085668295621872,
+      "learning_rate": 7.38226834403272e-05,
+      "loss": 0.1739,
+      "step": 16735
+    },
+    {
+      "epoch": 2.1083225745504928,
+      "grad_norm": 0.21690475940704346,
+      "learning_rate": 7.372801135008622e-05,
+      "loss": 0.1738,
+      "step": 16740
+    },
+    {
+      "epoch": 2.1089523569606703,
+      "grad_norm": 0.263988733291626,
+      "learning_rate": 7.363338021964545e-05,
+      "loss": 0.1951,
+      "step": 16745
+    },
+    {
+      "epoch": 2.1095821393708474,
+      "grad_norm": 0.24228844046592712,
+      "learning_rate": 7.353879009982377e-05,
+      "loss": 0.1775,
+      "step": 16750
+    },
+    {
+      "epoch": 2.1102119217810245,
+      "grad_norm": 0.2030615508556366,
+      "learning_rate": 7.344424104141843e-05,
+      "loss": 0.1754,
+      "step": 16755
+    },
+    {
+      "epoch": 2.110841704191202,
+      "grad_norm": 0.22505883872509003,
+      "learning_rate": 7.334973309520438e-05,
+      "loss": 0.1814,
+      "step": 16760
+    },
+    {
+      "epoch": 2.111471486601379,
+      "grad_norm": 0.28446871042251587,
+      "learning_rate": 7.32552663119345e-05,
+      "loss": 0.2009,
+      "step": 16765
+    },
+    {
+      "epoch": 2.1121012690115566,
+      "grad_norm": 0.2320084124803543,
+      "learning_rate": 7.316084074233968e-05,
+      "loss": 0.1866,
+      "step": 16770
+    },
+    {
+      "epoch": 2.1127310514217337,
+      "grad_norm": 0.23432306945323944,
+      "learning_rate": 7.306645643712851e-05,
+      "loss": 0.1838,
+      "step": 16775
+    },
+    {
+      "epoch": 2.113360833831911,
+      "grad_norm": 0.20252206921577454,
+      "learning_rate": 7.297211344698769e-05,
+      "loss": 0.1753,
+      "step": 16780
+    },
+    {
+      "epoch": 2.1139906162420883,
+      "grad_norm": 0.25251004099845886,
+      "learning_rate": 7.28778118225814e-05,
+      "loss": 0.1836,
+      "step": 16785
+    },
+    {
+      "epoch": 2.114620398652266,
+      "grad_norm": 0.2514311373233795,
+      "learning_rate": 7.278355161455176e-05,
+      "loss": 0.1838,
+      "step": 16790
+    },
+    {
+      "epoch": 2.115250181062443,
+      "grad_norm": 0.21513232588768005,
+      "learning_rate": 7.268933287351876e-05,
+      "loss": 0.1745,
+      "step": 16795
+    },
+    {
+      "epoch": 2.1158799634726204,
+      "grad_norm": 0.2200087606906891,
+      "learning_rate": 7.259515565007999e-05,
+      "loss": 0.1839,
+      "step": 16800
+    },
+    {
+      "epoch": 2.1165097458827975,
+      "grad_norm": 0.22383321821689606,
+      "learning_rate": 7.250101999481073e-05,
+      "loss": 0.1865,
+      "step": 16805
+    },
+    {
+      "epoch": 2.1171395282929746,
+      "grad_norm": 0.2382001131772995,
+      "learning_rate": 7.2406925958264e-05,
+      "loss": 0.1862,
+      "step": 16810
+    },
+    {
+      "epoch": 2.117769310703152,
+      "grad_norm": 0.2178415209054947,
+      "learning_rate": 7.231287359097045e-05,
+      "loss": 0.1799,
+      "step": 16815
+    },
+    {
+      "epoch": 2.118399093113329,
+      "grad_norm": 0.22616611421108246,
+      "learning_rate": 7.221886294343834e-05,
+      "loss": 0.1819,
+      "step": 16820
+    },
+    {
+      "epoch": 2.1190288755235067,
+      "grad_norm": 0.24810658395290375,
+      "learning_rate": 7.212489406615355e-05,
+      "loss": 0.181,
+      "step": 16825
+    },
+    {
+      "epoch": 2.119658657933684,
+      "grad_norm": 0.2408507764339447,
+      "learning_rate": 7.20309670095795e-05,
+      "loss": 0.1867,
+      "step": 16830
+    },
+    {
+      "epoch": 2.1202884403438613,
+      "grad_norm": 0.20721390843391418,
+      "learning_rate": 7.19370818241571e-05,
+      "loss": 0.175,
+      "step": 16835
+    },
+    {
+      "epoch": 2.1209182227540384,
+      "grad_norm": 0.22691728174686432,
+      "learning_rate": 7.184323856030497e-05,
+      "loss": 0.1753,
+      "step": 16840
+    },
+    {
+      "epoch": 2.121548005164216,
+      "grad_norm": 0.22788456082344055,
+      "learning_rate": 7.174943726841902e-05,
+      "loss": 0.1829,
+      "step": 16845
+    },
+    {
+      "epoch": 2.122177787574393,
+      "grad_norm": 0.21744227409362793,
+      "learning_rate": 7.165567799887268e-05,
+      "loss": 0.1797,
+      "step": 16850
+    },
+    {
+      "epoch": 2.1228075699845705,
+      "grad_norm": 0.211074560880661,
+      "learning_rate": 7.156196080201685e-05,
+      "loss": 0.1875,
+      "step": 16855
+    },
+    {
+      "epoch": 2.1234373523947476,
+      "grad_norm": 0.27859583497047424,
+      "learning_rate": 7.146828572817975e-05,
+      "loss": 0.1791,
+      "step": 16860
+    },
+    {
+      "epoch": 2.1240671348049247,
+      "grad_norm": 0.202862948179245,
+      "learning_rate": 7.13746528276671e-05,
+      "loss": 0.1752,
+      "step": 16865
+    },
+    {
+      "epoch": 2.1246969172151022,
+      "grad_norm": 0.2529730498790741,
+      "learning_rate": 7.128106215076187e-05,
+      "loss": 0.1734,
+      "step": 16870
+    },
+    {
+      "epoch": 2.1253266996252793,
+      "grad_norm": 0.22796177864074707,
+      "learning_rate": 7.118751374772433e-05,
+      "loss": 0.1807,
+      "step": 16875
+    },
+    {
+      "epoch": 2.125956482035457,
+      "grad_norm": 0.20112904906272888,
+      "learning_rate": 7.109400766879223e-05,
+      "loss": 0.1711,
+      "step": 16880
+    },
+    {
+      "epoch": 2.126586264445634,
+      "grad_norm": 0.22492708265781403,
+      "learning_rate": 7.100054396418048e-05,
+      "loss": 0.1784,
+      "step": 16885
+    },
+    {
+      "epoch": 2.1272160468558114,
+      "grad_norm": 0.25224363803863525,
+      "learning_rate": 7.09071226840811e-05,
+      "loss": 0.185,
+      "step": 16890
+    },
+    {
+      "epoch": 2.1278458292659885,
+      "grad_norm": 0.24734210968017578,
+      "learning_rate": 7.081374387866346e-05,
+      "loss": 0.1739,
+      "step": 16895
+    },
+    {
+      "epoch": 2.128475611676166,
+      "grad_norm": 0.21726474165916443,
+      "learning_rate": 7.07204075980742e-05,
+      "loss": 0.1695,
+      "step": 16900
+    },
+    {
+      "epoch": 2.129105394086343,
+      "grad_norm": 0.2073916345834732,
+      "learning_rate": 7.062711389243703e-05,
+      "loss": 0.1782,
+      "step": 16905
+    },
+    {
+      "epoch": 2.1297351764965207,
+      "grad_norm": 0.2361113578081131,
+      "learning_rate": 7.053386281185274e-05,
+      "loss": 0.1787,
+      "step": 16910
+    },
+    {
+      "epoch": 2.1303649589066977,
+      "grad_norm": 0.22586499154567719,
+      "learning_rate": 7.044065440639933e-05,
+      "loss": 0.1738,
+      "step": 16915
+    },
+    {
+      "epoch": 2.130994741316875,
+      "grad_norm": 0.23469188809394836,
+      "learning_rate": 7.034748872613184e-05,
+      "loss": 0.1805,
+      "step": 16920
+    },
+    {
+      "epoch": 2.1316245237270524,
+      "grad_norm": 0.1897682100534439,
+      "learning_rate": 7.025436582108234e-05,
+      "loss": 0.171,
+      "step": 16925
+    },
+    {
+      "epoch": 2.1322543061372294,
+      "grad_norm": 0.22100795805454254,
+      "learning_rate": 7.016128574126e-05,
+      "loss": 0.1736,
+      "step": 16930
+    },
+    {
+      "epoch": 2.132884088547407,
+      "grad_norm": 0.2332223504781723,
+      "learning_rate": 7.006824853665085e-05,
+      "loss": 0.1729,
+      "step": 16935
+    },
+    {
+      "epoch": 2.133513870957584,
+      "grad_norm": 0.23929065465927124,
+      "learning_rate": 6.997525425721814e-05,
+      "loss": 0.1736,
+      "step": 16940
+    },
+    {
+      "epoch": 2.1341436533677616,
+      "grad_norm": 0.26240813732147217,
+      "learning_rate": 6.988230295290185e-05,
+      "loss": 0.1798,
+      "step": 16945
+    },
+    {
+      "epoch": 2.1347734357779387,
+      "grad_norm": 0.22387517988681793,
+      "learning_rate": 6.978939467361895e-05,
+      "loss": 0.1734,
+      "step": 16950
+    },
+    {
+      "epoch": 2.135403218188116,
+      "grad_norm": 0.246952623128891,
+      "learning_rate": 6.969652946926332e-05,
+      "loss": 0.1834,
+      "step": 16955
+    },
+    {
+      "epoch": 2.1360330005982933,
+      "grad_norm": 0.25226834416389465,
+      "learning_rate": 6.960370738970568e-05,
+      "loss": 0.1798,
+      "step": 16960
+    },
+    {
+      "epoch": 2.136662783008471,
+      "grad_norm": 0.22118602693080902,
+      "learning_rate": 6.951092848479364e-05,
+      "loss": 0.1863,
+      "step": 16965
+    },
+    {
+      "epoch": 2.137292565418648,
+      "grad_norm": 0.2567583918571472,
+      "learning_rate": 6.941819280435155e-05,
+      "loss": 0.1828,
+      "step": 16970
+    },
+    {
+      "epoch": 2.137922347828825,
+      "grad_norm": 0.28791603446006775,
+      "learning_rate": 6.93255003981806e-05,
+      "loss": 0.1817,
+      "step": 16975
+    },
+    {
+      "epoch": 2.1385521302390025,
+      "grad_norm": 0.2655430734157562,
+      "learning_rate": 6.923285131605871e-05,
+      "loss": 0.1789,
+      "step": 16980
+    },
+    {
+      "epoch": 2.1391819126491796,
+      "grad_norm": 0.24513307213783264,
+      "learning_rate": 6.914024560774061e-05,
+      "loss": 0.1885,
+      "step": 16985
+    },
+    {
+      "epoch": 2.139811695059357,
+      "grad_norm": 0.211643248796463,
+      "learning_rate": 6.904768332295772e-05,
+      "loss": 0.188,
+      "step": 16990
+    },
+    {
+      "epoch": 2.140441477469534,
+      "grad_norm": 0.2373894900083542,
+      "learning_rate": 6.895516451141791e-05,
+      "loss": 0.1819,
+      "step": 16995
+    },
+    {
+      "epoch": 2.1410712598797117,
+      "grad_norm": 0.22991600632667542,
+      "learning_rate": 6.88626892228061e-05,
+      "loss": 0.189,
+      "step": 17000
+    },
+    {
+      "epoch": 2.1410712598797117,
+      "eval_loss": 0.3501429557800293,
+      "eval_runtime": 6.1606,
+      "eval_samples_per_second": 162.322,
+      "eval_steps_per_second": 10.226,
+      "step": 17000
+    },
+    {
+      "epoch": 2.141701042289889,
+      "grad_norm": 0.23578788340091705,
+      "learning_rate": 6.877025750678352e-05,
+      "loss": 0.1804,
+      "step": 17005
+    },
+    {
+      "epoch": 2.1423308247000663,
+      "grad_norm": 0.20814631879329681,
+      "learning_rate": 6.867786941298816e-05,
+      "loss": 0.1776,
+      "step": 17010
+    },
+    {
+      "epoch": 2.1429606071102434,
+      "grad_norm": 0.24113385379314423,
+      "learning_rate": 6.858552499103451e-05,
+      "loss": 0.171,
+      "step": 17015
+    },
+    {
+      "epoch": 2.143590389520421,
+      "grad_norm": 0.2317270189523697,
+      "learning_rate": 6.84932242905136e-05,
+      "loss": 0.1881,
+      "step": 17020
+    },
+    {
+      "epoch": 2.144220171930598,
+      "grad_norm": 0.26681753993034363,
+      "learning_rate": 6.840096736099314e-05,
+      "loss": 0.1792,
+      "step": 17025
+    },
+    {
+      "epoch": 2.144849954340775,
+      "grad_norm": 0.2119479924440384,
+      "learning_rate": 6.83087542520171e-05,
+      "loss": 0.178,
+      "step": 17030
+    },
+    {
+      "epoch": 2.1454797367509526,
+      "grad_norm": 0.20759105682373047,
+      "learning_rate": 6.821658501310604e-05,
+      "loss": 0.1754,
+      "step": 17035
+    },
+    {
+      "epoch": 2.1461095191611297,
+      "grad_norm": 0.23515643179416656,
+      "learning_rate": 6.812445969375691e-05,
+      "loss": 0.1854,
+      "step": 17040
+    },
+    {
+      "epoch": 2.146739301571307,
+      "grad_norm": 0.20694191753864288,
+      "learning_rate": 6.803237834344322e-05,
+      "loss": 0.1801,
+      "step": 17045
+    },
+    {
+      "epoch": 2.1473690839814843,
+      "grad_norm": 0.21541932225227356,
+      "learning_rate": 6.794034101161469e-05,
+      "loss": 0.1752,
+      "step": 17050
+    },
+    {
+      "epoch": 2.147998866391662,
+      "grad_norm": 0.20586980879306793,
+      "learning_rate": 6.784834774769748e-05,
+      "loss": 0.1803,
+      "step": 17055
+    },
+    {
+      "epoch": 2.148628648801839,
+      "grad_norm": 0.23750190436840057,
+      "learning_rate": 6.775639860109406e-05,
+      "loss": 0.1842,
+      "step": 17060
+    },
+    {
+      "epoch": 2.1492584312120164,
+      "grad_norm": 0.2041424959897995,
+      "learning_rate": 6.766449362118324e-05,
+      "loss": 0.1729,
+      "step": 17065
+    },
+    {
+      "epoch": 2.1498882136221935,
+      "grad_norm": 0.24630430340766907,
+      "learning_rate": 6.757263285732009e-05,
+      "loss": 0.1821,
+      "step": 17070
+    },
+    {
+      "epoch": 2.150517996032371,
+      "grad_norm": 0.23113587498664856,
+      "learning_rate": 6.748081635883594e-05,
+      "loss": 0.1821,
+      "step": 17075
+    },
+    {
+      "epoch": 2.151147778442548,
+      "grad_norm": 0.203240305185318,
+      "learning_rate": 6.738904417503829e-05,
+      "loss": 0.1767,
+      "step": 17080
+    },
+    {
+      "epoch": 2.151777560852725,
+      "grad_norm": 0.2500320374965668,
+      "learning_rate": 6.7297316355211e-05,
+      "loss": 0.1852,
+      "step": 17085
+    },
+    {
+      "epoch": 2.1524073432629027,
+      "grad_norm": 0.2349621206521988,
+      "learning_rate": 6.720563294861403e-05,
+      "loss": 0.1764,
+      "step": 17090
+    },
+    {
+      "epoch": 2.15303712567308,
+      "grad_norm": 0.2351408451795578,
+      "learning_rate": 6.71139940044833e-05,
+      "loss": 0.1835,
+      "step": 17095
+    },
+    {
+      "epoch": 2.1536669080832573,
+      "grad_norm": 0.2078278511762619,
+      "learning_rate": 6.702239957203108e-05,
+      "loss": 0.1783,
+      "step": 17100
+    },
+    {
+      "epoch": 2.1542966904934344,
+      "grad_norm": 0.23805204033851624,
+      "learning_rate": 6.693084970044574e-05,
+      "loss": 0.1858,
+      "step": 17105
+    },
+    {
+      "epoch": 2.154926472903612,
+      "grad_norm": 0.22789132595062256,
+      "learning_rate": 6.683934443889161e-05,
+      "loss": 0.1839,
+      "step": 17110
+    },
+    {
+      "epoch": 2.155556255313789,
+      "grad_norm": 0.27035263180732727,
+      "learning_rate": 6.674788383650911e-05,
+      "loss": 0.1878,
+      "step": 17115
+    },
+    {
+      "epoch": 2.1561860377239666,
+      "grad_norm": 0.21787506341934204,
+      "learning_rate": 6.665646794241468e-05,
+      "loss": 0.1854,
+      "step": 17120
+    },
+    {
+      "epoch": 2.1568158201341436,
+      "grad_norm": 0.2302270233631134,
+      "learning_rate": 6.656509680570073e-05,
+      "loss": 0.1822,
+      "step": 17125
+    },
+    {
+      "epoch": 2.157445602544321,
+      "grad_norm": 0.21228045225143433,
+      "learning_rate": 6.647377047543563e-05,
+      "loss": 0.1855,
+      "step": 17130
+    },
+    {
+      "epoch": 2.1580753849544982,
+      "grad_norm": 0.22131386399269104,
+      "learning_rate": 6.638248900066375e-05,
+      "loss": 0.1763,
+      "step": 17135
+    },
+    {
+      "epoch": 2.1587051673646753,
+      "grad_norm": 0.2691584527492523,
+      "learning_rate": 6.629125243040524e-05,
+      "loss": 0.1815,
+      "step": 17140
+    },
+    {
+      "epoch": 2.159334949774853,
+      "grad_norm": 0.22926035523414612,
+      "learning_rate": 6.620006081365634e-05,
+      "loss": 0.1833,
+      "step": 17145
+    },
+    {
+      "epoch": 2.15996473218503,
+      "grad_norm": 0.20654956996440887,
+      "learning_rate": 6.610891419938899e-05,
+      "loss": 0.1755,
+      "step": 17150
+    },
+    {
+      "epoch": 2.1605945145952075,
+      "grad_norm": 0.22390377521514893,
+      "learning_rate": 6.601781263655096e-05,
+      "loss": 0.1839,
+      "step": 17155
+    },
+    {
+      "epoch": 2.1612242970053845,
+      "grad_norm": 0.23877164721488953,
+      "learning_rate": 6.592675617406593e-05,
+      "loss": 0.1739,
+      "step": 17160
+    },
+    {
+      "epoch": 2.161854079415562,
+      "grad_norm": 0.24347762763500214,
+      "learning_rate": 6.583574486083325e-05,
+      "loss": 0.1863,
+      "step": 17165
+    },
+    {
+      "epoch": 2.162483861825739,
+      "grad_norm": 0.23407521843910217,
+      "learning_rate": 6.574477874572811e-05,
+      "loss": 0.1741,
+      "step": 17170
+    },
+    {
+      "epoch": 2.1631136442359167,
+      "grad_norm": 0.23338505625724792,
+      "learning_rate": 6.565385787760137e-05,
+      "loss": 0.1754,
+      "step": 17175
+    },
+    {
+      "epoch": 2.1637434266460938,
+      "grad_norm": 0.2206541895866394,
+      "learning_rate": 6.556298230527962e-05,
+      "loss": 0.1706,
+      "step": 17180
+    },
+    {
+      "epoch": 2.1643732090562713,
+      "grad_norm": 0.20819810032844543,
+      "learning_rate": 6.547215207756504e-05,
+      "loss": 0.1735,
+      "step": 17185
+    },
+    {
+      "epoch": 2.1650029914664484,
+      "grad_norm": 0.22891941666603088,
+      "learning_rate": 6.53813672432357e-05,
+      "loss": 0.187,
+      "step": 17190
+    },
+    {
+      "epoch": 2.1656327738766254,
+      "grad_norm": 0.2094859778881073,
+      "learning_rate": 6.52906278510451e-05,
+      "loss": 0.1795,
+      "step": 17195
+    },
+    {
+      "epoch": 2.166262556286803,
+      "grad_norm": 0.20969723165035248,
+      "learning_rate": 6.519993394972219e-05,
+      "loss": 0.1679,
+      "step": 17200
+    },
+    {
+      "epoch": 2.16689233869698,
+      "grad_norm": 0.25252285599708557,
+      "learning_rate": 6.510928558797185e-05,
+      "loss": 0.183,
+      "step": 17205
+    },
+    {
+      "epoch": 2.1675221211071576,
+      "grad_norm": 0.22556447982788086,
+      "learning_rate": 6.501868281447424e-05,
+      "loss": 0.1694,
+      "step": 17210
+    },
+    {
+      "epoch": 2.1681519035173347,
+      "grad_norm": 0.2429586797952652,
+      "learning_rate": 6.492812567788516e-05,
+      "loss": 0.18,
+      "step": 17215
+    },
+    {
+      "epoch": 2.168781685927512,
+      "grad_norm": 0.2400483787059784,
+      "learning_rate": 6.483761422683582e-05,
+      "loss": 0.1818,
+      "step": 17220
+    },
+    {
+      "epoch": 2.1694114683376893,
+      "grad_norm": 0.228154718875885,
+      "learning_rate": 6.47471485099329e-05,
+      "loss": 0.1744,
+      "step": 17225
+    },
+    {
+      "epoch": 2.170041250747867,
+      "grad_norm": 0.21748559176921844,
+      "learning_rate": 6.465672857575875e-05,
+      "loss": 0.1765,
+      "step": 17230
+    },
+    {
+      "epoch": 2.170671033158044,
+      "grad_norm": 0.2296319603919983,
+      "learning_rate": 6.456635447287073e-05,
+      "loss": 0.1881,
+      "step": 17235
+    },
+    {
+      "epoch": 2.1713008155682214,
+      "grad_norm": 0.2402602881193161,
+      "learning_rate": 6.447602624980186e-05,
+      "loss": 0.1769,
+      "step": 17240
+    },
+    {
+      "epoch": 2.1719305979783985,
+      "grad_norm": 0.2783866226673126,
+      "learning_rate": 6.438574395506043e-05,
+      "loss": 0.1836,
+      "step": 17245
+    },
+    {
+      "epoch": 2.1725603803885756,
+      "grad_norm": 0.20301677286624908,
+      "learning_rate": 6.429550763713017e-05,
+      "loss": 0.1655,
+      "step": 17250
+    },
+    {
+      "epoch": 2.173190162798753,
+      "grad_norm": 0.21163971722126007,
+      "learning_rate": 6.420531734447e-05,
+      "loss": 0.1764,
+      "step": 17255
+    },
+    {
+      "epoch": 2.17381994520893,
+      "grad_norm": 0.24942253530025482,
+      "learning_rate": 6.41151731255142e-05,
+      "loss": 0.1833,
+      "step": 17260
+    },
+    {
+      "epoch": 2.1744497276191077,
+      "grad_norm": 0.22958967089653015,
+      "learning_rate": 6.402507502867222e-05,
+      "loss": 0.1703,
+      "step": 17265
+    },
+    {
+      "epoch": 2.175079510029285,
+      "grad_norm": 0.21424312889575958,
+      "learning_rate": 6.393502310232886e-05,
+      "loss": 0.1757,
+      "step": 17270
+    },
+    {
+      "epoch": 2.1757092924394623,
+      "grad_norm": 0.20825864374637604,
+      "learning_rate": 6.384501739484401e-05,
+      "loss": 0.1715,
+      "step": 17275
+    },
+    {
+      "epoch": 2.1763390748496394,
+      "grad_norm": 0.21387939155101776,
+      "learning_rate": 6.375505795455281e-05,
+      "loss": 0.1697,
+      "step": 17280
+    },
+    {
+      "epoch": 2.176968857259817,
+      "grad_norm": 0.2073564976453781,
+      "learning_rate": 6.366514482976546e-05,
+      "loss": 0.1846,
+      "step": 17285
+    },
+    {
+      "epoch": 2.177598639669994,
+      "grad_norm": 0.21405762434005737,
+      "learning_rate": 6.35752780687675e-05,
+      "loss": 0.1777,
+      "step": 17290
+    },
+    {
+      "epoch": 2.1782284220801715,
+      "grad_norm": 0.22343981266021729,
+      "learning_rate": 6.348545771981938e-05,
+      "loss": 0.1801,
+      "step": 17295
+    },
+    {
+      "epoch": 2.1788582044903486,
+      "grad_norm": 0.22697073221206665,
+      "learning_rate": 6.339568383115668e-05,
+      "loss": 0.1829,
+      "step": 17300
+    },
+    {
+      "epoch": 2.1794879869005257,
+      "grad_norm": 0.2561056613922119,
+      "learning_rate": 6.330595645098996e-05,
+      "loss": 0.185,
+      "step": 17305
+    },
+    {
+      "epoch": 2.1801177693107032,
+      "grad_norm": 0.2563771903514862,
+      "learning_rate": 6.321627562750495e-05,
+      "loss": 0.1752,
+      "step": 17310
+    },
+    {
+      "epoch": 2.1807475517208803,
+      "grad_norm": 0.21171104907989502,
+      "learning_rate": 6.312664140886228e-05,
+      "loss": 0.166,
+      "step": 17315
+    },
+    {
+      "epoch": 2.181377334131058,
+      "grad_norm": 0.23899543285369873,
+      "learning_rate": 6.303705384319757e-05,
+      "loss": 0.1828,
+      "step": 17320
+    },
+    {
+      "epoch": 2.182007116541235,
+      "grad_norm": 0.26108884811401367,
+      "learning_rate": 6.29475129786214e-05,
+      "loss": 0.1829,
+      "step": 17325
+    },
+    {
+      "epoch": 2.1826368989514124,
+      "grad_norm": 0.2397276908159256,
+      "learning_rate": 6.285801886321919e-05,
+      "loss": 0.1733,
+      "step": 17330
+    },
+    {
+      "epoch": 2.1832666813615895,
+      "grad_norm": 0.22638286650180817,
+      "learning_rate": 6.27685715450515e-05,
+      "loss": 0.1719,
+      "step": 17335
+    },
+    {
+      "epoch": 2.183896463771767,
+      "grad_norm": 0.2424623966217041,
+      "learning_rate": 6.26791710721534e-05,
+      "loss": 0.1749,
+      "step": 17340
+    },
+    {
+      "epoch": 2.184526246181944,
+      "grad_norm": 0.23895704746246338,
+      "learning_rate": 6.2589817492535e-05,
+      "loss": 0.178,
+      "step": 17345
+    },
+    {
+      "epoch": 2.1851560285921217,
+      "grad_norm": 0.2223139852285385,
+      "learning_rate": 6.250051085418133e-05,
+      "loss": 0.1872,
+      "step": 17350
+    },
+    {
+      "epoch": 2.1857858110022987,
+      "grad_norm": 0.22255347669124603,
+      "learning_rate": 6.241125120505204e-05,
+      "loss": 0.1791,
+      "step": 17355
+    },
+    {
+      "epoch": 2.186415593412476,
+      "grad_norm": 0.23792186379432678,
+      "learning_rate": 6.232203859308157e-05,
+      "loss": 0.1738,
+      "step": 17360
+    },
+    {
+      "epoch": 2.1870453758226533,
+      "grad_norm": 0.24884961545467377,
+      "learning_rate": 6.223287306617915e-05,
+      "loss": 0.1778,
+      "step": 17365
+    },
+    {
+      "epoch": 2.1876751582328304,
+      "grad_norm": 0.2130117118358612,
+      "learning_rate": 6.214375467222873e-05,
+      "loss": 0.1666,
+      "step": 17370
+    },
+    {
+      "epoch": 2.188304940643008,
+      "grad_norm": 0.20538979768753052,
+      "learning_rate": 6.205468345908888e-05,
+      "loss": 0.1716,
+      "step": 17375
+    },
+    {
+      "epoch": 2.188934723053185,
+      "grad_norm": 0.2519354224205017,
+      "learning_rate": 6.196565947459292e-05,
+      "loss": 0.1885,
+      "step": 17380
+    },
+    {
+      "epoch": 2.1895645054633626,
+      "grad_norm": 0.2644721567630768,
+      "learning_rate": 6.187668276654872e-05,
+      "loss": 0.1923,
+      "step": 17385
+    },
+    {
+      "epoch": 2.1901942878735396,
+      "grad_norm": 0.22676245868206024,
+      "learning_rate": 6.178775338273876e-05,
+      "loss": 0.1745,
+      "step": 17390
+    },
+    {
+      "epoch": 2.190824070283717,
+      "grad_norm": 0.21329110860824585,
+      "learning_rate": 6.169887137092029e-05,
+      "loss": 0.1782,
+      "step": 17395
+    },
+    {
+      "epoch": 2.1914538526938943,
+      "grad_norm": 0.2096760869026184,
+      "learning_rate": 6.161003677882489e-05,
+      "loss": 0.1705,
+      "step": 17400
+    },
+    {
+      "epoch": 2.1920836351040713,
+      "grad_norm": 0.20192061364650726,
+      "learning_rate": 6.15212496541588e-05,
+      "loss": 0.1662,
+      "step": 17405
+    },
+    {
+      "epoch": 2.192713417514249,
+      "grad_norm": 0.2351575493812561,
+      "learning_rate": 6.14325100446027e-05,
+      "loss": 0.1716,
+      "step": 17410
+    },
+    {
+      "epoch": 2.193343199924426,
+      "grad_norm": 0.23202987015247345,
+      "learning_rate": 6.13438179978118e-05,
+      "loss": 0.1848,
+      "step": 17415
+    },
+    {
+      "epoch": 2.1939729823346035,
+      "grad_norm": 0.22229251265525818,
+      "learning_rate": 6.125517356141576e-05,
+      "loss": 0.1757,
+      "step": 17420
+    },
+    {
+      "epoch": 2.1946027647447806,
+      "grad_norm": 0.20741891860961914,
+      "learning_rate": 6.116657678301868e-05,
+      "loss": 0.1804,
+      "step": 17425
+    },
+    {
+      "epoch": 2.195232547154958,
+      "grad_norm": 0.2023356705904007,
+      "learning_rate": 6.107802771019895e-05,
+      "loss": 0.168,
+      "step": 17430
+    },
+    {
+      "epoch": 2.195862329565135,
+      "grad_norm": 0.30032244324684143,
+      "learning_rate": 6.098952639050961e-05,
+      "loss": 0.176,
+      "step": 17435
+    },
+    {
+      "epoch": 2.1964921119753127,
+      "grad_norm": 0.2093886286020279,
+      "learning_rate": 6.090107287147786e-05,
+      "loss": 0.171,
+      "step": 17440
+    },
+    {
+      "epoch": 2.1971218943854898,
+      "grad_norm": 0.20918086171150208,
+      "learning_rate": 6.081266720060517e-05,
+      "loss": 0.1705,
+      "step": 17445
+    },
+    {
+      "epoch": 2.1977516767956673,
+      "grad_norm": 0.2089412659406662,
+      "learning_rate": 6.072430942536737e-05,
+      "loss": 0.1797,
+      "step": 17450
+    },
+    {
+      "epoch": 2.1983814592058444,
+      "grad_norm": 0.2460128515958786,
+      "learning_rate": 6.0635999593214765e-05,
+      "loss": 0.1752,
+      "step": 17455
+    },
+    {
+      "epoch": 2.1990112416160215,
+      "grad_norm": 0.25952646136283875,
+      "learning_rate": 6.0547737751571654e-05,
+      "loss": 0.1784,
+      "step": 17460
+    },
+    {
+      "epoch": 2.199641024026199,
+      "grad_norm": 0.2011132687330246,
+      "learning_rate": 6.0459523947836674e-05,
+      "loss": 0.1714,
+      "step": 17465
+    },
+    {
+      "epoch": 2.200270806436376,
+      "grad_norm": 0.19077162444591522,
+      "learning_rate": 6.03713582293826e-05,
+      "loss": 0.174,
+      "step": 17470
+    },
+    {
+      "epoch": 2.2009005888465536,
+      "grad_norm": 0.22354647517204285,
+      "learning_rate": 6.02832406435566e-05,
+      "loss": 0.1754,
+      "step": 17475
+    },
+    {
+      "epoch": 2.2015303712567307,
+      "grad_norm": 0.22434799373149872,
+      "learning_rate": 6.019517123767968e-05,
+      "loss": 0.1747,
+      "step": 17480
+    },
+    {
+      "epoch": 2.202160153666908,
+      "grad_norm": 0.22911998629570007,
+      "learning_rate": 6.010715005904716e-05,
+      "loss": 0.1812,
+      "step": 17485
+    },
+    {
+      "epoch": 2.2027899360770853,
+      "grad_norm": 0.23919759690761566,
+      "learning_rate": 6.0019177154928364e-05,
+      "loss": 0.1771,
+      "step": 17490
+    },
+    {
+      "epoch": 2.203419718487263,
+      "grad_norm": 0.21539629995822906,
+      "learning_rate": 5.993125257256687e-05,
+      "loss": 0.1799,
+      "step": 17495
+    },
+    {
+      "epoch": 2.20404950089744,
+      "grad_norm": 0.22069337964057922,
+      "learning_rate": 5.984337635918014e-05,
+      "loss": 0.177,
+      "step": 17500
+    },
+    {
+      "epoch": 2.2046792833076174,
+      "grad_norm": 0.20763671398162842,
+      "learning_rate": 5.97555485619597e-05,
+      "loss": 0.1664,
+      "step": 17505
+    },
+    {
+      "epoch": 2.2053090657177945,
+      "grad_norm": 0.1950199007987976,
+      "learning_rate": 5.966776922807109e-05,
+      "loss": 0.1648,
+      "step": 17510
+    },
+    {
+      "epoch": 2.2059388481279716,
+      "grad_norm": 0.25142478942871094,
+      "learning_rate": 5.95800384046538e-05,
+      "loss": 0.1754,
+      "step": 17515
+    },
+    {
+      "epoch": 2.206568630538149,
+      "grad_norm": 0.2232702225446701,
+      "learning_rate": 5.94923561388213e-05,
+      "loss": 0.1716,
+      "step": 17520
+    },
+    {
+      "epoch": 2.207198412948326,
+      "grad_norm": 0.27657321095466614,
+      "learning_rate": 5.940472247766097e-05,
+      "loss": 0.1878,
+      "step": 17525
+    },
+    {
+      "epoch": 2.2078281953585037,
+      "grad_norm": 0.21436183154582977,
+      "learning_rate": 5.9317137468234083e-05,
+      "loss": 0.1727,
+      "step": 17530
+    },
+    {
+      "epoch": 2.208457977768681,
+      "grad_norm": 0.19741742312908173,
+      "learning_rate": 5.9229601157575744e-05,
+      "loss": 0.1694,
+      "step": 17535
+    },
+    {
+      "epoch": 2.2090877601788583,
+      "grad_norm": 0.2042321413755417,
+      "learning_rate": 5.914211359269509e-05,
+      "loss": 0.17,
+      "step": 17540
+    },
+    {
+      "epoch": 2.2097175425890354,
+      "grad_norm": 0.21126088500022888,
+      "learning_rate": 5.9054674820574814e-05,
+      "loss": 0.1703,
+      "step": 17545
+    },
+    {
+      "epoch": 2.210347324999213,
+      "grad_norm": 0.20463821291923523,
+      "learning_rate": 5.896728488817151e-05,
+      "loss": 0.172,
+      "step": 17550
+    },
+    {
+      "epoch": 2.21097710740939,
+      "grad_norm": 0.204604834318161,
+      "learning_rate": 5.887994384241569e-05,
+      "loss": 0.1723,
+      "step": 17555
+    },
+    {
+      "epoch": 2.2116068898195675,
+      "grad_norm": 0.18806815147399902,
+      "learning_rate": 5.879265173021141e-05,
+      "loss": 0.161,
+      "step": 17560
+    },
+    {
+      "epoch": 2.2122366722297446,
+      "grad_norm": 0.22745926678180695,
+      "learning_rate": 5.870540859843656e-05,
+      "loss": 0.1653,
+      "step": 17565
+    },
+    {
+      "epoch": 2.2128664546399217,
+      "grad_norm": 0.1888933777809143,
+      "learning_rate": 5.8618214493942675e-05,
+      "loss": 0.1685,
+      "step": 17570
+    },
+    {
+      "epoch": 2.2134962370500992,
+      "grad_norm": 0.19480280578136444,
+      "learning_rate": 5.853106946355501e-05,
+      "loss": 0.1676,
+      "step": 17575
+    },
+    {
+      "epoch": 2.2141260194602763,
+      "grad_norm": 0.2703428864479065,
+      "learning_rate": 5.8443973554072383e-05,
+      "loss": 0.1788,
+      "step": 17580
+    },
+    {
+      "epoch": 2.214755801870454,
+      "grad_norm": 0.21035927534103394,
+      "learning_rate": 5.8356926812267335e-05,
+      "loss": 0.1806,
+      "step": 17585
+    },
+    {
+      "epoch": 2.215385584280631,
+      "grad_norm": 0.21794281899929047,
+      "learning_rate": 5.826992928488594e-05,
+      "loss": 0.1641,
+      "step": 17590
+    },
+    {
+      "epoch": 2.2160153666908085,
+      "grad_norm": 0.2512260675430298,
+      "learning_rate": 5.818298101864779e-05,
+      "loss": 0.1697,
+      "step": 17595
+    },
+    {
+      "epoch": 2.2166451491009855,
+      "grad_norm": 0.2089598923921585,
+      "learning_rate": 5.8096082060246226e-05,
+      "loss": 0.1656,
+      "step": 17600
+    },
+    {
+      "epoch": 2.217274931511163,
+      "grad_norm": 0.2160467952489853,
+      "learning_rate": 5.80092324563479e-05,
+      "loss": 0.185,
+      "step": 17605
+    },
+    {
+      "epoch": 2.21790471392134,
+      "grad_norm": 0.20858334004878998,
+      "learning_rate": 5.7922432253593025e-05,
+      "loss": 0.1721,
+      "step": 17610
+    },
+    {
+      "epoch": 2.2185344963315172,
+      "grad_norm": 0.2090991735458374,
+      "learning_rate": 5.7835681498595327e-05,
+      "loss": 0.1706,
+      "step": 17615
+    },
+    {
+      "epoch": 2.2191642787416948,
+      "grad_norm": 0.21040284633636475,
+      "learning_rate": 5.77489802379419e-05,
+      "loss": 0.1789,
+      "step": 17620
+    },
+    {
+      "epoch": 2.219794061151872,
+      "grad_norm": 0.22497640550136566,
+      "learning_rate": 5.766232851819332e-05,
+      "loss": 0.1779,
+      "step": 17625
+    },
+    {
+      "epoch": 2.2204238435620494,
+      "grad_norm": 0.2845938801765442,
+      "learning_rate": 5.757572638588356e-05,
+      "loss": 0.1771,
+      "step": 17630
+    },
+    {
+      "epoch": 2.2210536259722264,
+      "grad_norm": 0.21166571974754333,
+      "learning_rate": 5.748917388751985e-05,
+      "loss": 0.1741,
+      "step": 17635
+    },
+    {
+      "epoch": 2.221683408382404,
+      "grad_norm": 0.26706454157829285,
+      "learning_rate": 5.7402671069583004e-05,
+      "loss": 0.1715,
+      "step": 17640
+    },
+    {
+      "epoch": 2.222313190792581,
+      "grad_norm": 0.2745297849178314,
+      "learning_rate": 5.731621797852698e-05,
+      "loss": 0.1843,
+      "step": 17645
+    },
+    {
+      "epoch": 2.2229429732027586,
+      "grad_norm": 0.2507629990577698,
+      "learning_rate": 5.7229814660778985e-05,
+      "loss": 0.186,
+      "step": 17650
+    },
+    {
+      "epoch": 2.2235727556129357,
+      "grad_norm": 0.21768365800380707,
+      "learning_rate": 5.7143461162739545e-05,
+      "loss": 0.1731,
+      "step": 17655
+    },
+    {
+      "epoch": 2.224202538023113,
+      "grad_norm": 0.22099876403808594,
+      "learning_rate": 5.705715753078259e-05,
+      "loss": 0.1802,
+      "step": 17660
+    },
+    {
+      "epoch": 2.2248323204332903,
+      "grad_norm": 0.20643608272075653,
+      "learning_rate": 5.697090381125507e-05,
+      "loss": 0.1769,
+      "step": 17665
+    },
+    {
+      "epoch": 2.2254621028434673,
+      "grad_norm": 0.2723044455051422,
+      "learning_rate": 5.688470005047722e-05,
+      "loss": 0.1882,
+      "step": 17670
+    },
+    {
+      "epoch": 2.226091885253645,
+      "grad_norm": 0.23548351228237152,
+      "learning_rate": 5.679854629474238e-05,
+      "loss": 0.1702,
+      "step": 17675
+    },
+    {
+      "epoch": 2.226721667663822,
+      "grad_norm": 0.24578404426574707,
+      "learning_rate": 5.671244259031722e-05,
+      "loss": 0.1736,
+      "step": 17680
+    },
+    {
+      "epoch": 2.2273514500739995,
+      "grad_norm": 0.21030524373054504,
+      "learning_rate": 5.662638898344125e-05,
+      "loss": 0.1711,
+      "step": 17685
+    },
+    {
+      "epoch": 2.2279812324841766,
+      "grad_norm": 0.24249999225139618,
+      "learning_rate": 5.6540385520327275e-05,
+      "loss": 0.1742,
+      "step": 17690
+    },
+    {
+      "epoch": 2.228611014894354,
+      "grad_norm": 0.23971515893936157,
+      "learning_rate": 5.645443224716106e-05,
+      "loss": 0.1655,
+      "step": 17695
+    },
+    {
+      "epoch": 2.229240797304531,
+      "grad_norm": 0.2133120596408844,
+      "learning_rate": 5.636852921010161e-05,
+      "loss": 0.1786,
+      "step": 17700
+    },
+    {
+      "epoch": 2.2298705797147087,
+      "grad_norm": 0.23475615680217743,
+      "learning_rate": 5.628267645528073e-05,
+      "loss": 0.1753,
+      "step": 17705
+    },
+    {
+      "epoch": 2.230500362124886,
+      "grad_norm": 0.22111907601356506,
+      "learning_rate": 5.619687402880332e-05,
+      "loss": 0.1617,
+      "step": 17710
+    },
+    {
+      "epoch": 2.2311301445350633,
+      "grad_norm": 0.2323450744152069,
+      "learning_rate": 5.611112197674725e-05,
+      "loss": 0.167,
+      "step": 17715
+    },
+    {
+      "epoch": 2.2317599269452404,
+      "grad_norm": 0.18698996305465698,
+      "learning_rate": 5.602542034516333e-05,
+      "loss": 0.1632,
+      "step": 17720
+    },
+    {
+      "epoch": 2.2323897093554175,
+      "grad_norm": 0.2252064198255539,
+      "learning_rate": 5.5939769180075286e-05,
+      "loss": 0.1709,
+      "step": 17725
+    },
+    {
+      "epoch": 2.233019491765595,
+      "grad_norm": 0.2561705410480499,
+      "learning_rate": 5.5854168527479756e-05,
+      "loss": 0.1826,
+      "step": 17730
+    },
+    {
+      "epoch": 2.233649274175772,
+      "grad_norm": 0.2448531985282898,
+      "learning_rate": 5.576861843334625e-05,
+      "loss": 0.1819,
+      "step": 17735
+    },
+    {
+      "epoch": 2.2342790565859496,
+      "grad_norm": 0.238671213388443,
+      "learning_rate": 5.568311894361707e-05,
+      "loss": 0.1839,
+      "step": 17740
+    },
+    {
+      "epoch": 2.2349088389961267,
+      "grad_norm": 0.22651298344135284,
+      "learning_rate": 5.5597670104207485e-05,
+      "loss": 0.172,
+      "step": 17745
+    },
+    {
+      "epoch": 2.235538621406304,
+      "grad_norm": 0.23881249129772186,
+      "learning_rate": 5.551227196100549e-05,
+      "loss": 0.1698,
+      "step": 17750
+    },
+    {
+      "epoch": 2.2361684038164813,
+      "grad_norm": 0.23065873980522156,
+      "learning_rate": 5.542692455987167e-05,
+      "loss": 0.1727,
+      "step": 17755
+    },
+    {
+      "epoch": 2.236798186226659,
+      "grad_norm": 0.19607169926166534,
+      "learning_rate": 5.534162794663969e-05,
+      "loss": 0.1719,
+      "step": 17760
+    },
+    {
+      "epoch": 2.237427968636836,
+      "grad_norm": 0.2033766806125641,
+      "learning_rate": 5.525638216711573e-05,
+      "loss": 0.171,
+      "step": 17765
+    },
+    {
+      "epoch": 2.2380577510470134,
+      "grad_norm": 0.20412589609622955,
+      "learning_rate": 5.5171187267078733e-05,
+      "loss": 0.1633,
+      "step": 17770
+    },
+    {
+      "epoch": 2.2386875334571905,
+      "grad_norm": 0.21895913779735565,
+      "learning_rate": 5.508604329228028e-05,
+      "loss": 0.1801,
+      "step": 17775
+    },
+    {
+      "epoch": 2.2393173158673676,
+      "grad_norm": 0.19198501110076904,
+      "learning_rate": 5.50009502884446e-05,
+      "loss": 0.1764,
+      "step": 17780
+    },
+    {
+      "epoch": 2.239947098277545,
+      "grad_norm": 0.21897682547569275,
+      "learning_rate": 5.4915908301268724e-05,
+      "loss": 0.1719,
+      "step": 17785
+    },
+    {
+      "epoch": 2.240576880687722,
+      "grad_norm": 0.22070536017417908,
+      "learning_rate": 5.483091737642198e-05,
+      "loss": 0.1678,
+      "step": 17790
+    },
+    {
+      "epoch": 2.2412066630978997,
+      "grad_norm": 0.2158748209476471,
+      "learning_rate": 5.474597755954651e-05,
+      "loss": 0.1703,
+      "step": 17795
+    },
+    {
+      "epoch": 2.241836445508077,
+      "grad_norm": 0.21174906194210052,
+      "learning_rate": 5.466108889625687e-05,
+      "loss": 0.1698,
+      "step": 17800
+    },
+    {
+      "epoch": 2.2424662279182543,
+      "grad_norm": 0.23331063985824585,
+      "learning_rate": 5.457625143214029e-05,
+      "loss": 0.1855,
+      "step": 17805
+    },
+    {
+      "epoch": 2.2430960103284314,
+      "grad_norm": 0.2186896651983261,
+      "learning_rate": 5.449146521275643e-05,
+      "loss": 0.1629,
+      "step": 17810
+    },
+    {
+      "epoch": 2.243725792738609,
+      "grad_norm": 0.22406966984272003,
+      "learning_rate": 5.440673028363738e-05,
+      "loss": 0.1731,
+      "step": 17815
+    },
+    {
+      "epoch": 2.244355575148786,
+      "grad_norm": 0.21894322335720062,
+      "learning_rate": 5.432204669028777e-05,
+      "loss": 0.1671,
+      "step": 17820
+    },
+    {
+      "epoch": 2.2449853575589636,
+      "grad_norm": 0.19151312112808228,
+      "learning_rate": 5.4237414478184585e-05,
+      "loss": 0.1634,
+      "step": 17825
+    },
+    {
+      "epoch": 2.2456151399691406,
+      "grad_norm": 0.20597226917743683,
+      "learning_rate": 5.415283369277729e-05,
+      "loss": 0.1594,
+      "step": 17830
+    },
+    {
+      "epoch": 2.2462449223793177,
+      "grad_norm": 0.23415236175060272,
+      "learning_rate": 5.406830437948767e-05,
+      "loss": 0.1667,
+      "step": 17835
+    },
+    {
+      "epoch": 2.2468747047894952,
+      "grad_norm": 0.21160747110843658,
+      "learning_rate": 5.398382658370986e-05,
+      "loss": 0.1694,
+      "step": 17840
+    },
+    {
+      "epoch": 2.2475044871996723,
+      "grad_norm": 0.2644958198070526,
+      "learning_rate": 5.3899400350810466e-05,
+      "loss": 0.1767,
+      "step": 17845
+    },
+    {
+      "epoch": 2.24813426960985,
+      "grad_norm": 0.23654960095882416,
+      "learning_rate": 5.381502572612826e-05,
+      "loss": 0.1684,
+      "step": 17850
+    },
+    {
+      "epoch": 2.248764052020027,
+      "grad_norm": 0.22581151127815247,
+      "learning_rate": 5.373070275497439e-05,
+      "loss": 0.1805,
+      "step": 17855
+    },
+    {
+      "epoch": 2.2493938344302045,
+      "grad_norm": 0.21524479985237122,
+      "learning_rate": 5.364643148263205e-05,
+      "loss": 0.1753,
+      "step": 17860
+    },
+    {
+      "epoch": 2.2500236168403815,
+      "grad_norm": 0.22853802144527435,
+      "learning_rate": 5.3562211954357006e-05,
+      "loss": 0.1752,
+      "step": 17865
+    },
+    {
+      "epoch": 2.250653399250559,
+      "grad_norm": 0.19708101451396942,
+      "learning_rate": 5.347804421537701e-05,
+      "loss": 0.1701,
+      "step": 17870
+    },
+    {
+      "epoch": 2.251283181660736,
+      "grad_norm": 0.22857971489429474,
+      "learning_rate": 5.339392831089209e-05,
+      "loss": 0.1662,
+      "step": 17875
+    },
+    {
+      "epoch": 2.2519129640709137,
+      "grad_norm": 0.2373005598783493,
+      "learning_rate": 5.33098642860743e-05,
+      "loss": 0.1878,
+      "step": 17880
+    },
+    {
+      "epoch": 2.2525427464810908,
+      "grad_norm": 0.22458739578723907,
+      "learning_rate": 5.322585218606811e-05,
+      "loss": 0.1711,
+      "step": 17885
+    },
+    {
+      "epoch": 2.253172528891268,
+      "grad_norm": 0.24684731662273407,
+      "learning_rate": 5.314189205598987e-05,
+      "loss": 0.1833,
+      "step": 17890
+    },
+    {
+      "epoch": 2.2538023113014454,
+      "grad_norm": 0.22604569792747498,
+      "learning_rate": 5.3057983940928046e-05,
+      "loss": 0.1683,
+      "step": 17895
+    },
+    {
+      "epoch": 2.2544320937116225,
+      "grad_norm": 0.23015649616718292,
+      "learning_rate": 5.2974127885943166e-05,
+      "loss": 0.1793,
+      "step": 17900
+    },
+    {
+      "epoch": 2.2550618761218,
+      "grad_norm": 0.2156984657049179,
+      "learning_rate": 5.289032393606797e-05,
+      "loss": 0.1816,
+      "step": 17905
+    },
+    {
+      "epoch": 2.255691658531977,
+      "grad_norm": 0.2468300610780716,
+      "learning_rate": 5.280657213630704e-05,
+      "loss": 0.1795,
+      "step": 17910
+    },
+    {
+      "epoch": 2.2563214409421546,
+      "grad_norm": 0.19326730072498322,
+      "learning_rate": 5.2722872531637024e-05,
+      "loss": 0.1726,
+      "step": 17915
+    },
+    {
+      "epoch": 2.2569512233523317,
+      "grad_norm": 0.19111455976963043,
+      "learning_rate": 5.2639225167006475e-05,
+      "loss": 0.1709,
+      "step": 17920
+    },
+    {
+      "epoch": 2.257581005762509,
+      "grad_norm": 0.24302569031715393,
+      "learning_rate": 5.255563008733599e-05,
+      "loss": 0.1752,
+      "step": 17925
+    },
+    {
+      "epoch": 2.2582107881726863,
+      "grad_norm": 0.20797547698020935,
+      "learning_rate": 5.247208733751801e-05,
+      "loss": 0.1792,
+      "step": 17930
+    },
+    {
+      "epoch": 2.258840570582864,
+      "grad_norm": 0.21642006933689117,
+      "learning_rate": 5.238859696241689e-05,
+      "loss": 0.1673,
+      "step": 17935
+    },
+    {
+      "epoch": 2.259470352993041,
+      "grad_norm": 0.22728614509105682,
+      "learning_rate": 5.2305159006868885e-05,
+      "loss": 0.1793,
+      "step": 17940
+    },
+    {
+      "epoch": 2.260100135403218,
+      "grad_norm": 0.24052174389362335,
+      "learning_rate": 5.2221773515682035e-05,
+      "loss": 0.1791,
+      "step": 17945
+    },
+    {
+      "epoch": 2.2607299178133955,
+      "grad_norm": 0.21312139928340912,
+      "learning_rate": 5.213844053363635e-05,
+      "loss": 0.177,
+      "step": 17950
+    },
+    {
+      "epoch": 2.2613597002235726,
+      "grad_norm": 0.22087723016738892,
+      "learning_rate": 5.205516010548349e-05,
+      "loss": 0.1764,
+      "step": 17955
+    },
+    {
+      "epoch": 2.26198948263375,
+      "grad_norm": 0.24077439308166504,
+      "learning_rate": 5.1971932275946967e-05,
+      "loss": 0.1884,
+      "step": 17960
+    },
+    {
+      "epoch": 2.262619265043927,
+      "grad_norm": 0.2120356261730194,
+      "learning_rate": 5.188875708972198e-05,
+      "loss": 0.173,
+      "step": 17965
+    },
+    {
+      "epoch": 2.2632490474541047,
+      "grad_norm": 0.24573729932308197,
+      "learning_rate": 5.1805634591475555e-05,
+      "loss": 0.1824,
+      "step": 17970
+    },
+    {
+      "epoch": 2.263878829864282,
+      "grad_norm": 0.20354896783828735,
+      "learning_rate": 5.1722564825846336e-05,
+      "loss": 0.1738,
+      "step": 17975
+    },
+    {
+      "epoch": 2.2645086122744593,
+      "grad_norm": 0.2105248123407364,
+      "learning_rate": 5.1639547837444725e-05,
+      "loss": 0.1694,
+      "step": 17980
+    },
+    {
+      "epoch": 2.2651383946846364,
+      "grad_norm": 0.21009747684001923,
+      "learning_rate": 5.1556583670852636e-05,
+      "loss": 0.1773,
+      "step": 17985
+    },
+    {
+      "epoch": 2.265768177094814,
+      "grad_norm": 0.21542850136756897,
+      "learning_rate": 5.147367237062387e-05,
+      "loss": 0.1682,
+      "step": 17990
+    },
+    {
+      "epoch": 2.266397959504991,
+      "grad_norm": 0.20584627985954285,
+      "learning_rate": 5.1390813981283676e-05,
+      "loss": 0.1734,
+      "step": 17995
+    },
+    {
+      "epoch": 2.267027741915168,
+      "grad_norm": 0.2486305981874466,
+      "learning_rate": 5.130800854732877e-05,
+      "loss": 0.1825,
+      "step": 18000
+    },
+    {
+      "epoch": 2.267027741915168,
+      "eval_loss": 0.35427358746528625,
+      "eval_runtime": 6.1591,
+      "eval_samples_per_second": 162.361,
+      "eval_steps_per_second": 10.229,
+      "step": 18000
+    },
+    {
+      "epoch": 2.2676575243253456,
+      "grad_norm": 0.19808907806873322,
+      "learning_rate": 5.122525611322761e-05,
+      "loss": 0.1625,
+      "step": 18005
+    },
+    {
+      "epoch": 2.2682873067355227,
+      "grad_norm": 0.24098962545394897,
+      "learning_rate": 5.114255672342022e-05,
+      "loss": 0.1687,
+      "step": 18010
+    },
+    {
+      "epoch": 2.2689170891457002,
+      "grad_norm": 0.22834831476211548,
+      "learning_rate": 5.105991042231799e-05,
+      "loss": 0.1695,
+      "step": 18015
+    },
+    {
+      "epoch": 2.2695468715558773,
+      "grad_norm": 0.19950784742832184,
+      "learning_rate": 5.097731725430392e-05,
+      "loss": 0.1692,
+      "step": 18020
+    },
+    {
+      "epoch": 2.270176653966055,
+      "grad_norm": 0.23613286018371582,
+      "learning_rate": 5.0894777263732405e-05,
+      "loss": 0.176,
+      "step": 18025
+    },
+    {
+      "epoch": 2.270806436376232,
+      "grad_norm": 0.2248247116804123,
+      "learning_rate": 5.081229049492929e-05,
+      "loss": 0.1638,
+      "step": 18030
+    },
+    {
+      "epoch": 2.2714362187864094,
+      "grad_norm": 0.21063442528247833,
+      "learning_rate": 5.072985699219186e-05,
+      "loss": 0.1696,
+      "step": 18035
+    },
+    {
+      "epoch": 2.2720660011965865,
+      "grad_norm": 0.26251456141471863,
+      "learning_rate": 5.064747679978881e-05,
+      "loss": 0.1784,
+      "step": 18040
+    },
+    {
+      "epoch": 2.272695783606764,
+      "grad_norm": 0.20396436750888824,
+      "learning_rate": 5.056514996196011e-05,
+      "loss": 0.1733,
+      "step": 18045
+    },
+    {
+      "epoch": 2.273325566016941,
+      "grad_norm": 0.21515126526355743,
+      "learning_rate": 5.048287652291728e-05,
+      "loss": 0.1625,
+      "step": 18050
+    },
+    {
+      "epoch": 2.273955348427118,
+      "grad_norm": 0.24371370673179626,
+      "learning_rate": 5.0400656526842946e-05,
+      "loss": 0.1739,
+      "step": 18055
+    },
+    {
+      "epoch": 2.2745851308372957,
+      "grad_norm": 0.22852087020874023,
+      "learning_rate": 5.03184900178912e-05,
+      "loss": 0.171,
+      "step": 18060
+    },
+    {
+      "epoch": 2.275214913247473,
+      "grad_norm": 0.22659562528133392,
+      "learning_rate": 5.023637704018719e-05,
+      "loss": 0.1769,
+      "step": 18065
+    },
+    {
+      "epoch": 2.2758446956576504,
+      "grad_norm": 0.2462269514799118,
+      "learning_rate": 5.01543176378276e-05,
+      "loss": 0.1731,
+      "step": 18070
+    },
+    {
+      "epoch": 2.2764744780678274,
+      "grad_norm": 0.21395175158977509,
+      "learning_rate": 5.007231185488016e-05,
+      "loss": 0.1705,
+      "step": 18075
+    },
+    {
+      "epoch": 2.277104260478005,
+      "grad_norm": 0.2166956514120102,
+      "learning_rate": 4.9990359735383837e-05,
+      "loss": 0.1671,
+      "step": 18080
+    },
+    {
+      "epoch": 2.277734042888182,
+      "grad_norm": 0.23139755427837372,
+      "learning_rate": 4.9908461323348754e-05,
+      "loss": 0.1785,
+      "step": 18085
+    },
+    {
+      "epoch": 2.2783638252983596,
+      "grad_norm": 0.23193643987178802,
+      "learning_rate": 4.982661666275632e-05,
+      "loss": 0.1746,
+      "step": 18090
+    },
+    {
+      "epoch": 2.2789936077085367,
+      "grad_norm": 0.21008536219596863,
+      "learning_rate": 4.974482579755899e-05,
+      "loss": 0.1784,
+      "step": 18095
+    },
+    {
+      "epoch": 2.279623390118714,
+      "grad_norm": 0.23688139021396637,
+      "learning_rate": 4.9663088771680235e-05,
+      "loss": 0.1812,
+      "step": 18100
+    },
+    {
+      "epoch": 2.2802531725288913,
+      "grad_norm": 0.20811019837856293,
+      "learning_rate": 4.958140562901468e-05,
+      "loss": 0.1721,
+      "step": 18105
+    },
+    {
+      "epoch": 2.2808829549390683,
+      "grad_norm": 0.2096734642982483,
+      "learning_rate": 4.9499776413428167e-05,
+      "loss": 0.1697,
+      "step": 18110
+    },
+    {
+      "epoch": 2.281512737349246,
+      "grad_norm": 0.22839121520519257,
+      "learning_rate": 4.9418201168757386e-05,
+      "loss": 0.1729,
+      "step": 18115
+    },
+    {
+      "epoch": 2.282142519759423,
+      "grad_norm": 0.21908484399318695,
+      "learning_rate": 4.9336679938810106e-05,
+      "loss": 0.1659,
+      "step": 18120
+    },
+    {
+      "epoch": 2.2827723021696005,
+      "grad_norm": 0.20620904862880707,
+      "learning_rate": 4.925521276736511e-05,
+      "loss": 0.1636,
+      "step": 18125
+    },
+    {
+      "epoch": 2.2834020845797776,
+      "grad_norm": 0.28344854712486267,
+      "learning_rate": 4.9173799698172095e-05,
+      "loss": 0.1753,
+      "step": 18130
+    },
+    {
+      "epoch": 2.284031866989955,
+      "grad_norm": 0.2172774374485016,
+      "learning_rate": 4.909244077495175e-05,
+      "loss": 0.1702,
+      "step": 18135
+    },
+    {
+      "epoch": 2.284661649400132,
+      "grad_norm": 0.19668060541152954,
+      "learning_rate": 4.90111360413957e-05,
+      "loss": 0.1715,
+      "step": 18140
+    },
+    {
+      "epoch": 2.2852914318103097,
+      "grad_norm": 0.19766007363796234,
+      "learning_rate": 4.892988554116642e-05,
+      "loss": 0.1608,
+      "step": 18145
+    },
+    {
+      "epoch": 2.2859212142204868,
+      "grad_norm": 0.2108301967382431,
+      "learning_rate": 4.884868931789724e-05,
+      "loss": 0.1633,
+      "step": 18150
+    },
+    {
+      "epoch": 2.2865509966306643,
+      "grad_norm": 0.25781720876693726,
+      "learning_rate": 4.8767547415192476e-05,
+      "loss": 0.1634,
+      "step": 18155
+    },
+    {
+      "epoch": 2.2871807790408414,
+      "grad_norm": 0.21515868604183197,
+      "learning_rate": 4.8686459876627164e-05,
+      "loss": 0.1687,
+      "step": 18160
+    },
+    {
+      "epoch": 2.2878105614510185,
+      "grad_norm": 0.23936854302883148,
+      "learning_rate": 4.860542674574713e-05,
+      "loss": 0.1786,
+      "step": 18165
+    },
+    {
+      "epoch": 2.288440343861196,
+      "grad_norm": 0.2083710879087448,
+      "learning_rate": 4.852444806606904e-05,
+      "loss": 0.1727,
+      "step": 18170
+    },
+    {
+      "epoch": 2.289070126271373,
+      "grad_norm": 0.24087072908878326,
+      "learning_rate": 4.844352388108028e-05,
+      "loss": 0.1646,
+      "step": 18175
+    },
+    {
+      "epoch": 2.2896999086815506,
+      "grad_norm": 0.22956833243370056,
+      "learning_rate": 4.836265423423898e-05,
+      "loss": 0.1667,
+      "step": 18180
+    },
+    {
+      "epoch": 2.2903296910917277,
+      "grad_norm": 0.2500525414943695,
+      "learning_rate": 4.828183916897402e-05,
+      "loss": 0.1788,
+      "step": 18185
+    },
+    {
+      "epoch": 2.290959473501905,
+      "grad_norm": 0.23779354989528656,
+      "learning_rate": 4.820107872868486e-05,
+      "loss": 0.1687,
+      "step": 18190
+    },
+    {
+      "epoch": 2.2915892559120823,
+      "grad_norm": 0.21519017219543457,
+      "learning_rate": 4.81203729567418e-05,
+      "loss": 0.173,
+      "step": 18195
+    },
+    {
+      "epoch": 2.29221903832226,
+      "grad_norm": 0.2123459428548813,
+      "learning_rate": 4.803972189648568e-05,
+      "loss": 0.1648,
+      "step": 18200
+    },
+    {
+      "epoch": 2.292848820732437,
+      "grad_norm": 0.2364078015089035,
+      "learning_rate": 4.795912559122789e-05,
+      "loss": 0.1743,
+      "step": 18205
+    },
+    {
+      "epoch": 2.2934786031426144,
+      "grad_norm": 0.23717305064201355,
+      "learning_rate": 4.787858408425045e-05,
+      "loss": 0.1827,
+      "step": 18210
+    },
+    {
+      "epoch": 2.2941083855527915,
+      "grad_norm": 0.197091206908226,
+      "learning_rate": 4.7798097418806134e-05,
+      "loss": 0.1713,
+      "step": 18215
+    },
+    {
+      "epoch": 2.2947381679629686,
+      "grad_norm": 0.19760344922542572,
+      "learning_rate": 4.771766563811803e-05,
+      "loss": 0.1612,
+      "step": 18220
+    },
+    {
+      "epoch": 2.295367950373146,
+      "grad_norm": 0.22046242654323578,
+      "learning_rate": 4.763728878537984e-05,
+      "loss": 0.1691,
+      "step": 18225
+    },
+    {
+      "epoch": 2.295997732783323,
+      "grad_norm": 0.22356641292572021,
+      "learning_rate": 4.755696690375574e-05,
+      "loss": 0.1684,
+      "step": 18230
+    },
+    {
+      "epoch": 2.2966275151935007,
+      "grad_norm": 0.20664890110492706,
+      "learning_rate": 4.7476700036380565e-05,
+      "loss": 0.1656,
+      "step": 18235
+    },
+    {
+      "epoch": 2.297257297603678,
+      "grad_norm": 0.2873956859111786,
+      "learning_rate": 4.73964882263593e-05,
+      "loss": 0.1811,
+      "step": 18240
+    },
+    {
+      "epoch": 2.2978870800138553,
+      "grad_norm": 0.23324726521968842,
+      "learning_rate": 4.7316331516767575e-05,
+      "loss": 0.17,
+      "step": 18245
+    },
+    {
+      "epoch": 2.2985168624240324,
+      "grad_norm": 0.22407886385917664,
+      "learning_rate": 4.7236229950651314e-05,
+      "loss": 0.1589,
+      "step": 18250
+    },
+    {
+      "epoch": 2.29914664483421,
+      "grad_norm": 0.2202986776828766,
+      "learning_rate": 4.7156183571026985e-05,
+      "loss": 0.1806,
+      "step": 18255
+    },
+    {
+      "epoch": 2.299776427244387,
+      "grad_norm": 0.1998445987701416,
+      "learning_rate": 4.707619242088129e-05,
+      "loss": 0.1571,
+      "step": 18260
+    },
+    {
+      "epoch": 2.3004062096545645,
+      "grad_norm": 0.24477636814117432,
+      "learning_rate": 4.69962565431713e-05,
+      "loss": 0.1788,
+      "step": 18265
+    },
+    {
+      "epoch": 2.3010359920647416,
+      "grad_norm": 0.2186649590730667,
+      "learning_rate": 4.691637598082439e-05,
+      "loss": 0.1837,
+      "step": 18270
+    },
+    {
+      "epoch": 2.3016657744749187,
+      "grad_norm": 0.19296254217624664,
+      "learning_rate": 4.683655077673826e-05,
+      "loss": 0.1609,
+      "step": 18275
+    },
+    {
+      "epoch": 2.3022955568850962,
+      "grad_norm": 0.234447181224823,
+      "learning_rate": 4.675678097378086e-05,
+      "loss": 0.1711,
+      "step": 18280
+    },
+    {
+      "epoch": 2.3029253392952733,
+      "grad_norm": 0.19974513351917267,
+      "learning_rate": 4.667706661479041e-05,
+      "loss": 0.1666,
+      "step": 18285
+    },
+    {
+      "epoch": 2.303555121705451,
+      "grad_norm": 0.23064357042312622,
+      "learning_rate": 4.659740774257527e-05,
+      "loss": 0.1684,
+      "step": 18290
+    },
+    {
+      "epoch": 2.304184904115628,
+      "grad_norm": 0.19428302347660065,
+      "learning_rate": 4.6517804399914214e-05,
+      "loss": 0.166,
+      "step": 18295
+    },
+    {
+      "epoch": 2.3048146865258055,
+      "grad_norm": 0.23040397465229034,
+      "learning_rate": 4.6438256629555956e-05,
+      "loss": 0.1687,
+      "step": 18300
+    },
+    {
+      "epoch": 2.3054444689359825,
+      "grad_norm": 0.22161847352981567,
+      "learning_rate": 4.635876447421955e-05,
+      "loss": 0.1784,
+      "step": 18305
+    },
+    {
+      "epoch": 2.30607425134616,
+      "grad_norm": 0.22831936180591583,
+      "learning_rate": 4.6279327976593924e-05,
+      "loss": 0.1731,
+      "step": 18310
+    },
+    {
+      "epoch": 2.306704033756337,
+      "grad_norm": 0.25957801938056946,
+      "learning_rate": 4.619994717933848e-05,
+      "loss": 0.1823,
+      "step": 18315
+    },
+    {
+      "epoch": 2.3073338161665147,
+      "grad_norm": 0.23449194431304932,
+      "learning_rate": 4.6120622125082426e-05,
+      "loss": 0.1725,
+      "step": 18320
+    },
+    {
+      "epoch": 2.3079635985766918,
+      "grad_norm": 0.24584275484085083,
+      "learning_rate": 4.604135285642514e-05,
+      "loss": 0.1857,
+      "step": 18325
+    },
+    {
+      "epoch": 2.308593380986869,
+      "grad_norm": 0.21245352923870087,
+      "learning_rate": 4.5962139415936056e-05,
+      "loss": 0.164,
+      "step": 18330
+    },
+    {
+      "epoch": 2.3092231633970464,
+      "grad_norm": 0.2068212777376175,
+      "learning_rate": 4.588298184615453e-05,
+      "loss": 0.1661,
+      "step": 18335
+    },
+    {
+      "epoch": 2.3098529458072234,
+      "grad_norm": 0.21349553763866425,
+      "learning_rate": 4.580388018959013e-05,
+      "loss": 0.1707,
+      "step": 18340
+    },
+    {
+      "epoch": 2.310482728217401,
+      "grad_norm": 0.2073366641998291,
+      "learning_rate": 4.5724834488722106e-05,
+      "loss": 0.1608,
+      "step": 18345
+    },
+    {
+      "epoch": 2.311112510627578,
+      "grad_norm": 0.2493850737810135,
+      "learning_rate": 4.564584478599982e-05,
+      "loss": 0.176,
+      "step": 18350
+    },
+    {
+      "epoch": 2.3117422930377556,
+      "grad_norm": 0.25253990292549133,
+      "learning_rate": 4.556691112384262e-05,
+      "loss": 0.1744,
+      "step": 18355
+    },
+    {
+      "epoch": 2.3123720754479327,
+      "grad_norm": 0.24499280750751495,
+      "learning_rate": 4.548803354463967e-05,
+      "loss": 0.1755,
+      "step": 18360
+    },
+    {
+      "epoch": 2.31300185785811,
+      "grad_norm": 0.21188803017139435,
+      "learning_rate": 4.540921209075e-05,
+      "loss": 0.1675,
+      "step": 18365
+    },
+    {
+      "epoch": 2.3136316402682873,
+      "grad_norm": 0.2255249321460724,
+      "learning_rate": 4.5330446804502543e-05,
+      "loss": 0.1668,
+      "step": 18370
+    },
+    {
+      "epoch": 2.314261422678465,
+      "grad_norm": 0.2088666409254074,
+      "learning_rate": 4.525173772819606e-05,
+      "loss": 0.173,
+      "step": 18375
+    },
+    {
+      "epoch": 2.314891205088642,
+      "grad_norm": 0.24474313855171204,
+      "learning_rate": 4.517308490409912e-05,
+      "loss": 0.1672,
+      "step": 18380
+    },
+    {
+      "epoch": 2.315520987498819,
+      "grad_norm": 0.2033611238002777,
+      "learning_rate": 4.5094488374450085e-05,
+      "loss": 0.1677,
+      "step": 18385
+    },
+    {
+      "epoch": 2.3161507699089965,
+      "grad_norm": 0.22693341970443726,
+      "learning_rate": 4.50159481814571e-05,
+      "loss": 0.1653,
+      "step": 18390
+    },
+    {
+      "epoch": 2.3167805523191736,
+      "grad_norm": 0.24162709712982178,
+      "learning_rate": 4.493746436729797e-05,
+      "loss": 0.1668,
+      "step": 18395
+    },
+    {
+      "epoch": 2.317410334729351,
+      "grad_norm": 0.21281133592128754,
+      "learning_rate": 4.485903697412041e-05,
+      "loss": 0.167,
+      "step": 18400
+    },
+    {
+      "epoch": 2.318040117139528,
+      "grad_norm": 0.2348182648420334,
+      "learning_rate": 4.478066604404168e-05,
+      "loss": 0.1683,
+      "step": 18405
+    },
+    {
+      "epoch": 2.3186698995497057,
+      "grad_norm": 0.2391456663608551,
+      "learning_rate": 4.470235161914878e-05,
+      "loss": 0.1708,
+      "step": 18410
+    },
+    {
+      "epoch": 2.319299681959883,
+      "grad_norm": 0.2014867216348648,
+      "learning_rate": 4.462409374149822e-05,
+      "loss": 0.1679,
+      "step": 18415
+    },
+    {
+      "epoch": 2.3199294643700603,
+      "grad_norm": 0.19464534521102905,
+      "learning_rate": 4.4545892453116414e-05,
+      "loss": 0.167,
+      "step": 18420
+    },
+    {
+      "epoch": 2.3205592467802374,
+      "grad_norm": 0.18525034189224243,
+      "learning_rate": 4.446774779599918e-05,
+      "loss": 0.16,
+      "step": 18425
+    },
+    {
+      "epoch": 2.321189029190415,
+      "grad_norm": 0.220379039645195,
+      "learning_rate": 4.438965981211201e-05,
+      "loss": 0.1728,
+      "step": 18430
+    },
+    {
+      "epoch": 2.321818811600592,
+      "grad_norm": 0.22186563909053802,
+      "learning_rate": 4.431162854338985e-05,
+      "loss": 0.1651,
+      "step": 18435
+    },
+    {
+      "epoch": 2.322448594010769,
+      "grad_norm": 0.22272159159183502,
+      "learning_rate": 4.423365403173739e-05,
+      "loss": 0.171,
+      "step": 18440
+    },
+    {
+      "epoch": 2.3230783764209466,
+      "grad_norm": 0.220636785030365,
+      "learning_rate": 4.4155736319028725e-05,
+      "loss": 0.1691,
+      "step": 18445
+    },
+    {
+      "epoch": 2.3237081588311237,
+      "grad_norm": 0.22500810027122498,
+      "learning_rate": 4.4077875447107356e-05,
+      "loss": 0.1648,
+      "step": 18450
+    },
+    {
+      "epoch": 2.324337941241301,
+      "grad_norm": 0.2163766771554947,
+      "learning_rate": 4.4000071457786335e-05,
+      "loss": 0.1655,
+      "step": 18455
+    },
+    {
+      "epoch": 2.3249677236514783,
+      "grad_norm": 0.2258923053741455,
+      "learning_rate": 4.392232439284829e-05,
+      "loss": 0.1704,
+      "step": 18460
+    },
+    {
+      "epoch": 2.325597506061656,
+      "grad_norm": 0.23461341857910156,
+      "learning_rate": 4.384463429404511e-05,
+      "loss": 0.1686,
+      "step": 18465
+    },
+    {
+      "epoch": 2.326227288471833,
+      "grad_norm": 0.22406549751758575,
+      "learning_rate": 4.376700120309816e-05,
+      "loss": 0.1655,
+      "step": 18470
+    },
+    {
+      "epoch": 2.3268570708820104,
+      "grad_norm": 0.21646642684936523,
+      "learning_rate": 4.368942516169819e-05,
+      "loss": 0.1682,
+      "step": 18475
+    },
+    {
+      "epoch": 2.3274868532921875,
+      "grad_norm": 0.23925819993019104,
+      "learning_rate": 4.3611906211505284e-05,
+      "loss": 0.1746,
+      "step": 18480
+    },
+    {
+      "epoch": 2.328116635702365,
+      "grad_norm": 0.19920630753040314,
+      "learning_rate": 4.35344443941489e-05,
+      "loss": 0.158,
+      "step": 18485
+    },
+    {
+      "epoch": 2.328746418112542,
+      "grad_norm": 0.2575379014015198,
+      "learning_rate": 4.345703975122783e-05,
+      "loss": 0.1708,
+      "step": 18490
+    },
+    {
+      "epoch": 2.329376200522719,
+      "grad_norm": 0.19556741416454315,
+      "learning_rate": 4.3379692324310056e-05,
+      "loss": 0.1677,
+      "step": 18495
+    },
+    {
+      "epoch": 2.3300059829328967,
+      "grad_norm": 0.2595387101173401,
+      "learning_rate": 4.3302402154933005e-05,
+      "loss": 0.1705,
+      "step": 18500
+    },
+    {
+      "epoch": 2.330635765343074,
+      "grad_norm": 0.21318422257900238,
+      "learning_rate": 4.322516928460325e-05,
+      "loss": 0.1676,
+      "step": 18505
+    },
+    {
+      "epoch": 2.3312655477532513,
+      "grad_norm": 0.2212359607219696,
+      "learning_rate": 4.3147993754796624e-05,
+      "loss": 0.1661,
+      "step": 18510
+    },
+    {
+      "epoch": 2.3318953301634284,
+      "grad_norm": 0.1886136680841446,
+      "learning_rate": 4.3070875606958006e-05,
+      "loss": 0.1613,
+      "step": 18515
+    },
+    {
+      "epoch": 2.332525112573606,
+      "grad_norm": 0.23505628108978271,
+      "learning_rate": 4.2993814882501754e-05,
+      "loss": 0.1687,
+      "step": 18520
+    },
+    {
+      "epoch": 2.333154894983783,
+      "grad_norm": 0.18686296045780182,
+      "learning_rate": 4.2916811622811195e-05,
+      "loss": 0.1613,
+      "step": 18525
+    },
+    {
+      "epoch": 2.3337846773939606,
+      "grad_norm": 0.21165959537029266,
+      "learning_rate": 4.2839865869238845e-05,
+      "loss": 0.1604,
+      "step": 18530
+    },
+    {
+      "epoch": 2.3344144598041376,
+      "grad_norm": 0.29806169867515564,
+      "learning_rate": 4.27629776631063e-05,
+      "loss": 0.1682,
+      "step": 18535
+    },
+    {
+      "epoch": 2.335044242214315,
+      "grad_norm": 0.2488899528980255,
+      "learning_rate": 4.268614704570426e-05,
+      "loss": 0.1758,
+      "step": 18540
+    },
+    {
+      "epoch": 2.3356740246244923,
+      "grad_norm": 0.21834008395671844,
+      "learning_rate": 4.2609374058292666e-05,
+      "loss": 0.1587,
+      "step": 18545
+    },
+    {
+      "epoch": 2.3363038070346693,
+      "grad_norm": 0.22900566458702087,
+      "learning_rate": 4.253265874210022e-05,
+      "loss": 0.1798,
+      "step": 18550
+    },
+    {
+      "epoch": 2.336933589444847,
+      "grad_norm": 0.22346030175685883,
+      "learning_rate": 4.2456001138324794e-05,
+      "loss": 0.1656,
+      "step": 18555
+    },
+    {
+      "epoch": 2.337563371855024,
+      "grad_norm": 0.22244654595851898,
+      "learning_rate": 4.237940128813336e-05,
+      "loss": 0.1734,
+      "step": 18560
+    },
+    {
+      "epoch": 2.3381931542652015,
+      "grad_norm": 0.19254350662231445,
+      "learning_rate": 4.230285923266175e-05,
+      "loss": 0.1619,
+      "step": 18565
+    },
+    {
+      "epoch": 2.3388229366753785,
+      "grad_norm": 0.22871673107147217,
+      "learning_rate": 4.222637501301481e-05,
+      "loss": 0.166,
+      "step": 18570
+    },
+    {
+      "epoch": 2.339452719085556,
+      "grad_norm": 0.20270411670207977,
+      "learning_rate": 4.2149948670266284e-05,
+      "loss": 0.1637,
+      "step": 18575
+    },
+    {
+      "epoch": 2.340082501495733,
+      "grad_norm": 0.23636558651924133,
+      "learning_rate": 4.2073580245458874e-05,
+      "loss": 0.1839,
+      "step": 18580
+    },
+    {
+      "epoch": 2.3407122839059107,
+      "grad_norm": 0.24934862554073334,
+      "learning_rate": 4.1997269779604185e-05,
+      "loss": 0.1661,
+      "step": 18585
+    },
+    {
+      "epoch": 2.3413420663160878,
+      "grad_norm": 0.2234071046113968,
+      "learning_rate": 4.192101731368267e-05,
+      "loss": 0.1699,
+      "step": 18590
+    },
+    {
+      "epoch": 2.3419718487262653,
+      "grad_norm": 0.20725548267364502,
+      "learning_rate": 4.1844822888643634e-05,
+      "loss": 0.1663,
+      "step": 18595
+    },
+    {
+      "epoch": 2.3426016311364424,
+      "grad_norm": 0.22668230533599854,
+      "learning_rate": 4.1768686545405186e-05,
+      "loss": 0.1647,
+      "step": 18600
+    },
+    {
+      "epoch": 2.3432314135466195,
+      "grad_norm": 0.23123641312122345,
+      "learning_rate": 4.1692608324854384e-05,
+      "loss": 0.171,
+      "step": 18605
+    },
+    {
+      "epoch": 2.343861195956797,
+      "grad_norm": 0.21715596318244934,
+      "learning_rate": 4.161658826784692e-05,
+      "loss": 0.1631,
+      "step": 18610
+    },
+    {
+      "epoch": 2.344490978366974,
+      "grad_norm": 0.24206319451332092,
+      "learning_rate": 4.154062641520732e-05,
+      "loss": 0.1724,
+      "step": 18615
+    },
+    {
+      "epoch": 2.3451207607771516,
+      "grad_norm": 0.21535861492156982,
+      "learning_rate": 4.1464722807728724e-05,
+      "loss": 0.1673,
+      "step": 18620
+    },
+    {
+      "epoch": 2.3457505431873287,
+      "grad_norm": 0.24345341324806213,
+      "learning_rate": 4.1388877486173245e-05,
+      "loss": 0.1648,
+      "step": 18625
+    },
+    {
+      "epoch": 2.346380325597506,
+      "grad_norm": 0.2361554056406021,
+      "learning_rate": 4.131309049127149e-05,
+      "loss": 0.1624,
+      "step": 18630
+    },
+    {
+      "epoch": 2.3470101080076833,
+      "grad_norm": 0.20666177570819855,
+      "learning_rate": 4.1237361863722816e-05,
+      "loss": 0.1662,
+      "step": 18635
+    },
+    {
+      "epoch": 2.347639890417861,
+      "grad_norm": 0.22876566648483276,
+      "learning_rate": 4.1161691644195165e-05,
+      "loss": 0.1767,
+      "step": 18640
+    },
+    {
+      "epoch": 2.348269672828038,
+      "grad_norm": 0.19370432198047638,
+      "learning_rate": 4.108607987332529e-05,
+      "loss": 0.1604,
+      "step": 18645
+    },
+    {
+      "epoch": 2.3488994552382154,
+      "grad_norm": 0.22485142946243286,
+      "learning_rate": 4.101052659171842e-05,
+      "loss": 0.1667,
+      "step": 18650
+    },
+    {
+      "epoch": 2.3495292376483925,
+      "grad_norm": 0.2446049600839615,
+      "learning_rate": 4.0935031839948315e-05,
+      "loss": 0.1719,
+      "step": 18655
+    },
+    {
+      "epoch": 2.3501590200585696,
+      "grad_norm": 0.22652800381183624,
+      "learning_rate": 4.0859595658557367e-05,
+      "loss": 0.1666,
+      "step": 18660
+    },
+    {
+      "epoch": 2.350788802468747,
+      "grad_norm": 0.1760840266942978,
+      "learning_rate": 4.078421808805663e-05,
+      "loss": 0.1516,
+      "step": 18665
+    },
+    {
+      "epoch": 2.351418584878924,
+      "grad_norm": 0.20791617035865784,
+      "learning_rate": 4.070889916892553e-05,
+      "loss": 0.164,
+      "step": 18670
+    },
+    {
+      "epoch": 2.3520483672891017,
+      "grad_norm": 0.2205626517534256,
+      "learning_rate": 4.063363894161206e-05,
+      "loss": 0.1669,
+      "step": 18675
+    },
+    {
+      "epoch": 2.352678149699279,
+      "grad_norm": 0.23379269242286682,
+      "learning_rate": 4.055843744653266e-05,
+      "loss": 0.1593,
+      "step": 18680
+    },
+    {
+      "epoch": 2.3533079321094563,
+      "grad_norm": 0.23451068997383118,
+      "learning_rate": 4.0483294724072254e-05,
+      "loss": 0.1633,
+      "step": 18685
+    },
+    {
+      "epoch": 2.3539377145196334,
+      "grad_norm": 0.28889602422714233,
+      "learning_rate": 4.040821081458422e-05,
+      "loss": 0.1752,
+      "step": 18690
+    },
+    {
+      "epoch": 2.354567496929811,
+      "grad_norm": 0.2054235339164734,
+      "learning_rate": 4.0333185758390307e-05,
+      "loss": 0.1666,
+      "step": 18695
+    },
+    {
+      "epoch": 2.355197279339988,
+      "grad_norm": 0.19770711660385132,
+      "learning_rate": 4.025821959578067e-05,
+      "loss": 0.1701,
+      "step": 18700
+    },
+    {
+      "epoch": 2.3558270617501655,
+      "grad_norm": 0.25233033299446106,
+      "learning_rate": 4.0183312367013906e-05,
+      "loss": 0.1722,
+      "step": 18705
+    },
+    {
+      "epoch": 2.3564568441603426,
+      "grad_norm": 0.20867769420146942,
+      "learning_rate": 4.010846411231689e-05,
+      "loss": 0.1601,
+      "step": 18710
+    },
+    {
+      "epoch": 2.3570866265705197,
+      "grad_norm": 0.21671661734580994,
+      "learning_rate": 4.003367487188483e-05,
+      "loss": 0.1658,
+      "step": 18715
+    },
+    {
+      "epoch": 2.3577164089806972,
+      "grad_norm": 0.17957130074501038,
+      "learning_rate": 3.9958944685881265e-05,
+      "loss": 0.1619,
+      "step": 18720
+    },
+    {
+      "epoch": 2.3583461913908743,
+      "grad_norm": 0.21048414707183838,
+      "learning_rate": 3.988427359443802e-05,
+      "loss": 0.1668,
+      "step": 18725
+    },
+    {
+      "epoch": 2.358975973801052,
+      "grad_norm": 0.21969716250896454,
+      "learning_rate": 3.980966163765513e-05,
+      "loss": 0.1619,
+      "step": 18730
+    },
+    {
+      "epoch": 2.359605756211229,
+      "grad_norm": 0.22368858754634857,
+      "learning_rate": 3.9735108855600984e-05,
+      "loss": 0.168,
+      "step": 18735
+    },
+    {
+      "epoch": 2.3602355386214064,
+      "grad_norm": 0.2626504600048065,
+      "learning_rate": 3.966061528831209e-05,
+      "loss": 0.1651,
+      "step": 18740
+    },
+    {
+      "epoch": 2.3608653210315835,
+      "grad_norm": 0.21985310316085815,
+      "learning_rate": 3.958618097579316e-05,
+      "loss": 0.1671,
+      "step": 18745
+    },
+    {
+      "epoch": 2.361495103441761,
+      "grad_norm": 0.22451792657375336,
+      "learning_rate": 3.9511805958017205e-05,
+      "loss": 0.1609,
+      "step": 18750
+    },
+    {
+      "epoch": 2.362124885851938,
+      "grad_norm": 0.2123977243900299,
+      "learning_rate": 3.943749027492532e-05,
+      "loss": 0.1719,
+      "step": 18755
+    },
+    {
+      "epoch": 2.3627546682621157,
+      "grad_norm": 0.2234313040971756,
+      "learning_rate": 3.936323396642658e-05,
+      "loss": 0.1556,
+      "step": 18760
+    },
+    {
+      "epoch": 2.3633844506722927,
+      "grad_norm": 0.19645099341869354,
+      "learning_rate": 3.928903707239846e-05,
+      "loss": 0.1673,
+      "step": 18765
+    },
+    {
+      "epoch": 2.36401423308247,
+      "grad_norm": 0.22249870002269745,
+      "learning_rate": 3.9214899632686334e-05,
+      "loss": 0.1589,
+      "step": 18770
+    },
+    {
+      "epoch": 2.3646440154926474,
+      "grad_norm": 0.2180803418159485,
+      "learning_rate": 3.914082168710369e-05,
+      "loss": 0.1685,
+      "step": 18775
+    },
+    {
+      "epoch": 2.3652737979028244,
+      "grad_norm": 0.2156234085559845,
+      "learning_rate": 3.906680327543212e-05,
+      "loss": 0.1613,
+      "step": 18780
+    },
+    {
+      "epoch": 2.365903580313002,
+      "grad_norm": 0.2180781066417694,
+      "learning_rate": 3.899284443742112e-05,
+      "loss": 0.1654,
+      "step": 18785
+    },
+    {
+      "epoch": 2.366533362723179,
+      "grad_norm": 0.2102290391921997,
+      "learning_rate": 3.89189452127884e-05,
+      "loss": 0.1635,
+      "step": 18790
+    },
+    {
+      "epoch": 2.3671631451333566,
+      "grad_norm": 0.26211512088775635,
+      "learning_rate": 3.884510564121944e-05,
+      "loss": 0.174,
+      "step": 18795
+    },
+    {
+      "epoch": 2.3677929275435337,
+      "grad_norm": 0.1999218463897705,
+      "learning_rate": 3.877132576236778e-05,
+      "loss": 0.1619,
+      "step": 18800
+    },
+    {
+      "epoch": 2.368422709953711,
+      "grad_norm": 0.21774223446846008,
+      "learning_rate": 3.8697605615854875e-05,
+      "loss": 0.1616,
+      "step": 18805
+    },
+    {
+      "epoch": 2.3690524923638883,
+      "grad_norm": 0.2304651439189911,
+      "learning_rate": 3.862394524127023e-05,
+      "loss": 0.1705,
+      "step": 18810
+    },
+    {
+      "epoch": 2.369682274774066,
+      "grad_norm": 0.24826854467391968,
+      "learning_rate": 3.8550344678171084e-05,
+      "loss": 0.1734,
+      "step": 18815
+    },
+    {
+      "epoch": 2.370312057184243,
+      "grad_norm": 0.21676623821258545,
+      "learning_rate": 3.847680396608262e-05,
+      "loss": 0.1669,
+      "step": 18820
+    },
+    {
+      "epoch": 2.37094183959442,
+      "grad_norm": 0.21203717589378357,
+      "learning_rate": 3.840332314449788e-05,
+      "loss": 0.1633,
+      "step": 18825
+    },
+    {
+      "epoch": 2.3715716220045975,
+      "grad_norm": 0.2409755140542984,
+      "learning_rate": 3.832990225287776e-05,
+      "loss": 0.1687,
+      "step": 18830
+    },
+    {
+      "epoch": 2.3722014044147746,
+      "grad_norm": 0.19998323917388916,
+      "learning_rate": 3.825654133065094e-05,
+      "loss": 0.1578,
+      "step": 18835
+    },
+    {
+      "epoch": 2.372831186824952,
+      "grad_norm": 0.22083517909049988,
+      "learning_rate": 3.818324041721391e-05,
+      "loss": 0.1721,
+      "step": 18840
+    },
+    {
+      "epoch": 2.373460969235129,
+      "grad_norm": 0.19865678250789642,
+      "learning_rate": 3.8109999551930914e-05,
+      "loss": 0.1613,
+      "step": 18845
+    },
+    {
+      "epoch": 2.3740907516453067,
+      "grad_norm": 0.2167719304561615,
+      "learning_rate": 3.8036818774134037e-05,
+      "loss": 0.1569,
+      "step": 18850
+    },
+    {
+      "epoch": 2.374720534055484,
+      "grad_norm": 0.2173914611339569,
+      "learning_rate": 3.796369812312298e-05,
+      "loss": 0.1676,
+      "step": 18855
+    },
+    {
+      "epoch": 2.3753503164656613,
+      "grad_norm": 0.22559495270252228,
+      "learning_rate": 3.7890637638165255e-05,
+      "loss": 0.169,
+      "step": 18860
+    },
+    {
+      "epoch": 2.3759800988758384,
+      "grad_norm": 0.2124035507440567,
+      "learning_rate": 3.781763735849589e-05,
+      "loss": 0.1715,
+      "step": 18865
+    },
+    {
+      "epoch": 2.376609881286016,
+      "grad_norm": 0.23133991658687592,
+      "learning_rate": 3.774469732331782e-05,
+      "loss": 0.162,
+      "step": 18870
+    },
+    {
+      "epoch": 2.377239663696193,
+      "grad_norm": 0.1754513680934906,
+      "learning_rate": 3.7671817571801464e-05,
+      "loss": 0.1602,
+      "step": 18875
+    },
+    {
+      "epoch": 2.37786944610637,
+      "grad_norm": 0.2158019244670868,
+      "learning_rate": 3.7598998143084924e-05,
+      "loss": 0.1571,
+      "step": 18880
+    },
+    {
+      "epoch": 2.3784992285165476,
+      "grad_norm": 0.20694270730018616,
+      "learning_rate": 3.752623907627388e-05,
+      "loss": 0.162,
+      "step": 18885
+    },
+    {
+      "epoch": 2.3791290109267247,
+      "grad_norm": 0.250929057598114,
+      "learning_rate": 3.7453540410441604e-05,
+      "loss": 0.1744,
+      "step": 18890
+    },
+    {
+      "epoch": 2.379758793336902,
+      "grad_norm": 0.23653000593185425,
+      "learning_rate": 3.738090218462903e-05,
+      "loss": 0.1789,
+      "step": 18895
+    },
+    {
+      "epoch": 2.3803885757470793,
+      "grad_norm": 0.1936427801847458,
+      "learning_rate": 3.730832443784443e-05,
+      "loss": 0.1532,
+      "step": 18900
+    },
+    {
+      "epoch": 2.381018358157257,
+      "grad_norm": 0.188064306974411,
+      "learning_rate": 3.7235807209063716e-05,
+      "loss": 0.1629,
+      "step": 18905
+    },
+    {
+      "epoch": 2.381648140567434,
+      "grad_norm": 0.2069697082042694,
+      "learning_rate": 3.71633505372304e-05,
+      "loss": 0.1568,
+      "step": 18910
+    },
+    {
+      "epoch": 2.3822779229776114,
+      "grad_norm": 0.21809029579162598,
+      "learning_rate": 3.709095446125529e-05,
+      "loss": 0.1717,
+      "step": 18915
+    },
+    {
+      "epoch": 2.3829077053877885,
+      "grad_norm": 0.23560817539691925,
+      "learning_rate": 3.701861902001675e-05,
+      "loss": 0.1662,
+      "step": 18920
+    },
+    {
+      "epoch": 2.383537487797966,
+      "grad_norm": 0.19693933427333832,
+      "learning_rate": 3.694634425236057e-05,
+      "loss": 0.1558,
+      "step": 18925
+    },
+    {
+      "epoch": 2.384167270208143,
+      "grad_norm": 0.19060872495174408,
+      "learning_rate": 3.687413019709994e-05,
+      "loss": 0.1621,
+      "step": 18930
+    },
+    {
+      "epoch": 2.38479705261832,
+      "grad_norm": 0.2021481990814209,
+      "learning_rate": 3.680197689301548e-05,
+      "loss": 0.1551,
+      "step": 18935
+    },
+    {
+      "epoch": 2.3854268350284977,
+      "grad_norm": 0.22511224448680878,
+      "learning_rate": 3.672988437885512e-05,
+      "loss": 0.1587,
+      "step": 18940
+    },
+    {
+      "epoch": 2.386056617438675,
+      "grad_norm": 0.2018289864063263,
+      "learning_rate": 3.665785269333423e-05,
+      "loss": 0.1654,
+      "step": 18945
+    },
+    {
+      "epoch": 2.3866863998488523,
+      "grad_norm": 0.21350149810314178,
+      "learning_rate": 3.65858818751354e-05,
+      "loss": 0.1673,
+      "step": 18950
+    },
+    {
+      "epoch": 2.3873161822590294,
+      "grad_norm": 0.21213771402835846,
+      "learning_rate": 3.65139719629087e-05,
+      "loss": 0.1791,
+      "step": 18955
+    },
+    {
+      "epoch": 2.387945964669207,
+      "grad_norm": 0.24175149202346802,
+      "learning_rate": 3.644212299527139e-05,
+      "loss": 0.1714,
+      "step": 18960
+    },
+    {
+      "epoch": 2.388575747079384,
+      "grad_norm": 0.2541513741016388,
+      "learning_rate": 3.63703350108079e-05,
+      "loss": 0.1685,
+      "step": 18965
+    },
+    {
+      "epoch": 2.389205529489561,
+      "grad_norm": 0.24447733163833618,
+      "learning_rate": 3.629860804807011e-05,
+      "loss": 0.1728,
+      "step": 18970
+    },
+    {
+      "epoch": 2.3898353118997386,
+      "grad_norm": 0.1830032914876938,
+      "learning_rate": 3.622694214557702e-05,
+      "loss": 0.1698,
+      "step": 18975
+    },
+    {
+      "epoch": 2.390465094309916,
+      "grad_norm": 0.23851166665554047,
+      "learning_rate": 3.6155337341814844e-05,
+      "loss": 0.1754,
+      "step": 18980
+    },
+    {
+      "epoch": 2.3910948767200932,
+      "grad_norm": 0.1973876655101776,
+      "learning_rate": 3.608379367523702e-05,
+      "loss": 0.1703,
+      "step": 18985
+    },
+    {
+      "epoch": 2.3917246591302703,
+      "grad_norm": 0.2209198772907257,
+      "learning_rate": 3.6012311184264046e-05,
+      "loss": 0.1674,
+      "step": 18990
+    },
+    {
+      "epoch": 2.392354441540448,
+      "grad_norm": 0.216825932264328,
+      "learning_rate": 3.5940889907283834e-05,
+      "loss": 0.1677,
+      "step": 18995
+    },
+    {
+      "epoch": 2.392984223950625,
+      "grad_norm": 0.1855764538049698,
+      "learning_rate": 3.586952988265106e-05,
+      "loss": 0.1592,
+      "step": 19000
+    },
+    {
+      "epoch": 2.392984223950625,
+      "eval_loss": 0.35235053300857544,
+      "eval_runtime": 6.1677,
+      "eval_samples_per_second": 162.135,
+      "eval_steps_per_second": 10.215,
+      "step": 19000
+    },
+    {
+      "epoch": 2.3936140063608025,
+      "grad_norm": 0.25512510538101196,
+      "learning_rate": 3.579823114868778e-05,
+      "loss": 0.1649,
+      "step": 19005
+    },
+    {
+      "epoch": 2.3942437887709795,
+      "grad_norm": 0.20220012962818146,
+      "learning_rate": 3.572699374368296e-05,
+      "loss": 0.1638,
+      "step": 19010
+    },
+    {
+      "epoch": 2.394873571181157,
+      "grad_norm": 0.21141274273395538,
+      "learning_rate": 3.5655817705892814e-05,
+      "loss": 0.1697,
+      "step": 19015
+    },
+    {
+      "epoch": 2.395503353591334,
+      "grad_norm": 0.23368066549301147,
+      "learning_rate": 3.558470307354046e-05,
+      "loss": 0.1653,
+      "step": 19020
+    },
+    {
+      "epoch": 2.3961331360015112,
+      "grad_norm": 0.20436784625053406,
+      "learning_rate": 3.5513649884816064e-05,
+      "loss": 0.1561,
+      "step": 19025
+    },
+    {
+      "epoch": 2.3967629184116888,
+      "grad_norm": 0.223700612783432,
+      "learning_rate": 3.5442658177876835e-05,
+      "loss": 0.1693,
+      "step": 19030
+    },
+    {
+      "epoch": 2.3973927008218663,
+      "grad_norm": 0.26057204604148865,
+      "learning_rate": 3.5371727990846944e-05,
+      "loss": 0.1767,
+      "step": 19035
+    },
+    {
+      "epoch": 2.3980224832320434,
+      "grad_norm": 0.21637168526649475,
+      "learning_rate": 3.53008593618175e-05,
+      "loss": 0.1671,
+      "step": 19040
+    },
+    {
+      "epoch": 2.3986522656422204,
+      "grad_norm": 0.23995353281497955,
+      "learning_rate": 3.5230052328846585e-05,
+      "loss": 0.1788,
+      "step": 19045
+    },
+    {
+      "epoch": 2.399282048052398,
+      "grad_norm": 0.2069759964942932,
+      "learning_rate": 3.5159306929959144e-05,
+      "loss": 0.1655,
+      "step": 19050
+    },
+    {
+      "epoch": 2.399911830462575,
+      "grad_norm": 0.20498618483543396,
+      "learning_rate": 3.508862320314717e-05,
+      "loss": 0.1589,
+      "step": 19055
+    },
+    {
+      "epoch": 2.4005416128727526,
+      "grad_norm": 0.20835870504379272,
+      "learning_rate": 3.501800118636939e-05,
+      "loss": 0.1556,
+      "step": 19060
+    },
+    {
+      "epoch": 2.4011713952829297,
+      "grad_norm": 0.22261159121990204,
+      "learning_rate": 3.4947440917551475e-05,
+      "loss": 0.1645,
+      "step": 19065
+    },
+    {
+      "epoch": 2.401801177693107,
+      "grad_norm": 0.20514576137065887,
+      "learning_rate": 3.487694243458578e-05,
+      "loss": 0.1558,
+      "step": 19070
+    },
+    {
+      "epoch": 2.4024309601032843,
+      "grad_norm": 0.18798956274986267,
+      "learning_rate": 3.480650577533175e-05,
+      "loss": 0.1635,
+      "step": 19075
+    },
+    {
+      "epoch": 2.4030607425134614,
+      "grad_norm": 0.1777620017528534,
+      "learning_rate": 3.47361309776154e-05,
+      "loss": 0.1613,
+      "step": 19080
+    },
+    {
+      "epoch": 2.403690524923639,
+      "grad_norm": 0.22258161008358002,
+      "learning_rate": 3.466581807922962e-05,
+      "loss": 0.1657,
+      "step": 19085
+    },
+    {
+      "epoch": 2.4043203073338164,
+      "grad_norm": 0.20584604144096375,
+      "learning_rate": 3.4595567117934045e-05,
+      "loss": 0.1609,
+      "step": 19090
+    },
+    {
+      "epoch": 2.4049500897439935,
+      "grad_norm": 0.26850444078445435,
+      "learning_rate": 3.452537813145501e-05,
+      "loss": 0.165,
+      "step": 19095
+    },
+    {
+      "epoch": 2.4055798721541706,
+      "grad_norm": 0.1789688616991043,
+      "learning_rate": 3.4455251157485706e-05,
+      "loss": 0.1597,
+      "step": 19100
+    },
+    {
+      "epoch": 2.406209654564348,
+      "grad_norm": 0.24336829781532288,
+      "learning_rate": 3.438518623368581e-05,
+      "loss": 0.1582,
+      "step": 19105
+    },
+    {
+      "epoch": 2.406839436974525,
+      "grad_norm": 0.20159520208835602,
+      "learning_rate": 3.4315183397681806e-05,
+      "loss": 0.1572,
+      "step": 19110
+    },
+    {
+      "epoch": 2.4074692193847027,
+      "grad_norm": 0.22423477470874786,
+      "learning_rate": 3.424524268706686e-05,
+      "loss": 0.1611,
+      "step": 19115
+    },
+    {
+      "epoch": 2.40809900179488,
+      "grad_norm": 0.22861574590206146,
+      "learning_rate": 3.417536413940073e-05,
+      "loss": 0.1708,
+      "step": 19120
+    },
+    {
+      "epoch": 2.4087287842050573,
+      "grad_norm": 0.22517502307891846,
+      "learning_rate": 3.4105547792209766e-05,
+      "loss": 0.1498,
+      "step": 19125
+    },
+    {
+      "epoch": 2.4093585666152344,
+      "grad_norm": 0.22406402230262756,
+      "learning_rate": 3.403579368298694e-05,
+      "loss": 0.1722,
+      "step": 19130
+    },
+    {
+      "epoch": 2.4099883490254115,
+      "grad_norm": 0.21624189615249634,
+      "learning_rate": 3.3966101849191807e-05,
+      "loss": 0.165,
+      "step": 19135
+    },
+    {
+      "epoch": 2.410618131435589,
+      "grad_norm": 0.2186998724937439,
+      "learning_rate": 3.389647232825048e-05,
+      "loss": 0.1545,
+      "step": 19140
+    },
+    {
+      "epoch": 2.4112479138457665,
+      "grad_norm": 0.20615451037883759,
+      "learning_rate": 3.38269051575556e-05,
+      "loss": 0.1653,
+      "step": 19145
+    },
+    {
+      "epoch": 2.4118776962559436,
+      "grad_norm": 0.21351304650306702,
+      "learning_rate": 3.3757400374466323e-05,
+      "loss": 0.1667,
+      "step": 19150
+    },
+    {
+      "epoch": 2.4125074786661207,
+      "grad_norm": 0.2263455092906952,
+      "learning_rate": 3.368795801630826e-05,
+      "loss": 0.1635,
+      "step": 19155
+    },
+    {
+      "epoch": 2.4131372610762982,
+      "grad_norm": 0.20655429363250732,
+      "learning_rate": 3.361857812037365e-05,
+      "loss": 0.1657,
+      "step": 19160
+    },
+    {
+      "epoch": 2.4137670434864753,
+      "grad_norm": 0.1987982541322708,
+      "learning_rate": 3.354926072392101e-05,
+      "loss": 0.1554,
+      "step": 19165
+    },
+    {
+      "epoch": 2.414396825896653,
+      "grad_norm": 0.20431582629680634,
+      "learning_rate": 3.348000586417539e-05,
+      "loss": 0.1552,
+      "step": 19170
+    },
+    {
+      "epoch": 2.41502660830683,
+      "grad_norm": 0.241183340549469,
+      "learning_rate": 3.34108135783282e-05,
+      "loss": 0.1758,
+      "step": 19175
+    },
+    {
+      "epoch": 2.4156563907170074,
+      "grad_norm": 0.1910007894039154,
+      "learning_rate": 3.3341683903537295e-05,
+      "loss": 0.1609,
+      "step": 19180
+    },
+    {
+      "epoch": 2.4162861731271845,
+      "grad_norm": 0.2089349776506424,
+      "learning_rate": 3.3272616876926916e-05,
+      "loss": 0.1608,
+      "step": 19185
+    },
+    {
+      "epoch": 2.4169159555373616,
+      "grad_norm": 0.20799914002418518,
+      "learning_rate": 3.3203612535587594e-05,
+      "loss": 0.1636,
+      "step": 19190
+    },
+    {
+      "epoch": 2.417545737947539,
+      "grad_norm": 0.2071768194437027,
+      "learning_rate": 3.313467091657622e-05,
+      "loss": 0.1643,
+      "step": 19195
+    },
+    {
+      "epoch": 2.4181755203577167,
+      "grad_norm": 0.22981540858745575,
+      "learning_rate": 3.3065792056916077e-05,
+      "loss": 0.1749,
+      "step": 19200
+    },
+    {
+      "epoch": 2.4188053027678937,
+      "grad_norm": 0.22096100449562073,
+      "learning_rate": 3.2996975993596706e-05,
+      "loss": 0.1671,
+      "step": 19205
+    },
+    {
+      "epoch": 2.419435085178071,
+      "grad_norm": 0.19851092994213104,
+      "learning_rate": 3.292822276357382e-05,
+      "loss": 0.1605,
+      "step": 19210
+    },
+    {
+      "epoch": 2.4200648675882483,
+      "grad_norm": 0.21692755818367004,
+      "learning_rate": 3.285953240376947e-05,
+      "loss": 0.1629,
+      "step": 19215
+    },
+    {
+      "epoch": 2.4206946499984254,
+      "grad_norm": 0.1912955939769745,
+      "learning_rate": 3.279090495107204e-05,
+      "loss": 0.1626,
+      "step": 19220
+    },
+    {
+      "epoch": 2.421324432408603,
+      "grad_norm": 0.17941045761108398,
+      "learning_rate": 3.2722340442335993e-05,
+      "loss": 0.1528,
+      "step": 19225
+    },
+    {
+      "epoch": 2.42195421481878,
+      "grad_norm": 0.24879097938537598,
+      "learning_rate": 3.265383891438203e-05,
+      "loss": 0.1622,
+      "step": 19230
+    },
+    {
+      "epoch": 2.4225839972289576,
+      "grad_norm": 0.27064043283462524,
+      "learning_rate": 3.258540040399703e-05,
+      "loss": 0.1677,
+      "step": 19235
+    },
+    {
+      "epoch": 2.4232137796391346,
+      "grad_norm": 0.20788533985614777,
+      "learning_rate": 3.2517024947934046e-05,
+      "loss": 0.1742,
+      "step": 19240
+    },
+    {
+      "epoch": 2.4238435620493117,
+      "grad_norm": 0.20137952268123627,
+      "learning_rate": 3.2448712582912265e-05,
+      "loss": 0.1656,
+      "step": 19245
+    },
+    {
+      "epoch": 2.4244733444594893,
+      "grad_norm": 0.22439540922641754,
+      "learning_rate": 3.2380463345616986e-05,
+      "loss": 0.1704,
+      "step": 19250
+    },
+    {
+      "epoch": 2.4251031268696663,
+      "grad_norm": 0.22877377271652222,
+      "learning_rate": 3.231227727269956e-05,
+      "loss": 0.1655,
+      "step": 19255
+    },
+    {
+      "epoch": 2.425732909279844,
+      "grad_norm": 0.19454975426197052,
+      "learning_rate": 3.224415440077757e-05,
+      "loss": 0.1711,
+      "step": 19260
+    },
+    {
+      "epoch": 2.426362691690021,
+      "grad_norm": 0.21811099350452423,
+      "learning_rate": 3.217609476643447e-05,
+      "loss": 0.1602,
+      "step": 19265
+    },
+    {
+      "epoch": 2.4269924741001985,
+      "grad_norm": 0.21663612127304077,
+      "learning_rate": 3.2108098406219884e-05,
+      "loss": 0.1626,
+      "step": 19270
+    },
+    {
+      "epoch": 2.4276222565103756,
+      "grad_norm": 0.21124348044395447,
+      "learning_rate": 3.204016535664937e-05,
+      "loss": 0.1621,
+      "step": 19275
+    },
+    {
+      "epoch": 2.428252038920553,
+      "grad_norm": 0.20466844737529755,
+      "learning_rate": 3.1972295654204554e-05,
+      "loss": 0.1608,
+      "step": 19280
+    },
+    {
+      "epoch": 2.42888182133073,
+      "grad_norm": 0.20153960585594177,
+      "learning_rate": 3.1904489335333014e-05,
+      "loss": 0.1699,
+      "step": 19285
+    },
+    {
+      "epoch": 2.4295116037409077,
+      "grad_norm": 0.2766586244106293,
+      "learning_rate": 3.1836746436448294e-05,
+      "loss": 0.1716,
+      "step": 19290
+    },
+    {
+      "epoch": 2.4301413861510848,
+      "grad_norm": 0.1904149353504181,
+      "learning_rate": 3.176906699392986e-05,
+      "loss": 0.1756,
+      "step": 19295
+    },
+    {
+      "epoch": 2.430771168561262,
+      "grad_norm": 0.19312819838523865,
+      "learning_rate": 3.170145104412309e-05,
+      "loss": 0.1666,
+      "step": 19300
+    },
+    {
+      "epoch": 2.4314009509714394,
+      "grad_norm": 0.19623906910419464,
+      "learning_rate": 3.163389862333939e-05,
+      "loss": 0.1541,
+      "step": 19305
+    },
+    {
+      "epoch": 2.4320307333816165,
+      "grad_norm": 0.1920466423034668,
+      "learning_rate": 3.156640976785592e-05,
+      "loss": 0.1575,
+      "step": 19310
+    },
+    {
+      "epoch": 2.432660515791794,
+      "grad_norm": 0.2178039401769638,
+      "learning_rate": 3.149898451391565e-05,
+      "loss": 0.1533,
+      "step": 19315
+    },
+    {
+      "epoch": 2.433290298201971,
+      "grad_norm": 0.21117891371250153,
+      "learning_rate": 3.143162289772757e-05,
+      "loss": 0.1529,
+      "step": 19320
+    },
+    {
+      "epoch": 2.4339200806121486,
+      "grad_norm": 0.21997326612472534,
+      "learning_rate": 3.1364324955466405e-05,
+      "loss": 0.167,
+      "step": 19325
+    },
+    {
+      "epoch": 2.4345498630223257,
+      "grad_norm": 0.2015310823917389,
+      "learning_rate": 3.129709072327264e-05,
+      "loss": 0.1608,
+      "step": 19330
+    },
+    {
+      "epoch": 2.435179645432503,
+      "grad_norm": 0.21516267955303192,
+      "learning_rate": 3.122992023725263e-05,
+      "loss": 0.159,
+      "step": 19335
+    },
+    {
+      "epoch": 2.4358094278426803,
+      "grad_norm": 0.22670945525169373,
+      "learning_rate": 3.116281353347841e-05,
+      "loss": 0.1703,
+      "step": 19340
+    },
+    {
+      "epoch": 2.436439210252858,
+      "grad_norm": 0.19361603260040283,
+      "learning_rate": 3.109577064798793e-05,
+      "loss": 0.1647,
+      "step": 19345
+    },
+    {
+      "epoch": 2.437068992663035,
+      "grad_norm": 0.18795832991600037,
+      "learning_rate": 3.1028791616784624e-05,
+      "loss": 0.1532,
+      "step": 19350
+    },
+    {
+      "epoch": 2.437698775073212,
+      "grad_norm": 0.24311493337154388,
+      "learning_rate": 3.0961876475837814e-05,
+      "loss": 0.1599,
+      "step": 19355
+    },
+    {
+      "epoch": 2.4383285574833895,
+      "grad_norm": 0.20328237116336823,
+      "learning_rate": 3.089502526108242e-05,
+      "loss": 0.1604,
+      "step": 19360
+    },
+    {
+      "epoch": 2.4389583398935666,
+      "grad_norm": 0.2067318707704544,
+      "learning_rate": 3.082823800841914e-05,
+      "loss": 0.161,
+      "step": 19365
+    },
+    {
+      "epoch": 2.439588122303744,
+      "grad_norm": 0.22590148448944092,
+      "learning_rate": 3.0761514753714235e-05,
+      "loss": 0.1711,
+      "step": 19370
+    },
+    {
+      "epoch": 2.440217904713921,
+      "grad_norm": 0.2264234572649002,
+      "learning_rate": 3.069485553279958e-05,
+      "loss": 0.1625,
+      "step": 19375
+    },
+    {
+      "epoch": 2.4408476871240987,
+      "grad_norm": 0.21357667446136475,
+      "learning_rate": 3.062826038147274e-05,
+      "loss": 0.162,
+      "step": 19380
+    },
+    {
+      "epoch": 2.441477469534276,
+      "grad_norm": 0.19787681102752686,
+      "learning_rate": 3.0561729335496816e-05,
+      "loss": 0.1566,
+      "step": 19385
+    },
+    {
+      "epoch": 2.4421072519444533,
+      "grad_norm": 0.20055502653121948,
+      "learning_rate": 3.0495262430600487e-05,
+      "loss": 0.1612,
+      "step": 19390
+    },
+    {
+      "epoch": 2.4427370343546304,
+      "grad_norm": 0.2178819328546524,
+      "learning_rate": 3.0428859702478003e-05,
+      "loss": 0.1701,
+      "step": 19395
+    },
+    {
+      "epoch": 2.443366816764808,
+      "grad_norm": 0.2206692099571228,
+      "learning_rate": 3.0362521186789125e-05,
+      "loss": 0.1668,
+      "step": 19400
+    },
+    {
+      "epoch": 2.443996599174985,
+      "grad_norm": 0.22452767193317413,
+      "learning_rate": 3.0296246919159218e-05,
+      "loss": 0.1713,
+      "step": 19405
+    },
+    {
+      "epoch": 2.444626381585162,
+      "grad_norm": 0.23711274564266205,
+      "learning_rate": 3.023003693517908e-05,
+      "loss": 0.1637,
+      "step": 19410
+    },
+    {
+      "epoch": 2.4452561639953396,
+      "grad_norm": 0.2252456545829773,
+      "learning_rate": 3.0163891270404904e-05,
+      "loss": 0.1685,
+      "step": 19415
+    },
+    {
+      "epoch": 2.4458859464055167,
+      "grad_norm": 0.2477557361125946,
+      "learning_rate": 3.0097809960358427e-05,
+      "loss": 0.1669,
+      "step": 19420
+    },
+    {
+      "epoch": 2.4465157288156942,
+      "grad_norm": 0.21543872356414795,
+      "learning_rate": 3.003179304052689e-05,
+      "loss": 0.1624,
+      "step": 19425
+    },
+    {
+      "epoch": 2.4471455112258713,
+      "grad_norm": 0.18810850381851196,
+      "learning_rate": 2.9965840546362858e-05,
+      "loss": 0.1531,
+      "step": 19430
+    },
+    {
+      "epoch": 2.447775293636049,
+      "grad_norm": 0.2468540370464325,
+      "learning_rate": 2.9899952513284307e-05,
+      "loss": 0.1644,
+      "step": 19435
+    },
+    {
+      "epoch": 2.448405076046226,
+      "grad_norm": 0.2639712393283844,
+      "learning_rate": 2.9834128976674643e-05,
+      "loss": 0.166,
+      "step": 19440
+    },
+    {
+      "epoch": 2.4490348584564035,
+      "grad_norm": 0.18254607915878296,
+      "learning_rate": 2.9768369971882598e-05,
+      "loss": 0.1478,
+      "step": 19445
+    },
+    {
+      "epoch": 2.4496646408665805,
+      "grad_norm": 0.2060953974723816,
+      "learning_rate": 2.9702675534222265e-05,
+      "loss": 0.161,
+      "step": 19450
+    },
+    {
+      "epoch": 2.450294423276758,
+      "grad_norm": 0.20919503271579742,
+      "learning_rate": 2.963704569897305e-05,
+      "loss": 0.1635,
+      "step": 19455
+    },
+    {
+      "epoch": 2.450924205686935,
+      "grad_norm": 0.20381583273410797,
+      "learning_rate": 2.957148050137963e-05,
+      "loss": 0.1677,
+      "step": 19460
+    },
+    {
+      "epoch": 2.4515539880971122,
+      "grad_norm": 0.23379412293434143,
+      "learning_rate": 2.9505979976652106e-05,
+      "loss": 0.1669,
+      "step": 19465
+    },
+    {
+      "epoch": 2.4521837705072898,
+      "grad_norm": 0.21713408827781677,
+      "learning_rate": 2.9440544159965707e-05,
+      "loss": 0.1639,
+      "step": 19470
+    },
+    {
+      "epoch": 2.452813552917467,
+      "grad_norm": 0.2360960692167282,
+      "learning_rate": 2.9375173086460975e-05,
+      "loss": 0.1682,
+      "step": 19475
+    },
+    {
+      "epoch": 2.4534433353276444,
+      "grad_norm": 0.21496212482452393,
+      "learning_rate": 2.9309866791243643e-05,
+      "loss": 0.1508,
+      "step": 19480
+    },
+    {
+      "epoch": 2.4540731177378214,
+      "grad_norm": 0.19526614248752594,
+      "learning_rate": 2.9244625309384706e-05,
+      "loss": 0.1607,
+      "step": 19485
+    },
+    {
+      "epoch": 2.454702900147999,
+      "grad_norm": 0.2625288665294647,
+      "learning_rate": 2.917944867592031e-05,
+      "loss": 0.1708,
+      "step": 19490
+    },
+    {
+      "epoch": 2.455332682558176,
+      "grad_norm": 0.19196555018424988,
+      "learning_rate": 2.9114336925851818e-05,
+      "loss": 0.1715,
+      "step": 19495
+    },
+    {
+      "epoch": 2.4559624649683536,
+      "grad_norm": 0.21597431600093842,
+      "learning_rate": 2.9049290094145726e-05,
+      "loss": 0.1508,
+      "step": 19500
+    },
+    {
+      "epoch": 2.4565922473785307,
+      "grad_norm": 0.2433023750782013,
+      "learning_rate": 2.8984308215733615e-05,
+      "loss": 0.1568,
+      "step": 19505
+    },
+    {
+      "epoch": 2.457222029788708,
+      "grad_norm": 0.231834277510643,
+      "learning_rate": 2.8919391325512314e-05,
+      "loss": 0.1552,
+      "step": 19510
+    },
+    {
+      "epoch": 2.4578518121988853,
+      "grad_norm": 0.21281488239765167,
+      "learning_rate": 2.885453945834369e-05,
+      "loss": 0.161,
+      "step": 19515
+    },
+    {
+      "epoch": 2.4584815946090623,
+      "grad_norm": 0.21355679631233215,
+      "learning_rate": 2.878975264905455e-05,
+      "loss": 0.1515,
+      "step": 19520
+    },
+    {
+      "epoch": 2.45911137701924,
+      "grad_norm": 0.20718532800674438,
+      "learning_rate": 2.8725030932437025e-05,
+      "loss": 0.1622,
+      "step": 19525
+    },
+    {
+      "epoch": 2.459741159429417,
+      "grad_norm": 0.21609242260456085,
+      "learning_rate": 2.8660374343248087e-05,
+      "loss": 0.1531,
+      "step": 19530
+    },
+    {
+      "epoch": 2.4603709418395945,
+      "grad_norm": 0.2453998625278473,
+      "learning_rate": 2.8595782916209825e-05,
+      "loss": 0.1605,
+      "step": 19535
+    },
+    {
+      "epoch": 2.4610007242497716,
+      "grad_norm": 0.27632614970207214,
+      "learning_rate": 2.8531256686009306e-05,
+      "loss": 0.1598,
+      "step": 19540
+    },
+    {
+      "epoch": 2.461630506659949,
+      "grad_norm": 0.19357621669769287,
+      "learning_rate": 2.846679568729855e-05,
+      "loss": 0.1527,
+      "step": 19545
+    },
+    {
+      "epoch": 2.462260289070126,
+      "grad_norm": 0.19920161366462708,
+      "learning_rate": 2.8402399954694692e-05,
+      "loss": 0.1561,
+      "step": 19550
+    },
+    {
+      "epoch": 2.4628900714803037,
+      "grad_norm": 0.19081860780715942,
+      "learning_rate": 2.8338069522779595e-05,
+      "loss": 0.1524,
+      "step": 19555
+    },
+    {
+      "epoch": 2.463519853890481,
+      "grad_norm": 0.22451332211494446,
+      "learning_rate": 2.8273804426100234e-05,
+      "loss": 0.1628,
+      "step": 19560
+    },
+    {
+      "epoch": 2.4641496363006583,
+      "grad_norm": 0.19204290211200714,
+      "learning_rate": 2.820960469916837e-05,
+      "loss": 0.1499,
+      "step": 19565
+    },
+    {
+      "epoch": 2.4647794187108354,
+      "grad_norm": 0.20258976519107819,
+      "learning_rate": 2.814547037646081e-05,
+      "loss": 0.1514,
+      "step": 19570
+    },
+    {
+      "epoch": 2.4654092011210125,
+      "grad_norm": 0.21591047942638397,
+      "learning_rate": 2.8081401492419102e-05,
+      "loss": 0.1555,
+      "step": 19575
+    },
+    {
+      "epoch": 2.46603898353119,
+      "grad_norm": 0.20639857649803162,
+      "learning_rate": 2.8017398081449728e-05,
+      "loss": 0.1597,
+      "step": 19580
+    },
+    {
+      "epoch": 2.466668765941367,
+      "grad_norm": 0.18190859258174896,
+      "learning_rate": 2.7953460177923953e-05,
+      "loss": 0.1676,
+      "step": 19585
+    },
+    {
+      "epoch": 2.4672985483515446,
+      "grad_norm": 0.23196272552013397,
+      "learning_rate": 2.7889587816177884e-05,
+      "loss": 0.1644,
+      "step": 19590
+    },
+    {
+      "epoch": 2.4679283307617217,
+      "grad_norm": 0.23499402403831482,
+      "learning_rate": 2.782578103051248e-05,
+      "loss": 0.1596,
+      "step": 19595
+    },
+    {
+      "epoch": 2.468558113171899,
+      "grad_norm": 0.19516189396381378,
+      "learning_rate": 2.7762039855193398e-05,
+      "loss": 0.1592,
+      "step": 19600
+    },
+    {
+      "epoch": 2.4691878955820763,
+      "grad_norm": 0.25550252199172974,
+      "learning_rate": 2.769836432445109e-05,
+      "loss": 0.1652,
+      "step": 19605
+    },
+    {
+      "epoch": 2.469817677992254,
+      "grad_norm": 0.20900960266590118,
+      "learning_rate": 2.7634754472480852e-05,
+      "loss": 0.1576,
+      "step": 19610
+    },
+    {
+      "epoch": 2.470447460402431,
+      "grad_norm": 0.19483284652233124,
+      "learning_rate": 2.757121033344258e-05,
+      "loss": 0.1671,
+      "step": 19615
+    },
+    {
+      "epoch": 2.4710772428126084,
+      "grad_norm": 0.21054719388484955,
+      "learning_rate": 2.7507731941460952e-05,
+      "loss": 0.1572,
+      "step": 19620
+    },
+    {
+      "epoch": 2.4717070252227855,
+      "grad_norm": 0.23577210307121277,
+      "learning_rate": 2.7444319330625243e-05,
+      "loss": 0.1657,
+      "step": 19625
+    },
+    {
+      "epoch": 2.4723368076329626,
+      "grad_norm": 0.21099181473255157,
+      "learning_rate": 2.7380972534989538e-05,
+      "loss": 0.1696,
+      "step": 19630
+    },
+    {
+      "epoch": 2.47296659004314,
+      "grad_norm": 0.20165832340717316,
+      "learning_rate": 2.7317691588572495e-05,
+      "loss": 0.1529,
+      "step": 19635
+    },
+    {
+      "epoch": 2.473596372453317,
+      "grad_norm": 0.19725088775157928,
+      "learning_rate": 2.7254476525357443e-05,
+      "loss": 0.1503,
+      "step": 19640
+    },
+    {
+      "epoch": 2.4742261548634947,
+      "grad_norm": 0.23867055773735046,
+      "learning_rate": 2.7191327379292283e-05,
+      "loss": 0.1766,
+      "step": 19645
+    },
+    {
+      "epoch": 2.474855937273672,
+      "grad_norm": 0.21567271649837494,
+      "learning_rate": 2.712824418428955e-05,
+      "loss": 0.1562,
+      "step": 19650
+    },
+    {
+      "epoch": 2.4754857196838493,
+      "grad_norm": 0.2127571702003479,
+      "learning_rate": 2.7065226974226444e-05,
+      "loss": 0.1588,
+      "step": 19655
+    },
+    {
+      "epoch": 2.4761155020940264,
+      "grad_norm": 0.192424476146698,
+      "learning_rate": 2.700227578294455e-05,
+      "loss": 0.1632,
+      "step": 19660
+    },
+    {
+      "epoch": 2.476745284504204,
+      "grad_norm": 0.19549550116062164,
+      "learning_rate": 2.693939064425007e-05,
+      "loss": 0.1666,
+      "step": 19665
+    },
+    {
+      "epoch": 2.477375066914381,
+      "grad_norm": 0.21715867519378662,
+      "learning_rate": 2.6876571591913874e-05,
+      "loss": 0.1637,
+      "step": 19670
+    },
+    {
+      "epoch": 2.4780048493245586,
+      "grad_norm": 0.246476948261261,
+      "learning_rate": 2.6813818659671167e-05,
+      "loss": 0.1691,
+      "step": 19675
+    },
+    {
+      "epoch": 2.4786346317347356,
+      "grad_norm": 0.19329246878623962,
+      "learning_rate": 2.6751131881221698e-05,
+      "loss": 0.1576,
+      "step": 19680
+    },
+    {
+      "epoch": 2.4792644141449127,
+      "grad_norm": 0.1897173672914505,
+      "learning_rate": 2.6688511290229714e-05,
+      "loss": 0.1566,
+      "step": 19685
+    },
+    {
+      "epoch": 2.4798941965550902,
+      "grad_norm": 0.19795387983322144,
+      "learning_rate": 2.662595692032391e-05,
+      "loss": 0.159,
+      "step": 19690
+    },
+    {
+      "epoch": 2.4805239789652673,
+      "grad_norm": 0.19520628452301025,
+      "learning_rate": 2.65634688050974e-05,
+      "loss": 0.1577,
+      "step": 19695
+    },
+    {
+      "epoch": 2.481153761375445,
+      "grad_norm": 0.21223746240139008,
+      "learning_rate": 2.650104697810772e-05,
+      "loss": 0.1674,
+      "step": 19700
+    },
+    {
+      "epoch": 2.481783543785622,
+      "grad_norm": 0.19204822182655334,
+      "learning_rate": 2.6438691472876828e-05,
+      "loss": 0.1492,
+      "step": 19705
+    },
+    {
+      "epoch": 2.4824133261957995,
+      "grad_norm": 0.2568466067314148,
+      "learning_rate": 2.6376402322891032e-05,
+      "loss": 0.1557,
+      "step": 19710
+    },
+    {
+      "epoch": 2.4830431086059765,
+      "grad_norm": 0.21695761382579803,
+      "learning_rate": 2.6314179561601078e-05,
+      "loss": 0.1715,
+      "step": 19715
+    },
+    {
+      "epoch": 2.483672891016154,
+      "grad_norm": 0.21485815942287445,
+      "learning_rate": 2.625202322242197e-05,
+      "loss": 0.1599,
+      "step": 19720
+    },
+    {
+      "epoch": 2.484302673426331,
+      "grad_norm": 0.18373069167137146,
+      "learning_rate": 2.6189933338733122e-05,
+      "loss": 0.1636,
+      "step": 19725
+    },
+    {
+      "epoch": 2.4849324558365087,
+      "grad_norm": 0.2190975546836853,
+      "learning_rate": 2.6127909943878177e-05,
+      "loss": 0.1613,
+      "step": 19730
+    },
+    {
+      "epoch": 2.4855622382466858,
+      "grad_norm": 0.22146424651145935,
+      "learning_rate": 2.606595307116513e-05,
+      "loss": 0.1554,
+      "step": 19735
+    },
+    {
+      "epoch": 2.486192020656863,
+      "grad_norm": 0.22576889395713806,
+      "learning_rate": 2.6004062753866228e-05,
+      "loss": 0.1723,
+      "step": 19740
+    },
+    {
+      "epoch": 2.4868218030670404,
+      "grad_norm": 0.22661438584327698,
+      "learning_rate": 2.5942239025218004e-05,
+      "loss": 0.1616,
+      "step": 19745
+    },
+    {
+      "epoch": 2.4874515854772175,
+      "grad_norm": 0.20992781221866608,
+      "learning_rate": 2.588048191842118e-05,
+      "loss": 0.1666,
+      "step": 19750
+    },
+    {
+      "epoch": 2.488081367887395,
+      "grad_norm": 0.18685118854045868,
+      "learning_rate": 2.581879146664078e-05,
+      "loss": 0.163,
+      "step": 19755
+    },
+    {
+      "epoch": 2.488711150297572,
+      "grad_norm": 0.2547582983970642,
+      "learning_rate": 2.5757167703005987e-05,
+      "loss": 0.1683,
+      "step": 19760
+    },
+    {
+      "epoch": 2.4893409327077496,
+      "grad_norm": 0.18066510558128357,
+      "learning_rate": 2.569561066061013e-05,
+      "loss": 0.1581,
+      "step": 19765
+    },
+    {
+      "epoch": 2.4899707151179267,
+      "grad_norm": 0.22709952294826508,
+      "learning_rate": 2.5634120372510708e-05,
+      "loss": 0.1655,
+      "step": 19770
+    },
+    {
+      "epoch": 2.490600497528104,
+      "grad_norm": 0.18300481140613556,
+      "learning_rate": 2.5572696871729496e-05,
+      "loss": 0.1634,
+      "step": 19775
+    },
+    {
+      "epoch": 2.4912302799382813,
+      "grad_norm": 0.23889437317848206,
+      "learning_rate": 2.5511340191252294e-05,
+      "loss": 0.1653,
+      "step": 19780
+    },
+    {
+      "epoch": 2.491860062348459,
+      "grad_norm": 0.18972428143024445,
+      "learning_rate": 2.545005036402904e-05,
+      "loss": 0.1522,
+      "step": 19785
+    },
+    {
+      "epoch": 2.492489844758636,
+      "grad_norm": 0.1869877278804779,
+      "learning_rate": 2.5388827422973722e-05,
+      "loss": 0.1587,
+      "step": 19790
+    },
+    {
+      "epoch": 2.493119627168813,
+      "grad_norm": 0.200529083609581,
+      "learning_rate": 2.5327671400964562e-05,
+      "loss": 0.1621,
+      "step": 19795
+    },
+    {
+      "epoch": 2.4937494095789905,
+      "grad_norm": 0.20414294302463531,
+      "learning_rate": 2.526658233084365e-05,
+      "loss": 0.1619,
+      "step": 19800
+    },
+    {
+      "epoch": 2.4943791919891676,
+      "grad_norm": 0.2506503760814667,
+      "learning_rate": 2.5205560245417227e-05,
+      "loss": 0.1711,
+      "step": 19805
+    },
+    {
+      "epoch": 2.495008974399345,
+      "grad_norm": 0.2258518785238266,
+      "learning_rate": 2.5144605177455534e-05,
+      "loss": 0.1718,
+      "step": 19810
+    },
+    {
+      "epoch": 2.495638756809522,
+      "grad_norm": 0.22719348967075348,
+      "learning_rate": 2.5083717159692902e-05,
+      "loss": 0.1611,
+      "step": 19815
+    },
+    {
+      "epoch": 2.4962685392196997,
+      "grad_norm": 0.18670164048671722,
+      "learning_rate": 2.502289622482752e-05,
+      "loss": 0.155,
+      "step": 19820
+    },
+    {
+      "epoch": 2.496898321629877,
+      "grad_norm": 0.19051162898540497,
+      "learning_rate": 2.4962142405521666e-05,
+      "loss": 0.1528,
+      "step": 19825
+    },
+    {
+      "epoch": 2.4975281040400543,
+      "grad_norm": 0.2364228218793869,
+      "learning_rate": 2.4901455734401508e-05,
+      "loss": 0.1642,
+      "step": 19830
+    },
+    {
+      "epoch": 2.4981578864502314,
+      "grad_norm": 0.1748083382844925,
+      "learning_rate": 2.484083624405716e-05,
+      "loss": 0.1536,
+      "step": 19835
+    },
+    {
+      "epoch": 2.498787668860409,
+      "grad_norm": 0.21124523878097534,
+      "learning_rate": 2.4780283967042697e-05,
+      "loss": 0.1641,
+      "step": 19840
+    },
+    {
+      "epoch": 2.499417451270586,
+      "grad_norm": 0.21559958159923553,
+      "learning_rate": 2.4719798935876073e-05,
+      "loss": 0.1522,
+      "step": 19845
+    },
+    {
+      "epoch": 2.500047233680763,
+      "grad_norm": 0.20545977354049683,
+      "learning_rate": 2.4659381183039105e-05,
+      "loss": 0.1492,
+      "step": 19850
+    },
+    {
+      "epoch": 2.5006770160909406,
+      "grad_norm": 0.21759046614170074,
+      "learning_rate": 2.459903074097749e-05,
+      "loss": 0.1637,
+      "step": 19855
+    },
+    {
+      "epoch": 2.5013067985011177,
+      "grad_norm": 0.2807125151157379,
+      "learning_rate": 2.4538747642100927e-05,
+      "loss": 0.1701,
+      "step": 19860
+    },
+    {
+      "epoch": 2.5019365809112952,
+      "grad_norm": 0.1915740966796875,
+      "learning_rate": 2.4478531918782656e-05,
+      "loss": 0.1551,
+      "step": 19865
+    },
+    {
+      "epoch": 2.5025663633214723,
+      "grad_norm": 0.1929636150598526,
+      "learning_rate": 2.441838360335992e-05,
+      "loss": 0.1561,
+      "step": 19870
+    },
+    {
+      "epoch": 2.50319614573165,
+      "grad_norm": 0.23392513394355774,
+      "learning_rate": 2.4358302728133827e-05,
+      "loss": 0.1606,
+      "step": 19875
+    },
+    {
+      "epoch": 2.503825928141827,
+      "grad_norm": 0.21680179238319397,
+      "learning_rate": 2.4298289325369137e-05,
+      "loss": 0.166,
+      "step": 19880
+    },
+    {
+      "epoch": 2.5044557105520044,
+      "grad_norm": 0.20863774418830872,
+      "learning_rate": 2.42383434272944e-05,
+      "loss": 0.153,
+      "step": 19885
+    },
+    {
+      "epoch": 2.5050854929621815,
+      "grad_norm": 0.2562030851840973,
+      "learning_rate": 2.4178465066101933e-05,
+      "loss": 0.1591,
+      "step": 19890
+    },
+    {
+      "epoch": 2.505715275372359,
+      "grad_norm": 0.22802165150642395,
+      "learning_rate": 2.4118654273947796e-05,
+      "loss": 0.1664,
+      "step": 19895
+    },
+    {
+      "epoch": 2.506345057782536,
+      "grad_norm": 0.23240098357200623,
+      "learning_rate": 2.4058911082951764e-05,
+      "loss": 0.1585,
+      "step": 19900
+    },
+    {
+      "epoch": 2.506974840192713,
+      "grad_norm": 0.21342670917510986,
+      "learning_rate": 2.3999235525197275e-05,
+      "loss": 0.1471,
+      "step": 19905
+    },
+    {
+      "epoch": 2.5076046226028907,
+      "grad_norm": 0.20659485459327698,
+      "learning_rate": 2.3939627632731458e-05,
+      "loss": 0.1593,
+      "step": 19910
+    },
+    {
+      "epoch": 2.508234405013068,
+      "grad_norm": 0.21293510496616364,
+      "learning_rate": 2.3880087437565104e-05,
+      "loss": 0.1575,
+      "step": 19915
+    },
+    {
+      "epoch": 2.5088641874232454,
+      "grad_norm": 0.2346251904964447,
+      "learning_rate": 2.382061497167271e-05,
+      "loss": 0.1639,
+      "step": 19920
+    },
+    {
+      "epoch": 2.5094939698334224,
+      "grad_norm": 0.22029395401477814,
+      "learning_rate": 2.376121026699232e-05,
+      "loss": 0.1537,
+      "step": 19925
+    },
+    {
+      "epoch": 2.5101237522436,
+      "grad_norm": 0.1979423314332962,
+      "learning_rate": 2.3701873355425606e-05,
+      "loss": 0.154,
+      "step": 19930
+    },
+    {
+      "epoch": 2.510753534653777,
+      "grad_norm": 0.1969837099313736,
+      "learning_rate": 2.3642604268837873e-05,
+      "loss": 0.1623,
+      "step": 19935
+    },
+    {
+      "epoch": 2.5113833170639546,
+      "grad_norm": 0.23190250992774963,
+      "learning_rate": 2.3583403039057946e-05,
+      "loss": 0.1673,
+      "step": 19940
+    },
+    {
+      "epoch": 2.5120130994741316,
+      "grad_norm": 0.20579595863819122,
+      "learning_rate": 2.3524269697878244e-05,
+      "loss": 0.1638,
+      "step": 19945
+    },
+    {
+      "epoch": 2.512642881884309,
+      "grad_norm": 0.2181597501039505,
+      "learning_rate": 2.3465204277054734e-05,
+      "loss": 0.1535,
+      "step": 19950
+    },
+    {
+      "epoch": 2.5132726642944863,
+      "grad_norm": 0.27504584193229675,
+      "learning_rate": 2.3406206808306854e-05,
+      "loss": 0.1687,
+      "step": 19955
+    },
+    {
+      "epoch": 2.5139024467046633,
+      "grad_norm": 0.21288301050662994,
+      "learning_rate": 2.334727732331765e-05,
+      "loss": 0.1611,
+      "step": 19960
+    },
+    {
+      "epoch": 2.514532229114841,
+      "grad_norm": 0.20768193900585175,
+      "learning_rate": 2.3288415853733615e-05,
+      "loss": 0.1595,
+      "step": 19965
+    },
+    {
+      "epoch": 2.515162011525018,
+      "grad_norm": 0.1934243142604828,
+      "learning_rate": 2.322962243116464e-05,
+      "loss": 0.1573,
+      "step": 19970
+    },
+    {
+      "epoch": 2.5157917939351955,
+      "grad_norm": 0.21198545396327972,
+      "learning_rate": 2.3170897087184133e-05,
+      "loss": 0.1549,
+      "step": 19975
+    },
+    {
+      "epoch": 2.5164215763453726,
+      "grad_norm": 0.25898632407188416,
+      "learning_rate": 2.3112239853328996e-05,
+      "loss": 0.171,
+      "step": 19980
+    },
+    {
+      "epoch": 2.51705135875555,
+      "grad_norm": 0.2114986777305603,
+      "learning_rate": 2.3053650761099485e-05,
+      "loss": 0.1544,
+      "step": 19985
+    },
+    {
+      "epoch": 2.517681141165727,
+      "grad_norm": 0.21560825407505035,
+      "learning_rate": 2.2995129841959266e-05,
+      "loss": 0.1736,
+      "step": 19990
+    },
+    {
+      "epoch": 2.5183109235759042,
+      "grad_norm": 0.18198496103286743,
+      "learning_rate": 2.2936677127335395e-05,
+      "loss": 0.154,
+      "step": 19995
+    },
+    {
+      "epoch": 2.5189407059860818,
+      "grad_norm": 0.20246680080890656,
+      "learning_rate": 2.287829264861842e-05,
+      "loss": 0.1598,
+      "step": 20000
+    },
+    {
+      "epoch": 2.5189407059860818,
+      "eval_loss": 0.3499235212802887,
+      "eval_runtime": 6.1623,
+      "eval_samples_per_second": 162.278,
+      "eval_steps_per_second": 10.224,
+      "step": 20000
+    },
+    {
+      "epoch": 2.5195704883962593,
+      "grad_norm": 0.2162911742925644,
+      "learning_rate": 2.2819976437162e-05,
+      "loss": 0.1623,
+      "step": 20005
+    },
+    {
+      "epoch": 2.5202002708064364,
+      "grad_norm": 0.16897226870059967,
+      "learning_rate": 2.2761728524283344e-05,
+      "loss": 0.1511,
+      "step": 20010
+    },
+    {
+      "epoch": 2.5208300532166135,
+      "grad_norm": 0.19399495422840118,
+      "learning_rate": 2.2703548941262877e-05,
+      "loss": 0.1615,
+      "step": 20015
+    },
+    {
+      "epoch": 2.521459835626791,
+      "grad_norm": 0.24832330644130707,
+      "learning_rate": 2.2645437719344424e-05,
+      "loss": 0.1596,
+      "step": 20020
+    },
+    {
+      "epoch": 2.522089618036968,
+      "grad_norm": 0.1990746706724167,
+      "learning_rate": 2.2587394889734982e-05,
+      "loss": 0.1517,
+      "step": 20025
+    },
+    {
+      "epoch": 2.5227194004471456,
+      "grad_norm": 0.2356463521718979,
+      "learning_rate": 2.252942048360491e-05,
+      "loss": 0.1666,
+      "step": 20030
+    },
+    {
+      "epoch": 2.5233491828573227,
+      "grad_norm": 0.2032928168773651,
+      "learning_rate": 2.2471514532087766e-05,
+      "loss": 0.1463,
+      "step": 20035
+    },
+    {
+      "epoch": 2.5239789652675,
+      "grad_norm": 0.24515411257743835,
+      "learning_rate": 2.2413677066280388e-05,
+      "loss": 0.1629,
+      "step": 20040
+    },
+    {
+      "epoch": 2.5246087476776773,
+      "grad_norm": 0.24597153067588806,
+      "learning_rate": 2.2355908117242803e-05,
+      "loss": 0.1507,
+      "step": 20045
+    },
+    {
+      "epoch": 2.5252385300878544,
+      "grad_norm": 0.1958838254213333,
+      "learning_rate": 2.2298207715998246e-05,
+      "loss": 0.167,
+      "step": 20050
+    },
+    {
+      "epoch": 2.525868312498032,
+      "grad_norm": 0.18343359231948853,
+      "learning_rate": 2.2240575893533176e-05,
+      "loss": 0.1582,
+      "step": 20055
+    },
+    {
+      "epoch": 2.5264980949082094,
+      "grad_norm": 0.2554282248020172,
+      "learning_rate": 2.218301268079715e-05,
+      "loss": 0.1701,
+      "step": 20060
+    },
+    {
+      "epoch": 2.5271278773183865,
+      "grad_norm": 0.2655259072780609,
+      "learning_rate": 2.2125518108703e-05,
+      "loss": 0.1666,
+      "step": 20065
+    },
+    {
+      "epoch": 2.5277576597285636,
+      "grad_norm": 0.23147699236869812,
+      "learning_rate": 2.206809220812662e-05,
+      "loss": 0.1646,
+      "step": 20070
+    },
+    {
+      "epoch": 2.528387442138741,
+      "grad_norm": 0.19453732669353485,
+      "learning_rate": 2.2010735009906926e-05,
+      "loss": 0.1595,
+      "step": 20075
+    },
+    {
+      "epoch": 2.529017224548918,
+      "grad_norm": 0.21716727316379547,
+      "learning_rate": 2.195344654484615e-05,
+      "loss": 0.1656,
+      "step": 20080
+    },
+    {
+      "epoch": 2.5296470069590957,
+      "grad_norm": 0.19851936399936676,
+      "learning_rate": 2.1896226843709475e-05,
+      "loss": 0.1545,
+      "step": 20085
+    },
+    {
+      "epoch": 2.530276789369273,
+      "grad_norm": 0.20362606644630432,
+      "learning_rate": 2.1839075937225192e-05,
+      "loss": 0.1534,
+      "step": 20090
+    },
+    {
+      "epoch": 2.5309065717794503,
+      "grad_norm": 0.23197387158870697,
+      "learning_rate": 2.1781993856084633e-05,
+      "loss": 0.1624,
+      "step": 20095
+    },
+    {
+      "epoch": 2.5315363541896274,
+      "grad_norm": 0.2547961473464966,
+      "learning_rate": 2.1724980630942145e-05,
+      "loss": 0.1539,
+      "step": 20100
+    },
+    {
+      "epoch": 2.5321661365998045,
+      "grad_norm": 0.1867532879114151,
+      "learning_rate": 2.1668036292415237e-05,
+      "loss": 0.1518,
+      "step": 20105
+    },
+    {
+      "epoch": 2.532795919009982,
+      "grad_norm": 0.19402964413166046,
+      "learning_rate": 2.161116087108421e-05,
+      "loss": 0.1522,
+      "step": 20110
+    },
+    {
+      "epoch": 2.5334257014201595,
+      "grad_norm": 0.20450226962566376,
+      "learning_rate": 2.1554354397492517e-05,
+      "loss": 0.155,
+      "step": 20115
+    },
+    {
+      "epoch": 2.5340554838303366,
+      "grad_norm": 0.22179925441741943,
+      "learning_rate": 2.149761690214649e-05,
+      "loss": 0.1557,
+      "step": 20120
+    },
+    {
+      "epoch": 2.5346852662405137,
+      "grad_norm": 0.2105506807565689,
+      "learning_rate": 2.1440948415515524e-05,
+      "loss": 0.1668,
+      "step": 20125
+    },
+    {
+      "epoch": 2.5353150486506912,
+      "grad_norm": 0.24963414669036865,
+      "learning_rate": 2.1384348968031857e-05,
+      "loss": 0.1597,
+      "step": 20130
+    },
+    {
+      "epoch": 2.5359448310608683,
+      "grad_norm": 0.23433445394039154,
+      "learning_rate": 2.132781859009069e-05,
+      "loss": 0.1579,
+      "step": 20135
+    },
+    {
+      "epoch": 2.536574613471046,
+      "grad_norm": 0.19620360434055328,
+      "learning_rate": 2.1271357312050126e-05,
+      "loss": 0.1492,
+      "step": 20140
+    },
+    {
+      "epoch": 2.537204395881223,
+      "grad_norm": 0.23040203750133514,
+      "learning_rate": 2.1214965164231157e-05,
+      "loss": 0.1585,
+      "step": 20145
+    },
+    {
+      "epoch": 2.5378341782914005,
+      "grad_norm": 0.23273873329162598,
+      "learning_rate": 2.1158642176917647e-05,
+      "loss": 0.1589,
+      "step": 20150
+    },
+    {
+      "epoch": 2.5384639607015775,
+      "grad_norm": 0.2472730576992035,
+      "learning_rate": 2.1102388380356344e-05,
+      "loss": 0.1677,
+      "step": 20155
+    },
+    {
+      "epoch": 2.5390937431117546,
+      "grad_norm": 0.19982990622520447,
+      "learning_rate": 2.104620380475679e-05,
+      "loss": 0.1515,
+      "step": 20160
+    },
+    {
+      "epoch": 2.539723525521932,
+      "grad_norm": 0.21257297694683075,
+      "learning_rate": 2.099008848029143e-05,
+      "loss": 0.165,
+      "step": 20165
+    },
+    {
+      "epoch": 2.5403533079321097,
+      "grad_norm": 0.20112313330173492,
+      "learning_rate": 2.0934042437095457e-05,
+      "loss": 0.1497,
+      "step": 20170
+    },
+    {
+      "epoch": 2.5409830903422868,
+      "grad_norm": 0.24434730410575867,
+      "learning_rate": 2.087806570526691e-05,
+      "loss": 0.1583,
+      "step": 20175
+    },
+    {
+      "epoch": 2.541612872752464,
+      "grad_norm": 0.20866596698760986,
+      "learning_rate": 2.0822158314866467e-05,
+      "loss": 0.1584,
+      "step": 20180
+    },
+    {
+      "epoch": 2.5422426551626414,
+      "grad_norm": 0.1903751641511917,
+      "learning_rate": 2.076632029591777e-05,
+      "loss": 0.1447,
+      "step": 20185
+    },
+    {
+      "epoch": 2.5428724375728184,
+      "grad_norm": 0.24377766251564026,
+      "learning_rate": 2.071055167840709e-05,
+      "loss": 0.1636,
+      "step": 20190
+    },
+    {
+      "epoch": 2.543502219982996,
+      "grad_norm": 0.25960245728492737,
+      "learning_rate": 2.0654852492283446e-05,
+      "loss": 0.164,
+      "step": 20195
+    },
+    {
+      "epoch": 2.544132002393173,
+      "grad_norm": 0.23870185017585754,
+      "learning_rate": 2.0599222767458533e-05,
+      "loss": 0.1579,
+      "step": 20200
+    },
+    {
+      "epoch": 2.5447617848033506,
+      "grad_norm": 0.2245192676782608,
+      "learning_rate": 2.0543662533806855e-05,
+      "loss": 0.1655,
+      "step": 20205
+    },
+    {
+      "epoch": 2.5453915672135277,
+      "grad_norm": 0.23136839270591736,
+      "learning_rate": 2.048817182116554e-05,
+      "loss": 0.1591,
+      "step": 20210
+    },
+    {
+      "epoch": 2.5460213496237047,
+      "grad_norm": 0.21092520654201508,
+      "learning_rate": 2.043275065933427e-05,
+      "loss": 0.1536,
+      "step": 20215
+    },
+    {
+      "epoch": 2.5466511320338823,
+      "grad_norm": 0.18601630628108978,
+      "learning_rate": 2.0377399078075485e-05,
+      "loss": 0.1523,
+      "step": 20220
+    },
+    {
+      "epoch": 2.54728091444406,
+      "grad_norm": 0.21489211916923523,
+      "learning_rate": 2.0322117107114343e-05,
+      "loss": 0.1554,
+      "step": 20225
+    },
+    {
+      "epoch": 2.547910696854237,
+      "grad_norm": 0.2098049521446228,
+      "learning_rate": 2.026690477613845e-05,
+      "loss": 0.1522,
+      "step": 20230
+    },
+    {
+      "epoch": 2.548540479264414,
+      "grad_norm": 0.2187887281179428,
+      "learning_rate": 2.021176211479813e-05,
+      "loss": 0.15,
+      "step": 20235
+    },
+    {
+      "epoch": 2.5491702616745915,
+      "grad_norm": 0.2641262710094452,
+      "learning_rate": 2.0156689152706216e-05,
+      "loss": 0.1725,
+      "step": 20240
+    },
+    {
+      "epoch": 2.5498000440847686,
+      "grad_norm": 0.22713615000247955,
+      "learning_rate": 2.010168591943817e-05,
+      "loss": 0.1528,
+      "step": 20245
+    },
+    {
+      "epoch": 2.550429826494946,
+      "grad_norm": 0.20724020898342133,
+      "learning_rate": 2.0046752444531976e-05,
+      "loss": 0.1646,
+      "step": 20250
+    },
+    {
+      "epoch": 2.551059608905123,
+      "grad_norm": 0.19516219198703766,
+      "learning_rate": 1.9991888757488156e-05,
+      "loss": 0.1574,
+      "step": 20255
+    },
+    {
+      "epoch": 2.5516893913153007,
+      "grad_norm": 0.22299246490001678,
+      "learning_rate": 1.993709488776979e-05,
+      "loss": 0.1656,
+      "step": 20260
+    },
+    {
+      "epoch": 2.552319173725478,
+      "grad_norm": 0.1897648572921753,
+      "learning_rate": 1.9882370864802373e-05,
+      "loss": 0.1639,
+      "step": 20265
+    },
+    {
+      "epoch": 2.552948956135655,
+      "grad_norm": 0.23607775568962097,
+      "learning_rate": 1.9827716717974048e-05,
+      "loss": 0.1618,
+      "step": 20270
+    },
+    {
+      "epoch": 2.5535787385458324,
+      "grad_norm": 0.250823438167572,
+      "learning_rate": 1.9773132476635285e-05,
+      "loss": 0.1628,
+      "step": 20275
+    },
+    {
+      "epoch": 2.55420852095601,
+      "grad_norm": 0.2012414038181305,
+      "learning_rate": 1.9718618170099087e-05,
+      "loss": 0.1536,
+      "step": 20280
+    },
+    {
+      "epoch": 2.554838303366187,
+      "grad_norm": 0.17350980639457703,
+      "learning_rate": 1.9664173827640873e-05,
+      "loss": 0.1524,
+      "step": 20285
+    },
+    {
+      "epoch": 2.555468085776364,
+      "grad_norm": 0.18761439621448517,
+      "learning_rate": 1.96097994784985e-05,
+      "loss": 0.1452,
+      "step": 20290
+    },
+    {
+      "epoch": 2.5560978681865416,
+      "grad_norm": 0.2061910331249237,
+      "learning_rate": 1.955549515187223e-05,
+      "loss": 0.1507,
+      "step": 20295
+    },
+    {
+      "epoch": 2.5567276505967187,
+      "grad_norm": 0.20667202770709991,
+      "learning_rate": 1.9501260876924736e-05,
+      "loss": 0.1484,
+      "step": 20300
+    },
+    {
+      "epoch": 2.557357433006896,
+      "grad_norm": 0.19904933869838715,
+      "learning_rate": 1.9447096682781015e-05,
+      "loss": 0.1562,
+      "step": 20305
+    },
+    {
+      "epoch": 2.5579872154170733,
+      "grad_norm": 0.20500166714191437,
+      "learning_rate": 1.9393002598528555e-05,
+      "loss": 0.1505,
+      "step": 20310
+    },
+    {
+      "epoch": 2.558616997827251,
+      "grad_norm": 0.21382258832454681,
+      "learning_rate": 1.933897865321712e-05,
+      "loss": 0.1606,
+      "step": 20315
+    },
+    {
+      "epoch": 2.559246780237428,
+      "grad_norm": 0.22117263078689575,
+      "learning_rate": 1.928502487585873e-05,
+      "loss": 0.1626,
+      "step": 20320
+    },
+    {
+      "epoch": 2.559876562647605,
+      "grad_norm": 0.2301877737045288,
+      "learning_rate": 1.9231141295427794e-05,
+      "loss": 0.1566,
+      "step": 20325
+    },
+    {
+      "epoch": 2.5605063450577825,
+      "grad_norm": 0.23893754184246063,
+      "learning_rate": 1.917732794086108e-05,
+      "loss": 0.1571,
+      "step": 20330
+    },
+    {
+      "epoch": 2.56113612746796,
+      "grad_norm": 0.2627946734428406,
+      "learning_rate": 1.9123584841057578e-05,
+      "loss": 0.1658,
+      "step": 20335
+    },
+    {
+      "epoch": 2.561765909878137,
+      "grad_norm": 0.19281533360481262,
+      "learning_rate": 1.906991202487854e-05,
+      "loss": 0.1525,
+      "step": 20340
+    },
+    {
+      "epoch": 2.562395692288314,
+      "grad_norm": 0.2772383689880371,
+      "learning_rate": 1.901630952114752e-05,
+      "loss": 0.1661,
+      "step": 20345
+    },
+    {
+      "epoch": 2.5630254746984917,
+      "grad_norm": 0.216465026140213,
+      "learning_rate": 1.896277735865027e-05,
+      "loss": 0.1538,
+      "step": 20350
+    },
+    {
+      "epoch": 2.563655257108669,
+      "grad_norm": 0.23878604173660278,
+      "learning_rate": 1.8909315566134782e-05,
+      "loss": 0.1601,
+      "step": 20355
+    },
+    {
+      "epoch": 2.5642850395188463,
+      "grad_norm": 0.2141411155462265,
+      "learning_rate": 1.8855924172311248e-05,
+      "loss": 0.1631,
+      "step": 20360
+    },
+    {
+      "epoch": 2.5649148219290234,
+      "grad_norm": 0.2064596712589264,
+      "learning_rate": 1.8802603205852073e-05,
+      "loss": 0.1578,
+      "step": 20365
+    },
+    {
+      "epoch": 2.565544604339201,
+      "grad_norm": 0.19963258504867554,
+      "learning_rate": 1.8749352695391867e-05,
+      "loss": 0.1609,
+      "step": 20370
+    },
+    {
+      "epoch": 2.566174386749378,
+      "grad_norm": 0.1846475601196289,
+      "learning_rate": 1.8696172669527336e-05,
+      "loss": 0.1544,
+      "step": 20375
+    },
+    {
+      "epoch": 2.566804169159555,
+      "grad_norm": 0.1911933869123459,
+      "learning_rate": 1.8643063156817423e-05,
+      "loss": 0.1565,
+      "step": 20380
+    },
+    {
+      "epoch": 2.5674339515697326,
+      "grad_norm": 0.2263742834329605,
+      "learning_rate": 1.8590024185783042e-05,
+      "loss": 0.1673,
+      "step": 20385
+    },
+    {
+      "epoch": 2.56806373397991,
+      "grad_norm": 0.21621178090572357,
+      "learning_rate": 1.8537055784907413e-05,
+      "loss": 0.1587,
+      "step": 20390
+    },
+    {
+      "epoch": 2.5686935163900873,
+      "grad_norm": 0.20875446498394012,
+      "learning_rate": 1.848415798263576e-05,
+      "loss": 0.1564,
+      "step": 20395
+    },
+    {
+      "epoch": 2.5693232988002643,
+      "grad_norm": 0.20144003629684448,
+      "learning_rate": 1.8431330807375417e-05,
+      "loss": 0.156,
+      "step": 20400
+    },
+    {
+      "epoch": 2.569953081210442,
+      "grad_norm": 0.24883227050304413,
+      "learning_rate": 1.837857428749575e-05,
+      "loss": 0.1568,
+      "step": 20405
+    },
+    {
+      "epoch": 2.570582863620619,
+      "grad_norm": 0.18426820635795593,
+      "learning_rate": 1.832588845132827e-05,
+      "loss": 0.1549,
+      "step": 20410
+    },
+    {
+      "epoch": 2.5712126460307965,
+      "grad_norm": 0.2462303191423416,
+      "learning_rate": 1.827327332716649e-05,
+      "loss": 0.1625,
+      "step": 20415
+    },
+    {
+      "epoch": 2.5718424284409735,
+      "grad_norm": 0.21659249067306519,
+      "learning_rate": 1.8220728943265837e-05,
+      "loss": 0.1521,
+      "step": 20420
+    },
+    {
+      "epoch": 2.572472210851151,
+      "grad_norm": 0.17811377346515656,
+      "learning_rate": 1.8168255327843882e-05,
+      "loss": 0.1586,
+      "step": 20425
+    },
+    {
+      "epoch": 2.573101993261328,
+      "grad_norm": 0.19524861872196198,
+      "learning_rate": 1.8115852509080197e-05,
+      "loss": 0.1474,
+      "step": 20430
+    },
+    {
+      "epoch": 2.5737317756715052,
+      "grad_norm": 0.17528071999549866,
+      "learning_rate": 1.806352051511627e-05,
+      "loss": 0.1599,
+      "step": 20435
+    },
+    {
+      "epoch": 2.5743615580816828,
+      "grad_norm": 0.19344571232795715,
+      "learning_rate": 1.801125937405557e-05,
+      "loss": 0.1579,
+      "step": 20440
+    },
+    {
+      "epoch": 2.5749913404918603,
+      "grad_norm": 0.20909984409809113,
+      "learning_rate": 1.795906911396353e-05,
+      "loss": 0.1584,
+      "step": 20445
+    },
+    {
+      "epoch": 2.5756211229020374,
+      "grad_norm": 0.17548586428165436,
+      "learning_rate": 1.790694976286752e-05,
+      "loss": 0.1535,
+      "step": 20450
+    },
+    {
+      "epoch": 2.5762509053122145,
+      "grad_norm": 0.25540080666542053,
+      "learning_rate": 1.7854901348756807e-05,
+      "loss": 0.1637,
+      "step": 20455
+    },
+    {
+      "epoch": 2.576880687722392,
+      "grad_norm": 0.1803160160779953,
+      "learning_rate": 1.780292389958257e-05,
+      "loss": 0.1526,
+      "step": 20460
+    },
+    {
+      "epoch": 2.577510470132569,
+      "grad_norm": 0.260122686624527,
+      "learning_rate": 1.775101744325792e-05,
+      "loss": 0.1704,
+      "step": 20465
+    },
+    {
+      "epoch": 2.5781402525427466,
+      "grad_norm": 0.19697842001914978,
+      "learning_rate": 1.7699182007657736e-05,
+      "loss": 0.1568,
+      "step": 20470
+    },
+    {
+      "epoch": 2.5787700349529237,
+      "grad_norm": 0.2179180532693863,
+      "learning_rate": 1.7647417620618936e-05,
+      "loss": 0.1612,
+      "step": 20475
+    },
+    {
+      "epoch": 2.579399817363101,
+      "grad_norm": 0.2509031593799591,
+      "learning_rate": 1.7595724309940117e-05,
+      "loss": 0.1531,
+      "step": 20480
+    },
+    {
+      "epoch": 2.5800295997732783,
+      "grad_norm": 0.19090527296066284,
+      "learning_rate": 1.754410210338179e-05,
+      "loss": 0.1477,
+      "step": 20485
+    },
+    {
+      "epoch": 2.5806593821834554,
+      "grad_norm": 0.20662526786327362,
+      "learning_rate": 1.749255102866623e-05,
+      "loss": 0.1584,
+      "step": 20490
+    },
+    {
+      "epoch": 2.581289164593633,
+      "grad_norm": 0.2258034199476242,
+      "learning_rate": 1.7441071113477572e-05,
+      "loss": 0.1597,
+      "step": 20495
+    },
+    {
+      "epoch": 2.5819189470038104,
+      "grad_norm": 0.22661426663398743,
+      "learning_rate": 1.738966238546169e-05,
+      "loss": 0.1582,
+      "step": 20500
+    },
+    {
+      "epoch": 2.5825487294139875,
+      "grad_norm": 0.24459710717201233,
+      "learning_rate": 1.7338324872226227e-05,
+      "loss": 0.1523,
+      "step": 20505
+    },
+    {
+      "epoch": 2.5831785118241646,
+      "grad_norm": 0.18816480040550232,
+      "learning_rate": 1.728705860134062e-05,
+      "loss": 0.1554,
+      "step": 20510
+    },
+    {
+      "epoch": 2.583808294234342,
+      "grad_norm": 0.18354368209838867,
+      "learning_rate": 1.7235863600336042e-05,
+      "loss": 0.1413,
+      "step": 20515
+    },
+    {
+      "epoch": 2.584438076644519,
+      "grad_norm": 0.1984662562608719,
+      "learning_rate": 1.71847398967054e-05,
+      "loss": 0.1566,
+      "step": 20520
+    },
+    {
+      "epoch": 2.5850678590546967,
+      "grad_norm": 0.2628153860569,
+      "learning_rate": 1.713368751790322e-05,
+      "loss": 0.1592,
+      "step": 20525
+    },
+    {
+      "epoch": 2.585697641464874,
+      "grad_norm": 0.18952016532421112,
+      "learning_rate": 1.7082706491345806e-05,
+      "loss": 0.1531,
+      "step": 20530
+    },
+    {
+      "epoch": 2.5863274238750513,
+      "grad_norm": 0.16905049979686737,
+      "learning_rate": 1.7031796844411198e-05,
+      "loss": 0.1556,
+      "step": 20535
+    },
+    {
+      "epoch": 2.5869572062852284,
+      "grad_norm": 0.20969530940055847,
+      "learning_rate": 1.6980958604438988e-05,
+      "loss": 0.1585,
+      "step": 20540
+    },
+    {
+      "epoch": 2.5875869886954055,
+      "grad_norm": 0.2143043577671051,
+      "learning_rate": 1.693019179873048e-05,
+      "loss": 0.1636,
+      "step": 20545
+    },
+    {
+      "epoch": 2.588216771105583,
+      "grad_norm": 0.24208824336528778,
+      "learning_rate": 1.6879496454548585e-05,
+      "loss": 0.1572,
+      "step": 20550
+    },
+    {
+      "epoch": 2.5888465535157605,
+      "grad_norm": 0.22409161925315857,
+      "learning_rate": 1.6828872599117958e-05,
+      "loss": 0.162,
+      "step": 20555
+    },
+    {
+      "epoch": 2.5894763359259376,
+      "grad_norm": 0.20685546100139618,
+      "learning_rate": 1.6778320259624654e-05,
+      "loss": 0.1587,
+      "step": 20560
+    },
+    {
+      "epoch": 2.5901061183361147,
+      "grad_norm": 0.19393740594387054,
+      "learning_rate": 1.672783946321649e-05,
+      "loss": 0.1491,
+      "step": 20565
+    },
+    {
+      "epoch": 2.5907359007462922,
+      "grad_norm": 0.1944616734981537,
+      "learning_rate": 1.667743023700275e-05,
+      "loss": 0.1565,
+      "step": 20570
+    },
+    {
+      "epoch": 2.5913656831564693,
+      "grad_norm": 0.21134309470653534,
+      "learning_rate": 1.662709260805442e-05,
+      "loss": 0.1575,
+      "step": 20575
+    },
+    {
+      "epoch": 2.591995465566647,
+      "grad_norm": 0.20300306379795074,
+      "learning_rate": 1.657682660340392e-05,
+      "loss": 0.1509,
+      "step": 20580
+    },
+    {
+      "epoch": 2.592625247976824,
+      "grad_norm": 0.209407240152359,
+      "learning_rate": 1.6526632250045237e-05,
+      "loss": 0.1568,
+      "step": 20585
+    },
+    {
+      "epoch": 2.5932550303870014,
+      "grad_norm": 0.18960040807724,
+      "learning_rate": 1.6476509574933888e-05,
+      "loss": 0.1561,
+      "step": 20590
+    },
+    {
+      "epoch": 2.5938848127971785,
+      "grad_norm": 0.2009792923927307,
+      "learning_rate": 1.6426458604986897e-05,
+      "loss": 0.1584,
+      "step": 20595
+    },
+    {
+      "epoch": 2.5945145952073556,
+      "grad_norm": 0.2359851896762848,
+      "learning_rate": 1.6376479367082796e-05,
+      "loss": 0.1573,
+      "step": 20600
+    },
+    {
+      "epoch": 2.595144377617533,
+      "grad_norm": 0.2108912616968155,
+      "learning_rate": 1.632657188806153e-05,
+      "loss": 0.1537,
+      "step": 20605
+    },
+    {
+      "epoch": 2.5957741600277107,
+      "grad_norm": 0.22792066633701324,
+      "learning_rate": 1.6276736194724575e-05,
+      "loss": 0.1611,
+      "step": 20610
+    },
+    {
+      "epoch": 2.5964039424378877,
+      "grad_norm": 0.1896820068359375,
+      "learning_rate": 1.622697231383488e-05,
+      "loss": 0.1623,
+      "step": 20615
+    },
+    {
+      "epoch": 2.597033724848065,
+      "grad_norm": 0.19234326481819153,
+      "learning_rate": 1.6177280272116728e-05,
+      "loss": 0.1448,
+      "step": 20620
+    },
+    {
+      "epoch": 2.5976635072582424,
+      "grad_norm": 0.17547307908535004,
+      "learning_rate": 1.6127660096255955e-05,
+      "loss": 0.1479,
+      "step": 20625
+    },
+    {
+      "epoch": 2.5982932896684194,
+      "grad_norm": 0.20076265931129456,
+      "learning_rate": 1.6078111812899618e-05,
+      "loss": 0.1504,
+      "step": 20630
+    },
+    {
+      "epoch": 2.598923072078597,
+      "grad_norm": 0.1888744831085205,
+      "learning_rate": 1.6028635448656364e-05,
+      "loss": 0.1587,
+      "step": 20635
+    },
+    {
+      "epoch": 2.599552854488774,
+      "grad_norm": 0.2309001237154007,
+      "learning_rate": 1.59792310300961e-05,
+      "loss": 0.1662,
+      "step": 20640
+    },
+    {
+      "epoch": 2.6001826368989516,
+      "grad_norm": 0.29581940174102783,
+      "learning_rate": 1.592989858375013e-05,
+      "loss": 0.1708,
+      "step": 20645
+    },
+    {
+      "epoch": 2.6008124193091287,
+      "grad_norm": 0.19039921462535858,
+      "learning_rate": 1.588063813611112e-05,
+      "loss": 0.1548,
+      "step": 20650
+    },
+    {
+      "epoch": 2.6014422017193057,
+      "grad_norm": 0.18266427516937256,
+      "learning_rate": 1.5831449713632993e-05,
+      "loss": 0.1523,
+      "step": 20655
+    },
+    {
+      "epoch": 2.6020719841294833,
+      "grad_norm": 0.1932811439037323,
+      "learning_rate": 1.5782333342731174e-05,
+      "loss": 0.1516,
+      "step": 20660
+    },
+    {
+      "epoch": 2.602701766539661,
+      "grad_norm": 0.17890222370624542,
+      "learning_rate": 1.5733289049782177e-05,
+      "loss": 0.1511,
+      "step": 20665
+    },
+    {
+      "epoch": 2.603331548949838,
+      "grad_norm": 0.20573283731937408,
+      "learning_rate": 1.5684316861123935e-05,
+      "loss": 0.1525,
+      "step": 20670
+    },
+    {
+      "epoch": 2.603961331360015,
+      "grad_norm": 0.21194593608379364,
+      "learning_rate": 1.5635416803055596e-05,
+      "loss": 0.1599,
+      "step": 20675
+    },
+    {
+      "epoch": 2.6045911137701925,
+      "grad_norm": 0.17930278182029724,
+      "learning_rate": 1.558658890183768e-05,
+      "loss": 0.1571,
+      "step": 20680
+    },
+    {
+      "epoch": 2.6052208961803696,
+      "grad_norm": 0.1965799480676651,
+      "learning_rate": 1.5537833183691857e-05,
+      "loss": 0.1552,
+      "step": 20685
+    },
+    {
+      "epoch": 2.605850678590547,
+      "grad_norm": 0.20715682208538055,
+      "learning_rate": 1.5489149674801054e-05,
+      "loss": 0.1588,
+      "step": 20690
+    },
+    {
+      "epoch": 2.606480461000724,
+      "grad_norm": 0.1894584596157074,
+      "learning_rate": 1.544053840130943e-05,
+      "loss": 0.1547,
+      "step": 20695
+    },
+    {
+      "epoch": 2.6071102434109017,
+      "grad_norm": 0.20791690051555634,
+      "learning_rate": 1.539199938932234e-05,
+      "loss": 0.1479,
+      "step": 20700
+    },
+    {
+      "epoch": 2.6077400258210788,
+      "grad_norm": 0.20393605530261993,
+      "learning_rate": 1.534353266490636e-05,
+      "loss": 0.1491,
+      "step": 20705
+    },
+    {
+      "epoch": 2.608369808231256,
+      "grad_norm": 0.20201466977596283,
+      "learning_rate": 1.5295138254089206e-05,
+      "loss": 0.1664,
+      "step": 20710
+    },
+    {
+      "epoch": 2.6089995906414334,
+      "grad_norm": 0.220575213432312,
+      "learning_rate": 1.5246816182859773e-05,
+      "loss": 0.16,
+      "step": 20715
+    },
+    {
+      "epoch": 2.609629373051611,
+      "grad_norm": 0.1888882964849472,
+      "learning_rate": 1.5198566477168166e-05,
+      "loss": 0.1592,
+      "step": 20720
+    },
+    {
+      "epoch": 2.610259155461788,
+      "grad_norm": 0.2035285383462906,
+      "learning_rate": 1.5150389162925564e-05,
+      "loss": 0.149,
+      "step": 20725
+    },
+    {
+      "epoch": 2.610888937871965,
+      "grad_norm": 0.21430674195289612,
+      "learning_rate": 1.5102284266004282e-05,
+      "loss": 0.1568,
+      "step": 20730
+    },
+    {
+      "epoch": 2.6115187202821426,
+      "grad_norm": 0.2220098227262497,
+      "learning_rate": 1.5054251812237695e-05,
+      "loss": 0.1601,
+      "step": 20735
+    },
+    {
+      "epoch": 2.6121485026923197,
+      "grad_norm": 0.18914029002189636,
+      "learning_rate": 1.5006291827420397e-05,
+      "loss": 0.1524,
+      "step": 20740
+    },
+    {
+      "epoch": 2.612778285102497,
+      "grad_norm": 0.19741562008857727,
+      "learning_rate": 1.4958404337307972e-05,
+      "loss": 0.1418,
+      "step": 20745
+    },
+    {
+      "epoch": 2.6134080675126743,
+      "grad_norm": 0.22962430119514465,
+      "learning_rate": 1.49105893676171e-05,
+      "loss": 0.1523,
+      "step": 20750
+    },
+    {
+      "epoch": 2.614037849922852,
+      "grad_norm": 0.17770111560821533,
+      "learning_rate": 1.4862846944025469e-05,
+      "loss": 0.1585,
+      "step": 20755
+    },
+    {
+      "epoch": 2.614667632333029,
+      "grad_norm": 0.22975338995456696,
+      "learning_rate": 1.481517709217191e-05,
+      "loss": 0.1608,
+      "step": 20760
+    },
+    {
+      "epoch": 2.615297414743206,
+      "grad_norm": 0.21070002019405365,
+      "learning_rate": 1.476757983765624e-05,
+      "loss": 0.1442,
+      "step": 20765
+    },
+    {
+      "epoch": 2.6159271971533835,
+      "grad_norm": 0.19414427876472473,
+      "learning_rate": 1.47200552060392e-05,
+      "loss": 0.1445,
+      "step": 20770
+    },
+    {
+      "epoch": 2.616556979563561,
+      "grad_norm": 0.18657416105270386,
+      "learning_rate": 1.4672603222842605e-05,
+      "loss": 0.1534,
+      "step": 20775
+    },
+    {
+      "epoch": 2.617186761973738,
+      "grad_norm": 0.2389591485261917,
+      "learning_rate": 1.4625223913549323e-05,
+      "loss": 0.158,
+      "step": 20780
+    },
+    {
+      "epoch": 2.617816544383915,
+      "grad_norm": 0.19741186499595642,
+      "learning_rate": 1.4577917303603081e-05,
+      "loss": 0.1585,
+      "step": 20785
+    },
+    {
+      "epoch": 2.6184463267940927,
+      "grad_norm": 0.18730677664279938,
+      "learning_rate": 1.4530683418408612e-05,
+      "loss": 0.1487,
+      "step": 20790
+    },
+    {
+      "epoch": 2.61907610920427,
+      "grad_norm": 0.2060120701789856,
+      "learning_rate": 1.4483522283331606e-05,
+      "loss": 0.1499,
+      "step": 20795
+    },
+    {
+      "epoch": 2.6197058916144473,
+      "grad_norm": 0.2186814844608307,
+      "learning_rate": 1.4436433923698638e-05,
+      "loss": 0.1562,
+      "step": 20800
+    },
+    {
+      "epoch": 2.6203356740246244,
+      "grad_norm": 0.21503032743930817,
+      "learning_rate": 1.4389418364797279e-05,
+      "loss": 0.1456,
+      "step": 20805
+    },
+    {
+      "epoch": 2.620965456434802,
+      "grad_norm": 0.17447194457054138,
+      "learning_rate": 1.4342475631875916e-05,
+      "loss": 0.1454,
+      "step": 20810
+    },
+    {
+      "epoch": 2.621595238844979,
+      "grad_norm": 0.18272021412849426,
+      "learning_rate": 1.4295605750143851e-05,
+      "loss": 0.149,
+      "step": 20815
+    },
+    {
+      "epoch": 2.622225021255156,
+      "grad_norm": 0.2014734447002411,
+      "learning_rate": 1.424880874477135e-05,
+      "loss": 0.1582,
+      "step": 20820
+    },
+    {
+      "epoch": 2.6228548036653336,
+      "grad_norm": 0.21231862902641296,
+      "learning_rate": 1.4202084640889443e-05,
+      "loss": 0.152,
+      "step": 20825
+    },
+    {
+      "epoch": 2.623484586075511,
+      "grad_norm": 0.19817417860031128,
+      "learning_rate": 1.415543346359006e-05,
+      "loss": 0.1492,
+      "step": 20830
+    },
+    {
+      "epoch": 2.6241143684856882,
+      "grad_norm": 0.20216423273086548,
+      "learning_rate": 1.410885523792586e-05,
+      "loss": 0.1452,
+      "step": 20835
+    },
+    {
+      "epoch": 2.6247441508958653,
+      "grad_norm": 0.20939548313617706,
+      "learning_rate": 1.4062349988910515e-05,
+      "loss": 0.1512,
+      "step": 20840
+    },
+    {
+      "epoch": 2.625373933306043,
+      "grad_norm": 0.19018815457820892,
+      "learning_rate": 1.4015917741518384e-05,
+      "loss": 0.1579,
+      "step": 20845
+    },
+    {
+      "epoch": 2.62600371571622,
+      "grad_norm": 0.20512887835502625,
+      "learning_rate": 1.396955852068462e-05,
+      "loss": 0.1624,
+      "step": 20850
+    },
+    {
+      "epoch": 2.6266334981263975,
+      "grad_norm": 0.24390068650245667,
+      "learning_rate": 1.3923272351305193e-05,
+      "loss": 0.1663,
+      "step": 20855
+    },
+    {
+      "epoch": 2.6272632805365745,
+      "grad_norm": 0.21338611841201782,
+      "learning_rate": 1.38770592582368e-05,
+      "loss": 0.1695,
+      "step": 20860
+    },
+    {
+      "epoch": 2.627893062946752,
+      "grad_norm": 0.21631261706352234,
+      "learning_rate": 1.3830919266297025e-05,
+      "loss": 0.1598,
+      "step": 20865
+    },
+    {
+      "epoch": 2.628522845356929,
+      "grad_norm": 0.21297498047351837,
+      "learning_rate": 1.3784852400264013e-05,
+      "loss": 0.1554,
+      "step": 20870
+    },
+    {
+      "epoch": 2.6291526277671062,
+      "grad_norm": 0.22182469069957733,
+      "learning_rate": 1.3738858684876724e-05,
+      "loss": 0.1618,
+      "step": 20875
+    },
+    {
+      "epoch": 2.6297824101772838,
+      "grad_norm": 0.2602301239967346,
+      "learning_rate": 1.369293814483487e-05,
+      "loss": 0.1616,
+      "step": 20880
+    },
+    {
+      "epoch": 2.6304121925874613,
+      "grad_norm": 0.20623180270195007,
+      "learning_rate": 1.3647090804798822e-05,
+      "loss": 0.1574,
+      "step": 20885
+    },
+    {
+      "epoch": 2.6310419749976384,
+      "grad_norm": 0.22888506948947906,
+      "learning_rate": 1.3601316689389635e-05,
+      "loss": 0.1476,
+      "step": 20890
+    },
+    {
+      "epoch": 2.6316717574078154,
+      "grad_norm": 0.1814454197883606,
+      "learning_rate": 1.3555615823189065e-05,
+      "loss": 0.1505,
+      "step": 20895
+    },
+    {
+      "epoch": 2.632301539817993,
+      "grad_norm": 0.19201195240020752,
+      "learning_rate": 1.350998823073951e-05,
+      "loss": 0.1559,
+      "step": 20900
+    },
+    {
+      "epoch": 2.63293132222817,
+      "grad_norm": 0.21456103026866913,
+      "learning_rate": 1.3464433936544055e-05,
+      "loss": 0.1519,
+      "step": 20905
+    },
+    {
+      "epoch": 2.6335611046383476,
+      "grad_norm": 0.21142180263996124,
+      "learning_rate": 1.3418952965066365e-05,
+      "loss": 0.153,
+      "step": 20910
+    },
+    {
+      "epoch": 2.6341908870485247,
+      "grad_norm": 0.2231752574443817,
+      "learning_rate": 1.3373545340730785e-05,
+      "loss": 0.1641,
+      "step": 20915
+    },
+    {
+      "epoch": 2.634820669458702,
+      "grad_norm": 0.19116418063640594,
+      "learning_rate": 1.3328211087922192e-05,
+      "loss": 0.1503,
+      "step": 20920
+    },
+    {
+      "epoch": 2.6354504518688793,
+      "grad_norm": 0.18010330200195312,
+      "learning_rate": 1.3282950230986194e-05,
+      "loss": 0.1434,
+      "step": 20925
+    },
+    {
+      "epoch": 2.6360802342790564,
+      "grad_norm": 0.179785817861557,
+      "learning_rate": 1.3237762794228884e-05,
+      "loss": 0.1502,
+      "step": 20930
+    },
+    {
+      "epoch": 2.636710016689234,
+      "grad_norm": 0.1842677742242813,
+      "learning_rate": 1.319264880191695e-05,
+      "loss": 0.1421,
+      "step": 20935
+    },
+    {
+      "epoch": 2.6373397990994114,
+      "grad_norm": 0.22725196182727814,
+      "learning_rate": 1.314760827827756e-05,
+      "loss": 0.1464,
+      "step": 20940
+    },
+    {
+      "epoch": 2.6379695815095885,
+      "grad_norm": 0.21761812269687653,
+      "learning_rate": 1.3102641247498585e-05,
+      "loss": 0.1492,
+      "step": 20945
+    },
+    {
+      "epoch": 2.6385993639197656,
+      "grad_norm": 0.2054702490568161,
+      "learning_rate": 1.305774773372834e-05,
+      "loss": 0.1626,
+      "step": 20950
+    },
+    {
+      "epoch": 2.639229146329943,
+      "grad_norm": 0.20189544558525085,
+      "learning_rate": 1.3012927761075658e-05,
+      "loss": 0.1672,
+      "step": 20955
+    },
+    {
+      "epoch": 2.63985892874012,
+      "grad_norm": 0.2214374542236328,
+      "learning_rate": 1.2968181353609852e-05,
+      "loss": 0.159,
+      "step": 20960
+    },
+    {
+      "epoch": 2.6404887111502977,
+      "grad_norm": 0.20227175951004028,
+      "learning_rate": 1.2923508535360833e-05,
+      "loss": 0.1668,
+      "step": 20965
+    },
+    {
+      "epoch": 2.641118493560475,
+      "grad_norm": 0.2125934362411499,
+      "learning_rate": 1.2878909330318893e-05,
+      "loss": 0.1587,
+      "step": 20970
+    },
+    {
+      "epoch": 2.6417482759706523,
+      "grad_norm": 0.20071247220039368,
+      "learning_rate": 1.2834383762434807e-05,
+      "loss": 0.1515,
+      "step": 20975
+    },
+    {
+      "epoch": 2.6423780583808294,
+      "grad_norm": 0.18576478958129883,
+      "learning_rate": 1.2789931855619817e-05,
+      "loss": 0.152,
+      "step": 20980
+    },
+    {
+      "epoch": 2.6430078407910065,
+      "grad_norm": 0.2210751622915268,
+      "learning_rate": 1.2745553633745642e-05,
+      "loss": 0.1542,
+      "step": 20985
+    },
+    {
+      "epoch": 2.643637623201184,
+      "grad_norm": 0.20466601848602295,
+      "learning_rate": 1.2701249120644402e-05,
+      "loss": 0.1599,
+      "step": 20990
+    },
+    {
+      "epoch": 2.6442674056113615,
+      "grad_norm": 0.18101972341537476,
+      "learning_rate": 1.2657018340108616e-05,
+      "loss": 0.1434,
+      "step": 20995
+    },
+    {
+      "epoch": 2.6448971880215386,
+      "grad_norm": 0.19254201650619507,
+      "learning_rate": 1.2612861315891215e-05,
+      "loss": 0.1492,
+      "step": 21000
+    },
+    {
+      "epoch": 2.6448971880215386,
+      "eval_loss": 0.35662394762039185,
+      "eval_runtime": 6.1657,
+      "eval_samples_per_second": 162.189,
+      "eval_steps_per_second": 10.218,
+      "step": 21000
+    },
+    {
+      "epoch": 2.6455269704317157,
+      "grad_norm": 0.21622531116008759,
+      "learning_rate": 1.2568778071705564e-05,
+      "loss": 0.1508,
+      "step": 21005
+    },
+    {
+      "epoch": 2.6461567528418932,
+      "grad_norm": 0.20254169404506683,
+      "learning_rate": 1.2524768631225329e-05,
+      "loss": 0.1541,
+      "step": 21010
+    },
+    {
+      "epoch": 2.6467865352520703,
+      "grad_norm": 0.18706361949443817,
+      "learning_rate": 1.2480833018084619e-05,
+      "loss": 0.1554,
+      "step": 21015
+    },
+    {
+      "epoch": 2.647416317662248,
+      "grad_norm": 0.19682380557060242,
+      "learning_rate": 1.2436971255877825e-05,
+      "loss": 0.1527,
+      "step": 21020
+    },
+    {
+      "epoch": 2.648046100072425,
+      "grad_norm": 0.193098783493042,
+      "learning_rate": 1.2393183368159759e-05,
+      "loss": 0.1505,
+      "step": 21025
+    },
+    {
+      "epoch": 2.6486758824826024,
+      "grad_norm": 0.1954520344734192,
+      "learning_rate": 1.2349469378445493e-05,
+      "loss": 0.1463,
+      "step": 21030
+    },
+    {
+      "epoch": 2.6493056648927795,
+      "grad_norm": 0.2523531913757324,
+      "learning_rate": 1.2305829310210446e-05,
+      "loss": 0.1655,
+      "step": 21035
+    },
+    {
+      "epoch": 2.6499354473029566,
+      "grad_norm": 0.20331156253814697,
+      "learning_rate": 1.2262263186890325e-05,
+      "loss": 0.1514,
+      "step": 21040
+    },
+    {
+      "epoch": 2.650565229713134,
+      "grad_norm": 0.2400408834218979,
+      "learning_rate": 1.221877103188113e-05,
+      "loss": 0.1673,
+      "step": 21045
+    },
+    {
+      "epoch": 2.6511950121233117,
+      "grad_norm": 0.18541163206100464,
+      "learning_rate": 1.2175352868539162e-05,
+      "loss": 0.1582,
+      "step": 21050
+    },
+    {
+      "epoch": 2.6518247945334887,
+      "grad_norm": 0.24442121386528015,
+      "learning_rate": 1.2132008720180953e-05,
+      "loss": 0.1525,
+      "step": 21055
+    },
+    {
+      "epoch": 2.652454576943666,
+      "grad_norm": 0.23133227229118347,
+      "learning_rate": 1.2088738610083282e-05,
+      "loss": 0.1547,
+      "step": 21060
+    },
+    {
+      "epoch": 2.6530843593538433,
+      "grad_norm": 0.21159769594669342,
+      "learning_rate": 1.2045542561483196e-05,
+      "loss": 0.1451,
+      "step": 21065
+    },
+    {
+      "epoch": 2.6537141417640204,
+      "grad_norm": 0.19382759928703308,
+      "learning_rate": 1.2002420597577972e-05,
+      "loss": 0.1532,
+      "step": 21070
+    },
+    {
+      "epoch": 2.654343924174198,
+      "grad_norm": 0.18077696859836578,
+      "learning_rate": 1.1959372741525135e-05,
+      "loss": 0.1493,
+      "step": 21075
+    },
+    {
+      "epoch": 2.654973706584375,
+      "grad_norm": 0.24647746980190277,
+      "learning_rate": 1.1916399016442264e-05,
+      "loss": 0.1533,
+      "step": 21080
+    },
+    {
+      "epoch": 2.6556034889945526,
+      "grad_norm": 0.19929052889347076,
+      "learning_rate": 1.1873499445407291e-05,
+      "loss": 0.1418,
+      "step": 21085
+    },
+    {
+      "epoch": 2.6562332714047296,
+      "grad_norm": 0.2208561897277832,
+      "learning_rate": 1.1830674051458277e-05,
+      "loss": 0.1628,
+      "step": 21090
+    },
+    {
+      "epoch": 2.6568630538149067,
+      "grad_norm": 0.18301671743392944,
+      "learning_rate": 1.1787922857593406e-05,
+      "loss": 0.1495,
+      "step": 21095
+    },
+    {
+      "epoch": 2.6574928362250843,
+      "grad_norm": 0.1876724660396576,
+      "learning_rate": 1.1745245886771065e-05,
+      "loss": 0.1526,
+      "step": 21100
+    },
+    {
+      "epoch": 2.658122618635262,
+      "grad_norm": 0.18740510940551758,
+      "learning_rate": 1.1702643161909736e-05,
+      "loss": 0.1469,
+      "step": 21105
+    },
+    {
+      "epoch": 2.658752401045439,
+      "grad_norm": 0.2159937024116516,
+      "learning_rate": 1.1660114705888119e-05,
+      "loss": 0.1534,
+      "step": 21110
+    },
+    {
+      "epoch": 2.659382183455616,
+      "grad_norm": 0.19949863851070404,
+      "learning_rate": 1.1617660541544893e-05,
+      "loss": 0.1446,
+      "step": 21115
+    },
+    {
+      "epoch": 2.6600119658657935,
+      "grad_norm": 0.18760685622692108,
+      "learning_rate": 1.1575280691678956e-05,
+      "loss": 0.1495,
+      "step": 21120
+    },
+    {
+      "epoch": 2.6606417482759706,
+      "grad_norm": 0.1935281902551651,
+      "learning_rate": 1.153297517904922e-05,
+      "loss": 0.1563,
+      "step": 21125
+    },
+    {
+      "epoch": 2.661271530686148,
+      "grad_norm": 0.20625917613506317,
+      "learning_rate": 1.1490744026374743e-05,
+      "loss": 0.1527,
+      "step": 21130
+    },
+    {
+      "epoch": 2.661901313096325,
+      "grad_norm": 0.22293558716773987,
+      "learning_rate": 1.1448587256334618e-05,
+      "loss": 0.1573,
+      "step": 21135
+    },
+    {
+      "epoch": 2.6625310955065027,
+      "grad_norm": 0.22753605246543884,
+      "learning_rate": 1.1406504891567986e-05,
+      "loss": 0.1563,
+      "step": 21140
+    },
+    {
+      "epoch": 2.6631608779166798,
+      "grad_norm": 0.1857980489730835,
+      "learning_rate": 1.1364496954674035e-05,
+      "loss": 0.1542,
+      "step": 21145
+    },
+    {
+      "epoch": 2.663790660326857,
+      "grad_norm": 0.20376616716384888,
+      "learning_rate": 1.1322563468212003e-05,
+      "loss": 0.1533,
+      "step": 21150
+    },
+    {
+      "epoch": 2.6644204427370344,
+      "grad_norm": 0.18928895890712738,
+      "learning_rate": 1.1280704454701111e-05,
+      "loss": 0.151,
+      "step": 21155
+    },
+    {
+      "epoch": 2.665050225147212,
+      "grad_norm": 0.2215338945388794,
+      "learning_rate": 1.1238919936620593e-05,
+      "loss": 0.1484,
+      "step": 21160
+    },
+    {
+      "epoch": 2.665680007557389,
+      "grad_norm": 0.26164811849594116,
+      "learning_rate": 1.1197209936409702e-05,
+      "loss": 0.1604,
+      "step": 21165
+    },
+    {
+      "epoch": 2.666309789967566,
+      "grad_norm": 0.1930347979068756,
+      "learning_rate": 1.1155574476467682e-05,
+      "loss": 0.1578,
+      "step": 21170
+    },
+    {
+      "epoch": 2.6669395723777436,
+      "grad_norm": 0.18400873243808746,
+      "learning_rate": 1.1114013579153719e-05,
+      "loss": 0.1559,
+      "step": 21175
+    },
+    {
+      "epoch": 2.6675693547879207,
+      "grad_norm": 0.19113576412200928,
+      "learning_rate": 1.1072527266786974e-05,
+      "loss": 0.1583,
+      "step": 21180
+    },
+    {
+      "epoch": 2.668199137198098,
+      "grad_norm": 0.1980462521314621,
+      "learning_rate": 1.1031115561646476e-05,
+      "loss": 0.1516,
+      "step": 21185
+    },
+    {
+      "epoch": 2.6688289196082753,
+      "grad_norm": 0.24164487421512604,
+      "learning_rate": 1.0989778485971334e-05,
+      "loss": 0.1578,
+      "step": 21190
+    },
+    {
+      "epoch": 2.669458702018453,
+      "grad_norm": 0.18318641185760498,
+      "learning_rate": 1.0948516061960478e-05,
+      "loss": 0.1517,
+      "step": 21195
+    },
+    {
+      "epoch": 2.67008848442863,
+      "grad_norm": 0.20504848659038544,
+      "learning_rate": 1.0907328311772778e-05,
+      "loss": 0.1619,
+      "step": 21200
+    },
+    {
+      "epoch": 2.670718266838807,
+      "grad_norm": 0.214483380317688,
+      "learning_rate": 1.0866215257526978e-05,
+      "loss": 0.1445,
+      "step": 21205
+    },
+    {
+      "epoch": 2.6713480492489845,
+      "grad_norm": 0.24230434000492096,
+      "learning_rate": 1.0825176921301698e-05,
+      "loss": 0.1521,
+      "step": 21210
+    },
+    {
+      "epoch": 2.671977831659162,
+      "grad_norm": 0.22668616473674774,
+      "learning_rate": 1.0784213325135577e-05,
+      "loss": 0.1539,
+      "step": 21215
+    },
+    {
+      "epoch": 2.672607614069339,
+      "grad_norm": 0.18815076351165771,
+      "learning_rate": 1.0743324491026883e-05,
+      "loss": 0.1496,
+      "step": 21220
+    },
+    {
+      "epoch": 2.673237396479516,
+      "grad_norm": 0.24532602727413177,
+      "learning_rate": 1.070251044093387e-05,
+      "loss": 0.1572,
+      "step": 21225
+    },
+    {
+      "epoch": 2.6738671788896937,
+      "grad_norm": 0.2050776183605194,
+      "learning_rate": 1.066177119677467e-05,
+      "loss": 0.1585,
+      "step": 21230
+    },
+    {
+      "epoch": 2.674496961299871,
+      "grad_norm": 0.1992231011390686,
+      "learning_rate": 1.062110678042717e-05,
+      "loss": 0.1493,
+      "step": 21235
+    },
+    {
+      "epoch": 2.6751267437100483,
+      "grad_norm": 0.2188093513250351,
+      "learning_rate": 1.0580517213729062e-05,
+      "loss": 0.1526,
+      "step": 21240
+    },
+    {
+      "epoch": 2.6757565261202254,
+      "grad_norm": 0.17839093506336212,
+      "learning_rate": 1.0540002518477898e-05,
+      "loss": 0.146,
+      "step": 21245
+    },
+    {
+      "epoch": 2.676386308530403,
+      "grad_norm": 0.20759402215480804,
+      "learning_rate": 1.0499562716430987e-05,
+      "loss": 0.1527,
+      "step": 21250
+    },
+    {
+      "epoch": 2.67701609094058,
+      "grad_norm": 0.20209045708179474,
+      "learning_rate": 1.0459197829305427e-05,
+      "loss": 0.1507,
+      "step": 21255
+    },
+    {
+      "epoch": 2.677645873350757,
+      "grad_norm": 0.24553018808364868,
+      "learning_rate": 1.0418907878778077e-05,
+      "loss": 0.1568,
+      "step": 21260
+    },
+    {
+      "epoch": 2.6782756557609346,
+      "grad_norm": 0.24322043359279633,
+      "learning_rate": 1.0378692886485563e-05,
+      "loss": 0.1527,
+      "step": 21265
+    },
+    {
+      "epoch": 2.678905438171112,
+      "grad_norm": 0.20755696296691895,
+      "learning_rate": 1.0338552874024242e-05,
+      "loss": 0.1497,
+      "step": 21270
+    },
+    {
+      "epoch": 2.6795352205812892,
+      "grad_norm": 0.19075340032577515,
+      "learning_rate": 1.0298487862950256e-05,
+      "loss": 0.1514,
+      "step": 21275
+    },
+    {
+      "epoch": 2.6801650029914663,
+      "grad_norm": 0.20733466744422913,
+      "learning_rate": 1.0258497874779426e-05,
+      "loss": 0.1531,
+      "step": 21280
+    },
+    {
+      "epoch": 2.680794785401644,
+      "grad_norm": 0.20700550079345703,
+      "learning_rate": 1.0218582930987224e-05,
+      "loss": 0.1547,
+      "step": 21285
+    },
+    {
+      "epoch": 2.681424567811821,
+      "grad_norm": 0.1864207237958908,
+      "learning_rate": 1.0178743053008969e-05,
+      "loss": 0.1507,
+      "step": 21290
+    },
+    {
+      "epoch": 2.6820543502219985,
+      "grad_norm": 0.21107596158981323,
+      "learning_rate": 1.0138978262239532e-05,
+      "loss": 0.1511,
+      "step": 21295
+    },
+    {
+      "epoch": 2.6826841326321755,
+      "grad_norm": 0.25058072805404663,
+      "learning_rate": 1.0099288580033548e-05,
+      "loss": 0.1573,
+      "step": 21300
+    },
+    {
+      "epoch": 2.683313915042353,
+      "grad_norm": 0.18913084268569946,
+      "learning_rate": 1.005967402770525e-05,
+      "loss": 0.1415,
+      "step": 21305
+    },
+    {
+      "epoch": 2.68394369745253,
+      "grad_norm": 0.18435829877853394,
+      "learning_rate": 1.002013462652857e-05,
+      "loss": 0.1432,
+      "step": 21310
+    },
+    {
+      "epoch": 2.6845734798627072,
+      "grad_norm": 0.19929499924182892,
+      "learning_rate": 9.980670397737106e-06,
+      "loss": 0.1562,
+      "step": 21315
+    },
+    {
+      "epoch": 2.6852032622728847,
+      "grad_norm": 0.2412646859884262,
+      "learning_rate": 9.941281362524007e-06,
+      "loss": 0.1544,
+      "step": 21320
+    },
+    {
+      "epoch": 2.6858330446830623,
+      "grad_norm": 0.23384952545166016,
+      "learning_rate": 9.9019675420421e-06,
+      "loss": 0.1598,
+      "step": 21325
+    },
+    {
+      "epoch": 2.6864628270932394,
+      "grad_norm": 0.1778135895729065,
+      "learning_rate": 9.862728957403766e-06,
+      "loss": 0.1515,
+      "step": 21330
+    },
+    {
+      "epoch": 2.6870926095034164,
+      "grad_norm": 0.20782922208309174,
+      "learning_rate": 9.823565629681079e-06,
+      "loss": 0.1504,
+      "step": 21335
+    },
+    {
+      "epoch": 2.687722391913594,
+      "grad_norm": 0.18523196876049042,
+      "learning_rate": 9.78447757990562e-06,
+      "loss": 0.1425,
+      "step": 21340
+    },
+    {
+      "epoch": 2.688352174323771,
+      "grad_norm": 0.18965183198451996,
+      "learning_rate": 9.745464829068561e-06,
+      "loss": 0.1541,
+      "step": 21345
+    },
+    {
+      "epoch": 2.6889819567339486,
+      "grad_norm": 0.18834419548511505,
+      "learning_rate": 9.706527398120645e-06,
+      "loss": 0.1536,
+      "step": 21350
+    },
+    {
+      "epoch": 2.6896117391441257,
+      "grad_norm": 0.18705077469348907,
+      "learning_rate": 9.66766530797216e-06,
+      "loss": 0.1451,
+      "step": 21355
+    },
+    {
+      "epoch": 2.690241521554303,
+      "grad_norm": 0.1886008232831955,
+      "learning_rate": 9.628878579492932e-06,
+      "loss": 0.1484,
+      "step": 21360
+    },
+    {
+      "epoch": 2.6908713039644803,
+      "grad_norm": 0.19375784695148468,
+      "learning_rate": 9.590167233512314e-06,
+      "loss": 0.1554,
+      "step": 21365
+    },
+    {
+      "epoch": 2.6915010863746573,
+      "grad_norm": 0.18573135137557983,
+      "learning_rate": 9.551531290819192e-06,
+      "loss": 0.1608,
+      "step": 21370
+    },
+    {
+      "epoch": 2.692130868784835,
+      "grad_norm": 0.18215128779411316,
+      "learning_rate": 9.512970772161955e-06,
+      "loss": 0.1564,
+      "step": 21375
+    },
+    {
+      "epoch": 2.6927606511950124,
+      "grad_norm": 0.1941639482975006,
+      "learning_rate": 9.474485698248469e-06,
+      "loss": 0.1551,
+      "step": 21380
+    },
+    {
+      "epoch": 2.6933904336051895,
+      "grad_norm": 0.19289493560791016,
+      "learning_rate": 9.436076089746153e-06,
+      "loss": 0.1537,
+      "step": 21385
+    },
+    {
+      "epoch": 2.6940202160153666,
+      "grad_norm": 0.19275395572185516,
+      "learning_rate": 9.397741967281724e-06,
+      "loss": 0.1441,
+      "step": 21390
+    },
+    {
+      "epoch": 2.694649998425544,
+      "grad_norm": 0.18316887319087982,
+      "learning_rate": 9.359483351441599e-06,
+      "loss": 0.1496,
+      "step": 21395
+    },
+    {
+      "epoch": 2.695279780835721,
+      "grad_norm": 0.16594599187374115,
+      "learning_rate": 9.321300262771475e-06,
+      "loss": 0.1408,
+      "step": 21400
+    },
+    {
+      "epoch": 2.6959095632458987,
+      "grad_norm": 0.2479625940322876,
+      "learning_rate": 9.28319272177655e-06,
+      "loss": 0.1565,
+      "step": 21405
+    },
+    {
+      "epoch": 2.696539345656076,
+      "grad_norm": 0.18492808938026428,
+      "learning_rate": 9.245160748921454e-06,
+      "loss": 0.143,
+      "step": 21410
+    },
+    {
+      "epoch": 2.6971691280662533,
+      "grad_norm": 0.22853007912635803,
+      "learning_rate": 9.207204364630182e-06,
+      "loss": 0.1668,
+      "step": 21415
+    },
+    {
+      "epoch": 2.6977989104764304,
+      "grad_norm": 0.1997872143983841,
+      "learning_rate": 9.169323589286264e-06,
+      "loss": 0.1563,
+      "step": 21420
+    },
+    {
+      "epoch": 2.6984286928866075,
+      "grad_norm": 0.23863272368907928,
+      "learning_rate": 9.131518443232476e-06,
+      "loss": 0.1554,
+      "step": 21425
+    },
+    {
+      "epoch": 2.699058475296785,
+      "grad_norm": 0.17353664338588715,
+      "learning_rate": 9.09378894677103e-06,
+      "loss": 0.147,
+      "step": 21430
+    },
+    {
+      "epoch": 2.699688257706962,
+      "grad_norm": 0.2168291211128235,
+      "learning_rate": 9.056135120163582e-06,
+      "loss": 0.1553,
+      "step": 21435
+    },
+    {
+      "epoch": 2.7003180401171396,
+      "grad_norm": 0.23211082816123962,
+      "learning_rate": 9.018556983631076e-06,
+      "loss": 0.1493,
+      "step": 21440
+    },
+    {
+      "epoch": 2.7009478225273167,
+      "grad_norm": 0.22088773548603058,
+      "learning_rate": 8.981054557353834e-06,
+      "loss": 0.1567,
+      "step": 21445
+    },
+    {
+      "epoch": 2.701577604937494,
+      "grad_norm": 0.20668818056583405,
+      "learning_rate": 8.943627861471497e-06,
+      "loss": 0.1559,
+      "step": 21450
+    },
+    {
+      "epoch": 2.7022073873476713,
+      "grad_norm": 0.22993560135364532,
+      "learning_rate": 8.906276916083072e-06,
+      "loss": 0.1628,
+      "step": 21455
+    },
+    {
+      "epoch": 2.702837169757849,
+      "grad_norm": 0.214871346950531,
+      "learning_rate": 8.869001741246862e-06,
+      "loss": 0.1567,
+      "step": 21460
+    },
+    {
+      "epoch": 2.703466952168026,
+      "grad_norm": 0.20056143403053284,
+      "learning_rate": 8.831802356980505e-06,
+      "loss": 0.1494,
+      "step": 21465
+    },
+    {
+      "epoch": 2.7040967345782034,
+      "grad_norm": 0.18365876376628876,
+      "learning_rate": 8.79467878326089e-06,
+      "loss": 0.1547,
+      "step": 21470
+    },
+    {
+      "epoch": 2.7047265169883805,
+      "grad_norm": 0.1938326060771942,
+      "learning_rate": 8.757631040024215e-06,
+      "loss": 0.1591,
+      "step": 21475
+    },
+    {
+      "epoch": 2.7053562993985576,
+      "grad_norm": 0.207264244556427,
+      "learning_rate": 8.72065914716602e-06,
+      "loss": 0.1588,
+      "step": 21480
+    },
+    {
+      "epoch": 2.705986081808735,
+      "grad_norm": 0.23815831542015076,
+      "learning_rate": 8.683763124541021e-06,
+      "loss": 0.1551,
+      "step": 21485
+    },
+    {
+      "epoch": 2.706615864218912,
+      "grad_norm": 0.20644132792949677,
+      "learning_rate": 8.646942991963236e-06,
+      "loss": 0.1496,
+      "step": 21490
+    },
+    {
+      "epoch": 2.7072456466290897,
+      "grad_norm": 0.19380377233028412,
+      "learning_rate": 8.610198769205895e-06,
+      "loss": 0.1499,
+      "step": 21495
+    },
+    {
+      "epoch": 2.707875429039267,
+      "grad_norm": 0.1877509504556656,
+      "learning_rate": 8.5735304760015e-06,
+      "loss": 0.1582,
+      "step": 21500
+    },
+    {
+      "epoch": 2.7085052114494443,
+      "grad_norm": 0.20092125236988068,
+      "learning_rate": 8.536938132041781e-06,
+      "loss": 0.1541,
+      "step": 21505
+    },
+    {
+      "epoch": 2.7091349938596214,
+      "grad_norm": 0.20917046070098877,
+      "learning_rate": 8.500421756977637e-06,
+      "loss": 0.1555,
+      "step": 21510
+    },
+    {
+      "epoch": 2.709764776269799,
+      "grad_norm": 0.18814347684383392,
+      "learning_rate": 8.463981370419165e-06,
+      "loss": 0.1511,
+      "step": 21515
+    },
+    {
+      "epoch": 2.710394558679976,
+      "grad_norm": 0.2021394819021225,
+      "learning_rate": 8.427616991935759e-06,
+      "loss": 0.1539,
+      "step": 21520
+    },
+    {
+      "epoch": 2.7110243410901536,
+      "grad_norm": 0.19899116456508636,
+      "learning_rate": 8.3913286410559e-06,
+      "loss": 0.1553,
+      "step": 21525
+    },
+    {
+      "epoch": 2.7116541235003306,
+      "grad_norm": 0.2093294858932495,
+      "learning_rate": 8.355116337267231e-06,
+      "loss": 0.1581,
+      "step": 21530
+    },
+    {
+      "epoch": 2.7122839059105077,
+      "grad_norm": 0.215724378824234,
+      "learning_rate": 8.318980100016564e-06,
+      "loss": 0.1516,
+      "step": 21535
+    },
+    {
+      "epoch": 2.7129136883206852,
+      "grad_norm": 0.21019119024276733,
+      "learning_rate": 8.28291994870996e-06,
+      "loss": 0.1521,
+      "step": 21540
+    },
+    {
+      "epoch": 2.7135434707308623,
+      "grad_norm": 0.20992571115493774,
+      "learning_rate": 8.246935902712493e-06,
+      "loss": 0.1401,
+      "step": 21545
+    },
+    {
+      "epoch": 2.71417325314104,
+      "grad_norm": 0.1939440220594406,
+      "learning_rate": 8.21102798134844e-06,
+      "loss": 0.1572,
+      "step": 21550
+    },
+    {
+      "epoch": 2.714803035551217,
+      "grad_norm": 0.2128129005432129,
+      "learning_rate": 8.175196203901157e-06,
+      "loss": 0.1624,
+      "step": 21555
+    },
+    {
+      "epoch": 2.7154328179613945,
+      "grad_norm": 0.22001588344573975,
+      "learning_rate": 8.139440589613122e-06,
+      "loss": 0.1498,
+      "step": 21560
+    },
+    {
+      "epoch": 2.7160626003715715,
+      "grad_norm": 0.24958358705043793,
+      "learning_rate": 8.103761157685939e-06,
+      "loss": 0.1614,
+      "step": 21565
+    },
+    {
+      "epoch": 2.716692382781749,
+      "grad_norm": 0.21756353974342346,
+      "learning_rate": 8.068157927280284e-06,
+      "loss": 0.1515,
+      "step": 21570
+    },
+    {
+      "epoch": 2.717322165191926,
+      "grad_norm": 0.19753116369247437,
+      "learning_rate": 8.032630917515842e-06,
+      "loss": 0.1504,
+      "step": 21575
+    },
+    {
+      "epoch": 2.7179519476021037,
+      "grad_norm": 0.2083761841058731,
+      "learning_rate": 7.997180147471505e-06,
+      "loss": 0.1488,
+      "step": 21580
+    },
+    {
+      "epoch": 2.7185817300122808,
+      "grad_norm": 0.2009708434343338,
+      "learning_rate": 7.961805636185126e-06,
+      "loss": 0.1475,
+      "step": 21585
+    },
+    {
+      "epoch": 2.719211512422458,
+      "grad_norm": 0.23513175547122955,
+      "learning_rate": 7.926507402653609e-06,
+      "loss": 0.1479,
+      "step": 21590
+    },
+    {
+      "epoch": 2.7198412948326354,
+      "grad_norm": 0.1990012526512146,
+      "learning_rate": 7.891285465832909e-06,
+      "loss": 0.1498,
+      "step": 21595
+    },
+    {
+      "epoch": 2.7204710772428125,
+      "grad_norm": 0.2000730186700821,
+      "learning_rate": 7.856139844638044e-06,
+      "loss": 0.1553,
+      "step": 21600
+    },
+    {
+      "epoch": 2.72110085965299,
+      "grad_norm": 0.17009419202804565,
+      "learning_rate": 7.821070557942966e-06,
+      "loss": 0.138,
+      "step": 21605
+    },
+    {
+      "epoch": 2.721730642063167,
+      "grad_norm": 0.19666020572185516,
+      "learning_rate": 7.786077624580728e-06,
+      "loss": 0.1505,
+      "step": 21610
+    },
+    {
+      "epoch": 2.7223604244733446,
+      "grad_norm": 0.20230218768119812,
+      "learning_rate": 7.751161063343314e-06,
+      "loss": 0.1459,
+      "step": 21615
+    },
+    {
+      "epoch": 2.7229902068835217,
+      "grad_norm": 0.20249028503894806,
+      "learning_rate": 7.716320892981692e-06,
+      "loss": 0.1481,
+      "step": 21620
+    },
+    {
+      "epoch": 2.7236199892936988,
+      "grad_norm": 0.183380126953125,
+      "learning_rate": 7.681557132205861e-06,
+      "loss": 0.1513,
+      "step": 21625
+    },
+    {
+      "epoch": 2.7242497717038763,
+      "grad_norm": 0.22188283503055573,
+      "learning_rate": 7.646869799684791e-06,
+      "loss": 0.1534,
+      "step": 21630
+    },
+    {
+      "epoch": 2.724879554114054,
+      "grad_norm": 0.19538500905036926,
+      "learning_rate": 7.6122589140462766e-06,
+      "loss": 0.1524,
+      "step": 21635
+    },
+    {
+      "epoch": 2.725509336524231,
+      "grad_norm": 0.1824834644794464,
+      "learning_rate": 7.577724493877219e-06,
+      "loss": 0.1564,
+      "step": 21640
+    },
+    {
+      "epoch": 2.726139118934408,
+      "grad_norm": 0.18397974967956543,
+      "learning_rate": 7.543266557723398e-06,
+      "loss": 0.1467,
+      "step": 21645
+    },
+    {
+      "epoch": 2.7267689013445855,
+      "grad_norm": 0.22993116080760956,
+      "learning_rate": 7.508885124089481e-06,
+      "loss": 0.1546,
+      "step": 21650
+    },
+    {
+      "epoch": 2.7273986837547626,
+      "grad_norm": 0.18351049721240997,
+      "learning_rate": 7.47458021143908e-06,
+      "loss": 0.1616,
+      "step": 21655
+    },
+    {
+      "epoch": 2.72802846616494,
+      "grad_norm": 0.20072756707668304,
+      "learning_rate": 7.440351838194724e-06,
+      "loss": 0.1451,
+      "step": 21660
+    },
+    {
+      "epoch": 2.728658248575117,
+      "grad_norm": 0.19199103116989136,
+      "learning_rate": 7.406200022737879e-06,
+      "loss": 0.1518,
+      "step": 21665
+    },
+    {
+      "epoch": 2.7292880309852947,
+      "grad_norm": 0.21039634943008423,
+      "learning_rate": 7.372124783408789e-06,
+      "loss": 0.154,
+      "step": 21670
+    },
+    {
+      "epoch": 2.729917813395472,
+      "grad_norm": 0.2162015289068222,
+      "learning_rate": 7.33812613850665e-06,
+      "loss": 0.1459,
+      "step": 21675
+    },
+    {
+      "epoch": 2.730547595805649,
+      "grad_norm": 0.192021444439888,
+      "learning_rate": 7.304204106289507e-06,
+      "loss": 0.1547,
+      "step": 21680
+    },
+    {
+      "epoch": 2.7311773782158264,
+      "grad_norm": 0.20860375463962555,
+      "learning_rate": 7.270358704974289e-06,
+      "loss": 0.1501,
+      "step": 21685
+    },
+    {
+      "epoch": 2.731807160626004,
+      "grad_norm": 0.1841016709804535,
+      "learning_rate": 7.236589952736738e-06,
+      "loss": 0.1538,
+      "step": 21690
+    },
+    {
+      "epoch": 2.732436943036181,
+      "grad_norm": 0.23411309719085693,
+      "learning_rate": 7.202897867711449e-06,
+      "loss": 0.153,
+      "step": 21695
+    },
+    {
+      "epoch": 2.733066725446358,
+      "grad_norm": 0.2005651742219925,
+      "learning_rate": 7.1692824679918325e-06,
+      "loss": 0.1505,
+      "step": 21700
+    },
+    {
+      "epoch": 2.7336965078565356,
+      "grad_norm": 0.18157663941383362,
+      "learning_rate": 7.135743771630131e-06,
+      "loss": 0.1424,
+      "step": 21705
+    },
+    {
+      "epoch": 2.7343262902667127,
+      "grad_norm": 0.20939917862415314,
+      "learning_rate": 7.102281796637388e-06,
+      "loss": 0.1585,
+      "step": 21710
+    },
+    {
+      "epoch": 2.7349560726768902,
+      "grad_norm": 0.17006689310073853,
+      "learning_rate": 7.068896560983445e-06,
+      "loss": 0.1529,
+      "step": 21715
+    },
+    {
+      "epoch": 2.7355858550870673,
+      "grad_norm": 0.23061016201972961,
+      "learning_rate": 7.035588082596927e-06,
+      "loss": 0.1556,
+      "step": 21720
+    },
+    {
+      "epoch": 2.736215637497245,
+      "grad_norm": 0.20175643265247345,
+      "learning_rate": 7.002356379365276e-06,
+      "loss": 0.1559,
+      "step": 21725
+    },
+    {
+      "epoch": 2.736845419907422,
+      "grad_norm": 0.19943305850028992,
+      "learning_rate": 6.969201469134683e-06,
+      "loss": 0.147,
+      "step": 21730
+    },
+    {
+      "epoch": 2.737475202317599,
+      "grad_norm": 0.22196878492832184,
+      "learning_rate": 6.936123369710056e-06,
+      "loss": 0.1517,
+      "step": 21735
+    },
+    {
+      "epoch": 2.7381049847277765,
+      "grad_norm": 0.19505414366722107,
+      "learning_rate": 6.903122098855085e-06,
+      "loss": 0.1464,
+      "step": 21740
+    },
+    {
+      "epoch": 2.738734767137954,
+      "grad_norm": 0.19797982275485992,
+      "learning_rate": 6.870197674292227e-06,
+      "loss": 0.1407,
+      "step": 21745
+    },
+    {
+      "epoch": 2.739364549548131,
+      "grad_norm": 0.2223568856716156,
+      "learning_rate": 6.837350113702672e-06,
+      "loss": 0.1524,
+      "step": 21750
+    },
+    {
+      "epoch": 2.739994331958308,
+      "grad_norm": 0.22087423503398895,
+      "learning_rate": 6.804579434726276e-06,
+      "loss": 0.1578,
+      "step": 21755
+    },
+    {
+      "epoch": 2.7406241143684857,
+      "grad_norm": 0.19389192759990692,
+      "learning_rate": 6.771885654961662e-06,
+      "loss": 0.1445,
+      "step": 21760
+    },
+    {
+      "epoch": 2.741253896778663,
+      "grad_norm": 0.20979470014572144,
+      "learning_rate": 6.739268791966118e-06,
+      "loss": 0.1548,
+      "step": 21765
+    },
+    {
+      "epoch": 2.7418836791888404,
+      "grad_norm": 0.22365309298038483,
+      "learning_rate": 6.7067288632556505e-06,
+      "loss": 0.1471,
+      "step": 21770
+    },
+    {
+      "epoch": 2.7425134615990174,
+      "grad_norm": 0.20007841289043427,
+      "learning_rate": 6.674265886304964e-06,
+      "loss": 0.1548,
+      "step": 21775
+    },
+    {
+      "epoch": 2.743143244009195,
+      "grad_norm": 0.1756853312253952,
+      "learning_rate": 6.641879878547379e-06,
+      "loss": 0.1443,
+      "step": 21780
+    },
+    {
+      "epoch": 2.743773026419372,
+      "grad_norm": 0.21500404179096222,
+      "learning_rate": 6.609570857374952e-06,
+      "loss": 0.1584,
+      "step": 21785
+    },
+    {
+      "epoch": 2.744402808829549,
+      "grad_norm": 0.1938805729150772,
+      "learning_rate": 6.577338840138369e-06,
+      "loss": 0.155,
+      "step": 21790
+    },
+    {
+      "epoch": 2.7450325912397266,
+      "grad_norm": 0.20673929154872894,
+      "learning_rate": 6.545183844146951e-06,
+      "loss": 0.1526,
+      "step": 21795
+    },
+    {
+      "epoch": 2.745662373649904,
+      "grad_norm": 0.19749803841114044,
+      "learning_rate": 6.513105886668668e-06,
+      "loss": 0.1533,
+      "step": 21800
+    },
+    {
+      "epoch": 2.7462921560600813,
+      "grad_norm": 0.212607279419899,
+      "learning_rate": 6.481104984930107e-06,
+      "loss": 0.1565,
+      "step": 21805
+    },
+    {
+      "epoch": 2.7469219384702583,
+      "grad_norm": 0.1796950250864029,
+      "learning_rate": 6.449181156116473e-06,
+      "loss": 0.1464,
+      "step": 21810
+    },
+    {
+      "epoch": 2.747551720880436,
+      "grad_norm": 0.18281513452529907,
+      "learning_rate": 6.417334417371616e-06,
+      "loss": 0.1482,
+      "step": 21815
+    },
+    {
+      "epoch": 2.748181503290613,
+      "grad_norm": 0.23321060836315155,
+      "learning_rate": 6.385564785797958e-06,
+      "loss": 0.1489,
+      "step": 21820
+    },
+    {
+      "epoch": 2.7488112857007905,
+      "grad_norm": 0.2202220857143402,
+      "learning_rate": 6.353872278456501e-06,
+      "loss": 0.1477,
+      "step": 21825
+    },
+    {
+      "epoch": 2.7494410681109676,
+      "grad_norm": 0.283456951379776,
+      "learning_rate": 6.3222569123668635e-06,
+      "loss": 0.1582,
+      "step": 21830
+    },
+    {
+      "epoch": 2.750070850521145,
+      "grad_norm": 0.18883143365383148,
+      "learning_rate": 6.29071870450723e-06,
+      "loss": 0.1469,
+      "step": 21835
+    },
+    {
+      "epoch": 2.750700632931322,
+      "grad_norm": 0.20364224910736084,
+      "learning_rate": 6.259257671814272e-06,
+      "loss": 0.1567,
+      "step": 21840
+    },
+    {
+      "epoch": 2.7513304153414992,
+      "grad_norm": 0.19058570265769958,
+      "learning_rate": 6.227873831183355e-06,
+      "loss": 0.1449,
+      "step": 21845
+    },
+    {
+      "epoch": 2.7519601977516768,
+      "grad_norm": 0.20439192652702332,
+      "learning_rate": 6.196567199468299e-06,
+      "loss": 0.1486,
+      "step": 21850
+    },
+    {
+      "epoch": 2.7525899801618543,
+      "grad_norm": 0.1962665468454361,
+      "learning_rate": 6.165337793481473e-06,
+      "loss": 0.1499,
+      "step": 21855
+    },
+    {
+      "epoch": 2.7532197625720314,
+      "grad_norm": 0.22097113728523254,
+      "learning_rate": 6.134185629993793e-06,
+      "loss": 0.153,
+      "step": 21860
+    },
+    {
+      "epoch": 2.7538495449822085,
+      "grad_norm": 0.20070448517799377,
+      "learning_rate": 6.103110725734644e-06,
+      "loss": 0.1463,
+      "step": 21865
+    },
+    {
+      "epoch": 2.754479327392386,
+      "grad_norm": 0.20577707886695862,
+      "learning_rate": 6.072113097392028e-06,
+      "loss": 0.1549,
+      "step": 21870
+    },
+    {
+      "epoch": 2.755109109802563,
+      "grad_norm": 0.1798795461654663,
+      "learning_rate": 6.041192761612313e-06,
+      "loss": 0.1454,
+      "step": 21875
+    },
+    {
+      "epoch": 2.7557388922127406,
+      "grad_norm": 0.20694920420646667,
+      "learning_rate": 6.010349735000464e-06,
+      "loss": 0.1524,
+      "step": 21880
+    },
+    {
+      "epoch": 2.7563686746229177,
+      "grad_norm": 0.19873858988285065,
+      "learning_rate": 5.979584034119867e-06,
+      "loss": 0.1523,
+      "step": 21885
+    },
+    {
+      "epoch": 2.756998457033095,
+      "grad_norm": 0.2215358465909958,
+      "learning_rate": 5.948895675492421e-06,
+      "loss": 0.1508,
+      "step": 21890
+    },
+    {
+      "epoch": 2.7576282394432723,
+      "grad_norm": 0.21731533110141754,
+      "learning_rate": 5.918284675598478e-06,
+      "loss": 0.149,
+      "step": 21895
+    },
+    {
+      "epoch": 2.7582580218534494,
+      "grad_norm": 0.21298860013484955,
+      "learning_rate": 5.887751050876837e-06,
+      "loss": 0.156,
+      "step": 21900
+    },
+    {
+      "epoch": 2.758887804263627,
+      "grad_norm": 0.20131991803646088,
+      "learning_rate": 5.85729481772475e-06,
+      "loss": 0.1403,
+      "step": 21905
+    },
+    {
+      "epoch": 2.7595175866738044,
+      "grad_norm": 0.17870669066905975,
+      "learning_rate": 5.826915992497932e-06,
+      "loss": 0.1483,
+      "step": 21910
+    },
+    {
+      "epoch": 2.7601473690839815,
+      "grad_norm": 0.2430955022573471,
+      "learning_rate": 5.796614591510468e-06,
+      "loss": 0.1484,
+      "step": 21915
+    },
+    {
+      "epoch": 2.7607771514941586,
+      "grad_norm": 0.1986503154039383,
+      "learning_rate": 5.766390631034939e-06,
+      "loss": 0.1524,
+      "step": 21920
+    },
+    {
+      "epoch": 2.761406933904336,
+      "grad_norm": 0.1926422268152237,
+      "learning_rate": 5.7362441273022645e-06,
+      "loss": 0.1484,
+      "step": 21925
+    },
+    {
+      "epoch": 2.762036716314513,
+      "grad_norm": 0.23347438871860504,
+      "learning_rate": 5.706175096501825e-06,
+      "loss": 0.1512,
+      "step": 21930
+    },
+    {
+      "epoch": 2.7626664987246907,
+      "grad_norm": 0.20513305068016052,
+      "learning_rate": 5.676183554781405e-06,
+      "loss": 0.1518,
+      "step": 21935
+    },
+    {
+      "epoch": 2.763296281134868,
+      "grad_norm": 0.18283484876155853,
+      "learning_rate": 5.64626951824712e-06,
+      "loss": 0.1381,
+      "step": 21940
+    },
+    {
+      "epoch": 2.7639260635450453,
+      "grad_norm": 0.17075172066688538,
+      "learning_rate": 5.616433002963472e-06,
+      "loss": 0.1501,
+      "step": 21945
+    },
+    {
+      "epoch": 2.7645558459552224,
+      "grad_norm": 0.2107374221086502,
+      "learning_rate": 5.5866740249533746e-06,
+      "loss": 0.1581,
+      "step": 21950
+    },
+    {
+      "epoch": 2.7651856283653995,
+      "grad_norm": 0.23205851018428802,
+      "learning_rate": 5.556992600198079e-06,
+      "loss": 0.1467,
+      "step": 21955
+    },
+    {
+      "epoch": 2.765815410775577,
+      "grad_norm": 0.1973281055688858,
+      "learning_rate": 5.527388744637201e-06,
+      "loss": 0.1434,
+      "step": 21960
+    },
+    {
+      "epoch": 2.7664451931857545,
+      "grad_norm": 0.20235906541347504,
+      "learning_rate": 5.497862474168657e-06,
+      "loss": 0.1454,
+      "step": 21965
+    },
+    {
+      "epoch": 2.7670749755959316,
+      "grad_norm": 0.21266506612300873,
+      "learning_rate": 5.4684138046487134e-06,
+      "loss": 0.1454,
+      "step": 21970
+    },
+    {
+      "epoch": 2.7677047580061087,
+      "grad_norm": 0.1890571415424347,
+      "learning_rate": 5.43904275189207e-06,
+      "loss": 0.1414,
+      "step": 21975
+    },
+    {
+      "epoch": 2.7683345404162862,
+      "grad_norm": 0.1897963136434555,
+      "learning_rate": 5.409749331671559e-06,
+      "loss": 0.1493,
+      "step": 21980
+    },
+    {
+      "epoch": 2.7689643228264633,
+      "grad_norm": 0.18935035169124603,
+      "learning_rate": 5.380533559718414e-06,
+      "loss": 0.1543,
+      "step": 21985
+    },
+    {
+      "epoch": 2.769594105236641,
+      "grad_norm": 0.20879988372325897,
+      "learning_rate": 5.351395451722251e-06,
+      "loss": 0.151,
+      "step": 21990
+    },
+    {
+      "epoch": 2.770223887646818,
+      "grad_norm": 0.20008423924446106,
+      "learning_rate": 5.322335023330837e-06,
+      "loss": 0.1515,
+      "step": 21995
+    },
+    {
+      "epoch": 2.7708536700569955,
+      "grad_norm": 0.18473681807518005,
+      "learning_rate": 5.293352290150321e-06,
+      "loss": 0.1464,
+      "step": 22000
+    },
+    {
+      "epoch": 2.7708536700569955,
+      "eval_loss": 0.3584047257900238,
+      "eval_runtime": 6.1661,
+      "eval_samples_per_second": 162.176,
+      "eval_steps_per_second": 10.217,
+      "step": 22000
+    },
+    {
+      "epoch": 2.7714834524671725,
+      "grad_norm": 0.19308076798915863,
+      "learning_rate": 5.264447267745053e-06,
+      "loss": 0.1582,
+      "step": 22005
+    },
+    {
+      "epoch": 2.7721132348773496,
+      "grad_norm": 0.23008759319782257,
+      "learning_rate": 5.235619971637734e-06,
+      "loss": 0.1546,
+      "step": 22010
+    },
+    {
+      "epoch": 2.772743017287527,
+      "grad_norm": 0.21323955059051514,
+      "learning_rate": 5.206870417309245e-06,
+      "loss": 0.1536,
+      "step": 22015
+    },
+    {
+      "epoch": 2.7733727996977047,
+      "grad_norm": 0.23257404565811157,
+      "learning_rate": 5.17819862019877e-06,
+      "loss": 0.1516,
+      "step": 22020
+    },
+    {
+      "epoch": 2.7740025821078818,
+      "grad_norm": 0.22094878554344177,
+      "learning_rate": 5.14960459570371e-06,
+      "loss": 0.1546,
+      "step": 22025
+    },
+    {
+      "epoch": 2.774632364518059,
+      "grad_norm": 0.21868959069252014,
+      "learning_rate": 5.121088359179698e-06,
+      "loss": 0.1567,
+      "step": 22030
+    },
+    {
+      "epoch": 2.7752621469282364,
+      "grad_norm": 0.19147329032421112,
+      "learning_rate": 5.09264992594065e-06,
+      "loss": 0.1502,
+      "step": 22035
+    },
+    {
+      "epoch": 2.7758919293384134,
+      "grad_norm": 0.17076317965984344,
+      "learning_rate": 5.064289311258618e-06,
+      "loss": 0.1511,
+      "step": 22040
+    },
+    {
+      "epoch": 2.776521711748591,
+      "grad_norm": 0.23041397333145142,
+      "learning_rate": 5.036006530363917e-06,
+      "loss": 0.1611,
+      "step": 22045
+    },
+    {
+      "epoch": 2.777151494158768,
+      "grad_norm": 0.21972966194152832,
+      "learning_rate": 5.007801598445033e-06,
+      "loss": 0.1493,
+      "step": 22050
+    },
+    {
+      "epoch": 2.7777812765689456,
+      "grad_norm": 0.17348721623420715,
+      "learning_rate": 4.979674530648664e-06,
+      "loss": 0.1481,
+      "step": 22055
+    },
+    {
+      "epoch": 2.7784110589791227,
+      "grad_norm": 0.19225727021694183,
+      "learning_rate": 4.9516253420796795e-06,
+      "loss": 0.1493,
+      "step": 22060
+    },
+    {
+      "epoch": 2.7790408413892997,
+      "grad_norm": 0.19729195535182953,
+      "learning_rate": 4.9236540478011625e-06,
+      "loss": 0.1442,
+      "step": 22065
+    },
+    {
+      "epoch": 2.7796706237994773,
+      "grad_norm": 0.17798985540866852,
+      "learning_rate": 4.8957606628342805e-06,
+      "loss": 0.1507,
+      "step": 22070
+    },
+    {
+      "epoch": 2.780300406209655,
+      "grad_norm": 0.19311825931072235,
+      "learning_rate": 4.867945202158469e-06,
+      "loss": 0.149,
+      "step": 22075
+    },
+    {
+      "epoch": 2.780930188619832,
+      "grad_norm": 0.18525920808315277,
+      "learning_rate": 4.840207680711278e-06,
+      "loss": 0.1635,
+      "step": 22080
+    },
+    {
+      "epoch": 2.781559971030009,
+      "grad_norm": 0.18988420069217682,
+      "learning_rate": 4.812548113388342e-06,
+      "loss": 0.153,
+      "step": 22085
+    },
+    {
+      "epoch": 2.7821897534401865,
+      "grad_norm": 0.18699151277542114,
+      "learning_rate": 4.784966515043498e-06,
+      "loss": 0.147,
+      "step": 22090
+    },
+    {
+      "epoch": 2.7828195358503636,
+      "grad_norm": 0.23182329535484314,
+      "learning_rate": 4.757462900488695e-06,
+      "loss": 0.1496,
+      "step": 22095
+    },
+    {
+      "epoch": 2.783449318260541,
+      "grad_norm": 0.20079541206359863,
+      "learning_rate": 4.730037284494021e-06,
+      "loss": 0.1583,
+      "step": 22100
+    },
+    {
+      "epoch": 2.784079100670718,
+      "grad_norm": 0.21548844873905182,
+      "learning_rate": 4.702689681787625e-06,
+      "loss": 0.1481,
+      "step": 22105
+    },
+    {
+      "epoch": 2.7847088830808957,
+      "grad_norm": 0.1968826800584793,
+      "learning_rate": 4.6754201070558105e-06,
+      "loss": 0.1452,
+      "step": 22110
+    },
+    {
+      "epoch": 2.785338665491073,
+      "grad_norm": 0.20061470568180084,
+      "learning_rate": 4.648228574942997e-06,
+      "loss": 0.1472,
+      "step": 22115
+    },
+    {
+      "epoch": 2.78596844790125,
+      "grad_norm": 0.19061359763145447,
+      "learning_rate": 4.621115100051604e-06,
+      "loss": 0.1478,
+      "step": 22120
+    },
+    {
+      "epoch": 2.7865982303114274,
+      "grad_norm": 0.23252861201763153,
+      "learning_rate": 4.594079696942199e-06,
+      "loss": 0.1527,
+      "step": 22125
+    },
+    {
+      "epoch": 2.787228012721605,
+      "grad_norm": 0.1698002964258194,
+      "learning_rate": 4.56712238013342e-06,
+      "loss": 0.1379,
+      "step": 22130
+    },
+    {
+      "epoch": 2.787857795131782,
+      "grad_norm": 0.19811010360717773,
+      "learning_rate": 4.540243164101954e-06,
+      "loss": 0.1417,
+      "step": 22135
+    },
+    {
+      "epoch": 2.788487577541959,
+      "grad_norm": 0.2089819759130478,
+      "learning_rate": 4.513442063282585e-06,
+      "loss": 0.1517,
+      "step": 22140
+    },
+    {
+      "epoch": 2.7891173599521366,
+      "grad_norm": 0.21028514206409454,
+      "learning_rate": 4.486719092068086e-06,
+      "loss": 0.1536,
+      "step": 22145
+    },
+    {
+      "epoch": 2.7897471423623137,
+      "grad_norm": 0.20895244181156158,
+      "learning_rate": 4.46007426480931e-06,
+      "loss": 0.1421,
+      "step": 22150
+    },
+    {
+      "epoch": 2.790376924772491,
+      "grad_norm": 0.1925353854894638,
+      "learning_rate": 4.4335075958151275e-06,
+      "loss": 0.1506,
+      "step": 22155
+    },
+    {
+      "epoch": 2.7910067071826683,
+      "grad_norm": 0.21809720993041992,
+      "learning_rate": 4.407019099352477e-06,
+      "loss": 0.1537,
+      "step": 22160
+    },
+    {
+      "epoch": 2.791636489592846,
+      "grad_norm": 0.23316286504268646,
+      "learning_rate": 4.380608789646245e-06,
+      "loss": 0.1593,
+      "step": 22165
+    },
+    {
+      "epoch": 2.792266272003023,
+      "grad_norm": 0.20298117399215698,
+      "learning_rate": 4.354276680879404e-06,
+      "loss": 0.1469,
+      "step": 22170
+    },
+    {
+      "epoch": 2.7928960544132,
+      "grad_norm": 0.18828284740447998,
+      "learning_rate": 4.328022787192875e-06,
+      "loss": 0.1478,
+      "step": 22175
+    },
+    {
+      "epoch": 2.7935258368233775,
+      "grad_norm": 0.19351090490818024,
+      "learning_rate": 4.301847122685614e-06,
+      "loss": 0.139,
+      "step": 22180
+    },
+    {
+      "epoch": 2.794155619233555,
+      "grad_norm": 0.19426658749580383,
+      "learning_rate": 4.27574970141456e-06,
+      "loss": 0.148,
+      "step": 22185
+    },
+    {
+      "epoch": 2.794785401643732,
+      "grad_norm": 0.18554694950580597,
+      "learning_rate": 4.2497305373945855e-06,
+      "loss": 0.1484,
+      "step": 22190
+    },
+    {
+      "epoch": 2.795415184053909,
+      "grad_norm": 0.21555371582508087,
+      "learning_rate": 4.223789644598613e-06,
+      "loss": 0.1537,
+      "step": 22195
+    },
+    {
+      "epoch": 2.7960449664640867,
+      "grad_norm": 0.20736396312713623,
+      "learning_rate": 4.197927036957499e-06,
+      "loss": 0.1533,
+      "step": 22200
+    },
+    {
+      "epoch": 2.796674748874264,
+      "grad_norm": 0.2143113762140274,
+      "learning_rate": 4.172142728360017e-06,
+      "loss": 0.1509,
+      "step": 22205
+    },
+    {
+      "epoch": 2.7973045312844413,
+      "grad_norm": 0.1888829916715622,
+      "learning_rate": 4.146436732652958e-06,
+      "loss": 0.1507,
+      "step": 22210
+    },
+    {
+      "epoch": 2.7979343136946184,
+      "grad_norm": 0.19072696566581726,
+      "learning_rate": 4.1208090636410286e-06,
+      "loss": 0.153,
+      "step": 22215
+    },
+    {
+      "epoch": 2.798564096104796,
+      "grad_norm": 0.23674504458904266,
+      "learning_rate": 4.09525973508687e-06,
+      "loss": 0.1584,
+      "step": 22220
+    },
+    {
+      "epoch": 2.799193878514973,
+      "grad_norm": 0.23174551129341125,
+      "learning_rate": 4.06978876071104e-06,
+      "loss": 0.1475,
+      "step": 22225
+    },
+    {
+      "epoch": 2.79982366092515,
+      "grad_norm": 0.2185906022787094,
+      "learning_rate": 4.044396154192031e-06,
+      "loss": 0.1494,
+      "step": 22230
+    },
+    {
+      "epoch": 2.8004534433353276,
+      "grad_norm": 0.1940082162618637,
+      "learning_rate": 4.019081929166268e-06,
+      "loss": 0.1497,
+      "step": 22235
+    },
+    {
+      "epoch": 2.801083225745505,
+      "grad_norm": 0.1945921629667282,
+      "learning_rate": 3.993846099228093e-06,
+      "loss": 0.1524,
+      "step": 22240
+    },
+    {
+      "epoch": 2.8017130081556822,
+      "grad_norm": 0.21760894358158112,
+      "learning_rate": 3.968688677929682e-06,
+      "loss": 0.1459,
+      "step": 22245
+    },
+    {
+      "epoch": 2.8023427905658593,
+      "grad_norm": 0.19670112431049347,
+      "learning_rate": 3.943609678781162e-06,
+      "loss": 0.151,
+      "step": 22250
+    },
+    {
+      "epoch": 2.802972572976037,
+      "grad_norm": 0.2076457440853119,
+      "learning_rate": 3.918609115250509e-06,
+      "loss": 0.1515,
+      "step": 22255
+    },
+    {
+      "epoch": 2.803602355386214,
+      "grad_norm": 0.20138059556484222,
+      "learning_rate": 3.893687000763635e-06,
+      "loss": 0.1492,
+      "step": 22260
+    },
+    {
+      "epoch": 2.8042321377963915,
+      "grad_norm": 0.20619480311870575,
+      "learning_rate": 3.868843348704265e-06,
+      "loss": 0.1516,
+      "step": 22265
+    },
+    {
+      "epoch": 2.8048619202065685,
+      "grad_norm": 0.17885464429855347,
+      "learning_rate": 3.844078172413994e-06,
+      "loss": 0.1413,
+      "step": 22270
+    },
+    {
+      "epoch": 2.805491702616746,
+      "grad_norm": 0.17029553651809692,
+      "learning_rate": 3.8193914851922855e-06,
+      "loss": 0.143,
+      "step": 22275
+    },
+    {
+      "epoch": 2.806121485026923,
+      "grad_norm": 0.18624289333820343,
+      "learning_rate": 3.794783300296483e-06,
+      "loss": 0.1448,
+      "step": 22280
+    },
+    {
+      "epoch": 2.8067512674371002,
+      "grad_norm": 0.20082144439220428,
+      "learning_rate": 3.7702536309417497e-06,
+      "loss": 0.1498,
+      "step": 22285
+    },
+    {
+      "epoch": 2.8073810498472778,
+      "grad_norm": 0.213558167219162,
+      "learning_rate": 3.745802490301031e-06,
+      "loss": 0.165,
+      "step": 22290
+    },
+    {
+      "epoch": 2.8080108322574553,
+      "grad_norm": 0.23692555725574493,
+      "learning_rate": 3.721429891505173e-06,
+      "loss": 0.1568,
+      "step": 22295
+    },
+    {
+      "epoch": 2.8086406146676324,
+      "grad_norm": 0.18088509142398834,
+      "learning_rate": 3.6971358476428237e-06,
+      "loss": 0.1508,
+      "step": 22300
+    },
+    {
+      "epoch": 2.8092703970778095,
+      "grad_norm": 0.20369915664196014,
+      "learning_rate": 3.672920371760446e-06,
+      "loss": 0.1469,
+      "step": 22305
+    },
+    {
+      "epoch": 2.809900179487987,
+      "grad_norm": 0.18801896274089813,
+      "learning_rate": 3.6487834768622883e-06,
+      "loss": 0.1417,
+      "step": 22310
+    },
+    {
+      "epoch": 2.810529961898164,
+      "grad_norm": 0.2028091549873352,
+      "learning_rate": 3.6247251759104145e-06,
+      "loss": 0.157,
+      "step": 22315
+    },
+    {
+      "epoch": 2.8111597443083416,
+      "grad_norm": 0.16689006984233856,
+      "learning_rate": 3.600745481824707e-06,
+      "loss": 0.1393,
+      "step": 22320
+    },
+    {
+      "epoch": 2.8117895267185187,
+      "grad_norm": 0.20889881253242493,
+      "learning_rate": 3.576844407482765e-06,
+      "loss": 0.1586,
+      "step": 22325
+    },
+    {
+      "epoch": 2.812419309128696,
+      "grad_norm": 0.21049942076206207,
+      "learning_rate": 3.5530219657200543e-06,
+      "loss": 0.155,
+      "step": 22330
+    },
+    {
+      "epoch": 2.8130490915388733,
+      "grad_norm": 0.23332563042640686,
+      "learning_rate": 3.5292781693297247e-06,
+      "loss": 0.1557,
+      "step": 22335
+    },
+    {
+      "epoch": 2.8136788739490504,
+      "grad_norm": 0.1706390082836151,
+      "learning_rate": 3.505613031062776e-06,
+      "loss": 0.1421,
+      "step": 22340
+    },
+    {
+      "epoch": 2.814308656359228,
+      "grad_norm": 0.17925478518009186,
+      "learning_rate": 3.4820265636279265e-06,
+      "loss": 0.1433,
+      "step": 22345
+    },
+    {
+      "epoch": 2.8149384387694054,
+      "grad_norm": 0.15641047060489655,
+      "learning_rate": 3.458518779691627e-06,
+      "loss": 0.1423,
+      "step": 22350
+    },
+    {
+      "epoch": 2.8155682211795825,
+      "grad_norm": 0.18733102083206177,
+      "learning_rate": 3.435089691878112e-06,
+      "loss": 0.1533,
+      "step": 22355
+    },
+    {
+      "epoch": 2.8161980035897596,
+      "grad_norm": 0.22065778076648712,
+      "learning_rate": 3.4117393127693183e-06,
+      "loss": 0.157,
+      "step": 22360
+    },
+    {
+      "epoch": 2.816827785999937,
+      "grad_norm": 0.20604351162910461,
+      "learning_rate": 3.388467654904947e-06,
+      "loss": 0.1438,
+      "step": 22365
+    },
+    {
+      "epoch": 2.817457568410114,
+      "grad_norm": 0.17883001267910004,
+      "learning_rate": 3.365274730782419e-06,
+      "loss": 0.1465,
+      "step": 22370
+    },
+    {
+      "epoch": 2.8180873508202917,
+      "grad_norm": 0.18118852376937866,
+      "learning_rate": 3.3421605528568374e-06,
+      "loss": 0.1501,
+      "step": 22375
+    },
+    {
+      "epoch": 2.818717133230469,
+      "grad_norm": 0.2178465574979782,
+      "learning_rate": 3.3191251335410564e-06,
+      "loss": 0.1467,
+      "step": 22380
+    },
+    {
+      "epoch": 2.8193469156406463,
+      "grad_norm": 0.18433082103729248,
+      "learning_rate": 3.29616848520563e-06,
+      "loss": 0.1478,
+      "step": 22385
+    },
+    {
+      "epoch": 2.8199766980508234,
+      "grad_norm": 0.19671761989593506,
+      "learning_rate": 3.273290620178831e-06,
+      "loss": 0.144,
+      "step": 22390
+    },
+    {
+      "epoch": 2.8206064804610005,
+      "grad_norm": 0.2000323235988617,
+      "learning_rate": 3.2504915507465144e-06,
+      "loss": 0.1443,
+      "step": 22395
+    },
+    {
+      "epoch": 2.821236262871178,
+      "grad_norm": 0.19443731009960175,
+      "learning_rate": 3.22777128915237e-06,
+      "loss": 0.1537,
+      "step": 22400
+    },
+    {
+      "epoch": 2.8218660452813555,
+      "grad_norm": 0.19904273748397827,
+      "learning_rate": 3.2051298475976707e-06,
+      "loss": 0.1581,
+      "step": 22405
+    },
+    {
+      "epoch": 2.8224958276915326,
+      "grad_norm": 0.1972033828496933,
+      "learning_rate": 3.18256723824139e-06,
+      "loss": 0.1383,
+      "step": 22410
+    },
+    {
+      "epoch": 2.8231256101017097,
+      "grad_norm": 0.21138480305671692,
+      "learning_rate": 3.16008347320017e-06,
+      "loss": 0.1442,
+      "step": 22415
+    },
+    {
+      "epoch": 2.8237553925118872,
+      "grad_norm": 0.22747448086738586,
+      "learning_rate": 3.1376785645483016e-06,
+      "loss": 0.1485,
+      "step": 22420
+    },
+    {
+      "epoch": 2.8243851749220643,
+      "grad_norm": 0.23757314682006836,
+      "learning_rate": 3.11535252431776e-06,
+      "loss": 0.1568,
+      "step": 22425
+    },
+    {
+      "epoch": 2.825014957332242,
+      "grad_norm": 0.2193070352077484,
+      "learning_rate": 3.0931053644980885e-06,
+      "loss": 0.1605,
+      "step": 22430
+    },
+    {
+      "epoch": 2.825644739742419,
+      "grad_norm": 0.2223901003599167,
+      "learning_rate": 3.0709370970365464e-06,
+      "loss": 0.1453,
+      "step": 22435
+    },
+    {
+      "epoch": 2.8262745221525964,
+      "grad_norm": 0.23655427992343903,
+      "learning_rate": 3.0488477338379944e-06,
+      "loss": 0.1484,
+      "step": 22440
+    },
+    {
+      "epoch": 2.8269043045627735,
+      "grad_norm": 0.20859979093074799,
+      "learning_rate": 3.026837286764944e-06,
+      "loss": 0.154,
+      "step": 22445
+    },
+    {
+      "epoch": 2.8275340869729506,
+      "grad_norm": 0.1994808316230774,
+      "learning_rate": 3.004905767637472e-06,
+      "loss": 0.1634,
+      "step": 22450
+    },
+    {
+      "epoch": 2.828163869383128,
+      "grad_norm": 0.19530266523361206,
+      "learning_rate": 2.983053188233342e-06,
+      "loss": 0.1458,
+      "step": 22455
+    },
+    {
+      "epoch": 2.8287936517933057,
+      "grad_norm": 0.19528019428253174,
+      "learning_rate": 2.9612795602878827e-06,
+      "loss": 0.1472,
+      "step": 22460
+    },
+    {
+      "epoch": 2.8294234342034827,
+      "grad_norm": 0.20543955266475677,
+      "learning_rate": 2.939584895494007e-06,
+      "loss": 0.1544,
+      "step": 22465
+    },
+    {
+      "epoch": 2.83005321661366,
+      "grad_norm": 0.18907050788402557,
+      "learning_rate": 2.917969205502263e-06,
+      "loss": 0.1469,
+      "step": 22470
+    },
+    {
+      "epoch": 2.8306829990238374,
+      "grad_norm": 0.2009141594171524,
+      "learning_rate": 2.896432501920748e-06,
+      "loss": 0.1463,
+      "step": 22475
+    },
+    {
+      "epoch": 2.8313127814340144,
+      "grad_norm": 0.1845710128545761,
+      "learning_rate": 2.8749747963151937e-06,
+      "loss": 0.1523,
+      "step": 22480
+    },
+    {
+      "epoch": 2.831942563844192,
+      "grad_norm": 0.22671662271022797,
+      "learning_rate": 2.853596100208866e-06,
+      "loss": 0.1553,
+      "step": 22485
+    },
+    {
+      "epoch": 2.832572346254369,
+      "grad_norm": 0.1716582477092743,
+      "learning_rate": 2.832296425082614e-06,
+      "loss": 0.1423,
+      "step": 22490
+    },
+    {
+      "epoch": 2.8332021286645466,
+      "grad_norm": 0.17477920651435852,
+      "learning_rate": 2.8110757823748554e-06,
+      "loss": 0.142,
+      "step": 22495
+    },
+    {
+      "epoch": 2.8338319110747237,
+      "grad_norm": 0.22391197085380554,
+      "learning_rate": 2.7899341834815236e-06,
+      "loss": 0.1576,
+      "step": 22500
+    },
+    {
+      "epoch": 2.8344616934849007,
+      "grad_norm": 0.19235247373580933,
+      "learning_rate": 2.7688716397561874e-06,
+      "loss": 0.1432,
+      "step": 22505
+    },
+    {
+      "epoch": 2.8350914758950783,
+      "grad_norm": 0.21828468143939972,
+      "learning_rate": 2.747888162509898e-06,
+      "loss": 0.1461,
+      "step": 22510
+    },
+    {
+      "epoch": 2.835721258305256,
+      "grad_norm": 0.19712364673614502,
+      "learning_rate": 2.726983763011259e-06,
+      "loss": 0.1461,
+      "step": 22515
+    },
+    {
+      "epoch": 2.836351040715433,
+      "grad_norm": 0.22868654131889343,
+      "learning_rate": 2.7061584524864066e-06,
+      "loss": 0.1546,
+      "step": 22520
+    },
+    {
+      "epoch": 2.83698082312561,
+      "grad_norm": 0.18876421451568604,
+      "learning_rate": 2.685412242119012e-06,
+      "loss": 0.1481,
+      "step": 22525
+    },
+    {
+      "epoch": 2.8376106055357875,
+      "grad_norm": 0.21973784267902374,
+      "learning_rate": 2.664745143050295e-06,
+      "loss": 0.1568,
+      "step": 22530
+    },
+    {
+      "epoch": 2.8382403879459646,
+      "grad_norm": 0.24478502571582794,
+      "learning_rate": 2.6441571663788963e-06,
+      "loss": 0.1558,
+      "step": 22535
+    },
+    {
+      "epoch": 2.838870170356142,
+      "grad_norm": 0.18952693045139313,
+      "learning_rate": 2.6236483231610707e-06,
+      "loss": 0.1461,
+      "step": 22540
+    },
+    {
+      "epoch": 2.839499952766319,
+      "grad_norm": 0.20026876032352448,
+      "learning_rate": 2.603218624410525e-06,
+      "loss": 0.1466,
+      "step": 22545
+    },
+    {
+      "epoch": 2.8401297351764967,
+      "grad_norm": 0.21935871243476868,
+      "learning_rate": 2.5828680810984824e-06,
+      "loss": 0.1563,
+      "step": 22550
+    },
+    {
+      "epoch": 2.8407595175866738,
+      "grad_norm": 0.22446821630001068,
+      "learning_rate": 2.5625967041536354e-06,
+      "loss": 0.1462,
+      "step": 22555
+    },
+    {
+      "epoch": 2.841389299996851,
+      "grad_norm": 0.2072252631187439,
+      "learning_rate": 2.5424045044621922e-06,
+      "loss": 0.1505,
+      "step": 22560
+    },
+    {
+      "epoch": 2.8420190824070284,
+      "grad_norm": 0.19828562438488007,
+      "learning_rate": 2.5222914928678285e-06,
+      "loss": 0.1462,
+      "step": 22565
+    },
+    {
+      "epoch": 2.842648864817206,
+      "grad_norm": 0.18411174416542053,
+      "learning_rate": 2.502257680171671e-06,
+      "loss": 0.1415,
+      "step": 22570
+    },
+    {
+      "epoch": 2.843278647227383,
+      "grad_norm": 0.20017574727535248,
+      "learning_rate": 2.482303077132347e-06,
+      "loss": 0.1556,
+      "step": 22575
+    },
+    {
+      "epoch": 2.84390842963756,
+      "grad_norm": 0.1881314069032669,
+      "learning_rate": 2.462427694465935e-06,
+      "loss": 0.1464,
+      "step": 22580
+    },
+    {
+      "epoch": 2.8445382120477376,
+      "grad_norm": 0.2211647629737854,
+      "learning_rate": 2.4426315428459466e-06,
+      "loss": 0.1471,
+      "step": 22585
+    },
+    {
+      "epoch": 2.8451679944579147,
+      "grad_norm": 0.20288364589214325,
+      "learning_rate": 2.4229146329033944e-06,
+      "loss": 0.146,
+      "step": 22590
+    },
+    {
+      "epoch": 2.845797776868092,
+      "grad_norm": 0.22115926444530487,
+      "learning_rate": 2.4032769752267087e-06,
+      "loss": 0.1422,
+      "step": 22595
+    },
+    {
+      "epoch": 2.8464275592782693,
+      "grad_norm": 0.196670264005661,
+      "learning_rate": 2.3837185803617544e-06,
+      "loss": 0.153,
+      "step": 22600
+    },
+    {
+      "epoch": 2.847057341688447,
+      "grad_norm": 0.23514890670776367,
+      "learning_rate": 2.3642394588118285e-06,
+      "loss": 0.1573,
+      "step": 22605
+    },
+    {
+      "epoch": 2.847687124098624,
+      "grad_norm": 0.1987423151731491,
+      "learning_rate": 2.3448396210376807e-06,
+      "loss": 0.1457,
+      "step": 22610
+    },
+    {
+      "epoch": 2.848316906508801,
+      "grad_norm": 0.18859946727752686,
+      "learning_rate": 2.3255190774574605e-06,
+      "loss": 0.1533,
+      "step": 22615
+    },
+    {
+      "epoch": 2.8489466889189785,
+      "grad_norm": 0.21700045466423035,
+      "learning_rate": 2.306277838446735e-06,
+      "loss": 0.1416,
+      "step": 22620
+    },
+    {
+      "epoch": 2.849576471329156,
+      "grad_norm": 0.17610225081443787,
+      "learning_rate": 2.2871159143384723e-06,
+      "loss": 0.1498,
+      "step": 22625
+    },
+    {
+      "epoch": 2.850206253739333,
+      "grad_norm": 0.2066749781370163,
+      "learning_rate": 2.26803331542309e-06,
+      "loss": 0.1587,
+      "step": 22630
+    },
+    {
+      "epoch": 2.85083603614951,
+      "grad_norm": 0.19877871870994568,
+      "learning_rate": 2.2490300519484082e-06,
+      "loss": 0.1526,
+      "step": 22635
+    },
+    {
+      "epoch": 2.8514658185596877,
+      "grad_norm": 0.19332483410835266,
+      "learning_rate": 2.230106134119547e-06,
+      "loss": 0.1562,
+      "step": 22640
+    },
+    {
+      "epoch": 2.852095600969865,
+      "grad_norm": 0.21806974709033966,
+      "learning_rate": 2.21126157209911e-06,
+      "loss": 0.1508,
+      "step": 22645
+    },
+    {
+      "epoch": 2.8527253833800423,
+      "grad_norm": 0.20896165072917938,
+      "learning_rate": 2.192496376007069e-06,
+      "loss": 0.1612,
+      "step": 22650
+    },
+    {
+      "epoch": 2.8533551657902194,
+      "grad_norm": 0.2381521761417389,
+      "learning_rate": 2.1738105559207465e-06,
+      "loss": 0.1545,
+      "step": 22655
+    },
+    {
+      "epoch": 2.853984948200397,
+      "grad_norm": 0.24022352695465088,
+      "learning_rate": 2.155204121874882e-06,
+      "loss": 0.1548,
+      "step": 22660
+    },
+    {
+      "epoch": 2.854614730610574,
+      "grad_norm": 0.20042377710342407,
+      "learning_rate": 2.1366770838615322e-06,
+      "loss": 0.1423,
+      "step": 22665
+    },
+    {
+      "epoch": 2.855244513020751,
+      "grad_norm": 0.1943242996931076,
+      "learning_rate": 2.118229451830139e-06,
+      "loss": 0.1453,
+      "step": 22670
+    },
+    {
+      "epoch": 2.8558742954309286,
+      "grad_norm": 0.20173771679401398,
+      "learning_rate": 2.0998612356874944e-06,
+      "loss": 0.1406,
+      "step": 22675
+    },
+    {
+      "epoch": 2.856504077841106,
+      "grad_norm": 0.21339194476604462,
+      "learning_rate": 2.081572445297791e-06,
+      "loss": 0.1447,
+      "step": 22680
+    },
+    {
+      "epoch": 2.8571338602512832,
+      "grad_norm": 0.18814577162265778,
+      "learning_rate": 2.0633630904824727e-06,
+      "loss": 0.144,
+      "step": 22685
+    },
+    {
+      "epoch": 2.8577636426614603,
+      "grad_norm": 0.1956281065940857,
+      "learning_rate": 2.045233181020417e-06,
+      "loss": 0.1503,
+      "step": 22690
+    },
+    {
+      "epoch": 2.858393425071638,
+      "grad_norm": 0.22954149544239044,
+      "learning_rate": 2.027182726647786e-06,
+      "loss": 0.1491,
+      "step": 22695
+    },
+    {
+      "epoch": 2.859023207481815,
+      "grad_norm": 0.18004447221755981,
+      "learning_rate": 2.009211737058092e-06,
+      "loss": 0.1492,
+      "step": 22700
+    },
+    {
+      "epoch": 2.8596529898919925,
+      "grad_norm": 0.226220041513443,
+      "learning_rate": 1.991320221902165e-06,
+      "loss": 0.159,
+      "step": 22705
+    },
+    {
+      "epoch": 2.8602827723021695,
+      "grad_norm": 0.1808856725692749,
+      "learning_rate": 1.9735081907881367e-06,
+      "loss": 0.1473,
+      "step": 22710
+    },
+    {
+      "epoch": 2.860912554712347,
+      "grad_norm": 0.2538818418979645,
+      "learning_rate": 1.9557756532815216e-06,
+      "loss": 0.1523,
+      "step": 22715
+    },
+    {
+      "epoch": 2.861542337122524,
+      "grad_norm": 0.18744130432605743,
+      "learning_rate": 1.9381226189050524e-06,
+      "loss": 0.1501,
+      "step": 22720
+    },
+    {
+      "epoch": 2.8621721195327012,
+      "grad_norm": 0.2162604182958603,
+      "learning_rate": 1.920549097138813e-06,
+      "loss": 0.1505,
+      "step": 22725
+    },
+    {
+      "epoch": 2.8628019019428788,
+      "grad_norm": 0.2076927125453949,
+      "learning_rate": 1.9030550974202197e-06,
+      "loss": 0.1493,
+      "step": 22730
+    },
+    {
+      "epoch": 2.8634316843530563,
+      "grad_norm": 0.2153797596693039,
+      "learning_rate": 1.885640629143942e-06,
+      "loss": 0.1409,
+      "step": 22735
+    },
+    {
+      "epoch": 2.8640614667632334,
+      "grad_norm": 0.19790925085544586,
+      "learning_rate": 1.868305701661932e-06,
+      "loss": 0.1596,
+      "step": 22740
+    },
+    {
+      "epoch": 2.8646912491734104,
+      "grad_norm": 0.18141327798366547,
+      "learning_rate": 1.8510503242834263e-06,
+      "loss": 0.1459,
+      "step": 22745
+    },
+    {
+      "epoch": 2.865321031583588,
+      "grad_norm": 0.20295578241348267,
+      "learning_rate": 1.833874506274996e-06,
+      "loss": 0.1485,
+      "step": 22750
+    },
+    {
+      "epoch": 2.865950813993765,
+      "grad_norm": 0.20226307213306427,
+      "learning_rate": 1.8167782568604127e-06,
+      "loss": 0.1507,
+      "step": 22755
+    },
+    {
+      "epoch": 2.8665805964039426,
+      "grad_norm": 0.19584356248378754,
+      "learning_rate": 1.7997615852207825e-06,
+      "loss": 0.1526,
+      "step": 22760
+    },
+    {
+      "epoch": 2.8672103788141197,
+      "grad_norm": 0.17093075811862946,
+      "learning_rate": 1.7828245004944286e-06,
+      "loss": 0.1481,
+      "step": 22765
+    },
+    {
+      "epoch": 2.867840161224297,
+      "grad_norm": 0.17845821380615234,
+      "learning_rate": 1.7659670117769587e-06,
+      "loss": 0.1459,
+      "step": 22770
+    },
+    {
+      "epoch": 2.8684699436344743,
+      "grad_norm": 0.1874646097421646,
+      "learning_rate": 1.749189128121231e-06,
+      "loss": 0.1493,
+      "step": 22775
+    },
+    {
+      "epoch": 2.8690997260446514,
+      "grad_norm": 0.18263909220695496,
+      "learning_rate": 1.7324908585373387e-06,
+      "loss": 0.1438,
+      "step": 22780
+    },
+    {
+      "epoch": 2.869729508454829,
+      "grad_norm": 0.1789528727531433,
+      "learning_rate": 1.7158722119926583e-06,
+      "loss": 0.1476,
+      "step": 22785
+    },
+    {
+      "epoch": 2.8703592908650064,
+      "grad_norm": 0.21683086454868317,
+      "learning_rate": 1.6993331974117508e-06,
+      "loss": 0.1499,
+      "step": 22790
+    },
+    {
+      "epoch": 2.8709890732751835,
+      "grad_norm": 0.24182718992233276,
+      "learning_rate": 1.6828738236764617e-06,
+      "loss": 0.1541,
+      "step": 22795
+    },
+    {
+      "epoch": 2.8716188556853606,
+      "grad_norm": 0.21868962049484253,
+      "learning_rate": 1.6664940996258702e-06,
+      "loss": 0.1471,
+      "step": 22800
+    },
+    {
+      "epoch": 2.872248638095538,
+      "grad_norm": 0.1993272453546524,
+      "learning_rate": 1.6501940340562236e-06,
+      "loss": 0.1526,
+      "step": 22805
+    },
+    {
+      "epoch": 2.872878420505715,
+      "grad_norm": 0.201304093003273,
+      "learning_rate": 1.6339736357210697e-06,
+      "loss": 0.1516,
+      "step": 22810
+    },
+    {
+      "epoch": 2.8735082029158927,
+      "grad_norm": 0.25056761503219604,
+      "learning_rate": 1.6178329133310908e-06,
+      "loss": 0.151,
+      "step": 22815
+    },
+    {
+      "epoch": 2.87413798532607,
+      "grad_norm": 0.19561152160167694,
+      "learning_rate": 1.6017718755542696e-06,
+      "loss": 0.143,
+      "step": 22820
+    },
+    {
+      "epoch": 2.8747677677362473,
+      "grad_norm": 0.22097674012184143,
+      "learning_rate": 1.5857905310157071e-06,
+      "loss": 0.1512,
+      "step": 22825
+    },
+    {
+      "epoch": 2.8753975501464244,
+      "grad_norm": 0.21212686598300934,
+      "learning_rate": 1.5698888882977712e-06,
+      "loss": 0.1541,
+      "step": 22830
+    },
+    {
+      "epoch": 2.8760273325566015,
+      "grad_norm": 0.20324502885341644,
+      "learning_rate": 1.5540669559399977e-06,
+      "loss": 0.1533,
+      "step": 22835
+    },
+    {
+      "epoch": 2.876657114966779,
+      "grad_norm": 0.169882133603096,
+      "learning_rate": 1.5383247424391564e-06,
+      "loss": 0.1406,
+      "step": 22840
+    },
+    {
+      "epoch": 2.8772868973769565,
+      "grad_norm": 0.23402316868305206,
+      "learning_rate": 1.5226622562491352e-06,
+      "loss": 0.1569,
+      "step": 22845
+    },
+    {
+      "epoch": 2.8779166797871336,
+      "grad_norm": 0.17247354984283447,
+      "learning_rate": 1.5070795057810559e-06,
+      "loss": 0.1432,
+      "step": 22850
+    },
+    {
+      "epoch": 2.8785464621973107,
+      "grad_norm": 0.21733173727989197,
+      "learning_rate": 1.4915764994032409e-06,
+      "loss": 0.1523,
+      "step": 22855
+    },
+    {
+      "epoch": 2.879176244607488,
+      "grad_norm": 0.19319911301136017,
+      "learning_rate": 1.4761532454411306e-06,
+      "loss": 0.1391,
+      "step": 22860
+    },
+    {
+      "epoch": 2.8798060270176653,
+      "grad_norm": 0.19645391404628754,
+      "learning_rate": 1.4608097521773664e-06,
+      "loss": 0.1499,
+      "step": 22865
+    },
+    {
+      "epoch": 2.880435809427843,
+      "grad_norm": 0.18772046267986298,
+      "learning_rate": 1.4455460278517572e-06,
+      "loss": 0.1483,
+      "step": 22870
+    },
+    {
+      "epoch": 2.88106559183802,
+      "grad_norm": 0.22282320261001587,
+      "learning_rate": 1.4303620806612792e-06,
+      "loss": 0.1468,
+      "step": 22875
+    },
+    {
+      "epoch": 2.8816953742481974,
+      "grad_norm": 0.19287440180778503,
+      "learning_rate": 1.4152579187600599e-06,
+      "loss": 0.1495,
+      "step": 22880
+    },
+    {
+      "epoch": 2.8823251566583745,
+      "grad_norm": 0.1981481909751892,
+      "learning_rate": 1.400233550259361e-06,
+      "loss": 0.1506,
+      "step": 22885
+    },
+    {
+      "epoch": 2.8829549390685516,
+      "grad_norm": 0.21331623196601868,
+      "learning_rate": 1.385288983227628e-06,
+      "loss": 0.1483,
+      "step": 22890
+    },
+    {
+      "epoch": 2.883584721478729,
+      "grad_norm": 0.20138582587242126,
+      "learning_rate": 1.3704242256904252e-06,
+      "loss": 0.1504,
+      "step": 22895
+    },
+    {
+      "epoch": 2.8842145038889067,
+      "grad_norm": 0.18493309617042542,
+      "learning_rate": 1.3556392856304831e-06,
+      "loss": 0.1539,
+      "step": 22900
+    },
+    {
+      "epoch": 2.8848442862990837,
+      "grad_norm": 0.22465452551841736,
+      "learning_rate": 1.3409341709876343e-06,
+      "loss": 0.1511,
+      "step": 22905
+    },
+    {
+      "epoch": 2.885474068709261,
+      "grad_norm": 0.19788892567157745,
+      "learning_rate": 1.326308889658878e-06,
+      "loss": 0.1504,
+      "step": 22910
+    },
+    {
+      "epoch": 2.8861038511194383,
+      "grad_norm": 0.20821528136730194,
+      "learning_rate": 1.3117634494982986e-06,
+      "loss": 0.1511,
+      "step": 22915
+    },
+    {
+      "epoch": 2.8867336335296154,
+      "grad_norm": 0.24520978331565857,
+      "learning_rate": 1.2972978583171644e-06,
+      "loss": 0.1503,
+      "step": 22920
+    },
+    {
+      "epoch": 2.887363415939793,
+      "grad_norm": 0.18331633508205414,
+      "learning_rate": 1.2829121238837947e-06,
+      "loss": 0.1485,
+      "step": 22925
+    },
+    {
+      "epoch": 2.88799319834997,
+      "grad_norm": 0.23501911759376526,
+      "learning_rate": 1.2686062539236762e-06,
+      "loss": 0.1452,
+      "step": 22930
+    },
+    {
+      "epoch": 2.8886229807601476,
+      "grad_norm": 0.19778122007846832,
+      "learning_rate": 1.2543802561193806e-06,
+      "loss": 0.1523,
+      "step": 22935
+    },
+    {
+      "epoch": 2.8892527631703246,
+      "grad_norm": 0.21170009672641754,
+      "learning_rate": 1.2402341381105962e-06,
+      "loss": 0.1627,
+      "step": 22940
+    },
+    {
+      "epoch": 2.8898825455805017,
+      "grad_norm": 0.2205863893032074,
+      "learning_rate": 1.22616790749413e-06,
+      "loss": 0.1507,
+      "step": 22945
+    },
+    {
+      "epoch": 2.8905123279906793,
+      "grad_norm": 0.1905989944934845,
+      "learning_rate": 1.2121815718238393e-06,
+      "loss": 0.15,
+      "step": 22950
+    },
+    {
+      "epoch": 2.891142110400857,
+      "grad_norm": 0.19767914712429047,
+      "learning_rate": 1.1982751386107159e-06,
+      "loss": 0.1499,
+      "step": 22955
+    },
+    {
+      "epoch": 2.891771892811034,
+      "grad_norm": 0.23298701643943787,
+      "learning_rate": 1.1844486153228361e-06,
+      "loss": 0.1507,
+      "step": 22960
+    },
+    {
+      "epoch": 2.892401675221211,
+      "grad_norm": 0.17696991562843323,
+      "learning_rate": 1.1707020093853602e-06,
+      "loss": 0.1458,
+      "step": 22965
+    },
+    {
+      "epoch": 2.8930314576313885,
+      "grad_norm": 0.22472181916236877,
+      "learning_rate": 1.1570353281805334e-06,
+      "loss": 0.154,
+      "step": 22970
+    },
+    {
+      "epoch": 2.8936612400415656,
+      "grad_norm": 0.18597114086151123,
+      "learning_rate": 1.1434485790476512e-06,
+      "loss": 0.1531,
+      "step": 22975
+    },
+    {
+      "epoch": 2.894291022451743,
+      "grad_norm": 0.20187944173812866,
+      "learning_rate": 1.1299417692831436e-06,
+      "loss": 0.1558,
+      "step": 22980
+    },
+    {
+      "epoch": 2.89492080486192,
+      "grad_norm": 0.19914616644382477,
+      "learning_rate": 1.1165149061404422e-06,
+      "loss": 0.1418,
+      "step": 22985
+    },
+    {
+      "epoch": 2.8955505872720977,
+      "grad_norm": 0.2554416060447693,
+      "learning_rate": 1.1031679968301122e-06,
+      "loss": 0.1556,
+      "step": 22990
+    },
+    {
+      "epoch": 2.8961803696822748,
+      "grad_norm": 0.2221318930387497,
+      "learning_rate": 1.08990104851972e-06,
+      "loss": 0.1523,
+      "step": 22995
+    },
+    {
+      "epoch": 2.896810152092452,
+      "grad_norm": 0.2058124542236328,
+      "learning_rate": 1.0767140683339336e-06,
+      "loss": 0.1403,
+      "step": 23000
+    },
+    {
+      "epoch": 2.896810152092452,
+      "eval_loss": 0.36229029297828674,
+      "eval_runtime": 6.1614,
+      "eval_samples_per_second": 162.302,
+      "eval_steps_per_second": 10.225,
+      "step": 23000
+    },
+    {
+      "epoch": 2.8974399345026294,
+      "grad_norm": 0.2248660773038864,
+      "learning_rate": 1.0636070633544547e-06,
+      "loss": 0.1514,
+      "step": 23005
+    },
+    {
+      "epoch": 2.8980697169128065,
+      "grad_norm": 0.1835104525089264,
+      "learning_rate": 1.0505800406200526e-06,
+      "loss": 0.1405,
+      "step": 23010
+    },
+    {
+      "epoch": 2.898699499322984,
+      "grad_norm": 0.22200733423233032,
+      "learning_rate": 1.0376330071265482e-06,
+      "loss": 0.1503,
+      "step": 23015
+    },
+    {
+      "epoch": 2.899329281733161,
+      "grad_norm": 0.18144001066684723,
+      "learning_rate": 1.024765969826763e-06,
+      "loss": 0.1516,
+      "step": 23020
+    },
+    {
+      "epoch": 2.8999590641433386,
+      "grad_norm": 0.17609558999538422,
+      "learning_rate": 1.0119789356306196e-06,
+      "loss": 0.1517,
+      "step": 23025
+    },
+    {
+      "epoch": 2.9005888465535157,
+      "grad_norm": 0.2412068098783493,
+      "learning_rate": 9.99271911405025e-07,
+      "loss": 0.157,
+      "step": 23030
+    },
+    {
+      "epoch": 2.901218628963693,
+      "grad_norm": 0.19257797300815582,
+      "learning_rate": 9.866449039739544e-07,
+      "loss": 0.1486,
+      "step": 23035
+    },
+    {
+      "epoch": 2.9018484113738703,
+      "grad_norm": 0.22018341720104218,
+      "learning_rate": 9.74097920118383e-07,
+      "loss": 0.1517,
+      "step": 23040
+    },
+    {
+      "epoch": 2.902478193784048,
+      "grad_norm": 0.19569897651672363,
+      "learning_rate": 9.616309665763544e-07,
+      "loss": 0.144,
+      "step": 23045
+    },
+    {
+      "epoch": 2.903107976194225,
+      "grad_norm": 0.20541365444660187,
+      "learning_rate": 9.492440500428966e-07,
+      "loss": 0.1535,
+      "step": 23050
+    },
+    {
+      "epoch": 2.903737758604402,
+      "grad_norm": 0.1934703141450882,
+      "learning_rate": 9.369371771700552e-07,
+      "loss": 0.1369,
+      "step": 23055
+    },
+    {
+      "epoch": 2.9043675410145795,
+      "grad_norm": 0.16967949271202087,
+      "learning_rate": 9.247103545669266e-07,
+      "loss": 0.1447,
+      "step": 23060
+    },
+    {
+      "epoch": 2.9049973234247566,
+      "grad_norm": 0.15972602367401123,
+      "learning_rate": 9.125635887995586e-07,
+      "loss": 0.1484,
+      "step": 23065
+    },
+    {
+      "epoch": 2.905627105834934,
+      "grad_norm": 0.21369343996047974,
+      "learning_rate": 9.004968863910667e-07,
+      "loss": 0.153,
+      "step": 23070
+    },
+    {
+      "epoch": 2.906256888245111,
+      "grad_norm": 0.18001650273799896,
+      "learning_rate": 8.885102538215338e-07,
+      "loss": 0.1499,
+      "step": 23075
+    },
+    {
+      "epoch": 2.9068866706552887,
+      "grad_norm": 0.20539362728595734,
+      "learning_rate": 8.766036975280777e-07,
+      "loss": 0.1495,
+      "step": 23080
+    },
+    {
+      "epoch": 2.907516453065466,
+      "grad_norm": 0.2048049122095108,
+      "learning_rate": 8.647772239047667e-07,
+      "loss": 0.1477,
+      "step": 23085
+    },
+    {
+      "epoch": 2.9081462354756433,
+      "grad_norm": 0.2018105536699295,
+      "learning_rate": 8.530308393027041e-07,
+      "loss": 0.1457,
+      "step": 23090
+    },
+    {
+      "epoch": 2.9087760178858204,
+      "grad_norm": 0.21647921204566956,
+      "learning_rate": 8.413645500299437e-07,
+      "loss": 0.1511,
+      "step": 23095
+    },
+    {
+      "epoch": 2.909405800295998,
+      "grad_norm": 0.19362643361091614,
+      "learning_rate": 8.297783623515741e-07,
+      "loss": 0.1491,
+      "step": 23100
+    },
+    {
+      "epoch": 2.910035582706175,
+      "grad_norm": 0.18819986283779144,
+      "learning_rate": 8.182722824896182e-07,
+      "loss": 0.1482,
+      "step": 23105
+    },
+    {
+      "epoch": 2.910665365116352,
+      "grad_norm": 0.18341930210590363,
+      "learning_rate": 8.068463166231332e-07,
+      "loss": 0.1468,
+      "step": 23110
+    },
+    {
+      "epoch": 2.9112951475265296,
+      "grad_norm": 0.17555510997772217,
+      "learning_rate": 7.955004708881107e-07,
+      "loss": 0.1538,
+      "step": 23115
+    },
+    {
+      "epoch": 2.9119249299367067,
+      "grad_norm": 0.1862919181585312,
+      "learning_rate": 7.842347513775271e-07,
+      "loss": 0.1485,
+      "step": 23120
+    },
+    {
+      "epoch": 2.9125547123468842,
+      "grad_norm": 0.23576001822948456,
+      "learning_rate": 7.730491641413262e-07,
+      "loss": 0.15,
+      "step": 23125
+    },
+    {
+      "epoch": 2.9131844947570613,
+      "grad_norm": 0.20176522433757782,
+      "learning_rate": 7.619437151864194e-07,
+      "loss": 0.1528,
+      "step": 23130
+    },
+    {
+      "epoch": 2.913814277167239,
+      "grad_norm": 0.1988651603460312,
+      "learning_rate": 7.50918410476703e-07,
+      "loss": 0.1513,
+      "step": 23135
+    },
+    {
+      "epoch": 2.914444059577416,
+      "grad_norm": 0.19462084770202637,
+      "learning_rate": 7.399732559330074e-07,
+      "loss": 0.1471,
+      "step": 23140
+    },
+    {
+      "epoch": 2.9150738419875935,
+      "grad_norm": 0.2095441371202469,
+      "learning_rate": 7.291082574331309e-07,
+      "loss": 0.1488,
+      "step": 23145
+    },
+    {
+      "epoch": 2.9157036243977705,
+      "grad_norm": 0.19712376594543457,
+      "learning_rate": 7.18323420811856e-07,
+      "loss": 0.146,
+      "step": 23150
+    },
+    {
+      "epoch": 2.916333406807948,
+      "grad_norm": 0.20228298008441925,
+      "learning_rate": 7.076187518608168e-07,
+      "loss": 0.1408,
+      "step": 23155
+    },
+    {
+      "epoch": 2.916963189218125,
+      "grad_norm": 0.21097783744335175,
+      "learning_rate": 6.969942563287311e-07,
+      "loss": 0.1463,
+      "step": 23160
+    },
+    {
+      "epoch": 2.9175929716283022,
+      "grad_norm": 0.1911788433790207,
+      "learning_rate": 6.864499399211687e-07,
+      "loss": 0.1359,
+      "step": 23165
+    },
+    {
+      "epoch": 2.9182227540384797,
+      "grad_norm": 0.19333137571811676,
+      "learning_rate": 6.759858083006831e-07,
+      "loss": 0.1521,
+      "step": 23170
+    },
+    {
+      "epoch": 2.918852536448657,
+      "grad_norm": 0.20187996327877045,
+      "learning_rate": 6.656018670867125e-07,
+      "loss": 0.1391,
+      "step": 23175
+    },
+    {
+      "epoch": 2.9194823188588344,
+      "grad_norm": 0.26705697178840637,
+      "learning_rate": 6.55298121855713e-07,
+      "loss": 0.1531,
+      "step": 23180
+    },
+    {
+      "epoch": 2.9201121012690114,
+      "grad_norm": 0.2374356985092163,
+      "learning_rate": 6.450745781410249e-07,
+      "loss": 0.1604,
+      "step": 23185
+    },
+    {
+      "epoch": 2.920741883679189,
+      "grad_norm": 0.18227587640285492,
+      "learning_rate": 6.349312414329067e-07,
+      "loss": 0.1486,
+      "step": 23190
+    },
+    {
+      "epoch": 2.921371666089366,
+      "grad_norm": 0.22778551280498505,
+      "learning_rate": 6.248681171785675e-07,
+      "loss": 0.1475,
+      "step": 23195
+    },
+    {
+      "epoch": 2.922001448499543,
+      "grad_norm": 0.2099718153476715,
+      "learning_rate": 6.148852107821511e-07,
+      "loss": 0.1442,
+      "step": 23200
+    },
+    {
+      "epoch": 2.9226312309097207,
+      "grad_norm": 0.19987498223781586,
+      "learning_rate": 6.04982527604686e-07,
+      "loss": 0.1504,
+      "step": 23205
+    },
+    {
+      "epoch": 2.923261013319898,
+      "grad_norm": 0.1993655115365982,
+      "learning_rate": 5.951600729641515e-07,
+      "loss": 0.154,
+      "step": 23210
+    },
+    {
+      "epoch": 2.9238907957300753,
+      "grad_norm": 0.19234336912631989,
+      "learning_rate": 5.854178521354113e-07,
+      "loss": 0.1531,
+      "step": 23215
+    },
+    {
+      "epoch": 2.9245205781402523,
+      "grad_norm": 0.20532859861850739,
+      "learning_rate": 5.757558703502973e-07,
+      "loss": 0.1522,
+      "step": 23220
+    },
+    {
+      "epoch": 2.92515036055043,
+      "grad_norm": 0.23982007801532745,
+      "learning_rate": 5.661741327974755e-07,
+      "loss": 0.15,
+      "step": 23225
+    },
+    {
+      "epoch": 2.925780142960607,
+      "grad_norm": 0.18457266688346863,
+      "learning_rate": 5.5667264462258e-07,
+      "loss": 0.1457,
+      "step": 23230
+    },
+    {
+      "epoch": 2.9264099253707845,
+      "grad_norm": 0.1986248642206192,
+      "learning_rate": 5.472514109281123e-07,
+      "loss": 0.145,
+      "step": 23235
+    },
+    {
+      "epoch": 2.9270397077809616,
+      "grad_norm": 0.2265506535768509,
+      "learning_rate": 5.379104367735087e-07,
+      "loss": 0.147,
+      "step": 23240
+    },
+    {
+      "epoch": 2.927669490191139,
+      "grad_norm": 0.21421676874160767,
+      "learning_rate": 5.286497271750733e-07,
+      "loss": 0.1492,
+      "step": 23245
+    },
+    {
+      "epoch": 2.928299272601316,
+      "grad_norm": 0.1608610451221466,
+      "learning_rate": 5.19469287106028e-07,
+      "loss": 0.1339,
+      "step": 23250
+    },
+    {
+      "epoch": 2.9289290550114933,
+      "grad_norm": 0.18140849471092224,
+      "learning_rate": 5.103691214964789e-07,
+      "loss": 0.1469,
+      "step": 23255
+    },
+    {
+      "epoch": 2.929558837421671,
+      "grad_norm": 0.21651338040828705,
+      "learning_rate": 5.013492352334003e-07,
+      "loss": 0.1596,
+      "step": 23260
+    },
+    {
+      "epoch": 2.9301886198318483,
+      "grad_norm": 0.18771570920944214,
+      "learning_rate": 4.924096331607008e-07,
+      "loss": 0.1482,
+      "step": 23265
+    },
+    {
+      "epoch": 2.9308184022420254,
+      "grad_norm": 0.2401566356420517,
+      "learning_rate": 4.835503200791402e-07,
+      "loss": 0.1557,
+      "step": 23270
+    },
+    {
+      "epoch": 2.9314481846522025,
+      "grad_norm": 0.22064423561096191,
+      "learning_rate": 4.747713007463627e-07,
+      "loss": 0.1621,
+      "step": 23275
+    },
+    {
+      "epoch": 2.93207796706238,
+      "grad_norm": 0.18799829483032227,
+      "learning_rate": 4.660725798769305e-07,
+      "loss": 0.1517,
+      "step": 23280
+    },
+    {
+      "epoch": 2.932707749472557,
+      "grad_norm": 0.24584966897964478,
+      "learning_rate": 4.574541621422401e-07,
+      "loss": 0.1563,
+      "step": 23285
+    },
+    {
+      "epoch": 2.9333375318827346,
+      "grad_norm": 0.18013089895248413,
+      "learning_rate": 4.489160521705726e-07,
+      "loss": 0.1388,
+      "step": 23290
+    },
+    {
+      "epoch": 2.9339673142929117,
+      "grad_norm": 0.20351989567279816,
+      "learning_rate": 4.404582545470936e-07,
+      "loss": 0.1451,
+      "step": 23295
+    },
+    {
+      "epoch": 2.934597096703089,
+      "grad_norm": 0.17512726783752441,
+      "learning_rate": 4.3208077381383655e-07,
+      "loss": 0.1413,
+      "step": 23300
+    },
+    {
+      "epoch": 2.9352268791132663,
+      "grad_norm": 0.2058653086423874,
+      "learning_rate": 4.2378361446970267e-07,
+      "loss": 0.1578,
+      "step": 23305
+    },
+    {
+      "epoch": 2.9358566615234434,
+      "grad_norm": 0.23548051714897156,
+      "learning_rate": 4.155667809704444e-07,
+      "loss": 0.154,
+      "step": 23310
+    },
+    {
+      "epoch": 2.936486443933621,
+      "grad_norm": 0.19606271386146545,
+      "learning_rate": 4.074302777286986e-07,
+      "loss": 0.1523,
+      "step": 23315
+    },
+    {
+      "epoch": 2.9371162263437984,
+      "grad_norm": 0.19439321756362915,
+      "learning_rate": 3.993741091139369e-07,
+      "loss": 0.1447,
+      "step": 23320
+    },
+    {
+      "epoch": 2.9377460087539755,
+      "grad_norm": 0.20347769558429718,
+      "learning_rate": 3.9139827945253167e-07,
+      "loss": 0.1466,
+      "step": 23325
+    },
+    {
+      "epoch": 2.9383757911641526,
+      "grad_norm": 0.23724155128002167,
+      "learning_rate": 3.835027930276735e-07,
+      "loss": 0.1597,
+      "step": 23330
+    },
+    {
+      "epoch": 2.93900557357433,
+      "grad_norm": 0.1813487559556961,
+      "learning_rate": 3.7568765407940406e-07,
+      "loss": 0.1358,
+      "step": 23335
+    },
+    {
+      "epoch": 2.939635355984507,
+      "grad_norm": 0.21012306213378906,
+      "learning_rate": 3.679528668046494e-07,
+      "loss": 0.1508,
+      "step": 23340
+    },
+    {
+      "epoch": 2.9402651383946847,
+      "grad_norm": 0.21550029516220093,
+      "learning_rate": 3.602984353571703e-07,
+      "loss": 0.1449,
+      "step": 23345
+    },
+    {
+      "epoch": 2.940894920804862,
+      "grad_norm": 0.203638955950737,
+      "learning_rate": 3.5272436384756186e-07,
+      "loss": 0.1491,
+      "step": 23350
+    },
+    {
+      "epoch": 2.9415247032150393,
+      "grad_norm": 0.21173645555973053,
+      "learning_rate": 3.452306563432872e-07,
+      "loss": 0.1553,
+      "step": 23355
+    },
+    {
+      "epoch": 2.9421544856252164,
+      "grad_norm": 0.22810731828212738,
+      "learning_rate": 3.3781731686861047e-07,
+      "loss": 0.1606,
+      "step": 23360
+    },
+    {
+      "epoch": 2.9427842680353935,
+      "grad_norm": 0.19044247269630432,
+      "learning_rate": 3.3048434940469713e-07,
+      "loss": 0.1421,
+      "step": 23365
+    },
+    {
+      "epoch": 2.943414050445571,
+      "grad_norm": 0.21632073819637299,
+      "learning_rate": 3.232317578894805e-07,
+      "loss": 0.1473,
+      "step": 23370
+    },
+    {
+      "epoch": 2.9440438328557486,
+      "grad_norm": 0.17703349888324738,
+      "learning_rate": 3.160595462178117e-07,
+      "loss": 0.1503,
+      "step": 23375
+    },
+    {
+      "epoch": 2.9446736152659256,
+      "grad_norm": 0.19983936846256256,
+      "learning_rate": 3.089677182412931e-07,
+      "loss": 0.1427,
+      "step": 23380
+    },
+    {
+      "epoch": 2.9453033976761027,
+      "grad_norm": 0.18913906812667847,
+      "learning_rate": 3.019562777684115e-07,
+      "loss": 0.1443,
+      "step": 23385
+    },
+    {
+      "epoch": 2.9459331800862802,
+      "grad_norm": 0.2024787813425064,
+      "learning_rate": 2.950252285644883e-07,
+      "loss": 0.1501,
+      "step": 23390
+    },
+    {
+      "epoch": 2.9465629624964573,
+      "grad_norm": 0.22307011485099792,
+      "learning_rate": 2.8817457435164614e-07,
+      "loss": 0.1526,
+      "step": 23395
+    },
+    {
+      "epoch": 2.947192744906635,
+      "grad_norm": 0.23350244760513306,
+      "learning_rate": 2.814043188088255e-07,
+      "loss": 0.1583,
+      "step": 23400
+    },
+    {
+      "epoch": 2.947822527316812,
+      "grad_norm": 0.18705366551876068,
+      "learning_rate": 2.7471446557181807e-07,
+      "loss": 0.1515,
+      "step": 23405
+    },
+    {
+      "epoch": 2.9484523097269895,
+      "grad_norm": 0.18902996182441711,
+      "learning_rate": 2.681050182332334e-07,
+      "loss": 0.1489,
+      "step": 23410
+    },
+    {
+      "epoch": 2.9490820921371665,
+      "grad_norm": 0.18764075636863708,
+      "learning_rate": 2.6157598034249885e-07,
+      "loss": 0.1519,
+      "step": 23415
+    },
+    {
+      "epoch": 2.9497118745473436,
+      "grad_norm": 0.20529431104660034,
+      "learning_rate": 2.5512735540584305e-07,
+      "loss": 0.1504,
+      "step": 23420
+    },
+    {
+      "epoch": 2.950341656957521,
+      "grad_norm": 0.21828597784042358,
+      "learning_rate": 2.487591468863293e-07,
+      "loss": 0.147,
+      "step": 23425
+    },
+    {
+      "epoch": 2.9509714393676987,
+      "grad_norm": 0.2224801480770111,
+      "learning_rate": 2.424713582038551e-07,
+      "loss": 0.1542,
+      "step": 23430
+    },
+    {
+      "epoch": 2.9516012217778758,
+      "grad_norm": 0.1950775682926178,
+      "learning_rate": 2.3626399273506957e-07,
+      "loss": 0.1462,
+      "step": 23435
+    },
+    {
+      "epoch": 2.952231004188053,
+      "grad_norm": 0.17124883830547333,
+      "learning_rate": 2.3013705381348946e-07,
+      "loss": 0.1374,
+      "step": 23440
+    },
+    {
+      "epoch": 2.9528607865982304,
+      "grad_norm": 0.20156201720237732,
+      "learning_rate": 2.2409054472941613e-07,
+      "loss": 0.1471,
+      "step": 23445
+    },
+    {
+      "epoch": 2.9534905690084075,
+      "grad_norm": 0.19449764490127563,
+      "learning_rate": 2.1812446872995214e-07,
+      "loss": 0.158,
+      "step": 23450
+    },
+    {
+      "epoch": 2.954120351418585,
+      "grad_norm": 0.18049843609333038,
+      "learning_rate": 2.1223882901905132e-07,
+      "loss": 0.1518,
+      "step": 23455
+    },
+    {
+      "epoch": 2.954750133828762,
+      "grad_norm": 0.19583791494369507,
+      "learning_rate": 2.06433628757402e-07,
+      "loss": 0.1556,
+      "step": 23460
+    },
+    {
+      "epoch": 2.9553799162389396,
+      "grad_norm": 0.1842242181301117,
+      "learning_rate": 2.0070887106254373e-07,
+      "loss": 0.1494,
+      "step": 23465
+    },
+    {
+      "epoch": 2.9560096986491167,
+      "grad_norm": 0.22290416061878204,
+      "learning_rate": 1.950645590088007e-07,
+      "loss": 0.1454,
+      "step": 23470
+    },
+    {
+      "epoch": 2.9566394810592938,
+      "grad_norm": 0.18742886185646057,
+      "learning_rate": 1.895006956272982e-07,
+      "loss": 0.1432,
+      "step": 23475
+    },
+    {
+      "epoch": 2.9572692634694713,
+      "grad_norm": 0.181674525141716,
+      "learning_rate": 1.8401728390594617e-07,
+      "loss": 0.1534,
+      "step": 23480
+    },
+    {
+      "epoch": 2.957899045879649,
+      "grad_norm": 0.2029273808002472,
+      "learning_rate": 1.786143267894724e-07,
+      "loss": 0.1515,
+      "step": 23485
+    },
+    {
+      "epoch": 2.958528828289826,
+      "grad_norm": 0.23291803896427155,
+      "learning_rate": 1.7329182717940594e-07,
+      "loss": 0.144,
+      "step": 23490
+    },
+    {
+      "epoch": 2.959158610700003,
+      "grad_norm": 0.24005140364170074,
+      "learning_rate": 1.6804978793401036e-07,
+      "loss": 0.1508,
+      "step": 23495
+    },
+    {
+      "epoch": 2.9597883931101805,
+      "grad_norm": 0.18763796985149384,
+      "learning_rate": 1.6288821186841716e-07,
+      "loss": 0.1477,
+      "step": 23500
+    },
+    {
+      "epoch": 2.9604181755203576,
+      "grad_norm": 0.18949641287326813,
+      "learning_rate": 1.578071017544924e-07,
+      "loss": 0.1501,
+      "step": 23505
+    },
+    {
+      "epoch": 2.961047957930535,
+      "grad_norm": 0.21106213331222534,
+      "learning_rate": 1.5280646032092003e-07,
+      "loss": 0.152,
+      "step": 23510
+    },
+    {
+      "epoch": 2.961677740340712,
+      "grad_norm": 0.231742724776268,
+      "learning_rate": 1.4788629025313526e-07,
+      "loss": 0.1546,
+      "step": 23515
+    },
+    {
+      "epoch": 2.9623075227508897,
+      "grad_norm": 0.21462422609329224,
+      "learning_rate": 1.430465941934078e-07,
+      "loss": 0.149,
+      "step": 23520
+    },
+    {
+      "epoch": 2.962937305161067,
+      "grad_norm": 0.18480440974235535,
+      "learning_rate": 1.382873747407587e-07,
+      "loss": 0.1486,
+      "step": 23525
+    },
+    {
+      "epoch": 2.963567087571244,
+      "grad_norm": 0.18864907324314117,
+      "learning_rate": 1.3360863445097682e-07,
+      "loss": 0.1556,
+      "step": 23530
+    },
+    {
+      "epoch": 2.9641968699814214,
+      "grad_norm": 0.17919088900089264,
+      "learning_rate": 1.2901037583668562e-07,
+      "loss": 0.1435,
+      "step": 23535
+    },
+    {
+      "epoch": 2.964826652391599,
+      "grad_norm": 0.22338031232357025,
+      "learning_rate": 1.2449260136722649e-07,
+      "loss": 0.1538,
+      "step": 23540
+    },
+    {
+      "epoch": 2.965456434801776,
+      "grad_norm": 0.23020599782466888,
+      "learning_rate": 1.200553134687754e-07,
+      "loss": 0.1559,
+      "step": 23545
+    },
+    {
+      "epoch": 2.966086217211953,
+      "grad_norm": 0.20723643898963928,
+      "learning_rate": 1.1569851452422618e-07,
+      "loss": 0.1415,
+      "step": 23550
+    },
+    {
+      "epoch": 2.9667159996221306,
+      "grad_norm": 0.2118474692106247,
+      "learning_rate": 1.1142220687330727e-07,
+      "loss": 0.1567,
+      "step": 23555
+    },
+    {
+      "epoch": 2.9673457820323077,
+      "grad_norm": 0.20279090106487274,
+      "learning_rate": 1.0722639281246503e-07,
+      "loss": 0.145,
+      "step": 23560
+    },
+    {
+      "epoch": 2.9679755644424852,
+      "grad_norm": 0.21842657029628754,
+      "learning_rate": 1.0311107459498035e-07,
+      "loss": 0.1557,
+      "step": 23565
+    },
+    {
+      "epoch": 2.9686053468526623,
+      "grad_norm": 0.2776351571083069,
+      "learning_rate": 9.90762544308521e-08,
+      "loss": 0.1638,
+      "step": 23570
+    },
+    {
+      "epoch": 2.96923512926284,
+      "grad_norm": 0.19698885083198547,
+      "learning_rate": 9.512193448686367e-08,
+      "loss": 0.1457,
+      "step": 23575
+    },
+    {
+      "epoch": 2.969864911673017,
+      "grad_norm": 0.1835564374923706,
+      "learning_rate": 9.124811688659972e-08,
+      "loss": 0.1569,
+      "step": 23580
+    },
+    {
+      "epoch": 2.970494694083194,
+      "grad_norm": 0.20081757009029388,
+      "learning_rate": 8.745480371036284e-08,
+      "loss": 0.1451,
+      "step": 23585
+    },
+    {
+      "epoch": 2.9711244764933715,
+      "grad_norm": 0.2095508873462677,
+      "learning_rate": 8.37419969952735e-08,
+      "loss": 0.1461,
+      "step": 23590
+    },
+    {
+      "epoch": 2.971754258903549,
+      "grad_norm": 0.24606873095035553,
+      "learning_rate": 8.010969873517015e-08,
+      "loss": 0.1555,
+      "step": 23595
+    },
+    {
+      "epoch": 2.972384041313726,
+      "grad_norm": 0.2295389175415039,
+      "learning_rate": 7.65579108806924e-08,
+      "loss": 0.1534,
+      "step": 23600
+    },
+    {
+      "epoch": 2.973013823723903,
+      "grad_norm": 0.217758446931839,
+      "learning_rate": 7.308663533924786e-08,
+      "loss": 0.1647,
+      "step": 23605
+    },
+    {
+      "epoch": 2.9736436061340807,
+      "grad_norm": 0.21592317521572113,
+      "learning_rate": 6.969587397496201e-08,
+      "loss": 0.1482,
+      "step": 23610
+    },
+    {
+      "epoch": 2.974273388544258,
+      "grad_norm": 0.2219853401184082,
+      "learning_rate": 6.638562860876162e-08,
+      "loss": 0.151,
+      "step": 23615
+    },
+    {
+      "epoch": 2.9749031709544353,
+      "grad_norm": 0.21385011076927185,
+      "learning_rate": 6.315590101832468e-08,
+      "loss": 0.1477,
+      "step": 23620
+    },
+    {
+      "epoch": 2.9755329533646124,
+      "grad_norm": 0.18287068605422974,
+      "learning_rate": 6.000669293808048e-08,
+      "loss": 0.1478,
+      "step": 23625
+    },
+    {
+      "epoch": 2.97616273577479,
+      "grad_norm": 0.23458221554756165,
+      "learning_rate": 5.693800605924281e-08,
+      "loss": 0.1521,
+      "step": 23630
+    },
+    {
+      "epoch": 2.976792518184967,
+      "grad_norm": 0.2393937110900879,
+      "learning_rate": 5.394984202976016e-08,
+      "loss": 0.1443,
+      "step": 23635
+    },
+    {
+      "epoch": 2.977422300595144,
+      "grad_norm": 0.2523866891860962,
+      "learning_rate": 5.104220245434887e-08,
+      "loss": 0.16,
+      "step": 23640
+    },
+    {
+      "epoch": 2.9780520830053216,
+      "grad_norm": 0.20749832689762115,
+      "learning_rate": 4.821508889445991e-08,
+      "loss": 0.1493,
+      "step": 23645
+    },
+    {
+      "epoch": 2.978681865415499,
+      "grad_norm": 0.20586150884628296,
+      "learning_rate": 4.546850286834547e-08,
+      "loss": 0.1425,
+      "step": 23650
+    },
+    {
+      "epoch": 2.9793116478256763,
+      "grad_norm": 0.20742247998714447,
+      "learning_rate": 4.2802445850959046e-08,
+      "loss": 0.1556,
+      "step": 23655
+    },
+    {
+      "epoch": 2.9799414302358533,
+      "grad_norm": 0.178171768784523,
+      "learning_rate": 4.0216919274038696e-08,
+      "loss": 0.1532,
+      "step": 23660
+    },
+    {
+      "epoch": 2.980571212646031,
+      "grad_norm": 0.24389080703258514,
+      "learning_rate": 3.771192452607374e-08,
+      "loss": 0.1551,
+      "step": 23665
+    },
+    {
+      "epoch": 2.981200995056208,
+      "grad_norm": 0.18905188143253326,
+      "learning_rate": 3.528746295232143e-08,
+      "loss": 0.144,
+      "step": 23670
+    },
+    {
+      "epoch": 2.9818307774663855,
+      "grad_norm": 0.258633017539978,
+      "learning_rate": 3.2943535854756956e-08,
+      "loss": 0.1624,
+      "step": 23675
+    },
+    {
+      "epoch": 2.9824605598765626,
+      "grad_norm": 0.19491221010684967,
+      "learning_rate": 3.0680144492123416e-08,
+      "loss": 0.1429,
+      "step": 23680
+    },
+    {
+      "epoch": 2.98309034228674,
+      "grad_norm": 0.19454774260520935,
+      "learning_rate": 2.8497290079898537e-08,
+      "loss": 0.1439,
+      "step": 23685
+    },
+    {
+      "epoch": 2.983720124696917,
+      "grad_norm": 0.21658724546432495,
+      "learning_rate": 2.6394973790361262e-08,
+      "loss": 0.1397,
+      "step": 23690
+    },
+    {
+      "epoch": 2.9843499071070942,
+      "grad_norm": 0.18729209899902344,
+      "learning_rate": 2.4373196752475177e-08,
+      "loss": 0.1494,
+      "step": 23695
+    },
+    {
+      "epoch": 2.9849796895172718,
+      "grad_norm": 0.19640378654003143,
+      "learning_rate": 2.243196005198844e-08,
+      "loss": 0.1497,
+      "step": 23700
+    },
+    {
+      "epoch": 2.9856094719274493,
+      "grad_norm": 0.20427659153938293,
+      "learning_rate": 2.0571264731383817e-08,
+      "loss": 0.1533,
+      "step": 23705
+    },
+    {
+      "epoch": 2.9862392543376264,
+      "grad_norm": 0.217566579580307,
+      "learning_rate": 1.8791111789911995e-08,
+      "loss": 0.1539,
+      "step": 23710
+    },
+    {
+      "epoch": 2.9868690367478035,
+      "grad_norm": 0.2069326490163803,
+      "learning_rate": 1.7091502183541606e-08,
+      "loss": 0.1481,
+      "step": 23715
+    },
+    {
+      "epoch": 2.987498819157981,
+      "grad_norm": 0.17276856303215027,
+      "learning_rate": 1.5472436825009205e-08,
+      "loss": 0.1457,
+      "step": 23720
+    },
+    {
+      "epoch": 2.988128601568158,
+      "grad_norm": 0.21325160562992096,
+      "learning_rate": 1.3933916583785954e-08,
+      "loss": 0.1453,
+      "step": 23725
+    },
+    {
+      "epoch": 2.9887583839783356,
+      "grad_norm": 0.2071818858385086,
+      "learning_rate": 1.2475942286094275e-08,
+      "loss": 0.1522,
+      "step": 23730
+    },
+    {
+      "epoch": 2.9893881663885127,
+      "grad_norm": 0.19544194638729095,
+      "learning_rate": 1.1098514714891205e-08,
+      "loss": 0.1427,
+      "step": 23735
+    },
+    {
+      "epoch": 2.99001794879869,
+      "grad_norm": 0.2268587052822113,
+      "learning_rate": 9.801634609901688e-09,
+      "loss": 0.1574,
+      "step": 23740
+    },
+    {
+      "epoch": 2.9906477312088673,
+      "grad_norm": 0.1800483763217926,
+      "learning_rate": 8.585302667585281e-09,
+      "loss": 0.1491,
+      "step": 23745
+    },
+    {
+      "epoch": 2.9912775136190444,
+      "grad_norm": 0.1946074366569519,
+      "learning_rate": 7.449519541119498e-09,
+      "loss": 0.1504,
+      "step": 23750
+    },
+    {
+      "epoch": 2.991907296029222,
+      "grad_norm": 0.19125299155712128,
+      "learning_rate": 6.394285840449764e-09,
+      "loss": 0.148,
+      "step": 23755
+    },
+    {
+      "epoch": 2.9925370784393994,
+      "grad_norm": 0.18596796691417694,
+      "learning_rate": 5.419602132272771e-09,
+      "loss": 0.1418,
+      "step": 23760
+    },
+    {
+      "epoch": 2.9931668608495765,
+      "grad_norm": 0.20286522805690765,
+      "learning_rate": 4.525468940003163e-09,
+      "loss": 0.1425,
+      "step": 23765
+    },
+    {
+      "epoch": 2.9937966432597536,
+      "grad_norm": 0.1934228241443634,
+      "learning_rate": 3.7118867438068465e-09,
+      "loss": 0.1456,
+      "step": 23770
+    },
+    {
+      "epoch": 2.994426425669931,
+      "grad_norm": 0.19864603877067566,
+      "learning_rate": 2.9788559806176447e-09,
+      "loss": 0.141,
+      "step": 23775
+    },
+    {
+      "epoch": 2.995056208080108,
+      "grad_norm": 0.1812632828950882,
+      "learning_rate": 2.326377044070682e-09,
+      "loss": 0.1414,
+      "step": 23780
+    },
+    {
+      "epoch": 2.9956859904902857,
+      "grad_norm": 0.18837498128414154,
+      "learning_rate": 1.7544502845856512e-09,
+      "loss": 0.1466,
+      "step": 23785
+    },
+    {
+      "epoch": 2.996315772900463,
+      "grad_norm": 0.19479897618293762,
+      "learning_rate": 1.2630760092668946e-09,
+      "loss": 0.1434,
+      "step": 23790
+    },
+    {
+      "epoch": 2.9969455553106403,
+      "grad_norm": 0.2550322413444519,
+      "learning_rate": 8.522544820199761e-10,
+      "loss": 0.1527,
+      "step": 23795
+    },
+    {
+      "epoch": 2.9975753377208174,
+      "grad_norm": 0.22663426399230957,
+      "learning_rate": 5.21985923451762e-10,
+      "loss": 0.154,
+      "step": 23800
+    },
+    {
+      "epoch": 2.9982051201309945,
+      "grad_norm": 0.19043132662773132,
+      "learning_rate": 2.7227051092038043e-10,
+      "loss": 0.1413,
+      "step": 23805
+    },
+    {
+      "epoch": 2.998834902541172,
+      "grad_norm": 0.17332880198955536,
+      "learning_rate": 1.0310837855187492e-10,
+      "loss": 0.137,
+      "step": 23810
+    },
+    {
+      "epoch": 2.9994646849513495,
+      "grad_norm": 0.2147851288318634,
+      "learning_rate": 1.4499617156937815e-11,
+      "loss": 0.1501,
+      "step": 23815
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 23817,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.0258751160588435e+19,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/training_args.bin b/training_args.bin
new file mode 100644
index 0000000..bc1eadc
--- /dev/null
+++ b/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62d10522f577ccde7133c9f6fc2ec8af0e0b4a33068ba484c9e2478c4fd87588
+size 7633