Model save

README.md

@@ -0,0 +1,99 @@
+---
+base_model: ondevicellm/tinyllama_moe_v2
+tags:
+- trl
+- sft
+- generated_from_trainer
+datasets:
+- generator
+model-index:
+- name: tinyllama_moe_sft_routeraux_ep3
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# tinyllama_moe_sft_routeraux_ep3
+
+This model is a fine-tuned version of [ondevicellm/tinyllama_moe_v2](https://huggingface.co/ondevicellm/tinyllama_moe_v2) on the generator dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.2954
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 16
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 4
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 128
+- total_eval_batch_size: 32
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_steps: 120
+- num_epochs: 3
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 1.72          | 0.09  | 100  | 1.6775          |
+| 1.4985        | 0.18  | 200  | 1.4900          |
+| 1.4482        | 0.26  | 300  | 1.4473          |
+| 1.4152        | 0.35  | 400  | 1.4215          |
+| 1.3777        | 0.44  | 500  | 1.4031          |
+| 1.3932        | 0.53  | 600  | 1.3886          |
+| 1.375         | 0.61  | 700  | 1.3762          |
+| 1.3574        | 0.7   | 800  | 1.3657          |
+| 1.349         | 0.79  | 900  | 1.3563          |
+| 1.3276        | 0.88  | 1000 | 1.3481          |
+| 1.3491        | 0.96  | 1100 | 1.3409          |
+| 1.2812        | 1.05  | 1200 | 1.3358          |
+| 1.2831        | 1.14  | 1300 | 1.3308          |
+| 1.2917        | 1.23  | 1400 | 1.3258          |
+| 1.2812        | 1.31  | 1500 | 1.3219          |
+| 1.2819        | 1.4   | 1600 | 1.3178          |
+| 1.2756        | 1.49  | 1700 | 1.3145          |
+| 1.2584        | 1.58  | 1800 | 1.3107          |
+| 1.2806        | 1.66  | 1900 | 1.3083          |
+| 1.2815        | 1.75  | 2000 | 1.3054          |
+| 1.2676        | 1.84  | 2100 | 1.3031          |
+| 1.2388        | 1.93  | 2200 | 1.3011          |
+| 1.2385        | 2.01  | 2300 | 1.3015          |
+| 1.2459        | 2.1   | 2400 | 1.3000          |
+| 1.2349        | 2.19  | 2500 | 1.2989          |
+| 1.2277        | 2.28  | 2600 | 1.2981          |
+| 1.2243        | 2.37  | 2700 | 1.2973          |
+| 1.2298        | 2.45  | 2800 | 1.2967          |
+| 1.2362        | 2.54  | 2900 | 1.2961          |
+| 1.216         | 2.63  | 3000 | 1.2958          |
+| 1.2381        | 2.72  | 3100 | 1.2957          |
+| 1.2274        | 2.8   | 3200 | 1.2955          |
+| 1.2235        | 2.89  | 3300 | 1.2954          |
+| 1.2438        | 2.98  | 3400 | 1.2954          |
+
+
+### Framework versions
+
+- Transformers 4.37.0
+- Pytorch 2.1.2+cu118
+- Datasets 2.16.1
+- Tokenizers 0.15.0

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 1.2954297065734863,
+    "eval_runtime": 375.8508,
+    "eval_samples": 23110,
+    "eval_samples_per_second": 43.012,
+    "eval_steps_per_second": 1.346,
+    "train_loss": 1.4016100346188847,
+    "train_runtime": 49299.9861,
+    "train_samples": 207865,
+    "train_samples_per_second": 8.888,
+    "train_steps_per_second": 0.069
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 1.2954297065734863,
+    "eval_runtime": 375.8508,
+    "eval_samples": 23110,
+    "eval_samples_per_second": 43.012,
+    "eval_steps_per_second": 1.346
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.37.0",
+  "use_cache": false
+}

model-00001-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f3328c108c33bad94cb9beb2ef6d2787196ed5d93cd17fa3a714204d3db43a33
+size 4984275728

model-00002-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cebb47e97c9ae4ccb4010489ea70bc5169fb146d678536159c3e5127fc785a79
+size 4991625072

model-00003-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d9da2e99b18f4f7ff04f0e7e941448f3f1d0549216399c101947b3134140a20
+size 2882730416

model.safetensors.index.json

@@ -0,0 +1,692 @@
+{
+  "metadata": {
+    "total_size": 12858544128
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00003-of-00003.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.0.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.0.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.1.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.1.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.2.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.2.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.2.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.3.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.3.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.3.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.4.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.4.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.4.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.5.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.5.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.5.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.6.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.6.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.6.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.7.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.7.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.experts.7.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.block_sparse_moe.gate.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.0.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.0.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.1.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.1.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.2.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.2.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.2.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.3.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.3.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.3.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.4.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.4.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.4.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.5.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.5.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.5.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.6.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.6.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.6.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.7.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.7.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.experts.7.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.block_sparse_moe.gate.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.0.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.0.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.1.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.1.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.2.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.2.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.2.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.3.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.3.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.3.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.4.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.4.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.4.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.5.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.5.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.5.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.6.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.6.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.6.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.7.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.7.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.experts.7.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.block_sparse_moe.gate.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.0.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.0.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.1.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.1.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.2.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.2.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.2.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.3.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.3.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.3.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.4.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.4.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.4.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.5.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.5.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.5.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.6.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.6.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.6.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.7.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.7.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.experts.7.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.block_sparse_moe.gate.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.0.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.0.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.1.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.1.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.2.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.2.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.2.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.3.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.3.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.3.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.4.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.4.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.4.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.5.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.5.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.5.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.6.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.6.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.6.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.7.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.7.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.experts.7.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.block_sparse_moe.gate.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.0.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.0.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.1.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.1.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.2.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.2.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.2.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.3.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.3.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.3.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.4.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.4.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.4.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.5.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.5.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.5.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.6.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.6.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.6.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.7.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.7.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.experts.7.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.block_sparse_moe.gate.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.0.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.0.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.1.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.1.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.2.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.2.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.2.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.3.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.3.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.3.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.4.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.4.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.4.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.5.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.5.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.5.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.6.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.6.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.6.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.7.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.7.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.experts.7.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.block_sparse_moe.gate.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.0.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.0.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.1.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.1.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.2.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.2.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.2.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.3.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.3.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.3.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.4.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.4.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.4.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.5.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.5.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.5.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.6.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.6.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.6.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.7.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.7.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.experts.7.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.block_sparse_moe.gate.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.0.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.0.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.1.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.1.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.2.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.2.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.2.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.3.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.3.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.3.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.4.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.4.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.4.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.5.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.5.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.5.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.6.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.6.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.6.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.7.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.7.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.experts.7.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.block_sparse_moe.gate.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.0.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.0.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.1.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.1.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.2.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.2.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.2.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.3.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.3.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.3.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.4.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.4.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.4.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.5.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.5.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.5.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.6.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.6.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.6.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.7.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.7.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.experts.7.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.17.block_sparse_moe.gate.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.0.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.0.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.1.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.1.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.2.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.2.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.2.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.3.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.3.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.3.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.4.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.4.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.4.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.5.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.5.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.5.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.6.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.6.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.6.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.7.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.7.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.experts.7.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.block_sparse_moe.gate.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.0.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.0.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.1.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.1.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.2.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.2.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.2.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.3.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.3.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.3.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.4.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.4.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.4.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.5.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.5.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.5.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.6.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.6.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.6.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.7.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.7.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.experts.7.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.block_sparse_moe.gate.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.0.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.0.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.1.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.1.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.2.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.2.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.2.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.3.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.3.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.3.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.4.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.4.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.4.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.5.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.5.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.5.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.6.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.6.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.6.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.7.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.7.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.experts.7.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.block_sparse_moe.gate.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.0.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.0.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.1.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.1.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.2.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.2.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.2.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.3.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.3.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.3.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.4.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.4.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.4.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.5.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.5.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.5.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.6.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.6.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.6.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.7.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.7.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.experts.7.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.block_sparse_moe.gate.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.0.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.0.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.1.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.1.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.2.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.2.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.2.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.3.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.3.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.3.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.4.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.4.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.4.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.5.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.5.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.5.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.6.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.6.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.6.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.7.w1.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.7.w2.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.experts.7.w3.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.block_sparse_moe.gate.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.0.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.0.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.1.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.1.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.2.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.2.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.2.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.3.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.3.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.3.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.4.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.4.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.4.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.5.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.5.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.5.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.6.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.6.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.6.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.7.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.7.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.experts.7.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.block_sparse_moe.gate.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.0.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.0.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.1.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.1.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.2.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.2.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.2.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.3.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.3.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.3.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.4.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.4.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.4.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.5.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.5.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.5.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.6.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.6.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.6.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.7.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.7.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.experts.7.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.block_sparse_moe.gate.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.0.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.0.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.1.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.1.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.2.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.2.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.2.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.3.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.3.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.3.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.4.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.4.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.4.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.5.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.5.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.5.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.6.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.6.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.6.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.7.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.7.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.experts.7.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.block_sparse_moe.gate.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.0.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.0.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.1.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.1.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.2.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.2.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.2.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.3.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.3.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.3.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.4.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.4.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.4.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.5.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.5.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.5.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.6.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.6.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.6.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.7.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.7.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.experts.7.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.block_sparse_moe.gate.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.0.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.0.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.1.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.1.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.2.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.2.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.2.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.3.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.3.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.3.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.4.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.4.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.4.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.5.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.5.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.5.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.6.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.6.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.6.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.7.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.7.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.experts.7.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.block_sparse_moe.gate.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.0.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.0.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.1.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.1.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.2.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.2.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.2.w3.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.3.w1.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.3.w2.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.3.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.4.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.4.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.4.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.5.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.5.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.5.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.6.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.6.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.6.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.7.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.7.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.experts.7.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.8.block_sparse_moe.gate.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.0.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.0.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.1.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.1.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.2.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.2.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.2.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.3.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.3.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.3.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.4.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.4.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.4.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.5.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.5.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.5.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.6.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.6.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.6.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.7.w1.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.7.w2.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.experts.7.w3.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.block_sparse_moe.gate.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.norm.weight": "model-00003-of-00003.safetensors"
+  }
+}

runs/Jan30_12-40-46_main1/events.out.tfevents.1706618574.main1.85999.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e26e7b13f231043824ff2d25f59f9ca4272948d85270e7834ca4ea65059b7d1e
-size 120742
+oid sha256:bb535340903daec8bd3e6e5df91853d006900d7ad1702cbb699b86153a3a90f2
+size 121724

runs/Jan30_12-40-46_main1/events.out.tfevents.1706668242.main1.85999.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef7ecf8ad1e9efa2307be5469282c68ab0f2533dc2e0a54ebeb0d87dc447ee77
+size 359

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "train_loss": 1.4016100346188847,
+    "train_runtime": 49299.9861,
+    "train_samples": 207865,
+    "train_samples_per_second": 8.888,
+    "train_steps_per_second": 0.069
+}

trainer_state.json

@@ -0,0 +1,4412 @@
+{
+  "best_metric": 1.2954297065734863,
+  "best_model_checkpoint": "data/tinyllama_moe_sft_routeraux_ep3/checkpoint-3400",
+  "epoch": 2.9986859395532193,
+  "eval_steps": 100,
+  "global_step": 3423,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.6666666666666668e-07,
+      "loss": 9.6447,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.333333333333333e-07,
+      "loss": 9.6853,
+      "step": 5
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6666666666666667e-06,
+      "loss": 9.6114,
+      "step": 10
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5e-06,
+      "loss": 9.3307,
+      "step": 15
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3333333333333333e-06,
+      "loss": 8.6284,
+      "step": 20
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.166666666666667e-06,
+      "loss": 7.8161,
+      "step": 25
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 5e-06,
+      "loss": 7.1309,
+      "step": 30
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 5.833333333333334e-06,
+      "loss": 6.421,
+      "step": 35
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 5.8815,
+      "step": 40
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 7.500000000000001e-06,
+      "loss": 5.218,
+      "step": 45
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 4.4455,
+      "step": 50
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.166666666666666e-06,
+      "loss": 3.5677,
+      "step": 55
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1e-05,
+      "loss": 2.903,
+      "step": 60
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.0833333333333334e-05,
+      "loss": 2.5064,
+      "step": 65
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.1666666666666668e-05,
+      "loss": 2.2887,
+      "step": 70
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.25e-05,
+      "loss": 2.093,
+      "step": 75
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.3333333333333333e-05,
+      "loss": 1.9673,
+      "step": 80
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.416666666666667e-05,
+      "loss": 1.8795,
+      "step": 85
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.5000000000000002e-05,
+      "loss": 1.8095,
+      "step": 90
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.5833333333333333e-05,
+      "loss": 1.7257,
+      "step": 95
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 1.72,
+      "step": 100
+    },
+    {
+      "epoch": 0.09,
+      "eval_loss": 1.6774533987045288,
+      "eval_runtime": 375.767,
+      "eval_samples_per_second": 43.021,
+      "eval_steps_per_second": 1.347,
+      "step": 100
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.7500000000000002e-05,
+      "loss": 1.6633,
+      "step": 105
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.8333333333333333e-05,
+      "loss": 1.6355,
+      "step": 110
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.916666666666667e-05,
+      "loss": 1.6203,
+      "step": 115
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 2e-05,
+      "loss": 1.583,
+      "step": 120
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9999886918439637e-05,
+      "loss": 1.5778,
+      "step": 125
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9999547676316034e-05,
+      "loss": 1.5454,
+      "step": 130
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.99989822813016e-05,
+      "loss": 1.5557,
+      "step": 135
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.999819074618348e-05,
+      "loss": 1.5627,
+      "step": 140
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9997173088863285e-05,
+      "loss": 1.5378,
+      "step": 145
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9995929332356666e-05,
+      "loss": 1.535,
+      "step": 150
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.999445950479281e-05,
+      "loss": 1.5257,
+      "step": 155
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9992763639413796e-05,
+      "loss": 1.5167,
+      "step": 160
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9990841774573843e-05,
+      "loss": 1.5081,
+      "step": 165
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9988693953738446e-05,
+      "loss": 1.5118,
+      "step": 170
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9986320225483396e-05,
+      "loss": 1.5187,
+      "step": 175
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9983720643493665e-05,
+      "loss": 1.4918,
+      "step": 180
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9980895266562217e-05,
+      "loss": 1.4904,
+      "step": 185
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9977844158588655e-05,
+      "loss": 1.496,
+      "step": 190
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.997456738857779e-05,
+      "loss": 1.4835,
+      "step": 195
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9971065030638076e-05,
+      "loss": 1.4985,
+      "step": 200
+    },
+    {
+      "epoch": 0.18,
+      "eval_loss": 1.490039587020874,
+      "eval_runtime": 375.9267,
+      "eval_samples_per_second": 43.003,
+      "eval_steps_per_second": 1.346,
+      "step": 200
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.996733716397993e-05,
+      "loss": 1.4911,
+      "step": 205
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.996338387291395e-05,
+      "loss": 1.4623,
+      "step": 210
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9959205246849e-05,
+      "loss": 1.4942,
+      "step": 215
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9954801380290194e-05,
+      "loss": 1.4692,
+      "step": 220
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.995017237283675e-05,
+      "loss": 1.4524,
+      "step": 225
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.994531832917974e-05,
+      "loss": 1.4736,
+      "step": 230
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.994023935909974e-05,
+      "loss": 1.4811,
+      "step": 235
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9934935577464312e-05,
+      "loss": 1.4491,
+      "step": 240
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9929407104225444e-05,
+      "loss": 1.4787,
+      "step": 245
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9923654064416813e-05,
+      "loss": 1.4516,
+      "step": 250
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.991767658815096e-05,
+      "loss": 1.462,
+      "step": 255
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9911474810616348e-05,
+      "loss": 1.4635,
+      "step": 260
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9905048872074322e-05,
+      "loss": 1.4445,
+      "step": 265
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.989839891785591e-05,
+      "loss": 1.4452,
+      "step": 270
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9891525098358553e-05,
+      "loss": 1.4585,
+      "step": 275
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9884427569042693e-05,
+      "loss": 1.4335,
+      "step": 280
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9877106490428275e-05,
+      "loss": 1.4669,
+      "step": 285
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9869562028091092e-05,
+      "loss": 1.4322,
+      "step": 290
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.986179435265906e-05,
+      "loss": 1.4525,
+      "step": 295
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9853803639808357e-05,
+      "loss": 1.4482,
+      "step": 300
+    },
+    {
+      "epoch": 0.26,
+      "eval_loss": 1.4473077058792114,
+      "eval_runtime": 375.244,
+      "eval_samples_per_second": 43.081,
+      "eval_steps_per_second": 1.348,
+      "step": 300
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.984559007025943e-05,
+      "loss": 1.4502,
+      "step": 305
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.983715382977293e-05,
+      "loss": 1.4192,
+      "step": 310
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9828495109145516e-05,
+      "loss": 1.4365,
+      "step": 315
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9819614104205504e-05,
+      "loss": 1.4295,
+      "step": 320
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9810511015808477e-05,
+      "loss": 1.4294,
+      "step": 325
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.980118604983273e-05,
+      "loss": 1.4223,
+      "step": 330
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.979163941717459e-05,
+      "loss": 1.4764,
+      "step": 335
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9781871333743695e-05,
+      "loss": 1.4252,
+      "step": 340
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9771882020458055e-05,
+      "loss": 1.4293,
+      "step": 345
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9761671703239108e-05,
+      "loss": 1.4245,
+      "step": 350
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9751240613006568e-05,
+      "loss": 1.4255,
+      "step": 355
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9740588985673226e-05,
+      "loss": 1.43,
+      "step": 360
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9729717062139616e-05,
+      "loss": 1.4251,
+      "step": 365
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9718625088288544e-05,
+      "loss": 1.4169,
+      "step": 370
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.970731331497956e-05,
+      "loss": 1.4098,
+      "step": 375
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.969578199804326e-05,
+      "loss": 1.4246,
+      "step": 380
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.96840313982755e-05,
+      "loss": 1.4155,
+      "step": 385
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.967206178143152e-05,
+      "loss": 1.4069,
+      "step": 390
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.96598734182199e-05,
+      "loss": 1.4086,
+      "step": 395
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9647466584296474e-05,
+      "loss": 1.4152,
+      "step": 400
+    },
+    {
+      "epoch": 0.35,
+      "eval_loss": 1.4215178489685059,
+      "eval_runtime": 374.8674,
+      "eval_samples_per_second": 43.125,
+      "eval_steps_per_second": 1.35,
+      "step": 400
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9634841560258063e-05,
+      "loss": 1.4346,
+      "step": 405
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9621998631636156e-05,
+      "loss": 1.429,
+      "step": 410
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.960893808889043e-05,
+      "loss": 1.4134,
+      "step": 415
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9595660227402204e-05,
+      "loss": 1.4147,
+      "step": 420
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.958216534746773e-05,
+      "loss": 1.405,
+      "step": 425
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9568453754291424e-05,
+      "loss": 1.4129,
+      "step": 430
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9554525757978958e-05,
+      "loss": 1.3981,
+      "step": 435
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9540381673530247e-05,
+      "loss": 1.414,
+      "step": 440
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.952602182083231e-05,
+      "loss": 1.4022,
+      "step": 445
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9511446524652062e-05,
+      "loss": 1.4286,
+      "step": 450
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.949665611462895e-05,
+      "loss": 1.3967,
+      "step": 455
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9481650925267506e-05,
+      "loss": 1.3952,
+      "step": 460
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.946643129592977e-05,
+      "loss": 1.4088,
+      "step": 465
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.945099757082763e-05,
+      "loss": 1.4085,
+      "step": 470
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.9435350099015028e-05,
+      "loss": 1.4024,
+      "step": 475
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.9419489234380077e-05,
+      "loss": 1.4106,
+      "step": 480
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.940341533563703e-05,
+      "loss": 1.3763,
+      "step": 485
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9387128766318205e-05,
+      "loss": 1.4099,
+      "step": 490
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9370629894765737e-05,
+      "loss": 1.3923,
+      "step": 495
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.935391909412325e-05,
+      "loss": 1.3777,
+      "step": 500
+    },
+    {
+      "epoch": 0.44,
+      "eval_loss": 1.4030784368515015,
+      "eval_runtime": 375.2903,
+      "eval_samples_per_second": 43.076,
+      "eval_steps_per_second": 1.348,
+      "step": 500
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9336996742327424e-05,
+      "loss": 1.3927,
+      "step": 505
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.931986322209946e-05,
+      "loss": 1.3959,
+      "step": 510
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.930251892093638e-05,
+      "loss": 1.3991,
+      "step": 515
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.928496423110233e-05,
+      "loss": 1.3742,
+      "step": 520
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9267199549619643e-05,
+      "loss": 1.3918,
+      "step": 525
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.92492252782599e-05,
+      "loss": 1.3829,
+      "step": 530
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9231041823534835e-05,
+      "loss": 1.3787,
+      "step": 535
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9212649596687136e-05,
+      "loss": 1.3928,
+      "step": 540
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9194049013681134e-05,
+      "loss": 1.382,
+      "step": 545
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9175240495193433e-05,
+      "loss": 1.3934,
+      "step": 550
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.915622446660335e-05,
+      "loss": 1.3944,
+      "step": 555
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9137001357983323e-05,
+      "loss": 1.4003,
+      "step": 560
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9117571604089172e-05,
+      "loss": 1.3985,
+      "step": 565
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.9097935644350284e-05,
+      "loss": 1.3874,
+      "step": 570
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.9078093922859642e-05,
+      "loss": 1.3886,
+      "step": 575
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.9058046888363814e-05,
+      "loss": 1.3737,
+      "step": 580
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.9037794994252792e-05,
+      "loss": 1.3929,
+      "step": 585
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.901733869854973e-05,
+      "loss": 1.3811,
+      "step": 590
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.8996678463900596e-05,
+      "loss": 1.4045,
+      "step": 595
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.8975814757563707e-05,
+      "loss": 1.3932,
+      "step": 600
+    },
+    {
+      "epoch": 0.53,
+      "eval_loss": 1.388567328453064,
+      "eval_runtime": 375.3941,
+      "eval_samples_per_second": 43.064,
+      "eval_steps_per_second": 1.348,
+      "step": 600
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.8954748051399153e-05,
+      "loss": 1.363,
+      "step": 605
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.893347882185814e-05,
+      "loss": 1.3755,
+      "step": 610
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.891200754997219e-05,
+      "loss": 1.3963,
+      "step": 615
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.88903347213423e-05,
+      "loss": 1.3708,
+      "step": 620
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.886846082612792e-05,
+      "loss": 1.3761,
+      "step": 625
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8846386359035892e-05,
+      "loss": 1.3752,
+      "step": 630
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.8824111819309256e-05,
+      "loss": 1.3971,
+      "step": 635
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.8801637710715945e-05,
+      "loss": 1.3834,
+      "step": 640
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.8778964541537422e-05,
+      "loss": 1.3829,
+      "step": 645
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.8756092824557148e-05,
+      "loss": 1.371,
+      "step": 650
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.873302307704902e-05,
+      "loss": 1.378,
+      "step": 655
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.870975582076564e-05,
+      "loss": 1.3838,
+      "step": 660
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.8686291581926546e-05,
+      "loss": 1.3708,
+      "step": 665
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.8662630891206276e-05,
+      "loss": 1.3608,
+      "step": 670
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.86387742837224e-05,
+      "loss": 1.3831,
+      "step": 675
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.86147222990234e-05,
+      "loss": 1.3919,
+      "step": 680
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.8590475481076468e-05,
+      "loss": 1.3711,
+      "step": 685
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.8566034378255198e-05,
+      "loss": 1.3721,
+      "step": 690
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.8541399543327206e-05,
+      "loss": 1.3644,
+      "step": 695
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.8516571533441606e-05,
+      "loss": 1.375,
+      "step": 700
+    },
+    {
+      "epoch": 0.61,
+      "eval_loss": 1.3762472867965698,
+      "eval_runtime": 374.3925,
+      "eval_samples_per_second": 43.179,
+      "eval_steps_per_second": 1.352,
+      "step": 700
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.8491550910116415e-05,
+      "loss": 1.374,
+      "step": 705
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.8466338239225862e-05,
+      "loss": 1.3448,
+      "step": 710
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.8440934090987576e-05,
+      "loss": 1.3538,
+      "step": 715
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.8415339039949702e-05,
+      "loss": 1.3836,
+      "step": 720
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8389553664977905e-05,
+      "loss": 1.3501,
+      "step": 725
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8363578549242266e-05,
+      "loss": 1.3926,
+      "step": 730
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8337414280204116e-05,
+      "loss": 1.3764,
+      "step": 735
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.8311061449602725e-05,
+      "loss": 1.3672,
+      "step": 740
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.8284520653441936e-05,
+      "loss": 1.3794,
+      "step": 745
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.8257792491976676e-05,
+      "loss": 1.3628,
+      "step": 750
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.8230877569699387e-05,
+      "loss": 1.3773,
+      "step": 755
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8203776495326346e-05,
+      "loss": 1.3637,
+      "step": 760
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8176489881783915e-05,
+      "loss": 1.3485,
+      "step": 765
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8149018346194655e-05,
+      "loss": 1.3489,
+      "step": 770
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8121362509863397e-05,
+      "loss": 1.3558,
+      "step": 775
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8093522998263154e-05,
+      "loss": 1.3462,
+      "step": 780
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.8065500441021018e-05,
+      "loss": 1.3679,
+      "step": 785
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.803729547190389e-05,
+      "loss": 1.3412,
+      "step": 790
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.800890872880414e-05,
+      "loss": 1.3693,
+      "step": 795
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.7980340853725223e-05,
+      "loss": 1.3574,
+      "step": 800
+    },
+    {
+      "epoch": 0.7,
+      "eval_loss": 1.3657442331314087,
+      "eval_runtime": 374.2554,
+      "eval_samples_per_second": 43.195,
+      "eval_steps_per_second": 1.352,
+      "step": 800
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.795159249276711e-05,
+      "loss": 1.3407,
+      "step": 805
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.792266429611171e-05,
+      "loss": 1.3456,
+      "step": 810
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.7893556918008136e-05,
+      "loss": 1.3439,
+      "step": 815
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.7864271016757942e-05,
+      "loss": 1.3365,
+      "step": 820
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.7834807254700212e-05,
+      "loss": 1.3582,
+      "step": 825
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.7805166298196577e-05,
+      "loss": 1.354,
+      "step": 830
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.7775348817616164e-05,
+      "loss": 1.3784,
+      "step": 835
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7745355487320418e-05,
+      "loss": 1.3576,
+      "step": 840
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7715186985647857e-05,
+      "loss": 1.3698,
+      "step": 845
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.768484399489873e-05,
+      "loss": 1.3471,
+      "step": 850
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.7654327201319584e-05,
+      "loss": 1.3601,
+      "step": 855
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.762363729508775e-05,
+      "loss": 1.3537,
+      "step": 860
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.7592774970295714e-05,
+      "loss": 1.3701,
+      "step": 865
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.7561740924935456e-05,
+      "loss": 1.3599,
+      "step": 870
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.753053586088263e-05,
+      "loss": 1.3416,
+      "step": 875
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7499160483880694e-05,
+      "loss": 1.3558,
+      "step": 880
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.7467615503524973e-05,
+      "loss": 1.3402,
+      "step": 885
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.7435901633246585e-05,
+      "loss": 1.3397,
+      "step": 890
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.740401959029632e-05,
+      "loss": 1.3444,
+      "step": 895
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.7371970095728408e-05,
+      "loss": 1.349,
+      "step": 900
+    },
+    {
+      "epoch": 0.79,
+      "eval_loss": 1.3563435077667236,
+      "eval_runtime": 375.3168,
+      "eval_samples_per_second": 43.073,
+      "eval_steps_per_second": 1.348,
+      "step": 900
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.7339753874384218e-05,
+      "loss": 1.3553,
+      "step": 905
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.730737165487587e-05,
+      "loss": 1.3473,
+      "step": 910
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.7274824169569747e-05,
+      "loss": 1.3648,
+      "step": 915
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.7242112154569928e-05,
+      "loss": 1.339,
+      "step": 920
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.7209236349701553e-05,
+      "loss": 1.3486,
+      "step": 925
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.717619749849409e-05,
+      "loss": 1.3392,
+      "step": 930
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.7142996348164508e-05,
+      "loss": 1.3633,
+      "step": 935
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.710963364960038e-05,
+      "loss": 1.3542,
+      "step": 940
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.707611015734291e-05,
+      "loss": 1.3529,
+      "step": 945
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.704242662956986e-05,
+      "loss": 1.3645,
+      "step": 950
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.700858382807841e-05,
+      "loss": 1.3603,
+      "step": 955
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.6974582518267913e-05,
+      "loss": 1.3414,
+      "step": 960
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.694042346912261e-05,
+      "loss": 1.3347,
+      "step": 965
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.6906107453194207e-05,
+      "loss": 1.3453,
+      "step": 970
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.687163524658444e-05,
+      "loss": 1.3592,
+      "step": 975
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.6837007628927483e-05,
+      "loss": 1.3536,
+      "step": 980
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.680222538337235e-05,
+      "loss": 1.3368,
+      "step": 985
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.6767289296565155e-05,
+      "loss": 1.341,
+      "step": 990
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.6732200158631343e-05,
+      "loss": 1.3561,
+      "step": 995
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.6696958763157808e-05,
+      "loss": 1.3276,
+      "step": 1000
+    },
+    {
+      "epoch": 0.88,
+      "eval_loss": 1.3481124639511108,
+      "eval_runtime": 383.9537,
+      "eval_samples_per_second": 42.104,
+      "eval_steps_per_second": 1.318,
+      "step": 1000
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.666156590717495e-05,
+      "loss": 1.3295,
+      "step": 1005
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.6626022391138643e-05,
+      "loss": 1.3362,
+      "step": 1010
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.6590329018912134e-05,
+      "loss": 1.3302,
+      "step": 1015
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.655448659774787e-05,
+      "loss": 1.3257,
+      "step": 1020
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.6518495938269242e-05,
+      "loss": 1.3471,
+      "step": 1025
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.6482357854452223e-05,
+      "loss": 1.3254,
+      "step": 1030
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.6446073163607e-05,
+      "loss": 1.3365,
+      "step": 1035
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.6409642686359472e-05,
+      "loss": 1.3467,
+      "step": 1040
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.637306724663267e-05,
+      "loss": 1.3442,
+      "step": 1045
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.6336347671628162e-05,
+      "loss": 1.3343,
+      "step": 1050
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.629948479180731e-05,
+      "loss": 1.33,
+      "step": 1055
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.6262479440872505e-05,
+      "loss": 1.3753,
+      "step": 1060
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.622533245574832e-05,
+      "loss": 1.3425,
+      "step": 1065
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.618804467656256e-05,
+      "loss": 1.3339,
+      "step": 1070
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.6150616946627272e-05,
+      "loss": 1.3271,
+      "step": 1075
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.6113050112419683e-05,
+      "loss": 1.3349,
+      "step": 1080
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.6075345023563035e-05,
+      "loss": 1.3227,
+      "step": 1085
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.6037502532807382e-05,
+      "loss": 1.3191,
+      "step": 1090
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.599952349601031e-05,
+      "loss": 1.3371,
+      "step": 1095
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.5961408772117567e-05,
+      "loss": 1.3491,
+      "step": 1100
+    },
+    {
+      "epoch": 0.96,
+      "eval_loss": 1.3409212827682495,
+      "eval_runtime": 375.3257,
+      "eval_samples_per_second": 43.072,
+      "eval_steps_per_second": 1.348,
+      "step": 1100
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.592315922314364e-05,
+      "loss": 1.3436,
+      "step": 1105
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.588477571415226e-05,
+      "loss": 1.3165,
+      "step": 1110
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.5846259113236855e-05,
+      "loss": 1.3112,
+      "step": 1115
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.580761029150089e-05,
+      "loss": 1.3291,
+      "step": 1120
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.5768830123038172e-05,
+      "loss": 1.3204,
+      "step": 1125
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.57299194849131e-05,
+      "loss": 1.3414,
+      "step": 1130
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.5690879257140804e-05,
+      "loss": 1.3262,
+      "step": 1135
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.5651710322667262e-05,
+      "loss": 1.3397,
+      "step": 1140
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.5612413567349314e-05,
+      "loss": 1.3172,
+      "step": 1145
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.557298987993465e-05,
+      "loss": 1.2644,
+      "step": 1150
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.553344015204168e-05,
+      "loss": 1.296,
+      "step": 1155
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5493765278139397e-05,
+      "loss": 1.304,
+      "step": 1160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5453966155527133e-05,
+      "loss": 1.2875,
+      "step": 1165
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.541404368431426e-05,
+      "loss": 1.3043,
+      "step": 1170
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.537399876739985e-05,
+      "loss": 1.2975,
+      "step": 1175
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.5333832310452232e-05,
+      "loss": 1.2996,
+      "step": 1180
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.5293545221888542e-05,
+      "loss": 1.2983,
+      "step": 1185
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.525313841285414e-05,
+      "loss": 1.2958,
+      "step": 1190
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.5212612797202033e-05,
+      "loss": 1.3077,
+      "step": 1195
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.517196929147219e-05,
+      "loss": 1.2812,
+      "step": 1200
+    },
+    {
+      "epoch": 1.05,
+      "eval_loss": 1.3358310461044312,
+      "eval_runtime": 375.231,
+      "eval_samples_per_second": 43.083,
+      "eval_steps_per_second": 1.349,
+      "step": 1200
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.5131208814870822e-05,
+      "loss": 1.3096,
+      "step": 1205
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.5090332289249586e-05,
+      "loss": 1.2738,
+      "step": 1210
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.5049340639084742e-05,
+      "loss": 1.2954,
+      "step": 1215
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.5008234791456242e-05,
+      "loss": 1.2903,
+      "step": 1220
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.4967015676026768e-05,
+      "loss": 1.2997,
+      "step": 1225
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4925684225020694e-05,
+      "loss": 1.2809,
+      "step": 1230
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4884241373203014e-05,
+      "loss": 1.2796,
+      "step": 1235
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.4842688057858203e-05,
+      "loss": 1.2922,
+      "step": 1240
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.4801025218769001e-05,
+      "loss": 1.2904,
+      "step": 1245
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4759253798195183e-05,
+      "loss": 1.2916,
+      "step": 1250
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4717374740852236e-05,
+      "loss": 1.3073,
+      "step": 1255
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.467538899388998e-05,
+      "loss": 1.2965,
+      "step": 1260
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.463329750687118e-05,
+      "loss": 1.2792,
+      "step": 1265
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.459110123175004e-05,
+      "loss": 1.2711,
+      "step": 1270
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.4548801122850682e-05,
+      "loss": 1.287,
+      "step": 1275
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.450639813684558e-05,
+      "loss": 1.3018,
+      "step": 1280
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.4463893232733886e-05,
+      "loss": 1.2919,
+      "step": 1285
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.4421287371819781e-05,
+      "loss": 1.2953,
+      "step": 1290
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.4378581517690711e-05,
+      "loss": 1.2797,
+      "step": 1295
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4335776636195605e-05,
+      "loss": 1.2831,
+      "step": 1300
+    },
+    {
+      "epoch": 1.14,
+      "eval_loss": 1.3308433294296265,
+      "eval_runtime": 375.7621,
+      "eval_samples_per_second": 43.022,
+      "eval_steps_per_second": 1.347,
+      "step": 1300
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4292873695423012e-05,
+      "loss": 1.2892,
+      "step": 1305
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.4249873665679241e-05,
+      "loss": 1.2763,
+      "step": 1310
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.4206777519466375e-05,
+      "loss": 1.2928,
+      "step": 1315
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4163586231460307e-05,
+      "loss": 1.278,
+      "step": 1320
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4120300778488687e-05,
+      "loss": 1.2904,
+      "step": 1325
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.4076922139508828e-05,
+      "loss": 1.2751,
+      "step": 1330
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.4033451295585565e-05,
+      "loss": 1.2837,
+      "step": 1335
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.3989889229869071e-05,
+      "loss": 1.2852,
+      "step": 1340
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.394623692757262e-05,
+      "loss": 1.3134,
+      "step": 1345
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.3902495375950303e-05,
+      "loss": 1.2686,
+      "step": 1350
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.3858665564274699e-05,
+      "loss": 1.2935,
+      "step": 1355
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.3814748483814511e-05,
+      "loss": 1.2701,
+      "step": 1360
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3770745127812134e-05,
+      "loss": 1.2976,
+      "step": 1365
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3726656491461196e-05,
+      "loss": 1.2847,
+      "step": 1370
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3682483571884064e-05,
+      "loss": 1.2881,
+      "step": 1375
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.3638227368109268e-05,
+      "loss": 1.2876,
+      "step": 1380
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.3593888881048922e-05,
+      "loss": 1.281,
+      "step": 1385
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.3549469113476087e-05,
+      "loss": 1.2825,
+      "step": 1390
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.3504969070002091e-05,
+      "loss": 1.2827,
+      "step": 1395
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.3460389757053802e-05,
+      "loss": 1.2917,
+      "step": 1400
+    },
+    {
+      "epoch": 1.23,
+      "eval_loss": 1.3258081674575806,
+      "eval_runtime": 375.9298,
+      "eval_samples_per_second": 43.003,
+      "eval_steps_per_second": 1.346,
+      "step": 1400
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.3415732182850873e-05,
+      "loss": 1.2798,
+      "step": 1405
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.3370997357382943e-05,
+      "loss": 1.3174,
+      "step": 1410
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.3326186292386778e-05,
+      "loss": 1.2884,
+      "step": 1415
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.3281300001323416e-05,
+      "loss": 1.2908,
+      "step": 1420
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3236339499355217e-05,
+      "loss": 1.2922,
+      "step": 1425
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3191305803322929e-05,
+      "loss": 1.273,
+      "step": 1430
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.3146199931722674e-05,
+      "loss": 1.294,
+      "step": 1435
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.3101022904682918e-05,
+      "loss": 1.274,
+      "step": 1440
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.3055775743941409e-05,
+      "loss": 1.2893,
+      "step": 1445
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.3010459472822046e-05,
+      "loss": 1.266,
+      "step": 1450
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.2965075116211769e-05,
+      "loss": 1.2824,
+      "step": 1455
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.2919623700537342e-05,
+      "loss": 1.2908,
+      "step": 1460
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.287410625374217e-05,
+      "loss": 1.2824,
+      "step": 1465
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.282852380526303e-05,
+      "loss": 1.2735,
+      "step": 1470
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.2782877386006807e-05,
+      "loss": 1.2876,
+      "step": 1475
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.2737168028327163e-05,
+      "loss": 1.2694,
+      "step": 1480
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.2691396766001192e-05,
+      "loss": 1.2618,
+      "step": 1485
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.2645564634206054e-05,
+      "loss": 1.2735,
+      "step": 1490
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.2599672669495537e-05,
+      "loss": 1.2879,
+      "step": 1495
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.2553721909776644e-05,
+      "loss": 1.2812,
+      "step": 1500
+    },
+    {
+      "epoch": 1.31,
+      "eval_loss": 1.3218797445297241,
+      "eval_runtime": 376.8672,
+      "eval_samples_per_second": 42.896,
+      "eval_steps_per_second": 1.343,
+      "step": 1500
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.2507713394286088e-05,
+      "loss": 1.273,
+      "step": 1505
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.246164816356682e-05,
+      "loss": 1.2909,
+      "step": 1510
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2415527259444471e-05,
+      "loss": 1.2921,
+      "step": 1515
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2369351725003802e-05,
+      "loss": 1.3026,
+      "step": 1520
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.232312260456511e-05,
+      "loss": 1.2781,
+      "step": 1525
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2276840943660613e-05,
+      "loss": 1.2828,
+      "step": 1530
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2230507789010792e-05,
+      "loss": 1.2839,
+      "step": 1535
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.2184124188500735e-05,
+      "loss": 1.2748,
+      "step": 1540
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.2137691191156425e-05,
+      "loss": 1.2658,
+      "step": 1545
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.209120984712102e-05,
+      "loss": 1.2768,
+      "step": 1550
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.2044681207631104e-05,
+      "loss": 1.2783,
+      "step": 1555
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1998106324992906e-05,
+      "loss": 1.2956,
+      "step": 1560
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1951486252558508e-05,
+      "loss": 1.2782,
+      "step": 1565
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1904822044702017e-05,
+      "loss": 1.2811,
+      "step": 1570
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1858114756795718e-05,
+      "loss": 1.2914,
+      "step": 1575
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1811365445186213e-05,
+      "loss": 1.2897,
+      "step": 1580
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1764575167170525e-05,
+      "loss": 1.2713,
+      "step": 1585
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1717744980972178e-05,
+      "loss": 1.2905,
+      "step": 1590
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.1670875945717282e-05,
+      "loss": 1.2806,
+      "step": 1595
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.1623969121410563e-05,
+      "loss": 1.2819,
+      "step": 1600
+    },
+    {
+      "epoch": 1.4,
+      "eval_loss": 1.3177708387374878,
+      "eval_runtime": 376.3708,
+      "eval_samples_per_second": 42.952,
+      "eval_steps_per_second": 1.344,
+      "step": 1600
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.1577025568911395e-05,
+      "loss": 1.2868,
+      "step": 1605
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.1530046349909816e-05,
+      "loss": 1.2736,
+      "step": 1610
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.1483032526902502e-05,
+      "loss": 1.2815,
+      "step": 1615
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1435985163168745e-05,
+      "loss": 1.2666,
+      "step": 1620
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1388905322746406e-05,
+      "loss": 1.286,
+      "step": 1625
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.1341794070407847e-05,
+      "loss": 1.2794,
+      "step": 1630
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.1294652471635857e-05,
+      "loss": 1.2733,
+      "step": 1635
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.124748159259954e-05,
+      "loss": 1.2733,
+      "step": 1640
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.1200282500130222e-05,
+      "loss": 1.2743,
+      "step": 1645
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1153056261697303e-05,
+      "loss": 1.2795,
+      "step": 1650
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1105803945384134e-05,
+      "loss": 1.2673,
+      "step": 1655
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1058526619863846e-05,
+      "loss": 1.2712,
+      "step": 1660
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.1011225354375184e-05,
+      "loss": 1.2926,
+      "step": 1665
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.0963901218698331e-05,
+      "loss": 1.2763,
+      "step": 1670
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0916555283130714e-05,
+      "loss": 1.2727,
+      "step": 1675
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0869188618462778e-05,
+      "loss": 1.2832,
+      "step": 1680
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0821802295953795e-05,
+      "loss": 1.2569,
+      "step": 1685
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0774397387307628e-05,
+      "loss": 1.273,
+      "step": 1690
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0726974964648478e-05,
+      "loss": 1.2612,
+      "step": 1695
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.0679536100496661e-05,
+      "loss": 1.2756,
+      "step": 1700
+    },
+    {
+      "epoch": 1.49,
+      "eval_loss": 1.3144780397415161,
+      "eval_runtime": 375.995,
+      "eval_samples_per_second": 42.995,
+      "eval_steps_per_second": 1.346,
+      "step": 1700
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.063208186774433e-05,
+      "loss": 1.2714,
+      "step": 1705
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.0584613339631222e-05,
+      "loss": 1.28,
+      "step": 1710
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.0537131589720387e-05,
+      "loss": 1.2627,
+      "step": 1715
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.0489637691873889e-05,
+      "loss": 1.2909,
+      "step": 1720
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.0442132720228551e-05,
+      "loss": 1.285,
+      "step": 1725
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.0394617749171636e-05,
+      "loss": 1.2801,
+      "step": 1730
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.0347093853316555e-05,
+      "loss": 1.262,
+      "step": 1735
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.0299562107478569e-05,
+      "loss": 1.2623,
+      "step": 1740
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.0252023586650476e-05,
+      "loss": 1.2878,
+      "step": 1745
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.0204479365978298e-05,
+      "loss": 1.2575,
+      "step": 1750
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.0156930520736965e-05,
+      "loss": 1.2774,
+      "step": 1755
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.0109378126306002e-05,
+      "loss": 1.2881,
+      "step": 1760
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.00618232581452e-05,
+      "loss": 1.2763,
+      "step": 1765
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.0014266991770299e-05,
+      "loss": 1.2606,
+      "step": 1770
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.966710402728658e-06,
+      "loss": 1.2788,
+      "step": 1775
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.919154566574942e-06,
+      "loss": 1.2846,
+      "step": 1780
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.871600558846772e-06,
+      "loss": 1.2801,
+      "step": 1785
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.82404945504044e-06,
+      "loss": 1.2655,
+      "step": 1790
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.776502330586535e-06,
+      "loss": 1.2716,
+      "step": 1795
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.728960260825675e-06,
+      "loss": 1.2584,
+      "step": 1800
+    },
+    {
+      "epoch": 1.58,
+      "eval_loss": 1.3107465505599976,
+      "eval_runtime": 376.4867,
+      "eval_samples_per_second": 42.939,
+      "eval_steps_per_second": 1.344,
+      "step": 1800
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.681424320984136e-06,
+      "loss": 1.2737,
+      "step": 1805
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.633895586149575e-06,
+      "loss": 1.2553,
+      "step": 1810
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.586375131246688e-06,
+      "loss": 1.2596,
+      "step": 1815
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.538864031012913e-06,
+      "loss": 1.2914,
+      "step": 1820
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.491363359974121e-06,
+      "loss": 1.264,
+      "step": 1825
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.443874192420312e-06,
+      "loss": 1.2551,
+      "step": 1830
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.396397602381318e-06,
+      "loss": 1.2898,
+      "step": 1835
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.34893466360252e-06,
+      "loss": 1.2662,
+      "step": 1840
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.301486449520543e-06,
+      "loss": 1.2819,
+      "step": 1845
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.254054033239017e-06,
+      "loss": 1.2508,
+      "step": 1850
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.206638487504265e-06,
+      "loss": 1.2616,
+      "step": 1855
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.15924088468106e-06,
+      "loss": 1.2614,
+      "step": 1860
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.11186229672839e-06,
+      "loss": 1.2694,
+      "step": 1865
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.064503795175175e-06,
+      "loss": 1.2706,
+      "step": 1870
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.017166451096077e-06,
+      "loss": 1.2586,
+      "step": 1875
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.969851335087233e-06,
+      "loss": 1.2717,
+      "step": 1880
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.922559517242078e-06,
+      "loss": 1.2515,
+      "step": 1885
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.87529206712712e-06,
+      "loss": 1.2684,
+      "step": 1890
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.828050053757764e-06,
+      "loss": 1.2552,
+      "step": 1895
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.780834545574122e-06,
+      "loss": 1.2806,
+      "step": 1900
+    },
+    {
+      "epoch": 1.66,
+      "eval_loss": 1.3082817792892456,
+      "eval_runtime": 377.0389,
+      "eval_samples_per_second": 42.876,
+      "eval_steps_per_second": 1.342,
+      "step": 1900
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.73364661041687e-06,
+      "loss": 1.2815,
+      "step": 1905
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.686487315503066e-06,
+      "loss": 1.2433,
+      "step": 1910
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.63935772740205e-06,
+      "loss": 1.2517,
+      "step": 1915
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.59225891201129e-06,
+      "loss": 1.2571,
+      "step": 1920
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.545191934532294e-06,
+      "loss": 1.2717,
+      "step": 1925
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.498157859446512e-06,
+      "loss": 1.2492,
+      "step": 1930
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.451157750491265e-06,
+      "loss": 1.2748,
+      "step": 1935
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.404192670635683e-06,
+      "loss": 1.2949,
+      "step": 1940
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.35726368205667e-06,
+      "loss": 1.268,
+      "step": 1945
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.310371846114875e-06,
+      "loss": 1.2537,
+      "step": 1950
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.263518223330698e-06,
+      "loss": 1.2701,
+      "step": 1955
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.216703873360292e-06,
+      "loss": 1.2551,
+      "step": 1960
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.169929854971598e-06,
+      "loss": 1.276,
+      "step": 1965
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.123197226020426e-06,
+      "loss": 1.2448,
+      "step": 1970
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.076507043426482e-06,
+      "loss": 1.2855,
+      "step": 1975
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.02986036314952e-06,
+      "loss": 1.2564,
+      "step": 1980
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.983258240165406e-06,
+      "loss": 1.2744,
+      "step": 1985
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.936701728442308e-06,
+      "loss": 1.2761,
+      "step": 1990
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.890191880916813e-06,
+      "loss": 1.2697,
+      "step": 1995
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.84372974947016e-06,
+      "loss": 1.2815,
+      "step": 2000
+    },
+    {
+      "epoch": 1.75,
+      "eval_loss": 1.3053697347640991,
+      "eval_runtime": 376.1669,
+      "eval_samples_per_second": 42.976,
+      "eval_steps_per_second": 1.345,
+      "step": 2000
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.797316384904402e-06,
+      "loss": 1.2641,
+      "step": 2005
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.750952836918679e-06,
+      "loss": 1.2659,
+      "step": 2010
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.704640154085466e-06,
+      "loss": 1.2701,
+      "step": 2015
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.658379383826841e-06,
+      "loss": 1.2744,
+      "step": 2020
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.612171572390834e-06,
+      "loss": 1.2715,
+      "step": 2025
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.566017764827717e-06,
+      "loss": 1.2725,
+      "step": 2030
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.519919004966414e-06,
+      "loss": 1.2696,
+      "step": 2035
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.473876335390857e-06,
+      "loss": 1.2489,
+      "step": 2040
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.427890797416435e-06,
+      "loss": 1.2658,
+      "step": 2045
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.3819634310664224e-06,
+      "loss": 1.2506,
+      "step": 2050
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.336095275048474e-06,
+      "loss": 1.2709,
+      "step": 2055
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.29028736673111e-06,
+      "loss": 1.271,
+      "step": 2060
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.244540742120294e-06,
+      "loss": 1.241,
+      "step": 2065
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.1988564358359566e-06,
+      "loss": 1.2626,
+      "step": 2070
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.153235481088624e-06,
+      "loss": 1.2809,
+      "step": 2075
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.107678909656052e-06,
+      "loss": 1.2753,
+      "step": 2080
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 7.062187751859868e-06,
+      "loss": 1.2772,
+      "step": 2085
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 7.016763036542305e-06,
+      "loss": 1.2668,
+      "step": 2090
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.971405791042889e-06,
+      "loss": 1.2583,
+      "step": 2095
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.92611704117525e-06,
+      "loss": 1.2676,
+      "step": 2100
+    },
+    {
+      "epoch": 1.84,
+      "eval_loss": 1.3030564785003662,
+      "eval_runtime": 376.0505,
+      "eval_samples_per_second": 42.989,
+      "eval_steps_per_second": 1.346,
+      "step": 2100
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.880897811203877e-06,
+      "loss": 1.2739,
+      "step": 2105
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.835749123820997e-06,
+      "loss": 1.2599,
+      "step": 2110
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.790672000123405e-06,
+      "loss": 1.2895,
+      "step": 2115
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.7456674595894065e-06,
+      "loss": 1.2558,
+      "step": 2120
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.700736520055725e-06,
+      "loss": 1.2508,
+      "step": 2125
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.6558801976945206e-06,
+      "loss": 1.2658,
+      "step": 2130
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.611099506990372e-06,
+      "loss": 1.2708,
+      "step": 2135
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.566395460717356e-06,
+      "loss": 1.2627,
+      "step": 2140
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.521769069916136e-06,
+      "loss": 1.2523,
+      "step": 2145
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.477221343871088e-06,
+      "loss": 1.2676,
+      "step": 2150
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.4327532900874945e-06,
+      "loss": 1.2726,
+      "step": 2155
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.38836591426873e-06,
+      "loss": 1.2476,
+      "step": 2160
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.344060220293542e-06,
+      "loss": 1.2665,
+      "step": 2165
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.299837210193331e-06,
+      "loss": 1.2599,
+      "step": 2170
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.255697884129495e-06,
+      "loss": 1.2839,
+      "step": 2175
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.2116432403708015e-06,
+      "loss": 1.2707,
+      "step": 2180
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.167674275270832e-06,
+      "loss": 1.2695,
+      "step": 2185
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.123791983245411e-06,
+      "loss": 1.2592,
+      "step": 2190
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.0799973567501616e-06,
+      "loss": 1.2706,
+      "step": 2195
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 6.036291386258013e-06,
+      "loss": 1.2388,
+      "step": 2200
+    },
+    {
+      "epoch": 1.93,
+      "eval_loss": 1.301101803779602,
+      "eval_runtime": 375.7618,
+      "eval_samples_per_second": 43.022,
+      "eval_steps_per_second": 1.347,
+      "step": 2200
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.992675060236841e-06,
+      "loss": 1.2828,
+      "step": 2205
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.94914936512708e-06,
+      "loss": 1.2572,
+      "step": 2210
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.905715285319442e-06,
+      "loss": 1.2661,
+      "step": 2215
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.862373803132625e-06,
+      "loss": 1.2767,
+      "step": 2220
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.819125898791115e-06,
+      "loss": 1.2591,
+      "step": 2225
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.775972550403015e-06,
+      "loss": 1.2659,
+      "step": 2230
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.732914733937917e-06,
+      "loss": 1.2611,
+      "step": 2235
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.6899534232048395e-06,
+      "loss": 1.2865,
+      "step": 2240
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.647089589830186e-06,
+      "loss": 1.2694,
+      "step": 2245
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.604324203235798e-06,
+      "loss": 1.262,
+      "step": 2250
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.561658230616997e-06,
+      "loss": 1.2754,
+      "step": 2255
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.519092636920741e-06,
+      "loss": 1.2574,
+      "step": 2260
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.476628384823773e-06,
+      "loss": 1.2829,
+      "step": 2265
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.434266434710879e-06,
+      "loss": 1.2654,
+      "step": 2270
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.392007744653134e-06,
+      "loss": 1.2549,
+      "step": 2275
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 5.3498532703862685e-06,
+      "loss": 1.2573,
+      "step": 2280
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 5.307803965289023e-06,
+      "loss": 1.2533,
+      "step": 2285
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 5.265860780361602e-06,
+      "loss": 1.2174,
+      "step": 2290
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 5.2240246642041705e-06,
+      "loss": 1.2181,
+      "step": 2295
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 5.182296562995383e-06,
+      "loss": 1.2385,
+      "step": 2300
+    },
+    {
+      "epoch": 2.01,
+      "eval_loss": 1.3014659881591797,
+      "eval_runtime": 376.283,
+      "eval_samples_per_second": 42.962,
+      "eval_steps_per_second": 1.345,
+      "step": 2300
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 5.140677420471003e-06,
+      "loss": 1.2312,
+      "step": 2305
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 5.099168177902539e-06,
+      "loss": 1.2344,
+      "step": 2310
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 5.057769774075985e-06,
+      "loss": 1.2395,
+      "step": 2315
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 5.0164831452705494e-06,
+      "loss": 1.2408,
+      "step": 2320
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 4.9753092252375245e-06,
+      "loss": 1.2335,
+      "step": 2325
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 4.934248945179127e-06,
+      "loss": 1.2372,
+      "step": 2330
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 4.893303233727472e-06,
+      "loss": 1.226,
+      "step": 2335
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 4.8524730169235404e-06,
+      "loss": 1.2449,
+      "step": 2340
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 4.811759218196262e-06,
+      "loss": 1.2331,
+      "step": 2345
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 4.771162758341612e-06,
+      "loss": 1.2319,
+      "step": 2350
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 4.730684555501799e-06,
+      "loss": 1.2382,
+      "step": 2355
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 4.690325525144488e-06,
+      "loss": 1.2332,
+      "step": 2360
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 4.6500865800421015e-06,
+      "loss": 1.235,
+      "step": 2365
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 4.609968630251187e-06,
+      "loss": 1.2301,
+      "step": 2370
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 4.569972583091807e-06,
+      "loss": 1.2202,
+      "step": 2375
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 4.5300993431270565e-06,
+      "loss": 1.2414,
+      "step": 2380
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 4.490349812142564e-06,
+      "loss": 1.2474,
+      "step": 2385
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 4.450724889126135e-06,
+      "loss": 1.235,
+      "step": 2390
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 4.411225470247387e-06,
+      "loss": 1.2201,
+      "step": 2395
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 4.371852448837511e-06,
+      "loss": 1.2459,
+      "step": 2400
+    },
+    {
+      "epoch": 2.1,
+      "eval_loss": 1.2999837398529053,
+      "eval_runtime": 376.4391,
+      "eval_samples_per_second": 42.945,
+      "eval_steps_per_second": 1.344,
+      "step": 2400
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 4.332606715369041e-06,
+      "loss": 1.226,
+      "step": 2405
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 4.2934891574357375e-06,
+      "loss": 1.2191,
+      "step": 2410
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 4.254500659732496e-06,
+      "loss": 1.2408,
+      "step": 2415
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 4.2156421040353435e-06,
+      "loss": 1.246,
+      "step": 2420
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 4.1769143691815095e-06,
+      "loss": 1.2376,
+      "step": 2425
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 4.138318331049525e-06,
+      "loss": 1.2402,
+      "step": 2430
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 4.09985486253944e-06,
+      "loss": 1.2197,
+      "step": 2435
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 4.061524833553058e-06,
+      "loss": 1.2358,
+      "step": 2440
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 4.0233291109742726e-06,
+      "loss": 1.2316,
+      "step": 2445
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 3.985268558649472e-06,
+      "loss": 1.2146,
+      "step": 2450
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 3.947344037367983e-06,
+      "loss": 1.2318,
+      "step": 2455
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 3.909556404842609e-06,
+      "loss": 1.2212,
+      "step": 2460
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 3.871906515690249e-06,
+      "loss": 1.2507,
+      "step": 2465
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 3.834395221412537e-06,
+      "loss": 1.2274,
+      "step": 2470
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 3.797023370376618e-06,
+      "loss": 1.2426,
+      "step": 2475
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 3.7597918077959306e-06,
+      "loss": 1.2212,
+      "step": 2480
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 3.7227013757111197e-06,
+      "loss": 1.2449,
+      "step": 2485
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 3.6857529129709655e-06,
+      "loss": 1.2136,
+      "step": 2490
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 3.64894725521344e-06,
+      "loss": 1.2289,
+      "step": 2495
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 3.61228523484678e-06,
+      "loss": 1.2349,
+      "step": 2500
+    },
+    {
+      "epoch": 2.19,
+      "eval_loss": 1.2988718748092651,
+      "eval_runtime": 375.5011,
+      "eval_samples_per_second": 43.052,
+      "eval_steps_per_second": 1.348,
+      "step": 2500
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 3.5757676810306775e-06,
+      "loss": 1.2277,
+      "step": 2505
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 3.539395419657531e-06,
+      "loss": 1.2365,
+      "step": 2510
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 3.5031692733337475e-06,
+      "loss": 1.2266,
+      "step": 2515
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 3.4670900613611656e-06,
+      "loss": 1.2194,
+      "step": 2520
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 3.431158599718496e-06,
+      "loss": 1.2422,
+      "step": 2525
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 3.3953757010428946e-06,
+      "loss": 1.2485,
+      "step": 2530
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 3.359742174611558e-06,
+      "loss": 1.2313,
+      "step": 2535
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 3.3242588263234467e-06,
+      "loss": 1.2248,
+      "step": 2540
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 3.2889264586810323e-06,
+      "loss": 1.216,
+      "step": 2545
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 3.2537458707721735e-06,
+      "loss": 1.2314,
+      "step": 2550
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 3.2187178582520206e-06,
+      "loss": 1.2437,
+      "step": 2555
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 3.183843213325042e-06,
+      "loss": 1.2093,
+      "step": 2560
+    },
+    {
+      "epoch": 2.25,
+      "learning_rate": 3.149122724727083e-06,
+      "loss": 1.2423,
+      "step": 2565
+    },
+    {
+      "epoch": 2.25,
+      "learning_rate": 3.1145571777075577e-06,
+      "loss": 1.2248,
+      "step": 2570
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 3.080147354011659e-06,
+      "loss": 1.2143,
+      "step": 2575
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 3.0458940318626963e-06,
+      "loss": 1.2322,
+      "step": 2580
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 3.011797985944499e-06,
+      "loss": 1.2424,
+      "step": 2585
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 2.977859987383874e-06,
+      "loss": 1.2312,
+      "step": 2590
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 2.944080803733197e-06,
+      "loss": 1.2095,
+      "step": 2595
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 2.9104611989530196e-06,
+      "loss": 1.2277,
+      "step": 2600
+    },
+    {
+      "epoch": 2.28,
+      "eval_loss": 1.2980636358261108,
+      "eval_runtime": 375.9407,
+      "eval_samples_per_second": 43.001,
+      "eval_steps_per_second": 1.346,
+      "step": 2600
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 2.8770019333948197e-06,
+      "loss": 1.2374,
+      "step": 2605
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 2.843703763783785e-06,
+      "loss": 1.2321,
+      "step": 2610
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 2.810567443201717e-06,
+      "loss": 1.2271,
+      "step": 2615
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 2.7775937210699754e-06,
+      "loss": 1.2308,
+      "step": 2620
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 2.7447833431325566e-06,
+      "loss": 1.1992,
+      "step": 2625
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 2.712137051439202e-06,
+      "loss": 1.2405,
+      "step": 2630
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 2.6796555843286375e-06,
+      "loss": 1.2312,
+      "step": 2635
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 2.6473396764118575e-06,
+      "loss": 1.2339,
+      "step": 2640
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 2.6151900585555178e-06,
+      "loss": 1.2296,
+      "step": 2645
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 2.583207457865413e-06,
+      "loss": 1.2238,
+      "step": 2650
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 2.5513925976700217e-06,
+      "loss": 1.2285,
+      "step": 2655
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 2.519746197504144e-06,
+      "loss": 1.2407,
+      "step": 2660
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 2.488268973092649e-06,
+      "loss": 1.2322,
+      "step": 2665
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 2.456961636334265e-06,
+      "loss": 1.2544,
+      "step": 2670
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 2.425824895285488e-06,
+      "loss": 1.2244,
+      "step": 2675
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 2.3948594541445735e-06,
+      "loss": 1.2348,
+      "step": 2680
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 2.3640660132356e-06,
+      "loss": 1.2166,
+      "step": 2685
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 2.333445268992639e-06,
+      "loss": 1.2246,
+      "step": 2690
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 2.302997913943994e-06,
+      "loss": 1.2392,
+      "step": 2695
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 2.272724636696555e-06,
+      "loss": 1.2243,
+      "step": 2700
+    },
+    {
+      "epoch": 2.37,
+      "eval_loss": 1.2972701787948608,
+      "eval_runtime": 376.0146,
+      "eval_samples_per_second": 42.993,
+      "eval_steps_per_second": 1.346,
+      "step": 2700
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 2.2426261219202006e-06,
+      "loss": 1.2235,
+      "step": 2705
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 2.21270305033234e-06,
+      "loss": 1.2371,
+      "step": 2710
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 2.1829560986824937e-06,
+      "loss": 1.2404,
+      "step": 2715
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 2.1533859397370084e-06,
+      "loss": 1.231,
+      "step": 2720
+    },
+    {
+      "epoch": 2.39,
+      "learning_rate": 2.1239932422638234e-06,
+      "loss": 1.2418,
+      "step": 2725
+    },
+    {
+      "epoch": 2.39,
+      "learning_rate": 2.0947786710173545e-06,
+      "loss": 1.2331,
+      "step": 2730
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 2.06574288672347e-06,
+      "loss": 1.2313,
+      "step": 2735
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 2.0368865460645202e-06,
+      "loss": 1.2146,
+      "step": 2740
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 2.008210301664518e-06,
+      "loss": 1.2373,
+      "step": 2745
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 1.9797148020743496e-06,
+      "loss": 1.2273,
+      "step": 2750
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 1.951400691757133e-06,
+      "loss": 1.227,
+      "step": 2755
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 1.9232686110736165e-06,
+      "loss": 1.2312,
+      "step": 2760
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 1.895319196267722e-06,
+      "loss": 1.2415,
+      "step": 2765
+    },
+    {
+      "epoch": 2.43,
+      "learning_rate": 1.8675530794521312e-06,
+      "loss": 1.2207,
+      "step": 2770
+    },
+    {
+      "epoch": 2.43,
+      "learning_rate": 1.8399708885940136e-06,
+      "loss": 1.2316,
+      "step": 2775
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 1.8125732475007983e-06,
+      "loss": 1.2363,
+      "step": 2780
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 1.785360775806093e-06,
+      "loss": 1.2317,
+      "step": 2785
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 1.7583340889556456e-06,
+      "loss": 1.2155,
+      "step": 2790
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 1.7314937981934399e-06,
+      "loss": 1.2234,
+      "step": 2795
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 1.7048405105478717e-06,
+      "loss": 1.2298,
+      "step": 2800
+    },
+    {
+      "epoch": 2.45,
+      "eval_loss": 1.2966749668121338,
+      "eval_runtime": 375.7961,
+      "eval_samples_per_second": 43.018,
+      "eval_steps_per_second": 1.346,
+      "step": 2800
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 1.6783748288180058e-06,
+      "loss": 1.2287,
+      "step": 2805
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 1.652097351559967e-06,
+      "loss": 1.2325,
+      "step": 2810
+    },
+    {
+      "epoch": 2.47,
+      "learning_rate": 1.6260086730733749e-06,
+      "loss": 1.2389,
+      "step": 2815
+    },
+    {
+      "epoch": 2.47,
+      "learning_rate": 1.6001093833879288e-06,
+      "loss": 1.2271,
+      "step": 2820
+    },
+    {
+      "epoch": 2.47,
+      "learning_rate": 1.5744000682500426e-06,
+      "loss": 1.2189,
+      "step": 2825
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 1.5488813091096145e-06,
+      "loss": 1.2409,
+      "step": 2830
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 1.523553683106861e-06,
+      "loss": 1.2492,
+      "step": 2835
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 1.49841776305928e-06,
+      "loss": 1.2411,
+      "step": 2840
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 1.4734741174486788e-06,
+      "loss": 1.2224,
+      "step": 2845
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 1.4487233104083354e-06,
+      "loss": 1.2338,
+      "step": 2850
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 1.424165901710224e-06,
+      "loss": 1.2414,
+      "step": 2855
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 1.3998024467523596e-06,
+      "loss": 1.2141,
+      "step": 2860
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 1.3756334965462502e-06,
+      "loss": 1.2522,
+      "step": 2865
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 1.3516595977044112e-06,
+      "loss": 1.229,
+      "step": 2870
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 1.3278812924280192e-06,
+      "loss": 1.2113,
+      "step": 2875
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 1.304299118494652e-06,
+      "loss": 1.2278,
+      "step": 2880
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 1.2809136092461084e-06,
+      "loss": 1.2454,
+      "step": 2885
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 1.2577252935763695e-06,
+      "loss": 1.2515,
+      "step": 2890
+    },
+    {
+      "epoch": 2.54,
+      "learning_rate": 1.234734695919616e-06,
+      "loss": 1.2334,
+      "step": 2895
+    },
+    {
+      "epoch": 2.54,
+      "learning_rate": 1.2119423362383776e-06,
+      "loss": 1.2362,
+      "step": 2900
+    },
+    {
+      "epoch": 2.54,
+      "eval_loss": 1.2961491346359253,
+      "eval_runtime": 376.749,
+      "eval_samples_per_second": 42.909,
+      "eval_steps_per_second": 1.343,
+      "step": 2900
+    },
+    {
+      "epoch": 2.54,
+      "learning_rate": 1.189348730011778e-06,
+      "loss": 1.2152,
+      "step": 2905
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 1.166954388223862e-06,
+      "loss": 1.2258,
+      "step": 2910
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 1.1447598173520558e-06,
+      "loss": 1.233,
+      "step": 2915
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 1.1227655193556973e-06,
+      "loss": 1.2287,
+      "step": 2920
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 1.1009719916646977e-06,
+      "loss": 1.2297,
+      "step": 2925
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 1.079379727168276e-06,
+      "loss": 1.2245,
+      "step": 2930
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 1.0579892142038284e-06,
+      "loss": 1.2418,
+      "step": 2935
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 1.0368009365458697e-06,
+      "loss": 1.2276,
+      "step": 2940
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 1.0158153733950981e-06,
+      "loss": 1.2267,
+      "step": 2945
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 9.950329993675623e-07,
+      "loss": 1.2148,
+      "step": 2950
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 9.744542844839145e-07,
+      "loss": 1.2299,
+      "step": 2955
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 9.540796941587983e-07,
+      "loss": 1.218,
+      "step": 2960
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 9.33909689190301e-07,
+      "loss": 1.2303,
+      "step": 2965
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 9.139447257495537e-07,
+      "loss": 1.2468,
+      "step": 2970
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 8.941852553703966e-07,
+      "loss": 1.242,
+      "step": 2975
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 8.746317249391834e-07,
+      "loss": 1.2271,
+      "step": 2980
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 8.55284576684654e-07,
+      "loss": 1.2411,
+      "step": 2985
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 8.361442481679561e-07,
+      "loss": 1.2299,
+      "step": 2990
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 8.172111722727294e-07,
+      "loss": 1.2351,
+      "step": 2995
+    },
+    {
+      "epoch": 2.63,
+      "learning_rate": 7.984857771953303e-07,
+      "loss": 1.216,
+      "step": 3000
+    },
+    {
+      "epoch": 2.63,
+      "eval_loss": 1.2958320379257202,
+      "eval_runtime": 375.7361,
+      "eval_samples_per_second": 43.025,
+      "eval_steps_per_second": 1.347,
+      "step": 3000
+    },
+    {
+      "epoch": 2.63,
+      "learning_rate": 7.799684864351342e-07,
+      "loss": 1.2298,
+      "step": 3005
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 7.616597187849683e-07,
+      "loss": 1.2185,
+      "step": 3010
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 7.435598883216377e-07,
+      "loss": 1.2264,
+      "step": 3015
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 7.256694043965528e-07,
+      "loss": 1.2266,
+      "step": 3020
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 7.07988671626485e-07,
+      "loss": 1.2444,
+      "step": 3025
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 6.905180898844022e-07,
+      "loss": 1.2194,
+      "step": 3030
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 6.732580542904343e-07,
+      "loss": 1.2114,
+      "step": 3035
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 6.562089552029305e-07,
+      "loss": 1.2334,
+      "step": 3040
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 6.39371178209639e-07,
+      "loss": 1.2286,
+      "step": 3045
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 6.227451041189759e-07,
+      "loss": 1.2257,
+      "step": 3050
+    },
+    {
+      "epoch": 2.68,
+      "learning_rate": 6.063311089514256e-07,
+      "loss": 1.2396,
+      "step": 3055
+    },
+    {
+      "epoch": 2.68,
+      "learning_rate": 5.901295639310212e-07,
+      "loss": 1.2368,
+      "step": 3060
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 5.74140835476964e-07,
+      "loss": 1.2484,
+      "step": 3065
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 5.583652851953225e-07,
+      "loss": 1.2235,
+      "step": 3070
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 5.428032698708696e-07,
+      "loss": 1.22,
+      "step": 3075
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 5.274551414589979e-07,
+      "loss": 1.2334,
+      "step": 3080
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 5.123212470777684e-07,
+      "loss": 1.2464,
+      "step": 3085
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 4.97401929000062e-07,
+      "loss": 1.2303,
+      "step": 3090
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 4.826975246458299e-07,
+      "loss": 1.2323,
+      "step": 3095
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 4.6820836657446964e-07,
+      "loss": 1.2381,
+      "step": 3100
+    },
+    {
+      "epoch": 2.72,
+      "eval_loss": 1.295718789100647,
+      "eval_runtime": 376.6477,
+      "eval_samples_per_second": 42.921,
+      "eval_steps_per_second": 1.343,
+      "step": 3100
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 4.5393478247730436e-07,
+      "loss": 1.2417,
+      "step": 3105
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 4.398770951701647e-07,
+      "loss": 1.2284,
+      "step": 3110
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 4.2603562258609176e-07,
+      "loss": 1.2261,
+      "step": 3115
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 4.124106777681536e-07,
+      "loss": 1.2191,
+      "step": 3120
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 3.9900256886235e-07,
+      "loss": 1.2212,
+      "step": 3125
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 3.8581159911065926e-07,
+      "loss": 1.236,
+      "step": 3130
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 3.7283806684416777e-07,
+      "loss": 1.2396,
+      "step": 3135
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 3.600822654763314e-07,
+      "loss": 1.2157,
+      "step": 3140
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 3.4754448349633374e-07,
+      "loss": 1.2179,
+      "step": 3145
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 3.35225004462566e-07,
+      "loss": 1.2363,
+      "step": 3150
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 3.2312410699620986e-07,
+      "loss": 1.238,
+      "step": 3155
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 3.11242064774937e-07,
+      "loss": 1.2162,
+      "step": 3160
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 2.99579146526725e-07,
+      "loss": 1.216,
+      "step": 3165
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 2.8813561602377025e-07,
+      "loss": 1.2313,
+      "step": 3170
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 2.7691173207653355e-07,
+      "loss": 1.2357,
+      "step": 3175
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 2.659077485278716e-07,
+      "loss": 1.2357,
+      "step": 3180
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 2.551239142473161e-07,
+      "loss": 1.2245,
+      "step": 3185
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 2.4456047312542365e-07,
+      "loss": 1.2373,
+      "step": 3190
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 2.3421766406827807e-07,
+      "loss": 1.233,
+      "step": 3195
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 2.2409572099207576e-07,
+      "loss": 1.2274,
+      "step": 3200
+    },
+    {
+      "epoch": 2.8,
+      "eval_loss": 1.2954812049865723,
+      "eval_runtime": 376.94,
+      "eval_samples_per_second": 42.887,
+      "eval_steps_per_second": 1.342,
+      "step": 3200
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 2.1419487281784002e-07,
+      "loss": 1.2134,
+      "step": 3205
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 2.045153434662428e-07,
+      "loss": 1.2267,
+      "step": 3210
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 1.9505735185254226e-07,
+      "loss": 1.2198,
+      "step": 3215
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 1.8582111188162555e-07,
+      "loss": 1.2278,
+      "step": 3220
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 1.7680683244318154e-07,
+      "loss": 1.2239,
+      "step": 3225
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 1.6801471740696462e-07,
+      "loss": 1.2362,
+      "step": 3230
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 1.594449656181918e-07,
+      "loss": 1.2259,
+      "step": 3235
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 1.510977708930461e-07,
+      "loss": 1.241,
+      "step": 3240
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 1.4297332201428703e-07,
+      "loss": 1.2141,
+      "step": 3245
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 1.3507180272698594e-07,
+      "loss": 1.2515,
+      "step": 3250
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 1.2739339173436838e-07,
+      "loss": 1.2287,
+      "step": 3255
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 1.1993826269377506e-07,
+      "loss": 1.2405,
+      "step": 3260
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 1.1270658421273062e-07,
+      "loss": 1.2204,
+      "step": 3265
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 1.0569851984513102e-07,
+      "loss": 1.2257,
+      "step": 3270
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 9.89142280875477e-08,
+      "loss": 1.2322,
+      "step": 3275
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 9.235386237564148e-08,
+      "loss": 1.235,
+      "step": 3280
+    },
+    {
+      "epoch": 2.88,
+      "learning_rate": 8.601757108068876e-08,
+      "loss": 1.2182,
+      "step": 3285
+    },
+    {
+      "epoch": 2.88,
+      "learning_rate": 7.990549750623189e-08,
+      "loss": 1.2252,
+      "step": 3290
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 7.401777988483406e-08,
+      "loss": 1.2378,
+      "step": 3295
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 6.835455137495395e-08,
+      "loss": 1.2235,
+      "step": 3300
+    },
+    {
+      "epoch": 2.89,
+      "eval_loss": 1.295430064201355,
+      "eval_runtime": 376.7433,
+      "eval_samples_per_second": 42.91,
+      "eval_steps_per_second": 1.343,
+      "step": 3300
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 6.2915940057936e-08,
+      "loss": 1.2124,
+      "step": 3305
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 5.7702068935109324e-08,
+      "loss": 1.204,
+      "step": 3310
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 5.271305592501108e-08,
+      "loss": 1.2171,
+      "step": 3315
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 4.794901386071749e-08,
+      "loss": 1.2299,
+      "step": 3320
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 4.341005048728919e-08,
+      "loss": 1.2314,
+      "step": 3325
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 3.9096268459339893e-08,
+      "loss": 1.2288,
+      "step": 3330
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 3.50077653387082e-08,
+      "loss": 1.2327,
+      "step": 3335
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 3.114463359225717e-08,
+      "loss": 1.2305,
+      "step": 3340
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 2.7506960589781527e-08,
+      "loss": 1.2301,
+      "step": 3345
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 2.4094828602027052e-08,
+      "loss": 1.217,
+      "step": 3350
+    },
+    {
+      "epoch": 2.94,
+      "learning_rate": 2.0908314798836483e-08,
+      "loss": 1.2319,
+      "step": 3355
+    },
+    {
+      "epoch": 2.94,
+      "learning_rate": 1.7947491247399808e-08,
+      "loss": 1.2411,
+      "step": 3360
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 1.5212424910627797e-08,
+      "loss": 1.2352,
+      "step": 3365
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 1.2703177645634335e-08,
+      "loss": 1.2282,
+      "step": 3370
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 1.0419806202336403e-08,
+      "loss": 1.2282,
+      "step": 3375
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 8.362362222177345e-09,
+      "loss": 1.2319,
+      "step": 3380
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 6.530892236951136e-09,
+      "loss": 1.2159,
+      "step": 3385
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 4.925437667755439e-09,
+      "loss": 1.2303,
+      "step": 3390
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 3.5460348240501376e-09,
+      "loss": 1.2118,
+      "step": 3395
+    },
+    {
+      "epoch": 2.98,
+      "learning_rate": 2.39271490284243e-09,
+      "loss": 1.2438,
+      "step": 3400
+    },
+    {
+      "epoch": 2.98,
+      "eval_loss": 1.2954297065734863,
+      "eval_runtime": 376.4412,
+      "eval_samples_per_second": 42.944,
+      "eval_steps_per_second": 1.344,
+      "step": 3400
+    },
+    {
+      "epoch": 2.98,
+      "learning_rate": 1.4655039879740706e-09,
+      "loss": 1.2334,
+      "step": 3405
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 7.644230495373884e-10,
+      "loss": 1.2246,
+      "step": 3410
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 2.8948794339789255e-10,
+      "loss": 1.2263,
+      "step": 3415
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.07094108345607e-11,
+      "loss": 1.235,
+      "step": 3420
+    },
+    {
+      "epoch": 3.0,
+      "step": 3423,
+      "total_flos": 3.426180505951378e+19,
+      "train_loss": 1.4016100346188847,
+      "train_runtime": 49299.9861,
+      "train_samples_per_second": 8.888,
+      "train_steps_per_second": 0.069
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 3423,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 100,
+  "total_flos": 3.426180505951378e+19,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}