Model save

README.md

@@ -0,0 +1,100 @@
+---
+license: apache-2.0
+base_model: ondevicellm/tinyllama_moe
+tags:
+- trl
+- sft
+- generated_from_trainer
+datasets:
+- generator
+model-index:
+- name: tinyllama_moe_sft_ultrachat200k_v2_epochs3
+  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_ultrachat200k_v2_epochs3
+
+This model is a fine-tuned version of [ondevicellm/tinyllama_moe](https://huggingface.co/ondevicellm/tinyllama_moe) on the generator dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.1178
+
+## 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: 115
+- num_epochs: 3
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 1.336         | 0.09  | 100  | 1.3129          |
+| 1.2424        | 0.18  | 200  | 1.2363          |
+| 1.2079        | 0.26  | 300  | 1.2084          |
+| 1.185         | 0.35  | 400  | 1.1911          |
+| 1.1546        | 0.44  | 500  | 1.1787          |
+| 1.1741        | 0.53  | 600  | 1.1692          |
+| 1.1612        | 0.61  | 700  | 1.1613          |
+| 1.1453        | 0.7   | 800  | 1.1547          |
+| 1.141         | 0.79  | 900  | 1.1489          |
+| 1.1247        | 0.88  | 1000 | 1.1438          |
+| 1.1485        | 0.96  | 1100 | 1.1392          |
+| 1.067         | 1.05  | 1200 | 1.1387          |
+| 1.0694        | 1.14  | 1300 | 1.1368          |
+| 1.0814        | 1.23  | 1400 | 1.1341          |
+| 1.0727        | 1.31  | 1500 | 1.1316          |
+| 1.0769        | 1.4   | 1600 | 1.1292          |
+| 1.0728        | 1.49  | 1700 | 1.1270          |
+| 1.0558        | 1.58  | 1800 | 1.1247          |
+| 1.0753        | 1.66  | 1900 | 1.1229          |
+| 1.0799        | 1.75  | 2000 | 1.1209          |
+| 1.066         | 1.84  | 2100 | 1.1192          |
+| 1.0406        | 1.93  | 2200 | 1.1178          |
+| 1.0193        | 2.01  | 2300 | 1.1222          |
+| 1.0276        | 2.1   | 2400 | 1.1220          |
+| 1.0171        | 2.19  | 2500 | 1.1215          |
+| 1.0112        | 2.28  | 2600 | 1.1211          |
+| 1.0087        | 2.37  | 2700 | 1.1207          |
+| 1.0158        | 2.45  | 2800 | 1.1204          |
+| 1.0219        | 2.54  | 2900 | 1.1199          |
+| 1.0024        | 2.63  | 3000 | 1.1197          |
+| 1.019         | 2.72  | 3100 | 1.1197          |
+| 1.0135        | 2.8   | 3200 | 1.1194          |
+| 1.0094        | 2.89  | 3300 | 1.1194          |
+| 1.0284        | 2.98  | 3400 | 1.1194          |
+
+
+### Framework versions
+
+- Transformers 4.36.2
+- Pytorch 2.1.2+cu118
+- Datasets 2.14.6
+- Tokenizers 0.15.0

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 1.1177691221237183,
+    "eval_runtime": 426.5899,
+    "eval_samples": 23110,
+    "eval_samples_per_second": 37.896,
+    "eval_steps_per_second": 1.186,
+    "train_loss": 1.1024342694896863,
+    "train_runtime": 51497.7783,
+    "train_samples": 207865,
+    "train_samples_per_second": 8.508,
+    "train_steps_per_second": 0.066
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 1.1177691221237183,
+    "eval_runtime": 426.5899,
+    "eval_samples": 23110,
+    "eval_samples_per_second": 37.896,
+    "eval_steps_per_second": 1.186
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.36.2",
+  "use_cache": false
+}

model-00001-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c001402a18e8e178c1727401980da81545c07b1fc5e3433fc9521cc2437cfdad
+size 4984275728

model-00002-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f009665206ac49e54bf93808a84d2959c12c0272f2aba5f0672f0b7f4c39f73
+size 4991625072

model-00003-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:171d052cb8033b6dafd50c5b2139eff39a3ebc11d072ff6010e18027c1abb1db
+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/Jan17_21-13-19_main1/events.out.tfevents.1705526089.main1.11993.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d0150bb65644574cd94662863c2eab1bc5ec04846368036c9a8af771add5fded
-size 120785
+oid sha256:7c70132a0f5763eb38ae3af3b8f80cd7a78744e86c7992bd31834eafd9069d66
+size 121767

runs/Jan17_21-13-19_main1/events.out.tfevents.1705578011.main1.11993.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cefd06a18ab91b3c763a7c07b11cdf25883385e1467da3aee8e38ac13eb65e5a
+size 359

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "train_loss": 1.1024342694896863,
+    "train_runtime": 51497.7783,
+    "train_samples": 207865,
+    "train_samples_per_second": 8.508,
+    "train_steps_per_second": 0.066
+}

trainer_state.json

@@ -0,0 +1,4412 @@
+{
+  "best_metric": 1.1177691221237183,
+  "best_model_checkpoint": "data/tinyllama_moe_sft_ultrachat200k_v2_epochs3/checkpoint-2200",
+  "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": 1.7391304347826088e-07,
+      "loss": 2.7477,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.695652173913044e-07,
+      "loss": 2.6989,
+      "step": 5
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7391304347826088e-06,
+      "loss": 2.7026,
+      "step": 10
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6086956521739132e-06,
+      "loss": 2.5994,
+      "step": 15
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4782608695652175e-06,
+      "loss": 2.4352,
+      "step": 20
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.347826086956522e-06,
+      "loss": 2.0772,
+      "step": 25
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 5.2173913043478265e-06,
+      "loss": 1.8706,
+      "step": 30
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 6.086956521739132e-06,
+      "loss": 1.7352,
+      "step": 35
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 6.956521739130435e-06,
+      "loss": 1.6574,
+      "step": 40
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 7.82608695652174e-06,
+      "loss": 1.5879,
+      "step": 45
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 8.695652173913044e-06,
+      "loss": 1.5429,
+      "step": 50
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.565217391304349e-06,
+      "loss": 1.4805,
+      "step": 55
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.0434782608695653e-05,
+      "loss": 1.4633,
+      "step": 60
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.1304347826086957e-05,
+      "loss": 1.4199,
+      "step": 65
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.2173913043478263e-05,
+      "loss": 1.4117,
+      "step": 70
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.3043478260869566e-05,
+      "loss": 1.3668,
+      "step": 75
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.391304347826087e-05,
+      "loss": 1.359,
+      "step": 80
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.4782608695652174e-05,
+      "loss": 1.3563,
+      "step": 85
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.565217391304348e-05,
+      "loss": 1.3425,
+      "step": 90
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.6521739130434785e-05,
+      "loss": 1.3117,
+      "step": 95
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.739130434782609e-05,
+      "loss": 1.336,
+      "step": 100
+    },
+    {
+      "epoch": 0.09,
+      "eval_loss": 1.312857747077942,
+      "eval_runtime": 427.5314,
+      "eval_samples_per_second": 37.812,
+      "eval_steps_per_second": 1.184,
+      "step": 100
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.8260869565217393e-05,
+      "loss": 1.3086,
+      "step": 105
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9130434782608697e-05,
+      "loss": 1.2992,
+      "step": 110
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 2e-05,
+      "loss": 1.2999,
+      "step": 115
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9999887260023335e-05,
+      "loss": 1.2738,
+      "step": 120
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.99995490426354e-05,
+      "loss": 1.277,
+      "step": 125
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9998985355462318e-05,
+      "loss": 1.255,
+      "step": 130
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.999819621121411e-05,
+      "loss": 1.271,
+      "step": 135
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.999718162768439e-05,
+      "loss": 1.282,
+      "step": 140
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9995941627749983e-05,
+      "loss": 1.2597,
+      "step": 145
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9994476239370407e-05,
+      "loss": 1.2623,
+      "step": 150
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9992785495587226e-05,
+      "loss": 1.2541,
+      "step": 155
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.999086943452333e-05,
+      "loss": 1.248,
+      "step": 160
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.998872809938205e-05,
+      "loss": 1.2444,
+      "step": 165
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9986361538446202e-05,
+      "loss": 1.2482,
+      "step": 170
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.998376980507699e-05,
+      "loss": 1.2575,
+      "step": 175
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.998095295771281e-05,
+      "loss": 1.2307,
+      "step": 180
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9977911059867917e-05,
+      "loss": 1.2328,
+      "step": 185
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9974644180131016e-05,
+      "loss": 1.2423,
+      "step": 190
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.997115239216369e-05,
+      "loss": 1.2272,
+      "step": 195
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.996743577469876e-05,
+      "loss": 1.2424,
+      "step": 200
+    },
+    {
+      "epoch": 0.18,
+      "eval_loss": 1.236301302909851,
+      "eval_runtime": 425.2862,
+      "eval_samples_per_second": 38.012,
+      "eval_steps_per_second": 1.19,
+      "step": 200
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.99634944115385e-05,
+      "loss": 1.2403,
+      "step": 205
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.995932839155275e-05,
+      "loss": 1.2136,
+      "step": 210
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9954937808676906e-05,
+      "loss": 1.2461,
+      "step": 215
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.995032276190981e-05,
+      "loss": 1.2197,
+      "step": 220
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9945483355311515e-05,
+      "loss": 1.2035,
+      "step": 225
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9940419698000944e-05,
+      "loss": 1.2222,
+      "step": 230
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.993513190415341e-05,
+      "loss": 1.2323,
+      "step": 235
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.992962009299807e-05,
+      "loss": 1.2048,
+      "step": 240
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.992388438881521e-05,
+      "loss": 1.2339,
+      "step": 245
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9917924920933472e-05,
+      "loss": 1.2085,
+      "step": 250
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.99117418237269e-05,
+      "loss": 1.2169,
+      "step": 255
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.990533523661194e-05,
+      "loss": 1.2209,
+      "step": 260
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9898705304044303e-05,
+      "loss": 1.2016,
+      "step": 265
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.989185217551566e-05,
+      "loss": 1.2028,
+      "step": 270
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9884776005550326e-05,
+      "loss": 1.2171,
+      "step": 275
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9877476953701756e-05,
+      "loss": 1.1922,
+      "step": 280
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.986995518454893e-05,
+      "loss": 1.224,
+      "step": 285
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9862210867692665e-05,
+      "loss": 1.1925,
+      "step": 290
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9854244177751782e-05,
+      "loss": 1.2121,
+      "step": 295
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.984605529435917e-05,
+      "loss": 1.2079,
+      "step": 300
+    },
+    {
+      "epoch": 0.26,
+      "eval_loss": 1.208355188369751,
+      "eval_runtime": 425.2694,
+      "eval_samples_per_second": 38.014,
+      "eval_steps_per_second": 1.19,
+      "step": 300
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9837644402157726e-05,
+      "loss": 1.2104,
+      "step": 305
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9829011690796222e-05,
+      "loss": 1.1775,
+      "step": 310
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.982015735492498e-05,
+      "loss": 1.2005,
+      "step": 315
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.981108159419153e-05,
+      "loss": 1.1946,
+      "step": 320
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9801784613236082e-05,
+      "loss": 1.1911,
+      "step": 325
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9792266621686924e-05,
+      "loss": 1.1859,
+      "step": 330
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9782527834155674e-05,
+      "loss": 1.2373,
+      "step": 335
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.977256847023248e-05,
+      "loss": 1.1896,
+      "step": 340
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9762388754481024e-05,
+      "loss": 1.1971,
+      "step": 345
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9751988916433494e-05,
+      "loss": 1.1909,
+      "step": 350
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9741369190585383e-05,
+      "loss": 1.19,
+      "step": 355
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9730529816390232e-05,
+      "loss": 1.1955,
+      "step": 360
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.971947103825419e-05,
+      "loss": 1.1921,
+      "step": 365
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9708193105530537e-05,
+      "loss": 1.1863,
+      "step": 370
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.969669627251405e-05,
+      "loss": 1.1778,
+      "step": 375
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9684980798435266e-05,
+      "loss": 1.1933,
+      "step": 380
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9673046947454634e-05,
+      "loss": 1.1831,
+      "step": 385
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9660894988656575e-05,
+      "loss": 1.1763,
+      "step": 390
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9648525196043402e-05,
+      "loss": 1.1778,
+      "step": 395
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9635937848529133e-05,
+      "loss": 1.185,
+      "step": 400
+    },
+    {
+      "epoch": 0.35,
+      "eval_loss": 1.191055178642273,
+      "eval_runtime": 424.901,
+      "eval_samples_per_second": 38.047,
+      "eval_steps_per_second": 1.191,
+      "step": 400
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.962313322993323e-05,
+      "loss": 1.2061,
+      "step": 405
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.961011162897417e-05,
+      "loss": 1.1976,
+      "step": 410
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9596873339262948e-05,
+      "loss": 1.184,
+      "step": 415
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9583418659296462e-05,
+      "loss": 1.1855,
+      "step": 420
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9569747892450768e-05,
+      "loss": 1.1769,
+      "step": 425
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.955586134697426e-05,
+      "loss": 1.1843,
+      "step": 430
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9541759335980695e-05,
+      "loss": 1.1709,
+      "step": 435
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9527442177442154e-05,
+      "loss": 1.1852,
+      "step": 440
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.951291019418186e-05,
+      "loss": 1.1749,
+      "step": 445
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.94981637138669e-05,
+      "loss": 1.1996,
+      "step": 450
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9483203069000854e-05,
+      "loss": 1.1702,
+      "step": 455
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9468028596916265e-05,
+      "loss": 1.1689,
+      "step": 460
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9452640639767058e-05,
+      "loss": 1.1794,
+      "step": 465
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.943703954452082e-05,
+      "loss": 1.1833,
+      "step": 470
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.9421225662950975e-05,
+      "loss": 1.1773,
+      "step": 475
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.940519935162885e-05,
+      "loss": 1.1847,
+      "step": 480
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.9388960971915637e-05,
+      "loss": 1.151,
+      "step": 485
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9372510889954247e-05,
+      "loss": 1.1838,
+      "step": 490
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9355849476661055e-05,
+      "loss": 1.1672,
+      "step": 495
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.933897710771752e-05,
+      "loss": 1.1546,
+      "step": 500
+    },
+    {
+      "epoch": 0.44,
+      "eval_loss": 1.1787223815917969,
+      "eval_runtime": 427.1249,
+      "eval_samples_per_second": 37.848,
+      "eval_steps_per_second": 1.185,
+      "step": 500
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9321894163561753e-05,
+      "loss": 1.1683,
+      "step": 505
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9304601029379884e-05,
+      "loss": 1.1716,
+      "step": 510
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9287098095097434e-05,
+      "loss": 1.1793,
+      "step": 515
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.926938575537048e-05,
+      "loss": 1.1494,
+      "step": 520
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9251464409576766e-05,
+      "loss": 1.169,
+      "step": 525
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9233334461806724e-05,
+      "loss": 1.1595,
+      "step": 530
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.921499632085433e-05,
+      "loss": 1.1558,
+      "step": 535
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.91964504002079e-05,
+      "loss": 1.1704,
+      "step": 540
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9177697118040763e-05,
+      "loss": 1.1593,
+      "step": 545
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9158736897201843e-05,
+      "loss": 1.1715,
+      "step": 550
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9139570165206106e-05,
+      "loss": 1.1723,
+      "step": 555
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.912019735422494e-05,
+      "loss": 1.179,
+      "step": 560
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.910061890107639e-05,
+      "loss": 1.1761,
+      "step": 565
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.9080835247215335e-05,
+      "loss": 1.1675,
+      "step": 570
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.9060846838723505e-05,
+      "loss": 1.1686,
+      "step": 575
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.9040654126299443e-05,
+      "loss": 1.153,
+      "step": 580
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.902025756524833e-05,
+      "loss": 1.1687,
+      "step": 585
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.8999657615471737e-05,
+      "loss": 1.1599,
+      "step": 590
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.897885474145723e-05,
+      "loss": 1.1848,
+      "step": 595
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.8957849412267914e-05,
+      "loss": 1.1741,
+      "step": 600
+    },
+    {
+      "epoch": 0.53,
+      "eval_loss": 1.1691807508468628,
+      "eval_runtime": 426.8549,
+      "eval_samples_per_second": 37.872,
+      "eval_steps_per_second": 1.185,
+      "step": 600
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.893664210153186e-05,
+      "loss": 1.1447,
+      "step": 605
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.8915233287431403e-05,
+      "loss": 1.1569,
+      "step": 610
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.8893623452692392e-05,
+      "loss": 1.1776,
+      "step": 615
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.8871813084573275e-05,
+      "loss": 1.1536,
+      "step": 620
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.884980267485413e-05,
+      "loss": 1.1573,
+      "step": 625
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8827592719825577e-05,
+      "loss": 1.1575,
+      "step": 630
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.880518372027757e-05,
+      "loss": 1.177,
+      "step": 635
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.8782576181488135e-05,
+      "loss": 1.168,
+      "step": 640
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.8759770613211946e-05,
+      "loss": 1.1651,
+      "step": 645
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.8736767529668855e-05,
+      "loss": 1.1548,
+      "step": 650
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.8713567449532277e-05,
+      "loss": 1.1605,
+      "step": 655
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.8690170895917513e-05,
+      "loss": 1.1651,
+      "step": 660
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.8666578396369947e-05,
+      "loss": 1.1562,
+      "step": 665
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.8642790482853144e-05,
+      "loss": 1.1434,
+      "step": 670
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.8618807691736867e-05,
+      "loss": 1.1667,
+      "step": 675
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.8594630563784983e-05,
+      "loss": 1.1737,
+      "step": 680
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.8570259644143256e-05,
+      "loss": 1.1543,
+      "step": 685
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.854569548232707e-05,
+      "loss": 1.1558,
+      "step": 690
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.8520938632209032e-05,
+      "loss": 1.1488,
+      "step": 695
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.8495989652006485e-05,
+      "loss": 1.1612,
+      "step": 700
+    },
+    {
+      "epoch": 0.61,
+      "eval_loss": 1.1612565517425537,
+      "eval_runtime": 426.665,
+      "eval_samples_per_second": 37.889,
+      "eval_steps_per_second": 1.186,
+      "step": 700
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.847084910426891e-05,
+      "loss": 1.1588,
+      "step": 705
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.8445517555865265e-05,
+      "loss": 1.1313,
+      "step": 710
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.8419995577971186e-05,
+      "loss": 1.1387,
+      "step": 715
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.839428374605611e-05,
+      "loss": 1.1686,
+      "step": 720
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8368382639870307e-05,
+      "loss": 1.1371,
+      "step": 725
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8342292843431794e-05,
+      "loss": 1.1781,
+      "step": 730
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.831601494501318e-05,
+      "loss": 1.1627,
+      "step": 735
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.8289549537128395e-05,
+      "loss": 1.1508,
+      "step": 740
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.8262897216519334e-05,
+      "loss": 1.1641,
+      "step": 745
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.82360585841424e-05,
+      "loss": 1.1495,
+      "step": 750
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.8209034245154944e-05,
+      "loss": 1.1637,
+      "step": 755
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.818182480890164e-05,
+      "loss": 1.1523,
+      "step": 760
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8154430888900728e-05,
+      "loss": 1.1368,
+      "step": 765
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8126853102830195e-05,
+      "loss": 1.1374,
+      "step": 770
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8099092072513826e-05,
+      "loss": 1.1433,
+      "step": 775
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8071148423907198e-05,
+      "loss": 1.1357,
+      "step": 780
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.804302278708358e-05,
+      "loss": 1.1536,
+      "step": 785
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.8014715796219698e-05,
+      "loss": 1.1308,
+      "step": 790
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.7986228089581443e-05,
+      "loss": 1.1589,
+      "step": 795
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.7957560309509498e-05,
+      "loss": 1.1453,
+      "step": 800
+    },
+    {
+      "epoch": 0.7,
+      "eval_loss": 1.1546913385391235,
+      "eval_runtime": 427.2728,
+      "eval_samples_per_second": 37.835,
+      "eval_steps_per_second": 1.184,
+      "step": 800
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.7928713102404825e-05,
+      "loss": 1.1298,
+      "step": 805
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.789968711871413e-05,
+      "loss": 1.1348,
+      "step": 810
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.7870483012915146e-05,
+      "loss": 1.1322,
+      "step": 815
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.784110144350192e-05,
+      "loss": 1.1252,
+      "step": 820
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.7811543072969938e-05,
+      "loss": 1.1476,
+      "step": 825
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.7781808567801206e-05,
+      "loss": 1.1424,
+      "step": 830
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.77518985984492e-05,
+      "loss": 1.167,
+      "step": 835
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7721813839323778e-05,
+      "loss": 1.1445,
+      "step": 840
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7691554968775942e-05,
+      "loss": 1.1596,
+      "step": 845
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7661122669082566e-05,
+      "loss": 1.1383,
+      "step": 850
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.7630517626431002e-05,
+      "loss": 1.1507,
+      "step": 855
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.759974053090361e-05,
+      "loss": 1.1455,
+      "step": 860
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.7568792076462194e-05,
+      "loss": 1.1606,
+      "step": 865
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.7537672960932363e-05,
+      "loss": 1.152,
+      "step": 870
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.750638388598778e-05,
+      "loss": 1.1334,
+      "step": 875
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7474925557134375e-05,
+      "loss": 1.1466,
+      "step": 880
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.744329868369439e-05,
+      "loss": 1.1321,
+      "step": 885
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.741150397879042e-05,
+      "loss": 1.1321,
+      "step": 890
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.7379542159329334e-05,
+      "loss": 1.1377,
+      "step": 895
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.734741394598607e-05,
+      "loss": 1.141,
+      "step": 900
+    },
+    {
+      "epoch": 0.79,
+      "eval_loss": 1.1488687992095947,
+      "eval_runtime": 426.9549,
+      "eval_samples_per_second": 37.863,
+      "eval_steps_per_second": 1.185,
+      "step": 900
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.7315120063187446e-05,
+      "loss": 1.1465,
+      "step": 905
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.7282661239095774e-05,
+      "loss": 1.1435,
+      "step": 910
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.7250038205592474e-05,
+      "loss": 1.1562,
+      "step": 915
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.721725169826155e-05,
+      "loss": 1.1324,
+      "step": 920
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.7184302456373013e-05,
+      "loss": 1.1414,
+      "step": 925
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.7151191222866225e-05,
+      "loss": 1.1318,
+      "step": 930
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.711791874433311e-05,
+      "loss": 1.155,
+      "step": 935
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.7084485771001374e-05,
+      "loss": 1.1459,
+      "step": 940
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.705089305671753e-05,
+      "loss": 1.1466,
+      "step": 945
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.7017141358929953e-05,
+      "loss": 1.156,
+      "step": 950
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.6983231438671762e-05,
+      "loss": 1.1527,
+      "step": 955
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.6949164060543684e-05,
+      "loss": 1.1354,
+      "step": 960
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.69149399926968e-05,
+      "loss": 1.1307,
+      "step": 965
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.688056000681523e-05,
+      "loss": 1.1388,
+      "step": 970
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.6846024878098738e-05,
+      "loss": 1.1526,
+      "step": 975
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.6811335385245247e-05,
+      "loss": 1.1474,
+      "step": 980
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.6776492310433278e-05,
+      "loss": 1.131,
+      "step": 985
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.6741496439304315e-05,
+      "loss": 1.1343,
+      "step": 990
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.6706348560945107e-05,
+      "loss": 1.1503,
+      "step": 995
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.6671049467869844e-05,
+      "loss": 1.1247,
+      "step": 1000
+    },
+    {
+      "epoch": 0.88,
+      "eval_loss": 1.1437697410583496,
+      "eval_runtime": 426.4692,
+      "eval_samples_per_second": 37.907,
+      "eval_steps_per_second": 1.186,
+      "step": 1000
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.6635599956002307e-05,
+      "loss": 1.1257,
+      "step": 1005
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.6600000824657935e-05,
+      "loss": 1.1317,
+      "step": 1010
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.6564252876525766e-05,
+      "loss": 1.1264,
+      "step": 1015
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.6528356917650378e-05,
+      "loss": 1.1238,
+      "step": 1020
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.6492313757413673e-05,
+      "loss": 1.1424,
+      "step": 1025
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.645612420851667e-05,
+      "loss": 1.1222,
+      "step": 1030
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.641978908696114e-05,
+      "loss": 1.133,
+      "step": 1035
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.638330921203124e-05,
+      "loss": 1.1423,
+      "step": 1040
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.634668540627502e-05,
+      "loss": 1.1403,
+      "step": 1045
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.6309918495485875e-05,
+      "loss": 1.1327,
+      "step": 1050
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.6273009308683945e-05,
+      "loss": 1.1269,
+      "step": 1055
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.6235958678097396e-05,
+      "loss": 1.1708,
+      "step": 1060
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.6198767439143677e-05,
+      "loss": 1.1402,
+      "step": 1065
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.616143643041067e-05,
+      "loss": 1.13,
+      "step": 1070
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.6123966493637785e-05,
+      "loss": 1.125,
+      "step": 1075
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.608635847369698e-05,
+      "loss": 1.1341,
+      "step": 1080
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.6048613218573718e-05,
+      "loss": 1.12,
+      "step": 1085
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.6010731579347833e-05,
+      "loss": 1.1188,
+      "step": 1090
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.5972714410174345e-05,
+      "loss": 1.1362,
+      "step": 1095
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.5934562568264214e-05,
+      "loss": 1.1485,
+      "step": 1100
+    },
+    {
+      "epoch": 0.96,
+      "eval_loss": 1.1392369270324707,
+      "eval_runtime": 427.0259,
+      "eval_samples_per_second": 37.857,
+      "eval_steps_per_second": 1.185,
+      "step": 1100
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.5896276913864985e-05,
+      "loss": 1.1422,
+      "step": 1105
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.5857858310241424e-05,
+      "loss": 1.1151,
+      "step": 1110
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.5819307623656022e-05,
+      "loss": 1.1098,
+      "step": 1115
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.578062572334948e-05,
+      "loss": 1.1257,
+      "step": 1120
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.5741813481521108e-05,
+      "loss": 1.1198,
+      "step": 1125
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.5702871773309147e-05,
+      "loss": 1.1381,
+      "step": 1130
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.5663801476771058e-05,
+      "loss": 1.1263,
+      "step": 1135
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.5624603472863703e-05,
+      "loss": 1.137,
+      "step": 1140
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.558527864542349e-05,
+      "loss": 1.1037,
+      "step": 1145
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.554582788114645e-05,
+      "loss": 1.0475,
+      "step": 1150
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5506252069568225e-05,
+      "loss": 1.0786,
+      "step": 1155
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5466552103044035e-05,
+      "loss": 1.0866,
+      "step": 1160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.542672887672854e-05,
+      "loss": 1.0706,
+      "step": 1165
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5386783288555655e-05,
+      "loss": 1.0856,
+      "step": 1170
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.534671623921832e-05,
+      "loss": 1.0812,
+      "step": 1175
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.530652863214818e-05,
+      "loss": 1.0815,
+      "step": 1180
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.52662213734952e-05,
+      "loss": 1.0796,
+      "step": 1185
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.5225795372107275e-05,
+      "loss": 1.0812,
+      "step": 1190
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.5185251539509685e-05,
+      "loss": 1.0921,
+      "step": 1195
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.5144590789884581e-05,
+      "loss": 1.067,
+      "step": 1200
+    },
+    {
+      "epoch": 1.05,
+      "eval_loss": 1.138688564300537,
+      "eval_runtime": 427.7454,
+      "eval_samples_per_second": 37.794,
+      "eval_steps_per_second": 1.183,
+      "step": 1200
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.5103814040050358e-05,
+      "loss": 1.0931,
+      "step": 1205
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.5062922209440976e-05,
+      "loss": 1.0611,
+      "step": 1210
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.5021916220085246e-05,
+      "loss": 1.0814,
+      "step": 1215
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.4980796996586024e-05,
+      "loss": 1.0766,
+      "step": 1220
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.4939565466099365e-05,
+      "loss": 1.0847,
+      "step": 1225
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4898222558313633e-05,
+      "loss": 1.068,
+      "step": 1230
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4856769205428514e-05,
+      "loss": 1.066,
+      "step": 1235
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.4815206342134017e-05,
+      "loss": 1.0786,
+      "step": 1240
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.4773534905589388e-05,
+      "loss": 1.0753,
+      "step": 1245
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.473175583540199e-05,
+      "loss": 1.0793,
+      "step": 1250
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4689870073606093e-05,
+      "loss": 1.0931,
+      "step": 1255
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.464787856464167e-05,
+      "loss": 1.0829,
+      "step": 1260
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.4605782255333058e-05,
+      "loss": 1.0664,
+      "step": 1265
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.4563582094867649e-05,
+      "loss": 1.0578,
+      "step": 1270
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.4521279034774465e-05,
+      "loss": 1.0741,
+      "step": 1275
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.4478874028902702e-05,
+      "loss": 1.0875,
+      "step": 1280
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.443636803340024e-05,
+      "loss": 1.0798,
+      "step": 1285
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.4393762006692065e-05,
+      "loss": 1.0807,
+      "step": 1290
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.435105690945867e-05,
+      "loss": 1.0673,
+      "step": 1295
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4308253704614386e-05,
+      "loss": 1.0694,
+      "step": 1300
+    },
+    {
+      "epoch": 1.14,
+      "eval_loss": 1.1368060111999512,
+      "eval_runtime": 425.836,
+      "eval_samples_per_second": 37.963,
+      "eval_steps_per_second": 1.188,
+      "step": 1300
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4265353357285675e-05,
+      "loss": 1.0752,
+      "step": 1305
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.422235683478937e-05,
+      "loss": 1.0621,
+      "step": 1310
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.4179265106610863e-05,
+      "loss": 1.0819,
+      "step": 1315
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4136079144382234e-05,
+      "loss": 1.0662,
+      "step": 1320
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4092799921860364e-05,
+      "loss": 1.0783,
+      "step": 1325
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.404942841490495e-05,
+      "loss": 1.0638,
+      "step": 1330
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.400596560145654e-05,
+      "loss": 1.0746,
+      "step": 1335
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.3962412461514446e-05,
+      "loss": 1.0756,
+      "step": 1340
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.3918769977114655e-05,
+      "loss": 1.0994,
+      "step": 1345
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.3875039132307712e-05,
+      "loss": 1.0576,
+      "step": 1350
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.3831220913136503e-05,
+      "loss": 1.0816,
+      "step": 1355
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.3787316307614021e-05,
+      "loss": 1.0631,
+      "step": 1360
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3743326305701108e-05,
+      "loss": 1.0889,
+      "step": 1365
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3699251899284127e-05,
+      "loss": 1.0756,
+      "step": 1370
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3655094082152585e-05,
+      "loss": 1.0798,
+      "step": 1375
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.3610853849976736e-05,
+      "loss": 1.0778,
+      "step": 1380
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.356653220028513e-05,
+      "loss": 1.0706,
+      "step": 1385
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.3522130132442115e-05,
+      "loss": 1.0733,
+      "step": 1390
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.3477648647625312e-05,
+      "loss": 1.0749,
+      "step": 1395
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.3433088748803033e-05,
+      "loss": 1.0814,
+      "step": 1400
+    },
+    {
+      "epoch": 1.23,
+      "eval_loss": 1.1340500116348267,
+      "eval_runtime": 427.2423,
+      "eval_samples_per_second": 37.838,
+      "eval_steps_per_second": 1.184,
+      "step": 1400
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.3388451440711671e-05,
+      "loss": 1.0715,
+      "step": 1405
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.3343737729833032e-05,
+      "loss": 1.1067,
+      "step": 1410
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.3298948624371669e-05,
+      "loss": 1.0794,
+      "step": 1415
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.325408513423212e-05,
+      "loss": 1.079,
+      "step": 1420
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.320914827099615e-05,
+      "loss": 1.0811,
+      "step": 1425
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3164139047899942e-05,
+      "loss": 1.0649,
+      "step": 1430
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.3119058479811254e-05,
+      "loss": 1.083,
+      "step": 1435
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.3073907583206514e-05,
+      "loss": 1.0655,
+      "step": 1440
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.302868737614793e-05,
+      "loss": 1.0796,
+      "step": 1445
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.298339887826053e-05,
+      "loss": 1.0576,
+      "step": 1450
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.2938043110709146e-05,
+      "loss": 1.0742,
+      "step": 1455
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.2892621096175412e-05,
+      "loss": 1.0828,
+      "step": 1460
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.2847133858834701e-05,
+      "loss": 1.0714,
+      "step": 1465
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.280158242433303e-05,
+      "loss": 1.0664,
+      "step": 1470
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.2755967819763929e-05,
+      "loss": 1.0777,
+      "step": 1475
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.271029107364529e-05,
+      "loss": 1.0622,
+      "step": 1480
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.266455321589617e-05,
+      "loss": 1.0545,
+      "step": 1485
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.2618755277813575e-05,
+      "loss": 1.0691,
+      "step": 1490
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.2572898292049195e-05,
+      "loss": 1.0776,
+      "step": 1495
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.2526983292586136e-05,
+      "loss": 1.0727,
+      "step": 1500
+    },
+    {
+      "epoch": 1.31,
+      "eval_loss": 1.131594181060791,
+      "eval_runtime": 426.5085,
+      "eval_samples_per_second": 37.903,
+      "eval_steps_per_second": 1.186,
+      "step": 1500
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.248101131471559e-05,
+      "loss": 1.0636,
+      "step": 1505
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.2434983395013495e-05,
+      "loss": 1.0831,
+      "step": 1510
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2388900571317175e-05,
+      "loss": 1.085,
+      "step": 1515
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2342763882701922e-05,
+      "loss": 1.0965,
+      "step": 1520
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2296574369457575e-05,
+      "loss": 1.0723,
+      "step": 1525
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2250333073065066e-05,
+      "loss": 1.076,
+      "step": 1530
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2204041036172924e-05,
+      "loss": 1.0774,
+      "step": 1535
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.2157699302573787e-05,
+      "loss": 1.0683,
+      "step": 1540
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.2111308917180844e-05,
+      "loss": 1.0596,
+      "step": 1545
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.206487092600429e-05,
+      "loss": 1.0696,
+      "step": 1550
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.2018386376127733e-05,
+      "loss": 1.0732,
+      "step": 1555
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.197185631568459e-05,
+      "loss": 1.0894,
+      "step": 1560
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.192528179383444e-05,
+      "loss": 1.0723,
+      "step": 1565
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1878663860739388e-05,
+      "loss": 1.0746,
+      "step": 1570
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1832003567540372e-05,
+      "loss": 1.0837,
+      "step": 1575
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1785301966333465e-05,
+      "loss": 1.0849,
+      "step": 1580
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1738560110146154e-05,
+      "loss": 1.0658,
+      "step": 1585
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1691779052913592e-05,
+      "loss": 1.0834,
+      "step": 1590
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.1644959849454842e-05,
+      "loss": 1.0743,
+      "step": 1595
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.159810355544908e-05,
+      "loss": 1.0769,
+      "step": 1600
+    },
+    {
+      "epoch": 1.4,
+      "eval_loss": 1.129186749458313,
+      "eval_runtime": 426.3787,
+      "eval_samples_per_second": 37.915,
+      "eval_steps_per_second": 1.187,
+      "step": 1600
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.1551211227411811e-05,
+      "loss": 1.0804,
+      "step": 1605
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.1504283922671027e-05,
+      "loss": 1.0684,
+      "step": 1610
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.1457322699343371e-05,
+      "loss": 1.0753,
+      "step": 1615
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1410328616310293e-05,
+      "loss": 1.0641,
+      "step": 1620
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.136330273319416e-05,
+      "loss": 1.0814,
+      "step": 1625
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.1316246110334357e-05,
+      "loss": 1.0738,
+      "step": 1630
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.12691598087634e-05,
+      "loss": 1.0674,
+      "step": 1635
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.1222044890183003e-05,
+      "loss": 1.0688,
+      "step": 1640
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.117490241694012e-05,
+      "loss": 1.0685,
+      "step": 1645
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1127733452003027e-05,
+      "loss": 1.0736,
+      "step": 1650
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1080539058937323e-05,
+      "loss": 1.0623,
+      "step": 1655
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1033320301881958e-05,
+      "loss": 1.0653,
+      "step": 1660
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.0986078245525249e-05,
+      "loss": 1.0869,
+      "step": 1665
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.093881395508086e-05,
+      "loss": 1.0709,
+      "step": 1670
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0891528496263797e-05,
+      "loss": 1.0689,
+      "step": 1675
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0844222935266356e-05,
+      "loss": 1.0763,
+      "step": 1680
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0796898338734116e-05,
+      "loss": 1.0542,
+      "step": 1685
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0749555773741851e-05,
+      "loss": 1.0675,
+      "step": 1690
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0702196307769497e-05,
+      "loss": 1.0559,
+      "step": 1695
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.0654821008678072e-05,
+      "loss": 1.0728,
+      "step": 1700
+    },
+    {
+      "epoch": 1.49,
+      "eval_loss": 1.1270054578781128,
+      "eval_runtime": 425.1702,
+      "eval_samples_per_second": 38.022,
+      "eval_steps_per_second": 1.19,
+      "step": 1700
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.0607430944685602e-05,
+      "loss": 1.0686,
+      "step": 1705
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.0560027184343024e-05,
+      "loss": 1.072,
+      "step": 1710
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.0512610796510106e-05,
+      "loss": 1.0572,
+      "step": 1715
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.0465182850331343e-05,
+      "loss": 1.0861,
+      "step": 1720
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.0417744415211841e-05,
+      "loss": 1.0797,
+      "step": 1725
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.0370296560793213e-05,
+      "loss": 1.0759,
+      "step": 1730
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.0322840356929464e-05,
+      "loss": 1.0598,
+      "step": 1735
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.0275376873662853e-05,
+      "loss": 1.0584,
+      "step": 1740
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.0227907181199777e-05,
+      "loss": 1.0831,
+      "step": 1745
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.0180432349886645e-05,
+      "loss": 1.0547,
+      "step": 1750
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.0132953450185727e-05,
+      "loss": 1.0726,
+      "step": 1755
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.0085471552651037e-05,
+      "loss": 1.083,
+      "step": 1760
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.0037987727904178e-05,
+      "loss": 1.0723,
+      "step": 1765
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.990503046610206e-06,
+      "loss": 1.056,
+      "step": 1770
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.943018579453491e-06,
+      "loss": 1.0764,
+      "step": 1775
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.895535397113585e-06,
+      "loss": 1.0795,
+      "step": 1780
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.848054570241058e-06,
+      "loss": 1.0741,
+      "step": 1785
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.800577169433372e-06,
+      "loss": 1.0629,
+      "step": 1790
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.753104265210735e-06,
+      "loss": 1.0681,
+      "step": 1795
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.705636927991976e-06,
+      "loss": 1.0558,
+      "step": 1800
+    },
+    {
+      "epoch": 1.58,
+      "eval_loss": 1.124674916267395,
+      "eval_runtime": 425.5493,
+      "eval_samples_per_second": 37.989,
+      "eval_steps_per_second": 1.189,
+      "step": 1800
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.658176228070393e-06,
+      "loss": 1.0708,
+      "step": 1805
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.610723235589621e-06,
+      "loss": 1.053,
+      "step": 1810
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.56327902051952e-06,
+      "loss": 1.0571,
+      "step": 1815
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.515844652632021e-06,
+      "loss": 1.0861,
+      "step": 1820
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.468421201477039e-06,
+      "loss": 1.0576,
+      "step": 1825
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.421009736358324e-06,
+      "loss": 1.053,
+      "step": 1830
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.373611326309368e-06,
+      "loss": 1.0871,
+      "step": 1835
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.32622704006931e-06,
+      "loss": 1.0644,
+      "step": 1840
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.278857946058804e-06,
+      "loss": 1.0786,
+      "step": 1845
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.231505112355966e-06,
+      "loss": 1.0498,
+      "step": 1850
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.184169606672266e-06,
+      "loss": 1.0578,
+      "step": 1855
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.136852496328469e-06,
+      "loss": 1.0573,
+      "step": 1860
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.089554848230556e-06,
+      "loss": 1.0679,
+      "step": 1865
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.042277728845673e-06,
+      "loss": 1.0688,
+      "step": 1870
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 8.995022204178096e-06,
+      "loss": 1.0545,
+      "step": 1875
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.947789339745165e-06,
+      "loss": 1.0703,
+      "step": 1880
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.900580200553295e-06,
+      "loss": 1.0487,
+      "step": 1885
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.85339585107393e-06,
+      "loss": 1.067,
+      "step": 1890
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.806237355219563e-06,
+      "loss": 1.0533,
+      "step": 1895
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.759105776319737e-06,
+      "loss": 1.0753,
+      "step": 1900
+    },
+    {
+      "epoch": 1.66,
+      "eval_loss": 1.1228857040405273,
+      "eval_runtime": 426.9078,
+      "eval_samples_per_second": 37.868,
+      "eval_steps_per_second": 1.185,
+      "step": 1900
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.712002177097079e-06,
+      "loss": 1.0754,
+      "step": 1905
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.66492761964332e-06,
+      "loss": 1.0426,
+      "step": 1910
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.61788316539536e-06,
+      "loss": 1.0496,
+      "step": 1915
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.57086987511134e-06,
+      "loss": 1.0545,
+      "step": 1920
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.523888808846705e-06,
+      "loss": 1.069,
+      "step": 1925
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.47694102593032e-06,
+      "loss": 1.0474,
+      "step": 1930
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.430027584940572e-06,
+      "loss": 1.0719,
+      "step": 1935
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.383149543681511e-06,
+      "loss": 1.0891,
+      "step": 1940
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.336307959158997e-06,
+      "loss": 1.066,
+      "step": 1945
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.28950388755685e-06,
+      "loss": 1.0524,
+      "step": 1950
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.242738384213071e-06,
+      "loss": 1.0692,
+      "step": 1955
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.196012503596e-06,
+      "loss": 1.0545,
+      "step": 1960
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.149327299280576e-06,
+      "loss": 1.0739,
+      "step": 1965
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.102683823924575e-06,
+      "loss": 1.0435,
+      "step": 1970
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.056083129244857e-06,
+      "loss": 1.0817,
+      "step": 1975
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.00952626599367e-06,
+      "loss": 1.0556,
+      "step": 1980
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.963014283934944e-06,
+      "loss": 1.0693,
+      "step": 1985
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.916548231820642e-06,
+      "loss": 1.0732,
+      "step": 1990
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.870129157367079e-06,
+      "loss": 1.0674,
+      "step": 1995
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.823758107231338e-06,
+      "loss": 1.0799,
+      "step": 2000
+    },
+    {
+      "epoch": 1.75,
+      "eval_loss": 1.1209245920181274,
+      "eval_runtime": 427.1001,
+      "eval_samples_per_second": 37.851,
+      "eval_steps_per_second": 1.185,
+      "step": 2000
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.777436126987642e-06,
+      "loss": 1.0638,
+      "step": 2005
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.731164261103777e-06,
+      "loss": 1.0617,
+      "step": 2010
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.684943552917569e-06,
+      "loss": 1.0698,
+      "step": 2015
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.638775044613324e-06,
+      "loss": 1.072,
+      "step": 2020
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.592659777198354e-06,
+      "loss": 1.0673,
+      "step": 2025
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.546598790479489e-06,
+      "loss": 1.0734,
+      "step": 2030
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.500593123039649e-06,
+      "loss": 1.067,
+      "step": 2035
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.4546438122144085e-06,
+      "loss": 1.0486,
+      "step": 2040
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.408751894068607e-06,
+      "loss": 1.063,
+      "step": 2045
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.362918403373007e-06,
+      "loss": 1.0486,
+      "step": 2050
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.31714437358094e-06,
+      "loss": 1.0702,
+      "step": 2055
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.271430836805018e-06,
+      "loss": 1.0682,
+      "step": 2060
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.225778823793853e-06,
+      "loss": 1.0414,
+      "step": 2065
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.180189363908818e-06,
+      "loss": 1.0621,
+      "step": 2070
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.134663485100852e-06,
+      "loss": 1.0796,
+      "step": 2075
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.089202213887248e-06,
+      "loss": 1.0749,
+      "step": 2080
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 7.043806575328542e-06,
+      "loss": 1.078,
+      "step": 2085
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 6.998477593005377e-06,
+      "loss": 1.0667,
+      "step": 2090
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.95321628899544e-06,
+      "loss": 1.0593,
+      "step": 2095
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.908023683850398e-06,
+      "loss": 1.066,
+      "step": 2100
+    },
+    {
+      "epoch": 1.84,
+      "eval_loss": 1.1192409992218018,
+      "eval_runtime": 426.8527,
+      "eval_samples_per_second": 37.873,
+      "eval_steps_per_second": 1.185,
+      "step": 2100
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.8629007965729e-06,
+      "loss": 1.0725,
+      "step": 2105
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.817848644593603e-06,
+      "loss": 1.0578,
+      "step": 2110
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.7728682437482165e-06,
+      "loss": 1.0892,
+      "step": 2115
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.727960608254611e-06,
+      "loss": 1.0551,
+      "step": 2120
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.683126750689934e-06,
+      "loss": 1.0516,
+      "step": 2125
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.63836768196781e-06,
+      "loss": 1.0651,
+      "step": 2130
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.5936844113155e-06,
+      "loss": 1.0722,
+      "step": 2135
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.549077946251186e-06,
+      "loss": 1.0621,
+      "step": 2140
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.50454929256124e-06,
+      "loss": 1.0523,
+      "step": 2145
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.460099454277525e-06,
+      "loss": 1.0682,
+      "step": 2150
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.4157294336548e-06,
+      "loss": 1.0714,
+      "step": 2155
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.371440231148075e-06,
+      "loss": 1.047,
+      "step": 2160
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.3272328453900856e-06,
+      "loss": 1.0659,
+      "step": 2165
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.283108273168753e-06,
+      "loss": 1.0567,
+      "step": 2170
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.2390675094047326e-06,
+      "loss": 1.082,
+      "step": 2175
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.1951115471289615e-06,
+      "loss": 1.0704,
+      "step": 2180
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.1512413774602634e-06,
+      "loss": 1.0697,
+      "step": 2185
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.1074579895830275e-06,
+      "loss": 1.0588,
+      "step": 2190
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.063762370724876e-06,
+      "loss": 1.0697,
+      "step": 2195
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 6.020155506134416e-06,
+      "loss": 1.0406,
+      "step": 2200
+    },
+    {
+      "epoch": 1.93,
+      "eval_loss": 1.1177691221237183,
+      "eval_runtime": 425.6505,
+      "eval_samples_per_second": 37.98,
+      "eval_steps_per_second": 1.189,
+      "step": 2200
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.976638379059034e-06,
+      "loss": 1.0813,
+      "step": 2205
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.933211970722703e-06,
+      "loss": 1.0589,
+      "step": 2210
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.8898772603038775e-06,
+      "loss": 1.0653,
+      "step": 2215
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.846635224913402e-06,
+      "loss": 1.076,
+      "step": 2220
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.803486839572497e-06,
+      "loss": 1.0569,
+      "step": 2225
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.760433077190741e-06,
+      "loss": 1.0627,
+      "step": 2230
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.717474908544176e-06,
+      "loss": 1.0598,
+      "step": 2235
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.674613302253391e-06,
+      "loss": 1.085,
+      "step": 2240
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.631849224761681e-06,
+      "loss": 1.07,
+      "step": 2245
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.589183640313264e-06,
+      "loss": 1.0633,
+      "step": 2250
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.546617510931535e-06,
+      "loss": 1.0765,
+      "step": 2255
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.504151796397392e-06,
+      "loss": 1.0606,
+      "step": 2260
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.461787454227558e-06,
+      "loss": 1.0817,
+      "step": 2265
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.419525439653025e-06,
+      "loss": 1.0665,
+      "step": 2270
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.3773667055975066e-06,
+      "loss": 1.0536,
+      "step": 2275
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 5.335312202655936e-06,
+      "loss": 1.0574,
+      "step": 2280
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 5.29336287907305e-06,
+      "loss": 1.0435,
+      "step": 2285
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 5.251519680721999e-06,
+      "loss": 1.0025,
+      "step": 2290
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 5.209783551083028e-06,
+      "loss": 0.9986,
+      "step": 2295
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 5.168155431222188e-06,
+      "loss": 1.0193,
+      "step": 2300
+    },
+    {
+      "epoch": 2.01,
+      "eval_loss": 1.1221781969070435,
+      "eval_runtime": 426.9687,
+      "eval_samples_per_second": 37.862,
+      "eval_steps_per_second": 1.185,
+      "step": 2300
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 5.126636259770135e-06,
+      "loss": 1.0106,
+      "step": 2305
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 5.085226972900961e-06,
+      "loss": 1.0173,
+      "step": 2310
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 5.043928504311058e-06,
+      "loss": 1.0212,
+      "step": 2315
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 5.002741785198111e-06,
+      "loss": 1.0217,
+      "step": 2320
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 4.961667744240069e-06,
+      "loss": 1.0155,
+      "step": 2325
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 4.920707307574218e-06,
+      "loss": 1.0196,
+      "step": 2330
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 4.879861398776287e-06,
+      "loss": 1.0074,
+      "step": 2335
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 4.839130938839645e-06,
+      "loss": 1.0258,
+      "step": 2340
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 4.798516846154508e-06,
+      "loss": 1.0171,
+      "step": 2345
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 4.758020036487247e-06,
+      "loss": 1.0156,
+      "step": 2350
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 4.717641422959741e-06,
+      "loss": 1.0213,
+      "step": 2355
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 4.677381916028775e-06,
+      "loss": 1.0167,
+      "step": 2360
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 4.637242423465529e-06,
+      "loss": 1.0171,
+      "step": 2365
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 4.597223850335088e-06,
+      "loss": 1.0137,
+      "step": 2370
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 4.557327098976059e-06,
+      "loss": 1.0048,
+      "step": 2375
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 4.5175530689802e-06,
+      "loss": 1.0229,
+      "step": 2380
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 4.477902657172153e-06,
+      "loss": 1.027,
+      "step": 2385
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 4.438376757589223e-06,
+      "loss": 1.0161,
+      "step": 2390
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 4.398976261461205e-06,
+      "loss": 1.0052,
+      "step": 2395
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 4.359702057190307e-06,
+      "loss": 1.0276,
+      "step": 2400
+    },
+    {
+      "epoch": 2.1,
+      "eval_loss": 1.1220041513442993,
+      "eval_runtime": 426.2423,
+      "eval_samples_per_second": 37.927,
+      "eval_steps_per_second": 1.187,
+      "step": 2400
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 4.3205550303310996e-06,
+      "loss": 1.0106,
+      "step": 2405
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 4.28153606357056e-06,
+      "loss": 1.0026,
+      "step": 2410
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 4.242646036708174e-06,
+      "loss": 1.0218,
+      "step": 2415
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 4.203885826636081e-06,
+      "loss": 1.0281,
+      "step": 2420
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 4.165256307319321e-06,
+      "loss": 1.0198,
+      "step": 2425
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 4.126758349776109e-06,
+      "loss": 1.0209,
+      "step": 2430
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 4.08839282205822e-06,
+      "loss": 1.0044,
+      "step": 2435
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 4.0501605892313915e-06,
+      "loss": 1.0208,
+      "step": 2440
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 4.012062513355824e-06,
+      "loss": 1.0162,
+      "step": 2445
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 3.9740994534667645e-06,
+      "loss": 1.0009,
+      "step": 2450
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 3.936272265555101e-06,
+      "loss": 1.0148,
+      "step": 2455
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 3.898581802548099e-06,
+      "loss": 1.0043,
+      "step": 2460
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 3.861028914290135e-06,
+      "loss": 1.0332,
+      "step": 2465
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 3.8236144475235635e-06,
+      "loss": 1.0126,
+      "step": 2470
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 3.786339245869606e-06,
+      "loss": 1.0256,
+      "step": 2475
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 3.749204149809329e-06,
+      "loss": 1.0061,
+      "step": 2480
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 3.712209996664713e-06,
+      "loss": 1.0244,
+      "step": 2485
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 3.675357620579745e-06,
+      "loss": 0.9962,
+      "step": 2490
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 3.6386478525016346e-06,
+      "loss": 1.013,
+      "step": 2495
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 3.6020815201620574e-06,
+      "loss": 1.0171,
+      "step": 2500
+    },
+    {
+      "epoch": 2.19,
+      "eval_loss": 1.121544361114502,
+      "eval_runtime": 425.7468,
+      "eval_samples_per_second": 37.971,
+      "eval_steps_per_second": 1.188,
+      "step": 2500
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 3.5656594480585015e-06,
+      "loss": 1.0092,
+      "step": 2505
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 3.529382457435686e-06,
+      "loss": 1.0214,
+      "step": 2510
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 3.4932513662670175e-06,
+      "loss": 1.0146,
+      "step": 2515
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 3.457266989236181e-06,
+      "loss": 1.0051,
+      "step": 2520
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 3.421430137718732e-06,
+      "loss": 1.0251,
+      "step": 2525
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 3.3857416197638394e-06,
+      "loss": 1.0321,
+      "step": 2530
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 3.350202240076035e-06,
+      "loss": 1.0146,
+      "step": 2535
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 3.314812799997085e-06,
+      "loss": 1.0124,
+      "step": 2540
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 3.279574097487923e-06,
+      "loss": 1.0024,
+      "step": 2545
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 3.244486927110645e-06,
+      "loss": 1.0147,
+      "step": 2550
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 3.2095520800106095e-06,
+      "loss": 1.0273,
+      "step": 2555
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 3.174770343898581e-06,
+      "loss": 0.9955,
+      "step": 2560
+    },
+    {
+      "epoch": 2.25,
+      "learning_rate": 3.14014250303299e-06,
+      "loss": 1.0268,
+      "step": 2565
+    },
+    {
+      "epoch": 2.25,
+      "learning_rate": 3.1056693382022208e-06,
+      "loss": 1.0115,
+      "step": 2570
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 3.0713516267070375e-06,
+      "loss": 1.0007,
+      "step": 2575
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 3.037190142343044e-06,
+      "loss": 1.0164,
+      "step": 2580
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 3.003185655383222e-06,
+      "loss": 1.0246,
+      "step": 2585
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 2.9693389325605904e-06,
+      "loss": 1.0153,
+      "step": 2590
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 2.935650737050897e-06,
+      "loss": 0.9947,
+      "step": 2595
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 2.9021218284554154e-06,
+      "loss": 1.0112,
+      "step": 2600
+    },
+    {
+      "epoch": 2.28,
+      "eval_loss": 1.1211416721343994,
+      "eval_runtime": 426.0338,
+      "eval_samples_per_second": 37.945,
+      "eval_steps_per_second": 1.188,
+      "step": 2600
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 2.8687529627838174e-06,
+      "loss": 1.0227,
+      "step": 2605
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 2.835544892437133e-06,
+      "loss": 1.0153,
+      "step": 2610
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 2.8024983661907803e-06,
+      "loss": 1.0114,
+      "step": 2615
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 2.7696141291776737e-06,
+      "loss": 1.0165,
+      "step": 2620
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 2.7368929228714423e-06,
+      "loss": 0.9873,
+      "step": 2625
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 2.704335485069688e-06,
+      "loss": 1.0195,
+      "step": 2630
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 2.671942549877369e-06,
+      "loss": 1.0156,
+      "step": 2635
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 2.6397148476902324e-06,
+      "loss": 1.017,
+      "step": 2640
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 2.6076531051783594e-06,
+      "loss": 1.0145,
+      "step": 2645
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 2.575758045269775e-06,
+      "loss": 1.009,
+      "step": 2650
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 2.5440303871341365e-06,
+      "loss": 1.0121,
+      "step": 2655
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 2.5124708461665326e-06,
+      "loss": 1.0255,
+      "step": 2660
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 2.481080133971342e-06,
+      "loss": 1.0191,
+      "step": 2665
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 2.4498589583461998e-06,
+      "loss": 1.0386,
+      "step": 2670
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 2.4188080232660327e-06,
+      "loss": 1.009,
+      "step": 2675
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 2.3879280288671725e-06,
+      "loss": 1.0187,
+      "step": 2680
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 2.357219671431594e-06,
+      "loss": 1.0017,
+      "step": 2685
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 2.326683643371197e-06,
+      "loss": 1.012,
+      "step": 2690
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 2.296320633212198e-06,
+      "loss": 1.022,
+      "step": 2695
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 2.266131325579606e-06,
+      "loss": 1.0087,
+      "step": 2700
+    },
+    {
+      "epoch": 2.37,
+      "eval_loss": 1.1207467317581177,
+      "eval_runtime": 426.3134,
+      "eval_samples_per_second": 37.92,
+      "eval_steps_per_second": 1.187,
+      "step": 2700
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 2.2361164011817904e-06,
+      "loss": 1.0097,
+      "step": 2705
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 2.20627653679513e-06,
+      "loss": 1.0224,
+      "step": 2710
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 2.1766124052487436e-06,
+      "loss": 1.0247,
+      "step": 2715
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 2.1471246754093333e-06,
+      "loss": 1.0146,
+      "step": 2720
+    },
+    {
+      "epoch": 2.39,
+      "learning_rate": 2.117814012166094e-06,
+      "loss": 1.026,
+      "step": 2725
+    },
+    {
+      "epoch": 2.39,
+      "learning_rate": 2.0886810764157216e-06,
+      "loss": 1.0186,
+      "step": 2730
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 2.059726525047515e-06,
+      "loss": 1.0157,
+      "step": 2735
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 2.0309510109285633e-06,
+      "loss": 1.0008,
+      "step": 2740
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 2.0023551828890296e-06,
+      "loss": 1.0221,
+      "step": 2745
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 1.9739396857075077e-06,
+      "loss": 1.0119,
+      "step": 2750
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 1.9457051600964948e-06,
+      "loss": 1.0127,
+      "step": 2755
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 1.9176522426879395e-06,
+      "loss": 1.018,
+      "step": 2760
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 1.8897815660189e-06,
+      "loss": 1.0266,
+      "step": 2765
+    },
+    {
+      "epoch": 2.43,
+      "learning_rate": 1.8620937585172571e-06,
+      "loss": 1.0075,
+      "step": 2770
+    },
+    {
+      "epoch": 2.43,
+      "learning_rate": 1.8345894444875668e-06,
+      "loss": 1.0179,
+      "step": 2775
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 1.8072692440969786e-06,
+      "loss": 1.0182,
+      "step": 2780
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 1.780133773361239e-06,
+      "loss": 1.0154,
+      "step": 2785
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 1.7531836441308147e-06,
+      "loss": 1.0012,
+      "step": 2790
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 1.7264194640770937e-06,
+      "loss": 1.0101,
+      "step": 2795
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 1.6998418366786872e-06,
+      "loss": 1.0158,
+      "step": 2800
+    },
+    {
+      "epoch": 2.45,
+      "eval_loss": 1.1203794479370117,
+      "eval_runtime": 425.5194,
+      "eval_samples_per_second": 37.991,
+      "eval_steps_per_second": 1.189,
+      "step": 2800
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 1.6734513612078075e-06,
+      "loss": 1.0146,
+      "step": 2805
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 1.647248632716776e-06,
+      "loss": 1.0203,
+      "step": 2810
+    },
+    {
+      "epoch": 2.47,
+      "learning_rate": 1.6212342420245963e-06,
+      "loss": 1.0231,
+      "step": 2815
+    },
+    {
+      "epoch": 2.47,
+      "learning_rate": 1.5954087757036196e-06,
+      "loss": 1.0136,
+      "step": 2820
+    },
+    {
+      "epoch": 2.47,
+      "learning_rate": 1.5697728160663473e-06,
+      "loss": 1.0032,
+      "step": 2825
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 1.5443269411522722e-06,
+      "loss": 1.0221,
+      "step": 2830
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 1.5190717247148678e-06,
+      "loss": 1.0323,
+      "step": 2835
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 1.4940077362086324e-06,
+      "loss": 1.0261,
+      "step": 2840
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 1.4691355407762663e-06,
+      "loss": 1.0107,
+      "step": 2845
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 1.4444556992359127e-06,
+      "loss": 1.0204,
+      "step": 2850
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 1.4199687680685236e-06,
+      "loss": 1.0248,
+      "step": 2855
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 1.3956752994053102e-06,
+      "loss": 1.0003,
+      "step": 2860
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 1.371575841015289e-06,
+      "loss": 1.0359,
+      "step": 2865
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 1.3476709362929375e-06,
+      "loss": 1.014,
+      "step": 2870
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 1.3239611242459328e-06,
+      "loss": 0.9979,
+      "step": 2875
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 1.3004469394830089e-06,
+      "loss": 1.0152,
+      "step": 2880
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 1.2771289122018948e-06,
+      "loss": 1.0292,
+      "step": 2885
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 1.2540075681773557e-06,
+      "loss": 1.0347,
+      "step": 2890
+    },
+    {
+      "epoch": 2.54,
+      "learning_rate": 1.2310834287493546e-06,
+      "loss": 1.0203,
+      "step": 2895
+    },
+    {
+      "epoch": 2.54,
+      "learning_rate": 1.2083570108112764e-06,
+      "loss": 1.0219,
+      "step": 2900
+    },
+    {
+      "epoch": 2.54,
+      "eval_loss": 1.119923710823059,
+      "eval_runtime": 424.9645,
+      "eval_samples_per_second": 38.041,
+      "eval_steps_per_second": 1.191,
+      "step": 2900
+    },
+    {
+      "epoch": 2.54,
+      "learning_rate": 1.18582882679829e-06,
+      "loss": 1.0003,
+      "step": 2905
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 1.1634993846757803e-06,
+      "loss": 1.0112,
+      "step": 2910
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 1.1413691879279055e-06,
+      "loss": 1.0171,
+      "step": 2915
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 1.1194387355462389e-06,
+      "loss": 1.0143,
+      "step": 2920
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 1.0977085220185169e-06,
+      "loss": 1.0182,
+      "step": 2925
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 1.076179037317495e-06,
+      "loss": 1.0094,
+      "step": 2930
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 1.054850766889891e-06,
+      "loss": 1.0258,
+      "step": 2935
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 1.033724191645451e-06,
+      "loss": 1.0134,
+      "step": 2940
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 1.0127997879460916e-06,
+      "loss": 1.0129,
+      "step": 2945
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 9.920780275951703e-07,
+      "loss": 1.0007,
+      "step": 2950
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 9.71559377826844e-07,
+      "loss": 1.0153,
+      "step": 2955
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 9.512443012955286e-07,
+      "loss": 1.0032,
+      "step": 2960
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 9.311332560654806e-07,
+      "loss": 1.0144,
+      "step": 2965
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 9.11226695600449e-07,
+      "loss": 1.0304,
+      "step": 2970
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 8.915250687534693e-07,
+      "loss": 1.0236,
+      "step": 2975
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 8.720288197567306e-07,
+      "loss": 1.0137,
+      "step": 2980
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 8.527383882115625e-07,
+      "loss": 1.0259,
+      "step": 2985
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 8.336542090785282e-07,
+      "loss": 1.0137,
+      "step": 2990
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 8.147767126676076e-07,
+      "loss": 1.0194,
+      "step": 2995
+    },
+    {
+      "epoch": 2.63,
+      "learning_rate": 7.96106324628505e-07,
+      "loss": 1.0024,
+      "step": 3000
+    },
+    {
+      "epoch": 2.63,
+      "eval_loss": 1.1197476387023926,
+      "eval_runtime": 426.0129,
+      "eval_samples_per_second": 37.947,
+      "eval_steps_per_second": 1.188,
+      "step": 3000
+    },
+    {
+      "epoch": 2.63,
+      "learning_rate": 7.776434659410404e-07,
+      "loss": 1.0164,
+      "step": 3005
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 7.593885529056633e-07,
+      "loss": 1.0041,
+      "step": 3010
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 7.413419971340707e-07,
+      "loss": 1.0118,
+      "step": 3015
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 7.235042055399166e-07,
+      "loss": 1.0086,
+      "step": 3020
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 7.058755803296435e-07,
+      "loss": 1.0265,
+      "step": 3025
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 6.884565189934089e-07,
+      "loss": 1.005,
+      "step": 3030
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 6.712474142961289e-07,
+      "loss": 1.0006,
+      "step": 3035
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 6.542486542686155e-07,
+      "loss": 1.0181,
+      "step": 3040
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 6.374606221988266e-07,
+      "loss": 1.0163,
+      "step": 3045
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 6.208836966232356e-07,
+      "loss": 1.0124,
+      "step": 3050
+    },
+    {
+      "epoch": 2.68,
+      "learning_rate": 6.045182513182802e-07,
+      "loss": 1.024,
+      "step": 3055
+    },
+    {
+      "epoch": 2.68,
+      "learning_rate": 5.883646552919464e-07,
+      "loss": 1.0187,
+      "step": 3060
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 5.724232727754408e-07,
+      "loss": 1.031,
+      "step": 3065
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 5.566944632149851e-07,
+      "loss": 1.0107,
+      "step": 3070
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 5.411785812636983e-07,
+      "loss": 1.0061,
+      "step": 3075
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 5.258759767736176e-07,
+      "loss": 1.017,
+      "step": 3080
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 5.107869947877985e-07,
+      "loss": 1.0316,
+      "step": 3085
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 4.959119755325337e-07,
+      "loss": 1.0141,
+      "step": 3090
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 4.812512544096915e-07,
+      "loss": 1.0167,
+      "step": 3095
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 4.6680516198914004e-07,
+      "loss": 1.019,
+      "step": 3100
+    },
+    {
+      "epoch": 2.72,
+      "eval_loss": 1.11965012550354,
+      "eval_runtime": 426.3607,
+      "eval_samples_per_second": 37.916,
+      "eval_steps_per_second": 1.187,
+      "step": 3100
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 4.5257402400130657e-07,
+      "loss": 1.0237,
+      "step": 3105
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 4.385581613298195e-07,
+      "loss": 1.0134,
+      "step": 3110
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 4.2475789000428924e-07,
+      "loss": 1.0134,
+      "step": 3115
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 4.111735211931689e-07,
+      "loss": 1.0057,
+      "step": 3120
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 3.978053611967403e-07,
+      "loss": 1.0074,
+      "step": 3125
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 3.846537114402171e-07,
+      "loss": 1.0232,
+      "step": 3130
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 3.717188684669326e-07,
+      "loss": 1.0263,
+      "step": 3135
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 3.5900112393166707e-07,
+      "loss": 1.0025,
+      "step": 3140
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 3.4650076459406434e-07,
+      "loss": 1.0045,
+      "step": 3145
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 3.3421807231216905e-07,
+      "loss": 1.0212,
+      "step": 3150
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 3.221533240360708e-07,
+      "loss": 1.0244,
+      "step": 3155
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 3.103067918016556e-07,
+      "loss": 1.0024,
+      "step": 3160
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 2.986787427244775e-07,
+      "loss": 1.0014,
+      "step": 3165
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 2.872694389937325e-07,
+      "loss": 1.0153,
+      "step": 3170
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 2.7607913786634743e-07,
+      "loss": 1.0185,
+      "step": 3175
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 2.6510809166118257e-07,
+      "loss": 1.0194,
+      "step": 3180
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 2.5435654775333276e-07,
+      "loss": 1.0098,
+      "step": 3185
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 2.4382474856856296e-07,
+      "loss": 1.0244,
+      "step": 3190
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 2.3351293157783063e-07,
+      "loss": 1.0191,
+      "step": 3195
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 2.2342132929193737e-07,
+      "loss": 1.0135,
+      "step": 3200
+    },
+    {
+      "epoch": 2.8,
+      "eval_loss": 1.11944580078125,
+      "eval_runtime": 426.0948,
+      "eval_samples_per_second": 37.94,
+      "eval_steps_per_second": 1.188,
+      "step": 3200
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 2.135501692562836e-07,
+      "loss": 1.0034,
+      "step": 3205
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 2.0389967404573997e-07,
+      "loss": 1.0126,
+      "step": 3210
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 1.9447006125962952e-07,
+      "loss": 1.0073,
+      "step": 3215
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 1.852615435168148e-07,
+      "loss": 1.0135,
+      "step": 3220
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 1.7627432845091275e-07,
+      "loss": 1.0119,
+      "step": 3225
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 1.675086187056063e-07,
+      "loss": 1.0211,
+      "step": 3230
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 1.5896461193007694e-07,
+      "loss": 1.0098,
+      "step": 3235
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 1.5064250077454934e-07,
+      "loss": 1.0283,
+      "step": 3240
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 1.4254247288594813e-07,
+      "loss": 0.9987,
+      "step": 3245
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 1.3466471090366473e-07,
+      "loss": 1.0331,
+      "step": 3250
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 1.270093924554383e-07,
+      "loss": 1.0144,
+      "step": 3255
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 1.195766901533546e-07,
+      "loss": 1.0239,
+      "step": 3260
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 1.1236677158994902e-07,
+      "loss": 1.0065,
+      "step": 3265
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 1.0537979933443188e-07,
+      "loss": 1.0079,
+      "step": 3270
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 9.861593092902244e-08,
+      "loss": 1.0166,
+      "step": 3275
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 9.20753188853929e-08,
+      "loss": 1.0202,
+      "step": 3280
+    },
+    {
+      "epoch": 2.88,
+      "learning_rate": 8.57581106812333e-08,
+      "loss": 1.0039,
+      "step": 3285
+    },
+    {
+      "epoch": 2.88,
+      "learning_rate": 7.966444875692536e-08,
+      "loss": 1.0087,
+      "step": 3290
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 7.379447051232836e-08,
+      "loss": 1.022,
+      "step": 3295
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 6.814830830368491e-08,
+      "loss": 1.0094,
+      "step": 3300
+    },
+    {
+      "epoch": 2.89,
+      "eval_loss": 1.1193827390670776,
+      "eval_runtime": 425.5309,
+      "eval_samples_per_second": 37.99,
+      "eval_steps_per_second": 1.189,
+      "step": 3300
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 6.272608944063451e-08,
+      "loss": 0.9974,
+      "step": 3305
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 5.7527936183342515e-08,
+      "loss": 0.9925,
+      "step": 3310
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 5.255396573974447e-08,
+      "loss": 1.0052,
+      "step": 3315
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 4.7804290262903895e-08,
+      "loss": 1.0154,
+      "step": 3320
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 4.3279016848478685e-08,
+      "loss": 1.0163,
+      "step": 3325
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 3.897824753231416e-08,
+      "loss": 1.0158,
+      "step": 3330
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 3.490207928813605e-08,
+      "loss": 1.0208,
+      "step": 3335
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 3.105060402536886e-08,
+      "loss": 1.0156,
+      "step": 3340
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 2.7423908587056458e-08,
+      "loss": 1.0159,
+      "step": 3345
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 2.4022074747912515e-08,
+      "loss": 1.002,
+      "step": 3350
+    },
+    {
+      "epoch": 2.94,
+      "learning_rate": 2.0845179212468646e-08,
+      "loss": 1.017,
+      "step": 3355
+    },
+    {
+      "epoch": 2.94,
+      "learning_rate": 1.789329361335246e-08,
+      "loss": 1.0269,
+      "step": 3360
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 1.5166484509665512e-08,
+      "loss": 1.0201,
+      "step": 3365
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 1.2664813385486752e-08,
+      "loss": 1.0123,
+      "step": 3370
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 1.038833664848582e-08,
+      "loss": 1.0125,
+      "step": 3375
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 8.337105628648535e-09,
+      "loss": 1.018,
+      "step": 3380
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 6.511166577123362e-09,
+      "loss": 1.0043,
+      "step": 3385
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 4.91056066517448e-09,
+      "loss": 1.0165,
+      "step": 3390
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 3.5353239832580656e-09,
+      "loss": 0.9968,
+      "step": 3395
+    },
+    {
+      "epoch": 2.98,
+      "learning_rate": 2.3854875402029575e-09,
+      "loss": 1.0284,
+      "step": 3400
+    },
+    {
+      "epoch": 2.98,
+      "eval_loss": 1.119370937347412,
+      "eval_runtime": 426.6328,
+      "eval_samples_per_second": 37.892,
+      "eval_steps_per_second": 1.186,
+      "step": 3400
+    },
+    {
+      "epoch": 2.98,
+      "learning_rate": 1.4610772625156532e-09,
+      "loss": 1.0189,
+      "step": 3405
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 7.621139937963317e-10,
+      "loss": 1.0086,
+      "step": 3410
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 2.886134942647889e-10,
+      "loss": 1.0118,
+      "step": 3415
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.0586440407386354e-11,
+      "loss": 1.0181,
+      "step": 3420
+    },
+    {
+      "epoch": 3.0,
+      "step": 3423,
+      "total_flos": 3.426180505951378e+19,
+      "train_loss": 1.1024342694896863,
+      "train_runtime": 51497.7783,
+      "train_samples_per_second": 8.508,
+      "train_steps_per_second": 0.066
+    }
+  ],
+  "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
+}