Upload model

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,60 @@
+---
+library_name: transformers
+license: other
+base_model: Qwen/Qwen2.5-7B-Instruct
+tags:
+- llama-factory
+- full
+- generated_from_trainer
+model-index:
+- name: nemo_nano_100k
+  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. -->
+
+# nemo_nano_100k
+
+This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on the mlfoundations-dev/nemo_nano_100k dataset.
+
+## 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: 8e-05
+- train_batch_size: 1
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 512
+- total_train_batch_size: 512
+- total_eval_batch_size: 4096
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 5.0
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.46.1
+- Pytorch 2.6.0+cu124
+- Datasets 3.1.0
+- Tokenizers 0.20.3

added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 5.0,
+    "total_flos": 5116723388743680.0,
+    "train_loss": 0.5099891699728418,
+    "train_runtime": 4463.8352,
+    "train_samples_per_second": 34.825,
+    "train_steps_per_second": 0.068
+}

config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "/leonardo_work/EUHPC_E03_068/DCFT_shared/hub/models--Qwen--Qwen2.5-7B-Instruct/snapshots/a09a35458c702b33eeacc393d103063234e8bc28",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 3584,
+  "initializer_range": 0.02,
+  "intermediate_size": 18944,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 28,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 4,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.46.1",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 152064
+}

generation_config.json

@@ -0,0 +1,14 @@
+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.05,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "4.46.1"
+}

model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc060b339cc19fe870b72402bd6cefc82aab6e9d8b39ac278ab8bce97029875e
+size 4877660776

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:32c750ced6c81742141089e43e56b9e012732a0236cedc6b629d9b2b99ae8e0e
+size 4932751008

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b4aee495a093f139ab59a2dc21d750ec0bfff81afc0df1712470457e41ea208
+size 4330865200

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:129d5033e8a80a0f468a9adbd287188143a1453e0f1fb63888f4291eef84e367
+size 1089994880

model.safetensors.index.json

@@ -0,0 +1,346 @@
+{
+  "metadata": {
+    "total_size": 15231233024
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00004-of-00004.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.18.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.norm.weight": "model-00003-of-00004.safetensors"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

start_end.json

@@ -0,0 +1 @@
+{"start_time": "2025-04-28 09:33:23", "end_time": "2025-04-28 10:49:04"}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- messages[0]['content'] }}\n    {%- else %}\n        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}\n    {%- endif %}\n    {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n    {%- else %}\n        {{- '<|im_start|>system\\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + message.content }}\n        {%- endif %}\n        {%- for tool_call in message.tool_calls %}\n            {%- if tool_call.function is defined %}\n                {%- set tool_call = tool_call.function %}\n            {%- endif %}\n            {{- '\\n<tool_call>\\n{\"name\": \"' }}\n            {{- tool_call.name }}\n            {{- '\", \"arguments\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- '}\\n</tool_call>' }}\n        {%- endfor %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 5.0,
+    "total_flos": 5116723388743680.0,
+    "train_loss": 0.5099891699728418,
+    "train_runtime": 4463.8352,
+    "train_samples_per_second": 34.825,
+    "train_steps_per_second": 0.068
+}

trainer_log.jsonl

@@ -0,0 +1,306 @@
+{"current_steps": 1, "total_steps": 305, "loss": 0.9597, "lr": 2.580645161290323e-06, "epoch": 0.01639344262295082, "percentage": 0.33, "elapsed_time": "0:00:46", "remaining_time": "3:54:30"}
+{"current_steps": 2, "total_steps": 305, "loss": 0.9618, "lr": 5.161290322580646e-06, "epoch": 0.03278688524590164, "percentage": 0.66, "elapsed_time": "0:01:00", "remaining_time": "2:31:36"}
+{"current_steps": 3, "total_steps": 305, "loss": 0.9528, "lr": 7.741935483870968e-06, "epoch": 0.04918032786885246, "percentage": 0.98, "elapsed_time": "0:01:13", "remaining_time": "2:02:32"}
+{"current_steps": 4, "total_steps": 305, "loss": 0.8743, "lr": 1.0322580645161291e-05, "epoch": 0.06557377049180328, "percentage": 1.31, "elapsed_time": "0:01:26", "remaining_time": "1:48:10"}
+{"current_steps": 5, "total_steps": 305, "loss": 0.8371, "lr": 1.2903225806451613e-05, "epoch": 0.08196721311475409, "percentage": 1.64, "elapsed_time": "0:01:39", "remaining_time": "1:39:08"}
+{"current_steps": 6, "total_steps": 305, "loss": 0.8396, "lr": 1.5483870967741936e-05, "epoch": 0.09836065573770492, "percentage": 1.97, "elapsed_time": "0:01:51", "remaining_time": "1:32:30"}
+{"current_steps": 7, "total_steps": 305, "loss": 0.8132, "lr": 1.806451612903226e-05, "epoch": 0.11475409836065574, "percentage": 2.3, "elapsed_time": "0:02:04", "remaining_time": "1:27:59"}
+{"current_steps": 8, "total_steps": 305, "loss": 0.8073, "lr": 2.0645161290322582e-05, "epoch": 0.13114754098360656, "percentage": 2.62, "elapsed_time": "0:02:16", "remaining_time": "1:24:29"}
+{"current_steps": 9, "total_steps": 305, "loss": 0.7532, "lr": 2.3225806451612906e-05, "epoch": 0.14754098360655737, "percentage": 2.95, "elapsed_time": "0:02:28", "remaining_time": "1:21:35"}
+{"current_steps": 10, "total_steps": 305, "loss": 0.7392, "lr": 2.5806451612903226e-05, "epoch": 0.16393442622950818, "percentage": 3.28, "elapsed_time": "0:02:41", "remaining_time": "1:19:27"}
+{"current_steps": 11, "total_steps": 305, "loss": 0.7263, "lr": 2.8387096774193552e-05, "epoch": 0.18032786885245902, "percentage": 3.61, "elapsed_time": "0:02:56", "remaining_time": "1:18:29"}
+{"current_steps": 12, "total_steps": 305, "loss": 0.709, "lr": 3.096774193548387e-05, "epoch": 0.19672131147540983, "percentage": 3.93, "elapsed_time": "0:03:09", "remaining_time": "1:17:02"}
+{"current_steps": 13, "total_steps": 305, "loss": 0.6926, "lr": 3.3548387096774195e-05, "epoch": 0.21311475409836064, "percentage": 4.26, "elapsed_time": "0:03:22", "remaining_time": "1:15:44"}
+{"current_steps": 14, "total_steps": 305, "loss": 0.6882, "lr": 3.612903225806452e-05, "epoch": 0.22950819672131148, "percentage": 4.59, "elapsed_time": "0:03:35", "remaining_time": "1:14:39"}
+{"current_steps": 15, "total_steps": 305, "loss": 0.6705, "lr": 3.870967741935484e-05, "epoch": 0.2459016393442623, "percentage": 4.92, "elapsed_time": "0:03:48", "remaining_time": "1:13:36"}
+{"current_steps": 16, "total_steps": 305, "loss": 0.6702, "lr": 4.1290322580645165e-05, "epoch": 0.26229508196721313, "percentage": 5.25, "elapsed_time": "0:04:33", "remaining_time": "1:22:20"}
+{"current_steps": 17, "total_steps": 305, "loss": 0.6655, "lr": 4.387096774193548e-05, "epoch": 0.2786885245901639, "percentage": 5.57, "elapsed_time": "0:04:46", "remaining_time": "1:20:57"}
+{"current_steps": 18, "total_steps": 305, "loss": 0.6602, "lr": 4.645161290322581e-05, "epoch": 0.29508196721311475, "percentage": 5.9, "elapsed_time": "0:04:59", "remaining_time": "1:19:32"}
+{"current_steps": 19, "total_steps": 305, "loss": 0.6586, "lr": 4.9032258064516135e-05, "epoch": 0.3114754098360656, "percentage": 6.23, "elapsed_time": "0:05:12", "remaining_time": "1:18:19"}
+{"current_steps": 20, "total_steps": 305, "loss": 0.6283, "lr": 5.161290322580645e-05, "epoch": 0.32786885245901637, "percentage": 6.56, "elapsed_time": "0:05:25", "remaining_time": "1:17:14"}
+{"current_steps": 21, "total_steps": 305, "loss": 0.645, "lr": 5.4193548387096774e-05, "epoch": 0.3442622950819672, "percentage": 6.89, "elapsed_time": "0:05:38", "remaining_time": "1:16:21"}
+{"current_steps": 22, "total_steps": 305, "loss": 0.6402, "lr": 5.6774193548387104e-05, "epoch": 0.36065573770491804, "percentage": 7.21, "elapsed_time": "0:05:51", "remaining_time": "1:15:22"}
+{"current_steps": 23, "total_steps": 305, "loss": 0.6422, "lr": 5.935483870967743e-05, "epoch": 0.3770491803278688, "percentage": 7.54, "elapsed_time": "0:06:04", "remaining_time": "1:14:26"}
+{"current_steps": 24, "total_steps": 305, "loss": 0.6368, "lr": 6.193548387096774e-05, "epoch": 0.39344262295081966, "percentage": 7.87, "elapsed_time": "0:06:16", "remaining_time": "1:13:25"}
+{"current_steps": 25, "total_steps": 305, "loss": 0.6331, "lr": 6.451612903225807e-05, "epoch": 0.4098360655737705, "percentage": 8.2, "elapsed_time": "0:06:28", "remaining_time": "1:12:33"}
+{"current_steps": 26, "total_steps": 305, "loss": 0.6154, "lr": 6.709677419354839e-05, "epoch": 0.4262295081967213, "percentage": 8.52, "elapsed_time": "0:06:41", "remaining_time": "1:11:50"}
+{"current_steps": 27, "total_steps": 305, "loss": 0.6248, "lr": 6.967741935483871e-05, "epoch": 0.4426229508196721, "percentage": 8.85, "elapsed_time": "0:06:54", "remaining_time": "1:11:12"}
+{"current_steps": 28, "total_steps": 305, "loss": 0.6323, "lr": 7.225806451612904e-05, "epoch": 0.45901639344262296, "percentage": 9.18, "elapsed_time": "0:07:08", "remaining_time": "1:10:39"}
+{"current_steps": 29, "total_steps": 305, "loss": 0.6233, "lr": 7.483870967741936e-05, "epoch": 0.47540983606557374, "percentage": 9.51, "elapsed_time": "0:07:21", "remaining_time": "1:10:06"}
+{"current_steps": 30, "total_steps": 305, "loss": 0.6156, "lr": 7.741935483870968e-05, "epoch": 0.4918032786885246, "percentage": 9.84, "elapsed_time": "0:07:34", "remaining_time": "1:09:26"}
+{"current_steps": 31, "total_steps": 305, "loss": 0.6138, "lr": 8e-05, "epoch": 0.5081967213114754, "percentage": 10.16, "elapsed_time": "0:07:47", "remaining_time": "1:08:49"}
+{"current_steps": 32, "total_steps": 305, "loss": 0.6199, "lr": 7.99973707985831e-05, "epoch": 0.5245901639344263, "percentage": 10.49, "elapsed_time": "0:08:00", "remaining_time": "1:08:15"}
+{"current_steps": 33, "total_steps": 305, "loss": 0.6296, "lr": 7.998948353996737e-05, "epoch": 0.5409836065573771, "percentage": 10.82, "elapsed_time": "0:08:13", "remaining_time": "1:07:43"}
+{"current_steps": 34, "total_steps": 305, "loss": 0.6103, "lr": 7.99763392610124e-05, "epoch": 0.5573770491803278, "percentage": 11.15, "elapsed_time": "0:08:27", "remaining_time": "1:07:22"}
+{"current_steps": 35, "total_steps": 305, "loss": 0.6189, "lr": 7.995793968966601e-05, "epoch": 0.5737704918032787, "percentage": 11.48, "elapsed_time": "0:08:40", "remaining_time": "1:06:52"}
+{"current_steps": 36, "total_steps": 305, "loss": 0.604, "lr": 7.993428724473719e-05, "epoch": 0.5901639344262295, "percentage": 11.8, "elapsed_time": "0:08:52", "remaining_time": "1:06:19"}
+{"current_steps": 37, "total_steps": 305, "loss": 0.6133, "lr": 7.990538503557801e-05, "epoch": 0.6065573770491803, "percentage": 12.13, "elapsed_time": "0:09:05", "remaining_time": "1:05:50"}
+{"current_steps": 38, "total_steps": 305, "loss": 0.6041, "lr": 7.987123686167492e-05, "epoch": 0.6229508196721312, "percentage": 12.46, "elapsed_time": "0:09:18", "remaining_time": "1:05:21"}
+{"current_steps": 39, "total_steps": 305, "loss": 0.6172, "lr": 7.98318472121493e-05, "epoch": 0.639344262295082, "percentage": 12.79, "elapsed_time": "0:09:30", "remaining_time": "1:04:54"}
+{"current_steps": 40, "total_steps": 305, "loss": 0.6167, "lr": 7.978722126516727e-05, "epoch": 0.6557377049180327, "percentage": 13.11, "elapsed_time": "0:09:44", "remaining_time": "1:04:29"}
+{"current_steps": 41, "total_steps": 305, "loss": 0.5973, "lr": 7.973736488725896e-05, "epoch": 0.6721311475409836, "percentage": 13.44, "elapsed_time": "0:09:56", "remaining_time": "1:04:02"}
+{"current_steps": 42, "total_steps": 305, "loss": 0.6016, "lr": 7.968228463254735e-05, "epoch": 0.6885245901639344, "percentage": 13.77, "elapsed_time": "0:10:09", "remaining_time": "1:03:37"}
+{"current_steps": 43, "total_steps": 305, "loss": 0.5971, "lr": 7.962198774188664e-05, "epoch": 0.7049180327868853, "percentage": 14.1, "elapsed_time": "0:10:22", "remaining_time": "1:03:14"}
+{"current_steps": 44, "total_steps": 305, "loss": 0.5889, "lr": 7.955648214191034e-05, "epoch": 0.7213114754098361, "percentage": 14.43, "elapsed_time": "0:10:35", "remaining_time": "1:02:52"}
+{"current_steps": 45, "total_steps": 305, "loss": 0.5849, "lr": 7.948577644398924e-05, "epoch": 0.7377049180327869, "percentage": 14.75, "elapsed_time": "0:10:48", "remaining_time": "1:02:26"}
+{"current_steps": 46, "total_steps": 305, "loss": 0.5871, "lr": 7.940987994309944e-05, "epoch": 0.7540983606557377, "percentage": 15.08, "elapsed_time": "0:11:01", "remaining_time": "1:02:03"}
+{"current_steps": 47, "total_steps": 305, "loss": 0.577, "lr": 7.932880261660031e-05, "epoch": 0.7704918032786885, "percentage": 15.41, "elapsed_time": "0:11:13", "remaining_time": "1:01:38"}
+{"current_steps": 48, "total_steps": 305, "loss": 0.5856, "lr": 7.924255512292291e-05, "epoch": 0.7868852459016393, "percentage": 15.74, "elapsed_time": "0:11:26", "remaining_time": "1:01:17"}
+{"current_steps": 49, "total_steps": 305, "loss": 0.5931, "lr": 7.91511488001689e-05, "epoch": 0.8032786885245902, "percentage": 16.07, "elapsed_time": "0:11:40", "remaining_time": "1:01:01"}
+{"current_steps": 50, "total_steps": 305, "loss": 0.571, "lr": 7.905459566461993e-05, "epoch": 0.819672131147541, "percentage": 16.39, "elapsed_time": "0:11:55", "remaining_time": "1:00:51"}
+{"current_steps": 51, "total_steps": 305, "loss": 0.5915, "lr": 7.895290840915803e-05, "epoch": 0.8360655737704918, "percentage": 16.72, "elapsed_time": "0:12:08", "remaining_time": "1:00:27"}
+{"current_steps": 52, "total_steps": 305, "loss": 0.5879, "lr": 7.884610040159704e-05, "epoch": 0.8524590163934426, "percentage": 17.05, "elapsed_time": "0:12:20", "remaining_time": "1:00:02"}
+{"current_steps": 53, "total_steps": 305, "loss": 0.5779, "lr": 7.873418568292517e-05, "epoch": 0.8688524590163934, "percentage": 17.38, "elapsed_time": "0:12:33", "remaining_time": "0:59:43"}
+{"current_steps": 54, "total_steps": 305, "loss": 0.6193, "lr": 7.861717896545928e-05, "epoch": 0.8852459016393442, "percentage": 17.7, "elapsed_time": "0:12:46", "remaining_time": "0:59:25"}
+{"current_steps": 55, "total_steps": 305, "loss": 0.6092, "lr": 7.849509563091073e-05, "epoch": 0.9016393442622951, "percentage": 18.03, "elapsed_time": "0:13:00", "remaining_time": "0:59:09"}
+{"current_steps": 56, "total_steps": 305, "loss": 0.5895, "lr": 7.836795172836335e-05, "epoch": 0.9180327868852459, "percentage": 18.36, "elapsed_time": "0:13:13", "remaining_time": "0:58:48"}
+{"current_steps": 57, "total_steps": 305, "loss": 0.5821, "lr": 7.823576397216354e-05, "epoch": 0.9344262295081968, "percentage": 18.69, "elapsed_time": "0:13:26", "remaining_time": "0:58:30"}
+{"current_steps": 58, "total_steps": 305, "loss": 0.5988, "lr": 7.809854973972311e-05, "epoch": 0.9508196721311475, "percentage": 19.02, "elapsed_time": "0:13:39", "remaining_time": "0:58:11"}
+{"current_steps": 59, "total_steps": 305, "loss": 0.6166, "lr": 7.795632706923478e-05, "epoch": 0.9672131147540983, "percentage": 19.34, "elapsed_time": "0:13:52", "remaining_time": "0:57:50"}
+{"current_steps": 60, "total_steps": 305, "loss": 0.5992, "lr": 7.78091146573009e-05, "epoch": 0.9836065573770492, "percentage": 19.67, "elapsed_time": "0:14:04", "remaining_time": "0:57:30"}
+{"current_steps": 61, "total_steps": 305, "loss": 0.5826, "lr": 7.765693185647557e-05, "epoch": 1.0, "percentage": 20.0, "elapsed_time": "0:14:18", "remaining_time": "0:57:14"}
+{"current_steps": 62, "total_steps": 305, "loss": 0.5692, "lr": 7.749979867272055e-05, "epoch": 1.0163934426229508, "percentage": 20.33, "elapsed_time": "0:15:13", "remaining_time": "0:59:40"}
+{"current_steps": 63, "total_steps": 305, "loss": 0.5729, "lr": 7.733773576277533e-05, "epoch": 1.0327868852459017, "percentage": 20.66, "elapsed_time": "0:15:26", "remaining_time": "0:59:17"}
+{"current_steps": 64, "total_steps": 305, "loss": 0.5579, "lr": 7.71707644314415e-05, "epoch": 1.0491803278688525, "percentage": 20.98, "elapsed_time": "0:15:38", "remaining_time": "0:58:54"}
+{"current_steps": 65, "total_steps": 305, "loss": 0.5545, "lr": 7.699890662878214e-05, "epoch": 1.0655737704918034, "percentage": 21.31, "elapsed_time": "0:15:51", "remaining_time": "0:58:32"}
+{"current_steps": 66, "total_steps": 305, "loss": 0.5625, "lr": 7.682218494723615e-05, "epoch": 1.0819672131147542, "percentage": 21.64, "elapsed_time": "0:16:04", "remaining_time": "0:58:13"}
+{"current_steps": 67, "total_steps": 305, "loss": 0.5791, "lr": 7.66406226186483e-05, "epoch": 1.098360655737705, "percentage": 21.97, "elapsed_time": "0:16:18", "remaining_time": "0:57:55"}
+{"current_steps": 68, "total_steps": 305, "loss": 0.5604, "lr": 7.645424351121519e-05, "epoch": 1.1147540983606556, "percentage": 22.3, "elapsed_time": "0:16:31", "remaining_time": "0:57:37"}
+{"current_steps": 69, "total_steps": 305, "loss": 0.5585, "lr": 7.626307212634745e-05, "epoch": 1.1311475409836065, "percentage": 22.62, "elapsed_time": "0:16:44", "remaining_time": "0:57:17"}
+{"current_steps": 70, "total_steps": 305, "loss": 0.5563, "lr": 7.606713359544891e-05, "epoch": 1.1475409836065573, "percentage": 22.95, "elapsed_time": "0:16:57", "remaining_time": "0:56:56"}
+{"current_steps": 71, "total_steps": 305, "loss": 0.5581, "lr": 7.586645367661272e-05, "epoch": 1.1639344262295082, "percentage": 23.28, "elapsed_time": "0:17:11", "remaining_time": "0:56:38"}
+{"current_steps": 72, "total_steps": 305, "loss": 0.5361, "lr": 7.566105875123523e-05, "epoch": 1.180327868852459, "percentage": 23.61, "elapsed_time": "0:17:25", "remaining_time": "0:56:24"}
+{"current_steps": 73, "total_steps": 305, "loss": 0.555, "lr": 7.545097582054786e-05, "epoch": 1.1967213114754098, "percentage": 23.93, "elapsed_time": "0:17:39", "remaining_time": "0:56:06"}
+{"current_steps": 74, "total_steps": 305, "loss": 0.5514, "lr": 7.523623250206758e-05, "epoch": 1.2131147540983607, "percentage": 24.26, "elapsed_time": "0:17:57", "remaining_time": "0:56:03"}
+{"current_steps": 75, "total_steps": 305, "loss": 0.5514, "lr": 7.501685702596625e-05, "epoch": 1.2295081967213115, "percentage": 24.59, "elapsed_time": "0:18:10", "remaining_time": "0:55:44"}
+{"current_steps": 76, "total_steps": 305, "loss": 0.5494, "lr": 7.47928782313595e-05, "epoch": 1.2459016393442623, "percentage": 24.92, "elapsed_time": "0:18:24", "remaining_time": "0:55:27"}
+{"current_steps": 77, "total_steps": 305, "loss": 0.5369, "lr": 7.456432556251552e-05, "epoch": 1.2622950819672132, "percentage": 25.25, "elapsed_time": "0:18:39", "remaining_time": "0:55:13"}
+{"current_steps": 78, "total_steps": 305, "loss": 0.5454, "lr": 7.433122906498436e-05, "epoch": 1.278688524590164, "percentage": 25.57, "elapsed_time": "0:18:52", "remaining_time": "0:54:56"}
+{"current_steps": 79, "total_steps": 305, "loss": 0.5536, "lr": 7.409361938164811e-05, "epoch": 1.2950819672131146, "percentage": 25.9, "elapsed_time": "0:19:11", "remaining_time": "0:54:55"}
+{"current_steps": 80, "total_steps": 305, "loss": 0.5451, "lr": 7.385152774869257e-05, "epoch": 1.3114754098360657, "percentage": 26.23, "elapsed_time": "0:19:25", "remaining_time": "0:54:37"}
+{"current_steps": 81, "total_steps": 305, "loss": 0.5444, "lr": 7.360498599150095e-05, "epoch": 1.3278688524590163, "percentage": 26.56, "elapsed_time": "0:19:39", "remaining_time": "0:54:21"}
+{"current_steps": 82, "total_steps": 305, "loss": 0.5482, "lr": 7.335402652047011e-05, "epoch": 1.3442622950819672, "percentage": 26.89, "elapsed_time": "0:19:52", "remaining_time": "0:54:03"}
+{"current_steps": 83, "total_steps": 305, "loss": 0.5425, "lr": 7.309868232674992e-05, "epoch": 1.360655737704918, "percentage": 27.21, "elapsed_time": "0:20:04", "remaining_time": "0:53:42"}
+{"current_steps": 84, "total_steps": 305, "loss": 0.5326, "lr": 7.283898697790615e-05, "epoch": 1.3770491803278688, "percentage": 27.54, "elapsed_time": "0:20:17", "remaining_time": "0:53:23"}
+{"current_steps": 85, "total_steps": 305, "loss": 0.541, "lr": 7.257497461350778e-05, "epoch": 1.3934426229508197, "percentage": 27.87, "elapsed_time": "0:20:31", "remaining_time": "0:53:06"}
+{"current_steps": 86, "total_steps": 305, "loss": 0.5463, "lr": 7.23066799406389e-05, "epoch": 1.4098360655737705, "percentage": 28.2, "elapsed_time": "0:20:44", "remaining_time": "0:52:50"}
+{"current_steps": 87, "total_steps": 305, "loss": 0.5362, "lr": 7.203413822933626e-05, "epoch": 1.4262295081967213, "percentage": 28.52, "elapsed_time": "0:20:57", "remaining_time": "0:52:31"}
+{"current_steps": 88, "total_steps": 305, "loss": 0.5405, "lr": 7.17573853079525e-05, "epoch": 1.4426229508196722, "percentage": 28.85, "elapsed_time": "0:21:11", "remaining_time": "0:52:14"}
+{"current_steps": 89, "total_steps": 305, "loss": 0.5393, "lr": 7.14764575584463e-05, "epoch": 1.459016393442623, "percentage": 29.18, "elapsed_time": "0:21:23", "remaining_time": "0:51:54"}
+{"current_steps": 90, "total_steps": 305, "loss": 0.5329, "lr": 7.119139191159949e-05, "epoch": 1.4754098360655736, "percentage": 29.51, "elapsed_time": "0:21:36", "remaining_time": "0:51:36"}
+{"current_steps": 91, "total_steps": 305, "loss": 0.5369, "lr": 7.090222584216221e-05, "epoch": 1.4918032786885247, "percentage": 29.84, "elapsed_time": "0:21:49", "remaining_time": "0:51:20"}
+{"current_steps": 92, "total_steps": 305, "loss": 0.5404, "lr": 7.060899736392643e-05, "epoch": 1.5081967213114753, "percentage": 30.16, "elapsed_time": "0:22:03", "remaining_time": "0:51:03"}
+{"current_steps": 93, "total_steps": 305, "loss": 0.5445, "lr": 7.031174502472868e-05, "epoch": 1.5245901639344264, "percentage": 30.49, "elapsed_time": "0:22:16", "remaining_time": "0:50:47"}
+{"current_steps": 94, "total_steps": 305, "loss": 0.5315, "lr": 7.001050790138252e-05, "epoch": 1.540983606557377, "percentage": 30.82, "elapsed_time": "0:22:29", "remaining_time": "0:50:29"}
+{"current_steps": 95, "total_steps": 305, "loss": 0.5369, "lr": 6.970532559454152e-05, "epoch": 1.5573770491803278, "percentage": 31.15, "elapsed_time": "0:22:44", "remaining_time": "0:50:15"}
+{"current_steps": 96, "total_steps": 305, "loss": 0.5212, "lr": 6.939623822349339e-05, "epoch": 1.5737704918032787, "percentage": 31.48, "elapsed_time": "0:22:57", "remaining_time": "0:49:59"}
+{"current_steps": 97, "total_steps": 305, "loss": 0.5325, "lr": 6.908328642088579e-05, "epoch": 1.5901639344262295, "percentage": 31.8, "elapsed_time": "0:23:10", "remaining_time": "0:49:42"}
+{"current_steps": 98, "total_steps": 305, "loss": 0.5354, "lr": 6.87665113273849e-05, "epoch": 1.6065573770491803, "percentage": 32.13, "elapsed_time": "0:23:24", "remaining_time": "0:49:25"}
+{"current_steps": 99, "total_steps": 305, "loss": 0.5305, "lr": 6.844595458626691e-05, "epoch": 1.6229508196721312, "percentage": 32.46, "elapsed_time": "0:23:37", "remaining_time": "0:49:09"}
+{"current_steps": 100, "total_steps": 305, "loss": 0.534, "lr": 6.812165833794375e-05, "epoch": 1.639344262295082, "percentage": 32.79, "elapsed_time": "0:23:51", "remaining_time": "0:48:53"}
+{"current_steps": 101, "total_steps": 305, "loss": 0.5298, "lr": 6.779366521442318e-05, "epoch": 1.6557377049180326, "percentage": 33.11, "elapsed_time": "0:24:04", "remaining_time": "0:48:36"}
+{"current_steps": 102, "total_steps": 305, "loss": 0.5258, "lr": 6.746201833370447e-05, "epoch": 1.6721311475409837, "percentage": 33.44, "elapsed_time": "0:24:17", "remaining_time": "0:48:21"}
+{"current_steps": 103, "total_steps": 305, "loss": 0.5363, "lr": 6.712676129411005e-05, "epoch": 1.6885245901639343, "percentage": 33.77, "elapsed_time": "0:24:30", "remaining_time": "0:48:04"}
+{"current_steps": 104, "total_steps": 305, "loss": 0.5256, "lr": 6.678793816855408e-05, "epoch": 1.7049180327868854, "percentage": 34.1, "elapsed_time": "0:24:43", "remaining_time": "0:47:46"}
+{"current_steps": 105, "total_steps": 305, "loss": 0.5366, "lr": 6.644559349874869e-05, "epoch": 1.721311475409836, "percentage": 34.43, "elapsed_time": "0:24:56", "remaining_time": "0:47:30"}
+{"current_steps": 106, "total_steps": 305, "loss": 0.522, "lr": 6.609977228934838e-05, "epoch": 1.737704918032787, "percentage": 34.75, "elapsed_time": "0:25:10", "remaining_time": "0:47:15"}
+{"current_steps": 107, "total_steps": 305, "loss": 0.5321, "lr": 6.575052000203386e-05, "epoch": 1.7540983606557377, "percentage": 35.08, "elapsed_time": "0:25:24", "remaining_time": "0:47:00"}
+{"current_steps": 108, "total_steps": 305, "loss": 0.5292, "lr": 6.539788254953557e-05, "epoch": 1.7704918032786885, "percentage": 35.41, "elapsed_time": "0:25:37", "remaining_time": "0:46:44"}
+{"current_steps": 109, "total_steps": 305, "loss": 0.5266, "lr": 6.504190628959798e-05, "epoch": 1.7868852459016393, "percentage": 35.74, "elapsed_time": "0:25:49", "remaining_time": "0:46:26"}
+{"current_steps": 110, "total_steps": 305, "loss": 0.5249, "lr": 6.468263801888546e-05, "epoch": 1.8032786885245902, "percentage": 36.07, "elapsed_time": "0:26:02", "remaining_time": "0:46:09"}
+{"current_steps": 111, "total_steps": 305, "loss": 0.5252, "lr": 6.432012496683032e-05, "epoch": 1.819672131147541, "percentage": 36.39, "elapsed_time": "0:26:16", "remaining_time": "0:45:55"}
+{"current_steps": 112, "total_steps": 305, "loss": 0.5286, "lr": 6.395441478942404e-05, "epoch": 1.8360655737704918, "percentage": 36.72, "elapsed_time": "0:26:29", "remaining_time": "0:45:39"}
+{"current_steps": 113, "total_steps": 305, "loss": 0.5271, "lr": 6.358555556295252e-05, "epoch": 1.8524590163934427, "percentage": 37.05, "elapsed_time": "0:26:43", "remaining_time": "0:45:24"}
+{"current_steps": 114, "total_steps": 305, "loss": 0.5355, "lr": 6.321359577767574e-05, "epoch": 1.8688524590163933, "percentage": 37.38, "elapsed_time": "0:26:56", "remaining_time": "0:45:08"}
+{"current_steps": 115, "total_steps": 305, "loss": 0.5256, "lr": 6.283858433145345e-05, "epoch": 1.8852459016393444, "percentage": 37.7, "elapsed_time": "0:27:10", "remaining_time": "0:44:53"}
+{"current_steps": 116, "total_steps": 305, "loss": 0.524, "lr": 6.246057052331693e-05, "epoch": 1.901639344262295, "percentage": 38.03, "elapsed_time": "0:27:24", "remaining_time": "0:44:38"}
+{"current_steps": 117, "total_steps": 305, "loss": 0.5243, "lr": 6.207960404698818e-05, "epoch": 1.918032786885246, "percentage": 38.36, "elapsed_time": "0:27:37", "remaining_time": "0:44:23"}
+{"current_steps": 118, "total_steps": 305, "loss": 0.5257, "lr": 6.169573498434717e-05, "epoch": 1.9344262295081966, "percentage": 38.69, "elapsed_time": "0:27:50", "remaining_time": "0:44:07"}
+{"current_steps": 119, "total_steps": 305, "loss": 0.5202, "lr": 6.130901379884808e-05, "epoch": 1.9508196721311475, "percentage": 39.02, "elapsed_time": "0:28:04", "remaining_time": "0:43:52"}
+{"current_steps": 120, "total_steps": 305, "loss": 0.527, "lr": 6.091949132888534e-05, "epoch": 1.9672131147540983, "percentage": 39.34, "elapsed_time": "0:28:17", "remaining_time": "0:43:36"}
+{"current_steps": 121, "total_steps": 305, "loss": 0.5281, "lr": 6.052721878111045e-05, "epoch": 1.9836065573770492, "percentage": 39.67, "elapsed_time": "0:28:30", "remaining_time": "0:43:21"}
+{"current_steps": 122, "total_steps": 305, "loss": 0.5128, "lr": 6.0132247723700324e-05, "epoch": 2.0, "percentage": 40.0, "elapsed_time": "0:28:44", "remaining_time": "0:43:06"}
+{"current_steps": 123, "total_steps": 305, "loss": 0.4906, "lr": 5.973463007957817e-05, "epoch": 2.0163934426229506, "percentage": 40.33, "elapsed_time": "0:29:41", "remaining_time": "0:43:56"}
+{"current_steps": 124, "total_steps": 305, "loss": 0.4939, "lr": 5.933441811958764e-05, "epoch": 2.0327868852459017, "percentage": 40.66, "elapsed_time": "0:29:54", "remaining_time": "0:43:39"}
+{"current_steps": 125, "total_steps": 305, "loss": 0.4952, "lr": 5.8931664455621345e-05, "epoch": 2.0491803278688523, "percentage": 40.98, "elapsed_time": "0:30:07", "remaining_time": "0:43:22"}
+{"current_steps": 126, "total_steps": 305, "loss": 0.4892, "lr": 5.8526422033704494e-05, "epoch": 2.0655737704918034, "percentage": 41.31, "elapsed_time": "0:30:20", "remaining_time": "0:43:06"}
+{"current_steps": 127, "total_steps": 305, "loss": 0.4903, "lr": 5.811874412703457e-05, "epoch": 2.081967213114754, "percentage": 41.64, "elapsed_time": "0:30:34", "remaining_time": "0:42:50"}
+{"current_steps": 128, "total_steps": 305, "loss": 0.4874, "lr": 5.770868432897809e-05, "epoch": 2.098360655737705, "percentage": 41.97, "elapsed_time": "0:30:46", "remaining_time": "0:42:34"}
+{"current_steps": 129, "total_steps": 305, "loss": 0.4829, "lr": 5.729629654602513e-05, "epoch": 2.1147540983606556, "percentage": 42.3, "elapsed_time": "0:31:00", "remaining_time": "0:42:18"}
+{"current_steps": 130, "total_steps": 305, "loss": 0.4918, "lr": 5.6881634990702855e-05, "epoch": 2.1311475409836067, "percentage": 42.62, "elapsed_time": "0:31:14", "remaining_time": "0:42:04"}
+{"current_steps": 131, "total_steps": 305, "loss": 0.4901, "lr": 5.646475417444873e-05, "epoch": 2.1475409836065573, "percentage": 42.95, "elapsed_time": "0:31:29", "remaining_time": "0:41:49"}
+{"current_steps": 132, "total_steps": 305, "loss": 0.4949, "lr": 5.6045708900444364e-05, "epoch": 2.1639344262295084, "percentage": 43.28, "elapsed_time": "0:31:42", "remaining_time": "0:41:32"}
+{"current_steps": 133, "total_steps": 305, "loss": 0.4811, "lr": 5.5624554256411175e-05, "epoch": 2.180327868852459, "percentage": 43.61, "elapsed_time": "0:31:56", "remaining_time": "0:41:17"}
+{"current_steps": 134, "total_steps": 305, "loss": 0.4885, "lr": 5.520134560736852e-05, "epoch": 2.19672131147541, "percentage": 43.93, "elapsed_time": "0:32:09", "remaining_time": "0:41:02"}
+{"current_steps": 135, "total_steps": 305, "loss": 0.4872, "lr": 5.477613858835536e-05, "epoch": 2.2131147540983607, "percentage": 44.26, "elapsed_time": "0:32:22", "remaining_time": "0:40:46"}
+{"current_steps": 136, "total_steps": 305, "loss": 0.4805, "lr": 5.434898909711656e-05, "epoch": 2.2295081967213113, "percentage": 44.59, "elapsed_time": "0:32:37", "remaining_time": "0:40:32"}
+{"current_steps": 137, "total_steps": 305, "loss": 0.4825, "lr": 5.3919953286754477e-05, "epoch": 2.2459016393442623, "percentage": 44.92, "elapsed_time": "0:32:50", "remaining_time": "0:40:16"}
+{"current_steps": 138, "total_steps": 305, "loss": 0.4834, "lr": 5.348908755834716e-05, "epoch": 2.262295081967213, "percentage": 45.25, "elapsed_time": "0:33:04", "remaining_time": "0:40:02"}
+{"current_steps": 139, "total_steps": 305, "loss": 0.4833, "lr": 5.305644855353378e-05, "epoch": 2.278688524590164, "percentage": 45.57, "elapsed_time": "0:33:18", "remaining_time": "0:39:46"}
+{"current_steps": 140, "total_steps": 305, "loss": 0.4844, "lr": 5.262209314706856e-05, "epoch": 2.2950819672131146, "percentage": 45.9, "elapsed_time": "0:33:31", "remaining_time": "0:39:30"}
+{"current_steps": 141, "total_steps": 305, "loss": 0.487, "lr": 5.218607843934401e-05, "epoch": 2.3114754098360657, "percentage": 46.23, "elapsed_time": "0:33:44", "remaining_time": "0:39:14"}
+{"current_steps": 142, "total_steps": 305, "loss": 0.4774, "lr": 5.1748461748884476e-05, "epoch": 2.3278688524590163, "percentage": 46.56, "elapsed_time": "0:33:58", "remaining_time": "0:39:00"}
+{"current_steps": 143, "total_steps": 305, "loss": 0.4811, "lr": 5.130930060481113e-05, "epoch": 2.3442622950819674, "percentage": 46.89, "elapsed_time": "0:34:12", "remaining_time": "0:38:44"}
+{"current_steps": 144, "total_steps": 305, "loss": 0.4799, "lr": 5.0868652739279045e-05, "epoch": 2.360655737704918, "percentage": 47.21, "elapsed_time": "0:34:25", "remaining_time": "0:38:29"}
+{"current_steps": 145, "total_steps": 305, "loss": 0.4884, "lr": 5.042657607988787e-05, "epoch": 2.3770491803278686, "percentage": 47.54, "elapsed_time": "0:34:38", "remaining_time": "0:38:14"}
+{"current_steps": 146, "total_steps": 305, "loss": 0.4819, "lr": 4.9983128742066553e-05, "epoch": 2.3934426229508197, "percentage": 47.87, "elapsed_time": "0:34:52", "remaining_time": "0:37:58"}
+{"current_steps": 147, "total_steps": 305, "loss": 0.4817, "lr": 4.953836902143354e-05, "epoch": 2.4098360655737707, "percentage": 48.2, "elapsed_time": "0:35:05", "remaining_time": "0:37:42"}
+{"current_steps": 148, "total_steps": 305, "loss": 0.4817, "lr": 4.909235538613321e-05, "epoch": 2.4262295081967213, "percentage": 48.52, "elapsed_time": "0:35:18", "remaining_time": "0:37:27"}
+{"current_steps": 149, "total_steps": 305, "loss": 0.4855, "lr": 4.864514646914968e-05, "epoch": 2.442622950819672, "percentage": 48.85, "elapsed_time": "0:35:31", "remaining_time": "0:37:11"}
+{"current_steps": 150, "total_steps": 305, "loss": 0.4829, "lr": 4.819680106059884e-05, "epoch": 2.459016393442623, "percentage": 49.18, "elapsed_time": "0:35:45", "remaining_time": "0:36:56"}
+{"current_steps": 151, "total_steps": 305, "loss": 0.491, "lr": 4.7747378099999866e-05, "epoch": 2.4754098360655736, "percentage": 49.51, "elapsed_time": "0:35:58", "remaining_time": "0:36:41"}
+{"current_steps": 152, "total_steps": 305, "loss": 0.4884, "lr": 4.729693666852699e-05, "epoch": 2.4918032786885247, "percentage": 49.84, "elapsed_time": "0:36:12", "remaining_time": "0:36:26"}
+{"current_steps": 153, "total_steps": 305, "loss": 0.4851, "lr": 4.684553598124272e-05, "epoch": 2.5081967213114753, "percentage": 50.16, "elapsed_time": "0:36:26", "remaining_time": "0:36:12"}
+{"current_steps": 154, "total_steps": 305, "loss": 0.4819, "lr": 4.639323537931339e-05, "epoch": 2.5245901639344264, "percentage": 50.49, "elapsed_time": "0:36:39", "remaining_time": "0:35:56"}
+{"current_steps": 155, "total_steps": 305, "loss": 0.4905, "lr": 4.5940094322208156e-05, "epoch": 2.540983606557377, "percentage": 50.82, "elapsed_time": "0:36:53", "remaining_time": "0:35:41"}
+{"current_steps": 156, "total_steps": 305, "loss": 0.4765, "lr": 4.5486172379882495e-05, "epoch": 2.557377049180328, "percentage": 51.15, "elapsed_time": "0:37:06", "remaining_time": "0:35:26"}
+{"current_steps": 157, "total_steps": 305, "loss": 0.4852, "lr": 4.5031529224947106e-05, "epoch": 2.5737704918032787, "percentage": 51.48, "elapsed_time": "0:37:20", "remaining_time": "0:35:12"}
+{"current_steps": 158, "total_steps": 305, "loss": 0.4836, "lr": 4.4576224624823334e-05, "epoch": 2.5901639344262293, "percentage": 51.8, "elapsed_time": "0:37:33", "remaining_time": "0:34:57"}
+{"current_steps": 159, "total_steps": 305, "loss": 0.4889, "lr": 4.412031843388618e-05, "epoch": 2.6065573770491803, "percentage": 52.13, "elapsed_time": "0:37:47", "remaining_time": "0:34:41"}
+{"current_steps": 160, "total_steps": 305, "loss": 0.481, "lr": 4.36638705855958e-05, "epoch": 2.6229508196721314, "percentage": 52.46, "elapsed_time": "0:38:00", "remaining_time": "0:34:26"}
+{"current_steps": 161, "total_steps": 305, "loss": 0.4738, "lr": 4.320694108461868e-05, "epoch": 2.639344262295082, "percentage": 52.79, "elapsed_time": "0:38:13", "remaining_time": "0:34:11"}
+{"current_steps": 162, "total_steps": 305, "loss": 0.4783, "lr": 4.2749589998939366e-05, "epoch": 2.6557377049180326, "percentage": 53.11, "elapsed_time": "0:38:27", "remaining_time": "0:33:56"}
+{"current_steps": 163, "total_steps": 305, "loss": 0.4807, "lr": 4.2291877451964e-05, "epoch": 2.6721311475409837, "percentage": 53.44, "elapsed_time": "0:38:42", "remaining_time": "0:33:43"}
+{"current_steps": 164, "total_steps": 305, "loss": 0.4754, "lr": 4.183386361461643e-05, "epoch": 2.6885245901639343, "percentage": 53.77, "elapsed_time": "0:38:55", "remaining_time": "0:33:28"}
+{"current_steps": 165, "total_steps": 305, "loss": 0.4816, "lr": 4.137560869742816e-05, "epoch": 2.7049180327868854, "percentage": 54.1, "elapsed_time": "0:39:09", "remaining_time": "0:33:13"}
+{"current_steps": 166, "total_steps": 305, "loss": 0.477, "lr": 4.0917172942623074e-05, "epoch": 2.721311475409836, "percentage": 54.43, "elapsed_time": "0:39:21", "remaining_time": "0:32:57"}
+{"current_steps": 167, "total_steps": 305, "loss": 0.4863, "lr": 4.045861661619796e-05, "epoch": 2.737704918032787, "percentage": 54.75, "elapsed_time": "0:39:36", "remaining_time": "0:32:43"}
+{"current_steps": 168, "total_steps": 305, "loss": 0.4806, "lr": 4e-05, "epoch": 2.7540983606557377, "percentage": 55.08, "elapsed_time": "0:39:49", "remaining_time": "0:32:28"}
+{"current_steps": 169, "total_steps": 305, "loss": 0.4819, "lr": 3.954138338380204e-05, "epoch": 2.7704918032786887, "percentage": 55.41, "elapsed_time": "0:40:02", "remaining_time": "0:32:13"}
+{"current_steps": 170, "total_steps": 305, "loss": 0.4829, "lr": 3.908282705737694e-05, "epoch": 2.7868852459016393, "percentage": 55.74, "elapsed_time": "0:40:15", "remaining_time": "0:31:58"}
+{"current_steps": 171, "total_steps": 305, "loss": 0.4842, "lr": 3.862439130257184e-05, "epoch": 2.80327868852459, "percentage": 56.07, "elapsed_time": "0:40:30", "remaining_time": "0:31:44"}
+{"current_steps": 172, "total_steps": 305, "loss": 0.4848, "lr": 3.8166136385383585e-05, "epoch": 2.819672131147541, "percentage": 56.39, "elapsed_time": "0:40:44", "remaining_time": "0:31:30"}
+{"current_steps": 173, "total_steps": 305, "loss": 0.4814, "lr": 3.770812254803601e-05, "epoch": 2.836065573770492, "percentage": 56.72, "elapsed_time": "0:40:58", "remaining_time": "0:31:15"}
+{"current_steps": 174, "total_steps": 305, "loss": 0.4793, "lr": 3.725041000106065e-05, "epoch": 2.8524590163934427, "percentage": 57.05, "elapsed_time": "0:41:10", "remaining_time": "0:30:59"}
+{"current_steps": 175, "total_steps": 305, "loss": 0.466, "lr": 3.679305891538133e-05, "epoch": 2.8688524590163933, "percentage": 57.38, "elapsed_time": "0:41:24", "remaining_time": "0:30:45"}
+{"current_steps": 176, "total_steps": 305, "loss": 0.4818, "lr": 3.6336129414404214e-05, "epoch": 2.8852459016393444, "percentage": 57.7, "elapsed_time": "0:41:37", "remaining_time": "0:30:30"}
+{"current_steps": 177, "total_steps": 305, "loss": 0.4841, "lr": 3.5879681566113836e-05, "epoch": 2.901639344262295, "percentage": 58.03, "elapsed_time": "0:41:51", "remaining_time": "0:30:16"}
+{"current_steps": 178, "total_steps": 305, "loss": 0.4857, "lr": 3.5423775375176686e-05, "epoch": 2.918032786885246, "percentage": 58.36, "elapsed_time": "0:42:04", "remaining_time": "0:30:01"}
+{"current_steps": 179, "total_steps": 305, "loss": 0.475, "lr": 3.4968470775052914e-05, "epoch": 2.9344262295081966, "percentage": 58.69, "elapsed_time": "0:42:18", "remaining_time": "0:29:46"}
+{"current_steps": 180, "total_steps": 305, "loss": 0.4765, "lr": 3.451382762011753e-05, "epoch": 2.9508196721311473, "percentage": 59.02, "elapsed_time": "0:42:33", "remaining_time": "0:29:33"}
+{"current_steps": 181, "total_steps": 305, "loss": 0.4843, "lr": 3.405990567779185e-05, "epoch": 2.9672131147540983, "percentage": 59.34, "elapsed_time": "0:42:52", "remaining_time": "0:29:22"}
+{"current_steps": 182, "total_steps": 305, "loss": 0.4807, "lr": 3.3606764620686614e-05, "epoch": 2.9836065573770494, "percentage": 59.67, "elapsed_time": "0:43:05", "remaining_time": "0:29:07"}
+{"current_steps": 183, "total_steps": 305, "loss": 0.483, "lr": 3.3154464018757284e-05, "epoch": 3.0, "percentage": 60.0, "elapsed_time": "0:43:19", "remaining_time": "0:28:52"}
+{"current_steps": 184, "total_steps": 305, "loss": 0.4458, "lr": 3.270306333147301e-05, "epoch": 3.0163934426229506, "percentage": 60.33, "elapsed_time": "0:44:14", "remaining_time": "0:29:05"}
+{"current_steps": 185, "total_steps": 305, "loss": 0.46, "lr": 3.225262190000015e-05, "epoch": 3.0327868852459017, "percentage": 60.66, "elapsed_time": "0:44:27", "remaining_time": "0:28:50"}
+{"current_steps": 186, "total_steps": 305, "loss": 0.4488, "lr": 3.180319893940117e-05, "epoch": 3.0491803278688523, "percentage": 60.98, "elapsed_time": "0:44:41", "remaining_time": "0:28:35"}
+{"current_steps": 187, "total_steps": 305, "loss": 0.4487, "lr": 3.135485353085033e-05, "epoch": 3.0655737704918034, "percentage": 61.31, "elapsed_time": "0:44:54", "remaining_time": "0:28:20"}
+{"current_steps": 188, "total_steps": 305, "loss": 0.4466, "lr": 3.0907644613866795e-05, "epoch": 3.081967213114754, "percentage": 61.64, "elapsed_time": "0:45:08", "remaining_time": "0:28:05"}
+{"current_steps": 189, "total_steps": 305, "loss": 0.4423, "lr": 3.0461630978566473e-05, "epoch": 3.098360655737705, "percentage": 61.97, "elapsed_time": "0:45:22", "remaining_time": "0:27:50"}
+{"current_steps": 190, "total_steps": 305, "loss": 0.4438, "lr": 3.0016871257933457e-05, "epoch": 3.1147540983606556, "percentage": 62.3, "elapsed_time": "0:45:36", "remaining_time": "0:27:36"}
+{"current_steps": 191, "total_steps": 305, "loss": 0.4455, "lr": 2.9573423920112155e-05, "epoch": 3.1311475409836067, "percentage": 62.62, "elapsed_time": "0:45:50", "remaining_time": "0:27:21"}
+{"current_steps": 192, "total_steps": 305, "loss": 0.4463, "lr": 2.9131347260720965e-05, "epoch": 3.1475409836065573, "percentage": 62.95, "elapsed_time": "0:46:03", "remaining_time": "0:27:06"}
+{"current_steps": 193, "total_steps": 305, "loss": 0.4465, "lr": 2.8690699395188873e-05, "epoch": 3.1639344262295084, "percentage": 63.28, "elapsed_time": "0:46:17", "remaining_time": "0:26:51"}
+{"current_steps": 194, "total_steps": 305, "loss": 0.445, "lr": 2.825153825111553e-05, "epoch": 3.180327868852459, "percentage": 63.61, "elapsed_time": "0:46:30", "remaining_time": "0:26:36"}
+{"current_steps": 195, "total_steps": 305, "loss": 0.4472, "lr": 2.7813921560656003e-05, "epoch": 3.19672131147541, "percentage": 63.93, "elapsed_time": "0:46:44", "remaining_time": "0:26:22"}
+{"current_steps": 196, "total_steps": 305, "loss": 0.4384, "lr": 2.737790685293145e-05, "epoch": 3.2131147540983607, "percentage": 64.26, "elapsed_time": "0:46:58", "remaining_time": "0:26:07"}
+{"current_steps": 197, "total_steps": 305, "loss": 0.4474, "lr": 2.694355144646622e-05, "epoch": 3.2295081967213113, "percentage": 64.59, "elapsed_time": "0:47:11", "remaining_time": "0:25:52"}
+{"current_steps": 198, "total_steps": 305, "loss": 0.4476, "lr": 2.6510912441652855e-05, "epoch": 3.2459016393442623, "percentage": 64.92, "elapsed_time": "0:47:24", "remaining_time": "0:25:37"}
+{"current_steps": 199, "total_steps": 305, "loss": 0.4481, "lr": 2.6080046713245527e-05, "epoch": 3.262295081967213, "percentage": 65.25, "elapsed_time": "0:47:40", "remaining_time": "0:25:23"}
+{"current_steps": 200, "total_steps": 305, "loss": 0.4447, "lr": 2.5651010902883457e-05, "epoch": 3.278688524590164, "percentage": 65.57, "elapsed_time": "0:47:54", "remaining_time": "0:25:08"}
+{"current_steps": 201, "total_steps": 305, "loss": 0.4406, "lr": 2.5223861411644647e-05, "epoch": 3.2950819672131146, "percentage": 65.9, "elapsed_time": "0:48:06", "remaining_time": "0:24:53"}
+{"current_steps": 202, "total_steps": 305, "loss": 0.443, "lr": 2.4798654392631502e-05, "epoch": 3.3114754098360657, "percentage": 66.23, "elapsed_time": "0:48:21", "remaining_time": "0:24:39"}
+{"current_steps": 203, "total_steps": 305, "loss": 0.4481, "lr": 2.4375445743588832e-05, "epoch": 3.3278688524590163, "percentage": 66.56, "elapsed_time": "0:48:34", "remaining_time": "0:24:24"}
+{"current_steps": 204, "total_steps": 305, "loss": 0.4484, "lr": 2.3954291099555652e-05, "epoch": 3.3442622950819674, "percentage": 66.89, "elapsed_time": "0:48:47", "remaining_time": "0:24:09"}
+{"current_steps": 205, "total_steps": 305, "loss": 0.4407, "lr": 2.3535245825551285e-05, "epoch": 3.360655737704918, "percentage": 67.21, "elapsed_time": "0:49:02", "remaining_time": "0:23:55"}
+{"current_steps": 206, "total_steps": 305, "loss": 0.4474, "lr": 2.3118365009297145e-05, "epoch": 3.3770491803278686, "percentage": 67.54, "elapsed_time": "0:49:16", "remaining_time": "0:23:40"}
+{"current_steps": 207, "total_steps": 305, "loss": 0.4466, "lr": 2.2703703453974883e-05, "epoch": 3.3934426229508197, "percentage": 67.87, "elapsed_time": "0:49:31", "remaining_time": "0:23:26"}
+{"current_steps": 208, "total_steps": 305, "loss": 0.4417, "lr": 2.2291315671021922e-05, "epoch": 3.4098360655737707, "percentage": 68.2, "elapsed_time": "0:49:46", "remaining_time": "0:23:12"}
+{"current_steps": 209, "total_steps": 305, "loss": 0.4438, "lr": 2.1881255872965434e-05, "epoch": 3.4262295081967213, "percentage": 68.52, "elapsed_time": "0:50:01", "remaining_time": "0:22:58"}
+{"current_steps": 210, "total_steps": 305, "loss": 0.4479, "lr": 2.1473577966295506e-05, "epoch": 3.442622950819672, "percentage": 68.85, "elapsed_time": "0:50:14", "remaining_time": "0:22:43"}
+{"current_steps": 211, "total_steps": 305, "loss": 0.4511, "lr": 2.1068335544378672e-05, "epoch": 3.459016393442623, "percentage": 69.18, "elapsed_time": "0:50:27", "remaining_time": "0:22:28"}
+{"current_steps": 212, "total_steps": 305, "loss": 0.4447, "lr": 2.0665581880412366e-05, "epoch": 3.4754098360655736, "percentage": 69.51, "elapsed_time": "0:50:42", "remaining_time": "0:22:14"}
+{"current_steps": 213, "total_steps": 305, "loss": 0.4456, "lr": 2.026536992042184e-05, "epoch": 3.4918032786885247, "percentage": 69.84, "elapsed_time": "0:50:56", "remaining_time": "0:22:00"}
+{"current_steps": 214, "total_steps": 305, "loss": 0.4472, "lr": 1.9867752276299686e-05, "epoch": 3.5081967213114753, "percentage": 70.16, "elapsed_time": "0:51:10", "remaining_time": "0:21:45"}
+{"current_steps": 215, "total_steps": 305, "loss": 0.4491, "lr": 1.9472781218889565e-05, "epoch": 3.5245901639344264, "percentage": 70.49, "elapsed_time": "0:51:24", "remaining_time": "0:21:31"}
+{"current_steps": 216, "total_steps": 305, "loss": 0.447, "lr": 1.908050867111468e-05, "epoch": 3.540983606557377, "percentage": 70.82, "elapsed_time": "0:51:37", "remaining_time": "0:21:16"}
+{"current_steps": 217, "total_steps": 305, "loss": 0.4459, "lr": 1.869098620115194e-05, "epoch": 3.557377049180328, "percentage": 71.15, "elapsed_time": "0:51:52", "remaining_time": "0:21:02"}
+{"current_steps": 218, "total_steps": 305, "loss": 0.444, "lr": 1.8304265015652835e-05, "epoch": 3.5737704918032787, "percentage": 71.48, "elapsed_time": "0:52:05", "remaining_time": "0:20:47"}
+{"current_steps": 219, "total_steps": 305, "loss": 0.4521, "lr": 1.792039595301183e-05, "epoch": 3.5901639344262293, "percentage": 71.8, "elapsed_time": "0:52:19", "remaining_time": "0:20:32"}
+{"current_steps": 220, "total_steps": 305, "loss": 0.447, "lr": 1.7539429476683088e-05, "epoch": 3.6065573770491803, "percentage": 72.13, "elapsed_time": "0:52:33", "remaining_time": "0:20:18"}
+{"current_steps": 221, "total_steps": 305, "loss": 0.4411, "lr": 1.7161415668546557e-05, "epoch": 3.6229508196721314, "percentage": 72.46, "elapsed_time": "0:52:46", "remaining_time": "0:20:03"}
+{"current_steps": 222, "total_steps": 305, "loss": 0.442, "lr": 1.6786404222324278e-05, "epoch": 3.639344262295082, "percentage": 72.79, "elapsed_time": "0:53:00", "remaining_time": "0:19:48"}
+{"current_steps": 223, "total_steps": 305, "loss": 0.444, "lr": 1.641444443704749e-05, "epoch": 3.6557377049180326, "percentage": 73.11, "elapsed_time": "0:53:13", "remaining_time": "0:19:34"}
+{"current_steps": 224, "total_steps": 305, "loss": 0.448, "lr": 1.6045585210575955e-05, "epoch": 3.6721311475409837, "percentage": 73.44, "elapsed_time": "0:53:27", "remaining_time": "0:19:19"}
+{"current_steps": 225, "total_steps": 305, "loss": 0.4439, "lr": 1.56798750331697e-05, "epoch": 3.6885245901639343, "percentage": 73.77, "elapsed_time": "0:53:41", "remaining_time": "0:19:05"}
+{"current_steps": 226, "total_steps": 305, "loss": 0.4402, "lr": 1.531736198111456e-05, "epoch": 3.7049180327868854, "percentage": 74.1, "elapsed_time": "0:53:55", "remaining_time": "0:18:50"}
+{"current_steps": 227, "total_steps": 305, "loss": 0.4394, "lr": 1.4958093710402036e-05, "epoch": 3.721311475409836, "percentage": 74.43, "elapsed_time": "0:54:09", "remaining_time": "0:18:36"}
+{"current_steps": 228, "total_steps": 305, "loss": 0.4481, "lr": 1.4602117450464453e-05, "epoch": 3.737704918032787, "percentage": 74.75, "elapsed_time": "0:54:23", "remaining_time": "0:18:22"}
+{"current_steps": 229, "total_steps": 305, "loss": 0.4457, "lr": 1.4249479997966145e-05, "epoch": 3.7540983606557377, "percentage": 75.08, "elapsed_time": "0:54:38", "remaining_time": "0:18:08"}
+{"current_steps": 230, "total_steps": 305, "loss": 0.4445, "lr": 1.3900227710651647e-05, "epoch": 3.7704918032786887, "percentage": 75.41, "elapsed_time": "0:54:52", "remaining_time": "0:17:53"}
+{"current_steps": 231, "total_steps": 305, "loss": 0.4459, "lr": 1.3554406501251332e-05, "epoch": 3.7868852459016393, "percentage": 75.74, "elapsed_time": "0:55:05", "remaining_time": "0:17:38"}
+{"current_steps": 232, "total_steps": 305, "loss": 0.4382, "lr": 1.3212061831445913e-05, "epoch": 3.80327868852459, "percentage": 76.07, "elapsed_time": "0:55:19", "remaining_time": "0:17:24"}
+{"current_steps": 233, "total_steps": 305, "loss": 0.441, "lr": 1.2873238705889968e-05, "epoch": 3.819672131147541, "percentage": 76.39, "elapsed_time": "0:55:33", "remaining_time": "0:17:10"}
+{"current_steps": 234, "total_steps": 305, "loss": 0.442, "lr": 1.2537981666295536e-05, "epoch": 3.836065573770492, "percentage": 76.72, "elapsed_time": "0:55:46", "remaining_time": "0:16:55"}
+{"current_steps": 235, "total_steps": 305, "loss": 0.4381, "lr": 1.2206334785576822e-05, "epoch": 3.8524590163934427, "percentage": 77.05, "elapsed_time": "0:56:00", "remaining_time": "0:16:41"}
+{"current_steps": 236, "total_steps": 305, "loss": 0.4442, "lr": 1.1878341662056259e-05, "epoch": 3.8688524590163933, "percentage": 77.38, "elapsed_time": "0:56:14", "remaining_time": "0:16:26"}
+{"current_steps": 237, "total_steps": 305, "loss": 0.4435, "lr": 1.1554045413733097e-05, "epoch": 3.8852459016393444, "percentage": 77.7, "elapsed_time": "0:56:27", "remaining_time": "0:16:11"}
+{"current_steps": 238, "total_steps": 305, "loss": 0.4444, "lr": 1.123348867261512e-05, "epoch": 3.901639344262295, "percentage": 78.03, "elapsed_time": "0:56:40", "remaining_time": "0:15:57"}
+{"current_steps": 239, "total_steps": 305, "loss": 0.4425, "lr": 1.0916713579114218e-05, "epoch": 3.918032786885246, "percentage": 78.36, "elapsed_time": "0:56:55", "remaining_time": "0:15:43"}
+{"current_steps": 240, "total_steps": 305, "loss": 0.4448, "lr": 1.0603761776506616e-05, "epoch": 3.9344262295081966, "percentage": 78.69, "elapsed_time": "0:57:08", "remaining_time": "0:15:28"}
+{"current_steps": 241, "total_steps": 305, "loss": 0.444, "lr": 1.029467440545849e-05, "epoch": 3.9508196721311473, "percentage": 79.02, "elapsed_time": "0:57:22", "remaining_time": "0:15:14"}
+{"current_steps": 242, "total_steps": 305, "loss": 0.4462, "lr": 9.989492098617494e-06, "epoch": 3.9672131147540983, "percentage": 79.34, "elapsed_time": "0:57:35", "remaining_time": "0:14:59"}
+{"current_steps": 243, "total_steps": 305, "loss": 0.4439, "lr": 9.688254975271327e-06, "epoch": 3.9836065573770494, "percentage": 79.67, "elapsed_time": "0:57:48", "remaining_time": "0:14:45"}
+{"current_steps": 244, "total_steps": 305, "loss": 0.427, "lr": 9.391002636073589e-06, "epoch": 4.0, "percentage": 80.0, "elapsed_time": "0:58:02", "remaining_time": "0:14:30"}
+{"current_steps": 245, "total_steps": 305, "loss": 0.4269, "lr": 9.0977741578378e-06, "epoch": 4.016393442622951, "percentage": 80.33, "elapsed_time": "0:59:00", "remaining_time": "0:14:27"}
+{"current_steps": 246, "total_steps": 305, "loss": 0.4269, "lr": 8.808608088400522e-06, "epoch": 4.032786885245901, "percentage": 80.66, "elapsed_time": "0:59:14", "remaining_time": "0:14:12"}
+{"current_steps": 247, "total_steps": 305, "loss": 0.4212, "lr": 8.523542441553712e-06, "epoch": 4.049180327868853, "percentage": 80.98, "elapsed_time": "0:59:29", "remaining_time": "0:13:58"}
+{"current_steps": 248, "total_steps": 305, "loss": 0.4288, "lr": 8.242614692047507e-06, "epoch": 4.065573770491803, "percentage": 81.31, "elapsed_time": "0:59:43", "remaining_time": "0:13:43"}
+{"current_steps": 249, "total_steps": 305, "loss": 0.4234, "lr": 7.96586177066375e-06, "epoch": 4.081967213114754, "percentage": 81.64, "elapsed_time": "0:59:58", "remaining_time": "0:13:29"}
+{"current_steps": 250, "total_steps": 305, "loss": 0.4236, "lr": 7.693320059361107e-06, "epoch": 4.098360655737705, "percentage": 81.97, "elapsed_time": "1:00:12", "remaining_time": "0:13:14"}
+{"current_steps": 251, "total_steps": 305, "loss": 0.4324, "lr": 7.425025386492231e-06, "epoch": 4.114754098360656, "percentage": 82.3, "elapsed_time": "1:00:27", "remaining_time": "0:13:00"}
+{"current_steps": 252, "total_steps": 305, "loss": 0.4305, "lr": 7.161013022093861e-06, "epoch": 4.131147540983607, "percentage": 82.62, "elapsed_time": "1:00:41", "remaining_time": "0:12:45"}
+{"current_steps": 253, "total_steps": 305, "loss": 0.4198, "lr": 6.901317673250089e-06, "epoch": 4.147540983606557, "percentage": 82.95, "elapsed_time": "1:00:55", "remaining_time": "0:12:31"}
+{"current_steps": 254, "total_steps": 305, "loss": 0.4275, "lr": 6.645973479529892e-06, "epoch": 4.163934426229508, "percentage": 83.28, "elapsed_time": "1:01:09", "remaining_time": "0:12:16"}
+{"current_steps": 255, "total_steps": 305, "loss": 0.4286, "lr": 6.395014008499067e-06, "epoch": 4.180327868852459, "percentage": 83.61, "elapsed_time": "1:01:25", "remaining_time": "0:12:02"}
+{"current_steps": 256, "total_steps": 305, "loss": 0.4249, "lr": 6.148472251307436e-06, "epoch": 4.19672131147541, "percentage": 83.93, "elapsed_time": "1:01:38", "remaining_time": "0:11:47"}
+{"current_steps": 257, "total_steps": 305, "loss": 0.4266, "lr": 5.906380618351893e-06, "epoch": 4.213114754098361, "percentage": 84.26, "elapsed_time": "1:01:52", "remaining_time": "0:11:33"}
+{"current_steps": 258, "total_steps": 305, "loss": 0.4193, "lr": 5.668770935015641e-06, "epoch": 4.229508196721311, "percentage": 84.59, "elapsed_time": "1:02:06", "remaining_time": "0:11:18"}
+{"current_steps": 259, "total_steps": 305, "loss": 0.4213, "lr": 5.435674437484491e-06, "epoch": 4.245901639344262, "percentage": 84.92, "elapsed_time": "1:02:21", "remaining_time": "0:11:04"}
+{"current_steps": 260, "total_steps": 305, "loss": 0.415, "lr": 5.207121768640519e-06, "epoch": 4.262295081967213, "percentage": 85.25, "elapsed_time": "1:02:35", "remaining_time": "0:10:49"}
+{"current_steps": 261, "total_steps": 305, "loss": 0.4278, "lr": 4.983142974033764e-06, "epoch": 4.278688524590164, "percentage": 85.57, "elapsed_time": "1:02:49", "remaining_time": "0:10:35"}
+{"current_steps": 262, "total_steps": 305, "loss": 0.4186, "lr": 4.763767497932419e-06, "epoch": 4.295081967213115, "percentage": 85.9, "elapsed_time": "1:03:03", "remaining_time": "0:10:21"}
+{"current_steps": 263, "total_steps": 305, "loss": 0.4192, "lr": 4.549024179452151e-06, "epoch": 4.311475409836065, "percentage": 86.23, "elapsed_time": "1:03:19", "remaining_time": "0:10:06"}
+{"current_steps": 264, "total_steps": 305, "loss": 0.4221, "lr": 4.338941248764781e-06, "epoch": 4.327868852459017, "percentage": 86.56, "elapsed_time": "1:03:34", "remaining_time": "0:09:52"}
+{"current_steps": 265, "total_steps": 305, "loss": 0.4291, "lr": 4.133546323387281e-06, "epoch": 4.344262295081967, "percentage": 86.89, "elapsed_time": "1:03:49", "remaining_time": "0:09:38"}
+{"current_steps": 266, "total_steps": 305, "loss": 0.4183, "lr": 3.932866404551092e-06, "epoch": 4.360655737704918, "percentage": 87.21, "elapsed_time": "1:04:02", "remaining_time": "0:09:23"}
+{"current_steps": 267, "total_steps": 305, "loss": 0.4237, "lr": 3.7369278736525584e-06, "epoch": 4.377049180327869, "percentage": 87.54, "elapsed_time": "1:04:17", "remaining_time": "0:09:09"}
+{"current_steps": 268, "total_steps": 305, "loss": 0.4247, "lr": 3.545756488784826e-06, "epoch": 4.39344262295082, "percentage": 87.87, "elapsed_time": "1:04:30", "remaining_time": "0:08:54"}
+{"current_steps": 269, "total_steps": 305, "loss": 0.4162, "lr": 3.359377381351707e-06, "epoch": 4.409836065573771, "percentage": 88.2, "elapsed_time": "1:04:45", "remaining_time": "0:08:39"}
+{"current_steps": 270, "total_steps": 305, "loss": 0.425, "lr": 3.1778150527638664e-06, "epoch": 4.426229508196721, "percentage": 88.52, "elapsed_time": "1:05:00", "remaining_time": "0:08:25"}
+{"current_steps": 271, "total_steps": 305, "loss": 0.425, "lr": 3.0010933712178735e-06, "epoch": 4.442622950819672, "percentage": 88.85, "elapsed_time": "1:05:15", "remaining_time": "0:08:11"}
+{"current_steps": 272, "total_steps": 305, "loss": 0.431, "lr": 2.8292355685585147e-06, "epoch": 4.459016393442623, "percentage": 89.18, "elapsed_time": "1:05:28", "remaining_time": "0:07:56"}
+{"current_steps": 273, "total_steps": 305, "loss": 0.4202, "lr": 2.6622642372246874e-06, "epoch": 4.475409836065574, "percentage": 89.51, "elapsed_time": "1:05:42", "remaining_time": "0:07:42"}
+{"current_steps": 274, "total_steps": 305, "loss": 0.4199, "lr": 2.5002013272794654e-06, "epoch": 4.491803278688525, "percentage": 89.84, "elapsed_time": "1:05:57", "remaining_time": "0:07:27"}
+{"current_steps": 275, "total_steps": 305, "loss": 0.4268, "lr": 2.3430681435244386e-06, "epoch": 4.508196721311475, "percentage": 90.16, "elapsed_time": "1:06:12", "remaining_time": "0:07:13"}
+{"current_steps": 276, "total_steps": 305, "loss": 0.4251, "lr": 2.1908853426990983e-06, "epoch": 4.524590163934426, "percentage": 90.49, "elapsed_time": "1:06:26", "remaining_time": "0:06:58"}
+{"current_steps": 277, "total_steps": 305, "loss": 0.4213, "lr": 2.0436729307652213e-06, "epoch": 4.540983606557377, "percentage": 90.82, "elapsed_time": "1:06:41", "remaining_time": "0:06:44"}
+{"current_steps": 278, "total_steps": 305, "loss": 0.4245, "lr": 1.9014502602769003e-06, "epoch": 4.557377049180328, "percentage": 91.15, "elapsed_time": "1:06:54", "remaining_time": "0:06:29"}
+{"current_steps": 279, "total_steps": 305, "loss": 0.4233, "lr": 1.7642360278364767e-06, "epoch": 4.573770491803279, "percentage": 91.48, "elapsed_time": "1:07:08", "remaining_time": "0:06:15"}
+{"current_steps": 280, "total_steps": 305, "loss": 0.425, "lr": 1.6320482716366726e-06, "epoch": 4.590163934426229, "percentage": 91.8, "elapsed_time": "1:07:23", "remaining_time": "0:06:01"}
+{"current_steps": 281, "total_steps": 305, "loss": 0.4145, "lr": 1.5049043690892729e-06, "epoch": 4.60655737704918, "percentage": 92.13, "elapsed_time": "1:07:36", "remaining_time": "0:05:46"}
+{"current_steps": 282, "total_steps": 305, "loss": 0.4216, "lr": 1.3828210345407268e-06, "epoch": 4.622950819672131, "percentage": 92.46, "elapsed_time": "1:07:52", "remaining_time": "0:05:32"}
+{"current_steps": 283, "total_steps": 305, "loss": 0.4201, "lr": 1.2658143170748383e-06, "epoch": 4.639344262295082, "percentage": 92.79, "elapsed_time": "1:08:05", "remaining_time": "0:05:17"}
+{"current_steps": 284, "total_steps": 305, "loss": 0.4163, "lr": 1.1538995984029654e-06, "epoch": 4.655737704918033, "percentage": 93.11, "elapsed_time": "1:08:19", "remaining_time": "0:05:03"}
+{"current_steps": 285, "total_steps": 305, "loss": 0.4263, "lr": 1.047091590841971e-06, "epoch": 4.672131147540983, "percentage": 93.44, "elapsed_time": "1:08:32", "remaining_time": "0:04:48"}
+{"current_steps": 286, "total_steps": 305, "loss": 0.4221, "lr": 9.454043353800801e-07, "epoch": 4.688524590163935, "percentage": 93.77, "elapsed_time": "1:08:47", "remaining_time": "0:04:34"}
+{"current_steps": 287, "total_steps": 305, "loss": 0.422, "lr": 8.488511998311133e-07, "epoch": 4.704918032786885, "percentage": 94.1, "elapsed_time": "1:09:02", "remaining_time": "0:04:19"}
+{"current_steps": 288, "total_steps": 305, "loss": 0.4226, "lr": 7.574448770771003e-07, "epoch": 4.721311475409836, "percentage": 94.43, "elapsed_time": "1:09:16", "remaining_time": "0:04:05"}
+{"current_steps": 289, "total_steps": 305, "loss": 0.4179, "lr": 6.711973833997088e-07, "epoch": 4.737704918032787, "percentage": 94.75, "elapsed_time": "1:09:31", "remaining_time": "0:03:50"}
+{"current_steps": 290, "total_steps": 305, "loss": 0.415, "lr": 5.90120056900565e-07, "epoch": 4.754098360655737, "percentage": 95.08, "elapsed_time": "1:09:46", "remaining_time": "0:03:36"}
+{"current_steps": 291, "total_steps": 305, "loss": 0.4263, "lr": 5.142235560107623e-07, "epoch": 4.770491803278689, "percentage": 95.41, "elapsed_time": "1:10:00", "remaining_time": "0:03:22"}
+{"current_steps": 292, "total_steps": 305, "loss": 0.4272, "lr": 4.4351785808967485e-07, "epoch": 4.786885245901639, "percentage": 95.74, "elapsed_time": "1:10:14", "remaining_time": "0:03:07"}
+{"current_steps": 293, "total_steps": 305, "loss": 0.4223, "lr": 3.780122581133672e-07, "epoch": 4.80327868852459, "percentage": 96.07, "elapsed_time": "1:10:28", "remaining_time": "0:02:53"}
+{"current_steps": 294, "total_steps": 305, "loss": 0.4262, "lr": 3.177153674526556e-07, "epoch": 4.8196721311475414, "percentage": 96.39, "elapsed_time": "1:10:44", "remaining_time": "0:02:38"}
+{"current_steps": 295, "total_steps": 305, "loss": 0.4223, "lr": 2.626351127410498e-07, "epoch": 4.836065573770492, "percentage": 96.72, "elapsed_time": "1:10:57", "remaining_time": "0:02:24"}
+{"current_steps": 296, "total_steps": 305, "loss": 0.4259, "lr": 2.1277873483274148e-07, "epoch": 4.852459016393443, "percentage": 97.05, "elapsed_time": "1:11:11", "remaining_time": "0:02:09"}
+{"current_steps": 297, "total_steps": 305, "loss": 0.4235, "lr": 1.6815278785069944e-07, "epoch": 4.868852459016393, "percentage": 97.38, "elapsed_time": "1:11:26", "remaining_time": "0:01:55"}
+{"current_steps": 298, "total_steps": 305, "loss": 0.4193, "lr": 1.2876313832508314e-07, "epoch": 4.885245901639344, "percentage": 97.7, "elapsed_time": "1:11:39", "remaining_time": "0:01:40"}
+{"current_steps": 299, "total_steps": 305, "loss": 0.417, "lr": 9.461496442199735e-08, "epoch": 4.901639344262295, "percentage": 98.03, "elapsed_time": "1:11:53", "remaining_time": "0:01:26"}
+{"current_steps": 300, "total_steps": 305, "loss": 0.4301, "lr": 6.571275526281451e-08, "epoch": 4.918032786885246, "percentage": 98.36, "elapsed_time": "1:12:07", "remaining_time": "0:01:12"}
+{"current_steps": 301, "total_steps": 305, "loss": 0.4309, "lr": 4.206031033399338e-08, "epoch": 4.934426229508197, "percentage": 98.69, "elapsed_time": "1:12:23", "remaining_time": "0:00:57"}
+{"current_steps": 302, "total_steps": 305, "loss": 0.4143, "lr": 2.366073898761645e-08, "epoch": 4.950819672131147, "percentage": 99.02, "elapsed_time": "1:12:36", "remaining_time": "0:00:43"}
+{"current_steps": 303, "total_steps": 305, "loss": 0.4301, "lr": 1.0516460032641284e-08, "epoch": 4.967213114754099, "percentage": 99.34, "elapsed_time": "1:12:50", "remaining_time": "0:00:28"}
+{"current_steps": 304, "total_steps": 305, "loss": 0.418, "lr": 2.6292014169104764e-09, "epoch": 4.983606557377049, "percentage": 99.67, "elapsed_time": "1:13:04", "remaining_time": "0:00:14"}
+{"current_steps": 305, "total_steps": 305, "loss": 0.4146, "lr": 0.0, "epoch": 5.0, "percentage": 100.0, "elapsed_time": "1:13:19", "remaining_time": "0:00:00"}
+{"current_steps": 305, "total_steps": 305, "epoch": 5.0, "percentage": 100.0, "elapsed_time": "1:14:22", "remaining_time": "0:00:00"}

trainer_state.json

@@ -0,0 +1,2177 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 5.0,
+  "eval_steps": 500,
+  "global_step": 305,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01639344262295082,
+      "grad_norm": 5.58112545754218,
+      "learning_rate": 2.580645161290323e-06,
+      "loss": 0.9597,
+      "step": 1
+    },
+    {
+      "epoch": 0.03278688524590164,
+      "grad_norm": 5.5621177659627685,
+      "learning_rate": 5.161290322580646e-06,
+      "loss": 0.9618,
+      "step": 2
+    },
+    {
+      "epoch": 0.04918032786885246,
+      "grad_norm": 5.2002830302431375,
+      "learning_rate": 7.741935483870968e-06,
+      "loss": 0.9528,
+      "step": 3
+    },
+    {
+      "epoch": 0.06557377049180328,
+      "grad_norm": 2.429288619179718,
+      "learning_rate": 1.0322580645161291e-05,
+      "loss": 0.8743,
+      "step": 4
+    },
+    {
+      "epoch": 0.08196721311475409,
+      "grad_norm": 2.303682448858842,
+      "learning_rate": 1.2903225806451613e-05,
+      "loss": 0.8371,
+      "step": 5
+    },
+    {
+      "epoch": 0.09836065573770492,
+      "grad_norm": 3.7903444701868207,
+      "learning_rate": 1.5483870967741936e-05,
+      "loss": 0.8396,
+      "step": 6
+    },
+    {
+      "epoch": 0.11475409836065574,
+      "grad_norm": 4.276574856653076,
+      "learning_rate": 1.806451612903226e-05,
+      "loss": 0.8132,
+      "step": 7
+    },
+    {
+      "epoch": 0.13114754098360656,
+      "grad_norm": 3.2100580662432443,
+      "learning_rate": 2.0645161290322582e-05,
+      "loss": 0.8073,
+      "step": 8
+    },
+    {
+      "epoch": 0.14754098360655737,
+      "grad_norm": 2.400368523717908,
+      "learning_rate": 2.3225806451612906e-05,
+      "loss": 0.7532,
+      "step": 9
+    },
+    {
+      "epoch": 0.16393442622950818,
+      "grad_norm": 2.107929978755255,
+      "learning_rate": 2.5806451612903226e-05,
+      "loss": 0.7392,
+      "step": 10
+    },
+    {
+      "epoch": 0.18032786885245902,
+      "grad_norm": 1.706605478358681,
+      "learning_rate": 2.8387096774193552e-05,
+      "loss": 0.7263,
+      "step": 11
+    },
+    {
+      "epoch": 0.19672131147540983,
+      "grad_norm": 1.4882561058595487,
+      "learning_rate": 3.096774193548387e-05,
+      "loss": 0.709,
+      "step": 12
+    },
+    {
+      "epoch": 0.21311475409836064,
+      "grad_norm": 1.2623324140261931,
+      "learning_rate": 3.3548387096774195e-05,
+      "loss": 0.6926,
+      "step": 13
+    },
+    {
+      "epoch": 0.22950819672131148,
+      "grad_norm": 1.0042413251607623,
+      "learning_rate": 3.612903225806452e-05,
+      "loss": 0.6882,
+      "step": 14
+    },
+    {
+      "epoch": 0.2459016393442623,
+      "grad_norm": 1.2054555229493642,
+      "learning_rate": 3.870967741935484e-05,
+      "loss": 0.6705,
+      "step": 15
+    },
+    {
+      "epoch": 0.26229508196721313,
+      "grad_norm": 1.404103110949475,
+      "learning_rate": 4.1290322580645165e-05,
+      "loss": 0.6702,
+      "step": 16
+    },
+    {
+      "epoch": 0.2786885245901639,
+      "grad_norm": 1.4484944472411994,
+      "learning_rate": 4.387096774193548e-05,
+      "loss": 0.6655,
+      "step": 17
+    },
+    {
+      "epoch": 0.29508196721311475,
+      "grad_norm": 0.9845835703253549,
+      "learning_rate": 4.645161290322581e-05,
+      "loss": 0.6602,
+      "step": 18
+    },
+    {
+      "epoch": 0.3114754098360656,
+      "grad_norm": 1.8228708849031587,
+      "learning_rate": 4.9032258064516135e-05,
+      "loss": 0.6586,
+      "step": 19
+    },
+    {
+      "epoch": 0.32786885245901637,
+      "grad_norm": 1.0073875744998761,
+      "learning_rate": 5.161290322580645e-05,
+      "loss": 0.6283,
+      "step": 20
+    },
+    {
+      "epoch": 0.3442622950819672,
+      "grad_norm": 2.1477016621670657,
+      "learning_rate": 5.4193548387096774e-05,
+      "loss": 0.645,
+      "step": 21
+    },
+    {
+      "epoch": 0.36065573770491804,
+      "grad_norm": 1.4144920525799276,
+      "learning_rate": 5.6774193548387104e-05,
+      "loss": 0.6402,
+      "step": 22
+    },
+    {
+      "epoch": 0.3770491803278688,
+      "grad_norm": 1.7423289786096712,
+      "learning_rate": 5.935483870967743e-05,
+      "loss": 0.6422,
+      "step": 23
+    },
+    {
+      "epoch": 0.39344262295081966,
+      "grad_norm": 1.5173523064275696,
+      "learning_rate": 6.193548387096774e-05,
+      "loss": 0.6368,
+      "step": 24
+    },
+    {
+      "epoch": 0.4098360655737705,
+      "grad_norm": 1.400015090802773,
+      "learning_rate": 6.451612903225807e-05,
+      "loss": 0.6331,
+      "step": 25
+    },
+    {
+      "epoch": 0.4262295081967213,
+      "grad_norm": 1.4981731922098314,
+      "learning_rate": 6.709677419354839e-05,
+      "loss": 0.6154,
+      "step": 26
+    },
+    {
+      "epoch": 0.4426229508196721,
+      "grad_norm": 1.3910445077945457,
+      "learning_rate": 6.967741935483871e-05,
+      "loss": 0.6248,
+      "step": 27
+    },
+    {
+      "epoch": 0.45901639344262296,
+      "grad_norm": 2.287234175350141,
+      "learning_rate": 7.225806451612904e-05,
+      "loss": 0.6323,
+      "step": 28
+    },
+    {
+      "epoch": 0.47540983606557374,
+      "grad_norm": 1.8671369685906136,
+      "learning_rate": 7.483870967741936e-05,
+      "loss": 0.6233,
+      "step": 29
+    },
+    {
+      "epoch": 0.4918032786885246,
+      "grad_norm": 1.465283599887322,
+      "learning_rate": 7.741935483870968e-05,
+      "loss": 0.6156,
+      "step": 30
+    },
+    {
+      "epoch": 0.5081967213114754,
+      "grad_norm": 2.013369833154031,
+      "learning_rate": 8e-05,
+      "loss": 0.6138,
+      "step": 31
+    },
+    {
+      "epoch": 0.5245901639344263,
+      "grad_norm": 1.4445848293844874,
+      "learning_rate": 7.99973707985831e-05,
+      "loss": 0.6199,
+      "step": 32
+    },
+    {
+      "epoch": 0.5409836065573771,
+      "grad_norm": 2.008002806928902,
+      "learning_rate": 7.998948353996737e-05,
+      "loss": 0.6296,
+      "step": 33
+    },
+    {
+      "epoch": 0.5573770491803278,
+      "grad_norm": 1.6006692484547813,
+      "learning_rate": 7.99763392610124e-05,
+      "loss": 0.6103,
+      "step": 34
+    },
+    {
+      "epoch": 0.5737704918032787,
+      "grad_norm": 1.7952820693638532,
+      "learning_rate": 7.995793968966601e-05,
+      "loss": 0.6189,
+      "step": 35
+    },
+    {
+      "epoch": 0.5901639344262295,
+      "grad_norm": 1.260255034241788,
+      "learning_rate": 7.993428724473719e-05,
+      "loss": 0.604,
+      "step": 36
+    },
+    {
+      "epoch": 0.6065573770491803,
+      "grad_norm": 2.918515643648626,
+      "learning_rate": 7.990538503557801e-05,
+      "loss": 0.6133,
+      "step": 37
+    },
+    {
+      "epoch": 0.6229508196721312,
+      "grad_norm": 1.5936203819891444,
+      "learning_rate": 7.987123686167492e-05,
+      "loss": 0.6041,
+      "step": 38
+    },
+    {
+      "epoch": 0.639344262295082,
+      "grad_norm": 3.246485277477806,
+      "learning_rate": 7.98318472121493e-05,
+      "loss": 0.6172,
+      "step": 39
+    },
+    {
+      "epoch": 0.6557377049180327,
+      "grad_norm": 2.818701756393196,
+      "learning_rate": 7.978722126516727e-05,
+      "loss": 0.6167,
+      "step": 40
+    },
+    {
+      "epoch": 0.6721311475409836,
+      "grad_norm": 1.7221453757855867,
+      "learning_rate": 7.973736488725896e-05,
+      "loss": 0.5973,
+      "step": 41
+    },
+    {
+      "epoch": 0.6885245901639344,
+      "grad_norm": 1.802719546933628,
+      "learning_rate": 7.968228463254735e-05,
+      "loss": 0.6016,
+      "step": 42
+    },
+    {
+      "epoch": 0.7049180327868853,
+      "grad_norm": 1.471860307081128,
+      "learning_rate": 7.962198774188664e-05,
+      "loss": 0.5971,
+      "step": 43
+    },
+    {
+      "epoch": 0.7213114754098361,
+      "grad_norm": 1.2373741203327389,
+      "learning_rate": 7.955648214191034e-05,
+      "loss": 0.5889,
+      "step": 44
+    },
+    {
+      "epoch": 0.7377049180327869,
+      "grad_norm": 1.261142820035152,
+      "learning_rate": 7.948577644398924e-05,
+      "loss": 0.5849,
+      "step": 45
+    },
+    {
+      "epoch": 0.7540983606557377,
+      "grad_norm": 2.0142144188297166,
+      "learning_rate": 7.940987994309944e-05,
+      "loss": 0.5871,
+      "step": 46
+    },
+    {
+      "epoch": 0.7704918032786885,
+      "grad_norm": 1.1871765895293158,
+      "learning_rate": 7.932880261660031e-05,
+      "loss": 0.577,
+      "step": 47
+    },
+    {
+      "epoch": 0.7868852459016393,
+      "grad_norm": 1.2630851815058795,
+      "learning_rate": 7.924255512292291e-05,
+      "loss": 0.5856,
+      "step": 48
+    },
+    {
+      "epoch": 0.8032786885245902,
+      "grad_norm": 1.6849343569275448,
+      "learning_rate": 7.91511488001689e-05,
+      "loss": 0.5931,
+      "step": 49
+    },
+    {
+      "epoch": 0.819672131147541,
+      "grad_norm": 0.9859431187761973,
+      "learning_rate": 7.905459566461993e-05,
+      "loss": 0.571,
+      "step": 50
+    },
+    {
+      "epoch": 0.8360655737704918,
+      "grad_norm": 1.7623450712427522,
+      "learning_rate": 7.895290840915803e-05,
+      "loss": 0.5915,
+      "step": 51
+    },
+    {
+      "epoch": 0.8524590163934426,
+      "grad_norm": 1.2069052100241535,
+      "learning_rate": 7.884610040159704e-05,
+      "loss": 0.5879,
+      "step": 52
+    },
+    {
+      "epoch": 0.8688524590163934,
+      "grad_norm": 1.2236642927441974,
+      "learning_rate": 7.873418568292517e-05,
+      "loss": 0.5779,
+      "step": 53
+    },
+    {
+      "epoch": 0.8852459016393442,
+      "grad_norm": 5.906328363997742,
+      "learning_rate": 7.861717896545928e-05,
+      "loss": 0.6193,
+      "step": 54
+    },
+    {
+      "epoch": 0.9016393442622951,
+      "grad_norm": 2.8517859472775013,
+      "learning_rate": 7.849509563091073e-05,
+      "loss": 0.6092,
+      "step": 55
+    },
+    {
+      "epoch": 0.9180327868852459,
+      "grad_norm": 1.6402739036234257,
+      "learning_rate": 7.836795172836335e-05,
+      "loss": 0.5895,
+      "step": 56
+    },
+    {
+      "epoch": 0.9344262295081968,
+      "grad_norm": 1.071478048704771,
+      "learning_rate": 7.823576397216354e-05,
+      "loss": 0.5821,
+      "step": 57
+    },
+    {
+      "epoch": 0.9508196721311475,
+      "grad_norm": 2.6529573722112563,
+      "learning_rate": 7.809854973972311e-05,
+      "loss": 0.5988,
+      "step": 58
+    },
+    {
+      "epoch": 0.9672131147540983,
+      "grad_norm": 2.1179614996092426,
+      "learning_rate": 7.795632706923478e-05,
+      "loss": 0.6166,
+      "step": 59
+    },
+    {
+      "epoch": 0.9836065573770492,
+      "grad_norm": 1.9008009613755237,
+      "learning_rate": 7.78091146573009e-05,
+      "loss": 0.5992,
+      "step": 60
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.894221127540697,
+      "learning_rate": 7.765693185647557e-05,
+      "loss": 0.5826,
+      "step": 61
+    },
+    {
+      "epoch": 1.0163934426229508,
+      "grad_norm": 1.708613207423494,
+      "learning_rate": 7.749979867272055e-05,
+      "loss": 0.5692,
+      "step": 62
+    },
+    {
+      "epoch": 1.0327868852459017,
+      "grad_norm": 1.8649041648138827,
+      "learning_rate": 7.733773576277533e-05,
+      "loss": 0.5729,
+      "step": 63
+    },
+    {
+      "epoch": 1.0491803278688525,
+      "grad_norm": 1.41612070780247,
+      "learning_rate": 7.71707644314415e-05,
+      "loss": 0.5579,
+      "step": 64
+    },
+    {
+      "epoch": 1.0655737704918034,
+      "grad_norm": 1.9035698103601548,
+      "learning_rate": 7.699890662878214e-05,
+      "loss": 0.5545,
+      "step": 65
+    },
+    {
+      "epoch": 1.0819672131147542,
+      "grad_norm": 4.94355544144476,
+      "learning_rate": 7.682218494723615e-05,
+      "loss": 0.5625,
+      "step": 66
+    },
+    {
+      "epoch": 1.098360655737705,
+      "grad_norm": 1.4759904398074293,
+      "learning_rate": 7.66406226186483e-05,
+      "loss": 0.5791,
+      "step": 67
+    },
+    {
+      "epoch": 1.1147540983606556,
+      "grad_norm": 1.5331088373110235,
+      "learning_rate": 7.645424351121519e-05,
+      "loss": 0.5604,
+      "step": 68
+    },
+    {
+      "epoch": 1.1311475409836065,
+      "grad_norm": 1.289579106246813,
+      "learning_rate": 7.626307212634745e-05,
+      "loss": 0.5585,
+      "step": 69
+    },
+    {
+      "epoch": 1.1475409836065573,
+      "grad_norm": 1.3594377937672748,
+      "learning_rate": 7.606713359544891e-05,
+      "loss": 0.5563,
+      "step": 70
+    },
+    {
+      "epoch": 1.1639344262295082,
+      "grad_norm": 1.4718812627084297,
+      "learning_rate": 7.586645367661272e-05,
+      "loss": 0.5581,
+      "step": 71
+    },
+    {
+      "epoch": 1.180327868852459,
+      "grad_norm": 0.9356040160597947,
+      "learning_rate": 7.566105875123523e-05,
+      "loss": 0.5361,
+      "step": 72
+    },
+    {
+      "epoch": 1.1967213114754098,
+      "grad_norm": 0.801036937648957,
+      "learning_rate": 7.545097582054786e-05,
+      "loss": 0.555,
+      "step": 73
+    },
+    {
+      "epoch": 1.2131147540983607,
+      "grad_norm": 0.9230784392329119,
+      "learning_rate": 7.523623250206758e-05,
+      "loss": 0.5514,
+      "step": 74
+    },
+    {
+      "epoch": 1.2295081967213115,
+      "grad_norm": 1.27446883289555,
+      "learning_rate": 7.501685702596625e-05,
+      "loss": 0.5514,
+      "step": 75
+    },
+    {
+      "epoch": 1.2459016393442623,
+      "grad_norm": 1.089059775020075,
+      "learning_rate": 7.47928782313595e-05,
+      "loss": 0.5494,
+      "step": 76
+    },
+    {
+      "epoch": 1.2622950819672132,
+      "grad_norm": 0.9532311404835904,
+      "learning_rate": 7.456432556251552e-05,
+      "loss": 0.5369,
+      "step": 77
+    },
+    {
+      "epoch": 1.278688524590164,
+      "grad_norm": 1.1037696208490315,
+      "learning_rate": 7.433122906498436e-05,
+      "loss": 0.5454,
+      "step": 78
+    },
+    {
+      "epoch": 1.2950819672131146,
+      "grad_norm": 1.0044629833456924,
+      "learning_rate": 7.409361938164811e-05,
+      "loss": 0.5536,
+      "step": 79
+    },
+    {
+      "epoch": 1.3114754098360657,
+      "grad_norm": 1.139817816380103,
+      "learning_rate": 7.385152774869257e-05,
+      "loss": 0.5451,
+      "step": 80
+    },
+    {
+      "epoch": 1.3278688524590163,
+      "grad_norm": 0.9956186270275268,
+      "learning_rate": 7.360498599150095e-05,
+      "loss": 0.5444,
+      "step": 81
+    },
+    {
+      "epoch": 1.3442622950819672,
+      "grad_norm": 1.4777189178807848,
+      "learning_rate": 7.335402652047011e-05,
+      "loss": 0.5482,
+      "step": 82
+    },
+    {
+      "epoch": 1.360655737704918,
+      "grad_norm": 1.0684258214405637,
+      "learning_rate": 7.309868232674992e-05,
+      "loss": 0.5425,
+      "step": 83
+    },
+    {
+      "epoch": 1.3770491803278688,
+      "grad_norm": 1.045469949816082,
+      "learning_rate": 7.283898697790615e-05,
+      "loss": 0.5326,
+      "step": 84
+    },
+    {
+      "epoch": 1.3934426229508197,
+      "grad_norm": 1.2884574146396586,
+      "learning_rate": 7.257497461350778e-05,
+      "loss": 0.541,
+      "step": 85
+    },
+    {
+      "epoch": 1.4098360655737705,
+      "grad_norm": 1.2291748562831601,
+      "learning_rate": 7.23066799406389e-05,
+      "loss": 0.5463,
+      "step": 86
+    },
+    {
+      "epoch": 1.4262295081967213,
+      "grad_norm": 0.8801073224983168,
+      "learning_rate": 7.203413822933626e-05,
+      "loss": 0.5362,
+      "step": 87
+    },
+    {
+      "epoch": 1.4426229508196722,
+      "grad_norm": 0.9677124160991732,
+      "learning_rate": 7.17573853079525e-05,
+      "loss": 0.5405,
+      "step": 88
+    },
+    {
+      "epoch": 1.459016393442623,
+      "grad_norm": 1.3289631307779222,
+      "learning_rate": 7.14764575584463e-05,
+      "loss": 0.5393,
+      "step": 89
+    },
+    {
+      "epoch": 1.4754098360655736,
+      "grad_norm": 0.6274914847275181,
+      "learning_rate": 7.119139191159949e-05,
+      "loss": 0.5329,
+      "step": 90
+    },
+    {
+      "epoch": 1.4918032786885247,
+      "grad_norm": 0.5931193088659492,
+      "learning_rate": 7.090222584216221e-05,
+      "loss": 0.5369,
+      "step": 91
+    },
+    {
+      "epoch": 1.5081967213114753,
+      "grad_norm": 1.0213786001116576,
+      "learning_rate": 7.060899736392643e-05,
+      "loss": 0.5404,
+      "step": 92
+    },
+    {
+      "epoch": 1.5245901639344264,
+      "grad_norm": 1.2526087387777403,
+      "learning_rate": 7.031174502472868e-05,
+      "loss": 0.5445,
+      "step": 93
+    },
+    {
+      "epoch": 1.540983606557377,
+      "grad_norm": 0.9943578918521142,
+      "learning_rate": 7.001050790138252e-05,
+      "loss": 0.5315,
+      "step": 94
+    },
+    {
+      "epoch": 1.5573770491803278,
+      "grad_norm": 1.2583383738970864,
+      "learning_rate": 6.970532559454152e-05,
+      "loss": 0.5369,
+      "step": 95
+    },
+    {
+      "epoch": 1.5737704918032787,
+      "grad_norm": 0.8624167954684502,
+      "learning_rate": 6.939623822349339e-05,
+      "loss": 0.5212,
+      "step": 96
+    },
+    {
+      "epoch": 1.5901639344262295,
+      "grad_norm": 0.8714228266723913,
+      "learning_rate": 6.908328642088579e-05,
+      "loss": 0.5325,
+      "step": 97
+    },
+    {
+      "epoch": 1.6065573770491803,
+      "grad_norm": 0.9559074422174284,
+      "learning_rate": 6.87665113273849e-05,
+      "loss": 0.5354,
+      "step": 98
+    },
+    {
+      "epoch": 1.6229508196721312,
+      "grad_norm": 1.020705004388792,
+      "learning_rate": 6.844595458626691e-05,
+      "loss": 0.5305,
+      "step": 99
+    },
+    {
+      "epoch": 1.639344262295082,
+      "grad_norm": 1.2718848911597436,
+      "learning_rate": 6.812165833794375e-05,
+      "loss": 0.534,
+      "step": 100
+    },
+    {
+      "epoch": 1.6557377049180326,
+      "grad_norm": 0.8646369335110455,
+      "learning_rate": 6.779366521442318e-05,
+      "loss": 0.5298,
+      "step": 101
+    },
+    {
+      "epoch": 1.6721311475409837,
+      "grad_norm": 1.0431466383974872,
+      "learning_rate": 6.746201833370447e-05,
+      "loss": 0.5258,
+      "step": 102
+    },
+    {
+      "epoch": 1.6885245901639343,
+      "grad_norm": 1.2838018594279825,
+      "learning_rate": 6.712676129411005e-05,
+      "loss": 0.5363,
+      "step": 103
+    },
+    {
+      "epoch": 1.7049180327868854,
+      "grad_norm": 0.7735313109171527,
+      "learning_rate": 6.678793816855408e-05,
+      "loss": 0.5256,
+      "step": 104
+    },
+    {
+      "epoch": 1.721311475409836,
+      "grad_norm": 1.1292517886407496,
+      "learning_rate": 6.644559349874869e-05,
+      "loss": 0.5366,
+      "step": 105
+    },
+    {
+      "epoch": 1.737704918032787,
+      "grad_norm": 1.1330330672560118,
+      "learning_rate": 6.609977228934838e-05,
+      "loss": 0.522,
+      "step": 106
+    },
+    {
+      "epoch": 1.7540983606557377,
+      "grad_norm": 0.8960440091301521,
+      "learning_rate": 6.575052000203386e-05,
+      "loss": 0.5321,
+      "step": 107
+    },
+    {
+      "epoch": 1.7704918032786885,
+      "grad_norm": 0.8001970911153957,
+      "learning_rate": 6.539788254953557e-05,
+      "loss": 0.5292,
+      "step": 108
+    },
+    {
+      "epoch": 1.7868852459016393,
+      "grad_norm": 0.826768977068413,
+      "learning_rate": 6.504190628959798e-05,
+      "loss": 0.5266,
+      "step": 109
+    },
+    {
+      "epoch": 1.8032786885245902,
+      "grad_norm": 1.047081785970805,
+      "learning_rate": 6.468263801888546e-05,
+      "loss": 0.5249,
+      "step": 110
+    },
+    {
+      "epoch": 1.819672131147541,
+      "grad_norm": 1.2237949792369915,
+      "learning_rate": 6.432012496683032e-05,
+      "loss": 0.5252,
+      "step": 111
+    },
+    {
+      "epoch": 1.8360655737704918,
+      "grad_norm": 0.5326723853060618,
+      "learning_rate": 6.395441478942404e-05,
+      "loss": 0.5286,
+      "step": 112
+    },
+    {
+      "epoch": 1.8524590163934427,
+      "grad_norm": 0.8184475093441959,
+      "learning_rate": 6.358555556295252e-05,
+      "loss": 0.5271,
+      "step": 113
+    },
+    {
+      "epoch": 1.8688524590163933,
+      "grad_norm": 1.0030170729677295,
+      "learning_rate": 6.321359577767574e-05,
+      "loss": 0.5355,
+      "step": 114
+    },
+    {
+      "epoch": 1.8852459016393444,
+      "grad_norm": 0.8438447158790536,
+      "learning_rate": 6.283858433145345e-05,
+      "loss": 0.5256,
+      "step": 115
+    },
+    {
+      "epoch": 1.901639344262295,
+      "grad_norm": 0.6775475091945453,
+      "learning_rate": 6.246057052331693e-05,
+      "loss": 0.524,
+      "step": 116
+    },
+    {
+      "epoch": 1.918032786885246,
+      "grad_norm": 0.5277047106995992,
+      "learning_rate": 6.207960404698818e-05,
+      "loss": 0.5243,
+      "step": 117
+    },
+    {
+      "epoch": 1.9344262295081966,
+      "grad_norm": 0.5241956608413472,
+      "learning_rate": 6.169573498434717e-05,
+      "loss": 0.5257,
+      "step": 118
+    },
+    {
+      "epoch": 1.9508196721311475,
+      "grad_norm": 0.6035995248402194,
+      "learning_rate": 6.130901379884808e-05,
+      "loss": 0.5202,
+      "step": 119
+    },
+    {
+      "epoch": 1.9672131147540983,
+      "grad_norm": 0.779564589475285,
+      "learning_rate": 6.091949132888534e-05,
+      "loss": 0.527,
+      "step": 120
+    },
+    {
+      "epoch": 1.9836065573770492,
+      "grad_norm": 1.022288473196731,
+      "learning_rate": 6.052721878111045e-05,
+      "loss": 0.5281,
+      "step": 121
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.2764091263174406,
+      "learning_rate": 6.0132247723700324e-05,
+      "loss": 0.5128,
+      "step": 122
+    },
+    {
+      "epoch": 2.0163934426229506,
+      "grad_norm": 0.7708344672192273,
+      "learning_rate": 5.973463007957817e-05,
+      "loss": 0.4906,
+      "step": 123
+    },
+    {
+      "epoch": 2.0327868852459017,
+      "grad_norm": 0.7977537199838536,
+      "learning_rate": 5.933441811958764e-05,
+      "loss": 0.4939,
+      "step": 124
+    },
+    {
+      "epoch": 2.0491803278688523,
+      "grad_norm": 1.0793702753193162,
+      "learning_rate": 5.8931664455621345e-05,
+      "loss": 0.4952,
+      "step": 125
+    },
+    {
+      "epoch": 2.0655737704918034,
+      "grad_norm": 0.8986778009783392,
+      "learning_rate": 5.8526422033704494e-05,
+      "loss": 0.4892,
+      "step": 126
+    },
+    {
+      "epoch": 2.081967213114754,
+      "grad_norm": 1.030605766210909,
+      "learning_rate": 5.811874412703457e-05,
+      "loss": 0.4903,
+      "step": 127
+    },
+    {
+      "epoch": 2.098360655737705,
+      "grad_norm": 1.0660924489319643,
+      "learning_rate": 5.770868432897809e-05,
+      "loss": 0.4874,
+      "step": 128
+    },
+    {
+      "epoch": 2.1147540983606556,
+      "grad_norm": 0.940262946228767,
+      "learning_rate": 5.729629654602513e-05,
+      "loss": 0.4829,
+      "step": 129
+    },
+    {
+      "epoch": 2.1311475409836067,
+      "grad_norm": 0.9888397367489762,
+      "learning_rate": 5.6881634990702855e-05,
+      "loss": 0.4918,
+      "step": 130
+    },
+    {
+      "epoch": 2.1475409836065573,
+      "grad_norm": 0.8484560909641439,
+      "learning_rate": 5.646475417444873e-05,
+      "loss": 0.4901,
+      "step": 131
+    },
+    {
+      "epoch": 2.1639344262295084,
+      "grad_norm": 0.521192253144461,
+      "learning_rate": 5.6045708900444364e-05,
+      "loss": 0.4949,
+      "step": 132
+    },
+    {
+      "epoch": 2.180327868852459,
+      "grad_norm": 0.771777999740359,
+      "learning_rate": 5.5624554256411175e-05,
+      "loss": 0.4811,
+      "step": 133
+    },
+    {
+      "epoch": 2.19672131147541,
+      "grad_norm": 0.8227491310955285,
+      "learning_rate": 5.520134560736852e-05,
+      "loss": 0.4885,
+      "step": 134
+    },
+    {
+      "epoch": 2.2131147540983607,
+      "grad_norm": 0.5992373916210076,
+      "learning_rate": 5.477613858835536e-05,
+      "loss": 0.4872,
+      "step": 135
+    },
+    {
+      "epoch": 2.2295081967213113,
+      "grad_norm": 0.5052047355472387,
+      "learning_rate": 5.434898909711656e-05,
+      "loss": 0.4805,
+      "step": 136
+    },
+    {
+      "epoch": 2.2459016393442623,
+      "grad_norm": 0.7157168717143663,
+      "learning_rate": 5.3919953286754477e-05,
+      "loss": 0.4825,
+      "step": 137
+    },
+    {
+      "epoch": 2.262295081967213,
+      "grad_norm": 0.5410573992811272,
+      "learning_rate": 5.348908755834716e-05,
+      "loss": 0.4834,
+      "step": 138
+    },
+    {
+      "epoch": 2.278688524590164,
+      "grad_norm": 0.4166121333458553,
+      "learning_rate": 5.305644855353378e-05,
+      "loss": 0.4833,
+      "step": 139
+    },
+    {
+      "epoch": 2.2950819672131146,
+      "grad_norm": 0.5233638295240889,
+      "learning_rate": 5.262209314706856e-05,
+      "loss": 0.4844,
+      "step": 140
+    },
+    {
+      "epoch": 2.3114754098360657,
+      "grad_norm": 0.4090985726833338,
+      "learning_rate": 5.218607843934401e-05,
+      "loss": 0.487,
+      "step": 141
+    },
+    {
+      "epoch": 2.3278688524590163,
+      "grad_norm": 0.363420954496579,
+      "learning_rate": 5.1748461748884476e-05,
+      "loss": 0.4774,
+      "step": 142
+    },
+    {
+      "epoch": 2.3442622950819674,
+      "grad_norm": 0.4690278843251207,
+      "learning_rate": 5.130930060481113e-05,
+      "loss": 0.4811,
+      "step": 143
+    },
+    {
+      "epoch": 2.360655737704918,
+      "grad_norm": 0.46496866291592687,
+      "learning_rate": 5.0868652739279045e-05,
+      "loss": 0.4799,
+      "step": 144
+    },
+    {
+      "epoch": 2.3770491803278686,
+      "grad_norm": 0.4067588783945542,
+      "learning_rate": 5.042657607988787e-05,
+      "loss": 0.4884,
+      "step": 145
+    },
+    {
+      "epoch": 2.3934426229508197,
+      "grad_norm": 0.3249832363624369,
+      "learning_rate": 4.9983128742066553e-05,
+      "loss": 0.4819,
+      "step": 146
+    },
+    {
+      "epoch": 2.4098360655737707,
+      "grad_norm": 0.3155267675853147,
+      "learning_rate": 4.953836902143354e-05,
+      "loss": 0.4817,
+      "step": 147
+    },
+    {
+      "epoch": 2.4262295081967213,
+      "grad_norm": 0.30276002082546505,
+      "learning_rate": 4.909235538613321e-05,
+      "loss": 0.4817,
+      "step": 148
+    },
+    {
+      "epoch": 2.442622950819672,
+      "grad_norm": 0.3578858890853383,
+      "learning_rate": 4.864514646914968e-05,
+      "loss": 0.4855,
+      "step": 149
+    },
+    {
+      "epoch": 2.459016393442623,
+      "grad_norm": 0.28701615085250826,
+      "learning_rate": 4.819680106059884e-05,
+      "loss": 0.4829,
+      "step": 150
+    },
+    {
+      "epoch": 2.4754098360655736,
+      "grad_norm": 0.37224620973373557,
+      "learning_rate": 4.7747378099999866e-05,
+      "loss": 0.491,
+      "step": 151
+    },
+    {
+      "epoch": 2.4918032786885247,
+      "grad_norm": 0.34012579144317995,
+      "learning_rate": 4.729693666852699e-05,
+      "loss": 0.4884,
+      "step": 152
+    },
+    {
+      "epoch": 2.5081967213114753,
+      "grad_norm": 0.3774404172171865,
+      "learning_rate": 4.684553598124272e-05,
+      "loss": 0.4851,
+      "step": 153
+    },
+    {
+      "epoch": 2.5245901639344264,
+      "grad_norm": 0.3549426068916513,
+      "learning_rate": 4.639323537931339e-05,
+      "loss": 0.4819,
+      "step": 154
+    },
+    {
+      "epoch": 2.540983606557377,
+      "grad_norm": 0.39031877291102196,
+      "learning_rate": 4.5940094322208156e-05,
+      "loss": 0.4905,
+      "step": 155
+    },
+    {
+      "epoch": 2.557377049180328,
+      "grad_norm": 0.295463066655204,
+      "learning_rate": 4.5486172379882495e-05,
+      "loss": 0.4765,
+      "step": 156
+    },
+    {
+      "epoch": 2.5737704918032787,
+      "grad_norm": 0.29781720638326936,
+      "learning_rate": 4.5031529224947106e-05,
+      "loss": 0.4852,
+      "step": 157
+    },
+    {
+      "epoch": 2.5901639344262293,
+      "grad_norm": 0.2852567559187638,
+      "learning_rate": 4.4576224624823334e-05,
+      "loss": 0.4836,
+      "step": 158
+    },
+    {
+      "epoch": 2.6065573770491803,
+      "grad_norm": 0.239161233813967,
+      "learning_rate": 4.412031843388618e-05,
+      "loss": 0.4889,
+      "step": 159
+    },
+    {
+      "epoch": 2.6229508196721314,
+      "grad_norm": 0.37163233149883945,
+      "learning_rate": 4.36638705855958e-05,
+      "loss": 0.481,
+      "step": 160
+    },
+    {
+      "epoch": 2.639344262295082,
+      "grad_norm": 0.35373235092972494,
+      "learning_rate": 4.320694108461868e-05,
+      "loss": 0.4738,
+      "step": 161
+    },
+    {
+      "epoch": 2.6557377049180326,
+      "grad_norm": 0.3602393611325291,
+      "learning_rate": 4.2749589998939366e-05,
+      "loss": 0.4783,
+      "step": 162
+    },
+    {
+      "epoch": 2.6721311475409837,
+      "grad_norm": 0.33028586314976544,
+      "learning_rate": 4.2291877451964e-05,
+      "loss": 0.4807,
+      "step": 163
+    },
+    {
+      "epoch": 2.6885245901639343,
+      "grad_norm": 0.27454726727065487,
+      "learning_rate": 4.183386361461643e-05,
+      "loss": 0.4754,
+      "step": 164
+    },
+    {
+      "epoch": 2.7049180327868854,
+      "grad_norm": 0.2201982880819997,
+      "learning_rate": 4.137560869742816e-05,
+      "loss": 0.4816,
+      "step": 165
+    },
+    {
+      "epoch": 2.721311475409836,
+      "grad_norm": 0.24024736204090635,
+      "learning_rate": 4.0917172942623074e-05,
+      "loss": 0.477,
+      "step": 166
+    },
+    {
+      "epoch": 2.737704918032787,
+      "grad_norm": 0.2774042453950074,
+      "learning_rate": 4.045861661619796e-05,
+      "loss": 0.4863,
+      "step": 167
+    },
+    {
+      "epoch": 2.7540983606557377,
+      "grad_norm": 0.25031478188990913,
+      "learning_rate": 4e-05,
+      "loss": 0.4806,
+      "step": 168
+    },
+    {
+      "epoch": 2.7704918032786887,
+      "grad_norm": 0.3146343902104566,
+      "learning_rate": 3.954138338380204e-05,
+      "loss": 0.4819,
+      "step": 169
+    },
+    {
+      "epoch": 2.7868852459016393,
+      "grad_norm": 0.29476137692022925,
+      "learning_rate": 3.908282705737694e-05,
+      "loss": 0.4829,
+      "step": 170
+    },
+    {
+      "epoch": 2.80327868852459,
+      "grad_norm": 0.2781979864605831,
+      "learning_rate": 3.862439130257184e-05,
+      "loss": 0.4842,
+      "step": 171
+    },
+    {
+      "epoch": 2.819672131147541,
+      "grad_norm": 0.2812486506160709,
+      "learning_rate": 3.8166136385383585e-05,
+      "loss": 0.4848,
+      "step": 172
+    },
+    {
+      "epoch": 2.836065573770492,
+      "grad_norm": 0.2443137176068297,
+      "learning_rate": 3.770812254803601e-05,
+      "loss": 0.4814,
+      "step": 173
+    },
+    {
+      "epoch": 2.8524590163934427,
+      "grad_norm": 0.2139764225453824,
+      "learning_rate": 3.725041000106065e-05,
+      "loss": 0.4793,
+      "step": 174
+    },
+    {
+      "epoch": 2.8688524590163933,
+      "grad_norm": 0.2771342287766635,
+      "learning_rate": 3.679305891538133e-05,
+      "loss": 0.466,
+      "step": 175
+    },
+    {
+      "epoch": 2.8852459016393444,
+      "grad_norm": 0.26719263531966153,
+      "learning_rate": 3.6336129414404214e-05,
+      "loss": 0.4818,
+      "step": 176
+    },
+    {
+      "epoch": 2.901639344262295,
+      "grad_norm": 0.21705964839463288,
+      "learning_rate": 3.5879681566113836e-05,
+      "loss": 0.4841,
+      "step": 177
+    },
+    {
+      "epoch": 2.918032786885246,
+      "grad_norm": 0.20964241906287742,
+      "learning_rate": 3.5423775375176686e-05,
+      "loss": 0.4857,
+      "step": 178
+    },
+    {
+      "epoch": 2.9344262295081966,
+      "grad_norm": 0.19494840468268942,
+      "learning_rate": 3.4968470775052914e-05,
+      "loss": 0.475,
+      "step": 179
+    },
+    {
+      "epoch": 2.9508196721311473,
+      "grad_norm": 0.19411352485607267,
+      "learning_rate": 3.451382762011753e-05,
+      "loss": 0.4765,
+      "step": 180
+    },
+    {
+      "epoch": 2.9672131147540983,
+      "grad_norm": 0.2389081408757777,
+      "learning_rate": 3.405990567779185e-05,
+      "loss": 0.4843,
+      "step": 181
+    },
+    {
+      "epoch": 2.9836065573770494,
+      "grad_norm": 0.19939844528891387,
+      "learning_rate": 3.3606764620686614e-05,
+      "loss": 0.4807,
+      "step": 182
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.23959467510788,
+      "learning_rate": 3.3154464018757284e-05,
+      "loss": 0.483,
+      "step": 183
+    },
+    {
+      "epoch": 3.0163934426229506,
+      "grad_norm": 0.20927947022485016,
+      "learning_rate": 3.270306333147301e-05,
+      "loss": 0.4458,
+      "step": 184
+    },
+    {
+      "epoch": 3.0327868852459017,
+      "grad_norm": 0.2064164028547752,
+      "learning_rate": 3.225262190000015e-05,
+      "loss": 0.46,
+      "step": 185
+    },
+    {
+      "epoch": 3.0491803278688523,
+      "grad_norm": 0.22391411463107186,
+      "learning_rate": 3.180319893940117e-05,
+      "loss": 0.4488,
+      "step": 186
+    },
+    {
+      "epoch": 3.0655737704918034,
+      "grad_norm": 0.21718594350542614,
+      "learning_rate": 3.135485353085033e-05,
+      "loss": 0.4487,
+      "step": 187
+    },
+    {
+      "epoch": 3.081967213114754,
+      "grad_norm": 0.2467699050215421,
+      "learning_rate": 3.0907644613866795e-05,
+      "loss": 0.4466,
+      "step": 188
+    },
+    {
+      "epoch": 3.098360655737705,
+      "grad_norm": 0.23078659838474802,
+      "learning_rate": 3.0461630978566473e-05,
+      "loss": 0.4423,
+      "step": 189
+    },
+    {
+      "epoch": 3.1147540983606556,
+      "grad_norm": 0.2370592376095119,
+      "learning_rate": 3.0016871257933457e-05,
+      "loss": 0.4438,
+      "step": 190
+    },
+    {
+      "epoch": 3.1311475409836067,
+      "grad_norm": 0.21570100983782378,
+      "learning_rate": 2.9573423920112155e-05,
+      "loss": 0.4455,
+      "step": 191
+    },
+    {
+      "epoch": 3.1475409836065573,
+      "grad_norm": 0.21407395891268235,
+      "learning_rate": 2.9131347260720965e-05,
+      "loss": 0.4463,
+      "step": 192
+    },
+    {
+      "epoch": 3.1639344262295084,
+      "grad_norm": 0.20617008048319857,
+      "learning_rate": 2.8690699395188873e-05,
+      "loss": 0.4465,
+      "step": 193
+    },
+    {
+      "epoch": 3.180327868852459,
+      "grad_norm": 0.2125529545129492,
+      "learning_rate": 2.825153825111553e-05,
+      "loss": 0.445,
+      "step": 194
+    },
+    {
+      "epoch": 3.19672131147541,
+      "grad_norm": 0.2002948390041686,
+      "learning_rate": 2.7813921560656003e-05,
+      "loss": 0.4472,
+      "step": 195
+    },
+    {
+      "epoch": 3.2131147540983607,
+      "grad_norm": 0.19399060354592243,
+      "learning_rate": 2.737790685293145e-05,
+      "loss": 0.4384,
+      "step": 196
+    },
+    {
+      "epoch": 3.2295081967213113,
+      "grad_norm": 0.20413317551543775,
+      "learning_rate": 2.694355144646622e-05,
+      "loss": 0.4474,
+      "step": 197
+    },
+    {
+      "epoch": 3.2459016393442623,
+      "grad_norm": 0.178524571287303,
+      "learning_rate": 2.6510912441652855e-05,
+      "loss": 0.4476,
+      "step": 198
+    },
+    {
+      "epoch": 3.262295081967213,
+      "grad_norm": 0.21495762020528217,
+      "learning_rate": 2.6080046713245527e-05,
+      "loss": 0.4481,
+      "step": 199
+    },
+    {
+      "epoch": 3.278688524590164,
+      "grad_norm": 0.17779692741231456,
+      "learning_rate": 2.5651010902883457e-05,
+      "loss": 0.4447,
+      "step": 200
+    },
+    {
+      "epoch": 3.2950819672131146,
+      "grad_norm": 0.19536488579919992,
+      "learning_rate": 2.5223861411644647e-05,
+      "loss": 0.4406,
+      "step": 201
+    },
+    {
+      "epoch": 3.3114754098360657,
+      "grad_norm": 0.21534013434506138,
+      "learning_rate": 2.4798654392631502e-05,
+      "loss": 0.443,
+      "step": 202
+    },
+    {
+      "epoch": 3.3278688524590163,
+      "grad_norm": 0.17678888902216489,
+      "learning_rate": 2.4375445743588832e-05,
+      "loss": 0.4481,
+      "step": 203
+    },
+    {
+      "epoch": 3.3442622950819674,
+      "grad_norm": 0.21881378852102576,
+      "learning_rate": 2.3954291099555652e-05,
+      "loss": 0.4484,
+      "step": 204
+    },
+    {
+      "epoch": 3.360655737704918,
+      "grad_norm": 0.1911371536452111,
+      "learning_rate": 2.3535245825551285e-05,
+      "loss": 0.4407,
+      "step": 205
+    },
+    {
+      "epoch": 3.3770491803278686,
+      "grad_norm": 0.1822153835411314,
+      "learning_rate": 2.3118365009297145e-05,
+      "loss": 0.4474,
+      "step": 206
+    },
+    {
+      "epoch": 3.3934426229508197,
+      "grad_norm": 0.1967826759425408,
+      "learning_rate": 2.2703703453974883e-05,
+      "loss": 0.4466,
+      "step": 207
+    },
+    {
+      "epoch": 3.4098360655737707,
+      "grad_norm": 0.1787783354685364,
+      "learning_rate": 2.2291315671021922e-05,
+      "loss": 0.4417,
+      "step": 208
+    },
+    {
+      "epoch": 3.4262295081967213,
+      "grad_norm": 0.18368967889880147,
+      "learning_rate": 2.1881255872965434e-05,
+      "loss": 0.4438,
+      "step": 209
+    },
+    {
+      "epoch": 3.442622950819672,
+      "grad_norm": 0.16138571200714233,
+      "learning_rate": 2.1473577966295506e-05,
+      "loss": 0.4479,
+      "step": 210
+    },
+    {
+      "epoch": 3.459016393442623,
+      "grad_norm": 0.15851838029612553,
+      "learning_rate": 2.1068335544378672e-05,
+      "loss": 0.4511,
+      "step": 211
+    },
+    {
+      "epoch": 3.4754098360655736,
+      "grad_norm": 0.15179774543882815,
+      "learning_rate": 2.0665581880412366e-05,
+      "loss": 0.4447,
+      "step": 212
+    },
+    {
+      "epoch": 3.4918032786885247,
+      "grad_norm": 0.15563700688196921,
+      "learning_rate": 2.026536992042184e-05,
+      "loss": 0.4456,
+      "step": 213
+    },
+    {
+      "epoch": 3.5081967213114753,
+      "grad_norm": 0.14825148109958275,
+      "learning_rate": 1.9867752276299686e-05,
+      "loss": 0.4472,
+      "step": 214
+    },
+    {
+      "epoch": 3.5245901639344264,
+      "grad_norm": 0.12432896908408256,
+      "learning_rate": 1.9472781218889565e-05,
+      "loss": 0.4491,
+      "step": 215
+    },
+    {
+      "epoch": 3.540983606557377,
+      "grad_norm": 0.14986131261136762,
+      "learning_rate": 1.908050867111468e-05,
+      "loss": 0.447,
+      "step": 216
+    },
+    {
+      "epoch": 3.557377049180328,
+      "grad_norm": 0.1280996425182516,
+      "learning_rate": 1.869098620115194e-05,
+      "loss": 0.4459,
+      "step": 217
+    },
+    {
+      "epoch": 3.5737704918032787,
+      "grad_norm": 0.15181510097669454,
+      "learning_rate": 1.8304265015652835e-05,
+      "loss": 0.444,
+      "step": 218
+    },
+    {
+      "epoch": 3.5901639344262293,
+      "grad_norm": 0.1402130102169863,
+      "learning_rate": 1.792039595301183e-05,
+      "loss": 0.4521,
+      "step": 219
+    },
+    {
+      "epoch": 3.6065573770491803,
+      "grad_norm": 0.16291621368769718,
+      "learning_rate": 1.7539429476683088e-05,
+      "loss": 0.447,
+      "step": 220
+    },
+    {
+      "epoch": 3.6229508196721314,
+      "grad_norm": 0.13943515675692342,
+      "learning_rate": 1.7161415668546557e-05,
+      "loss": 0.4411,
+      "step": 221
+    },
+    {
+      "epoch": 3.639344262295082,
+      "grad_norm": 0.1419363610523589,
+      "learning_rate": 1.6786404222324278e-05,
+      "loss": 0.442,
+      "step": 222
+    },
+    {
+      "epoch": 3.6557377049180326,
+      "grad_norm": 0.13518707011783945,
+      "learning_rate": 1.641444443704749e-05,
+      "loss": 0.444,
+      "step": 223
+    },
+    {
+      "epoch": 3.6721311475409837,
+      "grad_norm": 0.12940286831893916,
+      "learning_rate": 1.6045585210575955e-05,
+      "loss": 0.448,
+      "step": 224
+    },
+    {
+      "epoch": 3.6885245901639343,
+      "grad_norm": 0.15612865445371793,
+      "learning_rate": 1.56798750331697e-05,
+      "loss": 0.4439,
+      "step": 225
+    },
+    {
+      "epoch": 3.7049180327868854,
+      "grad_norm": 0.13228999124818513,
+      "learning_rate": 1.531736198111456e-05,
+      "loss": 0.4402,
+      "step": 226
+    },
+    {
+      "epoch": 3.721311475409836,
+      "grad_norm": 0.14034136382445142,
+      "learning_rate": 1.4958093710402036e-05,
+      "loss": 0.4394,
+      "step": 227
+    },
+    {
+      "epoch": 3.737704918032787,
+      "grad_norm": 0.14905661892566827,
+      "learning_rate": 1.4602117450464453e-05,
+      "loss": 0.4481,
+      "step": 228
+    },
+    {
+      "epoch": 3.7540983606557377,
+      "grad_norm": 0.16177536977164272,
+      "learning_rate": 1.4249479997966145e-05,
+      "loss": 0.4457,
+      "step": 229
+    },
+    {
+      "epoch": 3.7704918032786887,
+      "grad_norm": 0.12677113397589643,
+      "learning_rate": 1.3900227710651647e-05,
+      "loss": 0.4445,
+      "step": 230
+    },
+    {
+      "epoch": 3.7868852459016393,
+      "grad_norm": 0.17192511217282086,
+      "learning_rate": 1.3554406501251332e-05,
+      "loss": 0.4459,
+      "step": 231
+    },
+    {
+      "epoch": 3.80327868852459,
+      "grad_norm": 0.11920529678713093,
+      "learning_rate": 1.3212061831445913e-05,
+      "loss": 0.4382,
+      "step": 232
+    },
+    {
+      "epoch": 3.819672131147541,
+      "grad_norm": 0.12712529888330631,
+      "learning_rate": 1.2873238705889968e-05,
+      "loss": 0.441,
+      "step": 233
+    },
+    {
+      "epoch": 3.836065573770492,
+      "grad_norm": 0.12199439925977827,
+      "learning_rate": 1.2537981666295536e-05,
+      "loss": 0.442,
+      "step": 234
+    },
+    {
+      "epoch": 3.8524590163934427,
+      "grad_norm": 0.13371713843483143,
+      "learning_rate": 1.2206334785576822e-05,
+      "loss": 0.4381,
+      "step": 235
+    },
+    {
+      "epoch": 3.8688524590163933,
+      "grad_norm": 0.11420461356422432,
+      "learning_rate": 1.1878341662056259e-05,
+      "loss": 0.4442,
+      "step": 236
+    },
+    {
+      "epoch": 3.8852459016393444,
+      "grad_norm": 0.11704524464444101,
+      "learning_rate": 1.1554045413733097e-05,
+      "loss": 0.4435,
+      "step": 237
+    },
+    {
+      "epoch": 3.901639344262295,
+      "grad_norm": 0.11799803709109379,
+      "learning_rate": 1.123348867261512e-05,
+      "loss": 0.4444,
+      "step": 238
+    },
+    {
+      "epoch": 3.918032786885246,
+      "grad_norm": 0.12095672068157937,
+      "learning_rate": 1.0916713579114218e-05,
+      "loss": 0.4425,
+      "step": 239
+    },
+    {
+      "epoch": 3.9344262295081966,
+      "grad_norm": 0.10579439481053957,
+      "learning_rate": 1.0603761776506616e-05,
+      "loss": 0.4448,
+      "step": 240
+    },
+    {
+      "epoch": 3.9508196721311473,
+      "grad_norm": 0.12927283153010521,
+      "learning_rate": 1.029467440545849e-05,
+      "loss": 0.444,
+      "step": 241
+    },
+    {
+      "epoch": 3.9672131147540983,
+      "grad_norm": 0.11979578396485492,
+      "learning_rate": 9.989492098617494e-06,
+      "loss": 0.4462,
+      "step": 242
+    },
+    {
+      "epoch": 3.9836065573770494,
+      "grad_norm": 0.11270512581429702,
+      "learning_rate": 9.688254975271327e-06,
+      "loss": 0.4439,
+      "step": 243
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.12212819186751334,
+      "learning_rate": 9.391002636073589e-06,
+      "loss": 0.427,
+      "step": 244
+    },
+    {
+      "epoch": 4.016393442622951,
+      "grad_norm": 0.15550079584648527,
+      "learning_rate": 9.0977741578378e-06,
+      "loss": 0.4269,
+      "step": 245
+    },
+    {
+      "epoch": 4.032786885245901,
+      "grad_norm": 0.13688670983262652,
+      "learning_rate": 8.808608088400522e-06,
+      "loss": 0.4269,
+      "step": 246
+    },
+    {
+      "epoch": 4.049180327868853,
+      "grad_norm": 0.11888941041824816,
+      "learning_rate": 8.523542441553712e-06,
+      "loss": 0.4212,
+      "step": 247
+    },
+    {
+      "epoch": 4.065573770491803,
+      "grad_norm": 0.13424666760515866,
+      "learning_rate": 8.242614692047507e-06,
+      "loss": 0.4288,
+      "step": 248
+    },
+    {
+      "epoch": 4.081967213114754,
+      "grad_norm": 0.15658242982421974,
+      "learning_rate": 7.96586177066375e-06,
+      "loss": 0.4234,
+      "step": 249
+    },
+    {
+      "epoch": 4.098360655737705,
+      "grad_norm": 0.14858315425443683,
+      "learning_rate": 7.693320059361107e-06,
+      "loss": 0.4236,
+      "step": 250
+    },
+    {
+      "epoch": 4.114754098360656,
+      "grad_norm": 0.1432177280213629,
+      "learning_rate": 7.425025386492231e-06,
+      "loss": 0.4324,
+      "step": 251
+    },
+    {
+      "epoch": 4.131147540983607,
+      "grad_norm": 0.1363328212537908,
+      "learning_rate": 7.161013022093861e-06,
+      "loss": 0.4305,
+      "step": 252
+    },
+    {
+      "epoch": 4.147540983606557,
+      "grad_norm": 0.13831059143305544,
+      "learning_rate": 6.901317673250089e-06,
+      "loss": 0.4198,
+      "step": 253
+    },
+    {
+      "epoch": 4.163934426229508,
+      "grad_norm": 0.1479605646882516,
+      "learning_rate": 6.645973479529892e-06,
+      "loss": 0.4275,
+      "step": 254
+    },
+    {
+      "epoch": 4.180327868852459,
+      "grad_norm": 0.12133684415312865,
+      "learning_rate": 6.395014008499067e-06,
+      "loss": 0.4286,
+      "step": 255
+    },
+    {
+      "epoch": 4.19672131147541,
+      "grad_norm": 0.11552099846590175,
+      "learning_rate": 6.148472251307436e-06,
+      "loss": 0.4249,
+      "step": 256
+    },
+    {
+      "epoch": 4.213114754098361,
+      "grad_norm": 0.1286344544278559,
+      "learning_rate": 5.906380618351893e-06,
+      "loss": 0.4266,
+      "step": 257
+    },
+    {
+      "epoch": 4.229508196721311,
+      "grad_norm": 0.12137839353942781,
+      "learning_rate": 5.668770935015641e-06,
+      "loss": 0.4193,
+      "step": 258
+    },
+    {
+      "epoch": 4.245901639344262,
+      "grad_norm": 0.10879456442085193,
+      "learning_rate": 5.435674437484491e-06,
+      "loss": 0.4213,
+      "step": 259
+    },
+    {
+      "epoch": 4.262295081967213,
+      "grad_norm": 0.10654250511384006,
+      "learning_rate": 5.207121768640519e-06,
+      "loss": 0.415,
+      "step": 260
+    },
+    {
+      "epoch": 4.278688524590164,
+      "grad_norm": 0.11634254635354951,
+      "learning_rate": 4.983142974033764e-06,
+      "loss": 0.4278,
+      "step": 261
+    },
+    {
+      "epoch": 4.295081967213115,
+      "grad_norm": 0.1188048854614297,
+      "learning_rate": 4.763767497932419e-06,
+      "loss": 0.4186,
+      "step": 262
+    },
+    {
+      "epoch": 4.311475409836065,
+      "grad_norm": 0.10499559306706431,
+      "learning_rate": 4.549024179452151e-06,
+      "loss": 0.4192,
+      "step": 263
+    },
+    {
+      "epoch": 4.327868852459017,
+      "grad_norm": 0.10258020778548946,
+      "learning_rate": 4.338941248764781e-06,
+      "loss": 0.4221,
+      "step": 264
+    },
+    {
+      "epoch": 4.344262295081967,
+      "grad_norm": 0.10174308704204628,
+      "learning_rate": 4.133546323387281e-06,
+      "loss": 0.4291,
+      "step": 265
+    },
+    {
+      "epoch": 4.360655737704918,
+      "grad_norm": 0.09609276817159343,
+      "learning_rate": 3.932866404551092e-06,
+      "loss": 0.4183,
+      "step": 266
+    },
+    {
+      "epoch": 4.377049180327869,
+      "grad_norm": 0.11045174230896027,
+      "learning_rate": 3.7369278736525584e-06,
+      "loss": 0.4237,
+      "step": 267
+    },
+    {
+      "epoch": 4.39344262295082,
+      "grad_norm": 0.09919053798520305,
+      "learning_rate": 3.545756488784826e-06,
+      "loss": 0.4247,
+      "step": 268
+    },
+    {
+      "epoch": 4.409836065573771,
+      "grad_norm": 0.10170621462280902,
+      "learning_rate": 3.359377381351707e-06,
+      "loss": 0.4162,
+      "step": 269
+    },
+    {
+      "epoch": 4.426229508196721,
+      "grad_norm": 0.09599940271102718,
+      "learning_rate": 3.1778150527638664e-06,
+      "loss": 0.425,
+      "step": 270
+    },
+    {
+      "epoch": 4.442622950819672,
+      "grad_norm": 0.09230359083631827,
+      "learning_rate": 3.0010933712178735e-06,
+      "loss": 0.425,
+      "step": 271
+    },
+    {
+      "epoch": 4.459016393442623,
+      "grad_norm": 0.09161534788091766,
+      "learning_rate": 2.8292355685585147e-06,
+      "loss": 0.431,
+      "step": 272
+    },
+    {
+      "epoch": 4.475409836065574,
+      "grad_norm": 0.09181661135879722,
+      "learning_rate": 2.6622642372246874e-06,
+      "loss": 0.4202,
+      "step": 273
+    },
+    {
+      "epoch": 4.491803278688525,
+      "grad_norm": 0.09201200891820971,
+      "learning_rate": 2.5002013272794654e-06,
+      "loss": 0.4199,
+      "step": 274
+    },
+    {
+      "epoch": 4.508196721311475,
+      "grad_norm": 0.09965795343039312,
+      "learning_rate": 2.3430681435244386e-06,
+      "loss": 0.4268,
+      "step": 275
+    },
+    {
+      "epoch": 4.524590163934426,
+      "grad_norm": 0.09118520040005619,
+      "learning_rate": 2.1908853426990983e-06,
+      "loss": 0.4251,
+      "step": 276
+    },
+    {
+      "epoch": 4.540983606557377,
+      "grad_norm": 0.0928501159245484,
+      "learning_rate": 2.0436729307652213e-06,
+      "loss": 0.4213,
+      "step": 277
+    },
+    {
+      "epoch": 4.557377049180328,
+      "grad_norm": 0.09183472059851619,
+      "learning_rate": 1.9014502602769003e-06,
+      "loss": 0.4245,
+      "step": 278
+    },
+    {
+      "epoch": 4.573770491803279,
+      "grad_norm": 0.08782980102005061,
+      "learning_rate": 1.7642360278364767e-06,
+      "loss": 0.4233,
+      "step": 279
+    },
+    {
+      "epoch": 4.590163934426229,
+      "grad_norm": 0.09375666869538167,
+      "learning_rate": 1.6320482716366726e-06,
+      "loss": 0.425,
+      "step": 280
+    },
+    {
+      "epoch": 4.60655737704918,
+      "grad_norm": 0.08785996758165288,
+      "learning_rate": 1.5049043690892729e-06,
+      "loss": 0.4145,
+      "step": 281
+    },
+    {
+      "epoch": 4.622950819672131,
+      "grad_norm": 0.09413477632311608,
+      "learning_rate": 1.3828210345407268e-06,
+      "loss": 0.4216,
+      "step": 282
+    },
+    {
+      "epoch": 4.639344262295082,
+      "grad_norm": 0.09447484104691518,
+      "learning_rate": 1.2658143170748383e-06,
+      "loss": 0.4201,
+      "step": 283
+    },
+    {
+      "epoch": 4.655737704918033,
+      "grad_norm": 0.09229151843804058,
+      "learning_rate": 1.1538995984029654e-06,
+      "loss": 0.4163,
+      "step": 284
+    },
+    {
+      "epoch": 4.672131147540983,
+      "grad_norm": 0.08568358284035806,
+      "learning_rate": 1.047091590841971e-06,
+      "loss": 0.4263,
+      "step": 285
+    },
+    {
+      "epoch": 4.688524590163935,
+      "grad_norm": 0.09023433045860603,
+      "learning_rate": 9.454043353800801e-07,
+      "loss": 0.4221,
+      "step": 286
+    },
+    {
+      "epoch": 4.704918032786885,
+      "grad_norm": 0.0844734735476981,
+      "learning_rate": 8.488511998311133e-07,
+      "loss": 0.422,
+      "step": 287
+    },
+    {
+      "epoch": 4.721311475409836,
+      "grad_norm": 0.08564054109423068,
+      "learning_rate": 7.574448770771003e-07,
+      "loss": 0.4226,
+      "step": 288
+    },
+    {
+      "epoch": 4.737704918032787,
+      "grad_norm": 0.08653160576715448,
+      "learning_rate": 6.711973833997088e-07,
+      "loss": 0.4179,
+      "step": 289
+    },
+    {
+      "epoch": 4.754098360655737,
+      "grad_norm": 0.08383004695421666,
+      "learning_rate": 5.90120056900565e-07,
+      "loss": 0.415,
+      "step": 290
+    },
+    {
+      "epoch": 4.770491803278689,
+      "grad_norm": 0.08834395449932624,
+      "learning_rate": 5.142235560107623e-07,
+      "loss": 0.4263,
+      "step": 291
+    },
+    {
+      "epoch": 4.786885245901639,
+      "grad_norm": 0.08320061631305875,
+      "learning_rate": 4.4351785808967485e-07,
+      "loss": 0.4272,
+      "step": 292
+    },
+    {
+      "epoch": 4.80327868852459,
+      "grad_norm": 0.08304223152666458,
+      "learning_rate": 3.780122581133672e-07,
+      "loss": 0.4223,
+      "step": 293
+    },
+    {
+      "epoch": 4.8196721311475414,
+      "grad_norm": 0.08927719387363141,
+      "learning_rate": 3.177153674526556e-07,
+      "loss": 0.4262,
+      "step": 294
+    },
+    {
+      "epoch": 4.836065573770492,
+      "grad_norm": 0.08297799431829674,
+      "learning_rate": 2.626351127410498e-07,
+      "loss": 0.4223,
+      "step": 295
+    },
+    {
+      "epoch": 4.852459016393443,
+      "grad_norm": 0.08318865729803561,
+      "learning_rate": 2.1277873483274148e-07,
+      "loss": 0.4259,
+      "step": 296
+    },
+    {
+      "epoch": 4.868852459016393,
+      "grad_norm": 0.0828464753884451,
+      "learning_rate": 1.6815278785069944e-07,
+      "loss": 0.4235,
+      "step": 297
+    },
+    {
+      "epoch": 4.885245901639344,
+      "grad_norm": 0.08134276640655788,
+      "learning_rate": 1.2876313832508314e-07,
+      "loss": 0.4193,
+      "step": 298
+    },
+    {
+      "epoch": 4.901639344262295,
+      "grad_norm": 0.08124195226061579,
+      "learning_rate": 9.461496442199735e-08,
+      "loss": 0.417,
+      "step": 299
+    },
+    {
+      "epoch": 4.918032786885246,
+      "grad_norm": 0.08525708385313023,
+      "learning_rate": 6.571275526281451e-08,
+      "loss": 0.4301,
+      "step": 300
+    },
+    {
+      "epoch": 4.934426229508197,
+      "grad_norm": 0.08475437381287206,
+      "learning_rate": 4.206031033399338e-08,
+      "loss": 0.4309,
+      "step": 301
+    },
+    {
+      "epoch": 4.950819672131147,
+      "grad_norm": 0.08111110985986,
+      "learning_rate": 2.366073898761645e-08,
+      "loss": 0.4143,
+      "step": 302
+    },
+    {
+      "epoch": 4.967213114754099,
+      "grad_norm": 0.08973114402396704,
+      "learning_rate": 1.0516460032641284e-08,
+      "loss": 0.4301,
+      "step": 303
+    },
+    {
+      "epoch": 4.983606557377049,
+      "grad_norm": 0.07995200669593748,
+      "learning_rate": 2.6292014169104764e-09,
+      "loss": 0.418,
+      "step": 304
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.0879869455035565,
+      "learning_rate": 0.0,
+      "loss": 0.4146,
+      "step": 305
+    },
+    {
+      "epoch": 5.0,
+      "step": 305,
+      "total_flos": 5116723388743680.0,
+      "train_loss": 0.5099891699728418,
+      "train_runtime": 4463.8352,
+      "train_samples_per_second": 34.825,
+      "train_steps_per_second": 0.068
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 305,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5116723388743680.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f6f4f46dfd408429588a977d00eca414716e9c0ed86bb0b2ee8dafa01ffcfcc2
+size 7352