Upload folder using huggingface_hub

.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: mlfoundations-dev_code-stratos-unverified-scaled-0.5_stratos_7b
+  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. -->
+
+# mlfoundations-dev_code-stratos-unverified-scaled-0.5_stratos_7b
+
+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/code-stratos-unverified-scaled-0.5 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: 1e-05
+- train_batch_size: 1
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 96
+- total_train_batch_size: 96
+- total_eval_batch_size: 768
+- 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: 3.0
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.46.1
+- Pytorch 2.5.0a0+b465a5843b.nv24.09
+- Datasets 3.0.2
+- 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": 3.0,
+    "total_flos": 582499086630912.0,
+    "train_loss": 0.6133231129019688,
+    "train_runtime": 4436.2424,
+    "train_samples_per_second": 6.73,
+    "train_steps_per_second": 0.07
+}

config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "/leonardo_work/EUHPC_E03_068/marianna/models/Qwen/Qwen2.5-7B-Instruct",
+  "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 sha256:35043dc131aa6d56737fd8ddb026fe65c49330646b5b0abdaf189e491048af78
+size 4877660776

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:abf0b60cd3cab0e38d810d801cd1a77fd891c1ccd841c13843de8cb3c1e3b1f1
+size 4932751008

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94a7510d142871ebf3e8a567e4c0468f9c091568801f9a2aa9769d5521e7b0f0
+size 4330865200

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ec42f92ee12619e8e6340eb4726171573e9200c9d5f2ae1b9f44b66062b3ac8
+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
+  }
+}

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": 3.0,
+    "total_flos": 582499086630912.0,
+    "train_loss": 0.6133231129019688,
+    "train_runtime": 4436.2424,
+    "train_samples_per_second": 6.73,
+    "train_steps_per_second": 0.07
+}

trainer_log.jsonl

@@ -0,0 +1,313 @@
+{"current_steps": 1, "total_steps": 312, "loss": 1.0512, "lr": 3.125e-07, "epoch": 0.009615384615384616, "percentage": 0.32, "elapsed_time": "0:00:19", "remaining_time": "1:40:15"}
+{"current_steps": 2, "total_steps": 312, "loss": 1.0592, "lr": 6.25e-07, "epoch": 0.019230769230769232, "percentage": 0.64, "elapsed_time": "0:00:33", "remaining_time": "1:25:42"}
+{"current_steps": 3, "total_steps": 312, "loss": 1.0391, "lr": 9.375000000000001e-07, "epoch": 0.028846153846153848, "percentage": 0.96, "elapsed_time": "0:00:48", "remaining_time": "1:23:24"}
+{"current_steps": 4, "total_steps": 312, "loss": 1.0671, "lr": 1.25e-06, "epoch": 0.038461538461538464, "percentage": 1.28, "elapsed_time": "0:01:02", "remaining_time": "1:20:50"}
+{"current_steps": 5, "total_steps": 312, "loss": 1.0551, "lr": 1.5625e-06, "epoch": 0.04807692307692308, "percentage": 1.6, "elapsed_time": "0:01:15", "remaining_time": "1:17:23"}
+{"current_steps": 6, "total_steps": 312, "loss": 1.0392, "lr": 1.8750000000000003e-06, "epoch": 0.057692307692307696, "percentage": 1.92, "elapsed_time": "0:01:28", "remaining_time": "1:15:07"}
+{"current_steps": 7, "total_steps": 312, "loss": 0.9965, "lr": 2.1875000000000002e-06, "epoch": 0.0673076923076923, "percentage": 2.24, "elapsed_time": "0:01:43", "remaining_time": "1:15:23"}
+{"current_steps": 8, "total_steps": 312, "loss": 0.9904, "lr": 2.5e-06, "epoch": 0.07692307692307693, "percentage": 2.56, "elapsed_time": "0:01:56", "remaining_time": "1:13:55"}
+{"current_steps": 9, "total_steps": 312, "loss": 0.9546, "lr": 2.8125e-06, "epoch": 0.08653846153846154, "percentage": 2.88, "elapsed_time": "0:02:09", "remaining_time": "1:12:35"}
+{"current_steps": 10, "total_steps": 312, "loss": 0.9474, "lr": 3.125e-06, "epoch": 0.09615384615384616, "percentage": 3.21, "elapsed_time": "0:02:21", "remaining_time": "1:11:16"}
+{"current_steps": 11, "total_steps": 312, "loss": 0.9387, "lr": 3.4375e-06, "epoch": 0.10576923076923077, "percentage": 3.53, "elapsed_time": "0:02:36", "remaining_time": "1:11:10"}
+{"current_steps": 12, "total_steps": 312, "loss": 0.9115, "lr": 3.7500000000000005e-06, "epoch": 0.11538461538461539, "percentage": 3.85, "elapsed_time": "0:02:51", "remaining_time": "1:11:30"}
+{"current_steps": 13, "total_steps": 312, "loss": 0.9166, "lr": 4.0625000000000005e-06, "epoch": 0.125, "percentage": 4.17, "elapsed_time": "0:03:06", "remaining_time": "1:11:31"}
+{"current_steps": 14, "total_steps": 312, "loss": 0.8974, "lr": 4.3750000000000005e-06, "epoch": 0.1346153846153846, "percentage": 4.49, "elapsed_time": "0:03:20", "remaining_time": "1:11:17"}
+{"current_steps": 15, "total_steps": 312, "loss": 0.8862, "lr": 4.6875000000000004e-06, "epoch": 0.14423076923076922, "percentage": 4.81, "elapsed_time": "0:03:36", "remaining_time": "1:11:24"}
+{"current_steps": 16, "total_steps": 312, "loss": 0.8618, "lr": 5e-06, "epoch": 0.15384615384615385, "percentage": 5.13, "elapsed_time": "0:03:50", "remaining_time": "1:11:06"}
+{"current_steps": 17, "total_steps": 312, "loss": 0.8466, "lr": 5.3125e-06, "epoch": 0.16346153846153846, "percentage": 5.45, "elapsed_time": "0:04:04", "remaining_time": "1:10:45"}
+{"current_steps": 18, "total_steps": 312, "loss": 0.8225, "lr": 5.625e-06, "epoch": 0.17307692307692307, "percentage": 5.77, "elapsed_time": "0:04:17", "remaining_time": "1:10:09"}
+{"current_steps": 19, "total_steps": 312, "loss": 0.7979, "lr": 5.9375e-06, "epoch": 0.18269230769230768, "percentage": 6.09, "elapsed_time": "0:04:30", "remaining_time": "1:09:30"}
+{"current_steps": 20, "total_steps": 312, "loss": 0.8085, "lr": 6.25e-06, "epoch": 0.19230769230769232, "percentage": 6.41, "elapsed_time": "0:04:43", "remaining_time": "1:08:59"}
+{"current_steps": 21, "total_steps": 312, "loss": 0.7795, "lr": 6.5625e-06, "epoch": 0.20192307692307693, "percentage": 6.73, "elapsed_time": "0:04:59", "remaining_time": "1:09:14"}
+{"current_steps": 22, "total_steps": 312, "loss": 0.7871, "lr": 6.875e-06, "epoch": 0.21153846153846154, "percentage": 7.05, "elapsed_time": "0:05:15", "remaining_time": "1:09:17"}
+{"current_steps": 23, "total_steps": 312, "loss": 0.7624, "lr": 7.1875e-06, "epoch": 0.22115384615384615, "percentage": 7.37, "elapsed_time": "0:05:28", "remaining_time": "1:08:49"}
+{"current_steps": 24, "total_steps": 312, "loss": 0.7557, "lr": 7.500000000000001e-06, "epoch": 0.23076923076923078, "percentage": 7.69, "elapsed_time": "0:05:42", "remaining_time": "1:08:34"}
+{"current_steps": 25, "total_steps": 312, "loss": 0.7342, "lr": 7.8125e-06, "epoch": 0.2403846153846154, "percentage": 8.01, "elapsed_time": "0:05:57", "remaining_time": "1:08:25"}
+{"current_steps": 26, "total_steps": 312, "loss": 0.7372, "lr": 8.125000000000001e-06, "epoch": 0.25, "percentage": 8.33, "elapsed_time": "0:06:10", "remaining_time": "1:07:56"}
+{"current_steps": 27, "total_steps": 312, "loss": 0.7375, "lr": 8.4375e-06, "epoch": 0.25961538461538464, "percentage": 8.65, "elapsed_time": "0:06:29", "remaining_time": "1:08:29"}
+{"current_steps": 28, "total_steps": 312, "loss": 0.7147, "lr": 8.750000000000001e-06, "epoch": 0.2692307692307692, "percentage": 8.97, "elapsed_time": "0:06:42", "remaining_time": "1:08:00"}
+{"current_steps": 29, "total_steps": 312, "loss": 0.7287, "lr": 9.0625e-06, "epoch": 0.27884615384615385, "percentage": 9.29, "elapsed_time": "0:06:56", "remaining_time": "1:07:46"}
+{"current_steps": 30, "total_steps": 312, "loss": 0.7046, "lr": 9.375000000000001e-06, "epoch": 0.28846153846153844, "percentage": 9.62, "elapsed_time": "0:07:09", "remaining_time": "1:07:19"}
+{"current_steps": 31, "total_steps": 312, "loss": 0.701, "lr": 9.6875e-06, "epoch": 0.2980769230769231, "percentage": 9.94, "elapsed_time": "0:07:22", "remaining_time": "1:06:48"}
+{"current_steps": 32, "total_steps": 312, "loss": 0.7091, "lr": 1e-05, "epoch": 0.3076923076923077, "percentage": 10.26, "elapsed_time": "0:07:34", "remaining_time": "1:06:21"}
+{"current_steps": 33, "total_steps": 312, "loss": 0.71, "lr": 9.999685283773504e-06, "epoch": 0.3173076923076923, "percentage": 10.58, "elapsed_time": "0:07:47", "remaining_time": "1:05:52"}
+{"current_steps": 34, "total_steps": 312, "loss": 0.6876, "lr": 9.998741174712534e-06, "epoch": 0.3269230769230769, "percentage": 10.9, "elapsed_time": "0:08:00", "remaining_time": "1:05:27"}
+{"current_steps": 35, "total_steps": 312, "loss": 0.7019, "lr": 9.997167791667668e-06, "epoch": 0.33653846153846156, "percentage": 11.22, "elapsed_time": "0:08:13", "remaining_time": "1:05:03"}
+{"current_steps": 36, "total_steps": 312, "loss": 0.6779, "lr": 9.994965332706574e-06, "epoch": 0.34615384615384615, "percentage": 11.54, "elapsed_time": "0:08:27", "remaining_time": "1:04:50"}
+{"current_steps": 37, "total_steps": 312, "loss": 0.6753, "lr": 9.992134075089085e-06, "epoch": 0.3557692307692308, "percentage": 11.86, "elapsed_time": "0:08:39", "remaining_time": "1:04:22"}
+{"current_steps": 38, "total_steps": 312, "loss": 0.7009, "lr": 9.98867437523228e-06, "epoch": 0.36538461538461536, "percentage": 12.18, "elapsed_time": "0:08:51", "remaining_time": "1:03:55"}
+{"current_steps": 39, "total_steps": 312, "loss": 0.6828, "lr": 9.984586668665641e-06, "epoch": 0.375, "percentage": 12.5, "elapsed_time": "0:09:06", "remaining_time": "1:03:42"}
+{"current_steps": 40, "total_steps": 312, "loss": 0.6833, "lr": 9.979871469976197e-06, "epoch": 0.38461538461538464, "percentage": 12.82, "elapsed_time": "0:09:21", "remaining_time": "1:03:36"}
+{"current_steps": 41, "total_steps": 312, "loss": 0.6711, "lr": 9.974529372743762e-06, "epoch": 0.3942307692307692, "percentage": 13.14, "elapsed_time": "0:09:41", "remaining_time": "1:04:04"}
+{"current_steps": 42, "total_steps": 312, "loss": 0.6767, "lr": 9.968561049466214e-06, "epoch": 0.40384615384615385, "percentage": 13.46, "elapsed_time": "0:09:54", "remaining_time": "1:03:42"}
+{"current_steps": 43, "total_steps": 312, "loss": 0.6519, "lr": 9.961967251474823e-06, "epoch": 0.41346153846153844, "percentage": 13.78, "elapsed_time": "0:10:07", "remaining_time": "1:03:21"}
+{"current_steps": 44, "total_steps": 312, "loss": 0.67, "lr": 9.954748808839675e-06, "epoch": 0.4230769230769231, "percentage": 14.1, "elapsed_time": "0:10:21", "remaining_time": "1:03:05"}
+{"current_steps": 45, "total_steps": 312, "loss": 0.6659, "lr": 9.946906630265184e-06, "epoch": 0.4326923076923077, "percentage": 14.42, "elapsed_time": "0:10:34", "remaining_time": "1:02:42"}
+{"current_steps": 46, "total_steps": 312, "loss": 0.6758, "lr": 9.938441702975689e-06, "epoch": 0.4423076923076923, "percentage": 14.74, "elapsed_time": "0:10:48", "remaining_time": "1:02:28"}
+{"current_steps": 47, "total_steps": 312, "loss": 0.6537, "lr": 9.92935509259118e-06, "epoch": 0.4519230769230769, "percentage": 15.06, "elapsed_time": "0:11:01", "remaining_time": "1:02:08"}
+{"current_steps": 48, "total_steps": 312, "loss": 0.6674, "lr": 9.91964794299315e-06, "epoch": 0.46153846153846156, "percentage": 15.38, "elapsed_time": "0:11:16", "remaining_time": "1:01:59"}
+{"current_steps": 49, "total_steps": 312, "loss": 0.6574, "lr": 9.909321476180594e-06, "epoch": 0.47115384615384615, "percentage": 15.71, "elapsed_time": "0:11:27", "remaining_time": "1:01:31"}
+{"current_steps": 50, "total_steps": 312, "loss": 0.6641, "lr": 9.898376992116179e-06, "epoch": 0.4807692307692308, "percentage": 16.03, "elapsed_time": "0:11:41", "remaining_time": "1:01:15"}
+{"current_steps": 51, "total_steps": 312, "loss": 0.6628, "lr": 9.886815868562596e-06, "epoch": 0.49038461538461536, "percentage": 16.35, "elapsed_time": "0:11:54", "remaining_time": "1:00:57"}
+{"current_steps": 52, "total_steps": 312, "loss": 0.6669, "lr": 9.874639560909118e-06, "epoch": 0.5, "percentage": 16.67, "elapsed_time": "0:12:08", "remaining_time": "1:00:41"}
+{"current_steps": 53, "total_steps": 312, "loss": 0.6642, "lr": 9.861849601988384e-06, "epoch": 0.5096153846153846, "percentage": 16.99, "elapsed_time": "0:12:23", "remaining_time": "1:00:33"}
+{"current_steps": 54, "total_steps": 312, "loss": 0.63, "lr": 9.848447601883436e-06, "epoch": 0.5192307692307693, "percentage": 17.31, "elapsed_time": "0:12:36", "remaining_time": "1:00:16"}
+{"current_steps": 55, "total_steps": 312, "loss": 0.6441, "lr": 9.834435247725032e-06, "epoch": 0.5288461538461539, "percentage": 17.63, "elapsed_time": "0:12:51", "remaining_time": "1:00:04"}
+{"current_steps": 56, "total_steps": 312, "loss": 0.6387, "lr": 9.819814303479268e-06, "epoch": 0.5384615384615384, "percentage": 17.95, "elapsed_time": "0:13:04", "remaining_time": "0:59:46"}
+{"current_steps": 57, "total_steps": 312, "loss": 0.653, "lr": 9.804586609725499e-06, "epoch": 0.5480769230769231, "percentage": 18.27, "elapsed_time": "0:13:17", "remaining_time": "0:59:28"}
+{"current_steps": 58, "total_steps": 312, "loss": 0.6394, "lr": 9.788754083424654e-06, "epoch": 0.5576923076923077, "percentage": 18.59, "elapsed_time": "0:13:31", "remaining_time": "0:59:14"}
+{"current_steps": 59, "total_steps": 312, "loss": 0.6415, "lr": 9.772318717677905e-06, "epoch": 0.5673076923076923, "percentage": 18.91, "elapsed_time": "0:13:46", "remaining_time": "0:59:03"}
+{"current_steps": 60, "total_steps": 312, "loss": 0.6371, "lr": 9.755282581475769e-06, "epoch": 0.5769230769230769, "percentage": 19.23, "elapsed_time": "0:13:59", "remaining_time": "0:58:43"}
+{"current_steps": 61, "total_steps": 312, "loss": 0.6474, "lr": 9.737647819437645e-06, "epoch": 0.5865384615384616, "percentage": 19.55, "elapsed_time": "0:14:11", "remaining_time": "0:58:23"}
+{"current_steps": 62, "total_steps": 312, "loss": 0.6524, "lr": 9.719416651541839e-06, "epoch": 0.5961538461538461, "percentage": 19.87, "elapsed_time": "0:14:24", "remaining_time": "0:58:06"}
+{"current_steps": 63, "total_steps": 312, "loss": 0.6444, "lr": 9.700591372846096e-06, "epoch": 0.6057692307692307, "percentage": 20.19, "elapsed_time": "0:14:37", "remaining_time": "0:57:46"}
+{"current_steps": 64, "total_steps": 312, "loss": 0.6601, "lr": 9.681174353198687e-06, "epoch": 0.6153846153846154, "percentage": 20.51, "elapsed_time": "0:14:52", "remaining_time": "0:57:37"}
+{"current_steps": 65, "total_steps": 312, "loss": 0.6596, "lr": 9.661168036940071e-06, "epoch": 0.625, "percentage": 20.83, "elapsed_time": "0:15:06", "remaining_time": "0:57:23"}
+{"current_steps": 66, "total_steps": 312, "loss": 0.6608, "lr": 9.640574942595195e-06, "epoch": 0.6346153846153846, "percentage": 21.15, "elapsed_time": "0:15:23", "remaining_time": "0:57:20"}
+{"current_steps": 67, "total_steps": 312, "loss": 0.6434, "lr": 9.619397662556434e-06, "epoch": 0.6442307692307693, "percentage": 21.47, "elapsed_time": "0:15:39", "remaining_time": "0:57:14"}
+{"current_steps": 68, "total_steps": 312, "loss": 0.6369, "lr": 9.597638862757255e-06, "epoch": 0.6538461538461539, "percentage": 21.79, "elapsed_time": "0:15:53", "remaining_time": "0:57:03"}
+{"current_steps": 69, "total_steps": 312, "loss": 0.6333, "lr": 9.5753012823366e-06, "epoch": 0.6634615384615384, "percentage": 22.12, "elapsed_time": "0:16:07", "remaining_time": "0:56:46"}
+{"current_steps": 70, "total_steps": 312, "loss": 0.647, "lr": 9.552387733294081e-06, "epoch": 0.6730769230769231, "percentage": 22.44, "elapsed_time": "0:16:20", "remaining_time": "0:56:30"}
+{"current_steps": 71, "total_steps": 312, "loss": 0.6293, "lr": 9.528901100135971e-06, "epoch": 0.6826923076923077, "percentage": 22.76, "elapsed_time": "0:16:34", "remaining_time": "0:56:15"}
+{"current_steps": 72, "total_steps": 312, "loss": 0.6329, "lr": 9.504844339512096e-06, "epoch": 0.6923076923076923, "percentage": 23.08, "elapsed_time": "0:16:49", "remaining_time": "0:56:06"}
+{"current_steps": 73, "total_steps": 312, "loss": 0.6373, "lr": 9.480220479843627e-06, "epoch": 0.7019230769230769, "percentage": 23.4, "elapsed_time": "0:17:02", "remaining_time": "0:55:48"}
+{"current_steps": 74, "total_steps": 312, "loss": 0.647, "lr": 9.45503262094184e-06, "epoch": 0.7115384615384616, "percentage": 23.72, "elapsed_time": "0:17:17", "remaining_time": "0:55:36"}
+{"current_steps": 75, "total_steps": 312, "loss": 0.6469, "lr": 9.4292839336179e-06, "epoch": 0.7211538461538461, "percentage": 24.04, "elapsed_time": "0:17:33", "remaining_time": "0:55:27"}
+{"current_steps": 76, "total_steps": 312, "loss": 0.6316, "lr": 9.40297765928369e-06, "epoch": 0.7307692307692307, "percentage": 24.36, "elapsed_time": "0:17:45", "remaining_time": "0:55:09"}
+{"current_steps": 77, "total_steps": 312, "loss": 0.6369, "lr": 9.376117109543769e-06, "epoch": 0.7403846153846154, "percentage": 24.68, "elapsed_time": "0:17:59", "remaining_time": "0:54:54"}
+{"current_steps": 78, "total_steps": 312, "loss": 0.6435, "lr": 9.348705665778479e-06, "epoch": 0.75, "percentage": 25.0, "elapsed_time": "0:18:12", "remaining_time": "0:54:36"}
+{"current_steps": 79, "total_steps": 312, "loss": 0.6237, "lr": 9.320746778718274e-06, "epoch": 0.7596153846153846, "percentage": 25.32, "elapsed_time": "0:18:26", "remaining_time": "0:54:22"}
+{"current_steps": 80, "total_steps": 312, "loss": 0.6413, "lr": 9.292243968009332e-06, "epoch": 0.7692307692307693, "percentage": 25.64, "elapsed_time": "0:18:42", "remaining_time": "0:54:15"}
+{"current_steps": 81, "total_steps": 312, "loss": 0.6491, "lr": 9.263200821770462e-06, "epoch": 0.7788461538461539, "percentage": 25.96, "elapsed_time": "0:18:57", "remaining_time": "0:54:03"}
+{"current_steps": 82, "total_steps": 312, "loss": 0.6537, "lr": 9.233620996141421e-06, "epoch": 0.7884615384615384, "percentage": 26.28, "elapsed_time": "0:19:10", "remaining_time": "0:53:47"}
+{"current_steps": 83, "total_steps": 312, "loss": 0.6395, "lr": 9.203508214822652e-06, "epoch": 0.7980769230769231, "percentage": 26.6, "elapsed_time": "0:19:24", "remaining_time": "0:53:33"}
+{"current_steps": 84, "total_steps": 312, "loss": 0.6428, "lr": 9.172866268606514e-06, "epoch": 0.8076923076923077, "percentage": 26.92, "elapsed_time": "0:19:36", "remaining_time": "0:53:13"}
+{"current_steps": 85, "total_steps": 312, "loss": 0.6321, "lr": 9.141699014900084e-06, "epoch": 0.8173076923076923, "percentage": 27.24, "elapsed_time": "0:19:49", "remaining_time": "0:52:57"}
+{"current_steps": 86, "total_steps": 312, "loss": 0.6139, "lr": 9.110010377239552e-06, "epoch": 0.8269230769230769, "percentage": 27.56, "elapsed_time": "0:20:02", "remaining_time": "0:52:40"}
+{"current_steps": 87, "total_steps": 312, "loss": 0.6539, "lr": 9.077804344796302e-06, "epoch": 0.8365384615384616, "percentage": 27.88, "elapsed_time": "0:20:19", "remaining_time": "0:52:32"}
+{"current_steps": 88, "total_steps": 312, "loss": 0.634, "lr": 9.045084971874738e-06, "epoch": 0.8461538461538461, "percentage": 28.21, "elapsed_time": "0:20:32", "remaining_time": "0:52:16"}
+{"current_steps": 89, "total_steps": 312, "loss": 0.6336, "lr": 9.011856377401891e-06, "epoch": 0.8557692307692307, "percentage": 28.53, "elapsed_time": "0:20:45", "remaining_time": "0:52:00"}
+{"current_steps": 90, "total_steps": 312, "loss": 0.6327, "lr": 8.978122744408905e-06, "epoch": 0.8653846153846154, "percentage": 28.85, "elapsed_time": "0:21:00", "remaining_time": "0:51:48"}
+{"current_steps": 91, "total_steps": 312, "loss": 0.6248, "lr": 8.943888319504456e-06, "epoch": 0.875, "percentage": 29.17, "elapsed_time": "0:21:13", "remaining_time": "0:51:33"}
+{"current_steps": 92, "total_steps": 312, "loss": 0.6259, "lr": 8.90915741234015e-06, "epoch": 0.8846153846153846, "percentage": 29.49, "elapsed_time": "0:21:29", "remaining_time": "0:51:23"}
+{"current_steps": 93, "total_steps": 312, "loss": 0.6248, "lr": 8.873934395068006e-06, "epoch": 0.8942307692307693, "percentage": 29.81, "elapsed_time": "0:21:42", "remaining_time": "0:51:07"}
+{"current_steps": 94, "total_steps": 312, "loss": 0.6125, "lr": 8.838223701790057e-06, "epoch": 0.9038461538461539, "percentage": 30.13, "elapsed_time": "0:21:55", "remaining_time": "0:50:51"}
+{"current_steps": 95, "total_steps": 312, "loss": 0.616, "lr": 8.802029828000157e-06, "epoch": 0.9134615384615384, "percentage": 30.45, "elapsed_time": "0:22:10", "remaining_time": "0:50:39"}
+{"current_steps": 96, "total_steps": 312, "loss": 0.6369, "lr": 8.765357330018056e-06, "epoch": 0.9230769230769231, "percentage": 30.77, "elapsed_time": "0:22:23", "remaining_time": "0:50:22"}
+{"current_steps": 97, "total_steps": 312, "loss": 0.6244, "lr": 8.728210824415829e-06, "epoch": 0.9326923076923077, "percentage": 31.09, "elapsed_time": "0:22:36", "remaining_time": "0:50:07"}
+{"current_steps": 98, "total_steps": 312, "loss": 0.6323, "lr": 8.690594987436705e-06, "epoch": 0.9423076923076923, "percentage": 31.41, "elapsed_time": "0:22:52", "remaining_time": "0:49:56"}
+{"current_steps": 99, "total_steps": 312, "loss": 0.6283, "lr": 8.652514554406388e-06, "epoch": 0.9519230769230769, "percentage": 31.73, "elapsed_time": "0:23:05", "remaining_time": "0:49:40"}
+{"current_steps": 100, "total_steps": 312, "loss": 0.6339, "lr": 8.613974319136959e-06, "epoch": 0.9615384615384616, "percentage": 32.05, "elapsed_time": "0:23:19", "remaining_time": "0:49:26"}
+{"current_steps": 101, "total_steps": 312, "loss": 0.6298, "lr": 8.574979133323378e-06, "epoch": 0.9711538461538461, "percentage": 32.37, "elapsed_time": "0:23:58", "remaining_time": "0:50:04"}
+{"current_steps": 102, "total_steps": 312, "loss": 0.6211, "lr": 8.535533905932739e-06, "epoch": 0.9807692307692307, "percentage": 32.69, "elapsed_time": "0:24:11", "remaining_time": "0:49:48"}
+{"current_steps": 103, "total_steps": 312, "loss": 0.6135, "lr": 8.495643602586287e-06, "epoch": 0.9903846153846154, "percentage": 33.01, "elapsed_time": "0:24:25", "remaining_time": "0:49:32"}
+{"current_steps": 104, "total_steps": 312, "loss": 0.6295, "lr": 8.455313244934324e-06, "epoch": 1.0, "percentage": 33.33, "elapsed_time": "0:24:39", "remaining_time": "0:49:19"}
+{"current_steps": 105, "total_steps": 312, "loss": 0.5924, "lr": 8.414547910024035e-06, "epoch": 1.0096153846153846, "percentage": 33.65, "elapsed_time": "0:24:55", "remaining_time": "0:49:07"}
+{"current_steps": 106, "total_steps": 312, "loss": 0.6012, "lr": 8.373352729660373e-06, "epoch": 1.0192307692307692, "percentage": 33.97, "elapsed_time": "0:25:08", "remaining_time": "0:48:51"}
+{"current_steps": 107, "total_steps": 312, "loss": 0.6092, "lr": 8.331732889760021e-06, "epoch": 1.0288461538461537, "percentage": 34.29, "elapsed_time": "0:25:21", "remaining_time": "0:48:35"}
+{"current_steps": 108, "total_steps": 312, "loss": 0.5799, "lr": 8.289693629698564e-06, "epoch": 1.0384615384615385, "percentage": 34.62, "elapsed_time": "0:25:35", "remaining_time": "0:48:20"}
+{"current_steps": 109, "total_steps": 312, "loss": 0.5634, "lr": 8.247240241650918e-06, "epoch": 1.0480769230769231, "percentage": 34.94, "elapsed_time": "0:25:48", "remaining_time": "0:48:03"}
+{"current_steps": 110, "total_steps": 312, "loss": 0.5949, "lr": 8.204378069925121e-06, "epoch": 1.0576923076923077, "percentage": 35.26, "elapsed_time": "0:26:00", "remaining_time": "0:47:46"}
+{"current_steps": 111, "total_steps": 312, "loss": 0.5997, "lr": 8.16111251028955e-06, "epoch": 1.0673076923076923, "percentage": 35.58, "elapsed_time": "0:26:13", "remaining_time": "0:47:30"}
+{"current_steps": 112, "total_steps": 312, "loss": 0.5972, "lr": 8.117449009293668e-06, "epoch": 1.0769230769230769, "percentage": 35.9, "elapsed_time": "0:26:26", "remaining_time": "0:47:13"}
+{"current_steps": 113, "total_steps": 312, "loss": 0.5684, "lr": 8.073393063582386e-06, "epoch": 1.0865384615384615, "percentage": 36.22, "elapsed_time": "0:26:39", "remaining_time": "0:46:56"}
+{"current_steps": 114, "total_steps": 312, "loss": 0.5967, "lr": 8.0289502192041e-06, "epoch": 1.0961538461538463, "percentage": 36.54, "elapsed_time": "0:26:51", "remaining_time": "0:46:39"}
+{"current_steps": 115, "total_steps": 312, "loss": 0.5956, "lr": 7.984126070912519e-06, "epoch": 1.1057692307692308, "percentage": 36.86, "elapsed_time": "0:27:05", "remaining_time": "0:46:24"}
+{"current_steps": 116, "total_steps": 312, "loss": 0.5787, "lr": 7.938926261462366e-06, "epoch": 1.1153846153846154, "percentage": 37.18, "elapsed_time": "0:27:17", "remaining_time": "0:46:06"}
+{"current_steps": 117, "total_steps": 312, "loss": 0.5913, "lr": 7.89335648089903e-06, "epoch": 1.125, "percentage": 37.5, "elapsed_time": "0:27:32", "remaining_time": "0:45:54"}
+{"current_steps": 118, "total_steps": 312, "loss": 0.5776, "lr": 7.84742246584226e-06, "epoch": 1.1346153846153846, "percentage": 37.82, "elapsed_time": "0:27:45", "remaining_time": "0:45:38"}
+{"current_steps": 119, "total_steps": 312, "loss": 0.6113, "lr": 7.801129998764014e-06, "epoch": 1.1442307692307692, "percentage": 38.14, "elapsed_time": "0:27:59", "remaining_time": "0:45:24"}
+{"current_steps": 120, "total_steps": 312, "loss": 0.58, "lr": 7.754484907260513e-06, "epoch": 1.1538461538461537, "percentage": 38.46, "elapsed_time": "0:28:14", "remaining_time": "0:45:11"}
+{"current_steps": 121, "total_steps": 312, "loss": 0.5996, "lr": 7.70749306331863e-06, "epoch": 1.1634615384615385, "percentage": 38.78, "elapsed_time": "0:28:27", "remaining_time": "0:44:56"}
+{"current_steps": 122, "total_steps": 312, "loss": 0.5876, "lr": 7.660160382576683e-06, "epoch": 1.1730769230769231, "percentage": 39.1, "elapsed_time": "0:28:42", "remaining_time": "0:44:42"}
+{"current_steps": 123, "total_steps": 312, "loss": 0.5733, "lr": 7.612492823579744e-06, "epoch": 1.1826923076923077, "percentage": 39.42, "elapsed_time": "0:28:55", "remaining_time": "0:44:25"}
+{"current_steps": 124, "total_steps": 312, "loss": 0.6047, "lr": 7.564496387029532e-06, "epoch": 1.1923076923076923, "percentage": 39.74, "elapsed_time": "0:29:07", "remaining_time": "0:44:10"}
+{"current_steps": 125, "total_steps": 312, "loss": 0.5975, "lr": 7.516177115029002e-06, "epoch": 1.2019230769230769, "percentage": 40.06, "elapsed_time": "0:29:20", "remaining_time": "0:43:53"}
+{"current_steps": 126, "total_steps": 312, "loss": 0.5921, "lr": 7.467541090321735e-06, "epoch": 1.2115384615384615, "percentage": 40.38, "elapsed_time": "0:29:34", "remaining_time": "0:43:39"}
+{"current_steps": 127, "total_steps": 312, "loss": 0.5726, "lr": 7.4185944355261996e-06, "epoch": 1.2211538461538463, "percentage": 40.71, "elapsed_time": "0:29:50", "remaining_time": "0:43:28"}
+{"current_steps": 128, "total_steps": 312, "loss": 0.5844, "lr": 7.369343312364994e-06, "epoch": 1.2307692307692308, "percentage": 41.03, "elapsed_time": "0:30:04", "remaining_time": "0:43:13"}
+{"current_steps": 129, "total_steps": 312, "loss": 0.5758, "lr": 7.319793920889171e-06, "epoch": 1.2403846153846154, "percentage": 41.35, "elapsed_time": "0:30:17", "remaining_time": "0:42:58"}
+{"current_steps": 130, "total_steps": 312, "loss": 0.5936, "lr": 7.269952498697734e-06, "epoch": 1.25, "percentage": 41.67, "elapsed_time": "0:30:30", "remaining_time": "0:42:42"}
+{"current_steps": 131, "total_steps": 312, "loss": 0.591, "lr": 7.219825320152411e-06, "epoch": 1.2596153846153846, "percentage": 41.99, "elapsed_time": "0:30:43", "remaining_time": "0:42:27"}
+{"current_steps": 132, "total_steps": 312, "loss": 0.6117, "lr": 7.169418695587791e-06, "epoch": 1.2692307692307692, "percentage": 42.31, "elapsed_time": "0:30:58", "remaining_time": "0:42:14"}
+{"current_steps": 133, "total_steps": 312, "loss": 0.5771, "lr": 7.118738970516944e-06, "epoch": 1.2788461538461537, "percentage": 42.63, "elapsed_time": "0:31:11", "remaining_time": "0:41:58"}
+{"current_steps": 134, "total_steps": 312, "loss": 0.6068, "lr": 7.067792524832604e-06, "epoch": 1.2884615384615383, "percentage": 42.95, "elapsed_time": "0:31:26", "remaining_time": "0:41:46"}
+{"current_steps": 135, "total_steps": 312, "loss": 0.5748, "lr": 7.016585772004026e-06, "epoch": 1.2980769230769231, "percentage": 43.27, "elapsed_time": "0:31:39", "remaining_time": "0:41:30"}
+{"current_steps": 136, "total_steps": 312, "loss": 0.5829, "lr": 6.965125158269619e-06, "epoch": 1.3076923076923077, "percentage": 43.59, "elapsed_time": "0:31:52", "remaining_time": "0:41:14"}
+{"current_steps": 137, "total_steps": 312, "loss": 0.5679, "lr": 6.913417161825449e-06, "epoch": 1.3173076923076923, "percentage": 43.91, "elapsed_time": "0:32:05", "remaining_time": "0:40:59"}
+{"current_steps": 138, "total_steps": 312, "loss": 0.5679, "lr": 6.8614682920097265e-06, "epoch": 1.3269230769230769, "percentage": 44.23, "elapsed_time": "0:32:19", "remaining_time": "0:40:45"}
+{"current_steps": 139, "total_steps": 312, "loss": 0.5587, "lr": 6.809285088483361e-06, "epoch": 1.3365384615384617, "percentage": 44.55, "elapsed_time": "0:32:34", "remaining_time": "0:40:32"}
+{"current_steps": 140, "total_steps": 312, "loss": 0.5717, "lr": 6.7568741204067145e-06, "epoch": 1.3461538461538463, "percentage": 44.87, "elapsed_time": "0:32:47", "remaining_time": "0:40:16"}
+{"current_steps": 141, "total_steps": 312, "loss": 0.6041, "lr": 6.704241985612625e-06, "epoch": 1.3557692307692308, "percentage": 45.19, "elapsed_time": "0:33:00", "remaining_time": "0:40:01"}
+{"current_steps": 142, "total_steps": 312, "loss": 0.5718, "lr": 6.651395309775837e-06, "epoch": 1.3653846153846154, "percentage": 45.51, "elapsed_time": "0:33:13", "remaining_time": "0:39:47"}
+{"current_steps": 143, "total_steps": 312, "loss": 0.5718, "lr": 6.598340745578908e-06, "epoch": 1.375, "percentage": 45.83, "elapsed_time": "0:33:27", "remaining_time": "0:39:32"}
+{"current_steps": 144, "total_steps": 312, "loss": 0.608, "lr": 6.545084971874738e-06, "epoch": 1.3846153846153846, "percentage": 46.15, "elapsed_time": "0:33:41", "remaining_time": "0:39:18"}
+{"current_steps": 145, "total_steps": 312, "loss": 0.5636, "lr": 6.491634692845781e-06, "epoch": 1.3942307692307692, "percentage": 46.47, "elapsed_time": "0:33:54", "remaining_time": "0:39:02"}
+{"current_steps": 146, "total_steps": 312, "loss": 0.5734, "lr": 6.437996637160086e-06, "epoch": 1.4038461538461537, "percentage": 46.79, "elapsed_time": "0:34:05", "remaining_time": "0:38:46"}
+{"current_steps": 147, "total_steps": 312, "loss": 0.6, "lr": 6.384177557124247e-06, "epoch": 1.4134615384615383, "percentage": 47.12, "elapsed_time": "0:34:19", "remaining_time": "0:38:31"}
+{"current_steps": 148, "total_steps": 312, "loss": 0.5843, "lr": 6.330184227833376e-06, "epoch": 1.4230769230769231, "percentage": 47.44, "elapsed_time": "0:34:31", "remaining_time": "0:38:15"}
+{"current_steps": 149, "total_steps": 312, "loss": 0.5755, "lr": 6.276023446318214e-06, "epoch": 1.4326923076923077, "percentage": 47.76, "elapsed_time": "0:34:44", "remaining_time": "0:38:00"}
+{"current_steps": 150, "total_steps": 312, "loss": 0.5776, "lr": 6.2217020306894705e-06, "epoch": 1.4423076923076923, "percentage": 48.08, "elapsed_time": "0:34:58", "remaining_time": "0:37:46"}
+{"current_steps": 151, "total_steps": 312, "loss": 0.5902, "lr": 6.1672268192795285e-06, "epoch": 1.4519230769230769, "percentage": 48.4, "elapsed_time": "0:35:11", "remaining_time": "0:37:31"}
+{"current_steps": 152, "total_steps": 312, "loss": 0.5825, "lr": 6.112604669781572e-06, "epoch": 1.4615384615384617, "percentage": 48.72, "elapsed_time": "0:35:25", "remaining_time": "0:37:17"}
+{"current_steps": 153, "total_steps": 312, "loss": 0.5854, "lr": 6.057842458386315e-06, "epoch": 1.4711538461538463, "percentage": 49.04, "elapsed_time": "0:35:38", "remaining_time": "0:37:02"}
+{"current_steps": 154, "total_steps": 312, "loss": 0.5766, "lr": 6.002947078916365e-06, "epoch": 1.4807692307692308, "percentage": 49.36, "elapsed_time": "0:35:53", "remaining_time": "0:36:49"}
+{"current_steps": 155, "total_steps": 312, "loss": 0.5834, "lr": 5.947925441958393e-06, "epoch": 1.4903846153846154, "percentage": 49.68, "elapsed_time": "0:36:06", "remaining_time": "0:36:34"}
+{"current_steps": 156, "total_steps": 312, "loss": 0.5856, "lr": 5.892784473993184e-06, "epoch": 1.5, "percentage": 50.0, "elapsed_time": "0:36:21", "remaining_time": "0:36:21"}
+{"current_steps": 157, "total_steps": 312, "loss": 0.5735, "lr": 5.837531116523683e-06, "epoch": 1.5096153846153846, "percentage": 50.32, "elapsed_time": "0:36:37", "remaining_time": "0:36:09"}
+{"current_steps": 158, "total_steps": 312, "loss": 0.5799, "lr": 5.782172325201155e-06, "epoch": 1.5192307692307692, "percentage": 50.64, "elapsed_time": "0:36:49", "remaining_time": "0:35:53"}
+{"current_steps": 159, "total_steps": 312, "loss": 0.6024, "lr": 5.726715068949564e-06, "epoch": 1.5288461538461537, "percentage": 50.96, "elapsed_time": "0:37:02", "remaining_time": "0:35:38"}
+{"current_steps": 160, "total_steps": 312, "loss": 0.5834, "lr": 5.671166329088278e-06, "epoch": 1.5384615384615383, "percentage": 51.28, "elapsed_time": "0:37:15", "remaining_time": "0:35:23"}
+{"current_steps": 161, "total_steps": 312, "loss": 0.5828, "lr": 5.615533098453215e-06, "epoch": 1.5480769230769231, "percentage": 51.6, "elapsed_time": "0:37:28", "remaining_time": "0:35:09"}
+{"current_steps": 162, "total_steps": 312, "loss": 0.5965, "lr": 5.559822380516539e-06, "epoch": 1.5576923076923077, "percentage": 51.92, "elapsed_time": "0:37:42", "remaining_time": "0:34:54"}
+{"current_steps": 163, "total_steps": 312, "loss": 0.572, "lr": 5.504041188505022e-06, "epoch": 1.5673076923076923, "percentage": 52.24, "elapsed_time": "0:37:55", "remaining_time": "0:34:39"}
+{"current_steps": 164, "total_steps": 312, "loss": 0.5698, "lr": 5.448196544517168e-06, "epoch": 1.5769230769230769, "percentage": 52.56, "elapsed_time": "0:38:08", "remaining_time": "0:34:25"}
+{"current_steps": 165, "total_steps": 312, "loss": 0.5943, "lr": 5.392295478639226e-06, "epoch": 1.5865384615384617, "percentage": 52.88, "elapsed_time": "0:38:20", "remaining_time": "0:34:09"}
+{"current_steps": 166, "total_steps": 312, "loss": 0.5881, "lr": 5.336345028060199e-06, "epoch": 1.5961538461538463, "percentage": 53.21, "elapsed_time": "0:38:34", "remaining_time": "0:33:55"}
+{"current_steps": 167, "total_steps": 312, "loss": 0.5482, "lr": 5.2803522361859596e-06, "epoch": 1.6057692307692308, "percentage": 53.53, "elapsed_time": "0:38:47", "remaining_time": "0:33:40"}
+{"current_steps": 168, "total_steps": 312, "loss": 0.5694, "lr": 5.224324151752575e-06, "epoch": 1.6153846153846154, "percentage": 53.85, "elapsed_time": "0:39:01", "remaining_time": "0:33:26"}
+{"current_steps": 169, "total_steps": 312, "loss": 0.5868, "lr": 5.168267827938971e-06, "epoch": 1.625, "percentage": 54.17, "elapsed_time": "0:39:13", "remaining_time": "0:33:11"}
+{"current_steps": 170, "total_steps": 312, "loss": 0.5903, "lr": 5.112190321479026e-06, "epoch": 1.6346153846153846, "percentage": 54.49, "elapsed_time": "0:39:27", "remaining_time": "0:32:57"}
+{"current_steps": 171, "total_steps": 312, "loss": 0.5673, "lr": 5.05609869177323e-06, "epoch": 1.6442307692307692, "percentage": 54.81, "elapsed_time": "0:39:41", "remaining_time": "0:32:43"}
+{"current_steps": 172, "total_steps": 312, "loss": 0.5749, "lr": 5e-06, "epoch": 1.6538461538461537, "percentage": 55.13, "elapsed_time": "0:39:53", "remaining_time": "0:32:28"}
+{"current_steps": 173, "total_steps": 312, "loss": 0.5734, "lr": 4.943901308226771e-06, "epoch": 1.6634615384615383, "percentage": 55.45, "elapsed_time": "0:40:07", "remaining_time": "0:32:14"}
+{"current_steps": 174, "total_steps": 312, "loss": 0.5902, "lr": 4.887809678520976e-06, "epoch": 1.6730769230769231, "percentage": 55.77, "elapsed_time": "0:40:22", "remaining_time": "0:32:01"}
+{"current_steps": 175, "total_steps": 312, "loss": 0.5726, "lr": 4.831732172061032e-06, "epoch": 1.6826923076923077, "percentage": 56.09, "elapsed_time": "0:40:35", "remaining_time": "0:31:46"}
+{"current_steps": 176, "total_steps": 312, "loss": 0.5868, "lr": 4.775675848247427e-06, "epoch": 1.6923076923076923, "percentage": 56.41, "elapsed_time": "0:40:54", "remaining_time": "0:31:36"}
+{"current_steps": 177, "total_steps": 312, "loss": 0.6001, "lr": 4.719647763814041e-06, "epoch": 1.7019230769230769, "percentage": 56.73, "elapsed_time": "0:41:08", "remaining_time": "0:31:22"}
+{"current_steps": 178, "total_steps": 312, "loss": 0.5661, "lr": 4.663654971939802e-06, "epoch": 1.7115384615384617, "percentage": 57.05, "elapsed_time": "0:41:23", "remaining_time": "0:31:09"}
+{"current_steps": 179, "total_steps": 312, "loss": 0.5749, "lr": 4.6077045213607765e-06, "epoch": 1.7211538461538463, "percentage": 57.37, "elapsed_time": "0:41:38", "remaining_time": "0:30:56"}
+{"current_steps": 180, "total_steps": 312, "loss": 0.5781, "lr": 4.551803455482833e-06, "epoch": 1.7307692307692308, "percentage": 57.69, "elapsed_time": "0:41:51", "remaining_time": "0:30:41"}
+{"current_steps": 181, "total_steps": 312, "loss": 0.5713, "lr": 4.4959588114949785e-06, "epoch": 1.7403846153846154, "percentage": 58.01, "elapsed_time": "0:42:05", "remaining_time": "0:30:27"}
+{"current_steps": 182, "total_steps": 312, "loss": 0.5971, "lr": 4.4401776194834615e-06, "epoch": 1.75, "percentage": 58.33, "elapsed_time": "0:42:20", "remaining_time": "0:30:14"}
+{"current_steps": 183, "total_steps": 312, "loss": 0.5757, "lr": 4.384466901546786e-06, "epoch": 1.7596153846153846, "percentage": 58.65, "elapsed_time": "0:42:34", "remaining_time": "0:30:00"}
+{"current_steps": 184, "total_steps": 312, "loss": 0.5823, "lr": 4.3288336709117246e-06, "epoch": 1.7692307692307692, "percentage": 58.97, "elapsed_time": "0:42:47", "remaining_time": "0:29:46"}
+{"current_steps": 185, "total_steps": 312, "loss": 0.5728, "lr": 4.273284931050438e-06, "epoch": 1.7788461538461537, "percentage": 59.29, "elapsed_time": "0:43:02", "remaining_time": "0:29:32"}
+{"current_steps": 186, "total_steps": 312, "loss": 0.5789, "lr": 4.217827674798845e-06, "epoch": 1.7884615384615383, "percentage": 59.62, "elapsed_time": "0:43:15", "remaining_time": "0:29:18"}
+{"current_steps": 187, "total_steps": 312, "loss": 0.5449, "lr": 4.162468883476319e-06, "epoch": 1.7980769230769231, "percentage": 59.94, "elapsed_time": "0:43:28", "remaining_time": "0:29:03"}
+{"current_steps": 188, "total_steps": 312, "loss": 0.5789, "lr": 4.107215526006818e-06, "epoch": 1.8076923076923077, "percentage": 60.26, "elapsed_time": "0:43:42", "remaining_time": "0:28:49"}
+{"current_steps": 189, "total_steps": 312, "loss": 0.5634, "lr": 4.052074558041608e-06, "epoch": 1.8173076923076923, "percentage": 60.58, "elapsed_time": "0:43:54", "remaining_time": "0:28:34"}
+{"current_steps": 190, "total_steps": 312, "loss": 0.5795, "lr": 3.997052921083637e-06, "epoch": 1.8269230769230769, "percentage": 60.9, "elapsed_time": "0:44:07", "remaining_time": "0:28:20"}
+{"current_steps": 191, "total_steps": 312, "loss": 0.5707, "lr": 3.9421575416136866e-06, "epoch": 1.8365384615384617, "percentage": 61.22, "elapsed_time": "0:44:22", "remaining_time": "0:28:06"}
+{"current_steps": 192, "total_steps": 312, "loss": 0.5764, "lr": 3.887395330218429e-06, "epoch": 1.8461538461538463, "percentage": 61.54, "elapsed_time": "0:44:35", "remaining_time": "0:27:52"}
+{"current_steps": 193, "total_steps": 312, "loss": 0.5847, "lr": 3.832773180720475e-06, "epoch": 1.8557692307692308, "percentage": 61.86, "elapsed_time": "0:44:47", "remaining_time": "0:27:37"}
+{"current_steps": 194, "total_steps": 312, "loss": 0.5776, "lr": 3.778297969310529e-06, "epoch": 1.8653846153846154, "percentage": 62.18, "elapsed_time": "0:44:59", "remaining_time": "0:27:22"}
+{"current_steps": 195, "total_steps": 312, "loss": 0.5834, "lr": 3.723976553681787e-06, "epoch": 1.875, "percentage": 62.5, "elapsed_time": "0:45:12", "remaining_time": "0:27:07"}
+{"current_steps": 196, "total_steps": 312, "loss": 0.5761, "lr": 3.669815772166625e-06, "epoch": 1.8846153846153846, "percentage": 62.82, "elapsed_time": "0:45:26", "remaining_time": "0:26:53"}
+{"current_steps": 197, "total_steps": 312, "loss": 0.5648, "lr": 3.6158224428757538e-06, "epoch": 1.8942307692307692, "percentage": 63.14, "elapsed_time": "0:45:40", "remaining_time": "0:26:39"}
+{"current_steps": 198, "total_steps": 312, "loss": 0.5709, "lr": 3.562003362839914e-06, "epoch": 1.9038461538461537, "percentage": 63.46, "elapsed_time": "0:45:52", "remaining_time": "0:26:24"}
+{"current_steps": 199, "total_steps": 312, "loss": 0.5691, "lr": 3.50836530715422e-06, "epoch": 1.9134615384615383, "percentage": 63.78, "elapsed_time": "0:46:05", "remaining_time": "0:26:10"}
+{"current_steps": 200, "total_steps": 312, "loss": 0.5632, "lr": 3.4549150281252635e-06, "epoch": 1.9230769230769231, "percentage": 64.1, "elapsed_time": "0:46:18", "remaining_time": "0:25:55"}
+{"current_steps": 201, "total_steps": 312, "loss": 0.5666, "lr": 3.4016592544210937e-06, "epoch": 1.9326923076923077, "percentage": 64.42, "elapsed_time": "0:47:57", "remaining_time": "0:26:29"}
+{"current_steps": 202, "total_steps": 312, "loss": 0.588, "lr": 3.3486046902241663e-06, "epoch": 1.9423076923076923, "percentage": 64.74, "elapsed_time": "0:48:12", "remaining_time": "0:26:15"}
+{"current_steps": 203, "total_steps": 312, "loss": 0.5787, "lr": 3.295758014387375e-06, "epoch": 1.9519230769230769, "percentage": 65.06, "elapsed_time": "0:48:27", "remaining_time": "0:26:00"}
+{"current_steps": 204, "total_steps": 312, "loss": 0.5766, "lr": 3.2431258795932863e-06, "epoch": 1.9615384615384617, "percentage": 65.38, "elapsed_time": "0:48:40", "remaining_time": "0:25:46"}
+{"current_steps": 205, "total_steps": 312, "loss": 0.5714, "lr": 3.1907149115166403e-06, "epoch": 1.9711538461538463, "percentage": 65.71, "elapsed_time": "0:48:53", "remaining_time": "0:25:31"}
+{"current_steps": 206, "total_steps": 312, "loss": 0.5858, "lr": 3.1385317079902743e-06, "epoch": 1.9807692307692308, "percentage": 66.03, "elapsed_time": "0:49:06", "remaining_time": "0:25:15"}
+{"current_steps": 207, "total_steps": 312, "loss": 0.5735, "lr": 3.0865828381745515e-06, "epoch": 1.9903846153846154, "percentage": 66.35, "elapsed_time": "0:49:19", "remaining_time": "0:25:01"}
+{"current_steps": 208, "total_steps": 312, "loss": 0.5677, "lr": 3.0348748417303826e-06, "epoch": 2.0, "percentage": 66.67, "elapsed_time": "0:49:32", "remaining_time": "0:24:46"}
+{"current_steps": 209, "total_steps": 312, "loss": 0.5413, "lr": 2.9834142279959754e-06, "epoch": 2.0096153846153846, "percentage": 66.99, "elapsed_time": "0:49:47", "remaining_time": "0:24:32"}
+{"current_steps": 210, "total_steps": 312, "loss": 0.548, "lr": 2.932207475167398e-06, "epoch": 2.019230769230769, "percentage": 67.31, "elapsed_time": "0:50:02", "remaining_time": "0:24:18"}
+{"current_steps": 211, "total_steps": 312, "loss": 0.5441, "lr": 2.8812610294830568e-06, "epoch": 2.0288461538461537, "percentage": 67.63, "elapsed_time": "0:50:17", "remaining_time": "0:24:04"}
+{"current_steps": 212, "total_steps": 312, "loss": 0.5317, "lr": 2.83058130441221e-06, "epoch": 2.0384615384615383, "percentage": 67.95, "elapsed_time": "0:50:30", "remaining_time": "0:23:49"}
+{"current_steps": 213, "total_steps": 312, "loss": 0.5275, "lr": 2.7801746798475905e-06, "epoch": 2.048076923076923, "percentage": 68.27, "elapsed_time": "0:50:43", "remaining_time": "0:23:34"}
+{"current_steps": 214, "total_steps": 312, "loss": 0.5631, "lr": 2.7300475013022666e-06, "epoch": 2.0576923076923075, "percentage": 68.59, "elapsed_time": "0:50:56", "remaining_time": "0:23:19"}
+{"current_steps": 215, "total_steps": 312, "loss": 0.5447, "lr": 2.6802060791108304e-06, "epoch": 2.0673076923076925, "percentage": 68.91, "elapsed_time": "0:51:09", "remaining_time": "0:23:04"}
+{"current_steps": 216, "total_steps": 312, "loss": 0.535, "lr": 2.6306566876350072e-06, "epoch": 2.076923076923077, "percentage": 69.23, "elapsed_time": "0:51:21", "remaining_time": "0:22:49"}
+{"current_steps": 217, "total_steps": 312, "loss": 0.5534, "lr": 2.5814055644738013e-06, "epoch": 2.0865384615384617, "percentage": 69.55, "elapsed_time": "0:51:33", "remaining_time": "0:22:34"}
+{"current_steps": 218, "total_steps": 312, "loss": 0.5499, "lr": 2.532458909678266e-06, "epoch": 2.0961538461538463, "percentage": 69.87, "elapsed_time": "0:51:46", "remaining_time": "0:22:19"}
+{"current_steps": 219, "total_steps": 312, "loss": 0.5393, "lr": 2.483822884971e-06, "epoch": 2.105769230769231, "percentage": 70.19, "elapsed_time": "0:51:58", "remaining_time": "0:22:04"}
+{"current_steps": 220, "total_steps": 312, "loss": 0.5432, "lr": 2.43550361297047e-06, "epoch": 2.1153846153846154, "percentage": 70.51, "elapsed_time": "0:52:11", "remaining_time": "0:21:49"}
+{"current_steps": 221, "total_steps": 312, "loss": 0.5167, "lr": 2.387507176420256e-06, "epoch": 2.125, "percentage": 70.83, "elapsed_time": "0:52:24", "remaining_time": "0:21:34"}
+{"current_steps": 222, "total_steps": 312, "loss": 0.5517, "lr": 2.339839617423318e-06, "epoch": 2.1346153846153846, "percentage": 71.15, "elapsed_time": "0:52:39", "remaining_time": "0:21:20"}
+{"current_steps": 223, "total_steps": 312, "loss": 0.5388, "lr": 2.2925069366813718e-06, "epoch": 2.144230769230769, "percentage": 71.47, "elapsed_time": "0:52:52", "remaining_time": "0:21:06"}
+{"current_steps": 224, "total_steps": 312, "loss": 0.5351, "lr": 2.245515092739488e-06, "epoch": 2.1538461538461537, "percentage": 71.79, "elapsed_time": "0:53:04", "remaining_time": "0:20:51"}
+{"current_steps": 225, "total_steps": 312, "loss": 0.5398, "lr": 2.1988700012359865e-06, "epoch": 2.1634615384615383, "percentage": 72.12, "elapsed_time": "0:53:20", "remaining_time": "0:20:37"}
+{"current_steps": 226, "total_steps": 312, "loss": 0.5399, "lr": 2.1525775341577404e-06, "epoch": 2.173076923076923, "percentage": 72.44, "elapsed_time": "0:53:33", "remaining_time": "0:20:22"}
+{"current_steps": 227, "total_steps": 312, "loss": 0.535, "lr": 2.1066435191009717e-06, "epoch": 2.1826923076923075, "percentage": 72.76, "elapsed_time": "0:53:45", "remaining_time": "0:20:07"}
+{"current_steps": 228, "total_steps": 312, "loss": 0.5203, "lr": 2.061073738537635e-06, "epoch": 2.1923076923076925, "percentage": 73.08, "elapsed_time": "0:54:01", "remaining_time": "0:19:54"}
+{"current_steps": 229, "total_steps": 312, "loss": 0.5592, "lr": 2.0158739290874822e-06, "epoch": 2.201923076923077, "percentage": 73.4, "elapsed_time": "0:54:13", "remaining_time": "0:19:39"}
+{"current_steps": 230, "total_steps": 312, "loss": 0.5399, "lr": 1.971049780795901e-06, "epoch": 2.2115384615384617, "percentage": 73.72, "elapsed_time": "0:54:26", "remaining_time": "0:19:24"}
+{"current_steps": 231, "total_steps": 312, "loss": 0.522, "lr": 1.9266069364176144e-06, "epoch": 2.2211538461538463, "percentage": 74.04, "elapsed_time": "0:54:38", "remaining_time": "0:19:09"}
+{"current_steps": 232, "total_steps": 312, "loss": 0.5227, "lr": 1.8825509907063328e-06, "epoch": 2.230769230769231, "percentage": 74.36, "elapsed_time": "0:54:50", "remaining_time": "0:18:54"}
+{"current_steps": 233, "total_steps": 312, "loss": 0.537, "lr": 1.838887489710452e-06, "epoch": 2.2403846153846154, "percentage": 74.68, "elapsed_time": "0:55:03", "remaining_time": "0:18:40"}
+{"current_steps": 234, "total_steps": 312, "loss": 0.5505, "lr": 1.7956219300748796e-06, "epoch": 2.25, "percentage": 75.0, "elapsed_time": "0:55:16", "remaining_time": "0:18:25"}
+{"current_steps": 235, "total_steps": 312, "loss": 0.5678, "lr": 1.7527597583490825e-06, "epoch": 2.2596153846153846, "percentage": 75.32, "elapsed_time": "0:55:29", "remaining_time": "0:18:10"}
+{"current_steps": 236, "total_steps": 312, "loss": 0.5369, "lr": 1.7103063703014372e-06, "epoch": 2.269230769230769, "percentage": 75.64, "elapsed_time": "0:55:44", "remaining_time": "0:17:56"}
+{"current_steps": 237, "total_steps": 312, "loss": 0.5319, "lr": 1.6682671102399806e-06, "epoch": 2.2788461538461537, "percentage": 75.96, "elapsed_time": "0:55:57", "remaining_time": "0:17:42"}
+{"current_steps": 238, "total_steps": 312, "loss": 0.5492, "lr": 1.6266472703396286e-06, "epoch": 2.2884615384615383, "percentage": 76.28, "elapsed_time": "0:56:10", "remaining_time": "0:17:28"}
+{"current_steps": 239, "total_steps": 312, "loss": 0.5322, "lr": 1.5854520899759656e-06, "epoch": 2.298076923076923, "percentage": 76.6, "elapsed_time": "0:56:25", "remaining_time": "0:17:13"}
+{"current_steps": 240, "total_steps": 312, "loss": 0.553, "lr": 1.544686755065677e-06, "epoch": 2.3076923076923075, "percentage": 76.92, "elapsed_time": "0:56:37", "remaining_time": "0:16:59"}
+{"current_steps": 241, "total_steps": 312, "loss": 0.546, "lr": 1.5043563974137132e-06, "epoch": 2.3173076923076925, "percentage": 77.24, "elapsed_time": "0:56:50", "remaining_time": "0:16:44"}
+{"current_steps": 242, "total_steps": 312, "loss": 0.5357, "lr": 1.4644660940672628e-06, "epoch": 2.326923076923077, "percentage": 77.56, "elapsed_time": "0:57:03", "remaining_time": "0:16:30"}
+{"current_steps": 243, "total_steps": 312, "loss": 0.5192, "lr": 1.4250208666766235e-06, "epoch": 2.3365384615384617, "percentage": 77.88, "elapsed_time": "0:57:18", "remaining_time": "0:16:16"}
+{"current_steps": 244, "total_steps": 312, "loss": 0.5457, "lr": 1.3860256808630429e-06, "epoch": 2.3461538461538463, "percentage": 78.21, "elapsed_time": "0:57:32", "remaining_time": "0:16:02"}
+{"current_steps": 245, "total_steps": 312, "loss": 0.5543, "lr": 1.3474854455936126e-06, "epoch": 2.355769230769231, "percentage": 78.53, "elapsed_time": "0:57:45", "remaining_time": "0:15:47"}
+{"current_steps": 246, "total_steps": 312, "loss": 0.5661, "lr": 1.3094050125632973e-06, "epoch": 2.3653846153846154, "percentage": 78.85, "elapsed_time": "0:58:00", "remaining_time": "0:15:33"}
+{"current_steps": 247, "total_steps": 312, "loss": 0.5461, "lr": 1.2717891755841722e-06, "epoch": 2.375, "percentage": 79.17, "elapsed_time": "0:58:14", "remaining_time": "0:15:19"}
+{"current_steps": 248, "total_steps": 312, "loss": 0.5263, "lr": 1.234642669981946e-06, "epoch": 2.3846153846153846, "percentage": 79.49, "elapsed_time": "0:58:27", "remaining_time": "0:15:05"}
+{"current_steps": 249, "total_steps": 312, "loss": 0.5478, "lr": 1.1979701719998454e-06, "epoch": 2.394230769230769, "percentage": 79.81, "elapsed_time": "0:58:40", "remaining_time": "0:14:50"}
+{"current_steps": 250, "total_steps": 312, "loss": 0.5534, "lr": 1.1617762982099446e-06, "epoch": 2.4038461538461537, "percentage": 80.13, "elapsed_time": "0:58:54", "remaining_time": "0:14:36"}
+{"current_steps": 251, "total_steps": 312, "loss": 0.5366, "lr": 1.1260656049319957e-06, "epoch": 2.4134615384615383, "percentage": 80.45, "elapsed_time": "0:59:07", "remaining_time": "0:14:22"}
+{"current_steps": 252, "total_steps": 312, "loss": 0.5351, "lr": 1.0908425876598512e-06, "epoch": 2.423076923076923, "percentage": 80.77, "elapsed_time": "0:59:21", "remaining_time": "0:14:07"}
+{"current_steps": 253, "total_steps": 312, "loss": 0.5445, "lr": 1.0561116804955451e-06, "epoch": 2.4326923076923075, "percentage": 81.09, "elapsed_time": "0:59:35", "remaining_time": "0:13:53"}
+{"current_steps": 254, "total_steps": 312, "loss": 0.5545, "lr": 1.0218772555910955e-06, "epoch": 2.4423076923076925, "percentage": 81.41, "elapsed_time": "0:59:48", "remaining_time": "0:13:39"}
+{"current_steps": 255, "total_steps": 312, "loss": 0.5489, "lr": 9.881436225981107e-07, "epoch": 2.451923076923077, "percentage": 81.73, "elapsed_time": "1:00:01", "remaining_time": "0:13:25"}
+{"current_steps": 256, "total_steps": 312, "loss": 0.524, "lr": 9.549150281252633e-07, "epoch": 2.4615384615384617, "percentage": 82.05, "elapsed_time": "1:00:15", "remaining_time": "0:13:10"}
+{"current_steps": 257, "total_steps": 312, "loss": 0.5328, "lr": 9.221956552036992e-07, "epoch": 2.4711538461538463, "percentage": 82.37, "elapsed_time": "1:00:28", "remaining_time": "0:12:56"}
+{"current_steps": 258, "total_steps": 312, "loss": 0.561, "lr": 8.899896227604509e-07, "epoch": 2.480769230769231, "percentage": 82.69, "elapsed_time": "1:00:41", "remaining_time": "0:12:42"}
+{"current_steps": 259, "total_steps": 312, "loss": 0.5525, "lr": 8.58300985099918e-07, "epoch": 2.4903846153846154, "percentage": 83.01, "elapsed_time": "1:00:54", "remaining_time": "0:12:27"}
+{"current_steps": 260, "total_steps": 312, "loss": 0.5313, "lr": 8.271337313934869e-07, "epoch": 2.5, "percentage": 83.33, "elapsed_time": "1:01:06", "remaining_time": "0:12:13"}
+{"current_steps": 261, "total_steps": 312, "loss": 0.5327, "lr": 7.964917851773496e-07, "epoch": 2.5096153846153846, "percentage": 83.65, "elapsed_time": "1:01:21", "remaining_time": "0:11:59"}
+{"current_steps": 262, "total_steps": 312, "loss": 0.5375, "lr": 7.663790038585794e-07, "epoch": 2.519230769230769, "percentage": 83.97, "elapsed_time": "1:01:34", "remaining_time": "0:11:45"}
+{"current_steps": 263, "total_steps": 312, "loss": 0.5497, "lr": 7.367991782295392e-07, "epoch": 2.5288461538461537, "percentage": 84.29, "elapsed_time": "1:01:49", "remaining_time": "0:11:31"}
+{"current_steps": 264, "total_steps": 312, "loss": 0.5425, "lr": 7.077560319906696e-07, "epoch": 2.5384615384615383, "percentage": 84.62, "elapsed_time": "1:02:03", "remaining_time": "0:11:17"}
+{"current_steps": 265, "total_steps": 312, "loss": 0.548, "lr": 6.792532212817271e-07, "epoch": 2.5480769230769234, "percentage": 84.94, "elapsed_time": "1:02:16", "remaining_time": "0:11:02"}
+{"current_steps": 266, "total_steps": 312, "loss": 0.5527, "lr": 6.512943342215234e-07, "epoch": 2.5576923076923075, "percentage": 85.26, "elapsed_time": "1:02:29", "remaining_time": "0:10:48"}
+{"current_steps": 267, "total_steps": 312, "loss": 0.5562, "lr": 6.238828904562316e-07, "epoch": 2.5673076923076925, "percentage": 85.58, "elapsed_time": "1:02:41", "remaining_time": "0:10:33"}
+{"current_steps": 268, "total_steps": 312, "loss": 0.5426, "lr": 5.9702234071631e-07, "epoch": 2.5769230769230766, "percentage": 85.9, "elapsed_time": "1:02:55", "remaining_time": "0:10:19"}
+{"current_steps": 269, "total_steps": 312, "loss": 0.5482, "lr": 5.707160663821009e-07, "epoch": 2.5865384615384617, "percentage": 86.22, "elapsed_time": "1:03:09", "remaining_time": "0:10:05"}
+{"current_steps": 270, "total_steps": 312, "loss": 0.5282, "lr": 5.449673790581611e-07, "epoch": 2.5961538461538463, "percentage": 86.54, "elapsed_time": "1:03:24", "remaining_time": "0:09:51"}
+{"current_steps": 271, "total_steps": 312, "loss": 0.5496, "lr": 5.197795201563744e-07, "epoch": 2.605769230769231, "percentage": 86.86, "elapsed_time": "1:03:37", "remaining_time": "0:09:37"}
+{"current_steps": 272, "total_steps": 312, "loss": 0.5476, "lr": 4.951556604879049e-07, "epoch": 2.6153846153846154, "percentage": 87.18, "elapsed_time": "1:03:51", "remaining_time": "0:09:23"}
+{"current_steps": 273, "total_steps": 312, "loss": 0.5586, "lr": 4.710988998640298e-07, "epoch": 2.625, "percentage": 87.5, "elapsed_time": "1:04:04", "remaining_time": "0:09:09"}
+{"current_steps": 274, "total_steps": 312, "loss": 0.5531, "lr": 4.4761226670592074e-07, "epoch": 2.6346153846153846, "percentage": 87.82, "elapsed_time": "1:04:16", "remaining_time": "0:08:54"}
+{"current_steps": 275, "total_steps": 312, "loss": 0.5454, "lr": 4.2469871766340096e-07, "epoch": 2.644230769230769, "percentage": 88.14, "elapsed_time": "1:04:31", "remaining_time": "0:08:40"}
+{"current_steps": 276, "total_steps": 312, "loss": 0.5415, "lr": 4.0236113724274716e-07, "epoch": 2.6538461538461537, "percentage": 88.46, "elapsed_time": "1:04:44", "remaining_time": "0:08:26"}
+{"current_steps": 277, "total_steps": 312, "loss": 0.5286, "lr": 3.8060233744356634e-07, "epoch": 2.6634615384615383, "percentage": 88.78, "elapsed_time": "1:04:57", "remaining_time": "0:08:12"}
+{"current_steps": 278, "total_steps": 312, "loss": 0.5483, "lr": 3.5942505740480583e-07, "epoch": 2.6730769230769234, "percentage": 89.1, "elapsed_time": "1:05:11", "remaining_time": "0:07:58"}
+{"current_steps": 279, "total_steps": 312, "loss": 0.5397, "lr": 3.3883196305992906e-07, "epoch": 2.6826923076923075, "percentage": 89.42, "elapsed_time": "1:05:24", "remaining_time": "0:07:44"}
+{"current_steps": 280, "total_steps": 312, "loss": 0.5813, "lr": 3.18825646801314e-07, "epoch": 2.6923076923076925, "percentage": 89.74, "elapsed_time": "1:05:38", "remaining_time": "0:07:30"}
+{"current_steps": 281, "total_steps": 312, "loss": 0.5477, "lr": 2.9940862715390483e-07, "epoch": 2.7019230769230766, "percentage": 90.06, "elapsed_time": "1:05:51", "remaining_time": "0:07:15"}
+{"current_steps": 282, "total_steps": 312, "loss": 0.5699, "lr": 2.8058334845816214e-07, "epoch": 2.7115384615384617, "percentage": 90.38, "elapsed_time": "1:06:06", "remaining_time": "0:07:01"}
+{"current_steps": 283, "total_steps": 312, "loss": 0.5141, "lr": 2.6235218056235633e-07, "epoch": 2.7211538461538463, "percentage": 90.71, "elapsed_time": "1:06:19", "remaining_time": "0:06:47"}
+{"current_steps": 284, "total_steps": 312, "loss": 0.5593, "lr": 2.447174185242324e-07, "epoch": 2.730769230769231, "percentage": 91.03, "elapsed_time": "1:06:34", "remaining_time": "0:06:33"}
+{"current_steps": 285, "total_steps": 312, "loss": 0.5575, "lr": 2.276812823220964e-07, "epoch": 2.7403846153846154, "percentage": 91.35, "elapsed_time": "1:06:48", "remaining_time": "0:06:19"}
+{"current_steps": 286, "total_steps": 312, "loss": 0.5249, "lr": 2.1124591657534776e-07, "epoch": 2.75, "percentage": 91.67, "elapsed_time": "1:07:02", "remaining_time": "0:06:05"}
+{"current_steps": 287, "total_steps": 312, "loss": 0.5234, "lr": 1.9541339027450256e-07, "epoch": 2.7596153846153846, "percentage": 91.99, "elapsed_time": "1:07:17", "remaining_time": "0:05:51"}
+{"current_steps": 288, "total_steps": 312, "loss": 0.5215, "lr": 1.801856965207338e-07, "epoch": 2.769230769230769, "percentage": 92.31, "elapsed_time": "1:07:31", "remaining_time": "0:05:37"}
+{"current_steps": 289, "total_steps": 312, "loss": 0.5574, "lr": 1.6556475227496816e-07, "epoch": 2.7788461538461537, "percentage": 92.63, "elapsed_time": "1:07:44", "remaining_time": "0:05:23"}
+{"current_steps": 290, "total_steps": 312, "loss": 0.5465, "lr": 1.5155239811656562e-07, "epoch": 2.7884615384615383, "percentage": 92.95, "elapsed_time": "1:07:57", "remaining_time": "0:05:09"}
+{"current_steps": 291, "total_steps": 312, "loss": 0.5469, "lr": 1.3815039801161723e-07, "epoch": 2.7980769230769234, "percentage": 93.27, "elapsed_time": "1:08:11", "remaining_time": "0:04:55"}
+{"current_steps": 292, "total_steps": 312, "loss": 0.5399, "lr": 1.253604390908819e-07, "epoch": 2.8076923076923075, "percentage": 93.59, "elapsed_time": "1:08:25", "remaining_time": "0:04:41"}
+{"current_steps": 293, "total_steps": 312, "loss": 0.5451, "lr": 1.1318413143740436e-07, "epoch": 2.8173076923076925, "percentage": 93.91, "elapsed_time": "1:08:38", "remaining_time": "0:04:27"}
+{"current_steps": 294, "total_steps": 312, "loss": 0.5584, "lr": 1.0162300788382263e-07, "epoch": 2.8269230769230766, "percentage": 94.23, "elapsed_time": "1:08:54", "remaining_time": "0:04:13"}
+{"current_steps": 295, "total_steps": 312, "loss": 0.5332, "lr": 9.0678523819408e-08, "epoch": 2.8365384615384617, "percentage": 94.55, "elapsed_time": "1:09:06", "remaining_time": "0:03:58"}
+{"current_steps": 296, "total_steps": 312, "loss": 0.5622, "lr": 8.035205700685167e-08, "epoch": 2.8461538461538463, "percentage": 94.87, "elapsed_time": "1:09:21", "remaining_time": "0:03:44"}
+{"current_steps": 297, "total_steps": 312, "loss": 0.5351, "lr": 7.064490740882057e-08, "epoch": 2.855769230769231, "percentage": 95.19, "elapsed_time": "1:09:36", "remaining_time": "0:03:30"}
+{"current_steps": 298, "total_steps": 312, "loss": 0.5532, "lr": 6.15582970243117e-08, "epoch": 2.8653846153846154, "percentage": 95.51, "elapsed_time": "1:09:50", "remaining_time": "0:03:16"}
+{"current_steps": 299, "total_steps": 312, "loss": 0.5441, "lr": 5.3093369734816824e-08, "epoch": 2.875, "percentage": 95.83, "elapsed_time": "1:10:05", "remaining_time": "0:03:02"}
+{"current_steps": 300, "total_steps": 312, "loss": 0.5482, "lr": 4.52511911603265e-08, "epoch": 2.8846153846153846, "percentage": 96.15, "elapsed_time": "1:10:19", "remaining_time": "0:02:48"}
+{"current_steps": 301, "total_steps": 312, "loss": 0.5473, "lr": 3.8032748525179684e-08, "epoch": 2.894230769230769, "percentage": 96.47, "elapsed_time": "1:10:57", "remaining_time": "0:02:35"}
+{"current_steps": 302, "total_steps": 312, "loss": 0.5371, "lr": 3.143895053378698e-08, "epoch": 2.9038461538461537, "percentage": 96.79, "elapsed_time": "1:11:10", "remaining_time": "0:02:21"}
+{"current_steps": 303, "total_steps": 312, "loss": 0.5424, "lr": 2.547062725623828e-08, "epoch": 2.9134615384615383, "percentage": 97.12, "elapsed_time": "1:11:25", "remaining_time": "0:02:07"}
+{"current_steps": 304, "total_steps": 312, "loss": 0.5553, "lr": 2.012853002380466e-08, "epoch": 2.9230769230769234, "percentage": 97.44, "elapsed_time": "1:11:38", "remaining_time": "0:01:53"}
+{"current_steps": 305, "total_steps": 312, "loss": 0.5281, "lr": 1.541333133436018e-08, "epoch": 2.9326923076923075, "percentage": 97.76, "elapsed_time": "1:11:51", "remaining_time": "0:01:38"}
+{"current_steps": 306, "total_steps": 312, "loss": 0.5465, "lr": 1.132562476771959e-08, "epoch": 2.9423076923076925, "percentage": 98.08, "elapsed_time": "1:12:05", "remaining_time": "0:01:24"}
+{"current_steps": 307, "total_steps": 312, "loss": 0.5219, "lr": 7.865924910916977e-09, "epoch": 2.9519230769230766, "percentage": 98.4, "elapsed_time": "1:12:19", "remaining_time": "0:01:10"}
+{"current_steps": 308, "total_steps": 312, "loss": 0.5476, "lr": 5.034667293427053e-09, "epoch": 2.9615384615384617, "percentage": 98.72, "elapsed_time": "1:12:32", "remaining_time": "0:00:56"}
+{"current_steps": 309, "total_steps": 312, "loss": 0.5605, "lr": 2.8322083323334417e-09, "epoch": 2.9711538461538463, "percentage": 99.04, "elapsed_time": "1:12:45", "remaining_time": "0:00:42"}
+{"current_steps": 310, "total_steps": 312, "loss": 0.5469, "lr": 1.2588252874673469e-09, "epoch": 2.980769230769231, "percentage": 99.36, "elapsed_time": "1:12:59", "remaining_time": "0:00:28"}
+{"current_steps": 311, "total_steps": 312, "loss": 0.548, "lr": 3.147162264971471e-10, "epoch": 2.9903846153846154, "percentage": 99.68, "elapsed_time": "1:13:12", "remaining_time": "0:00:14"}
+{"current_steps": 312, "total_steps": 312, "loss": 0.5313, "lr": 0.0, "epoch": 3.0, "percentage": 100.0, "elapsed_time": "1:13:26", "remaining_time": "0:00:00"}
+{"current_steps": 312, "total_steps": 312, "epoch": 3.0, "percentage": 100.0, "elapsed_time": "1:13:54", "remaining_time": "0:00:00"}

trainer_state.json

@@ -0,0 +1,2226 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 312,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.009615384615384616,
+      "grad_norm": 6.949608325958252,
+      "learning_rate": 3.125e-07,
+      "loss": 1.0512,
+      "step": 1
+    },
+    {
+      "epoch": 0.019230769230769232,
+      "grad_norm": 7.026703357696533,
+      "learning_rate": 6.25e-07,
+      "loss": 1.0592,
+      "step": 2
+    },
+    {
+      "epoch": 0.028846153846153848,
+      "grad_norm": 6.95483922958374,
+      "learning_rate": 9.375000000000001e-07,
+      "loss": 1.0391,
+      "step": 3
+    },
+    {
+      "epoch": 0.038461538461538464,
+      "grad_norm": 7.0581231117248535,
+      "learning_rate": 1.25e-06,
+      "loss": 1.0671,
+      "step": 4
+    },
+    {
+      "epoch": 0.04807692307692308,
+      "grad_norm": 6.7878737449646,
+      "learning_rate": 1.5625e-06,
+      "loss": 1.0551,
+      "step": 5
+    },
+    {
+      "epoch": 0.057692307692307696,
+      "grad_norm": 6.292112350463867,
+      "learning_rate": 1.8750000000000003e-06,
+      "loss": 1.0392,
+      "step": 6
+    },
+    {
+      "epoch": 0.0673076923076923,
+      "grad_norm": 4.9926958084106445,
+      "learning_rate": 2.1875000000000002e-06,
+      "loss": 0.9965,
+      "step": 7
+    },
+    {
+      "epoch": 0.07692307692307693,
+      "grad_norm": 4.684077739715576,
+      "learning_rate": 2.5e-06,
+      "loss": 0.9904,
+      "step": 8
+    },
+    {
+      "epoch": 0.08653846153846154,
+      "grad_norm": 2.9312803745269775,
+      "learning_rate": 2.8125e-06,
+      "loss": 0.9546,
+      "step": 9
+    },
+    {
+      "epoch": 0.09615384615384616,
+      "grad_norm": 2.813443183898926,
+      "learning_rate": 3.125e-06,
+      "loss": 0.9474,
+      "step": 10
+    },
+    {
+      "epoch": 0.10576923076923077,
+      "grad_norm": 2.5470032691955566,
+      "learning_rate": 3.4375e-06,
+      "loss": 0.9387,
+      "step": 11
+    },
+    {
+      "epoch": 0.11538461538461539,
+      "grad_norm": 3.7931878566741943,
+      "learning_rate": 3.7500000000000005e-06,
+      "loss": 0.9115,
+      "step": 12
+    },
+    {
+      "epoch": 0.125,
+      "grad_norm": 4.256772994995117,
+      "learning_rate": 4.0625000000000005e-06,
+      "loss": 0.9166,
+      "step": 13
+    },
+    {
+      "epoch": 0.1346153846153846,
+      "grad_norm": 4.049694538116455,
+      "learning_rate": 4.3750000000000005e-06,
+      "loss": 0.8974,
+      "step": 14
+    },
+    {
+      "epoch": 0.14423076923076922,
+      "grad_norm": 3.689324378967285,
+      "learning_rate": 4.6875000000000004e-06,
+      "loss": 0.8862,
+      "step": 15
+    },
+    {
+      "epoch": 0.15384615384615385,
+      "grad_norm": 2.5206212997436523,
+      "learning_rate": 5e-06,
+      "loss": 0.8618,
+      "step": 16
+    },
+    {
+      "epoch": 0.16346153846153846,
+      "grad_norm": 2.2595858573913574,
+      "learning_rate": 5.3125e-06,
+      "loss": 0.8466,
+      "step": 17
+    },
+    {
+      "epoch": 0.17307692307692307,
+      "grad_norm": 1.8435250520706177,
+      "learning_rate": 5.625e-06,
+      "loss": 0.8225,
+      "step": 18
+    },
+    {
+      "epoch": 0.18269230769230768,
+      "grad_norm": 1.453895092010498,
+      "learning_rate": 5.9375e-06,
+      "loss": 0.7979,
+      "step": 19
+    },
+    {
+      "epoch": 0.19230769230769232,
+      "grad_norm": 1.2612786293029785,
+      "learning_rate": 6.25e-06,
+      "loss": 0.8085,
+      "step": 20
+    },
+    {
+      "epoch": 0.20192307692307693,
+      "grad_norm": 1.2307376861572266,
+      "learning_rate": 6.5625e-06,
+      "loss": 0.7795,
+      "step": 21
+    },
+    {
+      "epoch": 0.21153846153846154,
+      "grad_norm": 1.2606533765792847,
+      "learning_rate": 6.875e-06,
+      "loss": 0.7871,
+      "step": 22
+    },
+    {
+      "epoch": 0.22115384615384615,
+      "grad_norm": 1.2339670658111572,
+      "learning_rate": 7.1875e-06,
+      "loss": 0.7624,
+      "step": 23
+    },
+    {
+      "epoch": 0.23076923076923078,
+      "grad_norm": 1.0207619667053223,
+      "learning_rate": 7.500000000000001e-06,
+      "loss": 0.7557,
+      "step": 24
+    },
+    {
+      "epoch": 0.2403846153846154,
+      "grad_norm": 0.8926437497138977,
+      "learning_rate": 7.8125e-06,
+      "loss": 0.7342,
+      "step": 25
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 0.9713419079780579,
+      "learning_rate": 8.125000000000001e-06,
+      "loss": 0.7372,
+      "step": 26
+    },
+    {
+      "epoch": 0.25961538461538464,
+      "grad_norm": 1.042178988456726,
+      "learning_rate": 8.4375e-06,
+      "loss": 0.7375,
+      "step": 27
+    },
+    {
+      "epoch": 0.2692307692307692,
+      "grad_norm": 0.8239858150482178,
+      "learning_rate": 8.750000000000001e-06,
+      "loss": 0.7147,
+      "step": 28
+    },
+    {
+      "epoch": 0.27884615384615385,
+      "grad_norm": 0.8827706575393677,
+      "learning_rate": 9.0625e-06,
+      "loss": 0.7287,
+      "step": 29
+    },
+    {
+      "epoch": 0.28846153846153844,
+      "grad_norm": 0.9991267323493958,
+      "learning_rate": 9.375000000000001e-06,
+      "loss": 0.7046,
+      "step": 30
+    },
+    {
+      "epoch": 0.2980769230769231,
+      "grad_norm": 0.8923914432525635,
+      "learning_rate": 9.6875e-06,
+      "loss": 0.701,
+      "step": 31
+    },
+    {
+      "epoch": 0.3076923076923077,
+      "grad_norm": 0.7739760875701904,
+      "learning_rate": 1e-05,
+      "loss": 0.7091,
+      "step": 32
+    },
+    {
+      "epoch": 0.3173076923076923,
+      "grad_norm": 0.9522834420204163,
+      "learning_rate": 9.999685283773504e-06,
+      "loss": 0.71,
+      "step": 33
+    },
+    {
+      "epoch": 0.3269230769230769,
+      "grad_norm": 0.7752206921577454,
+      "learning_rate": 9.998741174712534e-06,
+      "loss": 0.6876,
+      "step": 34
+    },
+    {
+      "epoch": 0.33653846153846156,
+      "grad_norm": 1.002880573272705,
+      "learning_rate": 9.997167791667668e-06,
+      "loss": 0.7019,
+      "step": 35
+    },
+    {
+      "epoch": 0.34615384615384615,
+      "grad_norm": 0.7485476136207581,
+      "learning_rate": 9.994965332706574e-06,
+      "loss": 0.6779,
+      "step": 36
+    },
+    {
+      "epoch": 0.3557692307692308,
+      "grad_norm": 0.8847535252571106,
+      "learning_rate": 9.992134075089085e-06,
+      "loss": 0.6753,
+      "step": 37
+    },
+    {
+      "epoch": 0.36538461538461536,
+      "grad_norm": 0.6897234320640564,
+      "learning_rate": 9.98867437523228e-06,
+      "loss": 0.7009,
+      "step": 38
+    },
+    {
+      "epoch": 0.375,
+      "grad_norm": 0.9160583019256592,
+      "learning_rate": 9.984586668665641e-06,
+      "loss": 0.6828,
+      "step": 39
+    },
+    {
+      "epoch": 0.38461538461538464,
+      "grad_norm": 0.7275518178939819,
+      "learning_rate": 9.979871469976197e-06,
+      "loss": 0.6833,
+      "step": 40
+    },
+    {
+      "epoch": 0.3942307692307692,
+      "grad_norm": 0.7628114223480225,
+      "learning_rate": 9.974529372743762e-06,
+      "loss": 0.6711,
+      "step": 41
+    },
+    {
+      "epoch": 0.40384615384615385,
+      "grad_norm": 0.7545289993286133,
+      "learning_rate": 9.968561049466214e-06,
+      "loss": 0.6767,
+      "step": 42
+    },
+    {
+      "epoch": 0.41346153846153844,
+      "grad_norm": 0.6503568291664124,
+      "learning_rate": 9.961967251474823e-06,
+      "loss": 0.6519,
+      "step": 43
+    },
+    {
+      "epoch": 0.4230769230769231,
+      "grad_norm": 0.7073460817337036,
+      "learning_rate": 9.954748808839675e-06,
+      "loss": 0.67,
+      "step": 44
+    },
+    {
+      "epoch": 0.4326923076923077,
+      "grad_norm": 0.6822070479393005,
+      "learning_rate": 9.946906630265184e-06,
+      "loss": 0.6659,
+      "step": 45
+    },
+    {
+      "epoch": 0.4423076923076923,
+      "grad_norm": 0.6585090160369873,
+      "learning_rate": 9.938441702975689e-06,
+      "loss": 0.6758,
+      "step": 46
+    },
+    {
+      "epoch": 0.4519230769230769,
+      "grad_norm": 0.7007927894592285,
+      "learning_rate": 9.92935509259118e-06,
+      "loss": 0.6537,
+      "step": 47
+    },
+    {
+      "epoch": 0.46153846153846156,
+      "grad_norm": 0.5756403207778931,
+      "learning_rate": 9.91964794299315e-06,
+      "loss": 0.6674,
+      "step": 48
+    },
+    {
+      "epoch": 0.47115384615384615,
+      "grad_norm": 0.673179566860199,
+      "learning_rate": 9.909321476180594e-06,
+      "loss": 0.6574,
+      "step": 49
+    },
+    {
+      "epoch": 0.4807692307692308,
+      "grad_norm": 0.5679298043251038,
+      "learning_rate": 9.898376992116179e-06,
+      "loss": 0.6641,
+      "step": 50
+    },
+    {
+      "epoch": 0.49038461538461536,
+      "grad_norm": 0.6427024602890015,
+      "learning_rate": 9.886815868562596e-06,
+      "loss": 0.6628,
+      "step": 51
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.6467263102531433,
+      "learning_rate": 9.874639560909118e-06,
+      "loss": 0.6669,
+      "step": 52
+    },
+    {
+      "epoch": 0.5096153846153846,
+      "grad_norm": 0.5870878100395203,
+      "learning_rate": 9.861849601988384e-06,
+      "loss": 0.6642,
+      "step": 53
+    },
+    {
+      "epoch": 0.5192307692307693,
+      "grad_norm": 0.5774953961372375,
+      "learning_rate": 9.848447601883436e-06,
+      "loss": 0.63,
+      "step": 54
+    },
+    {
+      "epoch": 0.5288461538461539,
+      "grad_norm": 0.6247491240501404,
+      "learning_rate": 9.834435247725032e-06,
+      "loss": 0.6441,
+      "step": 55
+    },
+    {
+      "epoch": 0.5384615384615384,
+      "grad_norm": 0.5765902996063232,
+      "learning_rate": 9.819814303479268e-06,
+      "loss": 0.6387,
+      "step": 56
+    },
+    {
+      "epoch": 0.5480769230769231,
+      "grad_norm": 0.6525144577026367,
+      "learning_rate": 9.804586609725499e-06,
+      "loss": 0.653,
+      "step": 57
+    },
+    {
+      "epoch": 0.5576923076923077,
+      "grad_norm": 0.6681458950042725,
+      "learning_rate": 9.788754083424654e-06,
+      "loss": 0.6394,
+      "step": 58
+    },
+    {
+      "epoch": 0.5673076923076923,
+      "grad_norm": 0.5383596420288086,
+      "learning_rate": 9.772318717677905e-06,
+      "loss": 0.6415,
+      "step": 59
+    },
+    {
+      "epoch": 0.5769230769230769,
+      "grad_norm": 0.6443718671798706,
+      "learning_rate": 9.755282581475769e-06,
+      "loss": 0.6371,
+      "step": 60
+    },
+    {
+      "epoch": 0.5865384615384616,
+      "grad_norm": 0.6086981296539307,
+      "learning_rate": 9.737647819437645e-06,
+      "loss": 0.6474,
+      "step": 61
+    },
+    {
+      "epoch": 0.5961538461538461,
+      "grad_norm": 0.557014524936676,
+      "learning_rate": 9.719416651541839e-06,
+      "loss": 0.6524,
+      "step": 62
+    },
+    {
+      "epoch": 0.6057692307692307,
+      "grad_norm": 0.6446460485458374,
+      "learning_rate": 9.700591372846096e-06,
+      "loss": 0.6444,
+      "step": 63
+    },
+    {
+      "epoch": 0.6153846153846154,
+      "grad_norm": 0.6329739689826965,
+      "learning_rate": 9.681174353198687e-06,
+      "loss": 0.6601,
+      "step": 64
+    },
+    {
+      "epoch": 0.625,
+      "grad_norm": 0.6192615628242493,
+      "learning_rate": 9.661168036940071e-06,
+      "loss": 0.6596,
+      "step": 65
+    },
+    {
+      "epoch": 0.6346153846153846,
+      "grad_norm": 0.5624246001243591,
+      "learning_rate": 9.640574942595195e-06,
+      "loss": 0.6608,
+      "step": 66
+    },
+    {
+      "epoch": 0.6442307692307693,
+      "grad_norm": 0.6665242910385132,
+      "learning_rate": 9.619397662556434e-06,
+      "loss": 0.6434,
+      "step": 67
+    },
+    {
+      "epoch": 0.6538461538461539,
+      "grad_norm": 0.5866331458091736,
+      "learning_rate": 9.597638862757255e-06,
+      "loss": 0.6369,
+      "step": 68
+    },
+    {
+      "epoch": 0.6634615384615384,
+      "grad_norm": 0.5087175369262695,
+      "learning_rate": 9.5753012823366e-06,
+      "loss": 0.6333,
+      "step": 69
+    },
+    {
+      "epoch": 0.6730769230769231,
+      "grad_norm": 0.6268306970596313,
+      "learning_rate": 9.552387733294081e-06,
+      "loss": 0.647,
+      "step": 70
+    },
+    {
+      "epoch": 0.6826923076923077,
+      "grad_norm": 0.5971682667732239,
+      "learning_rate": 9.528901100135971e-06,
+      "loss": 0.6293,
+      "step": 71
+    },
+    {
+      "epoch": 0.6923076923076923,
+      "grad_norm": 0.6260504722595215,
+      "learning_rate": 9.504844339512096e-06,
+      "loss": 0.6329,
+      "step": 72
+    },
+    {
+      "epoch": 0.7019230769230769,
+      "grad_norm": 0.5957349538803101,
+      "learning_rate": 9.480220479843627e-06,
+      "loss": 0.6373,
+      "step": 73
+    },
+    {
+      "epoch": 0.7115384615384616,
+      "grad_norm": 0.5238078236579895,
+      "learning_rate": 9.45503262094184e-06,
+      "loss": 0.647,
+      "step": 74
+    },
+    {
+      "epoch": 0.7211538461538461,
+      "grad_norm": 0.6080106496810913,
+      "learning_rate": 9.4292839336179e-06,
+      "loss": 0.6469,
+      "step": 75
+    },
+    {
+      "epoch": 0.7307692307692307,
+      "grad_norm": 0.6035282015800476,
+      "learning_rate": 9.40297765928369e-06,
+      "loss": 0.6316,
+      "step": 76
+    },
+    {
+      "epoch": 0.7403846153846154,
+      "grad_norm": 0.5384305715560913,
+      "learning_rate": 9.376117109543769e-06,
+      "loss": 0.6369,
+      "step": 77
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 0.6150159239768982,
+      "learning_rate": 9.348705665778479e-06,
+      "loss": 0.6435,
+      "step": 78
+    },
+    {
+      "epoch": 0.7596153846153846,
+      "grad_norm": 0.5395728945732117,
+      "learning_rate": 9.320746778718274e-06,
+      "loss": 0.6237,
+      "step": 79
+    },
+    {
+      "epoch": 0.7692307692307693,
+      "grad_norm": 0.5888304710388184,
+      "learning_rate": 9.292243968009332e-06,
+      "loss": 0.6413,
+      "step": 80
+    },
+    {
+      "epoch": 0.7788461538461539,
+      "grad_norm": 0.5334227681159973,
+      "learning_rate": 9.263200821770462e-06,
+      "loss": 0.6491,
+      "step": 81
+    },
+    {
+      "epoch": 0.7884615384615384,
+      "grad_norm": 0.5316609144210815,
+      "learning_rate": 9.233620996141421e-06,
+      "loss": 0.6537,
+      "step": 82
+    },
+    {
+      "epoch": 0.7980769230769231,
+      "grad_norm": 0.495564341545105,
+      "learning_rate": 9.203508214822652e-06,
+      "loss": 0.6395,
+      "step": 83
+    },
+    {
+      "epoch": 0.8076923076923077,
+      "grad_norm": 0.5694448351860046,
+      "learning_rate": 9.172866268606514e-06,
+      "loss": 0.6428,
+      "step": 84
+    },
+    {
+      "epoch": 0.8173076923076923,
+      "grad_norm": 0.49600711464881897,
+      "learning_rate": 9.141699014900084e-06,
+      "loss": 0.6321,
+      "step": 85
+    },
+    {
+      "epoch": 0.8269230769230769,
+      "grad_norm": 0.63581782579422,
+      "learning_rate": 9.110010377239552e-06,
+      "loss": 0.6139,
+      "step": 86
+    },
+    {
+      "epoch": 0.8365384615384616,
+      "grad_norm": 0.5633158683776855,
+      "learning_rate": 9.077804344796302e-06,
+      "loss": 0.6539,
+      "step": 87
+    },
+    {
+      "epoch": 0.8461538461538461,
+      "grad_norm": 0.6227946281433105,
+      "learning_rate": 9.045084971874738e-06,
+      "loss": 0.634,
+      "step": 88
+    },
+    {
+      "epoch": 0.8557692307692307,
+      "grad_norm": 0.5541588068008423,
+      "learning_rate": 9.011856377401891e-06,
+      "loss": 0.6336,
+      "step": 89
+    },
+    {
+      "epoch": 0.8653846153846154,
+      "grad_norm": 0.5803648829460144,
+      "learning_rate": 8.978122744408905e-06,
+      "loss": 0.6327,
+      "step": 90
+    },
+    {
+      "epoch": 0.875,
+      "grad_norm": 0.5258088111877441,
+      "learning_rate": 8.943888319504456e-06,
+      "loss": 0.6248,
+      "step": 91
+    },
+    {
+      "epoch": 0.8846153846153846,
+      "grad_norm": 0.6071212887763977,
+      "learning_rate": 8.90915741234015e-06,
+      "loss": 0.6259,
+      "step": 92
+    },
+    {
+      "epoch": 0.8942307692307693,
+      "grad_norm": 0.571426510810852,
+      "learning_rate": 8.873934395068006e-06,
+      "loss": 0.6248,
+      "step": 93
+    },
+    {
+      "epoch": 0.9038461538461539,
+      "grad_norm": 0.5370094776153564,
+      "learning_rate": 8.838223701790057e-06,
+      "loss": 0.6125,
+      "step": 94
+    },
+    {
+      "epoch": 0.9134615384615384,
+      "grad_norm": 0.5595990419387817,
+      "learning_rate": 8.802029828000157e-06,
+      "loss": 0.616,
+      "step": 95
+    },
+    {
+      "epoch": 0.9230769230769231,
+      "grad_norm": 0.5226612687110901,
+      "learning_rate": 8.765357330018056e-06,
+      "loss": 0.6369,
+      "step": 96
+    },
+    {
+      "epoch": 0.9326923076923077,
+      "grad_norm": 0.541027307510376,
+      "learning_rate": 8.728210824415829e-06,
+      "loss": 0.6244,
+      "step": 97
+    },
+    {
+      "epoch": 0.9423076923076923,
+      "grad_norm": 0.5068326592445374,
+      "learning_rate": 8.690594987436705e-06,
+      "loss": 0.6323,
+      "step": 98
+    },
+    {
+      "epoch": 0.9519230769230769,
+      "grad_norm": 0.5558856725692749,
+      "learning_rate": 8.652514554406388e-06,
+      "loss": 0.6283,
+      "step": 99
+    },
+    {
+      "epoch": 0.9615384615384616,
+      "grad_norm": 0.5843573212623596,
+      "learning_rate": 8.613974319136959e-06,
+      "loss": 0.6339,
+      "step": 100
+    },
+    {
+      "epoch": 0.9711538461538461,
+      "grad_norm": 0.4959486126899719,
+      "learning_rate": 8.574979133323378e-06,
+      "loss": 0.6298,
+      "step": 101
+    },
+    {
+      "epoch": 0.9807692307692307,
+      "grad_norm": 0.5050930380821228,
+      "learning_rate": 8.535533905932739e-06,
+      "loss": 0.6211,
+      "step": 102
+    },
+    {
+      "epoch": 0.9903846153846154,
+      "grad_norm": 0.5262631773948669,
+      "learning_rate": 8.495643602586287e-06,
+      "loss": 0.6135,
+      "step": 103
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.4946816563606262,
+      "learning_rate": 8.455313244934324e-06,
+      "loss": 0.6295,
+      "step": 104
+    },
+    {
+      "epoch": 1.0096153846153846,
+      "grad_norm": 0.5233300924301147,
+      "learning_rate": 8.414547910024035e-06,
+      "loss": 0.5924,
+      "step": 105
+    },
+    {
+      "epoch": 1.0192307692307692,
+      "grad_norm": 0.52397620677948,
+      "learning_rate": 8.373352729660373e-06,
+      "loss": 0.6012,
+      "step": 106
+    },
+    {
+      "epoch": 1.0288461538461537,
+      "grad_norm": 0.612706184387207,
+      "learning_rate": 8.331732889760021e-06,
+      "loss": 0.6092,
+      "step": 107
+    },
+    {
+      "epoch": 1.0384615384615385,
+      "grad_norm": 0.5767417550086975,
+      "learning_rate": 8.289693629698564e-06,
+      "loss": 0.5799,
+      "step": 108
+    },
+    {
+      "epoch": 1.0480769230769231,
+      "grad_norm": 0.6034297943115234,
+      "learning_rate": 8.247240241650918e-06,
+      "loss": 0.5634,
+      "step": 109
+    },
+    {
+      "epoch": 1.0576923076923077,
+      "grad_norm": 0.5869642496109009,
+      "learning_rate": 8.204378069925121e-06,
+      "loss": 0.5949,
+      "step": 110
+    },
+    {
+      "epoch": 1.0673076923076923,
+      "grad_norm": 0.5452908873558044,
+      "learning_rate": 8.16111251028955e-06,
+      "loss": 0.5997,
+      "step": 111
+    },
+    {
+      "epoch": 1.0769230769230769,
+      "grad_norm": 0.6273619532585144,
+      "learning_rate": 8.117449009293668e-06,
+      "loss": 0.5972,
+      "step": 112
+    },
+    {
+      "epoch": 1.0865384615384615,
+      "grad_norm": 0.5914397239685059,
+      "learning_rate": 8.073393063582386e-06,
+      "loss": 0.5684,
+      "step": 113
+    },
+    {
+      "epoch": 1.0961538461538463,
+      "grad_norm": 0.5578101277351379,
+      "learning_rate": 8.0289502192041e-06,
+      "loss": 0.5967,
+      "step": 114
+    },
+    {
+      "epoch": 1.1057692307692308,
+      "grad_norm": 0.5323030352592468,
+      "learning_rate": 7.984126070912519e-06,
+      "loss": 0.5956,
+      "step": 115
+    },
+    {
+      "epoch": 1.1153846153846154,
+      "grad_norm": 0.5913547873497009,
+      "learning_rate": 7.938926261462366e-06,
+      "loss": 0.5787,
+      "step": 116
+    },
+    {
+      "epoch": 1.125,
+      "grad_norm": 0.5356422662734985,
+      "learning_rate": 7.89335648089903e-06,
+      "loss": 0.5913,
+      "step": 117
+    },
+    {
+      "epoch": 1.1346153846153846,
+      "grad_norm": 0.5649756193161011,
+      "learning_rate": 7.84742246584226e-06,
+      "loss": 0.5776,
+      "step": 118
+    },
+    {
+      "epoch": 1.1442307692307692,
+      "grad_norm": 0.5341564416885376,
+      "learning_rate": 7.801129998764014e-06,
+      "loss": 0.6113,
+      "step": 119
+    },
+    {
+      "epoch": 1.1538461538461537,
+      "grad_norm": 0.6811894774436951,
+      "learning_rate": 7.754484907260513e-06,
+      "loss": 0.58,
+      "step": 120
+    },
+    {
+      "epoch": 1.1634615384615385,
+      "grad_norm": 0.6039808392524719,
+      "learning_rate": 7.70749306331863e-06,
+      "loss": 0.5996,
+      "step": 121
+    },
+    {
+      "epoch": 1.1730769230769231,
+      "grad_norm": 0.6511228680610657,
+      "learning_rate": 7.660160382576683e-06,
+      "loss": 0.5876,
+      "step": 122
+    },
+    {
+      "epoch": 1.1826923076923077,
+      "grad_norm": 0.5735483765602112,
+      "learning_rate": 7.612492823579744e-06,
+      "loss": 0.5733,
+      "step": 123
+    },
+    {
+      "epoch": 1.1923076923076923,
+      "grad_norm": 0.48792481422424316,
+      "learning_rate": 7.564496387029532e-06,
+      "loss": 0.6047,
+      "step": 124
+    },
+    {
+      "epoch": 1.2019230769230769,
+      "grad_norm": 0.6753615736961365,
+      "learning_rate": 7.516177115029002e-06,
+      "loss": 0.5975,
+      "step": 125
+    },
+    {
+      "epoch": 1.2115384615384615,
+      "grad_norm": 0.567205011844635,
+      "learning_rate": 7.467541090321735e-06,
+      "loss": 0.5921,
+      "step": 126
+    },
+    {
+      "epoch": 1.2211538461538463,
+      "grad_norm": 0.623609185218811,
+      "learning_rate": 7.4185944355261996e-06,
+      "loss": 0.5726,
+      "step": 127
+    },
+    {
+      "epoch": 1.2307692307692308,
+      "grad_norm": 0.4969348609447479,
+      "learning_rate": 7.369343312364994e-06,
+      "loss": 0.5844,
+      "step": 128
+    },
+    {
+      "epoch": 1.2403846153846154,
+      "grad_norm": 0.5775588750839233,
+      "learning_rate": 7.319793920889171e-06,
+      "loss": 0.5758,
+      "step": 129
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 0.5946178436279297,
+      "learning_rate": 7.269952498697734e-06,
+      "loss": 0.5936,
+      "step": 130
+    },
+    {
+      "epoch": 1.2596153846153846,
+      "grad_norm": 0.501392662525177,
+      "learning_rate": 7.219825320152411e-06,
+      "loss": 0.591,
+      "step": 131
+    },
+    {
+      "epoch": 1.2692307692307692,
+      "grad_norm": 0.6227275133132935,
+      "learning_rate": 7.169418695587791e-06,
+      "loss": 0.6117,
+      "step": 132
+    },
+    {
+      "epoch": 1.2788461538461537,
+      "grad_norm": 0.49991685152053833,
+      "learning_rate": 7.118738970516944e-06,
+      "loss": 0.5771,
+      "step": 133
+    },
+    {
+      "epoch": 1.2884615384615383,
+      "grad_norm": 0.4866224229335785,
+      "learning_rate": 7.067792524832604e-06,
+      "loss": 0.6068,
+      "step": 134
+    },
+    {
+      "epoch": 1.2980769230769231,
+      "grad_norm": 0.5264644622802734,
+      "learning_rate": 7.016585772004026e-06,
+      "loss": 0.5748,
+      "step": 135
+    },
+    {
+      "epoch": 1.3076923076923077,
+      "grad_norm": 0.5308865904808044,
+      "learning_rate": 6.965125158269619e-06,
+      "loss": 0.5829,
+      "step": 136
+    },
+    {
+      "epoch": 1.3173076923076923,
+      "grad_norm": 0.530244767665863,
+      "learning_rate": 6.913417161825449e-06,
+      "loss": 0.5679,
+      "step": 137
+    },
+    {
+      "epoch": 1.3269230769230769,
+      "grad_norm": 0.4928182065486908,
+      "learning_rate": 6.8614682920097265e-06,
+      "loss": 0.5679,
+      "step": 138
+    },
+    {
+      "epoch": 1.3365384615384617,
+      "grad_norm": 0.523705244064331,
+      "learning_rate": 6.809285088483361e-06,
+      "loss": 0.5587,
+      "step": 139
+    },
+    {
+      "epoch": 1.3461538461538463,
+      "grad_norm": 0.48403921723365784,
+      "learning_rate": 6.7568741204067145e-06,
+      "loss": 0.5717,
+      "step": 140
+    },
+    {
+      "epoch": 1.3557692307692308,
+      "grad_norm": 0.551434338092804,
+      "learning_rate": 6.704241985612625e-06,
+      "loss": 0.6041,
+      "step": 141
+    },
+    {
+      "epoch": 1.3653846153846154,
+      "grad_norm": 0.501440703868866,
+      "learning_rate": 6.651395309775837e-06,
+      "loss": 0.5718,
+      "step": 142
+    },
+    {
+      "epoch": 1.375,
+      "grad_norm": 0.46742498874664307,
+      "learning_rate": 6.598340745578908e-06,
+      "loss": 0.5718,
+      "step": 143
+    },
+    {
+      "epoch": 1.3846153846153846,
+      "grad_norm": 0.5607274770736694,
+      "learning_rate": 6.545084971874738e-06,
+      "loss": 0.608,
+      "step": 144
+    },
+    {
+      "epoch": 1.3942307692307692,
+      "grad_norm": 0.5115028023719788,
+      "learning_rate": 6.491634692845781e-06,
+      "loss": 0.5636,
+      "step": 145
+    },
+    {
+      "epoch": 1.4038461538461537,
+      "grad_norm": 0.5423773527145386,
+      "learning_rate": 6.437996637160086e-06,
+      "loss": 0.5734,
+      "step": 146
+    },
+    {
+      "epoch": 1.4134615384615383,
+      "grad_norm": 0.4873090982437134,
+      "learning_rate": 6.384177557124247e-06,
+      "loss": 0.6,
+      "step": 147
+    },
+    {
+      "epoch": 1.4230769230769231,
+      "grad_norm": 0.4671023190021515,
+      "learning_rate": 6.330184227833376e-06,
+      "loss": 0.5843,
+      "step": 148
+    },
+    {
+      "epoch": 1.4326923076923077,
+      "grad_norm": 0.5089084506034851,
+      "learning_rate": 6.276023446318214e-06,
+      "loss": 0.5755,
+      "step": 149
+    },
+    {
+      "epoch": 1.4423076923076923,
+      "grad_norm": 0.4525885283946991,
+      "learning_rate": 6.2217020306894705e-06,
+      "loss": 0.5776,
+      "step": 150
+    },
+    {
+      "epoch": 1.4519230769230769,
+      "grad_norm": 0.5407394170761108,
+      "learning_rate": 6.1672268192795285e-06,
+      "loss": 0.5902,
+      "step": 151
+    },
+    {
+      "epoch": 1.4615384615384617,
+      "grad_norm": 0.4763994812965393,
+      "learning_rate": 6.112604669781572e-06,
+      "loss": 0.5825,
+      "step": 152
+    },
+    {
+      "epoch": 1.4711538461538463,
+      "grad_norm": 0.480131059885025,
+      "learning_rate": 6.057842458386315e-06,
+      "loss": 0.5854,
+      "step": 153
+    },
+    {
+      "epoch": 1.4807692307692308,
+      "grad_norm": 0.48260608315467834,
+      "learning_rate": 6.002947078916365e-06,
+      "loss": 0.5766,
+      "step": 154
+    },
+    {
+      "epoch": 1.4903846153846154,
+      "grad_norm": 0.44361093640327454,
+      "learning_rate": 5.947925441958393e-06,
+      "loss": 0.5834,
+      "step": 155
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 0.4906068742275238,
+      "learning_rate": 5.892784473993184e-06,
+      "loss": 0.5856,
+      "step": 156
+    },
+    {
+      "epoch": 1.5096153846153846,
+      "grad_norm": 0.48007476329803467,
+      "learning_rate": 5.837531116523683e-06,
+      "loss": 0.5735,
+      "step": 157
+    },
+    {
+      "epoch": 1.5192307692307692,
+      "grad_norm": 0.4426267147064209,
+      "learning_rate": 5.782172325201155e-06,
+      "loss": 0.5799,
+      "step": 158
+    },
+    {
+      "epoch": 1.5288461538461537,
+      "grad_norm": 0.45236214995384216,
+      "learning_rate": 5.726715068949564e-06,
+      "loss": 0.6024,
+      "step": 159
+    },
+    {
+      "epoch": 1.5384615384615383,
+      "grad_norm": 0.42147666215896606,
+      "learning_rate": 5.671166329088278e-06,
+      "loss": 0.5834,
+      "step": 160
+    },
+    {
+      "epoch": 1.5480769230769231,
+      "grad_norm": 0.42501798272132874,
+      "learning_rate": 5.615533098453215e-06,
+      "loss": 0.5828,
+      "step": 161
+    },
+    {
+      "epoch": 1.5576923076923077,
+      "grad_norm": 0.4713789224624634,
+      "learning_rate": 5.559822380516539e-06,
+      "loss": 0.5965,
+      "step": 162
+    },
+    {
+      "epoch": 1.5673076923076923,
+      "grad_norm": 0.438809871673584,
+      "learning_rate": 5.504041188505022e-06,
+      "loss": 0.572,
+      "step": 163
+    },
+    {
+      "epoch": 1.5769230769230769,
+      "grad_norm": 0.5145413279533386,
+      "learning_rate": 5.448196544517168e-06,
+      "loss": 0.5698,
+      "step": 164
+    },
+    {
+      "epoch": 1.5865384615384617,
+      "grad_norm": 0.5000028610229492,
+      "learning_rate": 5.392295478639226e-06,
+      "loss": 0.5943,
+      "step": 165
+    },
+    {
+      "epoch": 1.5961538461538463,
+      "grad_norm": 0.5032768845558167,
+      "learning_rate": 5.336345028060199e-06,
+      "loss": 0.5881,
+      "step": 166
+    },
+    {
+      "epoch": 1.6057692307692308,
+      "grad_norm": 0.4803272783756256,
+      "learning_rate": 5.2803522361859596e-06,
+      "loss": 0.5482,
+      "step": 167
+    },
+    {
+      "epoch": 1.6153846153846154,
+      "grad_norm": 0.4768068790435791,
+      "learning_rate": 5.224324151752575e-06,
+      "loss": 0.5694,
+      "step": 168
+    },
+    {
+      "epoch": 1.625,
+      "grad_norm": 0.5475027561187744,
+      "learning_rate": 5.168267827938971e-06,
+      "loss": 0.5868,
+      "step": 169
+    },
+    {
+      "epoch": 1.6346153846153846,
+      "grad_norm": 0.5247706770896912,
+      "learning_rate": 5.112190321479026e-06,
+      "loss": 0.5903,
+      "step": 170
+    },
+    {
+      "epoch": 1.6442307692307692,
+      "grad_norm": 0.45129239559173584,
+      "learning_rate": 5.05609869177323e-06,
+      "loss": 0.5673,
+      "step": 171
+    },
+    {
+      "epoch": 1.6538461538461537,
+      "grad_norm": 0.4390396773815155,
+      "learning_rate": 5e-06,
+      "loss": 0.5749,
+      "step": 172
+    },
+    {
+      "epoch": 1.6634615384615383,
+      "grad_norm": 0.5230726003646851,
+      "learning_rate": 4.943901308226771e-06,
+      "loss": 0.5734,
+      "step": 173
+    },
+    {
+      "epoch": 1.6730769230769231,
+      "grad_norm": 0.5058849453926086,
+      "learning_rate": 4.887809678520976e-06,
+      "loss": 0.5902,
+      "step": 174
+    },
+    {
+      "epoch": 1.6826923076923077,
+      "grad_norm": 0.5016485452651978,
+      "learning_rate": 4.831732172061032e-06,
+      "loss": 0.5726,
+      "step": 175
+    },
+    {
+      "epoch": 1.6923076923076923,
+      "grad_norm": 0.5082340836524963,
+      "learning_rate": 4.775675848247427e-06,
+      "loss": 0.5868,
+      "step": 176
+    },
+    {
+      "epoch": 1.7019230769230769,
+      "grad_norm": 0.43624982237815857,
+      "learning_rate": 4.719647763814041e-06,
+      "loss": 0.6001,
+      "step": 177
+    },
+    {
+      "epoch": 1.7115384615384617,
+      "grad_norm": 0.4955120086669922,
+      "learning_rate": 4.663654971939802e-06,
+      "loss": 0.5661,
+      "step": 178
+    },
+    {
+      "epoch": 1.7211538461538463,
+      "grad_norm": 0.5228602886199951,
+      "learning_rate": 4.6077045213607765e-06,
+      "loss": 0.5749,
+      "step": 179
+    },
+    {
+      "epoch": 1.7307692307692308,
+      "grad_norm": 0.4157963693141937,
+      "learning_rate": 4.551803455482833e-06,
+      "loss": 0.5781,
+      "step": 180
+    },
+    {
+      "epoch": 1.7403846153846154,
+      "grad_norm": 0.4823377728462219,
+      "learning_rate": 4.4959588114949785e-06,
+      "loss": 0.5713,
+      "step": 181
+    },
+    {
+      "epoch": 1.75,
+      "grad_norm": 0.49718138575553894,
+      "learning_rate": 4.4401776194834615e-06,
+      "loss": 0.5971,
+      "step": 182
+    },
+    {
+      "epoch": 1.7596153846153846,
+      "grad_norm": 0.4611460566520691,
+      "learning_rate": 4.384466901546786e-06,
+      "loss": 0.5757,
+      "step": 183
+    },
+    {
+      "epoch": 1.7692307692307692,
+      "grad_norm": 0.4549019932746887,
+      "learning_rate": 4.3288336709117246e-06,
+      "loss": 0.5823,
+      "step": 184
+    },
+    {
+      "epoch": 1.7788461538461537,
+      "grad_norm": 0.45323625206947327,
+      "learning_rate": 4.273284931050438e-06,
+      "loss": 0.5728,
+      "step": 185
+    },
+    {
+      "epoch": 1.7884615384615383,
+      "grad_norm": 0.49955999851226807,
+      "learning_rate": 4.217827674798845e-06,
+      "loss": 0.5789,
+      "step": 186
+    },
+    {
+      "epoch": 1.7980769230769231,
+      "grad_norm": 0.5281660556793213,
+      "learning_rate": 4.162468883476319e-06,
+      "loss": 0.5449,
+      "step": 187
+    },
+    {
+      "epoch": 1.8076923076923077,
+      "grad_norm": 0.46816229820251465,
+      "learning_rate": 4.107215526006818e-06,
+      "loss": 0.5789,
+      "step": 188
+    },
+    {
+      "epoch": 1.8173076923076923,
+      "grad_norm": 0.5172679424285889,
+      "learning_rate": 4.052074558041608e-06,
+      "loss": 0.5634,
+      "step": 189
+    },
+    {
+      "epoch": 1.8269230769230769,
+      "grad_norm": 0.5312298536300659,
+      "learning_rate": 3.997052921083637e-06,
+      "loss": 0.5795,
+      "step": 190
+    },
+    {
+      "epoch": 1.8365384615384617,
+      "grad_norm": 0.40530455112457275,
+      "learning_rate": 3.9421575416136866e-06,
+      "loss": 0.5707,
+      "step": 191
+    },
+    {
+      "epoch": 1.8461538461538463,
+      "grad_norm": 0.4245462715625763,
+      "learning_rate": 3.887395330218429e-06,
+      "loss": 0.5764,
+      "step": 192
+    },
+    {
+      "epoch": 1.8557692307692308,
+      "grad_norm": 0.4844348430633545,
+      "learning_rate": 3.832773180720475e-06,
+      "loss": 0.5847,
+      "step": 193
+    },
+    {
+      "epoch": 1.8653846153846154,
+      "grad_norm": 0.3969320058822632,
+      "learning_rate": 3.778297969310529e-06,
+      "loss": 0.5776,
+      "step": 194
+    },
+    {
+      "epoch": 1.875,
+      "grad_norm": 0.4481745958328247,
+      "learning_rate": 3.723976553681787e-06,
+      "loss": 0.5834,
+      "step": 195
+    },
+    {
+      "epoch": 1.8846153846153846,
+      "grad_norm": 0.4347081482410431,
+      "learning_rate": 3.669815772166625e-06,
+      "loss": 0.5761,
+      "step": 196
+    },
+    {
+      "epoch": 1.8942307692307692,
+      "grad_norm": 0.4153216779232025,
+      "learning_rate": 3.6158224428757538e-06,
+      "loss": 0.5648,
+      "step": 197
+    },
+    {
+      "epoch": 1.9038461538461537,
+      "grad_norm": 0.4166128933429718,
+      "learning_rate": 3.562003362839914e-06,
+      "loss": 0.5709,
+      "step": 198
+    },
+    {
+      "epoch": 1.9134615384615383,
+      "grad_norm": 0.45052772760391235,
+      "learning_rate": 3.50836530715422e-06,
+      "loss": 0.5691,
+      "step": 199
+    },
+    {
+      "epoch": 1.9230769230769231,
+      "grad_norm": 0.43254920840263367,
+      "learning_rate": 3.4549150281252635e-06,
+      "loss": 0.5632,
+      "step": 200
+    },
+    {
+      "epoch": 1.9326923076923077,
+      "grad_norm": 0.4649442136287689,
+      "learning_rate": 3.4016592544210937e-06,
+      "loss": 0.5666,
+      "step": 201
+    },
+    {
+      "epoch": 1.9423076923076923,
+      "grad_norm": 0.4219006597995758,
+      "learning_rate": 3.3486046902241663e-06,
+      "loss": 0.588,
+      "step": 202
+    },
+    {
+      "epoch": 1.9519230769230769,
+      "grad_norm": 0.43250784277915955,
+      "learning_rate": 3.295758014387375e-06,
+      "loss": 0.5787,
+      "step": 203
+    },
+    {
+      "epoch": 1.9615384615384617,
+      "grad_norm": 0.4511379897594452,
+      "learning_rate": 3.2431258795932863e-06,
+      "loss": 0.5766,
+      "step": 204
+    },
+    {
+      "epoch": 1.9711538461538463,
+      "grad_norm": 0.41626155376434326,
+      "learning_rate": 3.1907149115166403e-06,
+      "loss": 0.5714,
+      "step": 205
+    },
+    {
+      "epoch": 1.9807692307692308,
+      "grad_norm": 0.4738382399082184,
+      "learning_rate": 3.1385317079902743e-06,
+      "loss": 0.5858,
+      "step": 206
+    },
+    {
+      "epoch": 1.9903846153846154,
+      "grad_norm": 0.4208838939666748,
+      "learning_rate": 3.0865828381745515e-06,
+      "loss": 0.5735,
+      "step": 207
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.4607163369655609,
+      "learning_rate": 3.0348748417303826e-06,
+      "loss": 0.5677,
+      "step": 208
+    },
+    {
+      "epoch": 2.0096153846153846,
+      "grad_norm": 0.42707422375679016,
+      "learning_rate": 2.9834142279959754e-06,
+      "loss": 0.5413,
+      "step": 209
+    },
+    {
+      "epoch": 2.019230769230769,
+      "grad_norm": 0.4337131381034851,
+      "learning_rate": 2.932207475167398e-06,
+      "loss": 0.548,
+      "step": 210
+    },
+    {
+      "epoch": 2.0288461538461537,
+      "grad_norm": 0.3963165283203125,
+      "learning_rate": 2.8812610294830568e-06,
+      "loss": 0.5441,
+      "step": 211
+    },
+    {
+      "epoch": 2.0384615384615383,
+      "grad_norm": 0.3808555006980896,
+      "learning_rate": 2.83058130441221e-06,
+      "loss": 0.5317,
+      "step": 212
+    },
+    {
+      "epoch": 2.048076923076923,
+      "grad_norm": 0.41117677092552185,
+      "learning_rate": 2.7801746798475905e-06,
+      "loss": 0.5275,
+      "step": 213
+    },
+    {
+      "epoch": 2.0576923076923075,
+      "grad_norm": 0.3876531422138214,
+      "learning_rate": 2.7300475013022666e-06,
+      "loss": 0.5631,
+      "step": 214
+    },
+    {
+      "epoch": 2.0673076923076925,
+      "grad_norm": 0.39290904998779297,
+      "learning_rate": 2.6802060791108304e-06,
+      "loss": 0.5447,
+      "step": 215
+    },
+    {
+      "epoch": 2.076923076923077,
+      "grad_norm": 0.38423505425453186,
+      "learning_rate": 2.6306566876350072e-06,
+      "loss": 0.535,
+      "step": 216
+    },
+    {
+      "epoch": 2.0865384615384617,
+      "grad_norm": 0.3680586814880371,
+      "learning_rate": 2.5814055644738013e-06,
+      "loss": 0.5534,
+      "step": 217
+    },
+    {
+      "epoch": 2.0961538461538463,
+      "grad_norm": 0.4062810242176056,
+      "learning_rate": 2.532458909678266e-06,
+      "loss": 0.5499,
+      "step": 218
+    },
+    {
+      "epoch": 2.105769230769231,
+      "grad_norm": 0.39526402950286865,
+      "learning_rate": 2.483822884971e-06,
+      "loss": 0.5393,
+      "step": 219
+    },
+    {
+      "epoch": 2.1153846153846154,
+      "grad_norm": 0.4416036903858185,
+      "learning_rate": 2.43550361297047e-06,
+      "loss": 0.5432,
+      "step": 220
+    },
+    {
+      "epoch": 2.125,
+      "grad_norm": 0.40109318494796753,
+      "learning_rate": 2.387507176420256e-06,
+      "loss": 0.5167,
+      "step": 221
+    },
+    {
+      "epoch": 2.1346153846153846,
+      "grad_norm": 0.37532350420951843,
+      "learning_rate": 2.339839617423318e-06,
+      "loss": 0.5517,
+      "step": 222
+    },
+    {
+      "epoch": 2.144230769230769,
+      "grad_norm": 0.36060187220573425,
+      "learning_rate": 2.2925069366813718e-06,
+      "loss": 0.5388,
+      "step": 223
+    },
+    {
+      "epoch": 2.1538461538461537,
+      "grad_norm": 0.41168034076690674,
+      "learning_rate": 2.245515092739488e-06,
+      "loss": 0.5351,
+      "step": 224
+    },
+    {
+      "epoch": 2.1634615384615383,
+      "grad_norm": 0.3630206286907196,
+      "learning_rate": 2.1988700012359865e-06,
+      "loss": 0.5398,
+      "step": 225
+    },
+    {
+      "epoch": 2.173076923076923,
+      "grad_norm": 0.36833643913269043,
+      "learning_rate": 2.1525775341577404e-06,
+      "loss": 0.5399,
+      "step": 226
+    },
+    {
+      "epoch": 2.1826923076923075,
+      "grad_norm": 0.3559540808200836,
+      "learning_rate": 2.1066435191009717e-06,
+      "loss": 0.535,
+      "step": 227
+    },
+    {
+      "epoch": 2.1923076923076925,
+      "grad_norm": 0.3911932706832886,
+      "learning_rate": 2.061073738537635e-06,
+      "loss": 0.5203,
+      "step": 228
+    },
+    {
+      "epoch": 2.201923076923077,
+      "grad_norm": 0.3908790647983551,
+      "learning_rate": 2.0158739290874822e-06,
+      "loss": 0.5592,
+      "step": 229
+    },
+    {
+      "epoch": 2.2115384615384617,
+      "grad_norm": 0.38927578926086426,
+      "learning_rate": 1.971049780795901e-06,
+      "loss": 0.5399,
+      "step": 230
+    },
+    {
+      "epoch": 2.2211538461538463,
+      "grad_norm": 0.3874114751815796,
+      "learning_rate": 1.9266069364176144e-06,
+      "loss": 0.522,
+      "step": 231
+    },
+    {
+      "epoch": 2.230769230769231,
+      "grad_norm": 0.37980931997299194,
+      "learning_rate": 1.8825509907063328e-06,
+      "loss": 0.5227,
+      "step": 232
+    },
+    {
+      "epoch": 2.2403846153846154,
+      "grad_norm": 0.3756473958492279,
+      "learning_rate": 1.838887489710452e-06,
+      "loss": 0.537,
+      "step": 233
+    },
+    {
+      "epoch": 2.25,
+      "grad_norm": 0.36786749958992004,
+      "learning_rate": 1.7956219300748796e-06,
+      "loss": 0.5505,
+      "step": 234
+    },
+    {
+      "epoch": 2.2596153846153846,
+      "grad_norm": 0.361908495426178,
+      "learning_rate": 1.7527597583490825e-06,
+      "loss": 0.5678,
+      "step": 235
+    },
+    {
+      "epoch": 2.269230769230769,
+      "grad_norm": 0.339290052652359,
+      "learning_rate": 1.7103063703014372e-06,
+      "loss": 0.5369,
+      "step": 236
+    },
+    {
+      "epoch": 2.2788461538461537,
+      "grad_norm": 0.412014365196228,
+      "learning_rate": 1.6682671102399806e-06,
+      "loss": 0.5319,
+      "step": 237
+    },
+    {
+      "epoch": 2.2884615384615383,
+      "grad_norm": 0.3827563524246216,
+      "learning_rate": 1.6266472703396286e-06,
+      "loss": 0.5492,
+      "step": 238
+    },
+    {
+      "epoch": 2.298076923076923,
+      "grad_norm": 0.35775306820869446,
+      "learning_rate": 1.5854520899759656e-06,
+      "loss": 0.5322,
+      "step": 239
+    },
+    {
+      "epoch": 2.3076923076923075,
+      "grad_norm": 0.3594055771827698,
+      "learning_rate": 1.544686755065677e-06,
+      "loss": 0.553,
+      "step": 240
+    },
+    {
+      "epoch": 2.3173076923076925,
+      "grad_norm": 0.3411286473274231,
+      "learning_rate": 1.5043563974137132e-06,
+      "loss": 0.546,
+      "step": 241
+    },
+    {
+      "epoch": 2.326923076923077,
+      "grad_norm": 0.42784562706947327,
+      "learning_rate": 1.4644660940672628e-06,
+      "loss": 0.5357,
+      "step": 242
+    },
+    {
+      "epoch": 2.3365384615384617,
+      "grad_norm": 0.387997567653656,
+      "learning_rate": 1.4250208666766235e-06,
+      "loss": 0.5192,
+      "step": 243
+    },
+    {
+      "epoch": 2.3461538461538463,
+      "grad_norm": 0.36996373534202576,
+      "learning_rate": 1.3860256808630429e-06,
+      "loss": 0.5457,
+      "step": 244
+    },
+    {
+      "epoch": 2.355769230769231,
+      "grad_norm": 0.3782255947589874,
+      "learning_rate": 1.3474854455936126e-06,
+      "loss": 0.5543,
+      "step": 245
+    },
+    {
+      "epoch": 2.3653846153846154,
+      "grad_norm": 0.38931819796562195,
+      "learning_rate": 1.3094050125632973e-06,
+      "loss": 0.5661,
+      "step": 246
+    },
+    {
+      "epoch": 2.375,
+      "grad_norm": 0.36775466799736023,
+      "learning_rate": 1.2717891755841722e-06,
+      "loss": 0.5461,
+      "step": 247
+    },
+    {
+      "epoch": 2.3846153846153846,
+      "grad_norm": 0.3692035675048828,
+      "learning_rate": 1.234642669981946e-06,
+      "loss": 0.5263,
+      "step": 248
+    },
+    {
+      "epoch": 2.394230769230769,
+      "grad_norm": 0.3765394389629364,
+      "learning_rate": 1.1979701719998454e-06,
+      "loss": 0.5478,
+      "step": 249
+    },
+    {
+      "epoch": 2.4038461538461537,
+      "grad_norm": 0.36372944712638855,
+      "learning_rate": 1.1617762982099446e-06,
+      "loss": 0.5534,
+      "step": 250
+    },
+    {
+      "epoch": 2.4134615384615383,
+      "grad_norm": 0.3634006083011627,
+      "learning_rate": 1.1260656049319957e-06,
+      "loss": 0.5366,
+      "step": 251
+    },
+    {
+      "epoch": 2.423076923076923,
+      "grad_norm": 0.3760945796966553,
+      "learning_rate": 1.0908425876598512e-06,
+      "loss": 0.5351,
+      "step": 252
+    },
+    {
+      "epoch": 2.4326923076923075,
+      "grad_norm": 0.3575971722602844,
+      "learning_rate": 1.0561116804955451e-06,
+      "loss": 0.5445,
+      "step": 253
+    },
+    {
+      "epoch": 2.4423076923076925,
+      "grad_norm": 0.3454532325267792,
+      "learning_rate": 1.0218772555910955e-06,
+      "loss": 0.5545,
+      "step": 254
+    },
+    {
+      "epoch": 2.451923076923077,
+      "grad_norm": 0.3429204225540161,
+      "learning_rate": 9.881436225981107e-07,
+      "loss": 0.5489,
+      "step": 255
+    },
+    {
+      "epoch": 2.4615384615384617,
+      "grad_norm": 0.37573421001434326,
+      "learning_rate": 9.549150281252633e-07,
+      "loss": 0.524,
+      "step": 256
+    },
+    {
+      "epoch": 2.4711538461538463,
+      "grad_norm": 0.3568671941757202,
+      "learning_rate": 9.221956552036992e-07,
+      "loss": 0.5328,
+      "step": 257
+    },
+    {
+      "epoch": 2.480769230769231,
+      "grad_norm": 0.3483606278896332,
+      "learning_rate": 8.899896227604509e-07,
+      "loss": 0.561,
+      "step": 258
+    },
+    {
+      "epoch": 2.4903846153846154,
+      "grad_norm": 0.34958329796791077,
+      "learning_rate": 8.58300985099918e-07,
+      "loss": 0.5525,
+      "step": 259
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 0.35783374309539795,
+      "learning_rate": 8.271337313934869e-07,
+      "loss": 0.5313,
+      "step": 260
+    },
+    {
+      "epoch": 2.5096153846153846,
+      "grad_norm": 0.34240663051605225,
+      "learning_rate": 7.964917851773496e-07,
+      "loss": 0.5327,
+      "step": 261
+    },
+    {
+      "epoch": 2.519230769230769,
+      "grad_norm": 0.3467879891395569,
+      "learning_rate": 7.663790038585794e-07,
+      "loss": 0.5375,
+      "step": 262
+    },
+    {
+      "epoch": 2.5288461538461537,
+      "grad_norm": 0.34128859639167786,
+      "learning_rate": 7.367991782295392e-07,
+      "loss": 0.5497,
+      "step": 263
+    },
+    {
+      "epoch": 2.5384615384615383,
+      "grad_norm": 0.35784712433815,
+      "learning_rate": 7.077560319906696e-07,
+      "loss": 0.5425,
+      "step": 264
+    },
+    {
+      "epoch": 2.5480769230769234,
+      "grad_norm": 0.359834223985672,
+      "learning_rate": 6.792532212817271e-07,
+      "loss": 0.548,
+      "step": 265
+    },
+    {
+      "epoch": 2.5576923076923075,
+      "grad_norm": 0.36030974984169006,
+      "learning_rate": 6.512943342215234e-07,
+      "loss": 0.5527,
+      "step": 266
+    },
+    {
+      "epoch": 2.5673076923076925,
+      "grad_norm": 0.3154827058315277,
+      "learning_rate": 6.238828904562316e-07,
+      "loss": 0.5562,
+      "step": 267
+    },
+    {
+      "epoch": 2.5769230769230766,
+      "grad_norm": 0.35393068194389343,
+      "learning_rate": 5.9702234071631e-07,
+      "loss": 0.5426,
+      "step": 268
+    },
+    {
+      "epoch": 2.5865384615384617,
+      "grad_norm": 0.3374582529067993,
+      "learning_rate": 5.707160663821009e-07,
+      "loss": 0.5482,
+      "step": 269
+    },
+    {
+      "epoch": 2.5961538461538463,
+      "grad_norm": 0.3574032485485077,
+      "learning_rate": 5.449673790581611e-07,
+      "loss": 0.5282,
+      "step": 270
+    },
+    {
+      "epoch": 2.605769230769231,
+      "grad_norm": 0.36037665605545044,
+      "learning_rate": 5.197795201563744e-07,
+      "loss": 0.5496,
+      "step": 271
+    },
+    {
+      "epoch": 2.6153846153846154,
+      "grad_norm": 0.3468829095363617,
+      "learning_rate": 4.951556604879049e-07,
+      "loss": 0.5476,
+      "step": 272
+    },
+    {
+      "epoch": 2.625,
+      "grad_norm": 0.3324459493160248,
+      "learning_rate": 4.710988998640298e-07,
+      "loss": 0.5586,
+      "step": 273
+    },
+    {
+      "epoch": 2.6346153846153846,
+      "grad_norm": 0.33732524514198303,
+      "learning_rate": 4.4761226670592074e-07,
+      "loss": 0.5531,
+      "step": 274
+    },
+    {
+      "epoch": 2.644230769230769,
+      "grad_norm": 0.35106703639030457,
+      "learning_rate": 4.2469871766340096e-07,
+      "loss": 0.5454,
+      "step": 275
+    },
+    {
+      "epoch": 2.6538461538461537,
+      "grad_norm": 0.36037111282348633,
+      "learning_rate": 4.0236113724274716e-07,
+      "loss": 0.5415,
+      "step": 276
+    },
+    {
+      "epoch": 2.6634615384615383,
+      "grad_norm": 0.3727104961872101,
+      "learning_rate": 3.8060233744356634e-07,
+      "loss": 0.5286,
+      "step": 277
+    },
+    {
+      "epoch": 2.6730769230769234,
+      "grad_norm": 0.3624821901321411,
+      "learning_rate": 3.5942505740480583e-07,
+      "loss": 0.5483,
+      "step": 278
+    },
+    {
+      "epoch": 2.6826923076923075,
+      "grad_norm": 0.34953874349594116,
+      "learning_rate": 3.3883196305992906e-07,
+      "loss": 0.5397,
+      "step": 279
+    },
+    {
+      "epoch": 2.6923076923076925,
+      "grad_norm": 0.31536760926246643,
+      "learning_rate": 3.18825646801314e-07,
+      "loss": 0.5813,
+      "step": 280
+    },
+    {
+      "epoch": 2.7019230769230766,
+      "grad_norm": 0.34410393238067627,
+      "learning_rate": 2.9940862715390483e-07,
+      "loss": 0.5477,
+      "step": 281
+    },
+    {
+      "epoch": 2.7115384615384617,
+      "grad_norm": 0.33748528361320496,
+      "learning_rate": 2.8058334845816214e-07,
+      "loss": 0.5699,
+      "step": 282
+    },
+    {
+      "epoch": 2.7211538461538463,
+      "grad_norm": 0.3506697416305542,
+      "learning_rate": 2.6235218056235633e-07,
+      "loss": 0.5141,
+      "step": 283
+    },
+    {
+      "epoch": 2.730769230769231,
+      "grad_norm": 0.3356882333755493,
+      "learning_rate": 2.447174185242324e-07,
+      "loss": 0.5593,
+      "step": 284
+    },
+    {
+      "epoch": 2.7403846153846154,
+      "grad_norm": 0.34487199783325195,
+      "learning_rate": 2.276812823220964e-07,
+      "loss": 0.5575,
+      "step": 285
+    },
+    {
+      "epoch": 2.75,
+      "grad_norm": 0.333774596452713,
+      "learning_rate": 2.1124591657534776e-07,
+      "loss": 0.5249,
+      "step": 286
+    },
+    {
+      "epoch": 2.7596153846153846,
+      "grad_norm": 0.3721064031124115,
+      "learning_rate": 1.9541339027450256e-07,
+      "loss": 0.5234,
+      "step": 287
+    },
+    {
+      "epoch": 2.769230769230769,
+      "grad_norm": 0.35415223240852356,
+      "learning_rate": 1.801856965207338e-07,
+      "loss": 0.5215,
+      "step": 288
+    },
+    {
+      "epoch": 2.7788461538461537,
+      "grad_norm": 0.3103128969669342,
+      "learning_rate": 1.6556475227496816e-07,
+      "loss": 0.5574,
+      "step": 289
+    },
+    {
+      "epoch": 2.7884615384615383,
+      "grad_norm": 0.35190919041633606,
+      "learning_rate": 1.5155239811656562e-07,
+      "loss": 0.5465,
+      "step": 290
+    },
+    {
+      "epoch": 2.7980769230769234,
+      "grad_norm": 0.3369409143924713,
+      "learning_rate": 1.3815039801161723e-07,
+      "loss": 0.5469,
+      "step": 291
+    },
+    {
+      "epoch": 2.8076923076923075,
+      "grad_norm": 0.3374597430229187,
+      "learning_rate": 1.253604390908819e-07,
+      "loss": 0.5399,
+      "step": 292
+    },
+    {
+      "epoch": 2.8173076923076925,
+      "grad_norm": 0.3631032407283783,
+      "learning_rate": 1.1318413143740436e-07,
+      "loss": 0.5451,
+      "step": 293
+    },
+    {
+      "epoch": 2.8269230769230766,
+      "grad_norm": 0.30834081768989563,
+      "learning_rate": 1.0162300788382263e-07,
+      "loss": 0.5584,
+      "step": 294
+    },
+    {
+      "epoch": 2.8365384615384617,
+      "grad_norm": 0.3387204706668854,
+      "learning_rate": 9.0678523819408e-08,
+      "loss": 0.5332,
+      "step": 295
+    },
+    {
+      "epoch": 2.8461538461538463,
+      "grad_norm": 0.35595011711120605,
+      "learning_rate": 8.035205700685167e-08,
+      "loss": 0.5622,
+      "step": 296
+    },
+    {
+      "epoch": 2.855769230769231,
+      "grad_norm": 0.33064529299736023,
+      "learning_rate": 7.064490740882057e-08,
+      "loss": 0.5351,
+      "step": 297
+    },
+    {
+      "epoch": 2.8653846153846154,
+      "grad_norm": 0.3262805938720703,
+      "learning_rate": 6.15582970243117e-08,
+      "loss": 0.5532,
+      "step": 298
+    },
+    {
+      "epoch": 2.875,
+      "grad_norm": 0.3150298297405243,
+      "learning_rate": 5.3093369734816824e-08,
+      "loss": 0.5441,
+      "step": 299
+    },
+    {
+      "epoch": 2.8846153846153846,
+      "grad_norm": 0.32384297251701355,
+      "learning_rate": 4.52511911603265e-08,
+      "loss": 0.5482,
+      "step": 300
+    },
+    {
+      "epoch": 2.894230769230769,
+      "grad_norm": 0.4312896728515625,
+      "learning_rate": 3.8032748525179684e-08,
+      "loss": 0.5473,
+      "step": 301
+    },
+    {
+      "epoch": 2.9038461538461537,
+      "grad_norm": 0.3357475697994232,
+      "learning_rate": 3.143895053378698e-08,
+      "loss": 0.5371,
+      "step": 302
+    },
+    {
+      "epoch": 2.9134615384615383,
+      "grad_norm": 0.34379279613494873,
+      "learning_rate": 2.547062725623828e-08,
+      "loss": 0.5424,
+      "step": 303
+    },
+    {
+      "epoch": 2.9230769230769234,
+      "grad_norm": 0.3341004252433777,
+      "learning_rate": 2.012853002380466e-08,
+      "loss": 0.5553,
+      "step": 304
+    },
+    {
+      "epoch": 2.9326923076923075,
+      "grad_norm": 0.3333733081817627,
+      "learning_rate": 1.541333133436018e-08,
+      "loss": 0.5281,
+      "step": 305
+    },
+    {
+      "epoch": 2.9423076923076925,
+      "grad_norm": 0.3348696529865265,
+      "learning_rate": 1.132562476771959e-08,
+      "loss": 0.5465,
+      "step": 306
+    },
+    {
+      "epoch": 2.9519230769230766,
+      "grad_norm": 0.35991808772087097,
+      "learning_rate": 7.865924910916977e-09,
+      "loss": 0.5219,
+      "step": 307
+    },
+    {
+      "epoch": 2.9615384615384617,
+      "grad_norm": 0.3342251479625702,
+      "learning_rate": 5.034667293427053e-09,
+      "loss": 0.5476,
+      "step": 308
+    },
+    {
+      "epoch": 2.9711538461538463,
+      "grad_norm": 0.32228055596351624,
+      "learning_rate": 2.8322083323334417e-09,
+      "loss": 0.5605,
+      "step": 309
+    },
+    {
+      "epoch": 2.980769230769231,
+      "grad_norm": 0.31481924653053284,
+      "learning_rate": 1.2588252874673469e-09,
+      "loss": 0.5469,
+      "step": 310
+    },
+    {
+      "epoch": 2.9903846153846154,
+      "grad_norm": 0.31989800930023193,
+      "learning_rate": 3.147162264971471e-10,
+      "loss": 0.548,
+      "step": 311
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.3253592550754547,
+      "learning_rate": 0.0,
+      "loss": 0.5313,
+      "step": 312
+    },
+    {
+      "epoch": 3.0,
+      "step": 312,
+      "total_flos": 582499086630912.0,
+      "train_loss": 0.6133231129019688,
+      "train_runtime": 4436.2424,
+      "train_samples_per_second": 6.73,
+      "train_steps_per_second": 0.07
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 312,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 582499086630912.0,
+  "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:cd60627509acb96a599d677f9e8d8c20d551809022b041e0a6ba9364649c05c8
+size 7672