update

config.json

@@ -0,0 +1,42 @@
+{
+  "_name_or_path": "/mnt/petrelfs/chenqin.p/dyh/Qwen-14B-Chat/",
+  "architectures": [
+    "QWenLMHeadModel"
+  ],
+  "attn_dropout_prob": 0.0,
+  "auto_map": {
+    "AutoConfig": "configuration_qwen.QWenConfig",
+    "AutoModelForCausalLM": "modeling_qwen.QWenLMHeadModel"
+  },
+  "bf16": false,
+  "emb_dropout_prob": 0.0,
+  "fp16": true,
+  "fp32": false,
+  "hidden_size": 5120,
+  "initializer_range": 0.02,
+  "intermediate_size": 27392,
+  "kv_channels": 128,
+  "layer_norm_epsilon": 1e-06,
+  "max_position_embeddings": 8192,
+  "model_type": "qwen",
+  "no_bias": true,
+  "num_attention_heads": 40,
+  "num_hidden_layers": 40,
+  "onnx_safe": null,
+  "rotary_emb_base": 10000,
+  "rotary_pct": 1.0,
+  "scale_attn_weights": true,
+  "seq_length": 2048,
+  "softmax_in_fp32": false,
+  "tie_word_embeddings": false,
+  "tokenizer_class": "QWenTokenizer",
+  "torch_dtype": "float16",
+  "transformers_version": "4.35.2",
+  "use_cache": true,
+  "use_cache_kernel": false,
+  "use_cache_quantization": false,
+  "use_dynamic_ntk": true,
+  "use_flash_attn": true,
+  "use_logn_attn": true,
+  "vocab_size": 151872
+}

generation_config.json

@@ -0,0 +1,13 @@
+{
+  "chat_format": "chatml",
+  "do_sample": true,
+  "eos_token_id": 151643,
+  "fp16": true,
+  "max_new_tokens": 512,
+  "max_window_size": 6144,
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.1,
+  "top_k": 0,
+  "top_p": 0.8,
+  "transformers_version": "4.35.2"
+}

model-00001-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c05be6c984126251b4d03ddf055a738345e8ba90b94ea75c9ba7f33d59904363
+size 4917478208

model-00002-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b304afcf8cd231002064e97fe1c4f64dd029968445dafcf46d35f47ceecc06f
+size 4991627808

model-00003-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:edad5e2fa01836573ab6f62d2b721eb1c276b2c46d45b0fc9ab3e48e2ca12ca5
+size 4886749760

model-00004-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d663b2745eb827aadc0e8cab37939f47fd44c21232de437684dbb5402ea89740
+size 4903809600

model-00005-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4edd574e2434c9d4671e20ce8945574972180f2be048304b1744d99bb0f351c1
+size 4903819952

model-00006-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:614e9b45984d060708cddc067003acea1a31cbf6369038908bf990fbfcc7aefd
+size 3727199208

model.safetensors.index.json

@@ -0,0 +1,330 @@
+{
+  "metadata": {
+    "total_size": 28330649600
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.0.attn.c_attn.bias": "model-00001-of-00006.safetensors",
+    "transformer.h.0.attn.c_attn.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.0.attn.c_proj.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.0.ln_1.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.0.ln_2.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.0.mlp.c_proj.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.0.mlp.w1.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.0.mlp.w2.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.1.attn.c_attn.bias": "model-00001-of-00006.safetensors",
+    "transformer.h.1.attn.c_attn.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.1.attn.c_proj.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.1.ln_1.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.1.ln_2.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.1.mlp.c_proj.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.1.mlp.w1.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.1.mlp.w2.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.10.attn.c_attn.bias": "model-00002-of-00006.safetensors",
+    "transformer.h.10.attn.c_attn.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.10.attn.c_proj.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.10.ln_1.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.10.ln_2.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.10.mlp.c_proj.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.10.mlp.w1.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.10.mlp.w2.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.11.attn.c_attn.bias": "model-00002-of-00006.safetensors",
+    "transformer.h.11.attn.c_attn.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.11.attn.c_proj.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.11.ln_1.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.11.ln_2.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.11.mlp.c_proj.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.11.mlp.w1.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.11.mlp.w2.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.12.attn.c_attn.bias": "model-00002-of-00006.safetensors",
+    "transformer.h.12.attn.c_attn.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.12.attn.c_proj.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.12.ln_1.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.12.ln_2.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.12.mlp.c_proj.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.12.mlp.w1.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.12.mlp.w2.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.13.attn.c_attn.bias": "model-00002-of-00006.safetensors",
+    "transformer.h.13.attn.c_attn.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.13.attn.c_proj.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.13.ln_1.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.13.ln_2.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.13.mlp.c_proj.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.13.mlp.w1.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.13.mlp.w2.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.14.attn.c_attn.bias": "model-00003-of-00006.safetensors",
+    "transformer.h.14.attn.c_attn.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.14.attn.c_proj.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.14.ln_1.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.14.ln_2.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.14.mlp.c_proj.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.14.mlp.w1.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.14.mlp.w2.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.15.attn.c_attn.bias": "model-00003-of-00006.safetensors",
+    "transformer.h.15.attn.c_attn.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.15.attn.c_proj.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.15.ln_1.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.15.ln_2.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.15.mlp.c_proj.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.15.mlp.w1.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.15.mlp.w2.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.16.attn.c_attn.bias": "model-00003-of-00006.safetensors",
+    "transformer.h.16.attn.c_attn.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.16.attn.c_proj.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.16.ln_1.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.16.ln_2.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.16.mlp.c_proj.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.16.mlp.w1.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.16.mlp.w2.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.17.attn.c_attn.bias": "model-00003-of-00006.safetensors",
+    "transformer.h.17.attn.c_attn.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.17.attn.c_proj.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.17.ln_1.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.17.ln_2.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.17.mlp.c_proj.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.17.mlp.w1.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.17.mlp.w2.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.18.attn.c_attn.bias": "model-00003-of-00006.safetensors",
+    "transformer.h.18.attn.c_attn.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.18.attn.c_proj.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.18.ln_1.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.18.ln_2.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.18.mlp.c_proj.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.18.mlp.w1.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.18.mlp.w2.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.19.attn.c_attn.bias": "model-00003-of-00006.safetensors",
+    "transformer.h.19.attn.c_attn.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.19.attn.c_proj.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.19.ln_1.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.19.ln_2.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.19.mlp.c_proj.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.19.mlp.w1.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.19.mlp.w2.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.2.attn.c_attn.bias": "model-00001-of-00006.safetensors",
+    "transformer.h.2.attn.c_attn.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.2.attn.c_proj.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.2.ln_1.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.2.ln_2.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.2.mlp.c_proj.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.2.mlp.w1.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.2.mlp.w2.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.20.attn.c_attn.bias": "model-00003-of-00006.safetensors",
+    "transformer.h.20.attn.c_attn.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.20.attn.c_proj.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.20.ln_1.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.20.ln_2.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.20.mlp.c_proj.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.20.mlp.w1.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.20.mlp.w2.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.21.attn.c_attn.bias": "model-00004-of-00006.safetensors",
+    "transformer.h.21.attn.c_attn.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.21.attn.c_proj.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.21.ln_1.weight": "model-00003-of-00006.safetensors",
+    "transformer.h.21.ln_2.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.21.mlp.c_proj.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.21.mlp.w1.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.21.mlp.w2.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.22.attn.c_attn.bias": "model-00004-of-00006.safetensors",
+    "transformer.h.22.attn.c_attn.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.22.attn.c_proj.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.22.ln_1.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.22.ln_2.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.22.mlp.c_proj.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.22.mlp.w1.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.22.mlp.w2.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.23.attn.c_attn.bias": "model-00004-of-00006.safetensors",
+    "transformer.h.23.attn.c_attn.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.23.attn.c_proj.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.23.ln_1.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.23.ln_2.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.23.mlp.c_proj.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.23.mlp.w1.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.23.mlp.w2.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.24.attn.c_attn.bias": "model-00004-of-00006.safetensors",
+    "transformer.h.24.attn.c_attn.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.24.attn.c_proj.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.24.ln_1.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.24.ln_2.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.24.mlp.c_proj.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.24.mlp.w1.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.24.mlp.w2.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.25.attn.c_attn.bias": "model-00004-of-00006.safetensors",
+    "transformer.h.25.attn.c_attn.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.25.attn.c_proj.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.25.ln_1.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.25.ln_2.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.25.mlp.c_proj.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.25.mlp.w1.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.25.mlp.w2.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.26.attn.c_attn.bias": "model-00004-of-00006.safetensors",
+    "transformer.h.26.attn.c_attn.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.26.attn.c_proj.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.26.ln_1.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.26.ln_2.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.26.mlp.c_proj.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.26.mlp.w1.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.26.mlp.w2.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.27.attn.c_attn.bias": "model-00004-of-00006.safetensors",
+    "transformer.h.27.attn.c_attn.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.27.attn.c_proj.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.27.ln_1.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.27.ln_2.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.27.mlp.c_proj.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.27.mlp.w1.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.27.mlp.w2.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.28.attn.c_attn.bias": "model-00004-of-00006.safetensors",
+    "transformer.h.28.attn.c_attn.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.28.attn.c_proj.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.28.ln_1.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.28.ln_2.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.28.mlp.c_proj.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.28.mlp.w1.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.28.mlp.w2.weight": "model-00004-of-00006.safetensors",
+    "transformer.h.29.attn.c_attn.bias": "model-00005-of-00006.safetensors",
+    "transformer.h.29.attn.c_attn.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.29.attn.c_proj.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.29.ln_1.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.29.ln_2.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.29.mlp.c_proj.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.29.mlp.w1.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.29.mlp.w2.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.3.attn.c_attn.bias": "model-00001-of-00006.safetensors",
+    "transformer.h.3.attn.c_attn.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.3.attn.c_proj.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.3.ln_1.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.3.ln_2.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.3.mlp.c_proj.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.3.mlp.w1.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.3.mlp.w2.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.30.attn.c_attn.bias": "model-00005-of-00006.safetensors",
+    "transformer.h.30.attn.c_attn.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.30.attn.c_proj.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.30.ln_1.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.30.ln_2.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.30.mlp.c_proj.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.30.mlp.w1.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.30.mlp.w2.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.31.attn.c_attn.bias": "model-00005-of-00006.safetensors",
+    "transformer.h.31.attn.c_attn.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.31.attn.c_proj.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.31.ln_1.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.31.ln_2.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.31.mlp.c_proj.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.31.mlp.w1.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.31.mlp.w2.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.32.attn.c_attn.bias": "model-00005-of-00006.safetensors",
+    "transformer.h.32.attn.c_attn.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.32.attn.c_proj.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.32.ln_1.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.32.ln_2.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.32.mlp.c_proj.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.32.mlp.w1.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.32.mlp.w2.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.33.attn.c_attn.bias": "model-00005-of-00006.safetensors",
+    "transformer.h.33.attn.c_attn.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.33.attn.c_proj.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.33.ln_1.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.33.ln_2.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.33.mlp.c_proj.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.33.mlp.w1.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.33.mlp.w2.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.34.attn.c_attn.bias": "model-00005-of-00006.safetensors",
+    "transformer.h.34.attn.c_attn.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.34.attn.c_proj.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.34.ln_1.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.34.ln_2.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.34.mlp.c_proj.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.34.mlp.w1.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.34.mlp.w2.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.35.attn.c_attn.bias": "model-00005-of-00006.safetensors",
+    "transformer.h.35.attn.c_attn.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.35.attn.c_proj.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.35.ln_1.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.35.ln_2.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.35.mlp.c_proj.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.35.mlp.w1.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.35.mlp.w2.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.36.attn.c_attn.bias": "model-00005-of-00006.safetensors",
+    "transformer.h.36.attn.c_attn.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.36.attn.c_proj.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.36.ln_1.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.36.ln_2.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.36.mlp.c_proj.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.36.mlp.w1.weight": "model-00005-of-00006.safetensors",
+    "transformer.h.36.mlp.w2.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.37.attn.c_attn.bias": "model-00006-of-00006.safetensors",
+    "transformer.h.37.attn.c_attn.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.37.attn.c_proj.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.37.ln_1.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.37.ln_2.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.37.mlp.c_proj.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.37.mlp.w1.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.37.mlp.w2.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.38.attn.c_attn.bias": "model-00006-of-00006.safetensors",
+    "transformer.h.38.attn.c_attn.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.38.attn.c_proj.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.38.ln_1.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.38.ln_2.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.38.mlp.c_proj.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.38.mlp.w1.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.38.mlp.w2.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.39.attn.c_attn.bias": "model-00006-of-00006.safetensors",
+    "transformer.h.39.attn.c_attn.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.39.attn.c_proj.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.39.ln_1.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.39.ln_2.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.39.mlp.c_proj.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.39.mlp.w1.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.39.mlp.w2.weight": "model-00006-of-00006.safetensors",
+    "transformer.h.4.attn.c_attn.bias": "model-00001-of-00006.safetensors",
+    "transformer.h.4.attn.c_attn.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.4.attn.c_proj.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.4.ln_1.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.4.ln_2.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.4.mlp.c_proj.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.4.mlp.w1.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.4.mlp.w2.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.5.attn.c_attn.bias": "model-00001-of-00006.safetensors",
+    "transformer.h.5.attn.c_attn.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.5.attn.c_proj.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.5.ln_1.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.5.ln_2.weight": "model-00001-of-00006.safetensors",
+    "transformer.h.5.mlp.c_proj.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.5.mlp.w1.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.5.mlp.w2.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.6.attn.c_attn.bias": "model-00002-of-00006.safetensors",
+    "transformer.h.6.attn.c_attn.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.6.attn.c_proj.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.6.ln_1.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.6.ln_2.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.6.mlp.c_proj.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.6.mlp.w1.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.6.mlp.w2.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.7.attn.c_attn.bias": "model-00002-of-00006.safetensors",
+    "transformer.h.7.attn.c_attn.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.7.attn.c_proj.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.7.ln_1.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.7.ln_2.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.7.mlp.c_proj.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.7.mlp.w1.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.7.mlp.w2.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.8.attn.c_attn.bias": "model-00002-of-00006.safetensors",
+    "transformer.h.8.attn.c_attn.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.8.attn.c_proj.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.8.ln_1.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.8.ln_2.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.8.mlp.c_proj.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.8.mlp.w1.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.8.mlp.w2.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.9.attn.c_attn.bias": "model-00002-of-00006.safetensors",
+    "transformer.h.9.attn.c_attn.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.9.attn.c_proj.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.9.ln_1.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.9.ln_2.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.9.mlp.c_proj.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.9.mlp.w1.weight": "model-00002-of-00006.safetensors",
+    "transformer.h.9.mlp.w2.weight": "model-00002-of-00006.safetensors",
+    "transformer.ln_f.weight": "model-00006-of-00006.safetensors",
+    "transformer.wte.weight": "model-00001-of-00006.safetensors"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,4 @@
+{
+  "eos_token": "<|endoftext|>",
+  "pad_token": "<|endoftext|>"
+}

tokenizer_config.json

@@ -0,0 +1,14 @@
+{
+  "added_tokens_decoder": {},
+  "auto_map": {
+    "AutoTokenizer": [
+      "tokenization_qwen.QWenTokenizer",
+      null
+    ]
+  },
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|endoftext|>",
+  "model_max_length": 8192,
+  "pad_token": "<|endoftext|>",
+  "tokenizer_class": "QWenTokenizer"
+}

trainer_state.json

@@ -0,0 +1,3979 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9997296404217609,
+  "eval_steps": 3120,
+  "global_step": 6240,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.953125e-07,
+      "loss": 2.7534,
+      "step": 10
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 5.859375000000001e-07,
+      "loss": 1.8637,
+      "step": 20
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.765625e-07,
+      "loss": 1.9021,
+      "step": 30
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3671875000000002e-06,
+      "loss": 1.7516,
+      "step": 40
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7578125000000002e-06,
+      "loss": 1.6276,
+      "step": 50
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1484375e-06,
+      "loss": 1.6556,
+      "step": 60
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5390625000000003e-06,
+      "loss": 1.6838,
+      "step": 70
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9296875e-06,
+      "loss": 1.6008,
+      "step": 80
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.3203125000000002e-06,
+      "loss": 1.5574,
+      "step": 90
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7109375000000004e-06,
+      "loss": 1.4339,
+      "step": 100
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.101562500000001e-06,
+      "loss": 1.4322,
+      "step": 110
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.492187500000001e-06,
+      "loss": 1.4284,
+      "step": 120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8828125e-06,
+      "loss": 1.494,
+      "step": 130
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.2734375e-06,
+      "loss": 1.3452,
+      "step": 140
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.6640625000000005e-06,
+      "loss": 1.3777,
+      "step": 150
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 6.054687500000001e-06,
+      "loss": 1.3711,
+      "step": 160
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 6.445312500000001e-06,
+      "loss": 1.2907,
+      "step": 170
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 6.8359375e-06,
+      "loss": 1.2957,
+      "step": 180
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 7.2265625e-06,
+      "loss": 1.3431,
+      "step": 190
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 7.6171875000000005e-06,
+      "loss": 1.2668,
+      "step": 200
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 8.0078125e-06,
+      "loss": 1.3359,
+      "step": 210
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 8.3984375e-06,
+      "loss": 1.3276,
+      "step": 220
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 8.789062500000001e-06,
+      "loss": 1.364,
+      "step": 230
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.1796875e-06,
+      "loss": 1.2738,
+      "step": 240
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.570312500000001e-06,
+      "loss": 1.3374,
+      "step": 250
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.9609375e-06,
+      "loss": 1.3623,
+      "step": 260
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1e-05,
+      "loss": 1.3141,
+      "step": 270
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1e-05,
+      "loss": 1.3332,
+      "step": 280
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1e-05,
+      "loss": 1.402,
+      "step": 290
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1e-05,
+      "loss": 1.1948,
+      "step": 300
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1e-05,
+      "loss": 1.4215,
+      "step": 310
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1e-05,
+      "loss": 1.4096,
+      "step": 320
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1e-05,
+      "loss": 1.3142,
+      "step": 330
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1e-05,
+      "loss": 1.3973,
+      "step": 340
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1e-05,
+      "loss": 1.3519,
+      "step": 350
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1e-05,
+      "loss": 1.3599,
+      "step": 360
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1e-05,
+      "loss": 1.2374,
+      "step": 370
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1e-05,
+      "loss": 1.4625,
+      "step": 380
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1e-05,
+      "loss": 1.3358,
+      "step": 390
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1e-05,
+      "loss": 1.3182,
+      "step": 400
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1e-05,
+      "loss": 1.3355,
+      "step": 410
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1e-05,
+      "loss": 1.3048,
+      "step": 420
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1e-05,
+      "loss": 1.4234,
+      "step": 430
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1e-05,
+      "loss": 1.3935,
+      "step": 440
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1e-05,
+      "loss": 1.3028,
+      "step": 450
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1e-05,
+      "loss": 1.4297,
+      "step": 460
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1e-05,
+      "loss": 1.3128,
+      "step": 470
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1e-05,
+      "loss": 1.3076,
+      "step": 480
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1e-05,
+      "loss": 1.3364,
+      "step": 490
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1e-05,
+      "loss": 1.2971,
+      "step": 500
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1e-05,
+      "loss": 1.3807,
+      "step": 510
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1e-05,
+      "loss": 1.3804,
+      "step": 520
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1e-05,
+      "loss": 1.3974,
+      "step": 530
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1e-05,
+      "loss": 1.3752,
+      "step": 540
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1e-05,
+      "loss": 1.3389,
+      "step": 550
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1e-05,
+      "loss": 1.3491,
+      "step": 560
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1e-05,
+      "loss": 1.3376,
+      "step": 570
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1e-05,
+      "loss": 1.2993,
+      "step": 580
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1e-05,
+      "loss": 1.4303,
+      "step": 590
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1e-05,
+      "loss": 1.37,
+      "step": 600
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1e-05,
+      "loss": 1.2767,
+      "step": 610
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1e-05,
+      "loss": 1.2692,
+      "step": 620
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1e-05,
+      "loss": 1.3421,
+      "step": 630
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1e-05,
+      "loss": 1.3308,
+      "step": 640
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1e-05,
+      "loss": 1.3539,
+      "step": 650
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1e-05,
+      "loss": 1.2961,
+      "step": 660
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1e-05,
+      "loss": 1.3026,
+      "step": 670
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1e-05,
+      "loss": 1.3851,
+      "step": 680
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1e-05,
+      "loss": 1.3756,
+      "step": 690
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1e-05,
+      "loss": 1.3534,
+      "step": 700
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1e-05,
+      "loss": 1.3732,
+      "step": 710
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1e-05,
+      "loss": 1.416,
+      "step": 720
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1e-05,
+      "loss": 1.343,
+      "step": 730
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1e-05,
+      "loss": 1.3733,
+      "step": 740
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1e-05,
+      "loss": 1.2798,
+      "step": 750
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1e-05,
+      "loss": 1.339,
+      "step": 760
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1e-05,
+      "loss": 1.2802,
+      "step": 770
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1e-05,
+      "loss": 1.3436,
+      "step": 780
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1e-05,
+      "loss": 1.4034,
+      "step": 790
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1e-05,
+      "loss": 1.3548,
+      "step": 800
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1e-05,
+      "loss": 1.3972,
+      "step": 810
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1e-05,
+      "loss": 1.3321,
+      "step": 820
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1e-05,
+      "loss": 1.2946,
+      "step": 830
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1e-05,
+      "loss": 1.318,
+      "step": 840
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1e-05,
+      "loss": 1.2935,
+      "step": 850
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1e-05,
+      "loss": 1.4297,
+      "step": 860
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1e-05,
+      "loss": 1.3144,
+      "step": 870
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1e-05,
+      "loss": 1.2091,
+      "step": 880
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1e-05,
+      "loss": 1.2659,
+      "step": 890
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1e-05,
+      "loss": 1.4349,
+      "step": 900
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1e-05,
+      "loss": 1.3532,
+      "step": 910
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1e-05,
+      "loss": 1.2869,
+      "step": 920
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1e-05,
+      "loss": 1.3122,
+      "step": 930
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1e-05,
+      "loss": 1.3867,
+      "step": 940
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1e-05,
+      "loss": 1.3813,
+      "step": 950
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1e-05,
+      "loss": 1.2613,
+      "step": 960
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1e-05,
+      "loss": 1.2536,
+      "step": 970
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1e-05,
+      "loss": 1.2798,
+      "step": 980
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1e-05,
+      "loss": 1.3477,
+      "step": 990
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1e-05,
+      "loss": 1.3286,
+      "step": 1000
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1e-05,
+      "loss": 1.3503,
+      "step": 1010
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1e-05,
+      "loss": 1.4265,
+      "step": 1020
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1e-05,
+      "loss": 1.3439,
+      "step": 1030
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1e-05,
+      "loss": 1.3606,
+      "step": 1040
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1e-05,
+      "loss": 1.274,
+      "step": 1050
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1e-05,
+      "loss": 1.3384,
+      "step": 1060
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1e-05,
+      "loss": 1.4208,
+      "step": 1070
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1e-05,
+      "loss": 1.233,
+      "step": 1080
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1e-05,
+      "loss": 1.434,
+      "step": 1090
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1e-05,
+      "loss": 1.3791,
+      "step": 1100
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1e-05,
+      "loss": 1.3318,
+      "step": 1110
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1e-05,
+      "loss": 1.3855,
+      "step": 1120
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1e-05,
+      "loss": 1.3216,
+      "step": 1130
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1e-05,
+      "loss": 1.3341,
+      "step": 1140
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1e-05,
+      "loss": 1.3289,
+      "step": 1150
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1e-05,
+      "loss": 1.3535,
+      "step": 1160
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1e-05,
+      "loss": 1.4895,
+      "step": 1170
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1e-05,
+      "loss": 1.4152,
+      "step": 1180
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1e-05,
+      "loss": 1.2835,
+      "step": 1190
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1e-05,
+      "loss": 1.3839,
+      "step": 1200
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1e-05,
+      "loss": 1.3066,
+      "step": 1210
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1e-05,
+      "loss": 1.2582,
+      "step": 1220
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1e-05,
+      "loss": 1.2839,
+      "step": 1230
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1e-05,
+      "loss": 1.2858,
+      "step": 1240
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1e-05,
+      "loss": 1.3237,
+      "step": 1250
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1e-05,
+      "loss": 1.3306,
+      "step": 1260
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1e-05,
+      "loss": 1.3599,
+      "step": 1270
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1e-05,
+      "loss": 1.3803,
+      "step": 1280
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1e-05,
+      "loss": 1.3499,
+      "step": 1290
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1e-05,
+      "loss": 1.2982,
+      "step": 1300
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1e-05,
+      "loss": 1.3694,
+      "step": 1310
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1e-05,
+      "loss": 1.3994,
+      "step": 1320
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1e-05,
+      "loss": 1.3638,
+      "step": 1330
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1e-05,
+      "loss": 1.348,
+      "step": 1340
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1e-05,
+      "loss": 1.325,
+      "step": 1350
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1e-05,
+      "loss": 1.4602,
+      "step": 1360
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1e-05,
+      "loss": 1.2748,
+      "step": 1370
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1e-05,
+      "loss": 1.3623,
+      "step": 1380
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1e-05,
+      "loss": 1.2561,
+      "step": 1390
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1e-05,
+      "loss": 1.3941,
+      "step": 1400
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1e-05,
+      "loss": 1.3173,
+      "step": 1410
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1e-05,
+      "loss": 1.3102,
+      "step": 1420
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1e-05,
+      "loss": 1.2749,
+      "step": 1430
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1e-05,
+      "loss": 1.3771,
+      "step": 1440
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1e-05,
+      "loss": 1.3097,
+      "step": 1450
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1e-05,
+      "loss": 1.2732,
+      "step": 1460
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1e-05,
+      "loss": 1.3611,
+      "step": 1470
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1e-05,
+      "loss": 1.3637,
+      "step": 1480
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1e-05,
+      "loss": 1.2719,
+      "step": 1490
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1e-05,
+      "loss": 1.3075,
+      "step": 1500
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1e-05,
+      "loss": 1.3522,
+      "step": 1510
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1e-05,
+      "loss": 1.3614,
+      "step": 1520
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1e-05,
+      "loss": 1.2614,
+      "step": 1530
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1e-05,
+      "loss": 1.3766,
+      "step": 1540
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1e-05,
+      "loss": 1.4132,
+      "step": 1550
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1e-05,
+      "loss": 1.324,
+      "step": 1560
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1e-05,
+      "loss": 1.3579,
+      "step": 1570
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1e-05,
+      "loss": 1.2906,
+      "step": 1580
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1e-05,
+      "loss": 1.3578,
+      "step": 1590
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1e-05,
+      "loss": 1.3727,
+      "step": 1600
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1e-05,
+      "loss": 1.363,
+      "step": 1610
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1e-05,
+      "loss": 1.3323,
+      "step": 1620
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1e-05,
+      "loss": 1.3705,
+      "step": 1630
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1e-05,
+      "loss": 1.3211,
+      "step": 1640
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1e-05,
+      "loss": 1.2708,
+      "step": 1650
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1e-05,
+      "loss": 1.2953,
+      "step": 1660
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1e-05,
+      "loss": 1.3302,
+      "step": 1670
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1e-05,
+      "loss": 1.373,
+      "step": 1680
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1e-05,
+      "loss": 1.3791,
+      "step": 1690
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1e-05,
+      "loss": 1.2198,
+      "step": 1700
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1e-05,
+      "loss": 1.4144,
+      "step": 1710
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1e-05,
+      "loss": 1.3198,
+      "step": 1720
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1e-05,
+      "loss": 1.2686,
+      "step": 1730
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1e-05,
+      "loss": 1.3298,
+      "step": 1740
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1e-05,
+      "loss": 1.3054,
+      "step": 1750
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1e-05,
+      "loss": 1.3626,
+      "step": 1760
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1e-05,
+      "loss": 1.3624,
+      "step": 1770
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1e-05,
+      "loss": 1.2992,
+      "step": 1780
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1e-05,
+      "loss": 1.3284,
+      "step": 1790
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1e-05,
+      "loss": 1.3361,
+      "step": 1800
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1e-05,
+      "loss": 1.2872,
+      "step": 1810
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1e-05,
+      "loss": 1.3051,
+      "step": 1820
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1e-05,
+      "loss": 1.258,
+      "step": 1830
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1e-05,
+      "loss": 1.4084,
+      "step": 1840
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1e-05,
+      "loss": 1.3256,
+      "step": 1850
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1e-05,
+      "loss": 1.2824,
+      "step": 1860
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1e-05,
+      "loss": 1.4216,
+      "step": 1870
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1e-05,
+      "loss": 1.393,
+      "step": 1880
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1e-05,
+      "loss": 1.3535,
+      "step": 1890
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1e-05,
+      "loss": 1.369,
+      "step": 1900
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1e-05,
+      "loss": 1.1903,
+      "step": 1910
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1e-05,
+      "loss": 1.3719,
+      "step": 1920
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1e-05,
+      "loss": 1.3007,
+      "step": 1930
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1e-05,
+      "loss": 1.3146,
+      "step": 1940
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1e-05,
+      "loss": 1.2968,
+      "step": 1950
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1e-05,
+      "loss": 1.3905,
+      "step": 1960
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1e-05,
+      "loss": 1.3402,
+      "step": 1970
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1e-05,
+      "loss": 1.3421,
+      "step": 1980
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1e-05,
+      "loss": 1.2939,
+      "step": 1990
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1e-05,
+      "loss": 1.4034,
+      "step": 2000
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1e-05,
+      "loss": 1.2797,
+      "step": 2010
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1e-05,
+      "loss": 1.3532,
+      "step": 2020
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1e-05,
+      "loss": 1.3656,
+      "step": 2030
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1e-05,
+      "loss": 1.3,
+      "step": 2040
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1e-05,
+      "loss": 1.4061,
+      "step": 2050
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1e-05,
+      "loss": 1.2581,
+      "step": 2060
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1e-05,
+      "loss": 1.2881,
+      "step": 2070
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1e-05,
+      "loss": 1.2648,
+      "step": 2080
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1e-05,
+      "loss": 1.3563,
+      "step": 2090
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1e-05,
+      "loss": 1.3747,
+      "step": 2100
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1e-05,
+      "loss": 1.291,
+      "step": 2110
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1e-05,
+      "loss": 1.3655,
+      "step": 2120
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1e-05,
+      "loss": 1.3245,
+      "step": 2130
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1e-05,
+      "loss": 1.3134,
+      "step": 2140
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1e-05,
+      "loss": 1.2537,
+      "step": 2150
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1e-05,
+      "loss": 1.2823,
+      "step": 2160
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1e-05,
+      "loss": 1.2754,
+      "step": 2170
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1e-05,
+      "loss": 1.3753,
+      "step": 2180
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1e-05,
+      "loss": 1.3884,
+      "step": 2190
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1e-05,
+      "loss": 1.3139,
+      "step": 2200
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1e-05,
+      "loss": 1.2683,
+      "step": 2210
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1e-05,
+      "loss": 1.2946,
+      "step": 2220
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1e-05,
+      "loss": 1.2891,
+      "step": 2230
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1e-05,
+      "loss": 1.3521,
+      "step": 2240
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1e-05,
+      "loss": 1.3707,
+      "step": 2250
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1e-05,
+      "loss": 1.3231,
+      "step": 2260
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1e-05,
+      "loss": 1.1757,
+      "step": 2270
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1e-05,
+      "loss": 1.3597,
+      "step": 2280
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1e-05,
+      "loss": 1.3267,
+      "step": 2290
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1e-05,
+      "loss": 1.2564,
+      "step": 2300
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1e-05,
+      "loss": 1.3465,
+      "step": 2310
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1e-05,
+      "loss": 1.2958,
+      "step": 2320
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1e-05,
+      "loss": 1.3529,
+      "step": 2330
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1e-05,
+      "loss": 1.4335,
+      "step": 2340
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1e-05,
+      "loss": 1.3121,
+      "step": 2350
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1e-05,
+      "loss": 1.3199,
+      "step": 2360
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1e-05,
+      "loss": 1.3018,
+      "step": 2370
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1e-05,
+      "loss": 1.2688,
+      "step": 2380
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1e-05,
+      "loss": 1.2845,
+      "step": 2390
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1e-05,
+      "loss": 1.365,
+      "step": 2400
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1e-05,
+      "loss": 1.3712,
+      "step": 2410
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1e-05,
+      "loss": 1.3573,
+      "step": 2420
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1e-05,
+      "loss": 1.3368,
+      "step": 2430
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1e-05,
+      "loss": 1.2725,
+      "step": 2440
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1e-05,
+      "loss": 1.2307,
+      "step": 2450
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1e-05,
+      "loss": 1.3313,
+      "step": 2460
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1e-05,
+      "loss": 1.2459,
+      "step": 2470
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1e-05,
+      "loss": 1.2745,
+      "step": 2480
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1e-05,
+      "loss": 1.2797,
+      "step": 2490
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1e-05,
+      "loss": 1.2905,
+      "step": 2500
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1e-05,
+      "loss": 1.3558,
+      "step": 2510
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1e-05,
+      "loss": 1.3634,
+      "step": 2520
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1e-05,
+      "loss": 1.3673,
+      "step": 2530
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1e-05,
+      "loss": 1.3956,
+      "step": 2540
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1e-05,
+      "loss": 1.3099,
+      "step": 2550
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1e-05,
+      "loss": 1.3629,
+      "step": 2560
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1e-05,
+      "loss": 1.3324,
+      "step": 2570
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1e-05,
+      "loss": 1.2867,
+      "step": 2580
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1e-05,
+      "loss": 1.2411,
+      "step": 2590
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1e-05,
+      "loss": 1.3078,
+      "step": 2600
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1e-05,
+      "loss": 1.3422,
+      "step": 2610
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1e-05,
+      "loss": 1.3304,
+      "step": 2620
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1e-05,
+      "loss": 1.3108,
+      "step": 2630
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1e-05,
+      "loss": 1.2523,
+      "step": 2640
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1e-05,
+      "loss": 1.3082,
+      "step": 2650
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1e-05,
+      "loss": 1.3143,
+      "step": 2660
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1e-05,
+      "loss": 1.289,
+      "step": 2670
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1e-05,
+      "loss": 1.4273,
+      "step": 2680
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1e-05,
+      "loss": 1.3225,
+      "step": 2690
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1e-05,
+      "loss": 1.3121,
+      "step": 2700
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1e-05,
+      "loss": 1.382,
+      "step": 2710
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1e-05,
+      "loss": 1.2803,
+      "step": 2720
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1e-05,
+      "loss": 1.3615,
+      "step": 2730
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1e-05,
+      "loss": 1.2941,
+      "step": 2740
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1e-05,
+      "loss": 1.3935,
+      "step": 2750
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1e-05,
+      "loss": 1.2838,
+      "step": 2760
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1e-05,
+      "loss": 1.2333,
+      "step": 2770
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1e-05,
+      "loss": 1.3408,
+      "step": 2780
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1e-05,
+      "loss": 1.3302,
+      "step": 2790
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1e-05,
+      "loss": 1.3572,
+      "step": 2800
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1e-05,
+      "loss": 1.299,
+      "step": 2810
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1e-05,
+      "loss": 1.2898,
+      "step": 2820
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1e-05,
+      "loss": 1.2634,
+      "step": 2830
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1e-05,
+      "loss": 1.2869,
+      "step": 2840
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1e-05,
+      "loss": 1.3214,
+      "step": 2850
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1e-05,
+      "loss": 1.3324,
+      "step": 2860
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1e-05,
+      "loss": 1.271,
+      "step": 2870
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1e-05,
+      "loss": 1.3104,
+      "step": 2880
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1e-05,
+      "loss": 1.3267,
+      "step": 2890
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1e-05,
+      "loss": 1.2713,
+      "step": 2900
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1e-05,
+      "loss": 1.2806,
+      "step": 2910
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1e-05,
+      "loss": 1.3255,
+      "step": 2920
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1e-05,
+      "loss": 1.2804,
+      "step": 2930
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1e-05,
+      "loss": 1.3617,
+      "step": 2940
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1e-05,
+      "loss": 1.3665,
+      "step": 2950
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1e-05,
+      "loss": 1.3241,
+      "step": 2960
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1e-05,
+      "loss": 1.2946,
+      "step": 2970
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1e-05,
+      "loss": 1.3756,
+      "step": 2980
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1e-05,
+      "loss": 1.3199,
+      "step": 2990
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1e-05,
+      "loss": 1.3389,
+      "step": 3000
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1e-05,
+      "loss": 1.3891,
+      "step": 3010
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1e-05,
+      "loss": 1.3202,
+      "step": 3020
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1e-05,
+      "loss": 1.313,
+      "step": 3030
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1e-05,
+      "loss": 1.3361,
+      "step": 3040
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1e-05,
+      "loss": 1.3605,
+      "step": 3050
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1e-05,
+      "loss": 1.2426,
+      "step": 3060
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1e-05,
+      "loss": 1.226,
+      "step": 3070
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1e-05,
+      "loss": 1.2828,
+      "step": 3080
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1e-05,
+      "loss": 1.3532,
+      "step": 3090
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1e-05,
+      "loss": 1.3328,
+      "step": 3100
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1e-05,
+      "loss": 1.3588,
+      "step": 3110
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1e-05,
+      "loss": 1.2656,
+      "step": 3120
+    },
+    {
+      "epoch": 0.5,
+      "eval_oasst_export_accuracy": 0.6519529382418359,
+      "eval_oasst_export_loss": 1.509765625,
+      "eval_oasst_export_runtime": 10.8282,
+      "eval_oasst_export_samples_per_second": 27.705,
+      "eval_oasst_export_steps_per_second": 3.509,
+      "step": 3120
+    },
+    {
+      "epoch": 0.5,
+      "eval_dolly15k_accuracy": 0.6600884046970577,
+      "eval_dolly15k_loss": 1.2177734375,
+      "eval_dolly15k_runtime": 10.3078,
+      "eval_dolly15k_samples_per_second": 29.104,
+      "eval_dolly15k_steps_per_second": 3.687,
+      "step": 3120
+    },
+    {
+      "epoch": 0.5,
+      "eval__ruozhiba_accuracy": 0.4932349323493235,
+      "eval__ruozhiba_loss": 1.822265625,
+      "eval__ruozhiba_runtime": 1.4729,
+      "eval__ruozhiba_samples_per_second": 4.752,
+      "eval__ruozhiba_steps_per_second": 0.679,
+      "step": 3120
+    },
+    {
+      "epoch": 0.5,
+      "eval__sharegpt_format_accuracy": 0.8631696577139595,
+      "eval__sharegpt_format_loss": 0.83935546875,
+      "eval__sharegpt_format_runtime": 27.558,
+      "eval__sharegpt_format_samples_per_second": 18.144,
+      "eval__sharegpt_format_steps_per_second": 2.286,
+      "step": 3120
+    },
+    {
+      "epoch": 0.5,
+      "eval__similar_question_accuracy": 0.5443645083932853,
+      "eval__similar_question_loss": 1.865234375,
+      "eval__similar_question_runtime": 0.7582,
+      "eval__similar_question_samples_per_second": 9.232,
+      "eval__similar_question_steps_per_second": 1.319,
+      "step": 3120
+    },
+    {
+      "epoch": 0.5,
+      "eval__open_domain_subject_accuracy": 0.588056546439424,
+      "eval__open_domain_subject_loss": 1.578125,
+      "eval__open_domain_subject_runtime": 1.795,
+      "eval__open_domain_subject_samples_per_second": 25.627,
+      "eval__open_domain_subject_steps_per_second": 3.343,
+      "step": 3120
+    },
+    {
+      "epoch": 0.5,
+      "eval__empdia_ly_07_17_accuracy": 0.6026972611933143,
+      "eval__empdia_ly_07_17_loss": 1.453125,
+      "eval__empdia_ly_07_17_runtime": 3.6438,
+      "eval__empdia_ly_07_17_samples_per_second": 25.249,
+      "eval__empdia_ly_07_17_steps_per_second": 3.293,
+      "step": 3120
+    },
+    {
+      "epoch": 0.5,
+      "eval__zuowen_accuracy": 0.8976080316913121,
+      "eval__zuowen_loss": 0.3671875,
+      "eval__zuowen_runtime": 4.879,
+      "eval__zuowen_samples_per_second": 20.496,
+      "eval__zuowen_steps_per_second": 2.664,
+      "step": 3120
+    },
+    {
+      "epoch": 0.5,
+      "eval__psychat_6_29_and_11_15_mix_accuracy": 0.4837894477317554,
+      "eval__psychat_6_29_and_11_15_mix_loss": 2.08984375,
+      "eval__psychat_6_29_and_11_15_mix_runtime": 5.5362,
+      "eval__psychat_6_29_and_11_15_mix_samples_per_second": 18.063,
+      "eval__psychat_6_29_and_11_15_mix_steps_per_second": 2.348,
+      "step": 3120
+    },
+    {
+      "epoch": 0.5,
+      "eval__socrates_teaching_accuracy": 0.6189571458729154,
+      "eval__socrates_teaching_loss": 1.517578125,
+      "eval__socrates_teaching_runtime": 3.3517,
+      "eval__socrates_teaching_samples_per_second": 25.659,
+      "eval__socrates_teaching_steps_per_second": 3.282,
+      "step": 3120
+    },
+    {
+      "epoch": 0.5,
+      "eval__socrates_teaching_math1115_accuracy": 0.6499228356648579,
+      "eval__socrates_teaching_math1115_loss": 1.380859375,
+      "eval__socrates_teaching_math1115_runtime": 3.8085,
+      "eval__socrates_teaching_math1115_samples_per_second": 26.257,
+      "eval__socrates_teaching_math1115_steps_per_second": 3.413,
+      "step": 3120
+    },
+    {
+      "epoch": 0.5,
+      "eval__search_accuracy": 0.6731122393864103,
+      "eval__search_loss": 1.248046875,
+      "eval__search_runtime": 4.7332,
+      "eval__search_samples_per_second": 21.127,
+      "eval__search_steps_per_second": 2.747,
+      "step": 3120
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1e-05,
+      "loss": 1.4882,
+      "step": 3130
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1e-05,
+      "loss": 1.315,
+      "step": 3140
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1e-05,
+      "loss": 1.298,
+      "step": 3150
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1e-05,
+      "loss": 1.2754,
+      "step": 3160
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1e-05,
+      "loss": 1.2802,
+      "step": 3170
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1e-05,
+      "loss": 1.3694,
+      "step": 3180
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1e-05,
+      "loss": 1.322,
+      "step": 3190
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1e-05,
+      "loss": 1.3168,
+      "step": 3200
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1e-05,
+      "loss": 1.3408,
+      "step": 3210
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1e-05,
+      "loss": 1.3508,
+      "step": 3220
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1e-05,
+      "loss": 1.3369,
+      "step": 3230
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1e-05,
+      "loss": 1.3589,
+      "step": 3240
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1e-05,
+      "loss": 1.295,
+      "step": 3250
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1e-05,
+      "loss": 1.3599,
+      "step": 3260
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1e-05,
+      "loss": 1.3575,
+      "step": 3270
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1e-05,
+      "loss": 1.3273,
+      "step": 3280
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1e-05,
+      "loss": 1.3955,
+      "step": 3290
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1e-05,
+      "loss": 1.3382,
+      "step": 3300
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1e-05,
+      "loss": 1.3612,
+      "step": 3310
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1e-05,
+      "loss": 1.273,
+      "step": 3320
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1e-05,
+      "loss": 1.2952,
+      "step": 3330
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1e-05,
+      "loss": 1.3171,
+      "step": 3340
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1e-05,
+      "loss": 1.3354,
+      "step": 3350
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1e-05,
+      "loss": 1.3262,
+      "step": 3360
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1e-05,
+      "loss": 1.2947,
+      "step": 3370
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1e-05,
+      "loss": 1.3744,
+      "step": 3380
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1e-05,
+      "loss": 1.3622,
+      "step": 3390
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1e-05,
+      "loss": 1.3435,
+      "step": 3400
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1e-05,
+      "loss": 1.3495,
+      "step": 3410
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1e-05,
+      "loss": 1.3364,
+      "step": 3420
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1e-05,
+      "loss": 1.2357,
+      "step": 3430
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1e-05,
+      "loss": 1.3601,
+      "step": 3440
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1e-05,
+      "loss": 1.2077,
+      "step": 3450
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1e-05,
+      "loss": 1.3991,
+      "step": 3460
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1e-05,
+      "loss": 1.2993,
+      "step": 3470
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1e-05,
+      "loss": 1.3304,
+      "step": 3480
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1e-05,
+      "loss": 1.3736,
+      "step": 3490
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1e-05,
+      "loss": 1.3294,
+      "step": 3500
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1e-05,
+      "loss": 1.3564,
+      "step": 3510
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1e-05,
+      "loss": 1.4073,
+      "step": 3520
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1e-05,
+      "loss": 1.2244,
+      "step": 3530
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1e-05,
+      "loss": 1.3093,
+      "step": 3540
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1e-05,
+      "loss": 1.2837,
+      "step": 3550
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1e-05,
+      "loss": 1.3459,
+      "step": 3560
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1e-05,
+      "loss": 1.2766,
+      "step": 3570
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1e-05,
+      "loss": 1.3402,
+      "step": 3580
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1e-05,
+      "loss": 1.2708,
+      "step": 3590
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1e-05,
+      "loss": 1.395,
+      "step": 3600
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1e-05,
+      "loss": 1.3093,
+      "step": 3610
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1e-05,
+      "loss": 1.3306,
+      "step": 3620
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1e-05,
+      "loss": 1.3122,
+      "step": 3630
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1e-05,
+      "loss": 1.2672,
+      "step": 3640
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1e-05,
+      "loss": 1.2791,
+      "step": 3650
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1e-05,
+      "loss": 1.2945,
+      "step": 3660
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1e-05,
+      "loss": 1.3702,
+      "step": 3670
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1e-05,
+      "loss": 1.2232,
+      "step": 3680
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1e-05,
+      "loss": 1.2489,
+      "step": 3690
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1e-05,
+      "loss": 1.3055,
+      "step": 3700
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1e-05,
+      "loss": 1.3682,
+      "step": 3710
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1e-05,
+      "loss": 1.354,
+      "step": 3720
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1e-05,
+      "loss": 1.2918,
+      "step": 3730
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1e-05,
+      "loss": 1.3138,
+      "step": 3740
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1e-05,
+      "loss": 1.2282,
+      "step": 3750
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1e-05,
+      "loss": 1.3364,
+      "step": 3760
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1e-05,
+      "loss": 1.3029,
+      "step": 3770
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1e-05,
+      "loss": 1.3159,
+      "step": 3780
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1e-05,
+      "loss": 1.3643,
+      "step": 3790
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1e-05,
+      "loss": 1.3279,
+      "step": 3800
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1e-05,
+      "loss": 1.3223,
+      "step": 3810
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1e-05,
+      "loss": 1.3564,
+      "step": 3820
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1e-05,
+      "loss": 1.2999,
+      "step": 3830
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1e-05,
+      "loss": 1.3029,
+      "step": 3840
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1e-05,
+      "loss": 1.4002,
+      "step": 3850
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1e-05,
+      "loss": 1.3306,
+      "step": 3860
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1e-05,
+      "loss": 1.3571,
+      "step": 3870
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1e-05,
+      "loss": 1.3563,
+      "step": 3880
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1e-05,
+      "loss": 1.3275,
+      "step": 3890
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1e-05,
+      "loss": 1.3562,
+      "step": 3900
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1e-05,
+      "loss": 1.2843,
+      "step": 3910
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1e-05,
+      "loss": 1.3117,
+      "step": 3920
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1e-05,
+      "loss": 1.3778,
+      "step": 3930
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1e-05,
+      "loss": 1.275,
+      "step": 3940
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1e-05,
+      "loss": 1.3636,
+      "step": 3950
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1e-05,
+      "loss": 1.2871,
+      "step": 3960
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1e-05,
+      "loss": 1.4067,
+      "step": 3970
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1e-05,
+      "loss": 1.3772,
+      "step": 3980
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1e-05,
+      "loss": 1.3758,
+      "step": 3990
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1e-05,
+      "loss": 1.4059,
+      "step": 4000
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1e-05,
+      "loss": 1.2997,
+      "step": 4010
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1e-05,
+      "loss": 1.2109,
+      "step": 4020
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1e-05,
+      "loss": 1.2904,
+      "step": 4030
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1e-05,
+      "loss": 1.3084,
+      "step": 4040
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1e-05,
+      "loss": 1.2771,
+      "step": 4050
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1e-05,
+      "loss": 1.3101,
+      "step": 4060
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1e-05,
+      "loss": 1.3255,
+      "step": 4070
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1e-05,
+      "loss": 1.2772,
+      "step": 4080
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1e-05,
+      "loss": 1.243,
+      "step": 4090
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1e-05,
+      "loss": 1.3625,
+      "step": 4100
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1e-05,
+      "loss": 1.3195,
+      "step": 4110
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1e-05,
+      "loss": 1.363,
+      "step": 4120
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1e-05,
+      "loss": 1.312,
+      "step": 4130
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1e-05,
+      "loss": 1.2725,
+      "step": 4140
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1e-05,
+      "loss": 1.2446,
+      "step": 4150
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1e-05,
+      "loss": 1.3995,
+      "step": 4160
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1e-05,
+      "loss": 1.3403,
+      "step": 4170
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1e-05,
+      "loss": 1.3188,
+      "step": 4180
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1e-05,
+      "loss": 1.2815,
+      "step": 4190
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1e-05,
+      "loss": 1.341,
+      "step": 4200
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1e-05,
+      "loss": 1.3022,
+      "step": 4210
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1e-05,
+      "loss": 1.3172,
+      "step": 4220
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1e-05,
+      "loss": 1.2955,
+      "step": 4230
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1e-05,
+      "loss": 1.3163,
+      "step": 4240
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1e-05,
+      "loss": 1.1067,
+      "step": 4250
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1e-05,
+      "loss": 1.1948,
+      "step": 4260
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1e-05,
+      "loss": 1.281,
+      "step": 4270
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1e-05,
+      "loss": 1.2997,
+      "step": 4280
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1e-05,
+      "loss": 1.3314,
+      "step": 4290
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1e-05,
+      "loss": 1.2844,
+      "step": 4300
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1e-05,
+      "loss": 1.3547,
+      "step": 4310
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1e-05,
+      "loss": 1.3876,
+      "step": 4320
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1e-05,
+      "loss": 1.2988,
+      "step": 4330
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1e-05,
+      "loss": 1.3251,
+      "step": 4340
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1e-05,
+      "loss": 1.4043,
+      "step": 4350
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1e-05,
+      "loss": 1.386,
+      "step": 4360
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1e-05,
+      "loss": 1.4246,
+      "step": 4370
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1e-05,
+      "loss": 1.3648,
+      "step": 4380
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1e-05,
+      "loss": 1.3301,
+      "step": 4390
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1e-05,
+      "loss": 1.2854,
+      "step": 4400
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1e-05,
+      "loss": 1.3074,
+      "step": 4410
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1e-05,
+      "loss": 1.3313,
+      "step": 4420
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1e-05,
+      "loss": 1.2956,
+      "step": 4430
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1e-05,
+      "loss": 1.2657,
+      "step": 4440
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1e-05,
+      "loss": 1.2583,
+      "step": 4450
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1e-05,
+      "loss": 1.2468,
+      "step": 4460
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1e-05,
+      "loss": 1.3461,
+      "step": 4470
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1e-05,
+      "loss": 1.3944,
+      "step": 4480
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1e-05,
+      "loss": 1.305,
+      "step": 4490
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1e-05,
+      "loss": 1.3913,
+      "step": 4500
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1e-05,
+      "loss": 1.3478,
+      "step": 4510
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1e-05,
+      "loss": 1.372,
+      "step": 4520
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1e-05,
+      "loss": 1.3084,
+      "step": 4530
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1e-05,
+      "loss": 1.2623,
+      "step": 4540
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1e-05,
+      "loss": 1.3701,
+      "step": 4550
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1e-05,
+      "loss": 1.3582,
+      "step": 4560
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1e-05,
+      "loss": 1.3493,
+      "step": 4570
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1e-05,
+      "loss": 1.3395,
+      "step": 4580
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1e-05,
+      "loss": 1.3172,
+      "step": 4590
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1e-05,
+      "loss": 1.305,
+      "step": 4600
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1e-05,
+      "loss": 1.3468,
+      "step": 4610
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1e-05,
+      "loss": 1.3385,
+      "step": 4620
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1e-05,
+      "loss": 1.3171,
+      "step": 4630
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1e-05,
+      "loss": 1.2915,
+      "step": 4640
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1e-05,
+      "loss": 1.2722,
+      "step": 4650
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1e-05,
+      "loss": 1.2707,
+      "step": 4660
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1e-05,
+      "loss": 1.3224,
+      "step": 4670
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1e-05,
+      "loss": 1.3347,
+      "step": 4680
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1e-05,
+      "loss": 1.2209,
+      "step": 4690
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1e-05,
+      "loss": 1.408,
+      "step": 4700
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1e-05,
+      "loss": 1.4326,
+      "step": 4710
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1e-05,
+      "loss": 1.3184,
+      "step": 4720
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1e-05,
+      "loss": 1.3201,
+      "step": 4730
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1e-05,
+      "loss": 1.3722,
+      "step": 4740
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1e-05,
+      "loss": 1.3726,
+      "step": 4750
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1e-05,
+      "loss": 1.3054,
+      "step": 4760
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1e-05,
+      "loss": 1.2453,
+      "step": 4770
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1e-05,
+      "loss": 1.4105,
+      "step": 4780
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1e-05,
+      "loss": 1.2654,
+      "step": 4790
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1e-05,
+      "loss": 1.2995,
+      "step": 4800
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1e-05,
+      "loss": 1.3178,
+      "step": 4810
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1e-05,
+      "loss": 1.2836,
+      "step": 4820
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1e-05,
+      "loss": 1.2861,
+      "step": 4830
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1e-05,
+      "loss": 1.2579,
+      "step": 4840
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1e-05,
+      "loss": 1.346,
+      "step": 4850
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1e-05,
+      "loss": 1.3142,
+      "step": 4860
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1e-05,
+      "loss": 1.3693,
+      "step": 4870
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1e-05,
+      "loss": 1.353,
+      "step": 4880
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1e-05,
+      "loss": 1.3163,
+      "step": 4890
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1e-05,
+      "loss": 1.2843,
+      "step": 4900
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1e-05,
+      "loss": 1.3381,
+      "step": 4910
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1e-05,
+      "loss": 1.3815,
+      "step": 4920
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1e-05,
+      "loss": 1.3667,
+      "step": 4930
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1e-05,
+      "loss": 1.3335,
+      "step": 4940
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1e-05,
+      "loss": 1.397,
+      "step": 4950
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1e-05,
+      "loss": 1.3446,
+      "step": 4960
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1e-05,
+      "loss": 1.3039,
+      "step": 4970
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1e-05,
+      "loss": 1.3248,
+      "step": 4980
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1e-05,
+      "loss": 1.2654,
+      "step": 4990
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1e-05,
+      "loss": 1.2685,
+      "step": 5000
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1e-05,
+      "loss": 1.3487,
+      "step": 5010
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1e-05,
+      "loss": 1.414,
+      "step": 5020
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1e-05,
+      "loss": 1.3735,
+      "step": 5030
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1e-05,
+      "loss": 1.3164,
+      "step": 5040
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1e-05,
+      "loss": 1.3472,
+      "step": 5050
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1e-05,
+      "loss": 1.2923,
+      "step": 5060
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1e-05,
+      "loss": 1.3314,
+      "step": 5070
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1e-05,
+      "loss": 1.2856,
+      "step": 5080
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1e-05,
+      "loss": 1.3573,
+      "step": 5090
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1e-05,
+      "loss": 1.2677,
+      "step": 5100
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1e-05,
+      "loss": 1.3191,
+      "step": 5110
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1e-05,
+      "loss": 1.3441,
+      "step": 5120
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1e-05,
+      "loss": 1.3753,
+      "step": 5130
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1e-05,
+      "loss": 1.3383,
+      "step": 5140
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1e-05,
+      "loss": 1.3634,
+      "step": 5150
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1e-05,
+      "loss": 1.3228,
+      "step": 5160
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1e-05,
+      "loss": 1.2332,
+      "step": 5170
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1e-05,
+      "loss": 1.4067,
+      "step": 5180
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1e-05,
+      "loss": 1.2971,
+      "step": 5190
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1e-05,
+      "loss": 1.3475,
+      "step": 5200
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1e-05,
+      "loss": 1.346,
+      "step": 5210
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1e-05,
+      "loss": 1.2515,
+      "step": 5220
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1e-05,
+      "loss": 1.2941,
+      "step": 5230
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1e-05,
+      "loss": 1.3747,
+      "step": 5240
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1e-05,
+      "loss": 1.2792,
+      "step": 5250
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1e-05,
+      "loss": 1.3299,
+      "step": 5260
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1e-05,
+      "loss": 1.3716,
+      "step": 5270
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1e-05,
+      "loss": 1.3374,
+      "step": 5280
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1e-05,
+      "loss": 1.2217,
+      "step": 5290
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1e-05,
+      "loss": 1.3765,
+      "step": 5300
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1e-05,
+      "loss": 1.3176,
+      "step": 5310
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1e-05,
+      "loss": 1.3696,
+      "step": 5320
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1e-05,
+      "loss": 1.1787,
+      "step": 5330
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1e-05,
+      "loss": 1.2461,
+      "step": 5340
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1e-05,
+      "loss": 1.2431,
+      "step": 5350
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1e-05,
+      "loss": 1.225,
+      "step": 5360
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1e-05,
+      "loss": 1.3083,
+      "step": 5370
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1e-05,
+      "loss": 1.2606,
+      "step": 5380
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1e-05,
+      "loss": 1.3286,
+      "step": 5390
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1e-05,
+      "loss": 1.2922,
+      "step": 5400
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1e-05,
+      "loss": 1.212,
+      "step": 5410
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1e-05,
+      "loss": 1.3419,
+      "step": 5420
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1e-05,
+      "loss": 1.3236,
+      "step": 5430
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1e-05,
+      "loss": 1.3958,
+      "step": 5440
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1e-05,
+      "loss": 1.3291,
+      "step": 5450
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1e-05,
+      "loss": 1.3905,
+      "step": 5460
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1e-05,
+      "loss": 1.3746,
+      "step": 5470
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1e-05,
+      "loss": 1.2668,
+      "step": 5480
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1e-05,
+      "loss": 1.2886,
+      "step": 5490
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1e-05,
+      "loss": 1.3079,
+      "step": 5500
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1e-05,
+      "loss": 1.2446,
+      "step": 5510
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1e-05,
+      "loss": 1.262,
+      "step": 5520
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1e-05,
+      "loss": 1.3532,
+      "step": 5530
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1e-05,
+      "loss": 1.3518,
+      "step": 5540
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1e-05,
+      "loss": 1.3072,
+      "step": 5550
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1e-05,
+      "loss": 1.3111,
+      "step": 5560
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1e-05,
+      "loss": 1.2842,
+      "step": 5570
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1e-05,
+      "loss": 1.3496,
+      "step": 5580
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1e-05,
+      "loss": 1.3192,
+      "step": 5590
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1e-05,
+      "loss": 1.3427,
+      "step": 5600
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1e-05,
+      "loss": 1.3451,
+      "step": 5610
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1e-05,
+      "loss": 1.1581,
+      "step": 5620
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1e-05,
+      "loss": 1.3222,
+      "step": 5630
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1e-05,
+      "loss": 1.2949,
+      "step": 5640
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1e-05,
+      "loss": 1.3879,
+      "step": 5650
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1e-05,
+      "loss": 1.3179,
+      "step": 5660
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1e-05,
+      "loss": 1.3071,
+      "step": 5670
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1e-05,
+      "loss": 1.2956,
+      "step": 5680
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1e-05,
+      "loss": 1.281,
+      "step": 5690
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1e-05,
+      "loss": 1.1717,
+      "step": 5700
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1e-05,
+      "loss": 1.2689,
+      "step": 5710
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1e-05,
+      "loss": 1.3562,
+      "step": 5720
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1e-05,
+      "loss": 1.3056,
+      "step": 5730
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1e-05,
+      "loss": 1.3551,
+      "step": 5740
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1e-05,
+      "loss": 1.293,
+      "step": 5750
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1e-05,
+      "loss": 1.3762,
+      "step": 5760
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1e-05,
+      "loss": 1.3105,
+      "step": 5770
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1e-05,
+      "loss": 1.3378,
+      "step": 5780
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1e-05,
+      "loss": 1.3054,
+      "step": 5790
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1e-05,
+      "loss": 1.2674,
+      "step": 5800
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1e-05,
+      "loss": 1.2965,
+      "step": 5810
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1e-05,
+      "loss": 1.3108,
+      "step": 5820
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1e-05,
+      "loss": 1.3587,
+      "step": 5830
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1e-05,
+      "loss": 1.323,
+      "step": 5840
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1e-05,
+      "loss": 1.2806,
+      "step": 5850
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1e-05,
+      "loss": 1.349,
+      "step": 5860
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1e-05,
+      "loss": 1.2239,
+      "step": 5870
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1e-05,
+      "loss": 1.3114,
+      "step": 5880
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1e-05,
+      "loss": 1.2964,
+      "step": 5890
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1e-05,
+      "loss": 1.3216,
+      "step": 5900
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1e-05,
+      "loss": 1.2054,
+      "step": 5910
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1e-05,
+      "loss": 1.2829,
+      "step": 5920
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1e-05,
+      "loss": 1.3919,
+      "step": 5930
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1e-05,
+      "loss": 1.3553,
+      "step": 5940
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1e-05,
+      "loss": 1.4125,
+      "step": 5950
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1e-05,
+      "loss": 1.4373,
+      "step": 5960
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1e-05,
+      "loss": 1.4466,
+      "step": 5970
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1e-05,
+      "loss": 1.2954,
+      "step": 5980
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1e-05,
+      "loss": 1.2423,
+      "step": 5990
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1e-05,
+      "loss": 1.4012,
+      "step": 6000
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1e-05,
+      "loss": 1.3476,
+      "step": 6010
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1e-05,
+      "loss": 1.3581,
+      "step": 6020
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1e-05,
+      "loss": 1.2769,
+      "step": 6030
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1e-05,
+      "loss": 1.3585,
+      "step": 6040
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1e-05,
+      "loss": 1.3423,
+      "step": 6050
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1e-05,
+      "loss": 1.1853,
+      "step": 6060
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1e-05,
+      "loss": 1.285,
+      "step": 6070
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1e-05,
+      "loss": 1.3281,
+      "step": 6080
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1e-05,
+      "loss": 1.2289,
+      "step": 6090
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1e-05,
+      "loss": 1.3121,
+      "step": 6100
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1e-05,
+      "loss": 1.2404,
+      "step": 6110
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1e-05,
+      "loss": 1.3526,
+      "step": 6120
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1e-05,
+      "loss": 1.2416,
+      "step": 6130
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1e-05,
+      "loss": 1.2881,
+      "step": 6140
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1e-05,
+      "loss": 1.3063,
+      "step": 6150
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1e-05,
+      "loss": 1.3502,
+      "step": 6160
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1e-05,
+      "loss": 1.2806,
+      "step": 6170
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1e-05,
+      "loss": 1.2861,
+      "step": 6180
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1e-05,
+      "loss": 1.3131,
+      "step": 6190
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1e-05,
+      "loss": 1.3383,
+      "step": 6200
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1e-05,
+      "loss": 1.1937,
+      "step": 6210
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1e-05,
+      "loss": 1.3753,
+      "step": 6220
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1e-05,
+      "loss": 1.2823,
+      "step": 6230
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1e-05,
+      "loss": 1.28,
+      "step": 6240
+    },
+    {
+      "epoch": 1.0,
+      "eval_oasst_export_accuracy": 0.6477435716215936,
+      "eval_oasst_export_loss": 1.578125,
+      "eval_oasst_export_runtime": 11.6714,
+      "eval_oasst_export_samples_per_second": 25.704,
+      "eval_oasst_export_steps_per_second": 3.256,
+      "step": 6240
+    },
+    {
+      "epoch": 1.0,
+      "eval_dolly15k_accuracy": 0.6555342233334822,
+      "eval_dolly15k_loss": 1.2724609375,
+      "eval_dolly15k_runtime": 9.1686,
+      "eval_dolly15k_samples_per_second": 32.72,
+      "eval_dolly15k_steps_per_second": 4.145,
+      "step": 6240
+    },
+    {
+      "epoch": 1.0,
+      "eval__ruozhiba_accuracy": 0.5006150061500615,
+      "eval__ruozhiba_loss": 1.8681640625,
+      "eval__ruozhiba_runtime": 1.3716,
+      "eval__ruozhiba_samples_per_second": 5.104,
+      "eval__ruozhiba_steps_per_second": 0.729,
+      "step": 6240
+    },
+    {
+      "epoch": 1.0,
+      "eval__sharegpt_format_accuracy": 0.8640188866555115,
+      "eval__sharegpt_format_loss": 0.783203125,
+      "eval__sharegpt_format_runtime": 27.1892,
+      "eval__sharegpt_format_samples_per_second": 18.39,
+      "eval__sharegpt_format_steps_per_second": 2.317,
+      "step": 6240
+    },
+    {
+      "epoch": 1.0,
+      "eval__similar_question_accuracy": 0.5383693045563549,
+      "eval__similar_question_loss": 1.943359375,
+      "eval__similar_question_runtime": 0.65,
+      "eval__similar_question_samples_per_second": 10.768,
+      "eval__similar_question_steps_per_second": 1.538,
+      "step": 6240
+    },
+    {
+      "epoch": 1.0,
+      "eval__open_domain_subject_accuracy": 0.5881886642885454,
+      "eval__open_domain_subject_loss": 1.6328125,
+      "eval__open_domain_subject_runtime": 1.7927,
+      "eval__open_domain_subject_samples_per_second": 25.66,
+      "eval__open_domain_subject_steps_per_second": 3.347,
+      "step": 6240
+    },
+    {
+      "epoch": 1.0,
+      "eval__empdia_ly_07_17_accuracy": 0.6118372695024128,
+      "eval__empdia_ly_07_17_loss": 1.4130859375,
+      "eval__empdia_ly_07_17_runtime": 4.1234,
+      "eval__empdia_ly_07_17_samples_per_second": 22.312,
+      "eval__empdia_ly_07_17_steps_per_second": 2.91,
+      "step": 6240
+    },
+    {
+      "epoch": 1.0,
+      "eval__zuowen_accuracy": 0.9063775741631135,
+      "eval__zuowen_loss": 0.32373046875,
+      "eval__zuowen_runtime": 4.8684,
+      "eval__zuowen_samples_per_second": 20.541,
+      "eval__zuowen_steps_per_second": 2.67,
+      "step": 6240
+    },
+    {
+      "epoch": 1.0,
+      "eval__psychat_6_29_and_11_15_mix_accuracy": 0.48304980276134124,
+      "eval__psychat_6_29_and_11_15_mix_loss": 2.087890625,
+      "eval__psychat_6_29_and_11_15_mix_runtime": 4.0357,
+      "eval__psychat_6_29_and_11_15_mix_samples_per_second": 24.779,
+      "eval__psychat_6_29_and_11_15_mix_steps_per_second": 3.221,
+      "step": 6240
+    },
+    {
+      "epoch": 1.0,
+      "eval__socrates_teaching_accuracy": 0.6203293223559214,
+      "eval__socrates_teaching_loss": 1.5263671875,
+      "eval__socrates_teaching_runtime": 3.9583,
+      "eval__socrates_teaching_samples_per_second": 21.727,
+      "eval__socrates_teaching_steps_per_second": 2.779,
+      "step": 6240
+    },
+    {
+      "epoch": 1.0,
+      "eval__socrates_teaching_math1115_accuracy": 0.6545029123313586,
+      "eval__socrates_teaching_math1115_loss": 1.361328125,
+      "eval__socrates_teaching_math1115_runtime": 4.4559,
+      "eval__socrates_teaching_math1115_samples_per_second": 22.442,
+      "eval__socrates_teaching_math1115_steps_per_second": 2.918,
+      "step": 6240
+    },
+    {
+      "epoch": 1.0,
+      "eval__search_accuracy": 0.6712757912930755,
+      "eval__search_loss": 1.2890625,
+      "eval__search_runtime": 4.8386,
+      "eval__search_samples_per_second": 20.667,
+      "eval__search_steps_per_second": 2.687,
+      "step": 6240
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 18723,
+  "num_train_epochs": 3,
+  "save_steps": 3120,
+  "total_flos": 8905482717626368.0,
+  "trial_name": null,
+  "trial_params": null
+}