update model

added_tokens.json

@@ -0,0 +1,7 @@
+{
+  "</s>": 2,
+  "<bot>": 39409,
+  "<human>": 39408,
+  "<s>": 1,
+  "<unk>": 0
+}

config.json

@@ -0,0 +1,32 @@
+{
+  "_name_or_path": "/cognitive_comp/ganruyi/Fengshenbang-LM/fengshen/workspace/ziya_llama2_13b_medical/output_exp5/checkpoint-158105",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 5120,
+  "initializer_range": 0.02,
+  "intermediate_size": 13824,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 4096,
+  "model_type": "llama",
+  "num_attention_heads": 40,
+  "num_hidden_layers": 40,
+  "num_key_value_heads": 40,
+  "pad_token_id": 0,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "rotary_emb_base": 10000,
+  "rotary_pct": 1,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.34.0.dev0",
+  "use_cache": false,
+  "use_parallel_residual": false,
+  "vocab_size": 39424
+}

generation_config.json

@@ -0,0 +1,8 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "pad_token_id": 0,
+  "transformers_version": "4.34.0.dev0",
+  "use_cache": false
+}

pytorch_model-00001-of-00003.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:77db05ae2e87ad132ebaade0d0d6204124c23ce9fced390c0353929539ae695e
+size 9972312054

pytorch_model-00002-of-00003.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a10797f65ea9d2819235a984c4fff7917488a5e8157dfd571267185a2f62869
+size 9956584547

pytorch_model-00003-of-00003.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a295ac580085065a471bb00998c49b1f8793605ef5ad5b19567543418aaeb73
+size 6254999967

pytorch_model.bin.index.json

@@ -0,0 +1,370 @@
+{
+  "metadata": {
+    "total_size": 26183772160
+  },
+  "weight_map": {
+    "lm_head.weight": "pytorch_model-00003-of-00003.bin",
+    "model.embed_tokens.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.0.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.0.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.0.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.0.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.0.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.0.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.0.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.0.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.0.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.1.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.1.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.1.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.1.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.1.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.1.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.1.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.1.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.1.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.10.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.10.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.10.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.10.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.10.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.10.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.10.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.10.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.10.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.11.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.11.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.11.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.11.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.11.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.11.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.11.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.11.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.11.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.12.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.12.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.12.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.12.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.12.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.12.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.12.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.12.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.12.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.13.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.13.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.13.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.13.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.13.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.13.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.13.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.13.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.13.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.14.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.14.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.14.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.14.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.14.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.14.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.14.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.14.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.14.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.15.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.15.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.15.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.15.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.15.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.15.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.15.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.15.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.15.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.16.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.16.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.16.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.16.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.16.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.16.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.16.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.16.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.16.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.17.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.17.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.17.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.17.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.17.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.17.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.17.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.17.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.17.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.18.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.18.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.18.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.18.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.18.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.18.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.18.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.18.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.18.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.19.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.19.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.19.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.19.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.19.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.19.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.19.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.19.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.19.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.2.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.2.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.2.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.2.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.2.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.2.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.2.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.2.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.2.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.20.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.20.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.20.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.20.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.20.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.20.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.20.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.20.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.20.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.21.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.21.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.21.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.21.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.21.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.21.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.21.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.21.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.21.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.22.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.22.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.22.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.22.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.22.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.22.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.22.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.22.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.22.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.23.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.23.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.23.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.23.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.23.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.23.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.23.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.23.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.23.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.24.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.24.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.24.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.24.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.24.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.24.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.24.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.24.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.24.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.25.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.25.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.25.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.25.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.25.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.25.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.25.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.25.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.25.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.26.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.26.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.26.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.26.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.26.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.26.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.26.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.26.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.26.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.27.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.27.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.27.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.27.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.27.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.27.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.27.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.27.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.27.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.28.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.28.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.28.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.28.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.28.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.28.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.28.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.28.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.28.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.29.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.29.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.29.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.29.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.29.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.29.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.29.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.29.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.29.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.3.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.3.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.3.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.3.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.3.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.3.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.3.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.3.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.3.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.30.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.30.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.30.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.30.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.30.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.30.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.30.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.30.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.30.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.31.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.31.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.31.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.31.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.31.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.31.self_attn.k_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.31.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.31.self_attn.q_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.31.self_attn.v_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.32.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.32.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.32.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.32.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.32.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.32.self_attn.k_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.32.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.32.self_attn.q_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.32.self_attn.v_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.33.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.33.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.33.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.33.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.33.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.33.self_attn.k_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.33.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.33.self_attn.q_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.33.self_attn.v_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.34.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.34.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.34.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.34.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.34.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.34.self_attn.k_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.34.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.34.self_attn.q_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.34.self_attn.v_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.35.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.35.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.35.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.35.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.35.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.35.self_attn.k_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.35.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.35.self_attn.q_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.35.self_attn.v_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.36.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.36.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.36.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.36.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.36.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.36.self_attn.k_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.36.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.36.self_attn.q_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.36.self_attn.v_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.37.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.37.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.37.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.37.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.37.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.37.self_attn.k_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.37.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.37.self_attn.q_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.37.self_attn.v_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.38.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.38.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.38.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.38.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.38.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.38.self_attn.k_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.38.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.38.self_attn.q_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.38.self_attn.v_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.39.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.39.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.39.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.39.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.39.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.39.self_attn.k_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.39.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.39.self_attn.q_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.39.self_attn.v_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.4.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.4.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.4.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.4.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.4.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.4.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.4.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.4.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.4.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.5.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.5.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.5.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.5.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.5.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.5.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.5.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.5.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.5.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.6.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.6.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.6.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.6.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.6.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.6.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.6.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.6.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.6.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.7.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.7.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.7.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.7.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.7.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.7.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.7.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.7.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.7.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.8.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.8.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.8.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.8.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.8.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.8.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.8.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.8.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.8.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.9.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.9.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.9.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.9.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.9.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.9.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.9.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.9.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.9.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.norm.weight": "pytorch_model-00003-of-00003.bin"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,10 @@
+{
+  "additional_special_tokens": [
+    "<human>",
+    "<bot>"
+  ],
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "pad_token": "</s>",
+  "unk_token": "<unk>"
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6fd7e445833dd0889206aba242c2a51ecbae2437fd328d1759a35475fd8c0423
+size 588619

tokenizer_config.json

@@ -0,0 +1,63 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "39408": {
+      "content": "<human>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "39409": {
+      "content": "<bot>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<human>",
+    "<bot>"
+  ],
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": true,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": null,
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "tokenizer_file": null,
+  "unk_token": "<unk>",
+  "use_default_system_prompt": true,
+  "use_fast": false
+}

trainer_state.json

@@ -0,0 +1,4075 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 1000,
+  "global_step": 676,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.142857142857143e-07,
+      "loss": 1.7121,
+      "step": 1
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4285714285714286e-06,
+      "loss": 1.6351,
+      "step": 2
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1428571428571427e-06,
+      "loss": 0.9576,
+      "step": 3
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8571428571428573e-06,
+      "loss": 0.6691,
+      "step": 4
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5714285714285718e-06,
+      "loss": 1.155,
+      "step": 5
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2857142857142855e-06,
+      "loss": 0.5538,
+      "step": 6
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5e-06,
+      "loss": 1.5242,
+      "step": 7
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.7142857142857145e-06,
+      "loss": 1.0917,
+      "step": 8
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 6.4285714285714295e-06,
+      "loss": 0.9385,
+      "step": 9
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 7.1428571428571436e-06,
+      "loss": 0.745,
+      "step": 10
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 7.857142857142858e-06,
+      "loss": 1.0332,
+      "step": 11
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 8.571428571428571e-06,
+      "loss": 0.8973,
+      "step": 12
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.285714285714288e-06,
+      "loss": 0.8048,
+      "step": 13
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1e-05,
+      "loss": 0.9936,
+      "step": 14
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.0714285714285714e-05,
+      "loss": 0.8611,
+      "step": 15
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.1428571428571429e-05,
+      "loss": 0.951,
+      "step": 16
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.2142857142857142e-05,
+      "loss": 0.636,
+      "step": 17
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.2857142857142859e-05,
+      "loss": 0.2592,
+      "step": 18
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.3571428571428574e-05,
+      "loss": 0.7124,
+      "step": 19
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.4285714285714287e-05,
+      "loss": 0.6169,
+      "step": 20
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.5000000000000002e-05,
+      "loss": 0.6877,
+      "step": 21
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.5714285714285715e-05,
+      "loss": 0.5239,
+      "step": 22
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.642857142857143e-05,
+      "loss": 0.7335,
+      "step": 23
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.7142857142857142e-05,
+      "loss": 0.7291,
+      "step": 24
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.785714285714286e-05,
+      "loss": 0.6331,
+      "step": 25
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.8571428571428575e-05,
+      "loss": 0.3293,
+      "step": 26
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.928571428571429e-05,
+      "loss": 0.5166,
+      "step": 27
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 2e-05,
+      "loss": 0.3995,
+      "step": 28
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.999252615844544e-05,
+      "loss": 0.4698,
+      "step": 29
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9985052316890884e-05,
+      "loss": 0.599,
+      "step": 30
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9977578475336326e-05,
+      "loss": 0.3244,
+      "step": 31
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9970104633781765e-05,
+      "loss": 0.4321,
+      "step": 32
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9962630792227208e-05,
+      "loss": 0.2851,
+      "step": 33
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9955156950672647e-05,
+      "loss": 0.5091,
+      "step": 34
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9947683109118087e-05,
+      "loss": 0.4354,
+      "step": 35
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.994020926756353e-05,
+      "loss": 0.4124,
+      "step": 36
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9932735426008972e-05,
+      "loss": 0.4631,
+      "step": 37
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.992526158445441e-05,
+      "loss": 0.2247,
+      "step": 38
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9917787742899854e-05,
+      "loss": 0.3505,
+      "step": 39
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9910313901345293e-05,
+      "loss": 0.2736,
+      "step": 40
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9902840059790732e-05,
+      "loss": 0.3274,
+      "step": 41
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9895366218236175e-05,
+      "loss": 0.4126,
+      "step": 42
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9887892376681618e-05,
+      "loss": 0.4884,
+      "step": 43
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9880418535127057e-05,
+      "loss": 0.3895,
+      "step": 44
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.98729446935725e-05,
+      "loss": 0.3829,
+      "step": 45
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.986547085201794e-05,
+      "loss": 0.2123,
+      "step": 46
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9857997010463378e-05,
+      "loss": 0.3708,
+      "step": 47
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.985052316890882e-05,
+      "loss": 0.3577,
+      "step": 48
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9843049327354263e-05,
+      "loss": 0.4285,
+      "step": 49
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9835575485799703e-05,
+      "loss": 0.495,
+      "step": 50
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9828101644245142e-05,
+      "loss": 0.3795,
+      "step": 51
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9820627802690585e-05,
+      "loss": 0.254,
+      "step": 52
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9813153961136024e-05,
+      "loss": 0.2853,
+      "step": 53
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9805680119581466e-05,
+      "loss": 0.2242,
+      "step": 54
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.979820627802691e-05,
+      "loss": 0.3997,
+      "step": 55
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.979073243647235e-05,
+      "loss": 0.3478,
+      "step": 56
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9783258594917788e-05,
+      "loss": 0.3803,
+      "step": 57
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.977578475336323e-05,
+      "loss": 0.4557,
+      "step": 58
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.976831091180867e-05,
+      "loss": 0.4193,
+      "step": 59
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9760837070254112e-05,
+      "loss": 0.2113,
+      "step": 60
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9753363228699555e-05,
+      "loss": 0.3107,
+      "step": 61
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9745889387144994e-05,
+      "loss": 0.337,
+      "step": 62
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9738415545590433e-05,
+      "loss": 0.375,
+      "step": 63
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9730941704035876e-05,
+      "loss": 0.3449,
+      "step": 64
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.972346786248132e-05,
+      "loss": 0.2696,
+      "step": 65
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9715994020926758e-05,
+      "loss": 0.3429,
+      "step": 66
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.97085201793722e-05,
+      "loss": 0.3534,
+      "step": 67
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.970104633781764e-05,
+      "loss": 0.2249,
+      "step": 68
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.969357249626308e-05,
+      "loss": 0.2489,
+      "step": 69
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9686098654708522e-05,
+      "loss": 0.4954,
+      "step": 70
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9678624813153964e-05,
+      "loss": 0.3308,
+      "step": 71
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9671150971599404e-05,
+      "loss": 0.2431,
+      "step": 72
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9663677130044846e-05,
+      "loss": 0.2689,
+      "step": 73
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9656203288490286e-05,
+      "loss": 0.2281,
+      "step": 74
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9648729446935725e-05,
+      "loss": 0.3062,
+      "step": 75
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9641255605381167e-05,
+      "loss": 0.339,
+      "step": 76
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.963378176382661e-05,
+      "loss": 0.2751,
+      "step": 77
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.962630792227205e-05,
+      "loss": 0.3185,
+      "step": 78
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9618834080717492e-05,
+      "loss": 0.3796,
+      "step": 79
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.961136023916293e-05,
+      "loss": 0.3768,
+      "step": 80
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.960388639760837e-05,
+      "loss": 0.2139,
+      "step": 81
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9596412556053813e-05,
+      "loss": 0.1631,
+      "step": 82
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9588938714499256e-05,
+      "loss": 0.2343,
+      "step": 83
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9581464872944695e-05,
+      "loss": 0.3333,
+      "step": 84
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9573991031390134e-05,
+      "loss": 0.2805,
+      "step": 85
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9566517189835577e-05,
+      "loss": 0.3305,
+      "step": 86
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9559043348281016e-05,
+      "loss": 0.2159,
+      "step": 87
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.955156950672646e-05,
+      "loss": 0.3456,
+      "step": 88
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.95440956651719e-05,
+      "loss": 0.2844,
+      "step": 89
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.953662182361734e-05,
+      "loss": 0.2367,
+      "step": 90
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.952914798206278e-05,
+      "loss": 0.2048,
+      "step": 91
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9521674140508223e-05,
+      "loss": 0.2589,
+      "step": 92
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9514200298953662e-05,
+      "loss": 0.2653,
+      "step": 93
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9506726457399105e-05,
+      "loss": 0.2004,
+      "step": 94
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9499252615844547e-05,
+      "loss": 0.2007,
+      "step": 95
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9491778774289987e-05,
+      "loss": 0.1923,
+      "step": 96
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9484304932735426e-05,
+      "loss": 0.1986,
+      "step": 97
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.947683109118087e-05,
+      "loss": 0.4276,
+      "step": 98
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.946935724962631e-05,
+      "loss": 0.2475,
+      "step": 99
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.946188340807175e-05,
+      "loss": 0.1941,
+      "step": 100
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9454409566517193e-05,
+      "loss": 0.2069,
+      "step": 101
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9446935724962632e-05,
+      "loss": 0.2331,
+      "step": 102
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.943946188340807e-05,
+      "loss": 0.2208,
+      "step": 103
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9431988041853514e-05,
+      "loss": 0.2585,
+      "step": 104
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9424514200298957e-05,
+      "loss": 0.2988,
+      "step": 105
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9417040358744396e-05,
+      "loss": 0.2965,
+      "step": 106
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.940956651718984e-05,
+      "loss": 0.2611,
+      "step": 107
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9402092675635278e-05,
+      "loss": 0.2909,
+      "step": 108
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9394618834080717e-05,
+      "loss": 0.1858,
+      "step": 109
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.938714499252616e-05,
+      "loss": 0.1401,
+      "step": 110
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9379671150971603e-05,
+      "loss": 0.2057,
+      "step": 111
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9372197309417042e-05,
+      "loss": 0.128,
+      "step": 112
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9364723467862484e-05,
+      "loss": 0.2158,
+      "step": 113
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9357249626307924e-05,
+      "loss": 0.2921,
+      "step": 114
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9349775784753363e-05,
+      "loss": 0.2217,
+      "step": 115
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9342301943198806e-05,
+      "loss": 0.1605,
+      "step": 116
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9334828101644248e-05,
+      "loss": 0.2123,
+      "step": 117
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9327354260089688e-05,
+      "loss": 0.1571,
+      "step": 118
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.931988041853513e-05,
+      "loss": 0.3153,
+      "step": 119
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.931240657698057e-05,
+      "loss": 0.1478,
+      "step": 120
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.930493273542601e-05,
+      "loss": 0.2461,
+      "step": 121
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.929745889387145e-05,
+      "loss": 0.1576,
+      "step": 122
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9289985052316894e-05,
+      "loss": 0.2529,
+      "step": 123
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9282511210762333e-05,
+      "loss": 0.1751,
+      "step": 124
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9275037369207773e-05,
+      "loss": 0.2999,
+      "step": 125
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9267563527653215e-05,
+      "loss": 0.3322,
+      "step": 126
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9260089686098654e-05,
+      "loss": 0.2252,
+      "step": 127
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9252615844544097e-05,
+      "loss": 0.1505,
+      "step": 128
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.924514200298954e-05,
+      "loss": 0.1555,
+      "step": 129
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.923766816143498e-05,
+      "loss": 0.1885,
+      "step": 130
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9230194319880418e-05,
+      "loss": 0.2109,
+      "step": 131
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.922272047832586e-05,
+      "loss": 0.1625,
+      "step": 132
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9215246636771304e-05,
+      "loss": 0.1517,
+      "step": 133
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9207772795216743e-05,
+      "loss": 0.2655,
+      "step": 134
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9200298953662185e-05,
+      "loss": 0.1839,
+      "step": 135
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9192825112107625e-05,
+      "loss": 0.258,
+      "step": 136
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9185351270553064e-05,
+      "loss": 0.1707,
+      "step": 137
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9177877428998507e-05,
+      "loss": 0.0959,
+      "step": 138
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.917040358744395e-05,
+      "loss": 0.1505,
+      "step": 139
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.916292974588939e-05,
+      "loss": 0.1793,
+      "step": 140
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.915545590433483e-05,
+      "loss": 0.1991,
+      "step": 141
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.914798206278027e-05,
+      "loss": 0.1646,
+      "step": 142
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.914050822122571e-05,
+      "loss": 0.1877,
+      "step": 143
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9133034379671152e-05,
+      "loss": 0.2178,
+      "step": 144
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9125560538116595e-05,
+      "loss": 0.1381,
+      "step": 145
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9118086696562034e-05,
+      "loss": 0.1962,
+      "step": 146
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9110612855007477e-05,
+      "loss": 0.2094,
+      "step": 147
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9103139013452916e-05,
+      "loss": 0.1103,
+      "step": 148
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9095665171898355e-05,
+      "loss": 0.1808,
+      "step": 149
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9088191330343798e-05,
+      "loss": 0.3039,
+      "step": 150
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.908071748878924e-05,
+      "loss": 0.1886,
+      "step": 151
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.907324364723468e-05,
+      "loss": 0.1659,
+      "step": 152
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9065769805680123e-05,
+      "loss": 0.1135,
+      "step": 153
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9058295964125562e-05,
+      "loss": 0.1233,
+      "step": 154
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9050822122571e-05,
+      "loss": 0.1556,
+      "step": 155
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9043348281016444e-05,
+      "loss": 0.1469,
+      "step": 156
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9035874439461886e-05,
+      "loss": 0.126,
+      "step": 157
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9028400597907326e-05,
+      "loss": 0.1184,
+      "step": 158
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9020926756352765e-05,
+      "loss": 0.1788,
+      "step": 159
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9013452914798208e-05,
+      "loss": 0.1624,
+      "step": 160
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9005979073243647e-05,
+      "loss": 0.3195,
+      "step": 161
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.899850523168909e-05,
+      "loss": 0.1057,
+      "step": 162
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.8991031390134532e-05,
+      "loss": 0.1392,
+      "step": 163
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.898355754857997e-05,
+      "loss": 0.1107,
+      "step": 164
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.897608370702541e-05,
+      "loss": 0.217,
+      "step": 165
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.8968609865470853e-05,
+      "loss": 0.161,
+      "step": 166
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.8961136023916296e-05,
+      "loss": 0.1969,
+      "step": 167
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.8953662182361735e-05,
+      "loss": 0.1386,
+      "step": 168
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.8946188340807178e-05,
+      "loss": 0.1529,
+      "step": 169
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.8938714499252617e-05,
+      "loss": 0.109,
+      "step": 170
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.8931240657698056e-05,
+      "loss": 0.1202,
+      "step": 171
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.89237668161435e-05,
+      "loss": 0.0875,
+      "step": 172
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.8916292974588942e-05,
+      "loss": 0.1248,
+      "step": 173
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.890881913303438e-05,
+      "loss": 0.0989,
+      "step": 174
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.8901345291479824e-05,
+      "loss": 0.1993,
+      "step": 175
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.8893871449925263e-05,
+      "loss": 0.094,
+      "step": 176
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.8886397608370702e-05,
+      "loss": 0.2574,
+      "step": 177
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.8878923766816145e-05,
+      "loss": 0.1246,
+      "step": 178
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.8871449925261588e-05,
+      "loss": 0.1293,
+      "step": 179
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.8863976083707027e-05,
+      "loss": 0.2807,
+      "step": 180
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.885650224215247e-05,
+      "loss": 0.0824,
+      "step": 181
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.884902840059791e-05,
+      "loss": 0.1072,
+      "step": 182
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.8841554559043348e-05,
+      "loss": 0.1424,
+      "step": 183
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.883408071748879e-05,
+      "loss": 0.136,
+      "step": 184
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8826606875934233e-05,
+      "loss": 0.1441,
+      "step": 185
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8819133034379672e-05,
+      "loss": 0.1195,
+      "step": 186
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8811659192825115e-05,
+      "loss": 0.1971,
+      "step": 187
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.8804185351270554e-05,
+      "loss": 0.1315,
+      "step": 188
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.8796711509715994e-05,
+      "loss": 0.2408,
+      "step": 189
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.8789237668161436e-05,
+      "loss": 0.1223,
+      "step": 190
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.878176382660688e-05,
+      "loss": 0.13,
+      "step": 191
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.8774289985052318e-05,
+      "loss": 0.1561,
+      "step": 192
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.8766816143497757e-05,
+      "loss": 0.174,
+      "step": 193
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.87593423019432e-05,
+      "loss": 0.1024,
+      "step": 194
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.875186846038864e-05,
+      "loss": 0.1497,
+      "step": 195
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.8744394618834082e-05,
+      "loss": 0.1025,
+      "step": 196
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.8736920777279525e-05,
+      "loss": 0.1955,
+      "step": 197
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.8729446935724964e-05,
+      "loss": 0.3608,
+      "step": 198
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.8721973094170403e-05,
+      "loss": 0.1002,
+      "step": 199
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.8714499252615846e-05,
+      "loss": 0.0897,
+      "step": 200
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.870702541106129e-05,
+      "loss": 0.1505,
+      "step": 201
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.8699551569506728e-05,
+      "loss": 0.1407,
+      "step": 202
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.869207772795217e-05,
+      "loss": 0.1,
+      "step": 203
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.868460388639761e-05,
+      "loss": 0.1155,
+      "step": 204
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.867713004484305e-05,
+      "loss": 0.1011,
+      "step": 205
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.866965620328849e-05,
+      "loss": 0.1389,
+      "step": 206
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.8662182361733934e-05,
+      "loss": 0.1635,
+      "step": 207
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.8654708520179373e-05,
+      "loss": 0.0981,
+      "step": 208
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.8647234678624816e-05,
+      "loss": 0.0854,
+      "step": 209
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.8639760837070255e-05,
+      "loss": 0.0481,
+      "step": 210
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.8632286995515695e-05,
+      "loss": 0.1173,
+      "step": 211
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.8624813153961137e-05,
+      "loss": 0.1232,
+      "step": 212
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.861733931240658e-05,
+      "loss": 0.1024,
+      "step": 213
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.860986547085202e-05,
+      "loss": 0.0921,
+      "step": 214
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8602391629297462e-05,
+      "loss": 0.0635,
+      "step": 215
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.85949177877429e-05,
+      "loss": 0.0682,
+      "step": 216
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.858744394618834e-05,
+      "loss": 0.1079,
+      "step": 217
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8579970104633783e-05,
+      "loss": 0.0951,
+      "step": 218
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.8572496263079226e-05,
+      "loss": 0.0614,
+      "step": 219
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.8565022421524665e-05,
+      "loss": 0.1315,
+      "step": 220
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.8557548579970108e-05,
+      "loss": 0.0428,
+      "step": 221
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.8550074738415547e-05,
+      "loss": 0.0627,
+      "step": 222
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.8542600896860986e-05,
+      "loss": 0.1626,
+      "step": 223
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.853512705530643e-05,
+      "loss": 0.2031,
+      "step": 224
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.852765321375187e-05,
+      "loss": 0.2312,
+      "step": 225
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.852017937219731e-05,
+      "loss": 0.077,
+      "step": 226
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8512705530642753e-05,
+      "loss": 0.0495,
+      "step": 227
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8505231689088193e-05,
+      "loss": 0.118,
+      "step": 228
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8497757847533632e-05,
+      "loss": 0.1213,
+      "step": 229
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8490284005979075e-05,
+      "loss": 0.2259,
+      "step": 230
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8482810164424517e-05,
+      "loss": 0.1215,
+      "step": 231
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.8475336322869956e-05,
+      "loss": 0.0947,
+      "step": 232
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.8467862481315396e-05,
+      "loss": 0.1236,
+      "step": 233
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.846038863976084e-05,
+      "loss": 0.1157,
+      "step": 234
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.845291479820628e-05,
+      "loss": 0.1132,
+      "step": 235
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.844544095665172e-05,
+      "loss": 0.0576,
+      "step": 236
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.8437967115097163e-05,
+      "loss": 0.262,
+      "step": 237
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.8430493273542602e-05,
+      "loss": 0.0824,
+      "step": 238
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.842301943198804e-05,
+      "loss": 0.1578,
+      "step": 239
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.8415545590433484e-05,
+      "loss": 0.0914,
+      "step": 240
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.8408071748878927e-05,
+      "loss": 0.1206,
+      "step": 241
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.8400597907324366e-05,
+      "loss": 0.1859,
+      "step": 242
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.839312406576981e-05,
+      "loss": 0.1447,
+      "step": 243
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.8385650224215248e-05,
+      "loss": 0.0675,
+      "step": 244
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.8378176382660687e-05,
+      "loss": 0.0735,
+      "step": 245
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.837070254110613e-05,
+      "loss": 0.0905,
+      "step": 246
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.8363228699551572e-05,
+      "loss": 0.0429,
+      "step": 247
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.835575485799701e-05,
+      "loss": 0.0729,
+      "step": 248
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.8348281016442454e-05,
+      "loss": 0.0806,
+      "step": 249
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.8340807174887894e-05,
+      "loss": 0.0601,
+      "step": 250
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.8333333333333333e-05,
+      "loss": 0.062,
+      "step": 251
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.8325859491778776e-05,
+      "loss": 0.1074,
+      "step": 252
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.8318385650224218e-05,
+      "loss": 0.0427,
+      "step": 253
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.8310911808669657e-05,
+      "loss": 0.1015,
+      "step": 254
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.83034379671151e-05,
+      "loss": 0.0521,
+      "step": 255
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.829596412556054e-05,
+      "loss": 0.0448,
+      "step": 256
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.828849028400598e-05,
+      "loss": 0.0585,
+      "step": 257
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.828101644245142e-05,
+      "loss": 0.0849,
+      "step": 258
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.8273542600896864e-05,
+      "loss": 0.0897,
+      "step": 259
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.8266068759342303e-05,
+      "loss": 0.0221,
+      "step": 260
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.8258594917787746e-05,
+      "loss": 0.0745,
+      "step": 261
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.8251121076233185e-05,
+      "loss": 0.0403,
+      "step": 262
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.8243647234678624e-05,
+      "loss": 0.0425,
+      "step": 263
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.8236173393124067e-05,
+      "loss": 0.1864,
+      "step": 264
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.822869955156951e-05,
+      "loss": 0.1687,
+      "step": 265
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.822122571001495e-05,
+      "loss": 0.1226,
+      "step": 266
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.8213751868460388e-05,
+      "loss": 0.0665,
+      "step": 267
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.820627802690583e-05,
+      "loss": 0.0278,
+      "step": 268
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.8198804185351273e-05,
+      "loss": 0.0579,
+      "step": 269
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.8191330343796713e-05,
+      "loss": 0.076,
+      "step": 270
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.8183856502242155e-05,
+      "loss": 0.1066,
+      "step": 271
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.8176382660687595e-05,
+      "loss": 0.0512,
+      "step": 272
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.8168908819133034e-05,
+      "loss": 0.0652,
+      "step": 273
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.8161434977578477e-05,
+      "loss": 0.0533,
+      "step": 274
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.815396113602392e-05,
+      "loss": 0.0637,
+      "step": 275
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.814648729446936e-05,
+      "loss": 0.042,
+      "step": 276
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.81390134529148e-05,
+      "loss": 0.031,
+      "step": 277
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.813153961136024e-05,
+      "loss": 0.0618,
+      "step": 278
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.812406576980568e-05,
+      "loss": 0.1812,
+      "step": 279
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.8116591928251122e-05,
+      "loss": 0.0516,
+      "step": 280
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.8109118086696565e-05,
+      "loss": 0.0574,
+      "step": 281
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.8101644245142004e-05,
+      "loss": 0.0619,
+      "step": 282
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.8094170403587447e-05,
+      "loss": 0.0776,
+      "step": 283
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.8086696562032886e-05,
+      "loss": 0.1249,
+      "step": 284
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.8079222720478325e-05,
+      "loss": 0.1777,
+      "step": 285
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.8071748878923768e-05,
+      "loss": 0.0577,
+      "step": 286
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.806427503736921e-05,
+      "loss": 0.0734,
+      "step": 287
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.805680119581465e-05,
+      "loss": 0.1135,
+      "step": 288
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.8049327354260093e-05,
+      "loss": 0.032,
+      "step": 289
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.8041853512705532e-05,
+      "loss": 0.1834,
+      "step": 290
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.803437967115097e-05,
+      "loss": 0.0563,
+      "step": 291
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.8026905829596414e-05,
+      "loss": 0.0525,
+      "step": 292
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.8019431988041856e-05,
+      "loss": 0.0602,
+      "step": 293
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.8011958146487296e-05,
+      "loss": 0.1962,
+      "step": 294
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.8004484304932738e-05,
+      "loss": 0.0732,
+      "step": 295
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.7997010463378178e-05,
+      "loss": 0.0252,
+      "step": 296
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.7989536621823617e-05,
+      "loss": 0.1243,
+      "step": 297
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.798206278026906e-05,
+      "loss": 0.0363,
+      "step": 298
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.7974588938714502e-05,
+      "loss": 0.0544,
+      "step": 299
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.796711509715994e-05,
+      "loss": 0.0315,
+      "step": 300
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.7959641255605384e-05,
+      "loss": 0.1768,
+      "step": 301
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.7952167414050823e-05,
+      "loss": 0.0601,
+      "step": 302
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.7944693572496266e-05,
+      "loss": 0.0393,
+      "step": 303
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.7937219730941705e-05,
+      "loss": 0.0289,
+      "step": 304
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.7929745889387148e-05,
+      "loss": 0.1323,
+      "step": 305
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.7922272047832587e-05,
+      "loss": 0.1477,
+      "step": 306
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.7914798206278026e-05,
+      "loss": 0.0374,
+      "step": 307
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.790732436472347e-05,
+      "loss": 0.1036,
+      "step": 308
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.789985052316891e-05,
+      "loss": 0.0417,
+      "step": 309
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.789237668161435e-05,
+      "loss": 0.0736,
+      "step": 310
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.7884902840059794e-05,
+      "loss": 0.0307,
+      "step": 311
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.7877428998505233e-05,
+      "loss": 0.0684,
+      "step": 312
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.7869955156950672e-05,
+      "loss": 0.0391,
+      "step": 313
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.7862481315396115e-05,
+      "loss": 0.0781,
+      "step": 314
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.7855007473841557e-05,
+      "loss": 0.2455,
+      "step": 315
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.7847533632286997e-05,
+      "loss": 0.1698,
+      "step": 316
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.784005979073244e-05,
+      "loss": 0.0173,
+      "step": 317
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.783258594917788e-05,
+      "loss": 0.1204,
+      "step": 318
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.7825112107623318e-05,
+      "loss": 0.0667,
+      "step": 319
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.781763826606876e-05,
+      "loss": 0.0214,
+      "step": 320
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.7810164424514203e-05,
+      "loss": 0.0351,
+      "step": 321
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.7802690582959642e-05,
+      "loss": 0.2676,
+      "step": 322
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.7795216741405085e-05,
+      "loss": 0.0633,
+      "step": 323
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.7787742899850524e-05,
+      "loss": 0.0249,
+      "step": 324
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.7780269058295964e-05,
+      "loss": 0.1116,
+      "step": 325
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.7772795216741406e-05,
+      "loss": 0.0439,
+      "step": 326
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.776532137518685e-05,
+      "loss": 0.0798,
+      "step": 327
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.7757847533632288e-05,
+      "loss": 0.0249,
+      "step": 328
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.775037369207773e-05,
+      "loss": 0.0285,
+      "step": 329
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.774289985052317e-05,
+      "loss": 0.0945,
+      "step": 330
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.773542600896861e-05,
+      "loss": 0.1555,
+      "step": 331
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.7727952167414052e-05,
+      "loss": 0.0378,
+      "step": 332
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.7720478325859495e-05,
+      "loss": 0.0234,
+      "step": 333
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.7713004484304934e-05,
+      "loss": 0.084,
+      "step": 334
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.7705530642750376e-05,
+      "loss": 0.0862,
+      "step": 335
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.7698056801195816e-05,
+      "loss": 0.0364,
+      "step": 336
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.769058295964126e-05,
+      "loss": 0.0281,
+      "step": 337
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7683109118086698e-05,
+      "loss": 0.0277,
+      "step": 338
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.767563527653214e-05,
+      "loss": 0.014,
+      "step": 339
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.766816143497758e-05,
+      "loss": 0.0129,
+      "step": 340
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.766068759342302e-05,
+      "loss": 0.0411,
+      "step": 341
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.765321375186846e-05,
+      "loss": 0.0182,
+      "step": 342
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.7645739910313904e-05,
+      "loss": 0.0105,
+      "step": 343
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7638266068759343e-05,
+      "loss": 0.0083,
+      "step": 344
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7630792227204786e-05,
+      "loss": 0.0113,
+      "step": 345
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7623318385650225e-05,
+      "loss": 0.0051,
+      "step": 346
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.7615844544095665e-05,
+      "loss": 0.01,
+      "step": 347
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.7608370702541107e-05,
+      "loss": 0.0148,
+      "step": 348
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.760089686098655e-05,
+      "loss": 0.0527,
+      "step": 349
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.759342301943199e-05,
+      "loss": 0.0112,
+      "step": 350
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.7585949177877432e-05,
+      "loss": 0.0759,
+      "step": 351
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.757847533632287e-05,
+      "loss": 0.0126,
+      "step": 352
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.757100149476831e-05,
+      "loss": 0.0165,
+      "step": 353
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.7563527653213753e-05,
+      "loss": 0.0113,
+      "step": 354
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.7556053811659196e-05,
+      "loss": 0.0394,
+      "step": 355
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.7548579970104635e-05,
+      "loss": 0.009,
+      "step": 356
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.7541106128550077e-05,
+      "loss": 0.0126,
+      "step": 357
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.7533632286995517e-05,
+      "loss": 0.0162,
+      "step": 358
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.7526158445440956e-05,
+      "loss": 0.0052,
+      "step": 359
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.75186846038864e-05,
+      "loss": 0.059,
+      "step": 360
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.751121076233184e-05,
+      "loss": 0.0444,
+      "step": 361
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.750373692077728e-05,
+      "loss": 0.006,
+      "step": 362
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.7496263079222723e-05,
+      "loss": 0.0109,
+      "step": 363
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.7488789237668162e-05,
+      "loss": 0.0062,
+      "step": 364
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.7481315396113602e-05,
+      "loss": 0.0102,
+      "step": 365
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.7473841554559044e-05,
+      "loss": 0.01,
+      "step": 366
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.7466367713004487e-05,
+      "loss": 0.0094,
+      "step": 367
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.7458893871449926e-05,
+      "loss": 0.0457,
+      "step": 368
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.745142002989537e-05,
+      "loss": 0.009,
+      "step": 369
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.7443946188340808e-05,
+      "loss": 0.0058,
+      "step": 370
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.7436472346786247e-05,
+      "loss": 0.0078,
+      "step": 371
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.742899850523169e-05,
+      "loss": 0.011,
+      "step": 372
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.7421524663677133e-05,
+      "loss": 0.0078,
+      "step": 373
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.7414050822122572e-05,
+      "loss": 0.0081,
+      "step": 374
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.740657698056801e-05,
+      "loss": 0.0059,
+      "step": 375
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.7399103139013454e-05,
+      "loss": 0.0101,
+      "step": 376
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.7391629297458897e-05,
+      "loss": 0.0474,
+      "step": 377
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.7384155455904336e-05,
+      "loss": 0.0085,
+      "step": 378
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.737668161434978e-05,
+      "loss": 0.0097,
+      "step": 379
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.7369207772795218e-05,
+      "loss": 0.0073,
+      "step": 380
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.7361733931240657e-05,
+      "loss": 0.0101,
+      "step": 381
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.73542600896861e-05,
+      "loss": 0.0569,
+      "step": 382
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.7346786248131542e-05,
+      "loss": 0.0099,
+      "step": 383
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.733931240657698e-05,
+      "loss": 0.0083,
+      "step": 384
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.7331838565022424e-05,
+      "loss": 0.0178,
+      "step": 385
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.7324364723467863e-05,
+      "loss": 0.0059,
+      "step": 386
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.7316890881913303e-05,
+      "loss": 0.0073,
+      "step": 387
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.7309417040358745e-05,
+      "loss": 0.0295,
+      "step": 388
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.7301943198804188e-05,
+      "loss": 0.0382,
+      "step": 389
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.7294469357249627e-05,
+      "loss": 0.0195,
+      "step": 390
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.728699551569507e-05,
+      "loss": 0.0097,
+      "step": 391
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.727952167414051e-05,
+      "loss": 0.0111,
+      "step": 392
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.727204783258595e-05,
+      "loss": 0.0214,
+      "step": 393
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.726457399103139e-05,
+      "loss": 0.0106,
+      "step": 394
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.7257100149476834e-05,
+      "loss": 0.0035,
+      "step": 395
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.7249626307922273e-05,
+      "loss": 0.0935,
+      "step": 396
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.7242152466367716e-05,
+      "loss": 0.0049,
+      "step": 397
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.7234678624813155e-05,
+      "loss": 0.0112,
+      "step": 398
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.7227204783258594e-05,
+      "loss": 0.0231,
+      "step": 399
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.7219730941704037e-05,
+      "loss": 0.016,
+      "step": 400
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.721225710014948e-05,
+      "loss": 0.0096,
+      "step": 401
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.720478325859492e-05,
+      "loss": 0.0199,
+      "step": 402
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.719730941704036e-05,
+      "loss": 0.0155,
+      "step": 403
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.71898355754858e-05,
+      "loss": 0.0723,
+      "step": 404
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.718236173393124e-05,
+      "loss": 0.0126,
+      "step": 405
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.7174887892376683e-05,
+      "loss": 0.0065,
+      "step": 406
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.7167414050822125e-05,
+      "loss": 0.0155,
+      "step": 407
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.7159940209267564e-05,
+      "loss": 0.0112,
+      "step": 408
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.7152466367713007e-05,
+      "loss": 0.0134,
+      "step": 409
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.7144992526158446e-05,
+      "loss": 0.0664,
+      "step": 410
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.713751868460389e-05,
+      "loss": 0.0233,
+      "step": 411
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.7130044843049328e-05,
+      "loss": 0.0259,
+      "step": 412
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.712257100149477e-05,
+      "loss": 0.1194,
+      "step": 413
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.711509715994021e-05,
+      "loss": 0.0067,
+      "step": 414
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.710762331838565e-05,
+      "loss": 0.0146,
+      "step": 415
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.7100149476831092e-05,
+      "loss": 0.0164,
+      "step": 416
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.7092675635276535e-05,
+      "loss": 0.0465,
+      "step": 417
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.7085201793721974e-05,
+      "loss": 0.0421,
+      "step": 418
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.7077727952167417e-05,
+      "loss": 0.0049,
+      "step": 419
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.7070254110612856e-05,
+      "loss": 0.028,
+      "step": 420
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.7062780269058295e-05,
+      "loss": 0.0055,
+      "step": 421
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.7055306427503738e-05,
+      "loss": 0.0451,
+      "step": 422
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.704783258594918e-05,
+      "loss": 0.0283,
+      "step": 423
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.704035874439462e-05,
+      "loss": 0.0129,
+      "step": 424
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.7032884902840062e-05,
+      "loss": 0.0117,
+      "step": 425
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.70254110612855e-05,
+      "loss": 0.0129,
+      "step": 426
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.701793721973094e-05,
+      "loss": 0.0109,
+      "step": 427
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.7010463378176384e-05,
+      "loss": 0.0095,
+      "step": 428
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.7002989536621826e-05,
+      "loss": 0.0711,
+      "step": 429
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.6995515695067265e-05,
+      "loss": 0.0716,
+      "step": 430
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.6988041853512708e-05,
+      "loss": 0.0669,
+      "step": 431
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.6980568011958147e-05,
+      "loss": 0.0067,
+      "step": 432
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.6973094170403587e-05,
+      "loss": 0.0241,
+      "step": 433
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.696562032884903e-05,
+      "loss": 0.0068,
+      "step": 434
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.6958146487294472e-05,
+      "loss": 0.0297,
+      "step": 435
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.695067264573991e-05,
+      "loss": 0.0581,
+      "step": 436
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.6943198804185354e-05,
+      "loss": 0.0109,
+      "step": 437
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.6935724962630793e-05,
+      "loss": 0.0059,
+      "step": 438
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.6928251121076232e-05,
+      "loss": 0.0265,
+      "step": 439
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.6920777279521675e-05,
+      "loss": 0.0597,
+      "step": 440
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.6913303437967118e-05,
+      "loss": 0.0061,
+      "step": 441
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.6905829596412557e-05,
+      "loss": 0.0114,
+      "step": 442
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.6898355754858e-05,
+      "loss": 0.0123,
+      "step": 443
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.689088191330344e-05,
+      "loss": 0.0064,
+      "step": 444
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.688340807174888e-05,
+      "loss": 0.0107,
+      "step": 445
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.687593423019432e-05,
+      "loss": 0.0153,
+      "step": 446
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.6868460388639763e-05,
+      "loss": 0.0122,
+      "step": 447
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.6860986547085203e-05,
+      "loss": 0.0212,
+      "step": 448
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.6853512705530642e-05,
+      "loss": 0.0412,
+      "step": 449
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.6846038863976085e-05,
+      "loss": 0.187,
+      "step": 450
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.6838565022421527e-05,
+      "loss": 0.0641,
+      "step": 451
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.6831091180866966e-05,
+      "loss": 0.0902,
+      "step": 452
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.682361733931241e-05,
+      "loss": 0.0073,
+      "step": 453
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.681614349775785e-05,
+      "loss": 0.033,
+      "step": 454
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.6808669656203288e-05,
+      "loss": 0.0104,
+      "step": 455
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.680119581464873e-05,
+      "loss": 0.0122,
+      "step": 456
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.6793721973094173e-05,
+      "loss": 0.0131,
+      "step": 457
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.6786248131539612e-05,
+      "loss": 0.0673,
+      "step": 458
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.6778774289985055e-05,
+      "loss": 0.0271,
+      "step": 459
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.6771300448430494e-05,
+      "loss": 0.0143,
+      "step": 460
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.6763826606875933e-05,
+      "loss": 0.01,
+      "step": 461
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.6756352765321376e-05,
+      "loss": 0.0201,
+      "step": 462
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.674887892376682e-05,
+      "loss": 0.0056,
+      "step": 463
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.6741405082212258e-05,
+      "loss": 0.0618,
+      "step": 464
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.67339312406577e-05,
+      "loss": 0.0141,
+      "step": 465
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.672645739910314e-05,
+      "loss": 0.0118,
+      "step": 466
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.671898355754858e-05,
+      "loss": 0.0105,
+      "step": 467
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.6711509715994022e-05,
+      "loss": 0.0082,
+      "step": 468
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.6704035874439464e-05,
+      "loss": 0.0104,
+      "step": 469
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.6696562032884904e-05,
+      "loss": 0.03,
+      "step": 470
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.6689088191330346e-05,
+      "loss": 0.0099,
+      "step": 471
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.6681614349775786e-05,
+      "loss": 0.0157,
+      "step": 472
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.6674140508221225e-05,
+      "loss": 0.0101,
+      "step": 473
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 0.0048,
+      "step": 474
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.665919282511211e-05,
+      "loss": 0.0157,
+      "step": 475
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.665171898355755e-05,
+      "loss": 0.0271,
+      "step": 476
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.6644245142002992e-05,
+      "loss": 0.0232,
+      "step": 477
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.663677130044843e-05,
+      "loss": 0.0088,
+      "step": 478
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.6629297458893874e-05,
+      "loss": 0.0147,
+      "step": 479
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.6621823617339313e-05,
+      "loss": 0.0054,
+      "step": 480
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.6614349775784756e-05,
+      "loss": 0.0254,
+      "step": 481
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.6606875934230195e-05,
+      "loss": 0.0089,
+      "step": 482
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.6599402092675638e-05,
+      "loss": 0.0425,
+      "step": 483
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.6591928251121077e-05,
+      "loss": 0.0813,
+      "step": 484
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.658445440956652e-05,
+      "loss": 0.029,
+      "step": 485
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.657698056801196e-05,
+      "loss": 0.0117,
+      "step": 486
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.65695067264574e-05,
+      "loss": 0.0259,
+      "step": 487
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.656203288490284e-05,
+      "loss": 0.011,
+      "step": 488
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.655455904334828e-05,
+      "loss": 0.012,
+      "step": 489
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.6547085201793723e-05,
+      "loss": 0.0072,
+      "step": 490
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.6539611360239165e-05,
+      "loss": 0.0224,
+      "step": 491
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.6532137518684605e-05,
+      "loss": 0.0073,
+      "step": 492
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.6524663677130047e-05,
+      "loss": 0.0607,
+      "step": 493
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.6517189835575487e-05,
+      "loss": 0.0037,
+      "step": 494
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.6509715994020926e-05,
+      "loss": 0.0111,
+      "step": 495
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.650224215246637e-05,
+      "loss": 0.0187,
+      "step": 496
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.649476831091181e-05,
+      "loss": 0.0161,
+      "step": 497
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.648729446935725e-05,
+      "loss": 0.0065,
+      "step": 498
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.6479820627802693e-05,
+      "loss": 0.0091,
+      "step": 499
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.6472346786248132e-05,
+      "loss": 0.009,
+      "step": 500
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.646487294469357e-05,
+      "loss": 0.0109,
+      "step": 501
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.6457399103139014e-05,
+      "loss": 0.0117,
+      "step": 502
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.6449925261584457e-05,
+      "loss": 0.0344,
+      "step": 503
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.6442451420029896e-05,
+      "loss": 0.0193,
+      "step": 504
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.643497757847534e-05,
+      "loss": 0.0116,
+      "step": 505
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.6427503736920778e-05,
+      "loss": 0.0151,
+      "step": 506
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.6420029895366217e-05,
+      "loss": 0.0116,
+      "step": 507
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.641255605381166e-05,
+      "loss": 0.0198,
+      "step": 508
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.6405082212257103e-05,
+      "loss": 0.0079,
+      "step": 509
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.6397608370702542e-05,
+      "loss": 0.0096,
+      "step": 510
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.6390134529147984e-05,
+      "loss": 0.0225,
+      "step": 511
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.6382660687593424e-05,
+      "loss": 0.0403,
+      "step": 512
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.6375186846038866e-05,
+      "loss": 0.0078,
+      "step": 513
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.6367713004484306e-05,
+      "loss": 0.0018,
+      "step": 514
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.6360239162929748e-05,
+      "loss": 0.0154,
+      "step": 515
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.6352765321375188e-05,
+      "loss": 0.0192,
+      "step": 516
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.634529147982063e-05,
+      "loss": 0.0578,
+      "step": 517
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.633781763826607e-05,
+      "loss": 0.0641,
+      "step": 518
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.6330343796711512e-05,
+      "loss": 0.0141,
+      "step": 519
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.632286995515695e-05,
+      "loss": 0.0174,
+      "step": 520
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.6315396113602394e-05,
+      "loss": 0.0095,
+      "step": 521
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.6307922272047833e-05,
+      "loss": 0.0068,
+      "step": 522
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.6300448430493273e-05,
+      "loss": 0.0154,
+      "step": 523
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.6292974588938715e-05,
+      "loss": 0.0706,
+      "step": 524
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.6285500747384158e-05,
+      "loss": 0.0065,
+      "step": 525
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.6278026905829597e-05,
+      "loss": 0.0174,
+      "step": 526
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.627055306427504e-05,
+      "loss": 0.0121,
+      "step": 527
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.626307922272048e-05,
+      "loss": 0.0078,
+      "step": 528
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 1.6255605381165918e-05,
+      "loss": 0.0217,
+      "step": 529
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 1.624813153961136e-05,
+      "loss": 0.049,
+      "step": 530
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 1.6240657698056804e-05,
+      "loss": 0.0124,
+      "step": 531
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 1.6233183856502243e-05,
+      "loss": 0.0085,
+      "step": 532
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.6225710014947685e-05,
+      "loss": 0.0038,
+      "step": 533
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.6218236173393125e-05,
+      "loss": 0.0382,
+      "step": 534
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.6210762331838564e-05,
+      "loss": 0.0235,
+      "step": 535
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 1.6203288490284007e-05,
+      "loss": 0.0067,
+      "step": 536
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 1.619581464872945e-05,
+      "loss": 0.0496,
+      "step": 537
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 1.618834080717489e-05,
+      "loss": 0.0433,
+      "step": 538
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 1.618086696562033e-05,
+      "loss": 0.0158,
+      "step": 539
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.617339312406577e-05,
+      "loss": 0.0294,
+      "step": 540
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.616591928251121e-05,
+      "loss": 0.0128,
+      "step": 541
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.6158445440956652e-05,
+      "loss": 0.0215,
+      "step": 542
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.6150971599402095e-05,
+      "loss": 0.0075,
+      "step": 543
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.6143497757847534e-05,
+      "loss": 0.0061,
+      "step": 544
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.6136023916292977e-05,
+      "loss": 0.0123,
+      "step": 545
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 1.6128550074738416e-05,
+      "loss": 0.0613,
+      "step": 546
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 1.612107623318386e-05,
+      "loss": 0.0182,
+      "step": 547
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 1.6113602391629298e-05,
+      "loss": 0.0121,
+      "step": 548
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 1.610612855007474e-05,
+      "loss": 0.0114,
+      "step": 549
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.609865470852018e-05,
+      "loss": 0.0084,
+      "step": 550
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.6091180866965623e-05,
+      "loss": 0.0035,
+      "step": 551
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.6083707025411062e-05,
+      "loss": 0.0063,
+      "step": 552
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.6076233183856505e-05,
+      "loss": 0.0732,
+      "step": 553
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.6068759342301944e-05,
+      "loss": 0.0442,
+      "step": 554
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.6061285500747386e-05,
+      "loss": 0.0308,
+      "step": 555
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.6053811659192826e-05,
+      "loss": 0.008,
+      "step": 556
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 1.6046337817638265e-05,
+      "loss": 0.0068,
+      "step": 557
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 1.6038863976083708e-05,
+      "loss": 0.0074,
+      "step": 558
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 1.603139013452915e-05,
+      "loss": 0.0077,
+      "step": 559
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 1.602391629297459e-05,
+      "loss": 0.0037,
+      "step": 560
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 1.6016442451420032e-05,
+      "loss": 0.0081,
+      "step": 561
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 1.600896860986547e-05,
+      "loss": 0.0736,
+      "step": 562
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 1.600149476831091e-05,
+      "loss": 0.0096,
+      "step": 563
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 1.5994020926756357e-05,
+      "loss": 0.004,
+      "step": 564
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 1.5986547085201796e-05,
+      "loss": 0.0095,
+      "step": 565
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 1.5979073243647235e-05,
+      "loss": 0.0211,
+      "step": 566
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.5971599402092678e-05,
+      "loss": 0.0105,
+      "step": 567
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.5964125560538117e-05,
+      "loss": 0.0087,
+      "step": 568
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.5956651718983556e-05,
+      "loss": 0.0046,
+      "step": 569
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 1.5949177877429e-05,
+      "loss": 0.0112,
+      "step": 570
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 1.5941704035874442e-05,
+      "loss": 0.0809,
+      "step": 571
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 1.593423019431988e-05,
+      "loss": 0.0059,
+      "step": 572
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 1.5926756352765324e-05,
+      "loss": 0.0195,
+      "step": 573
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 1.5919282511210763e-05,
+      "loss": 0.0064,
+      "step": 574
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 1.5911808669656202e-05,
+      "loss": 0.0082,
+      "step": 575
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 1.5904334828101645e-05,
+      "loss": 0.0197,
+      "step": 576
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.5896860986547088e-05,
+      "loss": 0.034,
+      "step": 577
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.5889387144992527e-05,
+      "loss": 0.0088,
+      "step": 578
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.588191330343797e-05,
+      "loss": 0.0076,
+      "step": 579
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 1.587443946188341e-05,
+      "loss": 0.0169,
+      "step": 580
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 1.586696562032885e-05,
+      "loss": 0.0064,
+      "step": 581
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 1.585949177877429e-05,
+      "loss": 0.0474,
+      "step": 582
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 1.5852017937219733e-05,
+      "loss": 0.0137,
+      "step": 583
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 1.5844544095665172e-05,
+      "loss": 0.0142,
+      "step": 584
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 1.5837070254110615e-05,
+      "loss": 0.0393,
+      "step": 585
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 1.5829596412556054e-05,
+      "loss": 0.008,
+      "step": 586
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.5822122571001497e-05,
+      "loss": 0.0087,
+      "step": 587
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.5814648729446936e-05,
+      "loss": 0.0078,
+      "step": 588
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.580717488789238e-05,
+      "loss": 0.0067,
+      "step": 589
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 1.5799701046337818e-05,
+      "loss": 0.0254,
+      "step": 590
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 1.579222720478326e-05,
+      "loss": 0.0087,
+      "step": 591
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 1.57847533632287e-05,
+      "loss": 0.1282,
+      "step": 592
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 1.5777279521674143e-05,
+      "loss": 0.0153,
+      "step": 593
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 1.5769805680119582e-05,
+      "loss": 0.0102,
+      "step": 594
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 1.5762331838565025e-05,
+      "loss": 0.0258,
+      "step": 595
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 1.5754857997010464e-05,
+      "loss": 0.0055,
+      "step": 596
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.5747384155455903e-05,
+      "loss": 0.008,
+      "step": 597
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.573991031390135e-05,
+      "loss": 0.0115,
+      "step": 598
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.573243647234679e-05,
+      "loss": 0.0089,
+      "step": 599
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 1.5724962630792228e-05,
+      "loss": 0.0061,
+      "step": 600
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 1.571748878923767e-05,
+      "loss": 0.0377,
+      "step": 601
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 1.571001494768311e-05,
+      "loss": 0.0102,
+      "step": 602
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 1.570254110612855e-05,
+      "loss": 0.0156,
+      "step": 603
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 1.569506726457399e-05,
+      "loss": 0.0353,
+      "step": 604
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 1.5687593423019434e-05,
+      "loss": 0.0088,
+      "step": 605
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 1.5680119581464873e-05,
+      "loss": 0.006,
+      "step": 606
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.5672645739910316e-05,
+      "loss": 0.0198,
+      "step": 607
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.5665171898355755e-05,
+      "loss": 0.0117,
+      "step": 608
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.5657698056801195e-05,
+      "loss": 0.0069,
+      "step": 609
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.5650224215246637e-05,
+      "loss": 0.0097,
+      "step": 610
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.564275037369208e-05,
+      "loss": 0.0097,
+      "step": 611
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.563527653213752e-05,
+      "loss": 0.0127,
+      "step": 612
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.5627802690582962e-05,
+      "loss": 0.0243,
+      "step": 613
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.56203288490284e-05,
+      "loss": 0.0062,
+      "step": 614
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.5612855007473844e-05,
+      "loss": 0.0056,
+      "step": 615
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.5605381165919283e-05,
+      "loss": 0.008,
+      "step": 616
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 1.5597907324364726e-05,
+      "loss": 0.0081,
+      "step": 617
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 1.5590433482810165e-05,
+      "loss": 0.0115,
+      "step": 618
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 1.5582959641255608e-05,
+      "loss": 0.0615,
+      "step": 619
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 1.5575485799701047e-05,
+      "loss": 0.0088,
+      "step": 620
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.556801195814649e-05,
+      "loss": 0.0067,
+      "step": 621
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.556053811659193e-05,
+      "loss": 0.0072,
+      "step": 622
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.555306427503737e-05,
+      "loss": 0.0066,
+      "step": 623
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 1.554559043348281e-05,
+      "loss": 0.0052,
+      "step": 624
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 1.5538116591928253e-05,
+      "loss": 0.0301,
+      "step": 625
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 1.5530642750373693e-05,
+      "loss": 0.0171,
+      "step": 626
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 1.5523168908819135e-05,
+      "loss": 0.0083,
+      "step": 627
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 1.5515695067264575e-05,
+      "loss": 0.0138,
+      "step": 628
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 1.5508221225710017e-05,
+      "loss": 0.0092,
+      "step": 629
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 1.5500747384155456e-05,
+      "loss": 0.0047,
+      "step": 630
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.5493273542600896e-05,
+      "loss": 0.0235,
+      "step": 631
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.5485799701046342e-05,
+      "loss": 0.0481,
+      "step": 632
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.547832585949178e-05,
+      "loss": 0.0129,
+      "step": 633
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 1.547085201793722e-05,
+      "loss": 0.013,
+      "step": 634
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 1.5463378176382663e-05,
+      "loss": 0.0256,
+      "step": 635
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 1.5455904334828102e-05,
+      "loss": 0.0046,
+      "step": 636
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 1.544843049327354e-05,
+      "loss": 0.0236,
+      "step": 637
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 1.5440956651718984e-05,
+      "loss": 0.0058,
+      "step": 638
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 1.5433482810164427e-05,
+      "loss": 0.0123,
+      "step": 639
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 1.5426008968609866e-05,
+      "loss": 0.0049,
+      "step": 640
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.541853512705531e-05,
+      "loss": 0.009,
+      "step": 641
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.5411061285500748e-05,
+      "loss": 0.0156,
+      "step": 642
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.5403587443946187e-05,
+      "loss": 0.0026,
+      "step": 643
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 1.539611360239163e-05,
+      "loss": 0.0942,
+      "step": 644
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 1.5388639760837072e-05,
+      "loss": 0.0079,
+      "step": 645
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 1.538116591928251e-05,
+      "loss": 0.0091,
+      "step": 646
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 1.5373692077727954e-05,
+      "loss": 0.009,
+      "step": 647
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 1.5366218236173394e-05,
+      "loss": 0.0098,
+      "step": 648
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 1.5358744394618836e-05,
+      "loss": 0.0077,
+      "step": 649
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 1.5351270553064276e-05,
+      "loss": 0.0083,
+      "step": 650
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.5343796711509718e-05,
+      "loss": 0.0068,
+      "step": 651
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.5336322869955157e-05,
+      "loss": 0.0117,
+      "step": 652
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.53288490284006e-05,
+      "loss": 0.008,
+      "step": 653
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.532137518684604e-05,
+      "loss": 0.0177,
+      "step": 654
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 1.5313901345291482e-05,
+      "loss": 0.0041,
+      "step": 655
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 1.530642750373692e-05,
+      "loss": 0.0104,
+      "step": 656
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 1.5298953662182364e-05,
+      "loss": 0.0213,
+      "step": 657
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 1.5291479820627803e-05,
+      "loss": 0.0491,
+      "step": 658
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 1.5284005979073246e-05,
+      "loss": 0.0739,
+      "step": 659
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 1.5276532137518685e-05,
+      "loss": 0.0083,
+      "step": 660
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 1.5269058295964128e-05,
+      "loss": 0.0078,
+      "step": 661
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 1.5261584454409567e-05,
+      "loss": 0.0103,
+      "step": 662
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 1.5254110612855008e-05,
+      "loss": 0.0163,
+      "step": 663
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 1.5246636771300449e-05,
+      "loss": 0.0066,
+      "step": 664
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 1.523916292974589e-05,
+      "loss": 0.0083,
+      "step": 665
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 1.5231689088191332e-05,
+      "loss": 0.0195,
+      "step": 666
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 1.5224215246636773e-05,
+      "loss": 0.0068,
+      "step": 667
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 1.5216741405082214e-05,
+      "loss": 0.008,
+      "step": 668
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 1.5209267563527654e-05,
+      "loss": 0.1062,
+      "step": 669
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 1.5201793721973095e-05,
+      "loss": 0.0096,
+      "step": 670
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 1.5194319880418536e-05,
+      "loss": 0.0138,
+      "step": 671
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 1.5186846038863978e-05,
+      "loss": 0.15,
+      "step": 672
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 1.517937219730942e-05,
+      "loss": 0.0048,
+      "step": 673
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 1.517189835575486e-05,
+      "loss": 0.0037,
+      "step": 674
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.51644245142003e-05,
+      "loss": 0.0078,
+      "step": 675
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.515695067264574e-05,
+      "loss": 0.0067,
+      "step": 676
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 2704,
+  "num_train_epochs": 8,
+  "save_steps": 500,
+  "total_flos": 8.565880284061368e+17,
+  "trial_name": null,
+  "trial_params": null
+}