Training in progress, step 300

.gitattributes

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

config.json

@@ -0,0 +1,36 @@
+{
+  "_name_or_path": "hishab/llama-3.2-3B-extended-4096-full",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 170241,
+  "eos_token_id": 170250,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 3072,
+  "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "max_position_embeddings": 131072,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 24,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 8,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": {
+    "factor": 32.0,
+    "high_freq_factor": 4.0,
+    "low_freq_factor": 1.0,
+    "original_max_position_embeddings": 8192,
+    "rope_type": "llama3"
+  },
+  "rope_theta": 500000.0,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.45.2",
+  "use_cache": false,
+  "vocab_size": 170497
+}

model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8800ba9e04ae12c86d3fb5b3b0604e74450d71241e0afe7642b9efd68dfef30a
+size 4973656136

model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7838a563459e8e8fa56d1a04c91db213314ecc7b80eba784a50082786bdee70
+size 1711401640

model.safetensors.index.json

@@ -0,0 +1,261 @@
+{
+  "metadata": {
+    "total_size": 6685028352
+  },
+  "weight_map": {
+    "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.norm.weight": "model-00002-of-00002.safetensors"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,23 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:365a0c265893f9ab04f4094f07a13a399b024ab4fc75aadecd0133570d543a6f
+size 29364254

tokenizer_config.json

@@ -0,0 +1,2065 @@
+{
+  "added_tokens_decoder": {
+    "170241": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170242": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170243": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170244": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170245": {
+      "content": "<|finetune_right_pad_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170246": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170247": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170248": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170249": {
+      "content": "<|eom_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170250": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170251": {
+      "content": "<|python_tag|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170252": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170253": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170254": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170255": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170256": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170257": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170258": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170259": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170260": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170261": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170262": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170263": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170264": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170265": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170266": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170267": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170268": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170269": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170270": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170271": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170272": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170273": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170274": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170275": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170276": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170277": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170278": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170279": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170280": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170281": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170282": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170283": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170284": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170285": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170286": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170287": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170288": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170289": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170290": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170291": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170292": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170293": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170294": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170295": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170296": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170297": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170298": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170299": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170300": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170301": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170302": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170303": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170304": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170305": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170306": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170307": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170308": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170309": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170310": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170311": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170312": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170313": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170314": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170315": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170316": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170317": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170318": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170319": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170320": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170321": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170322": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170323": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170324": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170325": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170326": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170327": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170328": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170329": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170330": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170331": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170332": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170333": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170334": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170335": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170336": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170337": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170338": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170339": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170340": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170341": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170342": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170343": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170344": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170345": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170346": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170347": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170348": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170349": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170350": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170351": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170352": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170353": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170354": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170355": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170356": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170357": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170358": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170359": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170360": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170361": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170362": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170363": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170364": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170365": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170366": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170367": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170368": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170369": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170370": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170371": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170372": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170373": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170374": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170375": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170376": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170377": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170378": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170379": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170380": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170381": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170382": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170383": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170384": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170385": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170386": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170387": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170388": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170389": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170390": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170391": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170392": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170393": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170394": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170395": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170396": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170397": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170398": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170399": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170400": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170401": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170402": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170403": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170404": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170405": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170406": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170407": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170408": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170409": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170410": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170411": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170412": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170413": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170414": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170415": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170416": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170417": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170418": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170419": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170420": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170421": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170422": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170423": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170424": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170425": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170426": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170427": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170428": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170429": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170430": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170431": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170432": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170433": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170434": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170435": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170436": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170437": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170438": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170439": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170440": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170441": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170442": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170443": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170444": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170445": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170446": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170447": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170448": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170449": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170450": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170451": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170452": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170453": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170454": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170455": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170456": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170457": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170458": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170459": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170460": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170461": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170462": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170463": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170464": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170465": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170466": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170467": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170468": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170469": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170470": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170471": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170472": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170473": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170474": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170475": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170476": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170477": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170478": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170479": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170480": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170481": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170482": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170483": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170484": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170485": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170486": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170487": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170488": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170489": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170490": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170491": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170492": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170493": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170494": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170495": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170496": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content }}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 131072,
+  "pad_token": "<|eot_id|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}

trainer_log.jsonl

@@ -0,0 +1,312 @@
+{"current_steps": 1, "total_steps": 34894, "loss": 2.3091, "learning_rate": 1.1461318051575931e-07, "epoch": 2.865811938256082e-05, "percentage": 0.0, "elapsed_time": "0:00:22", "remaining_time": "9 days, 2:11:27"}
+{"current_steps": 2, "total_steps": 34894, "loss": 2.332, "learning_rate": 2.2922636103151863e-07, "epoch": 5.731623876512164e-05, "percentage": 0.01, "elapsed_time": "0:00:36", "remaining_time": "7 days, 9:50:27"}
+{"current_steps": 3, "total_steps": 34894, "loss": 2.3238, "learning_rate": 3.4383954154727796e-07, "epoch": 8.597435814768245e-05, "percentage": 0.01, "elapsed_time": "0:00:50", "remaining_time": "6 days, 20:14:00"}
+{"current_steps": 4, "total_steps": 34894, "loss": 2.2763, "learning_rate": 4.5845272206303726e-07, "epoch": 0.00011463247753024328, "percentage": 0.01, "elapsed_time": "0:01:04", "remaining_time": "6 days, 13:08:32"}
+{"current_steps": 5, "total_steps": 34894, "loss": 2.3563, "learning_rate": 5.730659025787966e-07, "epoch": 0.0001432905969128041, "percentage": 0.01, "elapsed_time": "0:01:19", "remaining_time": "6 days, 9:13:55"}
+{"current_steps": 6, "total_steps": 34894, "loss": 2.357, "learning_rate": 6.876790830945559e-07, "epoch": 0.0001719487162953649, "percentage": 0.02, "elapsed_time": "0:01:33", "remaining_time": "6 days, 6:37:26"}
+{"current_steps": 7, "total_steps": 34894, "loss": 2.378, "learning_rate": 8.022922636103153e-07, "epoch": 0.00020060683567792571, "percentage": 0.02, "elapsed_time": "0:01:47", "remaining_time": "6 days, 4:37:56"}
+{"current_steps": 8, "total_steps": 34894, "loss": 2.3353, "learning_rate": 9.169054441260745e-07, "epoch": 0.00022926495506048655, "percentage": 0.02, "elapsed_time": "0:02:01", "remaining_time": "6 days, 3:02:51"}
+{"current_steps": 9, "total_steps": 34894, "loss": 2.3412, "learning_rate": 1.031518624641834e-06, "epoch": 0.00025792307444304734, "percentage": 0.03, "elapsed_time": "0:02:15", "remaining_time": "6 days, 1:47:23"}
+{"current_steps": 10, "total_steps": 34894, "loss": 2.3177, "learning_rate": 1.1461318051575932e-06, "epoch": 0.0002865811938256082, "percentage": 0.03, "elapsed_time": "0:02:29", "remaining_time": "6 days, 0:51:45"}
+{"current_steps": 11, "total_steps": 34894, "loss": 2.3157, "learning_rate": 1.2607449856733525e-06, "epoch": 0.000315239313208169, "percentage": 0.03, "elapsed_time": "0:02:43", "remaining_time": "6 days, 0:21:50"}
+{"current_steps": 12, "total_steps": 34894, "loss": 2.3448, "learning_rate": 1.3753581661891118e-06, "epoch": 0.0003438974325907298, "percentage": 0.03, "elapsed_time": "0:02:58", "remaining_time": "5 days, 23:52:20"}
+{"current_steps": 13, "total_steps": 34894, "loss": 2.3315, "learning_rate": 1.4899713467048713e-06, "epoch": 0.00037255555197329064, "percentage": 0.04, "elapsed_time": "0:03:12", "remaining_time": "5 days, 23:29:03"}
+{"current_steps": 14, "total_steps": 34894, "loss": 2.2923, "learning_rate": 1.6045845272206306e-06, "epoch": 0.00040121367135585143, "percentage": 0.04, "elapsed_time": "0:03:26", "remaining_time": "5 days, 23:05:13"}
+{"current_steps": 15, "total_steps": 34894, "loss": 2.3185, "learning_rate": 1.7191977077363897e-06, "epoch": 0.00042987179073841227, "percentage": 0.04, "elapsed_time": "0:03:40", "remaining_time": "5 days, 22:36:37"}
+{"current_steps": 16, "total_steps": 34894, "loss": 2.3466, "learning_rate": 1.833810888252149e-06, "epoch": 0.0004585299101209731, "percentage": 0.05, "elapsed_time": "0:03:54", "remaining_time": "5 days, 22:12:03"}
+{"current_steps": 17, "total_steps": 34894, "loss": 2.2327, "learning_rate": 1.9484240687679085e-06, "epoch": 0.0004871880295035339, "percentage": 0.05, "elapsed_time": "0:04:09", "remaining_time": "5 days, 22:01:13"}
+{"current_steps": 18, "total_steps": 34894, "loss": 2.2944, "learning_rate": 2.063037249283668e-06, "epoch": 0.0005158461488860947, "percentage": 0.05, "elapsed_time": "0:04:23", "remaining_time": "5 days, 21:46:42"}
+{"current_steps": 19, "total_steps": 34894, "loss": 2.3304, "learning_rate": 2.177650429799427e-06, "epoch": 0.0005445042682686556, "percentage": 0.05, "elapsed_time": "0:04:37", "remaining_time": "5 days, 21:28:06"}
+{"current_steps": 20, "total_steps": 34894, "loss": 2.3839, "learning_rate": 2.2922636103151864e-06, "epoch": 0.0005731623876512164, "percentage": 0.06, "elapsed_time": "0:04:51", "remaining_time": "5 days, 21:11:28"}
+{"current_steps": 21, "total_steps": 34894, "loss": 2.3639, "learning_rate": 2.4068767908309457e-06, "epoch": 0.0006018205070337771, "percentage": 0.06, "elapsed_time": "0:05:05", "remaining_time": "5 days, 20:56:22"}
+{"current_steps": 22, "total_steps": 34894, "loss": 2.3418, "learning_rate": 2.521489971346705e-06, "epoch": 0.000630478626416338, "percentage": 0.06, "elapsed_time": "0:05:19", "remaining_time": "5 days, 20:42:10"}
+{"current_steps": 23, "total_steps": 34894, "loss": 2.2913, "learning_rate": 2.636103151862464e-06, "epoch": 0.0006591367457988988, "percentage": 0.07, "elapsed_time": "0:05:33", "remaining_time": "5 days, 20:29:53"}
+{"current_steps": 24, "total_steps": 34894, "loss": 2.3362, "learning_rate": 2.7507163323782236e-06, "epoch": 0.0006877948651814596, "percentage": 0.07, "elapsed_time": "0:05:47", "remaining_time": "5 days, 20:18:12"}
+{"current_steps": 25, "total_steps": 34894, "loss": 2.3346, "learning_rate": 2.865329512893983e-06, "epoch": 0.0007164529845640205, "percentage": 0.07, "elapsed_time": "0:06:01", "remaining_time": "5 days, 20:06:53"}
+{"current_steps": 26, "total_steps": 34894, "loss": 2.3125, "learning_rate": 2.9799426934097427e-06, "epoch": 0.0007451111039465813, "percentage": 0.07, "elapsed_time": "0:06:15", "remaining_time": "5 days, 19:56:29"}
+{"current_steps": 27, "total_steps": 34894, "loss": 2.3354, "learning_rate": 3.0945558739255016e-06, "epoch": 0.0007737692233291421, "percentage": 0.08, "elapsed_time": "0:06:29", "remaining_time": "5 days, 19:48:45"}
+{"current_steps": 28, "total_steps": 34894, "loss": 2.3419, "learning_rate": 3.2091690544412613e-06, "epoch": 0.0008024273427117029, "percentage": 0.08, "elapsed_time": "0:06:43", "remaining_time": "5 days, 19:41:31"}
+{"current_steps": 29, "total_steps": 34894, "loss": 2.3596, "learning_rate": 3.32378223495702e-06, "epoch": 0.0008310854620942637, "percentage": 0.08, "elapsed_time": "0:06:57", "remaining_time": "5 days, 19:33:45"}
+{"current_steps": 30, "total_steps": 34894, "loss": 2.332, "learning_rate": 3.4383954154727795e-06, "epoch": 0.0008597435814768245, "percentage": 0.09, "elapsed_time": "0:07:11", "remaining_time": "5 days, 19:26:17"}
+{"current_steps": 31, "total_steps": 34894, "loss": 2.2682, "learning_rate": 3.553008595988539e-06, "epoch": 0.0008884017008593853, "percentage": 0.09, "elapsed_time": "0:07:25", "remaining_time": "5 days, 19:19:29"}
+{"current_steps": 32, "total_steps": 34894, "loss": 2.3083, "learning_rate": 3.667621776504298e-06, "epoch": 0.0009170598202419462, "percentage": 0.09, "elapsed_time": "0:07:40", "remaining_time": "5 days, 19:12:27"}
+{"current_steps": 33, "total_steps": 34894, "loss": 2.3466, "learning_rate": 3.7822349570200578e-06, "epoch": 0.000945717939624507, "percentage": 0.09, "elapsed_time": "0:07:54", "remaining_time": "5 days, 19:07:23"}
+{"current_steps": 34, "total_steps": 34894, "loss": 2.3218, "learning_rate": 3.896848137535817e-06, "epoch": 0.0009743760590070678, "percentage": 0.1, "elapsed_time": "0:08:08", "remaining_time": "5 days, 19:01:38"}
+{"current_steps": 35, "total_steps": 34894, "loss": 2.3585, "learning_rate": 4.011461318051576e-06, "epoch": 0.0010030341783896287, "percentage": 0.1, "elapsed_time": "0:08:22", "remaining_time": "5 days, 18:55:39"}
+{"current_steps": 36, "total_steps": 34894, "loss": 2.3243, "learning_rate": 4.126074498567336e-06, "epoch": 0.0010316922977721894, "percentage": 0.1, "elapsed_time": "0:08:36", "remaining_time": "5 days, 18:50:11"}
+{"current_steps": 37, "total_steps": 34894, "loss": 2.3772, "learning_rate": 4.2406876790830946e-06, "epoch": 0.0010603504171547502, "percentage": 0.11, "elapsed_time": "0:08:50", "remaining_time": "5 days, 18:45:41"}
+{"current_steps": 38, "total_steps": 34894, "loss": 2.3944, "learning_rate": 4.355300859598854e-06, "epoch": 0.0010890085365373111, "percentage": 0.11, "elapsed_time": "0:09:04", "remaining_time": "5 days, 18:40:48"}
+{"current_steps": 39, "total_steps": 34894, "loss": 2.3012, "learning_rate": 4.469914040114613e-06, "epoch": 0.0011176666559198718, "percentage": 0.11, "elapsed_time": "0:09:18", "remaining_time": "5 days, 18:36:15"}
+{"current_steps": 40, "total_steps": 34894, "loss": 2.288, "learning_rate": 4.584527220630373e-06, "epoch": 0.0011463247753024327, "percentage": 0.11, "elapsed_time": "0:09:32", "remaining_time": "5 days, 18:33:08"}
+{"current_steps": 41, "total_steps": 34894, "loss": 2.3772, "learning_rate": 4.699140401146133e-06, "epoch": 0.0011749828946849936, "percentage": 0.12, "elapsed_time": "0:09:46", "remaining_time": "5 days, 18:28:40"}
+{"current_steps": 42, "total_steps": 34894, "loss": 2.3394, "learning_rate": 4.8137535816618915e-06, "epoch": 0.0012036410140675543, "percentage": 0.12, "elapsed_time": "0:10:00", "remaining_time": "5 days, 18:24:42"}
+{"current_steps": 43, "total_steps": 34894, "loss": 2.2805, "learning_rate": 4.928366762177651e-06, "epoch": 0.0012322991334501152, "percentage": 0.12, "elapsed_time": "0:10:14", "remaining_time": "5 days, 18:21:24"}
+{"current_steps": 44, "total_steps": 34894, "loss": 2.3854, "learning_rate": 5.04297994269341e-06, "epoch": 0.001260957252832676, "percentage": 0.13, "elapsed_time": "0:10:28", "remaining_time": "5 days, 18:17:42"}
+{"current_steps": 45, "total_steps": 34894, "loss": 2.33, "learning_rate": 5.157593123209169e-06, "epoch": 0.0012896153722152367, "percentage": 0.13, "elapsed_time": "0:10:42", "remaining_time": "5 days, 18:15:14"}
+{"current_steps": 46, "total_steps": 34894, "loss": 2.3382, "learning_rate": 5.272206303724928e-06, "epoch": 0.0013182734915977976, "percentage": 0.13, "elapsed_time": "0:10:57", "remaining_time": "5 days, 18:15:27"}
+{"current_steps": 47, "total_steps": 34894, "loss": 2.292, "learning_rate": 5.386819484240688e-06, "epoch": 0.0013469316109803585, "percentage": 0.13, "elapsed_time": "0:11:11", "remaining_time": "5 days, 18:12:35"}
+{"current_steps": 48, "total_steps": 34894, "loss": 2.366, "learning_rate": 5.501432664756447e-06, "epoch": 0.0013755897303629192, "percentage": 0.14, "elapsed_time": "0:11:25", "remaining_time": "5 days, 18:09:26"}
+{"current_steps": 49, "total_steps": 34894, "loss": 2.3514, "learning_rate": 5.616045845272207e-06, "epoch": 0.00140424784974548, "percentage": 0.14, "elapsed_time": "0:11:39", "remaining_time": "5 days, 18:06:08"}
+{"current_steps": 50, "total_steps": 34894, "loss": 2.3204, "learning_rate": 5.730659025787966e-06, "epoch": 0.001432905969128041, "percentage": 0.14, "elapsed_time": "0:11:53", "remaining_time": "5 days, 18:03:12"}
+{"current_steps": 51, "total_steps": 34894, "loss": 2.309, "learning_rate": 5.845272206303726e-06, "epoch": 0.0014615640885106017, "percentage": 0.15, "elapsed_time": "0:12:07", "remaining_time": "5 days, 18:01:57"}
+{"current_steps": 52, "total_steps": 34894, "loss": 2.3611, "learning_rate": 5.959885386819485e-06, "epoch": 0.0014902222078931626, "percentage": 0.15, "elapsed_time": "0:12:21", "remaining_time": "5 days, 18:00:54"}
+{"current_steps": 53, "total_steps": 34894, "loss": 2.2942, "learning_rate": 6.074498567335244e-06, "epoch": 0.0015188803272757232, "percentage": 0.15, "elapsed_time": "0:12:35", "remaining_time": "5 days, 17:58:45"}
+{"current_steps": 54, "total_steps": 34894, "loss": 2.3292, "learning_rate": 6.189111747851003e-06, "epoch": 0.0015475384466582841, "percentage": 0.15, "elapsed_time": "0:12:49", "remaining_time": "5 days, 17:56:11"}
+{"current_steps": 55, "total_steps": 34894, "loss": 2.3859, "learning_rate": 6.303724928366762e-06, "epoch": 0.001576196566040845, "percentage": 0.16, "elapsed_time": "0:13:03", "remaining_time": "5 days, 17:53:40"}
+{"current_steps": 56, "total_steps": 34894, "loss": 2.3546, "learning_rate": 6.4183381088825226e-06, "epoch": 0.0016048546854234057, "percentage": 0.16, "elapsed_time": "0:13:17", "remaining_time": "5 days, 17:51:17"}
+{"current_steps": 57, "total_steps": 34894, "loss": 2.3615, "learning_rate": 6.5329512893982814e-06, "epoch": 0.0016335128048059666, "percentage": 0.16, "elapsed_time": "0:13:31", "remaining_time": "5 days, 17:48:40"}
+{"current_steps": 58, "total_steps": 34894, "loss": 2.2999, "learning_rate": 6.64756446991404e-06, "epoch": 0.0016621709241885275, "percentage": 0.17, "elapsed_time": "0:13:45", "remaining_time": "5 days, 17:46:21"}
+{"current_steps": 59, "total_steps": 34894, "loss": 2.3222, "learning_rate": 6.7621776504298e-06, "epoch": 0.0016908290435710882, "percentage": 0.17, "elapsed_time": "0:13:59", "remaining_time": "5 days, 17:44:05"}
+{"current_steps": 60, "total_steps": 34894, "loss": 2.3308, "learning_rate": 6.876790830945559e-06, "epoch": 0.001719487162953649, "percentage": 0.17, "elapsed_time": "0:14:13", "remaining_time": "5 days, 17:41:51"}
+{"current_steps": 61, "total_steps": 34894, "loss": 2.4213, "learning_rate": 6.9914040114613195e-06, "epoch": 0.00174814528233621, "percentage": 0.17, "elapsed_time": "0:14:27", "remaining_time": "5 days, 17:39:47"}
+{"current_steps": 62, "total_steps": 34894, "loss": 2.3196, "learning_rate": 7.106017191977078e-06, "epoch": 0.0017768034017187706, "percentage": 0.18, "elapsed_time": "0:14:41", "remaining_time": "5 days, 17:37:41"}
+{"current_steps": 63, "total_steps": 34894, "loss": 2.2865, "learning_rate": 7.220630372492837e-06, "epoch": 0.0018054615211013315, "percentage": 0.18, "elapsed_time": "0:14:55", "remaining_time": "5 days, 17:35:57"}
+{"current_steps": 64, "total_steps": 34894, "loss": 2.3665, "learning_rate": 7.335243553008596e-06, "epoch": 0.0018341196404838924, "percentage": 0.18, "elapsed_time": "0:15:09", "remaining_time": "5 days, 17:33:53"}
+{"current_steps": 65, "total_steps": 34894, "loss": 2.3233, "learning_rate": 7.449856733524356e-06, "epoch": 0.001862777759866453, "percentage": 0.19, "elapsed_time": "0:15:24", "remaining_time": "5 days, 17:32:15"}
+{"current_steps": 66, "total_steps": 34894, "loss": 2.3218, "learning_rate": 7.5644699140401156e-06, "epoch": 0.001891435879249014, "percentage": 0.19, "elapsed_time": "0:15:38", "remaining_time": "5 days, 17:30:23"}
+{"current_steps": 67, "total_steps": 34894, "loss": 2.3197, "learning_rate": 7.679083094555874e-06, "epoch": 0.0019200939986315749, "percentage": 0.19, "elapsed_time": "0:15:52", "remaining_time": "5 days, 17:28:29"}
+{"current_steps": 68, "total_steps": 34894, "loss": 2.3065, "learning_rate": 7.793696275071634e-06, "epoch": 0.0019487521180141356, "percentage": 0.19, "elapsed_time": "0:16:06", "remaining_time": "5 days, 17:26:45"}
+{"current_steps": 69, "total_steps": 34894, "loss": 2.2595, "learning_rate": 7.908309455587392e-06, "epoch": 0.0019774102373966962, "percentage": 0.2, "elapsed_time": "0:16:20", "remaining_time": "5 days, 17:25:06"}
+{"current_steps": 70, "total_steps": 34894, "loss": 2.3716, "learning_rate": 8.022922636103152e-06, "epoch": 0.0020060683567792574, "percentage": 0.2, "elapsed_time": "0:16:34", "remaining_time": "5 days, 17:23:35"}
+{"current_steps": 71, "total_steps": 34894, "loss": 2.3077, "learning_rate": 8.137535816618912e-06, "epoch": 0.002034726476161818, "percentage": 0.2, "elapsed_time": "0:16:48", "remaining_time": "5 days, 17:21:54"}
+{"current_steps": 72, "total_steps": 34894, "loss": 2.3497, "learning_rate": 8.252148997134671e-06, "epoch": 0.0020633845955443787, "percentage": 0.21, "elapsed_time": "0:17:02", "remaining_time": "5 days, 17:20:20"}
+{"current_steps": 73, "total_steps": 34894, "loss": 2.3429, "learning_rate": 8.366762177650431e-06, "epoch": 0.00209204271492694, "percentage": 0.21, "elapsed_time": "0:17:16", "remaining_time": "5 days, 17:18:47"}
+{"current_steps": 74, "total_steps": 34894, "loss": 2.3302, "learning_rate": 8.481375358166189e-06, "epoch": 0.0021207008343095005, "percentage": 0.21, "elapsed_time": "0:17:30", "remaining_time": "5 days, 17:17:14"}
+{"current_steps": 75, "total_steps": 34894, "loss": 2.3187, "learning_rate": 8.595988538681949e-06, "epoch": 0.002149358953692061, "percentage": 0.21, "elapsed_time": "0:17:44", "remaining_time": "5 days, 17:16:18"}
+{"current_steps": 76, "total_steps": 34894, "loss": 2.3239, "learning_rate": 8.710601719197709e-06, "epoch": 0.0021780170730746223, "percentage": 0.22, "elapsed_time": "0:17:58", "remaining_time": "5 days, 17:14:59"}
+{"current_steps": 77, "total_steps": 34894, "loss": 2.3371, "learning_rate": 8.825214899713468e-06, "epoch": 0.002206675192457183, "percentage": 0.22, "elapsed_time": "0:18:12", "remaining_time": "5 days, 17:13:27"}
+{"current_steps": 78, "total_steps": 34894, "loss": 2.3339, "learning_rate": 8.939828080229226e-06, "epoch": 0.0022353333118397436, "percentage": 0.22, "elapsed_time": "0:18:26", "remaining_time": "5 days, 17:12:28"}
+{"current_steps": 79, "total_steps": 34894, "loss": 2.3342, "learning_rate": 9.054441260744986e-06, "epoch": 0.0022639914312223047, "percentage": 0.23, "elapsed_time": "0:18:40", "remaining_time": "5 days, 17:11:08"}
+{"current_steps": 80, "total_steps": 34894, "loss": 2.3604, "learning_rate": 9.169054441260746e-06, "epoch": 0.0022926495506048654, "percentage": 0.23, "elapsed_time": "0:18:54", "remaining_time": "5 days, 17:09:50"}
+{"current_steps": 81, "total_steps": 34894, "loss": 2.2879, "learning_rate": 9.283667621776506e-06, "epoch": 0.002321307669987426, "percentage": 0.23, "elapsed_time": "0:19:08", "remaining_time": "5 days, 17:08:32"}
+{"current_steps": 82, "total_steps": 34894, "loss": 2.3336, "learning_rate": 9.398280802292265e-06, "epoch": 0.002349965789369987, "percentage": 0.23, "elapsed_time": "0:19:22", "remaining_time": "5 days, 17:07:44"}
+{"current_steps": 83, "total_steps": 34894, "loss": 2.389, "learning_rate": 9.512893982808023e-06, "epoch": 0.002378623908752548, "percentage": 0.24, "elapsed_time": "0:19:36", "remaining_time": "5 days, 17:06:51"}
+{"current_steps": 84, "total_steps": 34894, "loss": 2.384, "learning_rate": 9.627507163323783e-06, "epoch": 0.0024072820281351086, "percentage": 0.24, "elapsed_time": "0:19:51", "remaining_time": "5 days, 17:06:29"}
+{"current_steps": 85, "total_steps": 34894, "loss": 2.3745, "learning_rate": 9.742120343839543e-06, "epoch": 0.0024359401475176697, "percentage": 0.24, "elapsed_time": "0:20:05", "remaining_time": "5 days, 17:05:41"}
+{"current_steps": 86, "total_steps": 34894, "loss": 2.2846, "learning_rate": 9.856733524355302e-06, "epoch": 0.0024645982669002304, "percentage": 0.25, "elapsed_time": "0:20:19", "remaining_time": "5 days, 17:04:37"}
+{"current_steps": 87, "total_steps": 34894, "loss": 2.3586, "learning_rate": 9.97134670487106e-06, "epoch": 0.002493256386282791, "percentage": 0.25, "elapsed_time": "0:20:33", "remaining_time": "5 days, 17:05:14"}
+{"current_steps": 88, "total_steps": 34894, "loss": 2.3073, "learning_rate": 1.008595988538682e-05, "epoch": 0.002521914505665352, "percentage": 0.25, "elapsed_time": "0:20:47", "remaining_time": "5 days, 17:03:59"}
+{"current_steps": 89, "total_steps": 34894, "loss": 2.3551, "learning_rate": 1.020057306590258e-05, "epoch": 0.002550572625047913, "percentage": 0.26, "elapsed_time": "0:21:01", "remaining_time": "5 days, 17:02:39"}
+{"current_steps": 90, "total_steps": 34894, "loss": 2.325, "learning_rate": 1.0315186246418338e-05, "epoch": 0.0025792307444304735, "percentage": 0.26, "elapsed_time": "0:21:15", "remaining_time": "5 days, 17:01:28"}
+{"current_steps": 91, "total_steps": 34894, "loss": 2.3561, "learning_rate": 1.0429799426934098e-05, "epoch": 0.0026078888638130346, "percentage": 0.26, "elapsed_time": "0:21:29", "remaining_time": "5 days, 17:00:30"}
+{"current_steps": 92, "total_steps": 34894, "loss": 2.3096, "learning_rate": 1.0544412607449856e-05, "epoch": 0.0026365469831955953, "percentage": 0.26, "elapsed_time": "0:21:43", "remaining_time": "5 days, 16:59:31"}
+{"current_steps": 93, "total_steps": 34894, "loss": 2.3351, "learning_rate": 1.0659025787965617e-05, "epoch": 0.002665205102578156, "percentage": 0.27, "elapsed_time": "0:21:57", "remaining_time": "5 days, 16:58:30"}
+{"current_steps": 94, "total_steps": 34894, "loss": 2.3023, "learning_rate": 1.0773638968481377e-05, "epoch": 0.002693863221960717, "percentage": 0.27, "elapsed_time": "0:22:11", "remaining_time": "5 days, 16:57:20"}
+{"current_steps": 95, "total_steps": 34894, "loss": 2.3555, "learning_rate": 1.0888252148997137e-05, "epoch": 0.0027225213413432777, "percentage": 0.27, "elapsed_time": "0:22:25", "remaining_time": "5 days, 16:56:28"}
+{"current_steps": 96, "total_steps": 34894, "loss": 2.3371, "learning_rate": 1.1002865329512895e-05, "epoch": 0.0027511794607258384, "percentage": 0.28, "elapsed_time": "0:22:39", "remaining_time": "5 days, 16:55:26"}
+{"current_steps": 97, "total_steps": 34894, "loss": 2.3351, "learning_rate": 1.1117478510028654e-05, "epoch": 0.0027798375801083995, "percentage": 0.28, "elapsed_time": "0:22:53", "remaining_time": "5 days, 16:54:51"}
+{"current_steps": 98, "total_steps": 34894, "loss": 2.2488, "learning_rate": 1.1232091690544414e-05, "epoch": 0.00280849569949096, "percentage": 0.28, "elapsed_time": "0:23:08", "remaining_time": "5 days, 16:53:48"}
+{"current_steps": 99, "total_steps": 34894, "loss": 2.2853, "learning_rate": 1.1346704871060172e-05, "epoch": 0.002837153818873521, "percentage": 0.28, "elapsed_time": "0:23:22", "remaining_time": "5 days, 16:52:53"}
+{"current_steps": 100, "total_steps": 34894, "loss": 2.3251, "learning_rate": 1.1461318051575932e-05, "epoch": 0.002865811938256082, "percentage": 0.29, "elapsed_time": "0:23:36", "remaining_time": "5 days, 16:52:48"}
+{"current_steps": 101, "total_steps": 34894, "loss": 2.3481, "learning_rate": 1.157593123209169e-05, "epoch": 0.0028944700576386427, "percentage": 0.29, "elapsed_time": "0:23:50", "remaining_time": "5 days, 16:52:08"}
+{"current_steps": 102, "total_steps": 34894, "loss": 2.2905, "learning_rate": 1.1690544412607451e-05, "epoch": 0.0029231281770212033, "percentage": 0.29, "elapsed_time": "0:24:04", "remaining_time": "5 days, 16:51:41"}
+{"current_steps": 103, "total_steps": 34894, "loss": 2.3235, "learning_rate": 1.1805157593123211e-05, "epoch": 0.0029517862964037645, "percentage": 0.3, "elapsed_time": "0:24:18", "remaining_time": "5 days, 16:51:38"}
+{"current_steps": 104, "total_steps": 34894, "loss": 2.3039, "learning_rate": 1.191977077363897e-05, "epoch": 0.002980444415786325, "percentage": 0.3, "elapsed_time": "0:24:32", "remaining_time": "5 days, 16:50:59"}
+{"current_steps": 105, "total_steps": 34894, "loss": 2.3056, "learning_rate": 1.2034383954154729e-05, "epoch": 0.003009102535168886, "percentage": 0.3, "elapsed_time": "0:24:46", "remaining_time": "5 days, 16:50:52"}
+{"current_steps": 106, "total_steps": 34894, "loss": 2.3236, "learning_rate": 1.2148997134670488e-05, "epoch": 0.0030377606545514465, "percentage": 0.3, "elapsed_time": "0:25:00", "remaining_time": "5 days, 16:49:58"}
+{"current_steps": 107, "total_steps": 34894, "loss": 2.275, "learning_rate": 1.2263610315186248e-05, "epoch": 0.0030664187739340076, "percentage": 0.31, "elapsed_time": "0:25:14", "remaining_time": "5 days, 16:49:01"}
+{"current_steps": 108, "total_steps": 34894, "loss": 2.3818, "learning_rate": 1.2378223495702006e-05, "epoch": 0.0030950768933165683, "percentage": 0.31, "elapsed_time": "0:25:29", "remaining_time": "5 days, 16:48:16"}
+{"current_steps": 109, "total_steps": 34894, "loss": 2.3885, "learning_rate": 1.2492836676217766e-05, "epoch": 0.003123735012699129, "percentage": 0.31, "elapsed_time": "0:25:43", "remaining_time": "5 days, 16:47:27"}
+{"current_steps": 110, "total_steps": 34894, "loss": 2.3248, "learning_rate": 1.2607449856733524e-05, "epoch": 0.00315239313208169, "percentage": 0.32, "elapsed_time": "0:25:57", "remaining_time": "5 days, 16:47:32"}
+{"current_steps": 111, "total_steps": 34894, "loss": 2.2825, "learning_rate": 1.2722063037249284e-05, "epoch": 0.0031810512514642507, "percentage": 0.32, "elapsed_time": "0:26:11", "remaining_time": "5 days, 16:47:00"}
+{"current_steps": 112, "total_steps": 34894, "loss": 2.3469, "learning_rate": 1.2836676217765045e-05, "epoch": 0.0032097093708468114, "percentage": 0.32, "elapsed_time": "0:26:25", "remaining_time": "5 days, 16:46:09"}
+{"current_steps": 113, "total_steps": 34894, "loss": 2.3376, "learning_rate": 1.2951289398280805e-05, "epoch": 0.0032383674902293725, "percentage": 0.32, "elapsed_time": "0:26:39", "remaining_time": "5 days, 16:45:18"}
+{"current_steps": 114, "total_steps": 34894, "loss": 2.2726, "learning_rate": 1.3065902578796563e-05, "epoch": 0.003267025609611933, "percentage": 0.33, "elapsed_time": "0:26:53", "remaining_time": "5 days, 16:44:30"}
+{"current_steps": 115, "total_steps": 34894, "loss": 2.3645, "learning_rate": 1.3180515759312323e-05, "epoch": 0.003295683728994494, "percentage": 0.33, "elapsed_time": "0:27:07", "remaining_time": "5 days, 16:43:37"}
+{"current_steps": 116, "total_steps": 34894, "loss": 2.3306, "learning_rate": 1.329512893982808e-05, "epoch": 0.003324341848377055, "percentage": 0.33, "elapsed_time": "0:27:21", "remaining_time": "5 days, 16:42:49"}
+{"current_steps": 117, "total_steps": 34894, "loss": 2.3881, "learning_rate": 1.340974212034384e-05, "epoch": 0.0033529999677596157, "percentage": 0.34, "elapsed_time": "0:27:35", "remaining_time": "5 days, 16:42:05"}
+{"current_steps": 118, "total_steps": 34894, "loss": 2.3678, "learning_rate": 1.35243553008596e-05, "epoch": 0.0033816580871421763, "percentage": 0.34, "elapsed_time": "0:27:49", "remaining_time": "5 days, 16:41:21"}
+{"current_steps": 119, "total_steps": 34894, "loss": 2.2656, "learning_rate": 1.3638968481375358e-05, "epoch": 0.0034103162065247375, "percentage": 0.34, "elapsed_time": "0:28:03", "remaining_time": "5 days, 16:40:33"}
+{"current_steps": 120, "total_steps": 34894, "loss": 2.3778, "learning_rate": 1.3753581661891118e-05, "epoch": 0.003438974325907298, "percentage": 0.34, "elapsed_time": "0:28:17", "remaining_time": "5 days, 16:40:46"}
+{"current_steps": 121, "total_steps": 34894, "loss": 2.3635, "learning_rate": 1.3868194842406878e-05, "epoch": 0.003467632445289859, "percentage": 0.35, "elapsed_time": "0:28:32", "remaining_time": "5 days, 16:40:25"}
+{"current_steps": 122, "total_steps": 34894, "loss": 2.3877, "learning_rate": 1.3982808022922639e-05, "epoch": 0.00349629056467242, "percentage": 0.35, "elapsed_time": "0:28:46", "remaining_time": "5 days, 16:39:44"}
+{"current_steps": 123, "total_steps": 34894, "loss": 2.3766, "learning_rate": 1.4097421203438397e-05, "epoch": 0.0035249486840549806, "percentage": 0.35, "elapsed_time": "0:29:00", "remaining_time": "5 days, 16:39:17"}
+{"current_steps": 124, "total_steps": 34894, "loss": 2.3809, "learning_rate": 1.4212034383954157e-05, "epoch": 0.0035536068034375413, "percentage": 0.36, "elapsed_time": "0:29:14", "remaining_time": "5 days, 16:38:53"}
+{"current_steps": 125, "total_steps": 34894, "loss": 2.31, "learning_rate": 1.4326647564469915e-05, "epoch": 0.0035822649228201024, "percentage": 0.36, "elapsed_time": "0:29:28", "remaining_time": "5 days, 16:38:07"}
+{"current_steps": 126, "total_steps": 34894, "loss": 2.3367, "learning_rate": 1.4441260744985674e-05, "epoch": 0.003610923042202663, "percentage": 0.36, "elapsed_time": "0:29:42", "remaining_time": "5 days, 16:37:31"}
+{"current_steps": 127, "total_steps": 34894, "loss": 2.2856, "learning_rate": 1.4555873925501434e-05, "epoch": 0.0036395811615852237, "percentage": 0.36, "elapsed_time": "0:29:56", "remaining_time": "5 days, 16:36:55"}
+{"current_steps": 128, "total_steps": 34894, "loss": 2.3997, "learning_rate": 1.4670487106017192e-05, "epoch": 0.003668239280967785, "percentage": 0.37, "elapsed_time": "0:30:10", "remaining_time": "5 days, 16:36:08"}
+{"current_steps": 129, "total_steps": 34894, "loss": 2.2927, "learning_rate": 1.4785100286532952e-05, "epoch": 0.0036968974003503455, "percentage": 0.37, "elapsed_time": "0:30:24", "remaining_time": "5 days, 16:35:23"}
+{"current_steps": 130, "total_steps": 34894, "loss": 2.3425, "learning_rate": 1.4899713467048712e-05, "epoch": 0.003725555519732906, "percentage": 0.37, "elapsed_time": "0:30:38", "remaining_time": "5 days, 16:34:46"}
+{"current_steps": 131, "total_steps": 34894, "loss": 2.2955, "learning_rate": 1.5014326647564471e-05, "epoch": 0.0037542136391154673, "percentage": 0.38, "elapsed_time": "0:30:52", "remaining_time": "5 days, 16:34:00"}
+{"current_steps": 132, "total_steps": 34894, "loss": 2.319, "learning_rate": 1.5128939828080231e-05, "epoch": 0.003782871758498028, "percentage": 0.38, "elapsed_time": "0:31:06", "remaining_time": "5 days, 16:33:18"}
+{"current_steps": 133, "total_steps": 34894, "loss": 2.335, "learning_rate": 1.5243553008595991e-05, "epoch": 0.0038115298778805887, "percentage": 0.38, "elapsed_time": "0:31:20", "remaining_time": "5 days, 16:32:39"}
+{"current_steps": 134, "total_steps": 34894, "loss": 2.3252, "learning_rate": 1.535816618911175e-05, "epoch": 0.0038401879972631498, "percentage": 0.38, "elapsed_time": "0:31:34", "remaining_time": "5 days, 16:32:00"}
+{"current_steps": 135, "total_steps": 34894, "loss": 2.3332, "learning_rate": 1.5472779369627507e-05, "epoch": 0.0038688461166457105, "percentage": 0.39, "elapsed_time": "0:31:48", "remaining_time": "5 days, 16:31:20"}
+{"current_steps": 136, "total_steps": 34894, "loss": 2.3668, "learning_rate": 1.558739255014327e-05, "epoch": 0.003897504236028271, "percentage": 0.39, "elapsed_time": "0:32:02", "remaining_time": "5 days, 16:30:37"}
+{"current_steps": 137, "total_steps": 34894, "loss": 2.3265, "learning_rate": 1.5702005730659026e-05, "epoch": 0.003926162355410832, "percentage": 0.39, "elapsed_time": "0:32:16", "remaining_time": "5 days, 16:29:56"}
+{"current_steps": 138, "total_steps": 34894, "loss": 2.3671, "learning_rate": 1.5816618911174784e-05, "epoch": 0.0039548204747933925, "percentage": 0.4, "elapsed_time": "0:32:30", "remaining_time": "5 days, 16:29:17"}
+{"current_steps": 139, "total_steps": 34894, "loss": 2.3408, "learning_rate": 1.5931232091690546e-05, "epoch": 0.003983478594175954, "percentage": 0.4, "elapsed_time": "0:32:44", "remaining_time": "5 days, 16:28:36"}
+{"current_steps": 140, "total_steps": 34894, "loss": 2.3092, "learning_rate": 1.6045845272206304e-05, "epoch": 0.004012136713558515, "percentage": 0.4, "elapsed_time": "0:32:59", "remaining_time": "5 days, 16:27:56"}
+{"current_steps": 141, "total_steps": 34894, "loss": 2.3729, "learning_rate": 1.6160458452722065e-05, "epoch": 0.004040794832941075, "percentage": 0.4, "elapsed_time": "0:33:13", "remaining_time": "5 days, 16:27:21"}
+{"current_steps": 142, "total_steps": 34894, "loss": 2.3216, "learning_rate": 1.6275071633237823e-05, "epoch": 0.004069452952323636, "percentage": 0.41, "elapsed_time": "0:33:27", "remaining_time": "5 days, 16:26:43"}
+{"current_steps": 143, "total_steps": 34894, "loss": 2.3708, "learning_rate": 1.6389684813753585e-05, "epoch": 0.004098111071706197, "percentage": 0.41, "elapsed_time": "0:33:41", "remaining_time": "5 days, 16:26:07"}
+{"current_steps": 144, "total_steps": 34894, "loss": 2.2778, "learning_rate": 1.6504297994269343e-05, "epoch": 0.004126769191088757, "percentage": 0.41, "elapsed_time": "0:33:55", "remaining_time": "5 days, 16:25:26"}
+{"current_steps": 145, "total_steps": 34894, "loss": 2.3386, "learning_rate": 1.66189111747851e-05, "epoch": 0.0041554273104713185, "percentage": 0.42, "elapsed_time": "0:34:09", "remaining_time": "5 days, 16:24:50"}
+{"current_steps": 146, "total_steps": 34894, "loss": 2.3758, "learning_rate": 1.6733524355300862e-05, "epoch": 0.00418408542985388, "percentage": 0.42, "elapsed_time": "0:34:23", "remaining_time": "5 days, 16:24:15"}
+{"current_steps": 147, "total_steps": 34894, "loss": 2.3286, "learning_rate": 1.684813753581662e-05, "epoch": 0.00421274354923644, "percentage": 0.42, "elapsed_time": "0:34:37", "remaining_time": "5 days, 16:23:58"}
+{"current_steps": 148, "total_steps": 34894, "loss": 2.3759, "learning_rate": 1.6962750716332378e-05, "epoch": 0.004241401668619001, "percentage": 0.42, "elapsed_time": "0:34:51", "remaining_time": "5 days, 16:24:13"}
+{"current_steps": 149, "total_steps": 34894, "loss": 2.321, "learning_rate": 1.707736389684814e-05, "epoch": 0.004270059788001562, "percentage": 0.43, "elapsed_time": "0:35:05", "remaining_time": "5 days, 16:24:26"}
+{"current_steps": 150, "total_steps": 34894, "loss": 2.3727, "learning_rate": 1.7191977077363898e-05, "epoch": 0.004298717907384122, "percentage": 0.43, "elapsed_time": "0:35:19", "remaining_time": "5 days, 16:23:50"}
+{"current_steps": 151, "total_steps": 34894, "loss": 2.3241, "learning_rate": 1.730659025787966e-05, "epoch": 0.0043273760267666835, "percentage": 0.43, "elapsed_time": "0:35:33", "remaining_time": "5 days, 16:23:13"}
+{"current_steps": 152, "total_steps": 34894, "loss": 2.3495, "learning_rate": 1.7421203438395417e-05, "epoch": 0.0043560341461492446, "percentage": 0.44, "elapsed_time": "0:35:47", "remaining_time": "5 days, 16:22:37"}
+{"current_steps": 153, "total_steps": 34894, "loss": 2.3993, "learning_rate": 1.7535816618911175e-05, "epoch": 0.004384692265531805, "percentage": 0.44, "elapsed_time": "0:36:02", "remaining_time": "5 days, 16:21:59"}
+{"current_steps": 154, "total_steps": 34894, "loss": 2.3438, "learning_rate": 1.7650429799426937e-05, "epoch": 0.004413350384914366, "percentage": 0.44, "elapsed_time": "0:36:16", "remaining_time": "5 days, 16:21:23"}
+{"current_steps": 155, "total_steps": 34894, "loss": 2.3106, "learning_rate": 1.7765042979942695e-05, "epoch": 0.004442008504296927, "percentage": 0.44, "elapsed_time": "0:36:30", "remaining_time": "5 days, 16:20:45"}
+{"current_steps": 156, "total_steps": 34894, "loss": 2.2814, "learning_rate": 1.7879656160458453e-05, "epoch": 0.004470666623679487, "percentage": 0.45, "elapsed_time": "0:36:44", "remaining_time": "5 days, 16:20:11"}
+{"current_steps": 157, "total_steps": 34894, "loss": 2.3489, "learning_rate": 1.7994269340974214e-05, "epoch": 0.004499324743062048, "percentage": 0.45, "elapsed_time": "0:36:58", "remaining_time": "5 days, 16:19:35"}
+{"current_steps": 158, "total_steps": 34894, "loss": 2.3178, "learning_rate": 1.8108882521489972e-05, "epoch": 0.0045279828624446095, "percentage": 0.45, "elapsed_time": "0:37:12", "remaining_time": "5 days, 16:19:03"}
+{"current_steps": 159, "total_steps": 34894, "loss": 2.3475, "learning_rate": 1.822349570200573e-05, "epoch": 0.00455664098182717, "percentage": 0.46, "elapsed_time": "0:37:26", "remaining_time": "5 days, 16:18:28"}
+{"current_steps": 160, "total_steps": 34894, "loss": 2.3319, "learning_rate": 1.833810888252149e-05, "epoch": 0.004585299101209731, "percentage": 0.46, "elapsed_time": "0:37:40", "remaining_time": "5 days, 16:18:32"}
+{"current_steps": 161, "total_steps": 34894, "loss": 2.3231, "learning_rate": 1.8452722063037253e-05, "epoch": 0.004613957220592292, "percentage": 0.46, "elapsed_time": "0:37:54", "remaining_time": "5 days, 16:18:13"}
+{"current_steps": 162, "total_steps": 34894, "loss": 2.3653, "learning_rate": 1.856733524355301e-05, "epoch": 0.004642615339974852, "percentage": 0.46, "elapsed_time": "0:38:08", "remaining_time": "5 days, 16:18:53"}
+{"current_steps": 163, "total_steps": 34894, "loss": 2.3304, "learning_rate": 1.868194842406877e-05, "epoch": 0.004671273459357413, "percentage": 0.47, "elapsed_time": "0:38:23", "remaining_time": "5 days, 16:19:35"}
+{"current_steps": 164, "total_steps": 34894, "loss": 2.3613, "learning_rate": 1.879656160458453e-05, "epoch": 0.004699931578739974, "percentage": 0.47, "elapsed_time": "0:38:37", "remaining_time": "5 days, 16:20:11"}
+{"current_steps": 165, "total_steps": 34894, "loss": 2.3691, "learning_rate": 1.891117478510029e-05, "epoch": 0.004728589698122535, "percentage": 0.47, "elapsed_time": "0:38:52", "remaining_time": "5 days, 16:20:47"}
+{"current_steps": 166, "total_steps": 34894, "loss": 2.3465, "learning_rate": 1.9025787965616047e-05, "epoch": 0.004757247817505096, "percentage": 0.48, "elapsed_time": "0:39:06", "remaining_time": "5 days, 16:20:49"}
+{"current_steps": 167, "total_steps": 34894, "loss": 2.3117, "learning_rate": 1.9140401146131805e-05, "epoch": 0.004785905936887657, "percentage": 0.48, "elapsed_time": "0:39:20", "remaining_time": "5 days, 16:20:11"}
+{"current_steps": 168, "total_steps": 34894, "loss": 2.3846, "learning_rate": 1.9255014326647566e-05, "epoch": 0.004814564056270217, "percentage": 0.48, "elapsed_time": "0:39:34", "remaining_time": "5 days, 16:19:54"}
+{"current_steps": 169, "total_steps": 34894, "loss": 2.378, "learning_rate": 1.9369627507163324e-05, "epoch": 0.004843222175652778, "percentage": 0.48, "elapsed_time": "0:39:48", "remaining_time": "5 days, 16:20:13"}
+{"current_steps": 170, "total_steps": 34894, "loss": 2.3349, "learning_rate": 1.9484240687679085e-05, "epoch": 0.004871880295035339, "percentage": 0.49, "elapsed_time": "0:40:02", "remaining_time": "5 days, 16:19:42"}
+{"current_steps": 171, "total_steps": 34894, "loss": 2.3219, "learning_rate": 1.9598853868194843e-05, "epoch": 0.0049005384144179, "percentage": 0.49, "elapsed_time": "0:40:16", "remaining_time": "5 days, 16:19:06"}
+{"current_steps": 172, "total_steps": 34894, "loss": 2.4099, "learning_rate": 1.9713467048710605e-05, "epoch": 0.004929196533800461, "percentage": 0.49, "elapsed_time": "0:40:30", "remaining_time": "5 days, 16:18:32"}
+{"current_steps": 173, "total_steps": 34894, "loss": 2.3433, "learning_rate": 1.9828080229226363e-05, "epoch": 0.004957854653183022, "percentage": 0.5, "elapsed_time": "0:40:44", "remaining_time": "5 days, 16:18:00"}
+{"current_steps": 174, "total_steps": 34894, "loss": 2.3494, "learning_rate": 1.994269340974212e-05, "epoch": 0.004986512772565582, "percentage": 0.5, "elapsed_time": "0:40:59", "remaining_time": "5 days, 16:17:52"}
+{"current_steps": 175, "total_steps": 34894, "loss": 2.3801, "learning_rate": 2.0057306590257882e-05, "epoch": 0.005015170891948143, "percentage": 0.5, "elapsed_time": "0:41:13", "remaining_time": "5 days, 16:17:24"}
+{"current_steps": 176, "total_steps": 34894, "loss": 2.3184, "learning_rate": 2.017191977077364e-05, "epoch": 0.005043829011330704, "percentage": 0.5, "elapsed_time": "0:41:27", "remaining_time": "5 days, 16:16:53"}
+{"current_steps": 177, "total_steps": 34894, "loss": 2.3836, "learning_rate": 2.02865329512894e-05, "epoch": 0.0050724871307132645, "percentage": 0.51, "elapsed_time": "0:41:41", "remaining_time": "5 days, 16:16:20"}
+{"current_steps": 178, "total_steps": 34894, "loss": 2.3115, "learning_rate": 2.040114613180516e-05, "epoch": 0.005101145250095826, "percentage": 0.51, "elapsed_time": "0:41:55", "remaining_time": "5 days, 16:15:50"}
+{"current_steps": 179, "total_steps": 34894, "loss": 2.3413, "learning_rate": 2.0515759312320918e-05, "epoch": 0.005129803369478387, "percentage": 0.51, "elapsed_time": "0:42:09", "remaining_time": "5 days, 16:16:19"}
+{"current_steps": 180, "total_steps": 34894, "loss": 2.3467, "learning_rate": 2.0630372492836676e-05, "epoch": 0.005158461488860947, "percentage": 0.52, "elapsed_time": "0:42:23", "remaining_time": "5 days, 16:15:59"}
+{"current_steps": 181, "total_steps": 34894, "loss": 2.3666, "learning_rate": 2.0744985673352437e-05, "epoch": 0.005187119608243508, "percentage": 0.52, "elapsed_time": "0:42:37", "remaining_time": "5 days, 16:15:39"}
+{"current_steps": 182, "total_steps": 34894, "loss": 2.3165, "learning_rate": 2.0859598853868195e-05, "epoch": 0.005215777727626069, "percentage": 0.52, "elapsed_time": "0:42:51", "remaining_time": "5 days, 16:15:03"}
+{"current_steps": 183, "total_steps": 34894, "loss": 2.3322, "learning_rate": 2.0974212034383953e-05, "epoch": 0.0052444358470086294, "percentage": 0.52, "elapsed_time": "0:43:05", "remaining_time": "5 days, 16:14:33"}
+{"current_steps": 184, "total_steps": 34894, "loss": 2.3251, "learning_rate": 2.108882521489971e-05, "epoch": 0.0052730939663911906, "percentage": 0.53, "elapsed_time": "0:43:19", "remaining_time": "5 days, 16:13:59"}
+{"current_steps": 185, "total_steps": 34894, "loss": 2.3653, "learning_rate": 2.1203438395415476e-05, "epoch": 0.005301752085773752, "percentage": 0.53, "elapsed_time": "0:43:33", "remaining_time": "5 days, 16:13:46"}
+{"current_steps": 186, "total_steps": 34894, "loss": 2.3289, "learning_rate": 2.1318051575931234e-05, "epoch": 0.005330410205156312, "percentage": 0.53, "elapsed_time": "0:43:48", "remaining_time": "5 days, 16:13:22"}
+{"current_steps": 187, "total_steps": 34894, "loss": 2.4178, "learning_rate": 2.1432664756446996e-05, "epoch": 0.005359068324538873, "percentage": 0.54, "elapsed_time": "0:44:02", "remaining_time": "5 days, 16:12:46"}
+{"current_steps": 188, "total_steps": 34894, "loss": 2.3664, "learning_rate": 2.1547277936962754e-05, "epoch": 0.005387726443921434, "percentage": 0.54, "elapsed_time": "0:44:16", "remaining_time": "5 days, 16:12:32"}
+{"current_steps": 189, "total_steps": 34894, "loss": 2.3628, "learning_rate": 2.1661891117478512e-05, "epoch": 0.005416384563303994, "percentage": 0.54, "elapsed_time": "0:44:30", "remaining_time": "5 days, 16:13:07"}
+{"current_steps": 190, "total_steps": 34894, "loss": 2.4115, "learning_rate": 2.1776504297994273e-05, "epoch": 0.0054450426826865555, "percentage": 0.54, "elapsed_time": "0:44:45", "remaining_time": "5 days, 16:13:43"}
+{"current_steps": 191, "total_steps": 34894, "loss": 2.3666, "learning_rate": 2.189111747851003e-05, "epoch": 0.005473700802069117, "percentage": 0.55, "elapsed_time": "0:44:59", "remaining_time": "5 days, 16:14:10"}
+{"current_steps": 192, "total_steps": 34894, "loss": 2.3745, "learning_rate": 2.200573065902579e-05, "epoch": 0.005502358921451677, "percentage": 0.55, "elapsed_time": "0:45:13", "remaining_time": "5 days, 16:14:35"}
+{"current_steps": 193, "total_steps": 34894, "loss": 2.3446, "learning_rate": 2.212034383954155e-05, "epoch": 0.005531017040834238, "percentage": 0.55, "elapsed_time": "0:45:28", "remaining_time": "5 days, 16:14:59"}
+{"current_steps": 194, "total_steps": 34894, "loss": 2.2998, "learning_rate": 2.223495702005731e-05, "epoch": 0.005559675160216799, "percentage": 0.56, "elapsed_time": "0:45:42", "remaining_time": "5 days, 16:15:21"}
+{"current_steps": 195, "total_steps": 34894, "loss": 2.321, "learning_rate": 2.2349570200573067e-05, "epoch": 0.005588333279599359, "percentage": 0.56, "elapsed_time": "0:45:56", "remaining_time": "5 days, 16:15:51"}
+{"current_steps": 196, "total_steps": 34894, "loss": 2.3393, "learning_rate": 2.2464183381088828e-05, "epoch": 0.00561699139898192, "percentage": 0.56, "elapsed_time": "0:46:11", "remaining_time": "5 days, 16:16:19"}
+{"current_steps": 197, "total_steps": 34894, "loss": 2.3268, "learning_rate": 2.2578796561604586e-05, "epoch": 0.0056456495183644815, "percentage": 0.56, "elapsed_time": "0:46:25", "remaining_time": "5 days, 16:15:46"}
+{"current_steps": 198, "total_steps": 34894, "loss": 2.364, "learning_rate": 2.2693409742120344e-05, "epoch": 0.005674307637747042, "percentage": 0.57, "elapsed_time": "0:46:39", "remaining_time": "5 days, 16:15:46"}
+{"current_steps": 199, "total_steps": 34894, "loss": 2.376, "learning_rate": 2.2808022922636102e-05, "epoch": 0.005702965757129603, "percentage": 0.57, "elapsed_time": "0:46:53", "remaining_time": "5 days, 16:16:22"}
+{"current_steps": 200, "total_steps": 34894, "loss": 2.3263, "learning_rate": 2.2922636103151864e-05, "epoch": 0.005731623876512164, "percentage": 0.57, "elapsed_time": "0:47:08", "remaining_time": "5 days, 16:16:41"}
+{"current_steps": 201, "total_steps": 34894, "loss": 2.3389, "learning_rate": 2.303724928366762e-05, "epoch": 0.005760281995894724, "percentage": 0.58, "elapsed_time": "0:47:22", "remaining_time": "5 days, 16:17:13"}
+{"current_steps": 202, "total_steps": 34894, "loss": 2.3763, "learning_rate": 2.315186246418338e-05, "epoch": 0.005788940115277285, "percentage": 0.58, "elapsed_time": "0:47:36", "remaining_time": "5 days, 16:17:35"}
+{"current_steps": 203, "total_steps": 34894, "loss": 2.2697, "learning_rate": 2.326647564469914e-05, "epoch": 0.0058175982346598465, "percentage": 0.58, "elapsed_time": "0:47:51", "remaining_time": "5 days, 16:17:32"}
+{"current_steps": 204, "total_steps": 34894, "loss": 2.3543, "learning_rate": 2.3381088825214902e-05, "epoch": 0.005846256354042407, "percentage": 0.58, "elapsed_time": "0:48:05", "remaining_time": "5 days, 16:17:44"}
+{"current_steps": 205, "total_steps": 34894, "loss": 2.3578, "learning_rate": 2.3495702005730664e-05, "epoch": 0.005874914473424968, "percentage": 0.59, "elapsed_time": "0:48:19", "remaining_time": "5 days, 16:17:22"}
+{"current_steps": 206, "total_steps": 34894, "loss": 2.309, "learning_rate": 2.3610315186246422e-05, "epoch": 0.005903572592807529, "percentage": 0.59, "elapsed_time": "0:48:33", "remaining_time": "5 days, 16:17:57"}
+{"current_steps": 207, "total_steps": 34894, "loss": 2.3201, "learning_rate": 2.372492836676218e-05, "epoch": 0.005932230712190089, "percentage": 0.59, "elapsed_time": "0:48:48", "remaining_time": "5 days, 16:18:17"}
+{"current_steps": 208, "total_steps": 34894, "loss": 2.3687, "learning_rate": 2.383954154727794e-05, "epoch": 0.00596088883157265, "percentage": 0.6, "elapsed_time": "0:49:02", "remaining_time": "5 days, 16:18:21"}
+{"current_steps": 209, "total_steps": 34894, "loss": 2.377, "learning_rate": 2.39541547277937e-05, "epoch": 0.0059895469509552105, "percentage": 0.6, "elapsed_time": "0:49:16", "remaining_time": "5 days, 16:17:47"}
+{"current_steps": 210, "total_steps": 34894, "loss": 2.3935, "learning_rate": 2.4068767908309457e-05, "epoch": 0.006018205070337772, "percentage": 0.6, "elapsed_time": "0:49:30", "remaining_time": "5 days, 16:17:16"}
+{"current_steps": 211, "total_steps": 34894, "loss": 2.3318, "learning_rate": 2.418338108882522e-05, "epoch": 0.006046863189720333, "percentage": 0.6, "elapsed_time": "0:49:44", "remaining_time": "5 days, 16:16:57"}
+{"current_steps": 212, "total_steps": 34894, "loss": 2.3918, "learning_rate": 2.4297994269340977e-05, "epoch": 0.006075521309102893, "percentage": 0.61, "elapsed_time": "0:49:59", "remaining_time": "5 days, 16:17:04"}
+{"current_steps": 213, "total_steps": 34894, "loss": 2.3887, "learning_rate": 2.4412607449856735e-05, "epoch": 0.006104179428485454, "percentage": 0.61, "elapsed_time": "0:50:13", "remaining_time": "5 days, 16:17:23"}
+{"current_steps": 214, "total_steps": 34894, "loss": 2.3965, "learning_rate": 2.4527220630372496e-05, "epoch": 0.006132837547868015, "percentage": 0.61, "elapsed_time": "0:50:27", "remaining_time": "5 days, 16:17:45"}
+{"current_steps": 215, "total_steps": 34894, "loss": 2.3288, "learning_rate": 2.4641833810888254e-05, "epoch": 0.0061614956672505754, "percentage": 0.62, "elapsed_time": "0:50:42", "remaining_time": "5 days, 16:18:06"}
+{"current_steps": 216, "total_steps": 34894, "loss": 2.3287, "learning_rate": 2.4756446991404012e-05, "epoch": 0.0061901537866331366, "percentage": 0.62, "elapsed_time": "0:50:56", "remaining_time": "5 days, 16:18:32"}
+{"current_steps": 217, "total_steps": 34894, "loss": 2.3275, "learning_rate": 2.487106017191977e-05, "epoch": 0.006218811906015698, "percentage": 0.62, "elapsed_time": "0:51:10", "remaining_time": "5 days, 16:18:24"}
+{"current_steps": 218, "total_steps": 34894, "loss": 2.3445, "learning_rate": 2.4985673352435532e-05, "epoch": 0.006247470025398258, "percentage": 0.62, "elapsed_time": "0:51:24", "remaining_time": "5 days, 16:17:54"}
+{"current_steps": 219, "total_steps": 34894, "loss": 2.3291, "learning_rate": 2.510028653295129e-05, "epoch": 0.006276128144780819, "percentage": 0.63, "elapsed_time": "0:51:38", "remaining_time": "5 days, 16:17:24"}
+{"current_steps": 220, "total_steps": 34894, "loss": 2.3362, "learning_rate": 2.5214899713467048e-05, "epoch": 0.00630478626416338, "percentage": 0.63, "elapsed_time": "0:51:52", "remaining_time": "5 days, 16:16:52"}
+{"current_steps": 221, "total_steps": 34894, "loss": 2.3496, "learning_rate": 2.532951289398281e-05, "epoch": 0.00633344438354594, "percentage": 0.63, "elapsed_time": "0:52:06", "remaining_time": "5 days, 16:16:19"}
+{"current_steps": 222, "total_steps": 34894, "loss": 2.3279, "learning_rate": 2.5444126074498567e-05, "epoch": 0.0063621025029285015, "percentage": 0.64, "elapsed_time": "0:52:20", "remaining_time": "5 days, 16:15:45"}
+{"current_steps": 223, "total_steps": 34894, "loss": 2.2964, "learning_rate": 2.5558739255014332e-05, "epoch": 0.006390760622311063, "percentage": 0.64, "elapsed_time": "0:52:35", "remaining_time": "5 days, 16:15:56"}
+{"current_steps": 224, "total_steps": 34894, "loss": 2.372, "learning_rate": 2.567335243553009e-05, "epoch": 0.006419418741693623, "percentage": 0.64, "elapsed_time": "0:52:49", "remaining_time": "5 days, 16:15:56"}
+{"current_steps": 225, "total_steps": 34894, "loss": 2.3237, "learning_rate": 2.5787965616045848e-05, "epoch": 0.006448076861076184, "percentage": 0.64, "elapsed_time": "0:53:03", "remaining_time": "5 days, 16:15:59"}
+{"current_steps": 226, "total_steps": 34894, "loss": 2.4317, "learning_rate": 2.590257879656161e-05, "epoch": 0.006476734980458745, "percentage": 0.65, "elapsed_time": "0:53:17", "remaining_time": "5 days, 16:16:07"}
+{"current_steps": 227, "total_steps": 34894, "loss": 2.4151, "learning_rate": 2.6017191977077368e-05, "epoch": 0.006505393099841305, "percentage": 0.65, "elapsed_time": "0:53:32", "remaining_time": "5 days, 16:15:36"}
+{"current_steps": 228, "total_steps": 34894, "loss": 2.3075, "learning_rate": 2.6131805157593126e-05, "epoch": 0.006534051219223866, "percentage": 0.65, "elapsed_time": "0:53:46", "remaining_time": "5 days, 16:15:05"}
+{"current_steps": 229, "total_steps": 34894, "loss": 2.3838, "learning_rate": 2.6246418338108887e-05, "epoch": 0.0065627093386064275, "percentage": 0.66, "elapsed_time": "0:54:00", "remaining_time": "5 days, 16:14:31"}
+{"current_steps": 230, "total_steps": 34894, "loss": 2.3744, "learning_rate": 2.6361031518624645e-05, "epoch": 0.006591367457988988, "percentage": 0.66, "elapsed_time": "0:54:14", "remaining_time": "5 days, 16:14:00"}
+{"current_steps": 231, "total_steps": 34894, "loss": 2.3467, "learning_rate": 2.6475644699140403e-05, "epoch": 0.006620025577371549, "percentage": 0.66, "elapsed_time": "0:54:28", "remaining_time": "5 days, 16:13:29"}
+{"current_steps": 232, "total_steps": 34894, "loss": 2.3595, "learning_rate": 2.659025787965616e-05, "epoch": 0.00664868369675411, "percentage": 0.66, "elapsed_time": "0:54:42", "remaining_time": "5 days, 16:12:57"}
+{"current_steps": 233, "total_steps": 34894, "loss": 2.3474, "learning_rate": 2.6704871060171923e-05, "epoch": 0.00667734181613667, "percentage": 0.67, "elapsed_time": "0:54:56", "remaining_time": "5 days, 16:12:27"}
+{"current_steps": 234, "total_steps": 34894, "loss": 2.3544, "learning_rate": 2.681948424068768e-05, "epoch": 0.006705999935519231, "percentage": 0.67, "elapsed_time": "0:55:10", "remaining_time": "5 days, 16:11:56"}
+{"current_steps": 235, "total_steps": 34894, "loss": 2.3264, "learning_rate": 2.693409742120344e-05, "epoch": 0.0067346580549017925, "percentage": 0.67, "elapsed_time": "0:55:24", "remaining_time": "5 days, 16:11:28"}
+{"current_steps": 236, "total_steps": 34894, "loss": 2.3483, "learning_rate": 2.70487106017192e-05, "epoch": 0.006763316174284353, "percentage": 0.68, "elapsed_time": "0:55:38", "remaining_time": "5 days, 16:10:56"}
+{"current_steps": 237, "total_steps": 34894, "loss": 2.4108, "learning_rate": 2.7163323782234958e-05, "epoch": 0.006791974293666914, "percentage": 0.68, "elapsed_time": "0:55:52", "remaining_time": "5 days, 16:10:22"}
+{"current_steps": 238, "total_steps": 34894, "loss": 2.3492, "learning_rate": 2.7277936962750716e-05, "epoch": 0.006820632413049475, "percentage": 0.68, "elapsed_time": "0:56:06", "remaining_time": "5 days, 16:09:52"}
+{"current_steps": 239, "total_steps": 34894, "loss": 2.3414, "learning_rate": 2.7392550143266478e-05, "epoch": 0.006849290532432035, "percentage": 0.68, "elapsed_time": "0:56:20", "remaining_time": "5 days, 16:09:21"}
+{"current_steps": 240, "total_steps": 34894, "loss": 2.3358, "learning_rate": 2.7507163323782236e-05, "epoch": 0.006877948651814596, "percentage": 0.69, "elapsed_time": "0:56:34", "remaining_time": "5 days, 16:08:52"}
+{"current_steps": 241, "total_steps": 34894, "loss": 2.3688, "learning_rate": 2.7621776504297994e-05, "epoch": 0.006906606771197157, "percentage": 0.69, "elapsed_time": "0:56:48", "remaining_time": "5 days, 16:08:20"}
+{"current_steps": 242, "total_steps": 34894, "loss": 2.4096, "learning_rate": 2.7736389684813755e-05, "epoch": 0.006935264890579718, "percentage": 0.69, "elapsed_time": "0:57:02", "remaining_time": "5 days, 16:08:02"}
+{"current_steps": 243, "total_steps": 34894, "loss": 2.3668, "learning_rate": 2.7851002865329517e-05, "epoch": 0.006963923009962279, "percentage": 0.7, "elapsed_time": "0:57:16", "remaining_time": "5 days, 16:07:29"}
+{"current_steps": 244, "total_steps": 34894, "loss": 2.3293, "learning_rate": 2.7965616045845278e-05, "epoch": 0.00699258112934484, "percentage": 0.7, "elapsed_time": "0:57:30", "remaining_time": "5 days, 16:07:20"}
+{"current_steps": 245, "total_steps": 34894, "loss": 2.2976, "learning_rate": 2.8080229226361036e-05, "epoch": 0.0070212392487274, "percentage": 0.7, "elapsed_time": "0:57:44", "remaining_time": "5 days, 16:06:59"}
+{"current_steps": 246, "total_steps": 34894, "loss": 2.3352, "learning_rate": 2.8194842406876794e-05, "epoch": 0.007049897368109961, "percentage": 0.7, "elapsed_time": "0:57:58", "remaining_time": "5 days, 16:06:34"}
+{"current_steps": 247, "total_steps": 34894, "loss": 2.4041, "learning_rate": 2.8309455587392552e-05, "epoch": 0.007078555487492522, "percentage": 0.71, "elapsed_time": "0:58:12", "remaining_time": "5 days, 16:06:05"}
+{"current_steps": 248, "total_steps": 34894, "loss": 2.3241, "learning_rate": 2.8424068767908313e-05, "epoch": 0.0071072136068750826, "percentage": 0.71, "elapsed_time": "0:58:27", "remaining_time": "5 days, 16:05:35"}
+{"current_steps": 249, "total_steps": 34894, "loss": 2.3461, "learning_rate": 2.853868194842407e-05, "epoch": 0.007135871726257644, "percentage": 0.71, "elapsed_time": "0:58:41", "remaining_time": "5 days, 16:05:06"}
+{"current_steps": 250, "total_steps": 34894, "loss": 2.3622, "learning_rate": 2.865329512893983e-05, "epoch": 0.007164529845640205, "percentage": 0.72, "elapsed_time": "0:58:55", "remaining_time": "5 days, 16:04:36"}
+{"current_steps": 251, "total_steps": 34894, "loss": 2.3289, "learning_rate": 2.876790830945559e-05, "epoch": 0.007193187965022765, "percentage": 0.72, "elapsed_time": "0:59:09", "remaining_time": "5 days, 16:04:06"}
+{"current_steps": 252, "total_steps": 34894, "loss": 2.3454, "learning_rate": 2.888252148997135e-05, "epoch": 0.007221846084405326, "percentage": 0.72, "elapsed_time": "0:59:23", "remaining_time": "5 days, 16:03:42"}
+{"current_steps": 253, "total_steps": 34894, "loss": 2.3418, "learning_rate": 2.8997134670487107e-05, "epoch": 0.007250504203787887, "percentage": 0.73, "elapsed_time": "0:59:37", "remaining_time": "5 days, 16:03:13"}
+{"current_steps": 254, "total_steps": 34894, "loss": 2.3677, "learning_rate": 2.911174785100287e-05, "epoch": 0.0072791623231704475, "percentage": 0.73, "elapsed_time": "0:59:51", "remaining_time": "5 days, 16:02:56"}
+{"current_steps": 255, "total_steps": 34894, "loss": 2.2985, "learning_rate": 2.9226361031518626e-05, "epoch": 0.007307820442553009, "percentage": 0.73, "elapsed_time": "1:00:05", "remaining_time": "5 days, 16:02:27"}
+{"current_steps": 256, "total_steps": 34894, "loss": 2.3139, "learning_rate": 2.9340974212034384e-05, "epoch": 0.00733647856193557, "percentage": 0.73, "elapsed_time": "1:00:19", "remaining_time": "5 days, 16:01:59"}
+{"current_steps": 257, "total_steps": 34894, "loss": 2.3974, "learning_rate": 2.9455587392550146e-05, "epoch": 0.00736513668131813, "percentage": 0.74, "elapsed_time": "1:00:33", "remaining_time": "5 days, 16:01:33"}
+{"current_steps": 258, "total_steps": 34894, "loss": 2.368, "learning_rate": 2.9570200573065904e-05, "epoch": 0.007393794800700691, "percentage": 0.74, "elapsed_time": "1:00:47", "remaining_time": "5 days, 16:01:03"}
+{"current_steps": 259, "total_steps": 34894, "loss": 2.3227, "learning_rate": 2.9684813753581662e-05, "epoch": 0.007422452920083252, "percentage": 0.74, "elapsed_time": "1:01:01", "remaining_time": "5 days, 16:00:34"}
+{"current_steps": 260, "total_steps": 34894, "loss": 2.3164, "learning_rate": 2.9799426934097423e-05, "epoch": 0.007451111039465812, "percentage": 0.75, "elapsed_time": "1:01:15", "remaining_time": "5 days, 16:00:05"}
+{"current_steps": 261, "total_steps": 34894, "loss": 2.3206, "learning_rate": 2.991404011461318e-05, "epoch": 0.0074797691588483735, "percentage": 0.75, "elapsed_time": "1:01:29", "remaining_time": "5 days, 15:59:37"}
+{"current_steps": 262, "total_steps": 34894, "loss": 2.3721, "learning_rate": 3.0028653295128943e-05, "epoch": 0.007508427278230935, "percentage": 0.75, "elapsed_time": "1:01:43", "remaining_time": "5 days, 15:59:08"}
+{"current_steps": 263, "total_steps": 34894, "loss": 2.3136, "learning_rate": 3.0143266475644704e-05, "epoch": 0.007537085397613495, "percentage": 0.75, "elapsed_time": "1:01:57", "remaining_time": "5 days, 15:58:41"}
+{"current_steps": 264, "total_steps": 34894, "loss": 2.327, "learning_rate": 3.0257879656160462e-05, "epoch": 0.007565743516996056, "percentage": 0.76, "elapsed_time": "1:02:11", "remaining_time": "5 days, 15:58:12"}
+{"current_steps": 265, "total_steps": 34894, "loss": 2.3403, "learning_rate": 3.037249283667622e-05, "epoch": 0.007594401636378617, "percentage": 0.76, "elapsed_time": "1:02:25", "remaining_time": "5 days, 15:57:43"}
+{"current_steps": 266, "total_steps": 34894, "loss": 2.3574, "learning_rate": 3.0487106017191982e-05, "epoch": 0.007623059755761177, "percentage": 0.76, "elapsed_time": "1:02:39", "remaining_time": "5 days, 15:57:14"}
+{"current_steps": 267, "total_steps": 34894, "loss": 2.3452, "learning_rate": 3.060171919770774e-05, "epoch": 0.0076517178751437384, "percentage": 0.77, "elapsed_time": "1:02:53", "remaining_time": "5 days, 15:57:01"}
+{"current_steps": 268, "total_steps": 34894, "loss": 2.3248, "learning_rate": 3.07163323782235e-05, "epoch": 0.0076803759945262996, "percentage": 0.77, "elapsed_time": "1:03:07", "remaining_time": "5 days, 15:56:35"}
+{"current_steps": 269, "total_steps": 34894, "loss": 2.3767, "learning_rate": 3.083094555873926e-05, "epoch": 0.00770903411390886, "percentage": 0.77, "elapsed_time": "1:03:21", "remaining_time": "5 days, 15:56:05"}
+{"current_steps": 270, "total_steps": 34894, "loss": 2.3773, "learning_rate": 3.0945558739255014e-05, "epoch": 0.007737692233291421, "percentage": 0.77, "elapsed_time": "1:03:35", "remaining_time": "5 days, 15:55:46"}
+{"current_steps": 271, "total_steps": 34894, "loss": 2.4337, "learning_rate": 3.1060171919770775e-05, "epoch": 0.007766350352673982, "percentage": 0.78, "elapsed_time": "1:03:49", "remaining_time": "5 days, 15:55:18"}
+{"current_steps": 272, "total_steps": 34894, "loss": 2.3875, "learning_rate": 3.117478510028654e-05, "epoch": 0.007795008472056542, "percentage": 0.78, "elapsed_time": "1:04:03", "remaining_time": "5 days, 15:54:50"}
+{"current_steps": 273, "total_steps": 34894, "loss": 2.3888, "learning_rate": 3.128939828080229e-05, "epoch": 0.007823666591439103, "percentage": 0.78, "elapsed_time": "1:04:18", "remaining_time": "5 days, 15:54:30"}
+{"current_steps": 274, "total_steps": 34894, "loss": 2.3491, "learning_rate": 3.140401146131805e-05, "epoch": 0.007852324710821664, "percentage": 0.79, "elapsed_time": "1:04:32", "remaining_time": "5 days, 15:54:02"}
+{"current_steps": 275, "total_steps": 34894, "loss": 2.3995, "learning_rate": 3.1518624641833814e-05, "epoch": 0.007880982830204225, "percentage": 0.79, "elapsed_time": "1:04:46", "remaining_time": "5 days, 15:53:36"}
+{"current_steps": 276, "total_steps": 34894, "loss": 2.4123, "learning_rate": 3.163323782234957e-05, "epoch": 0.007909640949586785, "percentage": 0.79, "elapsed_time": "1:05:00", "remaining_time": "5 days, 15:53:10"}
+{"current_steps": 277, "total_steps": 34894, "loss": 2.3314, "learning_rate": 3.174785100286533e-05, "epoch": 0.007938299068969347, "percentage": 0.79, "elapsed_time": "1:05:14", "remaining_time": "5 days, 15:52:48"}
+{"current_steps": 278, "total_steps": 34894, "loss": 2.41, "learning_rate": 3.186246418338109e-05, "epoch": 0.007966957188351907, "percentage": 0.8, "elapsed_time": "1:05:28", "remaining_time": "5 days, 15:52:21"}
+{"current_steps": 279, "total_steps": 34894, "loss": 2.3223, "learning_rate": 3.1977077363896846e-05, "epoch": 0.007995615307734467, "percentage": 0.8, "elapsed_time": "1:05:42", "remaining_time": "5 days, 15:51:58"}
+{"current_steps": 280, "total_steps": 34894, "loss": 2.3055, "learning_rate": 3.209169054441261e-05, "epoch": 0.00802427342711703, "percentage": 0.8, "elapsed_time": "1:05:56", "remaining_time": "5 days, 15:51:33"}
+{"current_steps": 281, "total_steps": 34894, "loss": 2.3598, "learning_rate": 3.220630372492837e-05, "epoch": 0.00805293154649959, "percentage": 0.81, "elapsed_time": "1:06:10", "remaining_time": "5 days, 15:51:07"}
+{"current_steps": 282, "total_steps": 34894, "loss": 2.3222, "learning_rate": 3.232091690544413e-05, "epoch": 0.00808158966588215, "percentage": 0.81, "elapsed_time": "1:06:24", "remaining_time": "5 days, 15:50:40"}
+{"current_steps": 283, "total_steps": 34894, "loss": 2.392, "learning_rate": 3.243553008595989e-05, "epoch": 0.008110247785264712, "percentage": 0.81, "elapsed_time": "1:06:38", "remaining_time": "5 days, 15:50:12"}
+{"current_steps": 284, "total_steps": 34894, "loss": 2.4117, "learning_rate": 3.2550143266475647e-05, "epoch": 0.008138905904647272, "percentage": 0.81, "elapsed_time": "1:06:52", "remaining_time": "5 days, 15:49:48"}
+{"current_steps": 285, "total_steps": 34894, "loss": 2.3431, "learning_rate": 3.266475644699141e-05, "epoch": 0.008167564024029832, "percentage": 0.82, "elapsed_time": "1:07:06", "remaining_time": "5 days, 15:49:29"}
+{"current_steps": 286, "total_steps": 34894, "loss": 2.3079, "learning_rate": 3.277936962750717e-05, "epoch": 0.008196222143412394, "percentage": 0.82, "elapsed_time": "1:07:20", "remaining_time": "5 days, 15:49:03"}
+{"current_steps": 287, "total_steps": 34894, "loss": 2.391, "learning_rate": 3.2893982808022924e-05, "epoch": 0.008224880262794955, "percentage": 0.82, "elapsed_time": "1:07:34", "remaining_time": "5 days, 15:48:37"}
+{"current_steps": 288, "total_steps": 34894, "loss": 2.3584, "learning_rate": 3.3008595988538685e-05, "epoch": 0.008253538382177515, "percentage": 0.83, "elapsed_time": "1:07:48", "remaining_time": "5 days, 15:48:11"}
+{"current_steps": 289, "total_steps": 34894, "loss": 2.3035, "learning_rate": 3.312320916905445e-05, "epoch": 0.008282196501560077, "percentage": 0.83, "elapsed_time": "1:08:02", "remaining_time": "5 days, 15:47:48"}
+{"current_steps": 290, "total_steps": 34894, "loss": 2.3755, "learning_rate": 3.32378223495702e-05, "epoch": 0.008310854620942637, "percentage": 0.83, "elapsed_time": "1:08:16", "remaining_time": "5 days, 15:47:36"}
+{"current_steps": 291, "total_steps": 34894, "loss": 2.3719, "learning_rate": 3.335243553008596e-05, "epoch": 0.008339512740325197, "percentage": 0.83, "elapsed_time": "1:08:30", "remaining_time": "5 days, 15:47:09"}
+{"current_steps": 292, "total_steps": 34894, "loss": 2.3112, "learning_rate": 3.3467048710601724e-05, "epoch": 0.00836817085970776, "percentage": 0.84, "elapsed_time": "1:08:44", "remaining_time": "5 days, 15:46:45"}
+{"current_steps": 293, "total_steps": 34894, "loss": 2.3658, "learning_rate": 3.358166189111748e-05, "epoch": 0.00839682897909032, "percentage": 0.84, "elapsed_time": "1:08:58", "remaining_time": "5 days, 15:46:22"}
+{"current_steps": 294, "total_steps": 34894, "loss": 2.3463, "learning_rate": 3.369627507163324e-05, "epoch": 0.00842548709847288, "percentage": 0.84, "elapsed_time": "1:09:13", "remaining_time": "5 days, 15:45:58"}
+{"current_steps": 295, "total_steps": 34894, "loss": 2.4016, "learning_rate": 3.3810888252149e-05, "epoch": 0.008454145217855442, "percentage": 0.85, "elapsed_time": "1:09:27", "remaining_time": "5 days, 15:45:57"}
+{"current_steps": 296, "total_steps": 34894, "loss": 2.3239, "learning_rate": 3.3925501432664757e-05, "epoch": 0.008482803337238002, "percentage": 0.85, "elapsed_time": "1:09:41", "remaining_time": "5 days, 15:46:09"}
+{"current_steps": 297, "total_steps": 34894, "loss": 2.3157, "learning_rate": 3.404011461318052e-05, "epoch": 0.008511461456620562, "percentage": 0.85, "elapsed_time": "1:09:55", "remaining_time": "5 days, 15:46:19"}
+{"current_steps": 298, "total_steps": 34894, "loss": 2.3857, "learning_rate": 3.415472779369628e-05, "epoch": 0.008540119576003124, "percentage": 0.85, "elapsed_time": "1:10:10", "remaining_time": "5 days, 15:46:24"}
+{"current_steps": 299, "total_steps": 34894, "loss": 2.3478, "learning_rate": 3.4269340974212034e-05, "epoch": 0.008568777695385684, "percentage": 0.86, "elapsed_time": "1:10:24", "remaining_time": "5 days, 15:46:00"}
+{"current_steps": 300, "total_steps": 34894, "loss": 2.3317, "learning_rate": 3.4383954154727795e-05, "epoch": 0.008597435814768245, "percentage": 0.86, "elapsed_time": "1:10:38", "remaining_time": "5 days, 15:45:34"}
+{"current_steps": 301, "total_steps": 34894, "loss": 2.316, "learning_rate": 3.449856733524356e-05, "epoch": 0.008626093934150807, "percentage": 0.86, "elapsed_time": "1:11:11", "remaining_time": "5 days, 16:21:51"}
+{"current_steps": 302, "total_steps": 34894, "loss": 2.3583, "learning_rate": 3.461318051575932e-05, "epoch": 0.008654752053533367, "percentage": 0.87, "elapsed_time": "1:11:25", "remaining_time": "5 days, 16:21:35"}
+{"current_steps": 303, "total_steps": 34894, "loss": 2.3098, "learning_rate": 3.472779369627507e-05, "epoch": 0.008683410172915927, "percentage": 0.87, "elapsed_time": "1:11:39", "remaining_time": "5 days, 16:21:26"}
+{"current_steps": 304, "total_steps": 34894, "loss": 2.2969, "learning_rate": 3.4842406876790834e-05, "epoch": 0.008712068292298489, "percentage": 0.87, "elapsed_time": "1:11:54", "remaining_time": "5 days, 16:21:05"}
+{"current_steps": 305, "total_steps": 34894, "loss": 2.3771, "learning_rate": 3.4957020057306596e-05, "epoch": 0.00874072641168105, "percentage": 0.87, "elapsed_time": "1:12:08", "remaining_time": "5 days, 16:20:53"}
+{"current_steps": 306, "total_steps": 34894, "loss": 2.2933, "learning_rate": 3.507163323782235e-05, "epoch": 0.00876938453106361, "percentage": 0.88, "elapsed_time": "1:12:22", "remaining_time": "5 days, 16:20:23"}
+{"current_steps": 307, "total_steps": 34894, "loss": 2.3708, "learning_rate": 3.518624641833811e-05, "epoch": 0.008798042650446172, "percentage": 0.88, "elapsed_time": "1:12:36", "remaining_time": "5 days, 16:19:56"}
+{"current_steps": 308, "total_steps": 34894, "loss": 2.3717, "learning_rate": 3.530085959885387e-05, "epoch": 0.008826700769828732, "percentage": 0.88, "elapsed_time": "1:12:50", "remaining_time": "5 days, 16:19:28"}
+{"current_steps": 309, "total_steps": 34894, "loss": 2.3385, "learning_rate": 3.541547277936963e-05, "epoch": 0.008855358889211292, "percentage": 0.89, "elapsed_time": "1:13:04", "remaining_time": "5 days, 16:19:07"}
+{"current_steps": 310, "total_steps": 34894, "loss": 2.3863, "learning_rate": 3.553008595988539e-05, "epoch": 0.008884017008593854, "percentage": 0.89, "elapsed_time": "1:13:18", "remaining_time": "5 days, 16:18:57"}
+{"current_steps": 311, "total_steps": 34894, "loss": 2.3724, "learning_rate": 3.564469914040115e-05, "epoch": 0.008912675127976414, "percentage": 0.89, "elapsed_time": "1:13:33", "remaining_time": "5 days, 16:18:53"}
+{"current_steps": 312, "total_steps": 34894, "loss": 2.3632, "learning_rate": 3.5759312320916905e-05, "epoch": 0.008941333247358975, "percentage": 0.89, "elapsed_time": "1:13:47", "remaining_time": "5 days, 16:18:33"}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ff41096313459f0734bada6b286cbcc680b1b6e801419929719ba85e1aa1c3a
+size 7224