Training in progress, step 1908

.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

adapter_config.json

@@ -0,0 +1,17 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
+  "encoder_dropout": 0.1,
+  "encoder_hidden_size": 768,
+  "encoder_num_layers": 2,
+  "encoder_reparameterization_type": "MLP",
+  "inference_mode": true,
+  "num_attention_heads": 32,
+  "num_layers": 32,
+  "num_transformer_submodules": 1,
+  "num_virtual_tokens": 100,
+  "peft_type": "P_TUNING",
+  "revision": null,
+  "task_type": "CAUSAL_LM",
+  "token_dim": 4096
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b325886cca5e494139aa845feac2caeffc5df55a82bdbf161d9723cb651a148
+size 1638528

special_tokens_map.json

@@ -0,0 +1,26 @@
+{
+  "additional_special_tokens": [
+    {
+      "content": "<|eom_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    }
+  ],
+  "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": "<|eot_id|>"
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8c1dcab308e7cf5970ea38815e0a62887d705c5b436f869ca27a5dcdd40c36a6
+size 17210148

tokenizer_config.json

@@ -0,0 +1,2077 @@
+{
+  "added_tokens_decoder": {
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_248|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_249|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_250|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128256": {
+      "content": "<|eom_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|eom_id|>"
+  ],
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "extra_special_tokens": {},
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|eot_id|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "PreTrainedTokenizer"
+}

train.yaml

@@ -0,0 +1,60 @@
+seed: 42
+
+### model
+model_name_or_path: meta-llama/Meta-Llama-3-8B-Instruct
+trust_remote_code: true
+flash_attn: auto
+use_cache: false
+
+### method
+stage: sft
+do_train: true
+finetuning_type: p-tuning
+task_type: CAUSAL_LM
+num_virtual_tokens: 100
+encoder_reparameterization_type: MLP
+encoder_hidden_size: 768
+encoder_dropout: 0.1
+
+### dataset
+dataset: codealpacapy
+template: llama3
+cutoff_len: 2048
+overwrite_cache: true
+preprocessing_num_workers: 8
+dataloader_num_workers: 2
+packing: false
+
+### output
+output_dir: saves_multiple/p-tuning/llama-3-8b-instruct/train_codealpacapy_42_1760664611
+logging_steps: 5
+save_steps: 0.05
+overwrite_output_dir: true
+save_only_model: false
+plot_loss: true
+include_num_input_tokens_seen: true
+push_to_hub: true
+push_to_hub_organization: rbelanec
+load_best_model_at_end: true
+save_total_limit: 1
+
+### train
+per_device_train_batch_size: 4
+learning_rate: 1.0e-3
+num_train_epochs: 20
+weight_decay: 1.0e-5
+lr_scheduler_type: cosine
+bf16: true
+ddp_timeout: 180000000
+resume_from_checkpoint: null
+warmup_ratio: 0.1
+optim: adamw_torch
+report_to:
+- wandb
+run_name: p-tuning_llama-3-8b-instruct_train_codealpacapy_42_1760664611
+
+### eval
+per_device_eval_batch_size: 4
+eval_strategy: steps
+eval_steps: 0.05
+val_size: 0.1

trainer_log.jsonl

@@ -0,0 +1,388 @@
+{"current_steps": 5, "total_steps": 38160, "loss": 4.3366, "lr": 1.0482180293501048e-06, "epoch": 0.002620545073375262, "percentage": 0.01, "elapsed_time": "0:00:01", "remaining_time": "3:07:18", "throughput": 2216.28, "total_tokens": 3264}
+{"current_steps": 10, "total_steps": 38160, "loss": 4.3961, "lr": 2.358490566037736e-06, "epoch": 0.005241090146750524, "percentage": 0.03, "elapsed_time": "0:00:02", "remaining_time": "2:14:55", "throughput": 2774.8, "total_tokens": 5888}
+{"current_steps": 15, "total_steps": 38160, "loss": 4.3377, "lr": 3.668763102725367e-06, "epoch": 0.007861635220125786, "percentage": 0.04, "elapsed_time": "0:00:02", "remaining_time": "1:57:10", "throughput": 3067.47, "total_tokens": 8480}
+{"current_steps": 20, "total_steps": 38160, "loss": 3.2407, "lr": 4.979035639412998e-06, "epoch": 0.010482180293501049, "percentage": 0.05, "elapsed_time": "0:00:03", "remaining_time": "1:49:54", "throughput": 3322.21, "total_tokens": 11488}
+{"current_steps": 25, "total_steps": 38160, "loss": 2.08, "lr": 6.289308176100629e-06, "epoch": 0.01310272536687631, "percentage": 0.07, "elapsed_time": "0:00:04", "remaining_time": "1:46:51", "throughput": 3540.33, "total_tokens": 14880}
+{"current_steps": 30, "total_steps": 38160, "loss": 1.0257, "lr": 7.59958071278826e-06, "epoch": 0.015723270440251572, "percentage": 0.08, "elapsed_time": "0:00:04", "remaining_time": "1:45:36", "throughput": 3728.97, "total_tokens": 18592}
+{"current_steps": 35, "total_steps": 38160, "loss": 0.8789, "lr": 8.90985324947589e-06, "epoch": 0.018343815513626835, "percentage": 0.09, "elapsed_time": "0:00:05", "remaining_time": "1:42:52", "throughput": 3806.28, "total_tokens": 21568}
+{"current_steps": 40, "total_steps": 38160, "loss": 0.7281, "lr": 1.0220125786163522e-05, "epoch": 0.020964360587002098, "percentage": 0.1, "elapsed_time": "0:00:06", "remaining_time": "1:40:17", "throughput": 3826.35, "total_tokens": 24160}
+{"current_steps": 45, "total_steps": 38160, "loss": 0.7915, "lr": 1.1530398322851153e-05, "epoch": 0.02358490566037736, "percentage": 0.12, "elapsed_time": "0:00:07", "remaining_time": "1:38:53", "throughput": 3878.35, "total_tokens": 27168}
+{"current_steps": 50, "total_steps": 38160, "loss": 0.6498, "lr": 1.2840670859538784e-05, "epoch": 0.02620545073375262, "percentage": 0.13, "elapsed_time": "0:00:07", "remaining_time": "1:37:41", "throughput": 3911.3, "total_tokens": 30080}
+{"current_steps": 55, "total_steps": 38160, "loss": 0.7485, "lr": 1.4150943396226415e-05, "epoch": 0.028825995807127882, "percentage": 0.14, "elapsed_time": "0:00:08", "remaining_time": "1:36:10", "throughput": 3915.27, "total_tokens": 32608}
+{"current_steps": 60, "total_steps": 38160, "loss": 0.7152, "lr": 1.5461215932914046e-05, "epoch": 0.031446540880503145, "percentage": 0.16, "elapsed_time": "0:00:09", "remaining_time": "1:36:42", "throughput": 3988.73, "total_tokens": 36448}
+{"current_steps": 65, "total_steps": 38160, "loss": 0.4345, "lr": 1.6771488469601677e-05, "epoch": 0.034067085953878404, "percentage": 0.17, "elapsed_time": "0:00:09", "remaining_time": "1:35:21", "throughput": 3989.06, "total_tokens": 38944}
+{"current_steps": 70, "total_steps": 38160, "loss": 0.5775, "lr": 1.8081761006289308e-05, "epoch": 0.03668763102725367, "percentage": 0.18, "elapsed_time": "0:00:10", "remaining_time": "1:34:54", "throughput": 3996.32, "total_tokens": 41824}
+{"current_steps": 75, "total_steps": 38160, "loss": 0.5619, "lr": 1.9392033542976942e-05, "epoch": 0.03930817610062893, "percentage": 0.2, "elapsed_time": "0:00:11", "remaining_time": "1:34:30", "throughput": 4020.44, "total_tokens": 44896}
+{"current_steps": 80, "total_steps": 38160, "loss": 0.5432, "lr": 2.070230607966457e-05, "epoch": 0.041928721174004195, "percentage": 0.21, "elapsed_time": "0:00:11", "remaining_time": "1:33:44", "throughput": 4030.0, "total_tokens": 47616}
+{"current_steps": 85, "total_steps": 38160, "loss": 0.6556, "lr": 2.20125786163522e-05, "epoch": 0.044549266247379454, "percentage": 0.22, "elapsed_time": "0:00:12", "remaining_time": "1:33:08", "throughput": 4037.26, "total_tokens": 50368}
+{"current_steps": 90, "total_steps": 38160, "loss": 0.6779, "lr": 2.3322851153039832e-05, "epoch": 0.04716981132075472, "percentage": 0.24, "elapsed_time": "0:00:13", "remaining_time": "1:33:28", "throughput": 4073.76, "total_tokens": 54016}
+{"current_steps": 95, "total_steps": 38160, "loss": 0.4229, "lr": 2.4633123689727463e-05, "epoch": 0.04979035639412998, "percentage": 0.25, "elapsed_time": "0:00:14", "remaining_time": "1:33:52", "throughput": 4118.36, "total_tokens": 57888}
+{"current_steps": 100, "total_steps": 38160, "loss": 0.5745, "lr": 2.5943396226415097e-05, "epoch": 0.05241090146750524, "percentage": 0.26, "elapsed_time": "0:00:14", "remaining_time": "1:33:34", "throughput": 4136.39, "total_tokens": 61024}
+{"current_steps": 105, "total_steps": 38160, "loss": 0.5809, "lr": 2.7253668763102725e-05, "epoch": 0.055031446540880505, "percentage": 0.28, "elapsed_time": "0:00:15", "remaining_time": "1:33:42", "throughput": 4168.55, "total_tokens": 64672}
+{"current_steps": 110, "total_steps": 38160, "loss": 0.5085, "lr": 2.8563941299790356e-05, "epoch": 0.057651991614255764, "percentage": 0.29, "elapsed_time": "0:00:16", "remaining_time": "1:33:22", "throughput": 4174.8, "total_tokens": 67616}
+{"current_steps": 115, "total_steps": 38160, "loss": 0.521, "lr": 2.987421383647799e-05, "epoch": 0.06027253668763103, "percentage": 0.3, "elapsed_time": "0:00:16", "remaining_time": "1:33:30", "throughput": 4200.05, "total_tokens": 71232}
+{"current_steps": 120, "total_steps": 38160, "loss": 0.4112, "lr": 3.118448637316562e-05, "epoch": 0.06289308176100629, "percentage": 0.31, "elapsed_time": "0:00:17", "remaining_time": "1:33:41", "throughput": 4222.73, "total_tokens": 74880}
+{"current_steps": 125, "total_steps": 38160, "loss": 0.5647, "lr": 3.2494758909853245e-05, "epoch": 0.06551362683438156, "percentage": 0.33, "elapsed_time": "0:00:18", "remaining_time": "1:33:46", "throughput": 4241.63, "total_tokens": 78432}
+{"current_steps": 130, "total_steps": 38160, "loss": 0.4921, "lr": 3.380503144654088e-05, "epoch": 0.06813417190775681, "percentage": 0.34, "elapsed_time": "0:00:19", "remaining_time": "1:33:17", "throughput": 4237.51, "total_tokens": 81088}
+{"current_steps": 135, "total_steps": 38160, "loss": 0.5339, "lr": 3.5115303983228514e-05, "epoch": 0.07075471698113207, "percentage": 0.35, "elapsed_time": "0:00:19", "remaining_time": "1:33:29", "throughput": 4262.7, "total_tokens": 84896}
+{"current_steps": 140, "total_steps": 38160, "loss": 0.5942, "lr": 3.642557651991614e-05, "epoch": 0.07337526205450734, "percentage": 0.37, "elapsed_time": "0:00:20", "remaining_time": "1:33:50", "throughput": 4279.87, "total_tokens": 88736}
+{"current_steps": 145, "total_steps": 38160, "loss": 0.8808, "lr": 3.7735849056603776e-05, "epoch": 0.0759958071278826, "percentage": 0.38, "elapsed_time": "0:00:21", "remaining_time": "1:33:52", "throughput": 4292.92, "total_tokens": 92224}
+{"current_steps": 150, "total_steps": 38160, "loss": 0.5362, "lr": 3.90461215932914e-05, "epoch": 0.07861635220125786, "percentage": 0.39, "elapsed_time": "0:00:22", "remaining_time": "1:33:37", "throughput": 4293.23, "total_tokens": 95168}
+{"current_steps": 155, "total_steps": 38160, "loss": 0.5562, "lr": 4.035639412997904e-05, "epoch": 0.08123689727463312, "percentage": 0.41, "elapsed_time": "0:00:22", "remaining_time": "1:33:50", "throughput": 4310.07, "total_tokens": 98976}
+{"current_steps": 160, "total_steps": 38160, "loss": 0.469, "lr": 4.1666666666666665e-05, "epoch": 0.08385744234800839, "percentage": 0.42, "elapsed_time": "0:00:23", "remaining_time": "1:33:44", "throughput": 4314.39, "total_tokens": 102176}
+{"current_steps": 165, "total_steps": 38160, "loss": 0.464, "lr": 4.29769392033543e-05, "epoch": 0.08647798742138364, "percentage": 0.43, "elapsed_time": "0:00:24", "remaining_time": "1:33:52", "throughput": 4326.48, "total_tokens": 105824}
+{"current_steps": 170, "total_steps": 38160, "loss": 0.5105, "lr": 4.4287211740041934e-05, "epoch": 0.08909853249475891, "percentage": 0.45, "elapsed_time": "0:00:25", "remaining_time": "1:33:31", "throughput": 4321.19, "total_tokens": 108512}
+{"current_steps": 175, "total_steps": 38160, "loss": 0.523, "lr": 4.5597484276729555e-05, "epoch": 0.09171907756813417, "percentage": 0.46, "elapsed_time": "0:00:25", "remaining_time": "1:33:31", "throughput": 4327.47, "total_tokens": 111872}
+{"current_steps": 180, "total_steps": 38160, "loss": 0.4721, "lr": 4.690775681341719e-05, "epoch": 0.09433962264150944, "percentage": 0.47, "elapsed_time": "0:00:26", "remaining_time": "1:33:35", "throughput": 4341.91, "total_tokens": 115552}
+{"current_steps": 185, "total_steps": 38160, "loss": 0.6313, "lr": 4.8218029350104823e-05, "epoch": 0.09696016771488469, "percentage": 0.48, "elapsed_time": "0:00:27", "remaining_time": "1:33:31", "throughput": 4345.34, "total_tokens": 118784}
+{"current_steps": 190, "total_steps": 38160, "loss": 0.4303, "lr": 4.952830188679246e-05, "epoch": 0.09958071278825996, "percentage": 0.5, "elapsed_time": "0:00:28", "remaining_time": "1:33:39", "throughput": 4357.49, "total_tokens": 122528}
+{"current_steps": 195, "total_steps": 38160, "loss": 0.5259, "lr": 5.0838574423480085e-05, "epoch": 0.10220125786163523, "percentage": 0.51, "elapsed_time": "0:00:28", "remaining_time": "1:33:16", "throughput": 4349.25, "total_tokens": 125024}
+{"current_steps": 200, "total_steps": 38160, "loss": 0.5996, "lr": 5.214884696016771e-05, "epoch": 0.10482180293501048, "percentage": 0.52, "elapsed_time": "0:00:29", "remaining_time": "1:33:09", "throughput": 4349.31, "total_tokens": 128096}
+{"current_steps": 205, "total_steps": 38160, "loss": 0.6745, "lr": 5.345911949685535e-05, "epoch": 0.10744234800838574, "percentage": 0.54, "elapsed_time": "0:00:30", "remaining_time": "1:32:46", "throughput": 4340.7, "total_tokens": 130496}
+{"current_steps": 210, "total_steps": 38160, "loss": 0.5892, "lr": 5.4769392033542975e-05, "epoch": 0.11006289308176101, "percentage": 0.55, "elapsed_time": "0:00:30", "remaining_time": "1:32:29", "throughput": 4335.98, "total_tokens": 133152}
+{"current_steps": 215, "total_steps": 38160, "loss": 0.6939, "lr": 5.607966457023061e-05, "epoch": 0.11268343815513626, "percentage": 0.56, "elapsed_time": "0:00:31", "remaining_time": "1:32:26", "throughput": 4339.45, "total_tokens": 136384}
+{"current_steps": 220, "total_steps": 38160, "loss": 0.6202, "lr": 5.7389937106918244e-05, "epoch": 0.11530398322851153, "percentage": 0.58, "elapsed_time": "0:00:32", "remaining_time": "1:32:19", "throughput": 4341.18, "total_tokens": 139456}
+{"current_steps": 225, "total_steps": 38160, "loss": 0.626, "lr": 5.870020964360587e-05, "epoch": 0.1179245283018868, "percentage": 0.59, "elapsed_time": "0:00:32", "remaining_time": "1:32:19", "throughput": 4346.4, "total_tokens": 142816}
+{"current_steps": 230, "total_steps": 38160, "loss": 0.5322, "lr": 6.00104821802935e-05, "epoch": 0.12054507337526206, "percentage": 0.6, "elapsed_time": "0:00:33", "remaining_time": "1:32:09", "throughput": 4345.53, "total_tokens": 145696}
+{"current_steps": 235, "total_steps": 38160, "loss": 0.6419, "lr": 6.132075471698113e-05, "epoch": 0.12316561844863731, "percentage": 0.62, "elapsed_time": "0:00:34", "remaining_time": "1:32:08", "throughput": 4350.41, "total_tokens": 149024}
+{"current_steps": 240, "total_steps": 38160, "loss": 0.5641, "lr": 6.263102725366875e-05, "epoch": 0.12578616352201258, "percentage": 0.63, "elapsed_time": "0:00:34", "remaining_time": "1:31:58", "throughput": 4347.37, "total_tokens": 151840}
+{"current_steps": 245, "total_steps": 38160, "loss": 0.6886, "lr": 6.39412997903564e-05, "epoch": 0.12840670859538783, "percentage": 0.64, "elapsed_time": "0:00:35", "remaining_time": "1:32:04", "throughput": 4354.86, "total_tokens": 155456}
+{"current_steps": 250, "total_steps": 38160, "loss": 0.5696, "lr": 6.525157232704402e-05, "epoch": 0.1310272536687631, "percentage": 0.66, "elapsed_time": "0:00:36", "remaining_time": "1:31:59", "throughput": 4355.42, "total_tokens": 158528}
+{"current_steps": 255, "total_steps": 38160, "loss": 0.5468, "lr": 6.656184486373166e-05, "epoch": 0.13364779874213836, "percentage": 0.67, "elapsed_time": "0:00:37", "remaining_time": "1:33:36", "throughput": 4406.81, "total_tokens": 166496}
+{"current_steps": 260, "total_steps": 38160, "loss": 0.5205, "lr": 6.787211740041929e-05, "epoch": 0.13626834381551362, "percentage": 0.68, "elapsed_time": "0:00:38", "remaining_time": "1:33:29", "throughput": 4406.65, "total_tokens": 169568}
+{"current_steps": 265, "total_steps": 38160, "loss": 0.6078, "lr": 6.918238993710691e-05, "epoch": 0.1388888888888889, "percentage": 0.69, "elapsed_time": "0:00:39", "remaining_time": "1:33:24", "throughput": 4406.91, "total_tokens": 172704}
+{"current_steps": 270, "total_steps": 38160, "loss": 0.7384, "lr": 7.049266247379455e-05, "epoch": 0.14150943396226415, "percentage": 0.71, "elapsed_time": "0:00:39", "remaining_time": "1:33:30", "throughput": 4414.05, "total_tokens": 176480}
+{"current_steps": 275, "total_steps": 38160, "loss": 0.5469, "lr": 7.180293501048218e-05, "epoch": 0.1441299790356394, "percentage": 0.72, "elapsed_time": "0:00:40", "remaining_time": "1:33:57", "throughput": 4433.35, "total_tokens": 181408}
+{"current_steps": 280, "total_steps": 38160, "loss": 0.5923, "lr": 7.311320754716982e-05, "epoch": 0.14675052410901468, "percentage": 0.73, "elapsed_time": "0:00:41", "remaining_time": "1:33:54", "throughput": 4435.18, "total_tokens": 184736}
+{"current_steps": 285, "total_steps": 38160, "loss": 0.5122, "lr": 7.442348008385745e-05, "epoch": 0.14937106918238993, "percentage": 0.75, "elapsed_time": "0:00:42", "remaining_time": "1:33:56", "throughput": 4440.69, "total_tokens": 188352}
+{"current_steps": 290, "total_steps": 38160, "loss": 0.4922, "lr": 7.573375262054507e-05, "epoch": 0.1519916142557652, "percentage": 0.76, "elapsed_time": "0:00:43", "remaining_time": "1:34:26", "throughput": 4457.2, "total_tokens": 193408}
+{"current_steps": 295, "total_steps": 38160, "loss": 0.4207, "lr": 7.70440251572327e-05, "epoch": 0.15461215932914046, "percentage": 0.77, "elapsed_time": "0:00:44", "remaining_time": "1:34:17", "throughput": 4455.63, "total_tokens": 196384}
+{"current_steps": 300, "total_steps": 38160, "loss": 0.6115, "lr": 7.835429769392034e-05, "epoch": 0.15723270440251572, "percentage": 0.79, "elapsed_time": "0:00:44", "remaining_time": "1:34:09", "throughput": 4453.36, "total_tokens": 199360}
+{"current_steps": 305, "total_steps": 38160, "loss": 0.5228, "lr": 7.966457023060797e-05, "epoch": 0.159853249475891, "percentage": 0.8, "elapsed_time": "0:00:45", "remaining_time": "1:34:10", "throughput": 4455.43, "total_tokens": 202848}
+{"current_steps": 310, "total_steps": 38160, "loss": 0.4791, "lr": 8.09748427672956e-05, "epoch": 0.16247379454926625, "percentage": 0.81, "elapsed_time": "0:00:46", "remaining_time": "1:34:25", "throughput": 4463.62, "total_tokens": 207104}
+{"current_steps": 315, "total_steps": 38160, "loss": 0.637, "lr": 8.228511530398323e-05, "epoch": 0.1650943396226415, "percentage": 0.83, "elapsed_time": "0:00:47", "remaining_time": "1:34:09", "throughput": 4457.23, "total_tokens": 209600}
+{"current_steps": 320, "total_steps": 38160, "loss": 0.611, "lr": 8.359538784067086e-05, "epoch": 0.16771488469601678, "percentage": 0.84, "elapsed_time": "0:00:47", "remaining_time": "1:34:11", "throughput": 4459.71, "total_tokens": 213152}
+{"current_steps": 325, "total_steps": 38160, "loss": 0.7027, "lr": 8.490566037735848e-05, "epoch": 0.17033542976939203, "percentage": 0.85, "elapsed_time": "0:00:48", "remaining_time": "1:33:56", "throughput": 4453.48, "total_tokens": 215616}
+{"current_steps": 330, "total_steps": 38160, "loss": 0.6089, "lr": 8.621593291404613e-05, "epoch": 0.17295597484276728, "percentage": 0.86, "elapsed_time": "0:00:49", "remaining_time": "1:33:48", "throughput": 4451.0, "total_tokens": 218528}
+{"current_steps": 335, "total_steps": 38160, "loss": 0.4405, "lr": 8.752620545073375e-05, "epoch": 0.17557651991614256, "percentage": 0.88, "elapsed_time": "0:00:49", "remaining_time": "1:33:46", "throughput": 4451.53, "total_tokens": 221824}
+{"current_steps": 340, "total_steps": 38160, "loss": 0.5055, "lr": 8.883647798742137e-05, "epoch": 0.17819706498951782, "percentage": 0.89, "elapsed_time": "0:00:50", "remaining_time": "1:33:46", "throughput": 4453.62, "total_tokens": 225280}
+{"current_steps": 345, "total_steps": 38160, "loss": 0.4572, "lr": 9.014675052410902e-05, "epoch": 0.18081761006289307, "percentage": 0.9, "elapsed_time": "0:00:51", "remaining_time": "1:34:20", "throughput": 4471.0, "total_tokens": 230912}
+{"current_steps": 350, "total_steps": 38160, "loss": 0.5646, "lr": 9.145702306079664e-05, "epoch": 0.18343815513626835, "percentage": 0.92, "elapsed_time": "0:00:52", "remaining_time": "1:34:22", "throughput": 4475.13, "total_tokens": 234592}
+{"current_steps": 355, "total_steps": 38160, "loss": 0.493, "lr": 9.276729559748428e-05, "epoch": 0.1860587002096436, "percentage": 0.93, "elapsed_time": "0:00:53", "remaining_time": "1:34:16", "throughput": 4473.31, "total_tokens": 237600}
+{"current_steps": 360, "total_steps": 38160, "loss": 0.6188, "lr": 9.407756813417191e-05, "epoch": 0.18867924528301888, "percentage": 0.94, "elapsed_time": "0:00:53", "remaining_time": "1:34:16", "throughput": 4475.5, "total_tokens": 241088}
+{"current_steps": 365, "total_steps": 38160, "loss": 0.5569, "lr": 9.538784067085953e-05, "epoch": 0.19129979035639413, "percentage": 0.96, "elapsed_time": "0:00:54", "remaining_time": "1:34:01", "throughput": 4469.52, "total_tokens": 243520}
+{"current_steps": 370, "total_steps": 38160, "loss": 0.742, "lr": 9.669811320754718e-05, "epoch": 0.19392033542976939, "percentage": 0.97, "elapsed_time": "0:00:55", "remaining_time": "1:34:07", "throughput": 4474.94, "total_tokens": 247456}
+{"current_steps": 375, "total_steps": 38160, "loss": 0.4974, "lr": 9.80083857442348e-05, "epoch": 0.19654088050314467, "percentage": 0.98, "elapsed_time": "0:00:56", "remaining_time": "1:34:05", "throughput": 4476.09, "total_tokens": 250784}
+{"current_steps": 380, "total_steps": 38160, "loss": 0.542, "lr": 9.931865828092243e-05, "epoch": 0.19916142557651992, "percentage": 1.0, "elapsed_time": "0:00:56", "remaining_time": "1:34:15", "throughput": 4483.88, "total_tokens": 255040}
+{"current_steps": 385, "total_steps": 38160, "loss": 0.6188, "lr": 0.00010062893081761007, "epoch": 0.20178197064989517, "percentage": 1.01, "elapsed_time": "0:00:57", "remaining_time": "1:34:11", "throughput": 4483.57, "total_tokens": 258240}
+{"current_steps": 390, "total_steps": 38160, "loss": 0.4524, "lr": 0.0001019392033542977, "epoch": 0.20440251572327045, "percentage": 1.02, "elapsed_time": "0:00:58", "remaining_time": "1:34:04", "throughput": 4480.67, "total_tokens": 261152}
+{"current_steps": 395, "total_steps": 38160, "loss": 0.5425, "lr": 0.00010324947589098532, "epoch": 0.2070230607966457, "percentage": 1.04, "elapsed_time": "0:00:59", "remaining_time": "1:34:02", "throughput": 4482.12, "total_tokens": 264544}
+{"current_steps": 400, "total_steps": 38160, "loss": 0.6098, "lr": 0.00010455974842767296, "epoch": 0.20964360587002095, "percentage": 1.05, "elapsed_time": "0:00:59", "remaining_time": "1:33:52", "throughput": 4477.24, "total_tokens": 267136}
+{"current_steps": 405, "total_steps": 38160, "loss": 0.4776, "lr": 0.00010587002096436059, "epoch": 0.21226415094339623, "percentage": 1.06, "elapsed_time": "0:01:00", "remaining_time": "1:33:50", "throughput": 4477.83, "total_tokens": 270464}
+{"current_steps": 410, "total_steps": 38160, "loss": 0.5049, "lr": 0.00010718029350104821, "epoch": 0.2148846960167715, "percentage": 1.07, "elapsed_time": "0:01:01", "remaining_time": "1:33:57", "throughput": 4483.44, "total_tokens": 274528}
+{"current_steps": 415, "total_steps": 38160, "loss": 0.4909, "lr": 0.00010849056603773586, "epoch": 0.21750524109014674, "percentage": 1.09, "elapsed_time": "0:01:01", "remaining_time": "1:33:51", "throughput": 4479.8, "total_tokens": 277376}
+{"current_steps": 420, "total_steps": 38160, "loss": 0.395, "lr": 0.00010980083857442348, "epoch": 0.22012578616352202, "percentage": 1.1, "elapsed_time": "0:01:02", "remaining_time": "1:33:38", "throughput": 4472.27, "total_tokens": 279616}
+{"current_steps": 425, "total_steps": 38160, "loss": 0.4555, "lr": 0.0001111111111111111, "epoch": 0.22274633123689727, "percentage": 1.11, "elapsed_time": "0:01:03", "remaining_time": "1:33:33", "throughput": 4470.51, "total_tokens": 282656}
+{"current_steps": 430, "total_steps": 38160, "loss": 0.5687, "lr": 0.00011242138364779875, "epoch": 0.22536687631027252, "percentage": 1.13, "elapsed_time": "0:01:03", "remaining_time": "1:33:27", "throughput": 4469.01, "total_tokens": 285600}
+{"current_steps": 435, "total_steps": 38160, "loss": 0.395, "lr": 0.00011373165618448637, "epoch": 0.2279874213836478, "percentage": 1.14, "elapsed_time": "0:01:04", "remaining_time": "1:33:32", "throughput": 4474.17, "total_tokens": 289568}
+{"current_steps": 440, "total_steps": 38160, "loss": 0.592, "lr": 0.00011504192872117402, "epoch": 0.23060796645702306, "percentage": 1.15, "elapsed_time": "0:01:05", "remaining_time": "1:33:30", "throughput": 4473.79, "total_tokens": 292768}
+{"current_steps": 445, "total_steps": 38160, "loss": 0.6336, "lr": 0.00011635220125786164, "epoch": 0.23322851153039834, "percentage": 1.17, "elapsed_time": "0:01:06", "remaining_time": "1:33:21", "throughput": 4470.61, "total_tokens": 295456}
+{"current_steps": 450, "total_steps": 38160, "loss": 0.5672, "lr": 0.00011766247379454926, "epoch": 0.2358490566037736, "percentage": 1.18, "elapsed_time": "0:01:06", "remaining_time": "1:33:27", "throughput": 4475.96, "total_tokens": 299488}
+{"current_steps": 455, "total_steps": 38160, "loss": 0.4676, "lr": 0.00011897274633123691, "epoch": 0.23846960167714884, "percentage": 1.19, "elapsed_time": "0:01:07", "remaining_time": "1:33:28", "throughput": 4476.61, "total_tokens": 302976}
+{"current_steps": 460, "total_steps": 38160, "loss": 0.5342, "lr": 0.00012028301886792453, "epoch": 0.24109014675052412, "percentage": 1.21, "elapsed_time": "0:01:08", "remaining_time": "1:33:28", "throughput": 4479.35, "total_tokens": 306560}
+{"current_steps": 465, "total_steps": 38160, "loss": 0.4462, "lr": 0.00012159329140461216, "epoch": 0.24371069182389937, "percentage": 1.22, "elapsed_time": "0:01:09", "remaining_time": "1:33:23", "throughput": 4477.61, "total_tokens": 309536}
+{"current_steps": 470, "total_steps": 38160, "loss": 0.5008, "lr": 0.0001229035639412998, "epoch": 0.24633123689727462, "percentage": 1.23, "elapsed_time": "0:01:09", "remaining_time": "1:33:18", "throughput": 4477.11, "total_tokens": 312576}
+{"current_steps": 475, "total_steps": 38160, "loss": 0.4987, "lr": 0.00012421383647798743, "epoch": 0.2489517819706499, "percentage": 1.24, "elapsed_time": "0:01:10", "remaining_time": "1:33:13", "throughput": 4476.34, "total_tokens": 315616}
+{"current_steps": 480, "total_steps": 38160, "loss": 0.5167, "lr": 0.00012552410901467507, "epoch": 0.25157232704402516, "percentage": 1.26, "elapsed_time": "0:01:11", "remaining_time": "1:33:13", "throughput": 4478.42, "total_tokens": 319136}
+{"current_steps": 485, "total_steps": 38160, "loss": 0.6978, "lr": 0.00012683438155136267, "epoch": 0.25419287211740044, "percentage": 1.27, "elapsed_time": "0:01:11", "remaining_time": "1:33:02", "throughput": 4471.92, "total_tokens": 321376}
+{"current_steps": 490, "total_steps": 38160, "loss": 0.5501, "lr": 0.0001281446540880503, "epoch": 0.25681341719077566, "percentage": 1.28, "elapsed_time": "0:01:12", "remaining_time": "1:33:02", "throughput": 4474.76, "total_tokens": 324928}
+{"current_steps": 495, "total_steps": 38160, "loss": 0.6535, "lr": 0.00012945492662473794, "epoch": 0.25943396226415094, "percentage": 1.3, "elapsed_time": "0:01:13", "remaining_time": "1:33:11", "throughput": 4480.87, "total_tokens": 329248}
+{"current_steps": 500, "total_steps": 38160, "loss": 0.6412, "lr": 0.00013076519916142558, "epoch": 0.2620545073375262, "percentage": 1.31, "elapsed_time": "0:01:14", "remaining_time": "1:33:06", "throughput": 4480.1, "total_tokens": 332288}
+{"current_steps": 505, "total_steps": 38160, "loss": 0.5217, "lr": 0.0001320754716981132, "epoch": 0.26467505241090145, "percentage": 1.32, "elapsed_time": "0:01:14", "remaining_time": "1:32:56", "throughput": 4476.05, "total_tokens": 334752}
+{"current_steps": 510, "total_steps": 38160, "loss": 0.5572, "lr": 0.00013338574423480085, "epoch": 0.2672955974842767, "percentage": 1.34, "elapsed_time": "0:01:15", "remaining_time": "1:32:53", "throughput": 4475.13, "total_tokens": 337856}
+{"current_steps": 515, "total_steps": 38160, "loss": 0.6406, "lr": 0.00013469601677148845, "epoch": 0.269916142557652, "percentage": 1.35, "elapsed_time": "0:01:16", "remaining_time": "1:32:57", "throughput": 4479.04, "total_tokens": 341792}
+{"current_steps": 520, "total_steps": 38160, "loss": 0.6394, "lr": 0.0001360062893081761, "epoch": 0.27253668763102723, "percentage": 1.36, "elapsed_time": "0:01:16", "remaining_time": "1:32:53", "throughput": 4477.69, "total_tokens": 344768}
+{"current_steps": 525, "total_steps": 38160, "loss": 0.4626, "lr": 0.00013731656184486375, "epoch": 0.2751572327044025, "percentage": 1.38, "elapsed_time": "0:01:17", "remaining_time": "1:32:45", "throughput": 4474.22, "total_tokens": 347360}
+{"current_steps": 530, "total_steps": 38160, "loss": 0.4874, "lr": 0.00013862683438155136, "epoch": 0.2777777777777778, "percentage": 1.39, "elapsed_time": "0:01:18", "remaining_time": "1:32:43", "throughput": 4474.55, "total_tokens": 350592}
+{"current_steps": 535, "total_steps": 38160, "loss": 0.491, "lr": 0.000139937106918239, "epoch": 0.280398322851153, "percentage": 1.4, "elapsed_time": "0:01:19", "remaining_time": "1:32:39", "throughput": 4472.75, "total_tokens": 353568}
+{"current_steps": 540, "total_steps": 38160, "loss": 0.6103, "lr": 0.00014124737945492663, "epoch": 0.2830188679245283, "percentage": 1.42, "elapsed_time": "0:01:19", "remaining_time": "1:32:32", "throughput": 4470.57, "total_tokens": 356320}
+{"current_steps": 545, "total_steps": 38160, "loss": 0.5264, "lr": 0.00014255765199161423, "epoch": 0.2856394129979036, "percentage": 1.43, "elapsed_time": "0:01:20", "remaining_time": "1:32:33", "throughput": 4473.59, "total_tokens": 359968}
+{"current_steps": 550, "total_steps": 38160, "loss": 0.4495, "lr": 0.0001438679245283019, "epoch": 0.2882599580712788, "percentage": 1.44, "elapsed_time": "0:01:21", "remaining_time": "1:32:27", "throughput": 4471.39, "total_tokens": 362752}
+{"current_steps": 555, "total_steps": 38160, "loss": 0.5891, "lr": 0.00014517819706498953, "epoch": 0.2908805031446541, "percentage": 1.45, "elapsed_time": "0:01:21", "remaining_time": "1:32:19", "throughput": 4468.24, "total_tokens": 365312}
+{"current_steps": 560, "total_steps": 38160, "loss": 0.4835, "lr": 0.00014648846960167716, "epoch": 0.29350104821802936, "percentage": 1.47, "elapsed_time": "0:01:22", "remaining_time": "1:32:20", "throughput": 4470.41, "total_tokens": 368896}
+{"current_steps": 565, "total_steps": 38160, "loss": 0.5361, "lr": 0.00014779874213836477, "epoch": 0.29612159329140464, "percentage": 1.48, "elapsed_time": "0:01:23", "remaining_time": "1:32:14", "throughput": 4468.48, "total_tokens": 371648}
+{"current_steps": 570, "total_steps": 38160, "loss": 0.5799, "lr": 0.00014910901467505243, "epoch": 0.29874213836477986, "percentage": 1.49, "elapsed_time": "0:01:23", "remaining_time": "1:32:12", "throughput": 4468.2, "total_tokens": 374848}
+{"current_steps": 575, "total_steps": 38160, "loss": 0.5398, "lr": 0.00015041928721174007, "epoch": 0.30136268343815514, "percentage": 1.51, "elapsed_time": "0:01:24", "remaining_time": "1:32:03", "throughput": 4463.68, "total_tokens": 377184}
+{"current_steps": 580, "total_steps": 38160, "loss": 0.6253, "lr": 0.00015172955974842767, "epoch": 0.3039832285115304, "percentage": 1.52, "elapsed_time": "0:01:25", "remaining_time": "1:32:10", "throughput": 4469.9, "total_tokens": 381504}
+{"current_steps": 585, "total_steps": 38160, "loss": 0.5353, "lr": 0.0001530398322851153, "epoch": 0.30660377358490565, "percentage": 1.53, "elapsed_time": "0:01:26", "remaining_time": "1:32:07", "throughput": 4469.61, "total_tokens": 384672}
+{"current_steps": 590, "total_steps": 38160, "loss": 0.5229, "lr": 0.00015435010482180294, "epoch": 0.30922431865828093, "percentage": 1.55, "elapsed_time": "0:01:26", "remaining_time": "1:32:19", "throughput": 4476.18, "total_tokens": 389376}
+{"current_steps": 595, "total_steps": 38160, "loss": 0.589, "lr": 0.00015566037735849058, "epoch": 0.3118448637316562, "percentage": 1.56, "elapsed_time": "0:01:27", "remaining_time": "1:32:17", "throughput": 4475.4, "total_tokens": 392512}
+{"current_steps": 600, "total_steps": 38160, "loss": 0.5162, "lr": 0.0001569706498951782, "epoch": 0.31446540880503143, "percentage": 1.57, "elapsed_time": "0:01:28", "remaining_time": "1:32:23", "throughput": 4481.71, "total_tokens": 396864}
+{"current_steps": 605, "total_steps": 38160, "loss": 0.4766, "lr": 0.00015828092243186584, "epoch": 0.3170859538784067, "percentage": 1.59, "elapsed_time": "0:01:29", "remaining_time": "1:32:16", "throughput": 4479.38, "total_tokens": 399520}
+{"current_steps": 610, "total_steps": 38160, "loss": 0.4859, "lr": 0.00015959119496855345, "epoch": 0.319706498951782, "percentage": 1.6, "elapsed_time": "0:01:29", "remaining_time": "1:32:12", "throughput": 4478.77, "total_tokens": 402560}
+{"current_steps": 615, "total_steps": 38160, "loss": 0.6413, "lr": 0.00016090146750524109, "epoch": 0.3223270440251572, "percentage": 1.61, "elapsed_time": "0:01:30", "remaining_time": "1:32:14", "throughput": 4480.71, "total_tokens": 406240}
+{"current_steps": 620, "total_steps": 38160, "loss": 0.5483, "lr": 0.00016221174004192875, "epoch": 0.3249475890985325, "percentage": 1.62, "elapsed_time": "0:01:31", "remaining_time": "1:32:08", "throughput": 4478.58, "total_tokens": 408960}
+{"current_steps": 625, "total_steps": 38160, "loss": 0.6812, "lr": 0.00016352201257861635, "epoch": 0.3275681341719078, "percentage": 1.64, "elapsed_time": "0:01:32", "remaining_time": "1:32:07", "throughput": 4479.0, "total_tokens": 412224}
+{"current_steps": 630, "total_steps": 38160, "loss": 0.4919, "lr": 0.000164832285115304, "epoch": 0.330188679245283, "percentage": 1.65, "elapsed_time": "0:01:32", "remaining_time": "1:32:04", "throughput": 4478.0, "total_tokens": 415296}
+{"current_steps": 635, "total_steps": 38160, "loss": 0.4547, "lr": 0.00016614255765199162, "epoch": 0.3328092243186583, "percentage": 1.66, "elapsed_time": "0:01:33", "remaining_time": "1:31:55", "throughput": 4473.36, "total_tokens": 417504}
+{"current_steps": 640, "total_steps": 38160, "loss": 0.5495, "lr": 0.00016745283018867923, "epoch": 0.33542976939203356, "percentage": 1.68, "elapsed_time": "0:01:33", "remaining_time": "1:31:50", "throughput": 4471.4, "total_tokens": 420288}
+{"current_steps": 645, "total_steps": 38160, "loss": 0.6852, "lr": 0.0001687631027253669, "epoch": 0.3380503144654088, "percentage": 1.69, "elapsed_time": "0:01:34", "remaining_time": "1:31:41", "throughput": 4466.98, "total_tokens": 422560}
+{"current_steps": 650, "total_steps": 38160, "loss": 0.6101, "lr": 0.00017007337526205453, "epoch": 0.34067085953878407, "percentage": 1.7, "elapsed_time": "0:01:35", "remaining_time": "1:31:38", "throughput": 4465.8, "total_tokens": 425472}
+{"current_steps": 655, "total_steps": 38160, "loss": 0.5264, "lr": 0.00017138364779874213, "epoch": 0.34329140461215935, "percentage": 1.72, "elapsed_time": "0:01:36", "remaining_time": "1:31:38", "throughput": 4467.79, "total_tokens": 429056}
+{"current_steps": 660, "total_steps": 38160, "loss": 0.492, "lr": 0.00017269392033542977, "epoch": 0.34591194968553457, "percentage": 1.73, "elapsed_time": "0:01:36", "remaining_time": "1:31:42", "throughput": 4471.28, "total_tokens": 433024}
+{"current_steps": 665, "total_steps": 38160, "loss": 0.5435, "lr": 0.0001740041928721174, "epoch": 0.34853249475890985, "percentage": 1.74, "elapsed_time": "0:01:37", "remaining_time": "1:31:37", "throughput": 4469.45, "total_tokens": 435776}
+{"current_steps": 670, "total_steps": 38160, "loss": 0.5188, "lr": 0.00017531446540880504, "epoch": 0.35115303983228513, "percentage": 1.76, "elapsed_time": "0:01:38", "remaining_time": "1:31:31", "throughput": 4467.1, "total_tokens": 438432}
+{"current_steps": 675, "total_steps": 38160, "loss": 0.5188, "lr": 0.00017662473794549267, "epoch": 0.35377358490566035, "percentage": 1.77, "elapsed_time": "0:01:38", "remaining_time": "1:31:35", "throughput": 4470.71, "total_tokens": 442400}
+{"current_steps": 680, "total_steps": 38160, "loss": 0.6068, "lr": 0.0001779350104821803, "epoch": 0.35639412997903563, "percentage": 1.78, "elapsed_time": "0:01:39", "remaining_time": "1:31:35", "throughput": 4472.14, "total_tokens": 445888}
+{"current_steps": 685, "total_steps": 38160, "loss": 0.4774, "lr": 0.0001792452830188679, "epoch": 0.3590146750524109, "percentage": 1.8, "elapsed_time": "0:01:40", "remaining_time": "1:31:31", "throughput": 4471.35, "total_tokens": 448864}
+{"current_steps": 690, "total_steps": 38160, "loss": 0.6968, "lr": 0.00018055555555555555, "epoch": 0.36163522012578614, "percentage": 1.81, "elapsed_time": "0:01:41", "remaining_time": "1:31:34", "throughput": 4473.86, "total_tokens": 452640}
+{"current_steps": 695, "total_steps": 38160, "loss": 0.6116, "lr": 0.0001818658280922432, "epoch": 0.3642557651991614, "percentage": 1.82, "elapsed_time": "0:01:41", "remaining_time": "1:31:28", "throughput": 4471.43, "total_tokens": 455296}
+{"current_steps": 700, "total_steps": 38160, "loss": 0.5421, "lr": 0.00018317610062893082, "epoch": 0.3668763102725367, "percentage": 1.83, "elapsed_time": "0:01:42", "remaining_time": "1:31:23", "throughput": 4468.57, "total_tokens": 457888}
+{"current_steps": 705, "total_steps": 38160, "loss": 0.6139, "lr": 0.00018448637316561845, "epoch": 0.3694968553459119, "percentage": 1.85, "elapsed_time": "0:01:43", "remaining_time": "1:31:25", "throughput": 4470.91, "total_tokens": 461664}
+{"current_steps": 710, "total_steps": 38160, "loss": 0.6136, "lr": 0.00018579664570230608, "epoch": 0.3721174004192872, "percentage": 1.86, "elapsed_time": "0:01:43", "remaining_time": "1:31:20", "throughput": 4468.87, "total_tokens": 464352}
+{"current_steps": 715, "total_steps": 38160, "loss": 0.6077, "lr": 0.0001871069182389937, "epoch": 0.3747379454926625, "percentage": 1.87, "elapsed_time": "0:01:44", "remaining_time": "1:31:24", "throughput": 4471.21, "total_tokens": 468256}
+{"current_steps": 720, "total_steps": 38160, "loss": 0.5133, "lr": 0.00018841719077568135, "epoch": 0.37735849056603776, "percentage": 1.89, "elapsed_time": "0:01:45", "remaining_time": "1:31:20", "throughput": 4469.71, "total_tokens": 471104}
+{"current_steps": 725, "total_steps": 38160, "loss": 0.6115, "lr": 0.000189727463312369, "epoch": 0.379979035639413, "percentage": 1.9, "elapsed_time": "0:01:46", "remaining_time": "1:31:15", "throughput": 4466.88, "total_tokens": 473696}
+{"current_steps": 730, "total_steps": 38160, "loss": 0.5122, "lr": 0.00019103773584905662, "epoch": 0.38259958071278827, "percentage": 1.91, "elapsed_time": "0:01:46", "remaining_time": "1:31:11", "throughput": 4465.2, "total_tokens": 476512}
+{"current_steps": 735, "total_steps": 38160, "loss": 0.5904, "lr": 0.00019234800838574423, "epoch": 0.38522012578616355, "percentage": 1.93, "elapsed_time": "0:01:47", "remaining_time": "1:31:12", "throughput": 4466.92, "total_tokens": 480128}
+{"current_steps": 740, "total_steps": 38160, "loss": 0.6301, "lr": 0.00019365828092243186, "epoch": 0.38784067085953877, "percentage": 1.94, "elapsed_time": "0:01:48", "remaining_time": "1:31:14", "throughput": 4468.92, "total_tokens": 483808}
+{"current_steps": 745, "total_steps": 38160, "loss": 0.5939, "lr": 0.00019496855345911953, "epoch": 0.39046121593291405, "percentage": 1.95, "elapsed_time": "0:01:48", "remaining_time": "1:31:08", "throughput": 4465.7, "total_tokens": 486240}
+{"current_steps": 750, "total_steps": 38160, "loss": 0.4792, "lr": 0.00019627882599580713, "epoch": 0.39308176100628933, "percentage": 1.97, "elapsed_time": "0:01:49", "remaining_time": "1:31:00", "throughput": 4461.74, "total_tokens": 488448}
+{"current_steps": 755, "total_steps": 38160, "loss": 0.5778, "lr": 0.00019758909853249477, "epoch": 0.39570230607966456, "percentage": 1.98, "elapsed_time": "0:01:50", "remaining_time": "1:30:57", "throughput": 4460.65, "total_tokens": 491392}
+{"current_steps": 760, "total_steps": 38160, "loss": 0.5299, "lr": 0.0001988993710691824, "epoch": 0.39832285115303984, "percentage": 1.99, "elapsed_time": "0:01:50", "remaining_time": "1:31:00", "throughput": 4463.19, "total_tokens": 495232}
+{"current_steps": 765, "total_steps": 38160, "loss": 0.4787, "lr": 0.00020020964360587, "epoch": 0.4009433962264151, "percentage": 2.0, "elapsed_time": "0:01:51", "remaining_time": "1:30:56", "throughput": 4461.82, "total_tokens": 498016}
+{"current_steps": 770, "total_steps": 38160, "loss": 0.5109, "lr": 0.00020151991614255767, "epoch": 0.40356394129979034, "percentage": 2.02, "elapsed_time": "0:01:52", "remaining_time": "1:30:54", "throughput": 4461.63, "total_tokens": 501152}
+{"current_steps": 775, "total_steps": 38160, "loss": 0.496, "lr": 0.0002028301886792453, "epoch": 0.4061844863731656, "percentage": 2.03, "elapsed_time": "0:01:52", "remaining_time": "1:30:50", "throughput": 4459.66, "total_tokens": 503872}
+{"current_steps": 780, "total_steps": 38160, "loss": 0.5207, "lr": 0.0002041404612159329, "epoch": 0.4088050314465409, "percentage": 2.04, "elapsed_time": "0:01:53", "remaining_time": "1:30:50", "throughput": 4460.84, "total_tokens": 507328}
+{"current_steps": 785, "total_steps": 38160, "loss": 0.4149, "lr": 0.00020545073375262055, "epoch": 0.4114255765199161, "percentage": 2.06, "elapsed_time": "0:01:54", "remaining_time": "1:30:51", "throughput": 4462.02, "total_tokens": 510912}
+{"current_steps": 790, "total_steps": 38160, "loss": 0.5954, "lr": 0.00020676100628930818, "epoch": 0.4140461215932914, "percentage": 2.07, "elapsed_time": "0:01:55", "remaining_time": "1:30:51", "throughput": 4462.65, "total_tokens": 514304}
+{"current_steps": 795, "total_steps": 38160, "loss": 0.4442, "lr": 0.00020807127882599581, "epoch": 0.4166666666666667, "percentage": 2.08, "elapsed_time": "0:01:55", "remaining_time": "1:30:49", "throughput": 4462.44, "total_tokens": 517440}
+{"current_steps": 800, "total_steps": 38160, "loss": 0.6763, "lr": 0.00020938155136268345, "epoch": 0.4192872117400419, "percentage": 2.1, "elapsed_time": "0:01:56", "remaining_time": "1:30:47", "throughput": 4461.66, "total_tokens": 520416}
+{"current_steps": 805, "total_steps": 38160, "loss": 0.6537, "lr": 0.00021069182389937108, "epoch": 0.4219077568134172, "percentage": 2.11, "elapsed_time": "0:01:57", "remaining_time": "1:30:44", "throughput": 4461.23, "total_tokens": 523456}
+{"current_steps": 810, "total_steps": 38160, "loss": 0.5032, "lr": 0.0002120020964360587, "epoch": 0.42452830188679247, "percentage": 2.12, "elapsed_time": "0:01:58", "remaining_time": "1:30:43", "throughput": 4461.74, "total_tokens": 526720}
+{"current_steps": 815, "total_steps": 38160, "loss": 0.4356, "lr": 0.00021331236897274632, "epoch": 0.4271488469601677, "percentage": 2.14, "elapsed_time": "0:01:58", "remaining_time": "1:30:44", "throughput": 4463.29, "total_tokens": 530272}
+{"current_steps": 820, "total_steps": 38160, "loss": 0.6282, "lr": 0.00021462264150943399, "epoch": 0.429769392033543, "percentage": 2.15, "elapsed_time": "0:01:59", "remaining_time": "1:30:42", "throughput": 4463.14, "total_tokens": 533408}
+{"current_steps": 825, "total_steps": 38160, "loss": 0.6366, "lr": 0.0002159329140461216, "epoch": 0.43238993710691825, "percentage": 2.16, "elapsed_time": "0:02:00", "remaining_time": "1:30:42", "throughput": 4463.43, "total_tokens": 536768}
+{"current_steps": 830, "total_steps": 38160, "loss": 0.6108, "lr": 0.00021724318658280923, "epoch": 0.4350104821802935, "percentage": 2.18, "elapsed_time": "0:02:01", "remaining_time": "1:30:47", "throughput": 4467.78, "total_tokens": 541184}
+{"current_steps": 835, "total_steps": 38160, "loss": 0.4637, "lr": 0.00021855345911949686, "epoch": 0.43763102725366876, "percentage": 2.19, "elapsed_time": "0:02:01", "remaining_time": "1:30:48", "throughput": 4468.06, "total_tokens": 544576}
+{"current_steps": 840, "total_steps": 38160, "loss": 0.5865, "lr": 0.00021986373165618447, "epoch": 0.44025157232704404, "percentage": 2.2, "elapsed_time": "0:02:02", "remaining_time": "1:30:56", "throughput": 4473.09, "total_tokens": 549408}
+{"current_steps": 845, "total_steps": 38160, "loss": 0.468, "lr": 0.00022117400419287213, "epoch": 0.44287211740041926, "percentage": 2.21, "elapsed_time": "0:02:03", "remaining_time": "1:30:51", "throughput": 4471.4, "total_tokens": 552032}
+{"current_steps": 850, "total_steps": 38160, "loss": 0.5104, "lr": 0.00022248427672955977, "epoch": 0.44549266247379454, "percentage": 2.23, "elapsed_time": "0:02:04", "remaining_time": "1:30:53", "throughput": 4473.88, "total_tokens": 555872}
+{"current_steps": 855, "total_steps": 38160, "loss": 0.6376, "lr": 0.00022379454926624737, "epoch": 0.4481132075471698, "percentage": 2.24, "elapsed_time": "0:02:04", "remaining_time": "1:30:52", "throughput": 4474.18, "total_tokens": 559136}
+{"current_steps": 860, "total_steps": 38160, "loss": 0.5296, "lr": 0.000225104821802935, "epoch": 0.45073375262054505, "percentage": 2.25, "elapsed_time": "0:02:05", "remaining_time": "1:30:49", "throughput": 4472.88, "total_tokens": 561984}
+{"current_steps": 865, "total_steps": 38160, "loss": 0.4832, "lr": 0.00022641509433962264, "epoch": 0.4533542976939203, "percentage": 2.27, "elapsed_time": "0:02:06", "remaining_time": "1:30:44", "throughput": 4470.03, "total_tokens": 564448}
+{"current_steps": 870, "total_steps": 38160, "loss": 0.5687, "lr": 0.00022772536687631028, "epoch": 0.4559748427672956, "percentage": 2.28, "elapsed_time": "0:02:06", "remaining_time": "1:30:38", "throughput": 4466.79, "total_tokens": 566752}
+{"current_steps": 875, "total_steps": 38160, "loss": 0.5288, "lr": 0.0002290356394129979, "epoch": 0.4585953878406709, "percentage": 2.29, "elapsed_time": "0:02:07", "remaining_time": "1:30:36", "throughput": 4466.63, "total_tokens": 569856}
+{"current_steps": 880, "total_steps": 38160, "loss": 0.5104, "lr": 0.00023034591194968554, "epoch": 0.4612159329140461, "percentage": 2.31, "elapsed_time": "0:02:08", "remaining_time": "1:30:35", "throughput": 4466.67, "total_tokens": 573056}
+{"current_steps": 885, "total_steps": 38160, "loss": 0.5814, "lr": 0.00023165618448637318, "epoch": 0.4638364779874214, "percentage": 2.32, "elapsed_time": "0:02:08", "remaining_time": "1:30:30", "throughput": 4465.02, "total_tokens": 575712}
+{"current_steps": 890, "total_steps": 38160, "loss": 0.4417, "lr": 0.00023296645702306079, "epoch": 0.46645702306079667, "percentage": 2.33, "elapsed_time": "0:02:09", "remaining_time": "1:30:29", "throughput": 4465.16, "total_tokens": 578880}
+{"current_steps": 895, "total_steps": 38160, "loss": 0.4844, "lr": 0.00023427672955974845, "epoch": 0.4690775681341719, "percentage": 2.35, "elapsed_time": "0:02:10", "remaining_time": "1:30:27", "throughput": 4464.31, "total_tokens": 581920}
+{"current_steps": 900, "total_steps": 38160, "loss": 0.5446, "lr": 0.00023558700209643608, "epoch": 0.4716981132075472, "percentage": 2.36, "elapsed_time": "0:02:11", "remaining_time": "1:30:23", "throughput": 4462.95, "total_tokens": 584672}
+{"current_steps": 905, "total_steps": 38160, "loss": 0.4763, "lr": 0.0002368972746331237, "epoch": 0.47431865828092246, "percentage": 2.37, "elapsed_time": "0:02:11", "remaining_time": "1:30:26", "throughput": 4465.65, "total_tokens": 588704}
+{"current_steps": 910, "total_steps": 38160, "loss": 0.5383, "lr": 0.00023820754716981132, "epoch": 0.4769392033542977, "percentage": 2.38, "elapsed_time": "0:02:12", "remaining_time": "1:30:24", "throughput": 4465.08, "total_tokens": 591712}
+{"current_steps": 915, "total_steps": 38160, "loss": 0.5598, "lr": 0.00023951781970649896, "epoch": 0.47955974842767296, "percentage": 2.4, "elapsed_time": "0:02:13", "remaining_time": "1:30:23", "throughput": 4466.19, "total_tokens": 595104}
+{"current_steps": 920, "total_steps": 38160, "loss": 0.494, "lr": 0.0002408280922431866, "epoch": 0.48218029350104824, "percentage": 2.41, "elapsed_time": "0:02:14", "remaining_time": "1:30:30", "throughput": 4470.86, "total_tokens": 599808}
+{"current_steps": 925, "total_steps": 38160, "loss": 0.4524, "lr": 0.00024213836477987423, "epoch": 0.48480083857442346, "percentage": 2.42, "elapsed_time": "0:02:14", "remaining_time": "1:30:27", "throughput": 4470.1, "total_tokens": 602720}
+{"current_steps": 930, "total_steps": 38160, "loss": 0.4922, "lr": 0.00024344863731656186, "epoch": 0.48742138364779874, "percentage": 2.44, "elapsed_time": "0:02:15", "remaining_time": "1:30:27", "throughput": 4470.91, "total_tokens": 606176}
+{"current_steps": 935, "total_steps": 38160, "loss": 0.4674, "lr": 0.00024475890985324947, "epoch": 0.490041928721174, "percentage": 2.45, "elapsed_time": "0:02:16", "remaining_time": "1:30:23", "throughput": 4469.18, "total_tokens": 608864}
+{"current_steps": 940, "total_steps": 38160, "loss": 0.5117, "lr": 0.0002460691823899371, "epoch": 0.49266247379454925, "percentage": 2.46, "elapsed_time": "0:02:16", "remaining_time": "1:30:22", "throughput": 4468.79, "total_tokens": 611968}
+{"current_steps": 945, "total_steps": 38160, "loss": 0.6172, "lr": 0.00024737945492662474, "epoch": 0.49528301886792453, "percentage": 2.48, "elapsed_time": "0:02:17", "remaining_time": "1:30:19", "throughput": 4468.12, "total_tokens": 614848}
+{"current_steps": 950, "total_steps": 38160, "loss": 0.665, "lr": 0.00024868972746331237, "epoch": 0.4979035639412998, "percentage": 2.49, "elapsed_time": "0:02:18", "remaining_time": "1:30:18", "throughput": 4468.02, "total_tokens": 618080}
+{"current_steps": 955, "total_steps": 38160, "loss": 0.3781, "lr": 0.00025, "epoch": 0.500524109014675, "percentage": 2.5, "elapsed_time": "0:02:18", "remaining_time": "1:30:14", "throughput": 4465.99, "total_tokens": 620736}
+{"current_steps": 960, "total_steps": 38160, "loss": 0.4699, "lr": 0.00025131027253668764, "epoch": 0.5031446540880503, "percentage": 2.52, "elapsed_time": "0:02:19", "remaining_time": "1:30:16", "throughput": 4467.9, "total_tokens": 624480}
+{"current_steps": 965, "total_steps": 38160, "loss": 0.4593, "lr": 0.0002526205450733753, "epoch": 0.5057651991614256, "percentage": 2.53, "elapsed_time": "0:02:20", "remaining_time": "1:30:18", "throughput": 4470.45, "total_tokens": 628448}
+{"current_steps": 970, "total_steps": 38160, "loss": 0.6066, "lr": 0.0002539308176100629, "epoch": 0.5083857442348009, "percentage": 2.54, "elapsed_time": "0:02:21", "remaining_time": "1:30:18", "throughput": 4471.55, "total_tokens": 632000}
+{"current_steps": 975, "total_steps": 38160, "loss": 0.6094, "lr": 0.0002552410901467505, "epoch": 0.5110062893081762, "percentage": 2.56, "elapsed_time": "0:02:22", "remaining_time": "1:30:20", "throughput": 4472.73, "total_tokens": 635680}
+{"current_steps": 980, "total_steps": 38160, "loss": 0.7542, "lr": 0.0002565513626834381, "epoch": 0.5136268343815513, "percentage": 2.57, "elapsed_time": "0:02:22", "remaining_time": "1:30:16", "throughput": 4471.82, "total_tokens": 638496}
+{"current_steps": 985, "total_steps": 38160, "loss": 0.4697, "lr": 0.0002578616352201258, "epoch": 0.5162473794549266, "percentage": 2.58, "elapsed_time": "0:02:23", "remaining_time": "1:30:18", "throughput": 4473.41, "total_tokens": 642272}
+{"current_steps": 990, "total_steps": 38160, "loss": 0.5283, "lr": 0.00025917190775681345, "epoch": 0.5188679245283019, "percentage": 2.59, "elapsed_time": "0:02:24", "remaining_time": "1:30:19", "throughput": 4474.5, "total_tokens": 645888}
+{"current_steps": 995, "total_steps": 38160, "loss": 0.3726, "lr": 0.0002604821802935011, "epoch": 0.5214884696016772, "percentage": 2.61, "elapsed_time": "0:02:25", "remaining_time": "1:30:33", "throughput": 4481.07, "total_tokens": 651808}
+{"current_steps": 1000, "total_steps": 38160, "loss": 0.5356, "lr": 0.0002617924528301887, "epoch": 0.5241090146750524, "percentage": 2.62, "elapsed_time": "0:02:26", "remaining_time": "1:30:34", "throughput": 4482.18, "total_tokens": 655488}
+{"current_steps": 1005, "total_steps": 38160, "loss": 0.4621, "lr": 0.0002631027253668763, "epoch": 0.5267295597484277, "percentage": 2.63, "elapsed_time": "0:02:26", "remaining_time": "1:30:34", "throughput": 4482.92, "total_tokens": 658976}
+{"current_steps": 1010, "total_steps": 38160, "loss": 0.4489, "lr": 0.00026441299790356393, "epoch": 0.5293501048218029, "percentage": 2.65, "elapsed_time": "0:02:27", "remaining_time": "1:30:32", "throughput": 4482.63, "total_tokens": 662048}
+{"current_steps": 1015, "total_steps": 38160, "loss": 0.5845, "lr": 0.00026572327044025156, "epoch": 0.5319706498951782, "percentage": 2.66, "elapsed_time": "0:02:28", "remaining_time": "1:30:28", "throughput": 4480.97, "total_tokens": 664640}
+{"current_steps": 1020, "total_steps": 38160, "loss": 0.489, "lr": 0.0002670335429769392, "epoch": 0.5345911949685535, "percentage": 2.67, "elapsed_time": "0:02:29", "remaining_time": "1:30:32", "throughput": 4484.34, "total_tokens": 669056}
+{"current_steps": 1025, "total_steps": 38160, "loss": 0.5195, "lr": 0.00026834381551362683, "epoch": 0.5372117400419287, "percentage": 2.69, "elapsed_time": "0:02:29", "remaining_time": "1:30:30", "throughput": 4483.78, "total_tokens": 672128}
+{"current_steps": 1030, "total_steps": 38160, "loss": 0.4863, "lr": 0.0002696540880503145, "epoch": 0.539832285115304, "percentage": 2.7, "elapsed_time": "0:02:30", "remaining_time": "1:30:28", "throughput": 4483.18, "total_tokens": 675136}
+{"current_steps": 1035, "total_steps": 38160, "loss": 0.6039, "lr": 0.0002709643605870021, "epoch": 0.5424528301886793, "percentage": 2.71, "elapsed_time": "0:02:31", "remaining_time": "1:30:30", "throughput": 4485.1, "total_tokens": 678976}
+{"current_steps": 1040, "total_steps": 38160, "loss": 0.6647, "lr": 0.00027227463312368973, "epoch": 0.5450733752620545, "percentage": 2.73, "elapsed_time": "0:02:32", "remaining_time": "1:30:28", "throughput": 4485.1, "total_tokens": 682176}
+{"current_steps": 1045, "total_steps": 38160, "loss": 0.5717, "lr": 0.00027358490566037737, "epoch": 0.5476939203354297, "percentage": 2.74, "elapsed_time": "0:02:32", "remaining_time": "1:30:24", "throughput": 4483.44, "total_tokens": 684768}
+{"current_steps": 1050, "total_steps": 38160, "loss": 0.5619, "lr": 0.000274895178197065, "epoch": 0.550314465408805, "percentage": 2.75, "elapsed_time": "0:02:33", "remaining_time": "1:30:21", "throughput": 4481.49, "total_tokens": 687392}
+{"current_steps": 1055, "total_steps": 38160, "loss": 0.4905, "lr": 0.00027620545073375264, "epoch": 0.5529350104821803, "percentage": 2.76, "elapsed_time": "0:02:34", "remaining_time": "1:30:19", "throughput": 4481.31, "total_tokens": 690496}
+{"current_steps": 1060, "total_steps": 38160, "loss": 0.6701, "lr": 0.00027751572327044027, "epoch": 0.5555555555555556, "percentage": 2.78, "elapsed_time": "0:02:34", "remaining_time": "1:30:16", "throughput": 4480.38, "total_tokens": 693376}
+{"current_steps": 1065, "total_steps": 38160, "loss": 0.4841, "lr": 0.00027882599580712785, "epoch": 0.5581761006289309, "percentage": 2.79, "elapsed_time": "0:02:35", "remaining_time": "1:30:12", "throughput": 4478.11, "total_tokens": 695808}
+{"current_steps": 1070, "total_steps": 38160, "loss": 0.491, "lr": 0.0002801362683438155, "epoch": 0.560796645702306, "percentage": 2.8, "elapsed_time": "0:02:36", "remaining_time": "1:30:12", "throughput": 4479.54, "total_tokens": 699424}
+{"current_steps": 1075, "total_steps": 38160, "loss": 0.5349, "lr": 0.0002814465408805031, "epoch": 0.5634171907756813, "percentage": 2.82, "elapsed_time": "0:02:36", "remaining_time": "1:30:14", "throughput": 4482.58, "total_tokens": 703584}
+{"current_steps": 1080, "total_steps": 38160, "loss": 0.5218, "lr": 0.0002827568134171908, "epoch": 0.5660377358490566, "percentage": 2.83, "elapsed_time": "0:02:37", "remaining_time": "1:30:10", "throughput": 4480.87, "total_tokens": 706144}
+{"current_steps": 1085, "total_steps": 38160, "loss": 0.5281, "lr": 0.00028406708595387844, "epoch": 0.5686582809224319, "percentage": 2.84, "elapsed_time": "0:02:38", "remaining_time": "1:30:07", "throughput": 4479.65, "total_tokens": 708864}
+{"current_steps": 1090, "total_steps": 38160, "loss": 0.5503, "lr": 0.0002853773584905661, "epoch": 0.5712788259958071, "percentage": 2.86, "elapsed_time": "0:02:38", "remaining_time": "1:30:07", "throughput": 4480.43, "total_tokens": 712352}
+{"current_steps": 1095, "total_steps": 38160, "loss": 0.7352, "lr": 0.00028668763102725366, "epoch": 0.5738993710691824, "percentage": 2.87, "elapsed_time": "0:02:39", "remaining_time": "1:30:04", "throughput": 4479.64, "total_tokens": 715296}
+{"current_steps": 1100, "total_steps": 38160, "loss": 0.4537, "lr": 0.0002879979035639413, "epoch": 0.5765199161425576, "percentage": 2.88, "elapsed_time": "0:02:40", "remaining_time": "1:30:03", "throughput": 4479.9, "total_tokens": 718528}
+{"current_steps": 1105, "total_steps": 38160, "loss": 0.4627, "lr": 0.00028930817610062893, "epoch": 0.5791404612159329, "percentage": 2.9, "elapsed_time": "0:02:42", "remaining_time": "1:30:43", "throughput": 4452.51, "total_tokens": 722752}
+{"current_steps": 1110, "total_steps": 38160, "loss": 0.4673, "lr": 0.00029061844863731656, "epoch": 0.5817610062893082, "percentage": 2.91, "elapsed_time": "0:02:43", "remaining_time": "1:30:42", "throughput": 4451.78, "total_tokens": 725824}
+{"current_steps": 1115, "total_steps": 38160, "loss": 0.5829, "lr": 0.0002919287211740042, "epoch": 0.5843815513626834, "percentage": 2.92, "elapsed_time": "0:02:43", "remaining_time": "1:30:44", "throughput": 4454.85, "total_tokens": 730080}
+{"current_steps": 1120, "total_steps": 38160, "loss": 0.5032, "lr": 0.00029323899371069183, "epoch": 0.5870020964360587, "percentage": 2.94, "elapsed_time": "0:02:44", "remaining_time": "1:30:43", "throughput": 4454.7, "total_tokens": 733216}
+{"current_steps": 1125, "total_steps": 38160, "loss": 0.6646, "lr": 0.00029454926624737946, "epoch": 0.589622641509434, "percentage": 2.95, "elapsed_time": "0:02:45", "remaining_time": "1:30:41", "throughput": 4454.78, "total_tokens": 736416}
+{"current_steps": 1130, "total_steps": 38160, "loss": 0.4559, "lr": 0.0002958595387840671, "epoch": 0.5922431865828093, "percentage": 2.96, "elapsed_time": "0:02:46", "remaining_time": "1:30:41", "throughput": 4455.13, "total_tokens": 739808}
+{"current_steps": 1135, "total_steps": 38160, "loss": 0.5518, "lr": 0.00029716981132075473, "epoch": 0.5948637316561844, "percentage": 2.97, "elapsed_time": "0:02:46", "remaining_time": "1:30:37", "throughput": 4453.64, "total_tokens": 742400}
+{"current_steps": 1140, "total_steps": 38160, "loss": 0.4005, "lr": 0.00029848008385744237, "epoch": 0.5974842767295597, "percentage": 2.99, "elapsed_time": "0:02:47", "remaining_time": "1:30:37", "throughput": 4454.14, "total_tokens": 745760}
+{"current_steps": 1145, "total_steps": 38160, "loss": 0.5925, "lr": 0.00029979035639413, "epoch": 0.600104821802935, "percentage": 3.0, "elapsed_time": "0:02:48", "remaining_time": "1:30:34", "throughput": 4453.74, "total_tokens": 748736}
+{"current_steps": 1150, "total_steps": 38160, "loss": 0.5357, "lr": 0.00030110062893081764, "epoch": 0.6027253668763103, "percentage": 3.01, "elapsed_time": "0:02:48", "remaining_time": "1:30:33", "throughput": 4454.07, "total_tokens": 752032}
+{"current_steps": 1155, "total_steps": 38160, "loss": 0.491, "lr": 0.00030241090146750527, "epoch": 0.6053459119496856, "percentage": 3.03, "elapsed_time": "0:02:49", "remaining_time": "1:30:38", "throughput": 4457.88, "total_tokens": 756736}
+{"current_steps": 1160, "total_steps": 38160, "loss": 0.4877, "lr": 0.00030372117400419285, "epoch": 0.6079664570230608, "percentage": 3.04, "elapsed_time": "0:02:50", "remaining_time": "1:30:36", "throughput": 4457.18, "total_tokens": 759680}
+{"current_steps": 1165, "total_steps": 38160, "loss": 0.5316, "lr": 0.0003050314465408805, "epoch": 0.610587002096436, "percentage": 3.05, "elapsed_time": "0:02:51", "remaining_time": "1:30:33", "throughput": 4456.38, "total_tokens": 762560}
+{"current_steps": 1170, "total_steps": 38160, "loss": 0.4494, "lr": 0.0003063417190775681, "epoch": 0.6132075471698113, "percentage": 3.07, "elapsed_time": "0:02:51", "remaining_time": "1:30:31", "throughput": 4456.07, "total_tokens": 765600}
+{"current_steps": 1175, "total_steps": 38160, "loss": 0.4, "lr": 0.00030765199161425575, "epoch": 0.6158280922431866, "percentage": 3.08, "elapsed_time": "0:02:52", "remaining_time": "1:30:30", "throughput": 4456.09, "total_tokens": 768736}
+{"current_steps": 1180, "total_steps": 38160, "loss": 0.6684, "lr": 0.00030896226415094344, "epoch": 0.6184486373165619, "percentage": 3.09, "elapsed_time": "0:02:53", "remaining_time": "1:30:28", "throughput": 4455.83, "total_tokens": 771872}
+{"current_steps": 1185, "total_steps": 38160, "loss": 0.619, "lr": 0.0003102725366876311, "epoch": 0.6210691823899371, "percentage": 3.11, "elapsed_time": "0:02:53", "remaining_time": "1:30:27", "throughput": 4455.58, "total_tokens": 775008}
+{"current_steps": 1190, "total_steps": 38160, "loss": 0.4269, "lr": 0.00031158280922431866, "epoch": 0.6236897274633124, "percentage": 3.12, "elapsed_time": "0:02:54", "remaining_time": "1:30:25", "throughput": 4454.89, "total_tokens": 777920}
+{"current_steps": 1195, "total_steps": 38160, "loss": 0.5171, "lr": 0.0003128930817610063, "epoch": 0.6263102725366876, "percentage": 3.13, "elapsed_time": "0:02:55", "remaining_time": "1:30:23", "throughput": 4455.15, "total_tokens": 781056}
+{"current_steps": 1200, "total_steps": 38160, "loss": 0.5798, "lr": 0.0003142033542976939, "epoch": 0.6289308176100629, "percentage": 3.14, "elapsed_time": "0:02:56", "remaining_time": "1:30:21", "throughput": 4455.03, "total_tokens": 784128}
+{"current_steps": 1205, "total_steps": 38160, "loss": 0.5388, "lr": 0.00031551362683438156, "epoch": 0.6315513626834381, "percentage": 3.16, "elapsed_time": "0:02:56", "remaining_time": "1:30:17", "throughput": 4453.28, "total_tokens": 786624}
+{"current_steps": 1210, "total_steps": 38160, "loss": 0.5321, "lr": 0.0003168238993710692, "epoch": 0.6341719077568134, "percentage": 3.17, "elapsed_time": "0:02:57", "remaining_time": "1:30:15", "throughput": 4452.83, "total_tokens": 789664}
+{"current_steps": 1215, "total_steps": 38160, "loss": 0.4326, "lr": 0.00031813417190775683, "epoch": 0.6367924528301887, "percentage": 3.18, "elapsed_time": "0:02:58", "remaining_time": "1:30:18", "throughput": 4455.05, "total_tokens": 793824}
+{"current_steps": 1220, "total_steps": 38160, "loss": 0.6322, "lr": 0.0003194444444444444, "epoch": 0.639412997903564, "percentage": 3.2, "elapsed_time": "0:02:58", "remaining_time": "1:30:15", "throughput": 4453.98, "total_tokens": 796544}
+{"current_steps": 1225, "total_steps": 38160, "loss": 0.4127, "lr": 0.00032075471698113204, "epoch": 0.6420335429769392, "percentage": 3.21, "elapsed_time": "0:02:59", "remaining_time": "1:30:12", "throughput": 4453.66, "total_tokens": 799520}
+{"current_steps": 1230, "total_steps": 38160, "loss": 0.3456, "lr": 0.00032206498951781973, "epoch": 0.6446540880503144, "percentage": 3.22, "elapsed_time": "0:03:00", "remaining_time": "1:30:10", "throughput": 4452.98, "total_tokens": 802464}
+{"current_steps": 1235, "total_steps": 38160, "loss": 0.5828, "lr": 0.00032337526205450737, "epoch": 0.6472746331236897, "percentage": 3.24, "elapsed_time": "0:03:00", "remaining_time": "1:30:08", "throughput": 4452.71, "total_tokens": 805536}
+{"current_steps": 1240, "total_steps": 38160, "loss": 0.3669, "lr": 0.000324685534591195, "epoch": 0.649895178197065, "percentage": 3.25, "elapsed_time": "0:03:01", "remaining_time": "1:30:06", "throughput": 4452.15, "total_tokens": 808448}
+{"current_steps": 1245, "total_steps": 38160, "loss": 0.4455, "lr": 0.00032599580712788263, "epoch": 0.6525157232704403, "percentage": 3.26, "elapsed_time": "0:03:02", "remaining_time": "1:30:07", "throughput": 4453.45, "total_tokens": 812160}
+{"current_steps": 1250, "total_steps": 38160, "loss": 0.637, "lr": 0.0003273060796645702, "epoch": 0.6551362683438156, "percentage": 3.28, "elapsed_time": "0:03:03", "remaining_time": "1:30:03", "throughput": 4452.16, "total_tokens": 814752}
+{"current_steps": 1255, "total_steps": 38160, "loss": 0.4277, "lr": 0.00032861635220125785, "epoch": 0.6577568134171907, "percentage": 3.29, "elapsed_time": "0:03:03", "remaining_time": "1:30:01", "throughput": 4451.26, "total_tokens": 817568}
+{"current_steps": 1260, "total_steps": 38160, "loss": 0.5027, "lr": 0.0003299266247379455, "epoch": 0.660377358490566, "percentage": 3.3, "elapsed_time": "0:03:04", "remaining_time": "1:29:58", "throughput": 4450.57, "total_tokens": 820384}
+{"current_steps": 1265, "total_steps": 38160, "loss": 0.3046, "lr": 0.0003312368972746331, "epoch": 0.6629979035639413, "percentage": 3.31, "elapsed_time": "0:03:04", "remaining_time": "1:29:55", "throughput": 4449.32, "total_tokens": 823040}
+{"current_steps": 1270, "total_steps": 38160, "loss": 0.5936, "lr": 0.00033254716981132075, "epoch": 0.6656184486373166, "percentage": 3.33, "elapsed_time": "0:03:05", "remaining_time": "1:29:51", "throughput": 4447.62, "total_tokens": 825600}
+{"current_steps": 1275, "total_steps": 38160, "loss": 0.5951, "lr": 0.0003338574423480084, "epoch": 0.6682389937106918, "percentage": 3.34, "elapsed_time": "0:03:06", "remaining_time": "1:29:48", "throughput": 4445.83, "total_tokens": 828032}
+{"current_steps": 1280, "total_steps": 38160, "loss": 0.4522, "lr": 0.0003351677148846961, "epoch": 0.6708595387840671, "percentage": 3.35, "elapsed_time": "0:03:06", "remaining_time": "1:29:47", "throughput": 4445.96, "total_tokens": 831296}
+{"current_steps": 1285, "total_steps": 38160, "loss": 0.6076, "lr": 0.00033647798742138366, "epoch": 0.6734800838574424, "percentage": 3.37, "elapsed_time": "0:03:07", "remaining_time": "1:29:45", "throughput": 4445.25, "total_tokens": 834208}
+{"current_steps": 1290, "total_steps": 38160, "loss": 0.4781, "lr": 0.0003377882599580713, "epoch": 0.6761006289308176, "percentage": 3.38, "elapsed_time": "0:03:08", "remaining_time": "1:29:46", "throughput": 4447.04, "total_tokens": 838048}
+{"current_steps": 1295, "total_steps": 38160, "loss": 0.5903, "lr": 0.0003390985324947589, "epoch": 0.6787211740041929, "percentage": 3.39, "elapsed_time": "0:03:09", "remaining_time": "1:29:45", "throughput": 4447.97, "total_tokens": 841536}
+{"current_steps": 1300, "total_steps": 38160, "loss": 0.4476, "lr": 0.00034040880503144656, "epoch": 0.6813417190775681, "percentage": 3.41, "elapsed_time": "0:03:09", "remaining_time": "1:29:46", "throughput": 4448.92, "total_tokens": 845184}
+{"current_steps": 1305, "total_steps": 38160, "loss": 0.442, "lr": 0.0003417190775681342, "epoch": 0.6839622641509434, "percentage": 3.42, "elapsed_time": "0:03:10", "remaining_time": "1:29:43", "throughput": 4448.21, "total_tokens": 848000}
+{"current_steps": 1310, "total_steps": 38160, "loss": 0.5777, "lr": 0.00034302935010482183, "epoch": 0.6865828092243187, "percentage": 3.43, "elapsed_time": "0:03:11", "remaining_time": "1:29:42", "throughput": 4447.86, "total_tokens": 851040}
+{"current_steps": 1315, "total_steps": 38160, "loss": 0.5948, "lr": 0.0003443396226415094, "epoch": 0.689203354297694, "percentage": 3.45, "elapsed_time": "0:03:12", "remaining_time": "1:29:39", "throughput": 4447.38, "total_tokens": 853920}
+{"current_steps": 1320, "total_steps": 38160, "loss": 0.513, "lr": 0.00034564989517819704, "epoch": 0.6918238993710691, "percentage": 3.46, "elapsed_time": "0:03:12", "remaining_time": "1:29:38", "throughput": 4448.2, "total_tokens": 857312}
+{"current_steps": 1325, "total_steps": 38160, "loss": 0.4137, "lr": 0.0003469601677148847, "epoch": 0.6944444444444444, "percentage": 3.47, "elapsed_time": "0:03:13", "remaining_time": "1:29:35", "throughput": 4446.87, "total_tokens": 859904}
+{"current_steps": 1330, "total_steps": 38160, "loss": 0.5575, "lr": 0.00034827044025157236, "epoch": 0.6970649895178197, "percentage": 3.49, "elapsed_time": "0:03:14", "remaining_time": "1:29:36", "throughput": 4448.57, "total_tokens": 863680}
+{"current_steps": 1335, "total_steps": 38160, "loss": 0.5729, "lr": 0.00034958071278826, "epoch": 0.699685534591195, "percentage": 3.5, "elapsed_time": "0:03:14", "remaining_time": "1:29:35", "throughput": 4448.67, "total_tokens": 866880}
+{"current_steps": 1340, "total_steps": 38160, "loss": 0.4186, "lr": 0.00035089098532494763, "epoch": 0.7023060796645703, "percentage": 3.51, "elapsed_time": "0:03:15", "remaining_time": "1:29:33", "throughput": 4448.24, "total_tokens": 869856}
+{"current_steps": 1345, "total_steps": 38160, "loss": 0.5439, "lr": 0.0003522012578616352, "epoch": 0.7049266247379455, "percentage": 3.52, "elapsed_time": "0:03:16", "remaining_time": "1:29:33", "throughput": 4449.2, "total_tokens": 873472}
+{"current_steps": 1350, "total_steps": 38160, "loss": 0.4465, "lr": 0.00035351153039832285, "epoch": 0.7075471698113207, "percentage": 3.54, "elapsed_time": "0:03:17", "remaining_time": "1:29:32", "throughput": 4449.41, "total_tokens": 876704}
+{"current_steps": 1355, "total_steps": 38160, "loss": 0.4769, "lr": 0.0003548218029350105, "epoch": 0.710167714884696, "percentage": 3.55, "elapsed_time": "0:03:17", "remaining_time": "1:29:33", "throughput": 4450.96, "total_tokens": 880576}
+{"current_steps": 1360, "total_steps": 38160, "loss": 0.4906, "lr": 0.0003561320754716981, "epoch": 0.7127882599580713, "percentage": 3.56, "elapsed_time": "0:03:18", "remaining_time": "1:29:32", "throughput": 4450.95, "total_tokens": 883680}
+{"current_steps": 1365, "total_steps": 38160, "loss": 0.5653, "lr": 0.00035744234800838575, "epoch": 0.7154088050314465, "percentage": 3.58, "elapsed_time": "0:03:19", "remaining_time": "1:29:29", "throughput": 4450.03, "total_tokens": 886464}
+{"current_steps": 1370, "total_steps": 38160, "loss": 0.5247, "lr": 0.0003587526205450734, "epoch": 0.7180293501048218, "percentage": 3.59, "elapsed_time": "0:03:19", "remaining_time": "1:29:28", "throughput": 4449.7, "total_tokens": 889536}
+{"current_steps": 1375, "total_steps": 38160, "loss": 0.507, "lr": 0.00036006289308176097, "epoch": 0.7206498951781971, "percentage": 3.6, "elapsed_time": "0:03:20", "remaining_time": "1:29:32", "throughput": 4452.79, "total_tokens": 894272}
+{"current_steps": 1380, "total_steps": 38160, "loss": 0.4721, "lr": 0.00036137316561844865, "epoch": 0.7232704402515723, "percentage": 3.62, "elapsed_time": "0:03:21", "remaining_time": "1:29:30", "throughput": 4452.09, "total_tokens": 897120}
+{"current_steps": 1385, "total_steps": 38160, "loss": 0.4942, "lr": 0.0003626834381551363, "epoch": 0.7258909853249476, "percentage": 3.63, "elapsed_time": "0:03:22", "remaining_time": "1:29:27", "throughput": 4451.18, "total_tokens": 899808}
+{"current_steps": 1390, "total_steps": 38160, "loss": 0.4896, "lr": 0.0003639937106918239, "epoch": 0.7285115303983228, "percentage": 3.64, "elapsed_time": "0:03:22", "remaining_time": "1:29:28", "throughput": 4452.43, "total_tokens": 903520}
+{"current_steps": 1395, "total_steps": 38160, "loss": 0.5451, "lr": 0.00036530398322851156, "epoch": 0.7311320754716981, "percentage": 3.66, "elapsed_time": "0:03:23", "remaining_time": "1:29:28", "throughput": 4453.51, "total_tokens": 907200}
+{"current_steps": 1400, "total_steps": 38160, "loss": 0.4845, "lr": 0.0003666142557651992, "epoch": 0.7337526205450734, "percentage": 3.67, "elapsed_time": "0:03:24", "remaining_time": "1:29:30", "throughput": 4455.68, "total_tokens": 911392}
+{"current_steps": 1405, "total_steps": 38160, "loss": 0.5145, "lr": 0.00036792452830188677, "epoch": 0.7363731656184487, "percentage": 3.68, "elapsed_time": "0:03:25", "remaining_time": "1:29:26", "throughput": 4453.09, "total_tokens": 913440}
+{"current_steps": 1410, "total_steps": 38160, "loss": 0.4958, "lr": 0.0003692348008385744, "epoch": 0.7389937106918238, "percentage": 3.69, "elapsed_time": "0:03:26", "remaining_time": "1:29:29", "throughput": 4455.5, "total_tokens": 917888}
+{"current_steps": 1415, "total_steps": 38160, "loss": 0.5504, "lr": 0.00037054507337526204, "epoch": 0.7416142557651991, "percentage": 3.71, "elapsed_time": "0:03:26", "remaining_time": "1:29:29", "throughput": 4456.12, "total_tokens": 921376}
+{"current_steps": 1420, "total_steps": 38160, "loss": 0.5504, "lr": 0.0003718553459119497, "epoch": 0.7442348008385744, "percentage": 3.72, "elapsed_time": "0:03:27", "remaining_time": "1:29:28", "throughput": 4456.66, "total_tokens": 924800}
+{"current_steps": 1425, "total_steps": 38160, "loss": 0.64, "lr": 0.0003731656184486373, "epoch": 0.7468553459119497, "percentage": 3.73, "elapsed_time": "0:03:28", "remaining_time": "1:29:26", "throughput": 4455.72, "total_tokens": 927648}
+{"current_steps": 1430, "total_steps": 38160, "loss": 0.4149, "lr": 0.000374475890985325, "epoch": 0.749475890985325, "percentage": 3.75, "elapsed_time": "0:03:28", "remaining_time": "1:29:23", "throughput": 4454.31, "total_tokens": 930144}
+{"current_steps": 1435, "total_steps": 38160, "loss": 0.413, "lr": 0.00037578616352201263, "epoch": 0.7520964360587002, "percentage": 3.76, "elapsed_time": "0:03:29", "remaining_time": "1:29:22", "throughput": 4453.92, "total_tokens": 933184}
+{"current_steps": 1440, "total_steps": 38160, "loss": 0.5526, "lr": 0.0003770964360587002, "epoch": 0.7547169811320755, "percentage": 3.77, "elapsed_time": "0:03:30", "remaining_time": "1:29:24", "throughput": 4455.84, "total_tokens": 937376}
+{"current_steps": 1445, "total_steps": 38160, "loss": 0.4662, "lr": 0.00037840670859538785, "epoch": 0.7573375262054507, "percentage": 3.79, "elapsed_time": "0:03:31", "remaining_time": "1:29:22", "throughput": 4454.83, "total_tokens": 940160}
+{"current_steps": 1450, "total_steps": 38160, "loss": 0.5484, "lr": 0.0003797169811320755, "epoch": 0.759958071278826, "percentage": 3.8, "elapsed_time": "0:03:31", "remaining_time": "1:29:19", "throughput": 4453.68, "total_tokens": 942816}
+{"current_steps": 1455, "total_steps": 38160, "loss": 0.5784, "lr": 0.0003810272536687631, "epoch": 0.7625786163522013, "percentage": 3.81, "elapsed_time": "0:03:32", "remaining_time": "1:29:18", "throughput": 4453.63, "total_tokens": 945952}
+{"current_steps": 1460, "total_steps": 38160, "loss": 0.3904, "lr": 0.00038233752620545075, "epoch": 0.7651991614255765, "percentage": 3.83, "elapsed_time": "0:03:33", "remaining_time": "1:29:18", "throughput": 4454.24, "total_tokens": 949472}
+{"current_steps": 1465, "total_steps": 38160, "loss": 0.5353, "lr": 0.0003836477987421384, "epoch": 0.7678197064989518, "percentage": 3.84, "elapsed_time": "0:03:33", "remaining_time": "1:29:17", "throughput": 4454.27, "total_tokens": 952672}
+{"current_steps": 1470, "total_steps": 38160, "loss": 0.5603, "lr": 0.00038495807127882596, "epoch": 0.7704402515723271, "percentage": 3.85, "elapsed_time": "0:03:34", "remaining_time": "1:29:14", "throughput": 4453.67, "total_tokens": 955520}
+{"current_steps": 1475, "total_steps": 38160, "loss": 0.5678, "lr": 0.0003862683438155136, "epoch": 0.7730607966457023, "percentage": 3.87, "elapsed_time": "0:03:35", "remaining_time": "1:29:13", "throughput": 4453.2, "total_tokens": 958464}
+{"current_steps": 1480, "total_steps": 38160, "loss": 0.4988, "lr": 0.0003875786163522013, "epoch": 0.7756813417190775, "percentage": 3.88, "elapsed_time": "0:03:35", "remaining_time": "1:29:09", "throughput": 4451.38, "total_tokens": 960768}
+{"current_steps": 1485, "total_steps": 38160, "loss": 0.6904, "lr": 0.0003888888888888889, "epoch": 0.7783018867924528, "percentage": 3.89, "elapsed_time": "0:03:36", "remaining_time": "1:29:06", "throughput": 4450.86, "total_tokens": 963616}
+{"current_steps": 1490, "total_steps": 38160, "loss": 0.7019, "lr": 0.00039019916142557656, "epoch": 0.7809224318658281, "percentage": 3.9, "elapsed_time": "0:03:37", "remaining_time": "1:29:09", "throughput": 4453.32, "total_tokens": 967936}
+{"current_steps": 1495, "total_steps": 38160, "loss": 0.481, "lr": 0.0003915094339622642, "epoch": 0.7835429769392034, "percentage": 3.92, "elapsed_time": "0:03:38", "remaining_time": "1:29:10", "throughput": 4454.56, "total_tokens": 971744}
+{"current_steps": 1500, "total_steps": 38160, "loss": 0.5369, "lr": 0.00039281970649895177, "epoch": 0.7861635220125787, "percentage": 3.93, "elapsed_time": "0:03:38", "remaining_time": "1:29:06", "throughput": 4452.85, "total_tokens": 974080}
+{"current_steps": 1505, "total_steps": 38160, "loss": 0.4812, "lr": 0.0003941299790356394, "epoch": 0.7887840670859538, "percentage": 3.94, "elapsed_time": "0:03:39", "remaining_time": "1:29:04", "throughput": 4452.33, "total_tokens": 976992}
+{"current_steps": 1510, "total_steps": 38160, "loss": 0.5068, "lr": 0.00039544025157232704, "epoch": 0.7914046121593291, "percentage": 3.96, "elapsed_time": "0:03:40", "remaining_time": "1:29:01", "throughput": 4451.25, "total_tokens": 979648}
+{"current_steps": 1515, "total_steps": 38160, "loss": 0.5554, "lr": 0.0003967505241090147, "epoch": 0.7940251572327044, "percentage": 3.97, "elapsed_time": "0:03:40", "remaining_time": "1:29:01", "throughput": 4451.88, "total_tokens": 983136}
+{"current_steps": 1520, "total_steps": 38160, "loss": 0.4946, "lr": 0.0003980607966457023, "epoch": 0.7966457023060797, "percentage": 3.98, "elapsed_time": "0:03:41", "remaining_time": "1:29:02", "throughput": 4453.29, "total_tokens": 986976}
+{"current_steps": 1525, "total_steps": 38160, "loss": 0.4699, "lr": 0.00039937106918238994, "epoch": 0.799266247379455, "percentage": 4.0, "elapsed_time": "0:03:42", "remaining_time": "1:29:01", "throughput": 4453.47, "total_tokens": 990272}
+{"current_steps": 1530, "total_steps": 38160, "loss": 0.4808, "lr": 0.0004006813417190776, "epoch": 0.8018867924528302, "percentage": 4.01, "elapsed_time": "0:03:43", "remaining_time": "1:29:03", "throughput": 4455.29, "total_tokens": 994368}
+{"current_steps": 1535, "total_steps": 38160, "loss": 0.4407, "lr": 0.0004019916142557652, "epoch": 0.8045073375262054, "percentage": 4.02, "elapsed_time": "0:03:43", "remaining_time": "1:29:02", "throughput": 4455.27, "total_tokens": 997504}
+{"current_steps": 1540, "total_steps": 38160, "loss": 0.568, "lr": 0.00040330188679245284, "epoch": 0.8071278825995807, "percentage": 4.04, "elapsed_time": "0:03:44", "remaining_time": "1:28:59", "throughput": 4454.63, "total_tokens": 1000352}
+{"current_steps": 1545, "total_steps": 38160, "loss": 0.58, "lr": 0.0004046121593291405, "epoch": 0.809748427672956, "percentage": 4.05, "elapsed_time": "0:03:45", "remaining_time": "1:28:58", "throughput": 4454.91, "total_tokens": 1003616}
+{"current_steps": 1550, "total_steps": 38160, "loss": 0.5682, "lr": 0.0004059224318658281, "epoch": 0.8123689727463312, "percentage": 4.06, "elapsed_time": "0:03:45", "remaining_time": "1:28:56", "throughput": 4454.03, "total_tokens": 1006240}
+{"current_steps": 1555, "total_steps": 38160, "loss": 0.5371, "lr": 0.00040723270440251575, "epoch": 0.8149895178197065, "percentage": 4.07, "elapsed_time": "0:03:46", "remaining_time": "1:28:55", "throughput": 4454.0, "total_tokens": 1009440}
+{"current_steps": 1560, "total_steps": 38160, "loss": 0.5506, "lr": 0.00040854297693920333, "epoch": 0.8176100628930818, "percentage": 4.09, "elapsed_time": "0:03:47", "remaining_time": "1:28:51", "throughput": 4452.75, "total_tokens": 1011936}
+{"current_steps": 1565, "total_steps": 38160, "loss": 0.4716, "lr": 0.00040985324947589096, "epoch": 0.820230607966457, "percentage": 4.1, "elapsed_time": "0:03:47", "remaining_time": "1:28:50", "throughput": 4452.69, "total_tokens": 1015040}
+{"current_steps": 1570, "total_steps": 38160, "loss": 0.5998, "lr": 0.0004111635220125786, "epoch": 0.8228511530398323, "percentage": 4.11, "elapsed_time": "0:03:48", "remaining_time": "1:28:49", "throughput": 4452.66, "total_tokens": 1018240}
+{"current_steps": 1575, "total_steps": 38160, "loss": 0.503, "lr": 0.00041247379454926623, "epoch": 0.8254716981132075, "percentage": 4.13, "elapsed_time": "0:03:49", "remaining_time": "1:28:50", "throughput": 4454.63, "total_tokens": 1022336}
+{"current_steps": 1580, "total_steps": 38160, "loss": 0.5135, "lr": 0.0004137840670859539, "epoch": 0.8280922431865828, "percentage": 4.14, "elapsed_time": "0:03:50", "remaining_time": "1:28:48", "throughput": 4454.0, "total_tokens": 1025184}
+{"current_steps": 1585, "total_steps": 38160, "loss": 0.5793, "lr": 0.00041509433962264155, "epoch": 0.8307127882599581, "percentage": 4.15, "elapsed_time": "0:03:50", "remaining_time": "1:28:48", "throughput": 4454.21, "total_tokens": 1028448}
+{"current_steps": 1590, "total_steps": 38160, "loss": 0.5545, "lr": 0.00041640461215932913, "epoch": 0.8333333333333334, "percentage": 4.17, "elapsed_time": "0:03:51", "remaining_time": "1:28:45", "throughput": 4453.04, "total_tokens": 1031072}
+{"current_steps": 1595, "total_steps": 38160, "loss": 0.476, "lr": 0.00041771488469601677, "epoch": 0.8359538784067087, "percentage": 4.18, "elapsed_time": "0:03:52", "remaining_time": "1:28:44", "throughput": 4453.05, "total_tokens": 1034208}
+{"current_steps": 1600, "total_steps": 38160, "loss": 0.4526, "lr": 0.0004190251572327044, "epoch": 0.8385744234800838, "percentage": 4.19, "elapsed_time": "0:03:53", "remaining_time": "1:28:46", "throughput": 4455.19, "total_tokens": 1038624}
+{"current_steps": 1605, "total_steps": 38160, "loss": 0.5629, "lr": 0.00042033542976939204, "epoch": 0.8411949685534591, "percentage": 4.21, "elapsed_time": "0:03:53", "remaining_time": "1:28:44", "throughput": 4453.91, "total_tokens": 1041152}
+{"current_steps": 1610, "total_steps": 38160, "loss": 0.4165, "lr": 0.00042164570230607967, "epoch": 0.8438155136268344, "percentage": 4.22, "elapsed_time": "0:03:54", "remaining_time": "1:28:41", "throughput": 4453.21, "total_tokens": 1043904}
+{"current_steps": 1615, "total_steps": 38160, "loss": 0.5203, "lr": 0.0004229559748427673, "epoch": 0.8464360587002097, "percentage": 4.23, "elapsed_time": "0:03:55", "remaining_time": "1:28:41", "throughput": 4453.86, "total_tokens": 1047456}
+{"current_steps": 1620, "total_steps": 38160, "loss": 0.5663, "lr": 0.00042426624737945494, "epoch": 0.8490566037735849, "percentage": 4.25, "elapsed_time": "0:03:55", "remaining_time": "1:28:39", "throughput": 4453.07, "total_tokens": 1050304}
+{"current_steps": 1625, "total_steps": 38160, "loss": 0.5146, "lr": 0.0004255765199161425, "epoch": 0.8516771488469602, "percentage": 4.26, "elapsed_time": "0:03:56", "remaining_time": "1:28:39", "throughput": 4453.15, "total_tokens": 1053600}
+{"current_steps": 1630, "total_steps": 38160, "loss": 0.3679, "lr": 0.0004268867924528302, "epoch": 0.8542976939203354, "percentage": 4.27, "elapsed_time": "0:03:57", "remaining_time": "1:28:38", "throughput": 4452.98, "total_tokens": 1056672}
+{"current_steps": 1635, "total_steps": 38160, "loss": 0.3992, "lr": 0.00042819706498951784, "epoch": 0.8569182389937107, "percentage": 4.28, "elapsed_time": "0:03:58", "remaining_time": "1:28:37", "throughput": 4452.86, "total_tokens": 1059904}
+{"current_steps": 1640, "total_steps": 38160, "loss": 0.5048, "lr": 0.0004295073375262055, "epoch": 0.859538784067086, "percentage": 4.3, "elapsed_time": "0:03:58", "remaining_time": "1:28:36", "throughput": 4453.5, "total_tokens": 1063360}
+{"current_steps": 1645, "total_steps": 38160, "loss": 0.508, "lr": 0.0004308176100628931, "epoch": 0.8621593291404612, "percentage": 4.31, "elapsed_time": "0:03:59", "remaining_time": "1:28:33", "throughput": 4452.06, "total_tokens": 1065728}
+{"current_steps": 1650, "total_steps": 38160, "loss": 0.481, "lr": 0.00043212788259958075, "epoch": 0.8647798742138365, "percentage": 4.32, "elapsed_time": "0:04:00", "remaining_time": "1:28:31", "throughput": 4451.19, "total_tokens": 1068448}
+{"current_steps": 1655, "total_steps": 38160, "loss": 0.5102, "lr": 0.0004334381551362683, "epoch": 0.8674004192872118, "percentage": 4.34, "elapsed_time": "0:04:00", "remaining_time": "1:28:28", "throughput": 4450.29, "total_tokens": 1071104}
+{"current_steps": 1660, "total_steps": 38160, "loss": 0.4178, "lr": 0.00043474842767295596, "epoch": 0.870020964360587, "percentage": 4.35, "elapsed_time": "0:04:01", "remaining_time": "1:28:30", "throughput": 4451.6, "total_tokens": 1075072}
+{"current_steps": 1665, "total_steps": 38160, "loss": 0.5128, "lr": 0.0004360587002096436, "epoch": 0.8726415094339622, "percentage": 4.36, "elapsed_time": "0:04:02", "remaining_time": "1:28:27", "throughput": 4450.98, "total_tokens": 1077824}
+{"current_steps": 1670, "total_steps": 38160, "loss": 0.4771, "lr": 0.00043736897274633123, "epoch": 0.8752620545073375, "percentage": 4.38, "elapsed_time": "0:04:02", "remaining_time": "1:28:26", "throughput": 4450.73, "total_tokens": 1080832}
+{"current_steps": 1675, "total_steps": 38160, "loss": 0.543, "lr": 0.00043867924528301886, "epoch": 0.8778825995807128, "percentage": 4.39, "elapsed_time": "0:04:03", "remaining_time": "1:28:25", "throughput": 4450.83, "total_tokens": 1084096}
+{"current_steps": 1680, "total_steps": 38160, "loss": 0.5159, "lr": 0.00043998951781970655, "epoch": 0.8805031446540881, "percentage": 4.4, "elapsed_time": "0:04:04", "remaining_time": "1:28:24", "throughput": 4450.39, "total_tokens": 1087104}
+{"current_steps": 1685, "total_steps": 38160, "loss": 0.5715, "lr": 0.00044129979035639413, "epoch": 0.8831236897274634, "percentage": 4.42, "elapsed_time": "0:04:04", "remaining_time": "1:28:22", "throughput": 4450.63, "total_tokens": 1090304}
+{"current_steps": 1690, "total_steps": 38160, "loss": 0.7883, "lr": 0.00044261006289308177, "epoch": 0.8857442348008385, "percentage": 4.43, "elapsed_time": "0:04:05", "remaining_time": "1:28:23", "throughput": 4451.75, "total_tokens": 1094112}
+{"current_steps": 1695, "total_steps": 38160, "loss": 0.5333, "lr": 0.0004439203354297694, "epoch": 0.8883647798742138, "percentage": 4.44, "elapsed_time": "0:04:06", "remaining_time": "1:28:26", "throughput": 4453.65, "total_tokens": 1098496}
+{"current_steps": 1700, "total_steps": 38160, "loss": 0.4503, "lr": 0.00044523060796645704, "epoch": 0.8909853249475891, "percentage": 4.45, "elapsed_time": "0:04:07", "remaining_time": "1:28:23", "throughput": 4452.38, "total_tokens": 1100992}
+{"current_steps": 1705, "total_steps": 38160, "loss": 0.5451, "lr": 0.00044654088050314467, "epoch": 0.8936058700209644, "percentage": 4.47, "elapsed_time": "0:04:07", "remaining_time": "1:28:21", "throughput": 4451.74, "total_tokens": 1103776}
+{"current_steps": 1710, "total_steps": 38160, "loss": 0.452, "lr": 0.0004478511530398323, "epoch": 0.8962264150943396, "percentage": 4.48, "elapsed_time": "0:04:08", "remaining_time": "1:28:18", "throughput": 4450.71, "total_tokens": 1106368}
+{"current_steps": 1715, "total_steps": 38160, "loss": 0.4424, "lr": 0.0004491614255765199, "epoch": 0.8988469601677149, "percentage": 4.49, "elapsed_time": "0:04:09", "remaining_time": "1:28:17", "throughput": 4450.3, "total_tokens": 1109376}
+{"current_steps": 1720, "total_steps": 38160, "loss": 0.4653, "lr": 0.0004504716981132075, "epoch": 0.9014675052410901, "percentage": 4.51, "elapsed_time": "0:04:09", "remaining_time": "1:28:16", "throughput": 4449.96, "total_tokens": 1112448}
+{"current_steps": 1725, "total_steps": 38160, "loss": 0.4825, "lr": 0.00045178197064989515, "epoch": 0.9040880503144654, "percentage": 4.52, "elapsed_time": "0:04:10", "remaining_time": "1:28:15", "throughput": 4449.67, "total_tokens": 1115520}
+{"current_steps": 1730, "total_steps": 38160, "loss": 0.6451, "lr": 0.00045309224318658284, "epoch": 0.9067085953878407, "percentage": 4.53, "elapsed_time": "0:04:11", "remaining_time": "1:28:14", "throughput": 4450.01, "total_tokens": 1118784}
+{"current_steps": 1735, "total_steps": 38160, "loss": 0.7135, "lr": 0.0004544025157232705, "epoch": 0.9093291404612159, "percentage": 4.55, "elapsed_time": "0:04:12", "remaining_time": "1:28:13", "throughput": 4450.44, "total_tokens": 1122112}
+{"current_steps": 1740, "total_steps": 38160, "loss": 0.4702, "lr": 0.0004557127882599581, "epoch": 0.9119496855345912, "percentage": 4.56, "elapsed_time": "0:04:12", "remaining_time": "1:28:13", "throughput": 4451.72, "total_tokens": 1125952}
+{"current_steps": 1745, "total_steps": 38160, "loss": 0.657, "lr": 0.0004570230607966457, "epoch": 0.9145702306079665, "percentage": 4.57, "elapsed_time": "0:04:13", "remaining_time": "1:28:12", "throughput": 4451.66, "total_tokens": 1129024}
+{"current_steps": 1750, "total_steps": 38160, "loss": 0.4666, "lr": 0.0004583333333333333, "epoch": 0.9171907756813418, "percentage": 4.59, "elapsed_time": "0:04:14", "remaining_time": "1:28:09", "throughput": 4450.34, "total_tokens": 1131488}
+{"current_steps": 1755, "total_steps": 38160, "loss": 0.4721, "lr": 0.00045964360587002096, "epoch": 0.9198113207547169, "percentage": 4.6, "elapsed_time": "0:04:14", "remaining_time": "1:28:07", "throughput": 4449.54, "total_tokens": 1134176}
+{"current_steps": 1760, "total_steps": 38160, "loss": 0.5269, "lr": 0.0004609538784067086, "epoch": 0.9224318658280922, "percentage": 4.61, "elapsed_time": "0:04:15", "remaining_time": "1:28:07", "throughput": 4450.54, "total_tokens": 1137888}
+{"current_steps": 1765, "total_steps": 38160, "loss": 0.615, "lr": 0.00046226415094339623, "epoch": 0.9250524109014675, "percentage": 4.63, "elapsed_time": "0:04:16", "remaining_time": "1:28:06", "throughput": 4450.92, "total_tokens": 1141184}
+{"current_steps": 1770, "total_steps": 38160, "loss": 0.5396, "lr": 0.00046357442348008386, "epoch": 0.9276729559748428, "percentage": 4.64, "elapsed_time": "0:04:17", "remaining_time": "1:28:04", "throughput": 4450.55, "total_tokens": 1144064}
+{"current_steps": 1775, "total_steps": 38160, "loss": 0.5017, "lr": 0.00046488469601677155, "epoch": 0.9302935010482181, "percentage": 4.65, "elapsed_time": "0:04:17", "remaining_time": "1:28:04", "throughput": 4451.24, "total_tokens": 1147584}
+{"current_steps": 1780, "total_steps": 38160, "loss": 0.4822, "lr": 0.00046619496855345913, "epoch": 0.9329140461215933, "percentage": 4.66, "elapsed_time": "0:04:19", "remaining_time": "1:28:15", "throughput": 4457.0, "total_tokens": 1154784}
+{"current_steps": 1785, "total_steps": 38160, "loss": 0.4725, "lr": 0.00046750524109014677, "epoch": 0.9355345911949685, "percentage": 4.68, "elapsed_time": "0:04:19", "remaining_time": "1:28:12", "throughput": 4456.25, "total_tokens": 1157440}
+{"current_steps": 1790, "total_steps": 38160, "loss": 0.6405, "lr": 0.0004688155136268344, "epoch": 0.9381551362683438, "percentage": 4.69, "elapsed_time": "0:04:20", "remaining_time": "1:28:10", "throughput": 4455.5, "total_tokens": 1160224}
+{"current_steps": 1795, "total_steps": 38160, "loss": 0.5172, "lr": 0.00047012578616352203, "epoch": 0.9407756813417191, "percentage": 4.7, "elapsed_time": "0:04:21", "remaining_time": "1:28:12", "throughput": 4457.44, "total_tokens": 1164480}
+{"current_steps": 1800, "total_steps": 38160, "loss": 0.3977, "lr": 0.00047143605870020967, "epoch": 0.9433962264150944, "percentage": 4.72, "elapsed_time": "0:04:22", "remaining_time": "1:28:13", "throughput": 4459.39, "total_tokens": 1168704}
+{"current_steps": 1805, "total_steps": 38160, "loss": 0.4354, "lr": 0.0004727463312368973, "epoch": 0.9460167714884696, "percentage": 4.73, "elapsed_time": "0:04:22", "remaining_time": "1:28:14", "throughput": 4460.25, "total_tokens": 1172384}
+{"current_steps": 1810, "total_steps": 38160, "loss": 0.4943, "lr": 0.0004740566037735849, "epoch": 0.9486373165618449, "percentage": 4.74, "elapsed_time": "0:04:23", "remaining_time": "1:28:12", "throughput": 4460.03, "total_tokens": 1175392}
+{"current_steps": 1815, "total_steps": 38160, "loss": 0.4631, "lr": 0.0004753668763102725, "epoch": 0.9512578616352201, "percentage": 4.76, "elapsed_time": "0:04:24", "remaining_time": "1:28:11", "throughput": 4459.73, "total_tokens": 1178368}
+{"current_steps": 1820, "total_steps": 38160, "loss": 0.4071, "lr": 0.00047667714884696015, "epoch": 0.9538784067085954, "percentage": 4.77, "elapsed_time": "0:04:25", "remaining_time": "1:28:11", "throughput": 4460.77, "total_tokens": 1182240}
+{"current_steps": 1825, "total_steps": 38160, "loss": 0.4638, "lr": 0.00047798742138364784, "epoch": 0.9564989517819706, "percentage": 4.78, "elapsed_time": "0:04:25", "remaining_time": "1:28:11", "throughput": 4460.83, "total_tokens": 1185504}
+{"current_steps": 1830, "total_steps": 38160, "loss": 0.4824, "lr": 0.0004792976939203355, "epoch": 0.9591194968553459, "percentage": 4.8, "elapsed_time": "0:04:26", "remaining_time": "1:28:09", "throughput": 4460.21, "total_tokens": 1188288}
+{"current_steps": 1835, "total_steps": 38160, "loss": 0.457, "lr": 0.0004806079664570231, "epoch": 0.9617400419287212, "percentage": 4.81, "elapsed_time": "0:04:27", "remaining_time": "1:28:09", "throughput": 4460.92, "total_tokens": 1192064}
+{"current_steps": 1840, "total_steps": 38160, "loss": 0.461, "lr": 0.0004819182389937107, "epoch": 0.9643605870020965, "percentage": 4.82, "elapsed_time": "0:04:27", "remaining_time": "1:28:08", "throughput": 4460.82, "total_tokens": 1195104}
+{"current_steps": 1845, "total_steps": 38160, "loss": 0.6006, "lr": 0.0004832285115303983, "epoch": 0.9669811320754716, "percentage": 4.83, "elapsed_time": "0:04:28", "remaining_time": "1:28:06", "throughput": 4460.17, "total_tokens": 1197920}
+{"current_steps": 1850, "total_steps": 38160, "loss": 0.4591, "lr": 0.00048453878406708596, "epoch": 0.9696016771488469, "percentage": 4.85, "elapsed_time": "0:04:29", "remaining_time": "1:28:04", "throughput": 4459.81, "total_tokens": 1200896}
+{"current_steps": 1855, "total_steps": 38160, "loss": 0.4771, "lr": 0.0004858490566037736, "epoch": 0.9722222222222222, "percentage": 4.86, "elapsed_time": "0:04:29", "remaining_time": "1:28:03", "throughput": 4459.74, "total_tokens": 1204032}
+{"current_steps": 1860, "total_steps": 38160, "loss": 0.4047, "lr": 0.0004871593291404612, "epoch": 0.9748427672955975, "percentage": 4.87, "elapsed_time": "0:04:30", "remaining_time": "1:28:02", "throughput": 4459.83, "total_tokens": 1207264}
+{"current_steps": 1865, "total_steps": 38160, "loss": 0.5478, "lr": 0.0004884696016771489, "epoch": 0.9774633123689728, "percentage": 4.89, "elapsed_time": "0:04:31", "remaining_time": "1:28:00", "throughput": 4459.16, "total_tokens": 1209984}
+{"current_steps": 1870, "total_steps": 38160, "loss": 0.4362, "lr": 0.0004897798742138365, "epoch": 0.980083857442348, "percentage": 4.9, "elapsed_time": "0:04:32", "remaining_time": "1:28:01", "throughput": 4460.16, "total_tokens": 1213856}
+{"current_steps": 1875, "total_steps": 38160, "loss": 0.4379, "lr": 0.0004910901467505241, "epoch": 0.9827044025157232, "percentage": 4.91, "elapsed_time": "0:04:32", "remaining_time": "1:28:01", "throughput": 4460.61, "total_tokens": 1217408}
+{"current_steps": 1880, "total_steps": 38160, "loss": 0.5103, "lr": 0.0004924004192872118, "epoch": 0.9853249475890985, "percentage": 4.93, "elapsed_time": "0:04:33", "remaining_time": "1:28:01", "throughput": 4461.24, "total_tokens": 1221024}
+{"current_steps": 1885, "total_steps": 38160, "loss": 0.4078, "lr": 0.0004937106918238993, "epoch": 0.9879454926624738, "percentage": 4.94, "elapsed_time": "0:04:34", "remaining_time": "1:28:01", "throughput": 4461.37, "total_tokens": 1224320}
+{"current_steps": 1890, "total_steps": 38160, "loss": 0.3956, "lr": 0.000495020964360587, "epoch": 0.9905660377358491, "percentage": 4.95, "elapsed_time": "0:04:35", "remaining_time": "1:28:01", "throughput": 4462.27, "total_tokens": 1228000}
+{"current_steps": 1895, "total_steps": 38160, "loss": 0.4922, "lr": 0.0004963312368972746, "epoch": 0.9931865828092243, "percentage": 4.97, "elapsed_time": "0:04:36", "remaining_time": "1:28:02", "throughput": 4464.22, "total_tokens": 1232320}
+{"current_steps": 1900, "total_steps": 38160, "loss": 0.5838, "lr": 0.0004976415094339623, "epoch": 0.9958071278825996, "percentage": 4.98, "elapsed_time": "0:04:37", "remaining_time": "1:28:09", "throughput": 4467.79, "total_tokens": 1238272}
+{"current_steps": 1905, "total_steps": 38160, "loss": 0.4991, "lr": 0.0004989517819706499, "epoch": 0.9984276729559748, "percentage": 4.99, "elapsed_time": "0:04:37", "remaining_time": "1:28:07", "throughput": 4467.65, "total_tokens": 1241344}
+{"current_steps": 1908, "total_steps": 38160, "eval_loss": 0.47990021109580994, "epoch": 1.0, "percentage": 5.0, "elapsed_time": "0:04:51", "remaining_time": "1:32:25", "throughput": 4258.72, "total_tokens": 1243048}
+{"current_steps": 1910, "total_steps": 38160, "loss": 0.4387, "lr": 0.0005002620545073376, "epoch": 1.00104821802935, "percentage": 5.01, "elapsed_time": "0:04:54", "remaining_time": "1:33:02", "throughput": 4231.21, "total_tokens": 1244520}
+{"current_steps": 1915, "total_steps": 38160, "loss": 0.4998, "lr": 0.0005015723270440253, "epoch": 1.0036687631027255, "percentage": 5.02, "elapsed_time": "0:04:54", "remaining_time": "1:33:00", "throughput": 4231.48, "total_tokens": 1247528}
+{"current_steps": 1920, "total_steps": 38160, "loss": 0.5969, "lr": 0.0005028825995807128, "epoch": 1.0062893081761006, "percentage": 5.03, "elapsed_time": "0:04:55", "remaining_time": "1:32:58", "throughput": 4232.05, "total_tokens": 1250824}
+{"current_steps": 1925, "total_steps": 38160, "loss": 0.5149, "lr": 0.0005041928721174004, "epoch": 1.0089098532494758, "percentage": 5.04, "elapsed_time": "0:04:56", "remaining_time": "1:32:57", "throughput": 4232.52, "total_tokens": 1254024}
+{"current_steps": 1930, "total_steps": 38160, "loss": 0.4696, "lr": 0.0005055031446540881, "epoch": 1.0115303983228512, "percentage": 5.06, "elapsed_time": "0:04:57", "remaining_time": "1:33:03", "throughput": 4237.56, "total_tokens": 1260328}
+{"current_steps": 1935, "total_steps": 38160, "loss": 0.6895, "lr": 0.0005068134171907757, "epoch": 1.0141509433962264, "percentage": 5.07, "elapsed_time": "0:04:58", "remaining_time": "1:33:01", "throughput": 4238.34, "total_tokens": 1263656}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc90822302257416ab9eff3ad5cb4107ec07c36dd65012367a7f19c07e6e64f1
+size 6353