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 +1,15 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
+  "exclude_modules": null,
+  "inference_mode": true,
+  "modules_to_save": null,
+  "peft_type": "LN_TUNING",
+  "revision": null,
+  "target_modules": [
+    "post_attention_layernorm",
+    "input_layernorm",
+    "norm"
+  ],
+  "task_type": "CAUSAL_LM"
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:51da5a70cc9b38779901c89c68524c5d93c122f7daab15cce05791e371ebd55a
+size 1074144

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 +1,56 @@
+seed: 123
+
+### 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: lntuning
+task_type: CAUSAL_LM
+
+### 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/lntuning/llama-3-8b-instruct/train_codealpacapy_123_1762561812
+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: 5.0e-5
+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: lntuning_llama-3-8b-instruct_train_codealpacapy_123_1762561812
+
+### 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": 1.257, "lr": 5.241090146750525e-08, "epoch": 0.002620545073375262, "percentage": 0.01, "elapsed_time": "0:00:01", "remaining_time": "3:08:33", "throughput": 1985.75, "total_tokens": 2944}
+{"current_steps": 10, "total_steps": 38160, "loss": 1.1154, "lr": 1.1792452830188679e-07, "epoch": 0.005241090146750524, "percentage": 0.03, "elapsed_time": "0:00:02", "remaining_time": "2:27:36", "throughput": 2935.98, "total_tokens": 6816}
+{"current_steps": 15, "total_steps": 38160, "loss": 1.2213, "lr": 1.8343815513626837e-07, "epoch": 0.007861635220125786, "percentage": 0.04, "elapsed_time": "0:00:03", "remaining_time": "2:09:51", "throughput": 3185.3, "total_tokens": 9760}
+{"current_steps": 20, "total_steps": 38160, "loss": 1.1648, "lr": 2.489517819706499e-07, "epoch": 0.010482180293501049, "percentage": 0.05, "elapsed_time": "0:00:03", "remaining_time": "2:07:04", "throughput": 3369.4, "total_tokens": 13472}
+{"current_steps": 25, "total_steps": 38160, "loss": 1.2346, "lr": 3.144654088050315e-07, "epoch": 0.01310272536687631, "percentage": 0.07, "elapsed_time": "0:00:04", "remaining_time": "2:04:39", "throughput": 3439.38, "total_tokens": 16864}
+{"current_steps": 30, "total_steps": 38160, "loss": 1.3564, "lr": 3.79979035639413e-07, "epoch": 0.015723270440251572, "percentage": 0.08, "elapsed_time": "0:00:05", "remaining_time": "2:01:41", "throughput": 3542.68, "total_tokens": 20352}
+{"current_steps": 35, "total_steps": 38160, "loss": 1.3339, "lr": 4.454926624737945e-07, "epoch": 0.018343815513626835, "percentage": 0.09, "elapsed_time": "0:00:06", "remaining_time": "1:57:42", "throughput": 3548.55, "total_tokens": 23008}
+{"current_steps": 40, "total_steps": 38160, "loss": 1.0737, "lr": 5.110062893081761e-07, "epoch": 0.020964360587002098, "percentage": 0.1, "elapsed_time": "0:00:07", "remaining_time": "1:54:43", "throughput": 3588.47, "total_tokens": 25920}
+{"current_steps": 45, "total_steps": 38160, "loss": 1.2294, "lr": 5.765199161425577e-07, "epoch": 0.02358490566037736, "percentage": 0.12, "elapsed_time": "0:00:07", "remaining_time": "1:51:49", "throughput": 3619.76, "total_tokens": 28672}
+{"current_steps": 50, "total_steps": 38160, "loss": 1.0775, "lr": 6.420335429769392e-07, "epoch": 0.02620545073375262, "percentage": 0.13, "elapsed_time": "0:00:08", "remaining_time": "1:50:16", "throughput": 3627.15, "total_tokens": 31488}
+{"current_steps": 55, "total_steps": 38160, "loss": 1.2133, "lr": 7.075471698113208e-07, "epoch": 0.028825995807127882, "percentage": 0.14, "elapsed_time": "0:00:09", "remaining_time": "1:49:40", "throughput": 3631.78, "total_tokens": 34496}
+{"current_steps": 60, "total_steps": 38160, "loss": 1.2943, "lr": 7.730607966457024e-07, "epoch": 0.031446540880503145, "percentage": 0.16, "elapsed_time": "0:00:10", "remaining_time": "1:49:10", "throughput": 3694.53, "total_tokens": 38112}
+{"current_steps": 65, "total_steps": 38160, "loss": 1.3676, "lr": 8.38574423480084e-07, "epoch": 0.034067085953878404, "percentage": 0.17, "elapsed_time": "0:00:11", "remaining_time": "1:47:44", "throughput": 3704.74, "total_tokens": 40864}
+{"current_steps": 70, "total_steps": 38160, "loss": 1.2801, "lr": 9.040880503144654e-07, "epoch": 0.03668763102725367, "percentage": 0.18, "elapsed_time": "0:00:11", "remaining_time": "1:47:12", "throughput": 3738.6, "total_tokens": 44192}
+{"current_steps": 75, "total_steps": 38160, "loss": 1.4051, "lr": 9.69601677148847e-07, "epoch": 0.03930817610062893, "percentage": 0.2, "elapsed_time": "0:00:12", "remaining_time": "1:45:45", "throughput": 3743.82, "total_tokens": 46784}
+{"current_steps": 80, "total_steps": 38160, "loss": 1.3892, "lr": 1.0351153039832286e-06, "epoch": 0.041928721174004195, "percentage": 0.21, "elapsed_time": "0:00:13", "remaining_time": "1:45:43", "throughput": 3765.27, "total_tokens": 50176}
+{"current_steps": 85, "total_steps": 38160, "loss": 1.0898, "lr": 1.10062893081761e-06, "epoch": 0.044549266247379454, "percentage": 0.22, "elapsed_time": "0:00:14", "remaining_time": "1:46:21", "throughput": 3818.43, "total_tokens": 54400}
+{"current_steps": 90, "total_steps": 38160, "loss": 1.4386, "lr": 1.1661425576519918e-06, "epoch": 0.04716981132075472, "percentage": 0.24, "elapsed_time": "0:00:14", "remaining_time": "1:45:17", "throughput": 3815.93, "total_tokens": 56992}
+{"current_steps": 95, "total_steps": 38160, "loss": 1.0163, "lr": 1.2316561844863732e-06, "epoch": 0.04979035639412998, "percentage": 0.25, "elapsed_time": "0:00:15", "remaining_time": "1:45:36", "throughput": 3850.98, "total_tokens": 60896}
+{"current_steps": 100, "total_steps": 38160, "loss": 1.1498, "lr": 1.2971698113207548e-06, "epoch": 0.05241090146750524, "percentage": 0.26, "elapsed_time": "0:00:16", "remaining_time": "1:45:30", "throughput": 3882.61, "total_tokens": 64576}
+{"current_steps": 105, "total_steps": 38160, "loss": 1.3792, "lr": 1.3626834381551364e-06, "epoch": 0.055031446540880505, "percentage": 0.28, "elapsed_time": "0:00:17", "remaining_time": "1:45:09", "throughput": 3883.8, "total_tokens": 67616}
+{"current_steps": 110, "total_steps": 38160, "loss": 1.0257, "lr": 1.4281970649895177e-06, "epoch": 0.057651991614255764, "percentage": 0.29, "elapsed_time": "0:00:18", "remaining_time": "1:46:29", "throughput": 3927.39, "total_tokens": 72544}
+{"current_steps": 115, "total_steps": 38160, "loss": 1.2987, "lr": 1.4937106918238995e-06, "epoch": 0.06027253668763103, "percentage": 0.3, "elapsed_time": "0:00:19", "remaining_time": "1:46:10", "throughput": 3928.28, "total_tokens": 75648}
+{"current_steps": 120, "total_steps": 38160, "loss": 1.386, "lr": 1.559224318658281e-06, "epoch": 0.06289308176100629, "percentage": 0.31, "elapsed_time": "0:00:19", "remaining_time": "1:45:32", "throughput": 3923.31, "total_tokens": 78368}
+{"current_steps": 125, "total_steps": 38160, "loss": 1.239, "lr": 1.6247379454926625e-06, "epoch": 0.06551362683438156, "percentage": 0.33, "elapsed_time": "0:00:20", "remaining_time": "1:45:31", "throughput": 3938.35, "total_tokens": 81952}
+{"current_steps": 130, "total_steps": 38160, "loss": 1.1895, "lr": 1.690251572327044e-06, "epoch": 0.06813417190775681, "percentage": 0.34, "elapsed_time": "0:00:21", "remaining_time": "1:44:59", "throughput": 3945.11, "total_tokens": 84960}
+{"current_steps": 135, "total_steps": 38160, "loss": 1.3062, "lr": 1.7557651991614257e-06, "epoch": 0.07075471698113207, "percentage": 0.35, "elapsed_time": "0:00:22", "remaining_time": "1:44:58", "throughput": 3960.88, "total_tokens": 88576}
+{"current_steps": 140, "total_steps": 38160, "loss": 1.0192, "lr": 1.821278825995807e-06, "epoch": 0.07337526205450734, "percentage": 0.37, "elapsed_time": "0:00:23", "remaining_time": "1:44:42", "throughput": 3983.98, "total_tokens": 92160}
+{"current_steps": 145, "total_steps": 38160, "loss": 1.1054, "lr": 1.8867924528301887e-06, "epoch": 0.0759958071278826, "percentage": 0.38, "elapsed_time": "0:00:23", "remaining_time": "1:44:33", "throughput": 4005.39, "total_tokens": 95840}
+{"current_steps": 150, "total_steps": 38160, "loss": 1.1889, "lr": 1.9523060796645703e-06, "epoch": 0.07861635220125786, "percentage": 0.39, "elapsed_time": "0:00:24", "remaining_time": "1:44:10", "throughput": 4005.87, "total_tokens": 98816}
+{"current_steps": 155, "total_steps": 38160, "loss": 0.9874, "lr": 2.017819706498952e-06, "epoch": 0.08123689727463312, "percentage": 0.41, "elapsed_time": "0:00:25", "remaining_time": "1:44:24", "throughput": 4026.46, "total_tokens": 102880}
+{"current_steps": 160, "total_steps": 38160, "loss": 1.1657, "lr": 2.0833333333333334e-06, "epoch": 0.08385744234800839, "percentage": 0.42, "elapsed_time": "0:00:26", "remaining_time": "1:44:00", "throughput": 4031.36, "total_tokens": 105920}
+{"current_steps": 165, "total_steps": 38160, "loss": 1.3796, "lr": 2.148846960167715e-06, "epoch": 0.08647798742138364, "percentage": 0.43, "elapsed_time": "0:00:27", "remaining_time": "1:43:52", "throughput": 4019.57, "total_tokens": 108800}
+{"current_steps": 170, "total_steps": 38160, "loss": 1.088, "lr": 2.2143605870020966e-06, "epoch": 0.08909853249475891, "percentage": 0.45, "elapsed_time": "0:00:27", "remaining_time": "1:43:35", "throughput": 4025.74, "total_tokens": 111968}
+{"current_steps": 175, "total_steps": 38160, "loss": 1.1693, "lr": 2.279874213836478e-06, "epoch": 0.09171907756813417, "percentage": 0.46, "elapsed_time": "0:00:28", "remaining_time": "1:43:25", "throughput": 4021.64, "total_tokens": 114976}
+{"current_steps": 180, "total_steps": 38160, "loss": 1.4241, "lr": 2.34538784067086e-06, "epoch": 0.09433962264150944, "percentage": 0.47, "elapsed_time": "0:00:29", "remaining_time": "1:43:17", "throughput": 4002.63, "total_tokens": 117568}
+{"current_steps": 185, "total_steps": 38160, "loss": 1.0373, "lr": 2.410901467505241e-06, "epoch": 0.09696016771488469, "percentage": 0.48, "elapsed_time": "0:00:30", "remaining_time": "1:43:03", "throughput": 3996.37, "total_tokens": 120384}
+{"current_steps": 190, "total_steps": 38160, "loss": 1.0023, "lr": 2.476415094339623e-06, "epoch": 0.09958071278825996, "percentage": 0.5, "elapsed_time": "0:00:30", "remaining_time": "1:42:53", "throughput": 4003.33, "total_tokens": 123680}
+{"current_steps": 195, "total_steps": 38160, "loss": 1.2404, "lr": 2.5419287211740044e-06, "epoch": 0.10220125786163523, "percentage": 0.51, "elapsed_time": "0:00:31", "remaining_time": "1:42:53", "throughput": 4010.44, "total_tokens": 127168}
+{"current_steps": 200, "total_steps": 38160, "loss": 0.9805, "lr": 2.6074423480083857e-06, "epoch": 0.10482180293501048, "percentage": 0.52, "elapsed_time": "0:00:32", "remaining_time": "1:42:43", "throughput": 4008.36, "total_tokens": 130176}
+{"current_steps": 205, "total_steps": 38160, "loss": 1.1461, "lr": 2.6729559748427675e-06, "epoch": 0.10744234800838574, "percentage": 0.54, "elapsed_time": "0:00:33", "remaining_time": "1:42:38", "throughput": 4000.18, "total_tokens": 133056}
+{"current_steps": 210, "total_steps": 38160, "loss": 1.0589, "lr": 2.738469601677149e-06, "epoch": 0.11006289308176101, "percentage": 0.55, "elapsed_time": "0:00:34", "remaining_time": "1:42:59", "throughput": 4002.53, "total_tokens": 136864}
+{"current_steps": 215, "total_steps": 38160, "loss": 1.4862, "lr": 2.8039832285115303e-06, "epoch": 0.11268343815513626, "percentage": 0.56, "elapsed_time": "0:00:34", "remaining_time": "1:42:45", "throughput": 4009.15, "total_tokens": 140064}
+{"current_steps": 220, "total_steps": 38160, "loss": 1.054, "lr": 2.869496855345912e-06, "epoch": 0.11530398322851153, "percentage": 0.58, "elapsed_time": "0:00:35", "remaining_time": "1:42:47", "throughput": 4018.32, "total_tokens": 143712}
+{"current_steps": 225, "total_steps": 38160, "loss": 1.1625, "lr": 2.935010482180294e-06, "epoch": 0.1179245283018868, "percentage": 0.59, "elapsed_time": "0:00:36", "remaining_time": "1:42:48", "throughput": 4014.34, "total_tokens": 146880}
+{"current_steps": 230, "total_steps": 38160, "loss": 1.0283, "lr": 3.0005241090146753e-06, "epoch": 0.12054507337526206, "percentage": 0.6, "elapsed_time": "0:00:37", "remaining_time": "1:42:53", "throughput": 4021.94, "total_tokens": 150560}
+{"current_steps": 235, "total_steps": 38160, "loss": 1.2707, "lr": 3.0660377358490567e-06, "epoch": 0.12316561844863731, "percentage": 0.62, "elapsed_time": "0:00:38", "remaining_time": "1:42:48", "throughput": 4020.53, "total_tokens": 153664}
+{"current_steps": 240, "total_steps": 38160, "loss": 0.9751, "lr": 3.131551362683438e-06, "epoch": 0.12578616352201258, "percentage": 0.63, "elapsed_time": "0:00:39", "remaining_time": "1:42:52", "throughput": 4033.17, "total_tokens": 157568}
+{"current_steps": 245, "total_steps": 38160, "loss": 1.0809, "lr": 3.1970649895178203e-06, "epoch": 0.12840670859538783, "percentage": 0.64, "elapsed_time": "0:00:39", "remaining_time": "1:42:44", "throughput": 4029.23, "total_tokens": 160512}
+{"current_steps": 250, "total_steps": 38160, "loss": 0.9947, "lr": 3.2625786163522012e-06, "epoch": 0.1310272536687631, "percentage": 0.66, "elapsed_time": "0:00:40", "remaining_time": "1:42:59", "throughput": 4037.09, "total_tokens": 164512}
+{"current_steps": 255, "total_steps": 38160, "loss": 1.2133, "lr": 3.3280922431865826e-06, "epoch": 0.13364779874213836, "percentage": 0.67, "elapsed_time": "0:00:41", "remaining_time": "1:42:40", "throughput": 4034.07, "total_tokens": 167200}
+{"current_steps": 260, "total_steps": 38160, "loss": 1.2308, "lr": 3.393605870020965e-06, "epoch": 0.13626834381551362, "percentage": 0.68, "elapsed_time": "0:00:42", "remaining_time": "1:42:25", "throughput": 4032.25, "total_tokens": 169984}
+{"current_steps": 265, "total_steps": 38160, "loss": 1.0836, "lr": 3.459119496855346e-06, "epoch": 0.1388888888888889, "percentage": 0.69, "elapsed_time": "0:00:43", "remaining_time": "1:42:35", "throughput": 4039.15, "total_tokens": 173856}
+{"current_steps": 270, "total_steps": 38160, "loss": 1.4256, "lr": 3.524633123689727e-06, "epoch": 0.14150943396226415, "percentage": 0.71, "elapsed_time": "0:00:43", "remaining_time": "1:42:20", "throughput": 4026.68, "total_tokens": 176192}
+{"current_steps": 275, "total_steps": 38160, "loss": 0.8757, "lr": 3.5901467505241094e-06, "epoch": 0.1441299790356394, "percentage": 0.72, "elapsed_time": "0:00:44", "remaining_time": "1:42:28", "throughput": 4029.25, "total_tokens": 179840}
+{"current_steps": 280, "total_steps": 38160, "loss": 1.0033, "lr": 3.6556603773584908e-06, "epoch": 0.14675052410901468, "percentage": 0.73, "elapsed_time": "0:00:45", "remaining_time": "1:42:23", "throughput": 4029.48, "total_tokens": 182976}
+{"current_steps": 285, "total_steps": 38160, "loss": 1.0193, "lr": 3.7211740041928726e-06, "epoch": 0.14937106918238993, "percentage": 0.75, "elapsed_time": "0:00:46", "remaining_time": "1:42:19", "throughput": 4035.03, "total_tokens": 186400}
+{"current_steps": 290, "total_steps": 38160, "loss": 1.0793, "lr": 3.786687631027254e-06, "epoch": 0.1519916142557652, "percentage": 0.76, "elapsed_time": "0:00:47", "remaining_time": "1:42:24", "throughput": 4040.86, "total_tokens": 190144}
+{"current_steps": 295, "total_steps": 38160, "loss": 1.3219, "lr": 3.852201257861635e-06, "epoch": 0.15461215932914046, "percentage": 0.77, "elapsed_time": "0:00:47", "remaining_time": "1:42:28", "throughput": 4036.26, "total_tokens": 193344}
+{"current_steps": 300, "total_steps": 38160, "loss": 0.9976, "lr": 3.917714884696017e-06, "epoch": 0.15723270440251572, "percentage": 0.79, "elapsed_time": "0:00:48", "remaining_time": "1:42:15", "throughput": 4029.42, "total_tokens": 195904}
+{"current_steps": 305, "total_steps": 38160, "loss": 1.1408, "lr": 3.9832285115303985e-06, "epoch": 0.159853249475891, "percentage": 0.8, "elapsed_time": "0:00:49", "remaining_time": "1:42:08", "throughput": 4028.95, "total_tokens": 198944}
+{"current_steps": 310, "total_steps": 38160, "loss": 1.354, "lr": 4.04874213836478e-06, "epoch": 0.16247379454926625, "percentage": 0.81, "elapsed_time": "0:00:50", "remaining_time": "1:42:04", "throughput": 4023.11, "total_tokens": 201792}
+{"current_steps": 315, "total_steps": 38160, "loss": 0.9747, "lr": 4.114255765199162e-06, "epoch": 0.1650943396226415, "percentage": 0.83, "elapsed_time": "0:00:51", "remaining_time": "1:42:15", "throughput": 4035.32, "total_tokens": 206080}
+{"current_steps": 320, "total_steps": 38160, "loss": 1.1497, "lr": 4.1797693920335435e-06, "epoch": 0.16771488469601678, "percentage": 0.84, "elapsed_time": "0:00:51", "remaining_time": "1:42:09", "throughput": 4039.43, "total_tokens": 209376}
+{"current_steps": 325, "total_steps": 38160, "loss": 1.3004, "lr": 4.245283018867925e-06, "epoch": 0.17033542976939203, "percentage": 0.85, "elapsed_time": "0:00:52", "remaining_time": "1:41:56", "throughput": 4038.2, "total_tokens": 212160}
+{"current_steps": 330, "total_steps": 38160, "loss": 0.9316, "lr": 4.310796645702306e-06, "epoch": 0.17295597484276728, "percentage": 0.86, "elapsed_time": "0:00:53", "remaining_time": "1:42:00", "throughput": 4048.17, "total_tokens": 216128}
+{"current_steps": 335, "total_steps": 38160, "loss": 2.0594, "lr": 4.376310272536688e-06, "epoch": 0.17557651991614256, "percentage": 0.88, "elapsed_time": "0:00:54", "remaining_time": "1:41:59", "throughput": 4044.8, "total_tokens": 219232}
+{"current_steps": 340, "total_steps": 38160, "loss": 1.0318, "lr": 4.441823899371069e-06, "epoch": 0.17819706498951782, "percentage": 0.89, "elapsed_time": "0:00:55", "remaining_time": "1:42:03", "throughput": 4041.06, "total_tokens": 222464}
+{"current_steps": 345, "total_steps": 38160, "loss": 1.4151, "lr": 4.507337526205451e-06, "epoch": 0.18081761006289307, "percentage": 0.9, "elapsed_time": "0:00:55", "remaining_time": "1:41:52", "throughput": 4037.94, "total_tokens": 225184}
+{"current_steps": 350, "total_steps": 38160, "loss": 1.2186, "lr": 4.572851153039833e-06, "epoch": 0.18343815513626835, "percentage": 0.92, "elapsed_time": "0:00:56", "remaining_time": "1:41:43", "throughput": 4035.32, "total_tokens": 228000}
+{"current_steps": 355, "total_steps": 38160, "loss": 1.0458, "lr": 4.638364779874214e-06, "epoch": 0.1860587002096436, "percentage": 0.93, "elapsed_time": "0:00:57", "remaining_time": "1:41:36", "throughput": 4034.08, "total_tokens": 230944}
+{"current_steps": 360, "total_steps": 38160, "loss": 1.0896, "lr": 4.703878406708595e-06, "epoch": 0.18867924528301888, "percentage": 0.94, "elapsed_time": "0:00:57", "remaining_time": "1:41:27", "throughput": 4031.48, "total_tokens": 233728}
+{"current_steps": 365, "total_steps": 38160, "loss": 1.0326, "lr": 4.769392033542977e-06, "epoch": 0.19129979035639413, "percentage": 0.96, "elapsed_time": "0:00:58", "remaining_time": "1:41:18", "throughput": 4028.38, "total_tokens": 236480}
+{"current_steps": 370, "total_steps": 38160, "loss": 1.2213, "lr": 4.834905660377359e-06, "epoch": 0.19392033542976939, "percentage": 0.97, "elapsed_time": "0:00:59", "remaining_time": "1:41:18", "throughput": 4024.01, "total_tokens": 239488}
+{"current_steps": 375, "total_steps": 38160, "loss": 1.2331, "lr": 4.90041928721174e-06, "epoch": 0.19654088050314467, "percentage": 0.98, "elapsed_time": "0:01:00", "remaining_time": "1:41:07", "throughput": 4020.95, "total_tokens": 242144}
+{"current_steps": 380, "total_steps": 38160, "loss": 1.0927, "lr": 4.965932914046122e-06, "epoch": 0.19916142557651992, "percentage": 1.0, "elapsed_time": "0:01:00", "remaining_time": "1:41:00", "throughput": 4022.73, "total_tokens": 245216}
+{"current_steps": 385, "total_steps": 38160, "loss": 1.0174, "lr": 5.031446540880504e-06, "epoch": 0.20178197064989517, "percentage": 1.01, "elapsed_time": "0:01:01", "remaining_time": "1:41:17", "throughput": 4030.12, "total_tokens": 249632}
+{"current_steps": 390, "total_steps": 38160, "loss": 1.1441, "lr": 5.096960167714885e-06, "epoch": 0.20440251572327045, "percentage": 1.02, "elapsed_time": "0:01:02", "remaining_time": "1:41:21", "throughput": 4034.24, "total_tokens": 253312}
+{"current_steps": 395, "total_steps": 38160, "loss": 1.113, "lr": 5.162473794549266e-06, "epoch": 0.2070230607966457, "percentage": 1.04, "elapsed_time": "0:01:03", "remaining_time": "1:41:15", "throughput": 4034.77, "total_tokens": 256384}
+{"current_steps": 400, "total_steps": 38160, "loss": 1.3312, "lr": 5.227987421383648e-06, "epoch": 0.20964360587002095, "percentage": 1.05, "elapsed_time": "0:01:04", "remaining_time": "1:41:26", "throughput": 4028.46, "total_tokens": 259744}
+{"current_steps": 405, "total_steps": 38160, "loss": 1.3715, "lr": 5.2935010482180295e-06, "epoch": 0.21226415094339623, "percentage": 1.06, "elapsed_time": "0:01:05", "remaining_time": "1:41:32", "throughput": 4031.03, "total_tokens": 263456}
+{"current_steps": 410, "total_steps": 38160, "loss": 1.0528, "lr": 5.359014675052411e-06, "epoch": 0.2148846960167715, "percentage": 1.07, "elapsed_time": "0:01:06", "remaining_time": "1:41:30", "throughput": 4028.84, "total_tokens": 266496}
+{"current_steps": 415, "total_steps": 38160, "loss": 1.1782, "lr": 5.424528301886793e-06, "epoch": 0.21750524109014674, "percentage": 1.09, "elapsed_time": "0:01:06", "remaining_time": "1:41:24", "throughput": 4030.34, "total_tokens": 269600}
+{"current_steps": 420, "total_steps": 38160, "loss": 1.2919, "lr": 5.4900419287211744e-06, "epoch": 0.22012578616352202, "percentage": 1.1, "elapsed_time": "0:01:07", "remaining_time": "1:41:27", "throughput": 4032.13, "total_tokens": 273152}
+{"current_steps": 425, "total_steps": 38160, "loss": 1.5176, "lr": 5.555555555555556e-06, "epoch": 0.22274633123689727, "percentage": 1.11, "elapsed_time": "0:01:08", "remaining_time": "1:41:38", "throughput": 4032.99, "total_tokens": 276992}
+{"current_steps": 430, "total_steps": 38160, "loss": 1.4355, "lr": 5.621069182389937e-06, "epoch": 0.22536687631027252, "percentage": 1.13, "elapsed_time": "0:01:09", "remaining_time": "1:41:29", "throughput": 4024.56, "total_tokens": 279296}
+{"current_steps": 435, "total_steps": 38160, "loss": 0.93, "lr": 5.686582809224319e-06, "epoch": 0.2279874213836478, "percentage": 1.14, "elapsed_time": "0:01:10", "remaining_time": "1:41:38", "throughput": 4028.36, "total_tokens": 283296}
+{"current_steps": 440, "total_steps": 38160, "loss": 1.5187, "lr": 5.752096436058701e-06, "epoch": 0.23060796645702306, "percentage": 1.15, "elapsed_time": "0:01:11", "remaining_time": "1:41:31", "throughput": 4025.31, "total_tokens": 286016}
+{"current_steps": 445, "total_steps": 38160, "loss": 1.0657, "lr": 5.817610062893082e-06, "epoch": 0.23322851153039834, "percentage": 1.17, "elapsed_time": "0:01:11", "remaining_time": "1:41:27", "throughput": 4026.76, "total_tokens": 289248}
+{"current_steps": 450, "total_steps": 38160, "loss": 1.4625, "lr": 5.8831236897274636e-06, "epoch": 0.2358490566037736, "percentage": 1.18, "elapsed_time": "0:01:12", "remaining_time": "1:41:25", "throughput": 4022.92, "total_tokens": 292160}
+{"current_steps": 455, "total_steps": 38160, "loss": 1.1244, "lr": 5.948637316561846e-06, "epoch": 0.23846960167714884, "percentage": 1.19, "elapsed_time": "0:01:13", "remaining_time": "1:41:23", "throughput": 4022.33, "total_tokens": 295296}
+{"current_steps": 460, "total_steps": 38160, "loss": 1.0574, "lr": 6.014150943396227e-06, "epoch": 0.24109014675052412, "percentage": 1.21, "elapsed_time": "0:01:14", "remaining_time": "1:41:31", "throughput": 4027.81, "total_tokens": 299360}
+{"current_steps": 465, "total_steps": 38160, "loss": 0.9278, "lr": 6.079664570230608e-06, "epoch": 0.24371069182389937, "percentage": 1.22, "elapsed_time": "0:01:15", "remaining_time": "1:41:34", "throughput": 4031.37, "total_tokens": 303072}
+{"current_steps": 470, "total_steps": 38160, "loss": 0.898, "lr": 6.14517819706499e-06, "epoch": 0.24633123689727462, "percentage": 1.23, "elapsed_time": "0:01:16", "remaining_time": "1:41:47", "throughput": 4039.17, "total_tokens": 307616}
+{"current_steps": 475, "total_steps": 38160, "loss": 1.3183, "lr": 6.210691823899371e-06, "epoch": 0.2489517819706499, "percentage": 1.24, "elapsed_time": "0:01:16", "remaining_time": "1:41:41", "throughput": 4032.46, "total_tokens": 310112}
+{"current_steps": 480, "total_steps": 38160, "loss": 1.0244, "lr": 6.2762054507337535e-06, "epoch": 0.25157232704402516, "percentage": 1.26, "elapsed_time": "0:01:17", "remaining_time": "1:41:41", "throughput": 4033.19, "total_tokens": 313472}
+{"current_steps": 485, "total_steps": 38160, "loss": 1.1941, "lr": 6.341719077568134e-06, "epoch": 0.25419287211740044, "percentage": 1.27, "elapsed_time": "0:01:18", "remaining_time": "1:41:31", "throughput": 4031.05, "total_tokens": 316128}
+{"current_steps": 490, "total_steps": 38160, "loss": 0.9364, "lr": 6.407232704402516e-06, "epoch": 0.25681341719077566, "percentage": 1.28, "elapsed_time": "0:01:19", "remaining_time": "1:41:32", "throughput": 4032.79, "total_tokens": 319584}
+{"current_steps": 495, "total_steps": 38160, "loss": 0.9761, "lr": 6.472746331236898e-06, "epoch": 0.25943396226415094, "percentage": 1.3, "elapsed_time": "0:01:19", "remaining_time": "1:41:25", "throughput": 4032.35, "total_tokens": 322496}
+{"current_steps": 500, "total_steps": 38160, "loss": 1.5193, "lr": 6.538259958071278e-06, "epoch": 0.2620545073375262, "percentage": 1.31, "elapsed_time": "0:01:20", "remaining_time": "1:41:21", "throughput": 4026.54, "total_tokens": 325120}
+{"current_steps": 505, "total_steps": 38160, "loss": 1.255, "lr": 6.60377358490566e-06, "epoch": 0.26467505241090145, "percentage": 1.32, "elapsed_time": "0:01:21", "remaining_time": "1:41:17", "throughput": 4022.15, "total_tokens": 327840}
+{"current_steps": 510, "total_steps": 38160, "loss": 0.7886, "lr": 6.669287211740043e-06, "epoch": 0.2672955974842767, "percentage": 1.34, "elapsed_time": "0:01:22", "remaining_time": "1:41:19", "throughput": 4028.56, "total_tokens": 331776}
+{"current_steps": 515, "total_steps": 38160, "loss": 1.0223, "lr": 6.734800838574423e-06, "epoch": 0.269916142557652, "percentage": 1.35, "elapsed_time": "0:01:23", "remaining_time": "1:41:42", "throughput": 4031.0, "total_tokens": 336544}
+{"current_steps": 520, "total_steps": 38160, "loss": 1.0708, "lr": 6.800314465408805e-06, "epoch": 0.27253668763102723, "percentage": 1.36, "elapsed_time": "0:01:24", "remaining_time": "1:41:50", "throughput": 4036.57, "total_tokens": 340768}
+{"current_steps": 525, "total_steps": 38160, "loss": 1.3615, "lr": 6.865828092243188e-06, "epoch": 0.2751572327044025, "percentage": 1.38, "elapsed_time": "0:01:25", "remaining_time": "1:41:42", "throughput": 4032.46, "total_tokens": 343296}
+{"current_steps": 530, "total_steps": 38160, "loss": 0.9297, "lr": 6.931341719077568e-06, "epoch": 0.2777777777777778, "percentage": 1.39, "elapsed_time": "0:01:25", "remaining_time": "1:41:42", "throughput": 4035.13, "total_tokens": 346848}
+{"current_steps": 535, "total_steps": 38160, "loss": 1.2102, "lr": 6.9968553459119495e-06, "epoch": 0.280398322851153, "percentage": 1.4, "elapsed_time": "0:01:26", "remaining_time": "1:41:38", "throughput": 4031.63, "total_tokens": 349632}
+{"current_steps": 540, "total_steps": 38160, "loss": 0.941, "lr": 7.062368972746332e-06, "epoch": 0.2830188679245283, "percentage": 1.42, "elapsed_time": "0:01:27", "remaining_time": "1:41:39", "throughput": 4031.98, "total_tokens": 353024}
+{"current_steps": 545, "total_steps": 38160, "loss": 1.2362, "lr": 7.127882599580712e-06, "epoch": 0.2856394129979036, "percentage": 1.43, "elapsed_time": "0:01:28", "remaining_time": "1:41:30", "throughput": 4029.03, "total_tokens": 355552}
+{"current_steps": 550, "total_steps": 38160, "loss": 0.8085, "lr": 7.1933962264150945e-06, "epoch": 0.2882599580712788, "percentage": 1.44, "elapsed_time": "0:01:29", "remaining_time": "1:41:40", "throughput": 4036.43, "total_tokens": 360128}
+{"current_steps": 555, "total_steps": 38160, "loss": 0.9297, "lr": 7.258909853249477e-06, "epoch": 0.2908805031446541, "percentage": 1.45, "elapsed_time": "0:01:29", "remaining_time": "1:41:36", "throughput": 4038.23, "total_tokens": 363360}
+{"current_steps": 560, "total_steps": 38160, "loss": 1.1519, "lr": 7.324423480083858e-06, "epoch": 0.29350104821802936, "percentage": 1.47, "elapsed_time": "0:01:30", "remaining_time": "1:41:31", "throughput": 4035.54, "total_tokens": 366144}
+{"current_steps": 565, "total_steps": 38160, "loss": 0.9932, "lr": 7.3899371069182395e-06, "epoch": 0.29612159329140464, "percentage": 1.48, "elapsed_time": "0:01:31", "remaining_time": "1:41:31", "throughput": 4033.4, "total_tokens": 369216}
+{"current_steps": 570, "total_steps": 38160, "loss": 1.008, "lr": 7.455450733752621e-06, "epoch": 0.29874213836477986, "percentage": 1.49, "elapsed_time": "0:01:32", "remaining_time": "1:41:28", "throughput": 4034.74, "total_tokens": 372512}
+{"current_steps": 575, "total_steps": 38160, "loss": 1.2519, "lr": 7.520964360587003e-06, "epoch": 0.30136268343815514, "percentage": 1.51, "elapsed_time": "0:01:33", "remaining_time": "1:41:23", "throughput": 4028.7, "total_tokens": 374976}
+{"current_steps": 580, "total_steps": 38160, "loss": 1.0918, "lr": 7.586477987421384e-06, "epoch": 0.3039832285115304, "percentage": 1.52, "elapsed_time": "0:01:33", "remaining_time": "1:41:18", "throughput": 4025.12, "total_tokens": 377632}
+{"current_steps": 585, "total_steps": 38160, "loss": 1.1725, "lr": 7.651991614255765e-06, "epoch": 0.30660377358490565, "percentage": 1.53, "elapsed_time": "0:01:34", "remaining_time": "1:41:18", "throughput": 4023.57, "total_tokens": 380768}
+{"current_steps": 590, "total_steps": 38160, "loss": 1.0223, "lr": 7.717505241090148e-06, "epoch": 0.30922431865828093, "percentage": 1.55, "elapsed_time": "0:01:35", "remaining_time": "1:41:18", "throughput": 4017.93, "total_tokens": 383520}
+{"current_steps": 595, "total_steps": 38160, "loss": 1.2908, "lr": 7.783018867924528e-06, "epoch": 0.3118448637316562, "percentage": 1.56, "elapsed_time": "0:01:36", "remaining_time": "1:41:15", "throughput": 4016.23, "total_tokens": 386496}
+{"current_steps": 600, "total_steps": 38160, "loss": 0.982, "lr": 7.848532494758911e-06, "epoch": 0.31446540880503143, "percentage": 1.57, "elapsed_time": "0:01:37", "remaining_time": "1:41:20", "throughput": 4019.67, "total_tokens": 390464}
+{"current_steps": 605, "total_steps": 38160, "loss": 0.8656, "lr": 7.914046121593292e-06, "epoch": 0.3170859538784067, "percentage": 1.59, "elapsed_time": "0:01:37", "remaining_time": "1:41:18", "throughput": 4017.66, "total_tokens": 393440}
+{"current_steps": 610, "total_steps": 38160, "loss": 0.8299, "lr": 7.979559748427674e-06, "epoch": 0.319706498951782, "percentage": 1.6, "elapsed_time": "0:01:39", "remaining_time": "1:41:34", "throughput": 4023.36, "total_tokens": 398336}
+{"current_steps": 615, "total_steps": 38160, "loss": 0.9578, "lr": 8.045073375262055e-06, "epoch": 0.3223270440251572, "percentage": 1.61, "elapsed_time": "0:01:39", "remaining_time": "1:41:34", "throughput": 4021.97, "total_tokens": 401504}
+{"current_steps": 620, "total_steps": 38160, "loss": 0.9046, "lr": 8.110587002096436e-06, "epoch": 0.3249475890985325, "percentage": 1.62, "elapsed_time": "0:01:40", "remaining_time": "1:41:35", "throughput": 4021.48, "total_tokens": 404864}
+{"current_steps": 625, "total_steps": 38160, "loss": 0.9816, "lr": 8.176100628930818e-06, "epoch": 0.3275681341719078, "percentage": 1.64, "elapsed_time": "0:01:41", "remaining_time": "1:41:46", "throughput": 4023.14, "total_tokens": 409056}
+{"current_steps": 630, "total_steps": 38160, "loss": 0.974, "lr": 8.241614255765199e-06, "epoch": 0.330188679245283, "percentage": 1.65, "elapsed_time": "0:01:42", "remaining_time": "1:41:43", "throughput": 4026.29, "total_tokens": 412544}
+{"current_steps": 635, "total_steps": 38160, "loss": 0.9609, "lr": 8.307127882599582e-06, "epoch": 0.3328092243186583, "percentage": 1.66, "elapsed_time": "0:01:43", "remaining_time": "1:41:40", "throughput": 4025.25, "total_tokens": 415520}
+{"current_steps": 640, "total_steps": 38160, "loss": 1.6847, "lr": 8.372641509433962e-06, "epoch": 0.33542976939203356, "percentage": 1.68, "elapsed_time": "0:01:43", "remaining_time": "1:41:32", "throughput": 4019.66, "total_tokens": 417760}
+{"current_steps": 645, "total_steps": 38160, "loss": 0.9012, "lr": 8.438155136268345e-06, "epoch": 0.3380503144654088, "percentage": 1.69, "elapsed_time": "0:01:44", "remaining_time": "1:41:36", "throughput": 4023.34, "total_tokens": 421728}
+{"current_steps": 650, "total_steps": 38160, "loss": 1.0753, "lr": 8.503668763102726e-06, "epoch": 0.34067085953878407, "percentage": 1.7, "elapsed_time": "0:01:45", "remaining_time": "1:41:40", "throughput": 4026.49, "total_tokens": 425664}
+{"current_steps": 655, "total_steps": 38160, "loss": 0.8507, "lr": 8.569182389937106e-06, "epoch": 0.34329140461215935, "percentage": 1.72, "elapsed_time": "0:01:46", "remaining_time": "1:41:37", "throughput": 4027.84, "total_tokens": 428896}
+{"current_steps": 660, "total_steps": 38160, "loss": 1.0146, "lr": 8.634696016771489e-06, "epoch": 0.34591194968553457, "percentage": 1.73, "elapsed_time": "0:01:47", "remaining_time": "1:41:31", "throughput": 4026.8, "total_tokens": 431744}
+{"current_steps": 665, "total_steps": 38160, "loss": 0.8793, "lr": 8.70020964360587e-06, "epoch": 0.34853249475890985, "percentage": 1.74, "elapsed_time": "0:01:48", "remaining_time": "1:41:32", "throughput": 4030.65, "total_tokens": 435520}
+{"current_steps": 670, "total_steps": 38160, "loss": 0.8676, "lr": 8.765723270440252e-06, "epoch": 0.35115303983228513, "percentage": 1.76, "elapsed_time": "0:01:48", "remaining_time": "1:41:31", "throughput": 4028.67, "total_tokens": 438592}
+{"current_steps": 675, "total_steps": 38160, "loss": 0.8622, "lr": 8.831236897274633e-06, "epoch": 0.35377358490566035, "percentage": 1.77, "elapsed_time": "0:01:49", "remaining_time": "1:41:28", "throughput": 4028.31, "total_tokens": 441664}
+{"current_steps": 680, "total_steps": 38160, "loss": 1.0794, "lr": 8.896750524109015e-06, "epoch": 0.35639412997903563, "percentage": 1.78, "elapsed_time": "0:01:50", "remaining_time": "1:41:25", "throughput": 4029.0, "total_tokens": 444832}
+{"current_steps": 685, "total_steps": 38160, "loss": 1.2432, "lr": 8.962264150943396e-06, "epoch": 0.3590146750524109, "percentage": 1.8, "elapsed_time": "0:01:51", "remaining_time": "1:41:19", "throughput": 4025.94, "total_tokens": 447392}
+{"current_steps": 690, "total_steps": 38160, "loss": 0.9732, "lr": 9.027777777777777e-06, "epoch": 0.36163522012578614, "percentage": 1.81, "elapsed_time": "0:01:51", "remaining_time": "1:41:18", "throughput": 4025.15, "total_tokens": 450528}
+{"current_steps": 695, "total_steps": 38160, "loss": 0.8436, "lr": 9.09329140461216e-06, "epoch": 0.3642557651991614, "percentage": 1.82, "elapsed_time": "0:01:52", "remaining_time": "1:41:14", "throughput": 4024.74, "total_tokens": 453504}
+{"current_steps": 700, "total_steps": 38160, "loss": 1.8513, "lr": 9.15880503144654e-06, "epoch": 0.3668763102725367, "percentage": 1.83, "elapsed_time": "0:01:53", "remaining_time": "1:41:09", "throughput": 4023.35, "total_tokens": 456288}
+{"current_steps": 705, "total_steps": 38160, "loss": 0.7251, "lr": 9.224318658280923e-06, "epoch": 0.3694968553459119, "percentage": 1.85, "elapsed_time": "0:01:54", "remaining_time": "1:41:08", "throughput": 4025.67, "total_tokens": 459840}
+{"current_steps": 710, "total_steps": 38160, "loss": 0.8733, "lr": 9.289832285115305e-06, "epoch": 0.3721174004192872, "percentage": 1.86, "elapsed_time": "0:01:55", "remaining_time": "1:41:07", "throughput": 4026.35, "total_tokens": 463168}
+{"current_steps": 715, "total_steps": 38160, "loss": 0.9111, "lr": 9.355345911949686e-06, "epoch": 0.3747379454926625, "percentage": 1.87, "elapsed_time": "0:01:55", "remaining_time": "1:41:04", "throughput": 4025.43, "total_tokens": 466112}
+{"current_steps": 720, "total_steps": 38160, "loss": 0.9724, "lr": 9.420859538784067e-06, "epoch": 0.37735849056603776, "percentage": 1.89, "elapsed_time": "0:01:56", "remaining_time": "1:41:02", "throughput": 4027.67, "total_tokens": 469568}
+{"current_steps": 725, "total_steps": 38160, "loss": 0.7596, "lr": 9.486373165618449e-06, "epoch": 0.379979035639413, "percentage": 1.9, "elapsed_time": "0:01:57", "remaining_time": "1:41:02", "throughput": 4027.34, "total_tokens": 472896}
+{"current_steps": 730, "total_steps": 38160, "loss": 0.8202, "lr": 9.551886792452832e-06, "epoch": 0.38259958071278827, "percentage": 1.91, "elapsed_time": "0:01:58", "remaining_time": "1:41:01", "throughput": 4024.01, "total_tokens": 475744}
+{"current_steps": 735, "total_steps": 38160, "loss": 0.8974, "lr": 9.617400419287211e-06, "epoch": 0.38522012578616355, "percentage": 1.93, "elapsed_time": "0:01:59", "remaining_time": "1:41:29", "throughput": 4023.61, "total_tokens": 481216}
+{"current_steps": 740, "total_steps": 38160, "loss": 0.9847, "lr": 9.682914046121595e-06, "epoch": 0.38784067085953877, "percentage": 1.94, "elapsed_time": "0:02:00", "remaining_time": "1:41:24", "throughput": 4018.4, "total_tokens": 483488}
+{"current_steps": 745, "total_steps": 38160, "loss": 0.9643, "lr": 9.748427672955976e-06, "epoch": 0.39046121593291405, "percentage": 1.95, "elapsed_time": "0:02:01", "remaining_time": "1:41:25", "throughput": 4019.65, "total_tokens": 487072}
+{"current_steps": 750, "total_steps": 38160, "loss": 0.8574, "lr": 9.813941299790357e-06, "epoch": 0.39308176100628933, "percentage": 1.97, "elapsed_time": "0:02:01", "remaining_time": "1:41:23", "throughput": 4021.65, "total_tokens": 490496}
+{"current_steps": 755, "total_steps": 38160, "loss": 0.8111, "lr": 9.879454926624739e-06, "epoch": 0.39570230607966456, "percentage": 1.98, "elapsed_time": "0:02:02", "remaining_time": "1:41:21", "throughput": 4023.0, "total_tokens": 493824}
+{"current_steps": 760, "total_steps": 38160, "loss": 1.1684, "lr": 9.94496855345912e-06, "epoch": 0.39832285115303984, "percentage": 1.99, "elapsed_time": "0:02:03", "remaining_time": "1:41:14", "throughput": 4019.89, "total_tokens": 496192}
+{"current_steps": 765, "total_steps": 38160, "loss": 1.0089, "lr": 1.0010482180293501e-05, "epoch": 0.4009433962264151, "percentage": 2.0, "elapsed_time": "0:02:04", "remaining_time": "1:41:10", "throughput": 4017.19, "total_tokens": 498880}
+{"current_steps": 770, "total_steps": 38160, "loss": 0.8748, "lr": 1.0075995807127883e-05, "epoch": 0.40356394129979034, "percentage": 2.02, "elapsed_time": "0:02:04", "remaining_time": "1:41:05", "throughput": 4018.93, "total_tokens": 502016}
+{"current_steps": 775, "total_steps": 38160, "loss": 0.8919, "lr": 1.0141509433962266e-05, "epoch": 0.4061844863731656, "percentage": 2.03, "elapsed_time": "0:02:05", "remaining_time": "1:41:02", "throughput": 4019.47, "total_tokens": 505120}
+{"current_steps": 780, "total_steps": 38160, "loss": 0.8436, "lr": 1.0207023060796646e-05, "epoch": 0.4088050314465409, "percentage": 2.04, "elapsed_time": "0:02:06", "remaining_time": "1:40:57", "throughput": 4021.04, "total_tokens": 508256}
+{"current_steps": 785, "total_steps": 38160, "loss": 0.858, "lr": 1.0272536687631027e-05, "epoch": 0.4114255765199161, "percentage": 2.06, "elapsed_time": "0:02:07", "remaining_time": "1:40:56", "throughput": 4023.26, "total_tokens": 511776}
+{"current_steps": 790, "total_steps": 38160, "loss": 1.0142, "lr": 1.033805031446541e-05, "epoch": 0.4140461215932914, "percentage": 2.07, "elapsed_time": "0:02:07", "remaining_time": "1:40:51", "throughput": 4021.67, "total_tokens": 514464}
+{"current_steps": 795, "total_steps": 38160, "loss": 0.8255, "lr": 1.040356394129979e-05, "epoch": 0.4166666666666667, "percentage": 2.08, "elapsed_time": "0:02:08", "remaining_time": "1:40:51", "throughput": 4024.21, "total_tokens": 518144}
+{"current_steps": 800, "total_steps": 38160, "loss": 1.0494, "lr": 1.0469077568134173e-05, "epoch": 0.4192872117400419, "percentage": 2.1, "elapsed_time": "0:02:09", "remaining_time": "1:40:45", "throughput": 4022.54, "total_tokens": 520768}
+{"current_steps": 805, "total_steps": 38160, "loss": 1.008, "lr": 1.0534591194968554e-05, "epoch": 0.4219077568134172, "percentage": 2.11, "elapsed_time": "0:02:10", "remaining_time": "1:40:42", "throughput": 4022.5, "total_tokens": 523808}
+{"current_steps": 810, "total_steps": 38160, "loss": 0.6437, "lr": 1.0600104821802936e-05, "epoch": 0.42452830188679247, "percentage": 2.12, "elapsed_time": "0:02:10", "remaining_time": "1:40:39", "throughput": 4024.95, "total_tokens": 527136}
+{"current_steps": 815, "total_steps": 38160, "loss": 0.9278, "lr": 1.0665618448637317e-05, "epoch": 0.4271488469601677, "percentage": 2.14, "elapsed_time": "0:02:11", "remaining_time": "1:40:34", "throughput": 4026.06, "total_tokens": 530240}
+{"current_steps": 820, "total_steps": 38160, "loss": 0.8469, "lr": 1.0731132075471698e-05, "epoch": 0.429769392033543, "percentage": 2.15, "elapsed_time": "0:02:12", "remaining_time": "1:40:40", "throughput": 4029.4, "total_tokens": 534496}
+{"current_steps": 825, "total_steps": 38160, "loss": 0.8628, "lr": 1.079664570230608e-05, "epoch": 0.43238993710691825, "percentage": 2.16, "elapsed_time": "0:02:13", "remaining_time": "1:40:47", "throughput": 4030.38, "total_tokens": 538624}
+{"current_steps": 830, "total_steps": 38160, "loss": 0.8884, "lr": 1.0862159329140461e-05, "epoch": 0.4350104821802935, "percentage": 2.18, "elapsed_time": "0:02:14", "remaining_time": "1:40:49", "throughput": 4030.57, "total_tokens": 542112}
+{"current_steps": 835, "total_steps": 38160, "loss": 0.9381, "lr": 1.0927672955974844e-05, "epoch": 0.43763102725366876, "percentage": 2.19, "elapsed_time": "0:02:15", "remaining_time": "1:40:46", "throughput": 4031.96, "total_tokens": 545408}
+{"current_steps": 840, "total_steps": 38160, "loss": 0.7983, "lr": 1.0993186582809224e-05, "epoch": 0.44025157232704404, "percentage": 2.2, "elapsed_time": "0:02:16", "remaining_time": "1:40:42", "throughput": 4032.09, "total_tokens": 548416}
+{"current_steps": 845, "total_steps": 38160, "loss": 0.8325, "lr": 1.1058700209643607e-05, "epoch": 0.44287211740041926, "percentage": 2.21, "elapsed_time": "0:02:16", "remaining_time": "1:40:38", "throughput": 4031.3, "total_tokens": 551264}
+{"current_steps": 850, "total_steps": 38160, "loss": 0.6804, "lr": 1.1124213836477988e-05, "epoch": 0.44549266247379454, "percentage": 2.23, "elapsed_time": "0:02:17", "remaining_time": "1:40:37", "throughput": 4032.33, "total_tokens": 554592}
+{"current_steps": 855, "total_steps": 38160, "loss": 0.6151, "lr": 1.118972746331237e-05, "epoch": 0.4481132075471698, "percentage": 2.24, "elapsed_time": "0:02:18", "remaining_time": "1:40:36", "throughput": 4035.71, "total_tokens": 558304}
+{"current_steps": 860, "total_steps": 38160, "loss": 0.8777, "lr": 1.1255241090146751e-05, "epoch": 0.45073375262054505, "percentage": 2.25, "elapsed_time": "0:02:19", "remaining_time": "1:40:31", "throughput": 4036.15, "total_tokens": 561248}
+{"current_steps": 865, "total_steps": 38160, "loss": 0.6354, "lr": 1.1320754716981132e-05, "epoch": 0.4533542976939203, "percentage": 2.27, "elapsed_time": "0:02:19", "remaining_time": "1:40:26", "throughput": 4037.31, "total_tokens": 564352}
+{"current_steps": 870, "total_steps": 38160, "loss": 0.7525, "lr": 1.1386268343815514e-05, "epoch": 0.4559748427672956, "percentage": 2.28, "elapsed_time": "0:02:20", "remaining_time": "1:40:23", "throughput": 4039.96, "total_tokens": 567712}
+{"current_steps": 875, "total_steps": 38160, "loss": 0.8766, "lr": 1.1451781970649895e-05, "epoch": 0.4585953878406709, "percentage": 2.29, "elapsed_time": "0:02:21", "remaining_time": "1:40:16", "throughput": 4038.45, "total_tokens": 570240}
+{"current_steps": 880, "total_steps": 38160, "loss": 0.9219, "lr": 1.1517295597484278e-05, "epoch": 0.4612159329140461, "percentage": 2.31, "elapsed_time": "0:02:21", "remaining_time": "1:40:11", "throughput": 4036.19, "total_tokens": 572768}
+{"current_steps": 885, "total_steps": 38160, "loss": 0.7997, "lr": 1.158280922431866e-05, "epoch": 0.4638364779874214, "percentage": 2.32, "elapsed_time": "0:02:22", "remaining_time": "1:40:13", "throughput": 4037.92, "total_tokens": 576512}
+{"current_steps": 890, "total_steps": 38160, "loss": 0.8581, "lr": 1.164832285115304e-05, "epoch": 0.46645702306079667, "percentage": 2.33, "elapsed_time": "0:02:23", "remaining_time": "1:40:09", "throughput": 4036.69, "total_tokens": 579264}
+{"current_steps": 895, "total_steps": 38160, "loss": 0.9608, "lr": 1.1713836477987422e-05, "epoch": 0.4690775681341719, "percentage": 2.35, "elapsed_time": "0:02:24", "remaining_time": "1:40:04", "throughput": 4031.95, "total_tokens": 581440}
+{"current_steps": 900, "total_steps": 38160, "loss": 0.5699, "lr": 1.1779350104821804e-05, "epoch": 0.4716981132075472, "percentage": 2.36, "elapsed_time": "0:02:24", "remaining_time": "1:40:01", "throughput": 4032.34, "total_tokens": 584544}
+{"current_steps": 905, "total_steps": 38160, "loss": 0.787, "lr": 1.1844863731656185e-05, "epoch": 0.47431865828092246, "percentage": 2.37, "elapsed_time": "0:02:25", "remaining_time": "1:40:02", "throughput": 4033.77, "total_tokens": 588160}
+{"current_steps": 910, "total_steps": 38160, "loss": 0.8652, "lr": 1.1910377358490566e-05, "epoch": 0.4769392033542977, "percentage": 2.38, "elapsed_time": "0:02:26", "remaining_time": "1:39:58", "throughput": 4028.68, "total_tokens": 590400}
+{"current_steps": 915, "total_steps": 38160, "loss": 0.6771, "lr": 1.197589098532495e-05, "epoch": 0.47955974842767296, "percentage": 2.4, "elapsed_time": "0:02:27", "remaining_time": "1:39:55", "throughput": 4028.93, "total_tokens": 593440}
+{"current_steps": 920, "total_steps": 38160, "loss": 0.7917, "lr": 1.204140461215933e-05, "epoch": 0.48218029350104824, "percentage": 2.41, "elapsed_time": "0:02:28", "remaining_time": "1:39:52", "throughput": 4028.94, "total_tokens": 596480}
+{"current_steps": 925, "total_steps": 38160, "loss": 0.7332, "lr": 1.210691823899371e-05, "epoch": 0.48480083857442346, "percentage": 2.42, "elapsed_time": "0:02:28", "remaining_time": "1:39:48", "throughput": 4028.73, "total_tokens": 599392}
+{"current_steps": 930, "total_steps": 38160, "loss": 0.6215, "lr": 1.2172431865828094e-05, "epoch": 0.48742138364779874, "percentage": 2.44, "elapsed_time": "0:02:29", "remaining_time": "1:39:45", "throughput": 4029.15, "total_tokens": 602432}
+{"current_steps": 935, "total_steps": 38160, "loss": 0.8049, "lr": 1.2237945492662473e-05, "epoch": 0.490041928721174, "percentage": 2.45, "elapsed_time": "0:02:30", "remaining_time": "1:39:40", "throughput": 4024.63, "total_tokens": 604512}
+{"current_steps": 940, "total_steps": 38160, "loss": 0.6622, "lr": 1.2303459119496856e-05, "epoch": 0.49266247379454925, "percentage": 2.46, "elapsed_time": "0:02:31", "remaining_time": "1:39:45", "throughput": 4027.96, "total_tokens": 608864}
+{"current_steps": 945, "total_steps": 38160, "loss": 0.6706, "lr": 1.2368972746331238e-05, "epoch": 0.49528301886792453, "percentage": 2.48, "elapsed_time": "0:02:32", "remaining_time": "1:39:46", "throughput": 4028.97, "total_tokens": 612448}
+{"current_steps": 950, "total_steps": 38160, "loss": 0.9209, "lr": 1.243448637316562e-05, "epoch": 0.4979035639412998, "percentage": 2.49, "elapsed_time": "0:02:32", "remaining_time": "1:39:42", "throughput": 4025.96, "total_tokens": 614944}
+{"current_steps": 955, "total_steps": 38160, "loss": 0.7243, "lr": 1.25e-05, "epoch": 0.500524109014675, "percentage": 2.5, "elapsed_time": "0:02:33", "remaining_time": "1:39:38", "throughput": 4023.91, "total_tokens": 617472}
+{"current_steps": 960, "total_steps": 38160, "loss": 0.668, "lr": 1.2565513626834382e-05, "epoch": 0.5031446540880503, "percentage": 2.52, "elapsed_time": "0:02:34", "remaining_time": "1:39:34", "throughput": 4022.49, "total_tokens": 620192}
+{"current_steps": 965, "total_steps": 38160, "loss": 0.738, "lr": 1.2631027253668765e-05, "epoch": 0.5057651991614256, "percentage": 2.53, "elapsed_time": "0:02:34", "remaining_time": "1:39:30", "throughput": 4023.17, "total_tokens": 623232}
+{"current_steps": 970, "total_steps": 38160, "loss": 0.8766, "lr": 1.2696540880503146e-05, "epoch": 0.5083857442348009, "percentage": 2.54, "elapsed_time": "0:02:35", "remaining_time": "1:39:28", "throughput": 4021.36, "total_tokens": 626016}
+{"current_steps": 975, "total_steps": 38160, "loss": 0.658, "lr": 1.2762054507337526e-05, "epoch": 0.5110062893081762, "percentage": 2.56, "elapsed_time": "0:02:36", "remaining_time": "1:39:28", "throughput": 4022.27, "total_tokens": 629472}
+{"current_steps": 980, "total_steps": 38160, "loss": 0.8124, "lr": 1.2827568134171907e-05, "epoch": 0.5136268343815513, "percentage": 2.57, "elapsed_time": "0:02:37", "remaining_time": "1:39:25", "throughput": 4022.95, "total_tokens": 632608}
+{"current_steps": 985, "total_steps": 38160, "loss": 0.6304, "lr": 1.289308176100629e-05, "epoch": 0.5162473794549266, "percentage": 2.58, "elapsed_time": "0:02:38", "remaining_time": "1:39:26", "throughput": 4024.19, "total_tokens": 636192}
+{"current_steps": 990, "total_steps": 38160, "loss": 0.6376, "lr": 1.2958595387840672e-05, "epoch": 0.5188679245283019, "percentage": 2.59, "elapsed_time": "0:02:39", "remaining_time": "1:39:30", "throughput": 4028.08, "total_tokens": 640544}
+{"current_steps": 995, "total_steps": 38160, "loss": 0.6243, "lr": 1.3024109014675053e-05, "epoch": 0.5214884696016772, "percentage": 2.61, "elapsed_time": "0:02:39", "remaining_time": "1:39:32", "throughput": 4030.16, "total_tokens": 644448}
+{"current_steps": 1000, "total_steps": 38160, "loss": 0.6284, "lr": 1.3089622641509436e-05, "epoch": 0.5241090146750524, "percentage": 2.62, "elapsed_time": "0:02:40", "remaining_time": "1:39:34", "throughput": 4032.0, "total_tokens": 648288}
+{"current_steps": 1005, "total_steps": 38160, "loss": 0.8307, "lr": 1.3155136268343814e-05, "epoch": 0.5267295597484277, "percentage": 2.63, "elapsed_time": "0:02:41", "remaining_time": "1:39:31", "throughput": 4032.01, "total_tokens": 651296}
+{"current_steps": 1010, "total_steps": 38160, "loss": 0.6545, "lr": 1.3220649895178197e-05, "epoch": 0.5293501048218029, "percentage": 2.65, "elapsed_time": "0:02:42", "remaining_time": "1:39:27", "throughput": 4032.22, "total_tokens": 654176}
+{"current_steps": 1015, "total_steps": 38160, "loss": 0.6775, "lr": 1.3286163522012579e-05, "epoch": 0.5319706498951782, "percentage": 2.66, "elapsed_time": "0:02:43", "remaining_time": "1:39:26", "throughput": 4033.81, "total_tokens": 657664}
+{"current_steps": 1020, "total_steps": 38160, "loss": 0.8223, "lr": 1.3351677148846962e-05, "epoch": 0.5345911949685535, "percentage": 2.67, "elapsed_time": "0:02:43", "remaining_time": "1:39:24", "throughput": 4029.76, "total_tokens": 660064}
+{"current_steps": 1025, "total_steps": 38160, "loss": 0.7799, "lr": 1.3417190775681343e-05, "epoch": 0.5372117400419287, "percentage": 2.69, "elapsed_time": "0:02:44", "remaining_time": "1:39:21", "throughput": 4028.14, "total_tokens": 662880}
+{"current_steps": 1030, "total_steps": 38160, "loss": 0.6218, "lr": 1.3482704402515725e-05, "epoch": 0.539832285115304, "percentage": 2.7, "elapsed_time": "0:02:45", "remaining_time": "1:39:20", "throughput": 4027.69, "total_tokens": 665920}
+{"current_steps": 1035, "total_steps": 38160, "loss": 0.6189, "lr": 1.3548218029350104e-05, "epoch": 0.5424528301886793, "percentage": 2.71, "elapsed_time": "0:02:46", "remaining_time": "1:39:18", "throughput": 4026.41, "total_tokens": 668864}
+{"current_steps": 1040, "total_steps": 38160, "loss": 0.5946, "lr": 1.3613731656184486e-05, "epoch": 0.5450733752620545, "percentage": 2.73, "elapsed_time": "0:02:46", "remaining_time": "1:39:14", "throughput": 4026.74, "total_tokens": 671776}
+{"current_steps": 1045, "total_steps": 38160, "loss": 0.6877, "lr": 1.3679245283018869e-05, "epoch": 0.5476939203354297, "percentage": 2.74, "elapsed_time": "0:02:47", "remaining_time": "1:39:12", "throughput": 4027.03, "total_tokens": 674912}
+{"current_steps": 1050, "total_steps": 38160, "loss": 0.5254, "lr": 1.374475890985325e-05, "epoch": 0.550314465408805, "percentage": 2.75, "elapsed_time": "0:02:48", "remaining_time": "1:39:10", "throughput": 4028.94, "total_tokens": 678368}
+{"current_steps": 1055, "total_steps": 38160, "loss": 0.7425, "lr": 1.3810272536687632e-05, "epoch": 0.5529350104821803, "percentage": 2.76, "elapsed_time": "0:02:49", "remaining_time": "1:39:12", "throughput": 4031.45, "total_tokens": 682336}
+{"current_steps": 1060, "total_steps": 38160, "loss": 0.641, "lr": 1.3875786163522015e-05, "epoch": 0.5555555555555556, "percentage": 2.78, "elapsed_time": "0:02:50", "remaining_time": "1:39:12", "throughput": 4031.85, "total_tokens": 685760}
+{"current_steps": 1065, "total_steps": 38160, "loss": 0.5746, "lr": 1.3941299790356394e-05, "epoch": 0.5581761006289309, "percentage": 2.79, "elapsed_time": "0:02:50", "remaining_time": "1:39:11", "throughput": 4029.64, "total_tokens": 688480}
+{"current_steps": 1070, "total_steps": 38160, "loss": 0.79, "lr": 1.4006813417190776e-05, "epoch": 0.560796645702306, "percentage": 2.8, "elapsed_time": "0:02:51", "remaining_time": "1:39:11", "throughput": 4030.66, "total_tokens": 692064}
+{"current_steps": 1075, "total_steps": 38160, "loss": 0.7777, "lr": 1.4072327044025157e-05, "epoch": 0.5634171907756813, "percentage": 2.82, "elapsed_time": "0:02:52", "remaining_time": "1:39:08", "throughput": 4027.17, "total_tokens": 694464}
+{"current_steps": 1080, "total_steps": 38160, "loss": 0.6883, "lr": 1.413784067085954e-05, "epoch": 0.5660377358490566, "percentage": 2.83, "elapsed_time": "0:02:53", "remaining_time": "1:39:05", "throughput": 4027.26, "total_tokens": 697344}
+{"current_steps": 1085, "total_steps": 38160, "loss": 0.6932, "lr": 1.4203354297693922e-05, "epoch": 0.5686582809224319, "percentage": 2.84, "elapsed_time": "0:02:53", "remaining_time": "1:39:02", "throughput": 4026.42, "total_tokens": 700256}
+{"current_steps": 1090, "total_steps": 38160, "loss": 0.6559, "lr": 1.4268867924528303e-05, "epoch": 0.5712788259958071, "percentage": 2.86, "elapsed_time": "0:02:54", "remaining_time": "1:39:03", "throughput": 4028.17, "total_tokens": 703968}
+{"current_steps": 1095, "total_steps": 38160, "loss": 0.6365, "lr": 1.4334381551362683e-05, "epoch": 0.5738993710691824, "percentage": 2.87, "elapsed_time": "0:02:55", "remaining_time": "1:39:07", "throughput": 4030.19, "total_tokens": 708160}
+{"current_steps": 1100, "total_steps": 38160, "loss": 0.873, "lr": 1.4399895178197064e-05, "epoch": 0.5765199161425576, "percentage": 2.88, "elapsed_time": "0:02:56", "remaining_time": "1:39:04", "throughput": 4029.74, "total_tokens": 710976}
+{"current_steps": 1105, "total_steps": 38160, "loss": 0.8462, "lr": 1.4465408805031447e-05, "epoch": 0.5791404612159329, "percentage": 2.9, "elapsed_time": "0:02:57", "remaining_time": "1:39:00", "throughput": 4028.8, "total_tokens": 713696}
+{"current_steps": 1110, "total_steps": 38160, "loss": 0.6461, "lr": 1.4530922431865828e-05, "epoch": 0.5817610062893082, "percentage": 2.91, "elapsed_time": "0:02:57", "remaining_time": "1:38:59", "throughput": 4027.16, "total_tokens": 716608}
+{"current_steps": 1115, "total_steps": 38160, "loss": 0.6218, "lr": 1.4596436058700211e-05, "epoch": 0.5843815513626834, "percentage": 2.92, "elapsed_time": "0:02:58", "remaining_time": "1:38:55", "throughput": 4027.49, "total_tokens": 719488}
+{"current_steps": 1120, "total_steps": 38160, "loss": 0.549, "lr": 1.4661949685534593e-05, "epoch": 0.5870020964360587, "percentage": 2.94, "elapsed_time": "0:02:59", "remaining_time": "1:38:55", "throughput": 4024.63, "total_tokens": 722304}
+{"current_steps": 1125, "total_steps": 38160, "loss": 0.5871, "lr": 1.4727463312368974e-05, "epoch": 0.589622641509434, "percentage": 2.95, "elapsed_time": "0:03:00", "remaining_time": "1:38:53", "throughput": 4024.81, "total_tokens": 725472}
+{"current_steps": 1130, "total_steps": 38160, "loss": 0.4653, "lr": 1.4792976939203354e-05, "epoch": 0.5922431865828093, "percentage": 2.96, "elapsed_time": "0:03:01", "remaining_time": "1:38:54", "throughput": 4025.18, "total_tokens": 728896}
+{"current_steps": 1135, "total_steps": 38160, "loss": 0.5847, "lr": 1.4858490566037735e-05, "epoch": 0.5948637316561844, "percentage": 2.97, "elapsed_time": "0:03:01", "remaining_time": "1:38:52", "throughput": 4024.74, "total_tokens": 732000}
+{"current_steps": 1140, "total_steps": 38160, "loss": 0.6957, "lr": 1.4924004192872118e-05, "epoch": 0.5974842767295597, "percentage": 2.99, "elapsed_time": "0:03:02", "remaining_time": "1:38:52", "throughput": 4025.03, "total_tokens": 735328}
+{"current_steps": 1145, "total_steps": 38160, "loss": 0.5218, "lr": 1.49895178197065e-05, "epoch": 0.600104821802935, "percentage": 3.0, "elapsed_time": "0:03:03", "remaining_time": "1:38:51", "throughput": 4025.38, "total_tokens": 738528}
+{"current_steps": 1150, "total_steps": 38160, "loss": 0.7888, "lr": 1.5055031446540883e-05, "epoch": 0.6027253668763103, "percentage": 3.01, "elapsed_time": "0:03:04", "remaining_time": "1:38:49", "throughput": 4024.09, "total_tokens": 741376}
+{"current_steps": 1155, "total_steps": 38160, "loss": 0.6678, "lr": 1.5120545073375264e-05, "epoch": 0.6053459119496856, "percentage": 3.03, "elapsed_time": "0:03:04", "remaining_time": "1:38:45", "throughput": 4022.26, "total_tokens": 743872}
+{"current_steps": 1160, "total_steps": 38160, "loss": 0.7054, "lr": 1.5186058700209644e-05, "epoch": 0.6079664570230608, "percentage": 3.04, "elapsed_time": "0:03:05", "remaining_time": "1:38:43", "throughput": 4020.82, "total_tokens": 746720}
+{"current_steps": 1165, "total_steps": 38160, "loss": 0.7134, "lr": 1.5251572327044025e-05, "epoch": 0.610587002096436, "percentage": 3.05, "elapsed_time": "0:03:06", "remaining_time": "1:38:43", "throughput": 4022.11, "total_tokens": 750240}
+{"current_steps": 1170, "total_steps": 38160, "loss": 0.6216, "lr": 1.5317085953878407e-05, "epoch": 0.6132075471698113, "percentage": 3.07, "elapsed_time": "0:03:07", "remaining_time": "1:38:39", "throughput": 4023.31, "total_tokens": 753344}
+{"current_steps": 1175, "total_steps": 38160, "loss": 0.7061, "lr": 1.538259958071279e-05, "epoch": 0.6158280922431866, "percentage": 3.08, "elapsed_time": "0:03:07", "remaining_time": "1:38:36", "throughput": 4022.05, "total_tokens": 755968}
+{"current_steps": 1180, "total_steps": 38160, "loss": 0.6997, "lr": 1.5448113207547173e-05, "epoch": 0.6184486373165619, "percentage": 3.09, "elapsed_time": "0:03:08", "remaining_time": "1:38:37", "throughput": 4022.5, "total_tokens": 759520}
+{"current_steps": 1185, "total_steps": 38160, "loss": 0.6185, "lr": 1.5513626834381552e-05, "epoch": 0.6210691823899371, "percentage": 3.11, "elapsed_time": "0:03:09", "remaining_time": "1:38:33", "throughput": 4020.8, "total_tokens": 762048}
+{"current_steps": 1190, "total_steps": 38160, "loss": 0.8012, "lr": 1.5579140461215932e-05, "epoch": 0.6236897274633124, "percentage": 3.12, "elapsed_time": "0:03:10", "remaining_time": "1:38:33", "throughput": 4021.48, "total_tokens": 765504}
+{"current_steps": 1195, "total_steps": 38160, "loss": 0.5988, "lr": 1.5644654088050315e-05, "epoch": 0.6263102725366876, "percentage": 3.13, "elapsed_time": "0:03:11", "remaining_time": "1:38:29", "throughput": 4021.11, "total_tokens": 768192}
+{"current_steps": 1200, "total_steps": 38160, "loss": 0.655, "lr": 1.5710167714884695e-05, "epoch": 0.6289308176100629, "percentage": 3.14, "elapsed_time": "0:03:11", "remaining_time": "1:38:26", "throughput": 4021.05, "total_tokens": 771168}
+{"current_steps": 1205, "total_steps": 38160, "loss": 0.386, "lr": 1.5775681341719078e-05, "epoch": 0.6315513626834381, "percentage": 3.16, "elapsed_time": "0:03:13", "remaining_time": "1:38:47", "throughput": 4022.84, "total_tokens": 777504}
+{"current_steps": 1210, "total_steps": 38160, "loss": 0.6013, "lr": 1.584119496855346e-05, "epoch": 0.6341719077568134, "percentage": 3.17, "elapsed_time": "0:03:14", "remaining_time": "1:38:46", "throughput": 4021.76, "total_tokens": 780512}
+{"current_steps": 1215, "total_steps": 38160, "loss": 0.6492, "lr": 1.5906708595387844e-05, "epoch": 0.6367924528301887, "percentage": 3.18, "elapsed_time": "0:03:14", "remaining_time": "1:38:47", "throughput": 4022.5, "total_tokens": 784192}
+{"current_steps": 1220, "total_steps": 38160, "loss": 0.708, "lr": 1.597222222222222e-05, "epoch": 0.639412997903564, "percentage": 3.2, "elapsed_time": "0:03:15", "remaining_time": "1:38:47", "throughput": 4022.75, "total_tokens": 787552}
+{"current_steps": 1225, "total_steps": 38160, "loss": 0.7912, "lr": 1.6037735849056604e-05, "epoch": 0.6420335429769392, "percentage": 3.21, "elapsed_time": "0:03:16", "remaining_time": "1:38:46", "throughput": 4022.87, "total_tokens": 790720}
+{"current_steps": 1230, "total_steps": 38160, "loss": 0.6971, "lr": 1.6103249475890987e-05, "epoch": 0.6446540880503144, "percentage": 3.22, "elapsed_time": "0:03:17", "remaining_time": "1:38:44", "throughput": 4021.99, "total_tokens": 793696}
+{"current_steps": 1235, "total_steps": 38160, "loss": 0.6986, "lr": 1.6168763102725366e-05, "epoch": 0.6472746331236897, "percentage": 3.24, "elapsed_time": "0:03:18", "remaining_time": "1:38:42", "throughput": 4021.54, "total_tokens": 796640}
+{"current_steps": 1240, "total_steps": 38160, "loss": 0.615, "lr": 1.623427672955975e-05, "epoch": 0.649895178197065, "percentage": 3.25, "elapsed_time": "0:03:18", "remaining_time": "1:38:41", "throughput": 4021.05, "total_tokens": 799648}
+{"current_steps": 1245, "total_steps": 38160, "loss": 0.4334, "lr": 1.6299790356394132e-05, "epoch": 0.6525157232704403, "percentage": 3.26, "elapsed_time": "0:03:19", "remaining_time": "1:38:39", "throughput": 4019.69, "total_tokens": 802496}
+{"current_steps": 1250, "total_steps": 38160, "loss": 0.6735, "lr": 1.6365303983228512e-05, "epoch": 0.6551362683438156, "percentage": 3.28, "elapsed_time": "0:03:20", "remaining_time": "1:38:40", "throughput": 4018.86, "total_tokens": 805760}
+{"current_steps": 1255, "total_steps": 38160, "loss": 0.71, "lr": 1.6430817610062892e-05, "epoch": 0.6577568134171907, "percentage": 3.29, "elapsed_time": "0:03:21", "remaining_time": "1:38:37", "throughput": 4019.49, "total_tokens": 808832}
+{"current_steps": 1260, "total_steps": 38160, "loss": 0.5617, "lr": 1.6496331236897275e-05, "epoch": 0.660377358490566, "percentage": 3.3, "elapsed_time": "0:03:22", "remaining_time": "1:38:36", "throughput": 4020.28, "total_tokens": 812256}
+{"current_steps": 1265, "total_steps": 38160, "loss": 0.5755, "lr": 1.6561844863731658e-05, "epoch": 0.6629979035639413, "percentage": 3.31, "elapsed_time": "0:03:22", "remaining_time": "1:38:36", "throughput": 4020.7, "total_tokens": 815680}
+{"current_steps": 1270, "total_steps": 38160, "loss": 0.5148, "lr": 1.6627358490566038e-05, "epoch": 0.6656184486373166, "percentage": 3.33, "elapsed_time": "0:03:23", "remaining_time": "1:38:35", "throughput": 4018.45, "total_tokens": 818400}
+{"current_steps": 1275, "total_steps": 38160, "loss": 0.6219, "lr": 1.669287211740042e-05, "epoch": 0.6682389937106918, "percentage": 3.34, "elapsed_time": "0:03:24", "remaining_time": "1:38:35", "throughput": 4017.74, "total_tokens": 821536}
+{"current_steps": 1280, "total_steps": 38160, "loss": 0.549, "lr": 1.6758385744234804e-05, "epoch": 0.6708595387840671, "percentage": 3.35, "elapsed_time": "0:03:25", "remaining_time": "1:38:36", "throughput": 4017.79, "total_tokens": 825024}
+{"current_steps": 1285, "total_steps": 38160, "loss": 0.5221, "lr": 1.6823899371069183e-05, "epoch": 0.6734800838574424, "percentage": 3.37, "elapsed_time": "0:03:26", "remaining_time": "1:38:39", "throughput": 4019.84, "total_tokens": 829248}
+{"current_steps": 1290, "total_steps": 38160, "loss": 0.5741, "lr": 1.6889412997903563e-05, "epoch": 0.6761006289308176, "percentage": 3.38, "elapsed_time": "0:03:27", "remaining_time": "1:38:37", "throughput": 4018.24, "total_tokens": 831872}
+{"current_steps": 1295, "total_steps": 38160, "loss": 0.742, "lr": 1.6954926624737946e-05, "epoch": 0.6787211740041929, "percentage": 3.39, "elapsed_time": "0:03:27", "remaining_time": "1:38:35", "throughput": 4017.78, "total_tokens": 834880}
+{"current_steps": 1300, "total_steps": 38160, "loss": 0.7144, "lr": 1.702044025157233e-05, "epoch": 0.6813417190775681, "percentage": 3.41, "elapsed_time": "0:03:28", "remaining_time": "1:38:33", "throughput": 4017.21, "total_tokens": 837824}
+{"current_steps": 1305, "total_steps": 38160, "loss": 0.695, "lr": 1.708595387840671e-05, "epoch": 0.6839622641509434, "percentage": 3.42, "elapsed_time": "0:03:29", "remaining_time": "1:38:32", "throughput": 4016.74, "total_tokens": 840896}
+{"current_steps": 1310, "total_steps": 38160, "loss": 0.7697, "lr": 1.7151467505241092e-05, "epoch": 0.6865828092243187, "percentage": 3.43, "elapsed_time": "0:03:30", "remaining_time": "1:38:31", "throughput": 4016.08, "total_tokens": 844000}
+{"current_steps": 1315, "total_steps": 38160, "loss": 0.5579, "lr": 1.7216981132075472e-05, "epoch": 0.689203354297694, "percentage": 3.45, "elapsed_time": "0:03:31", "remaining_time": "1:38:32", "throughput": 4017.95, "total_tokens": 847840}
+{"current_steps": 1320, "total_steps": 38160, "loss": 0.7351, "lr": 1.7282494758909855e-05, "epoch": 0.6918238993710691, "percentage": 3.46, "elapsed_time": "0:03:31", "remaining_time": "1:38:32", "throughput": 4016.82, "total_tokens": 850976}
+{"current_steps": 1325, "total_steps": 38160, "loss": 0.5582, "lr": 1.7348008385744234e-05, "epoch": 0.6944444444444444, "percentage": 3.47, "elapsed_time": "0:03:32", "remaining_time": "1:38:37", "throughput": 4019.93, "total_tokens": 855616}
+{"current_steps": 1330, "total_steps": 38160, "loss": 0.3995, "lr": 1.7413522012578618e-05, "epoch": 0.6970649895178197, "percentage": 3.49, "elapsed_time": "0:03:33", "remaining_time": "1:38:34", "throughput": 4019.35, "total_tokens": 858528}
+{"current_steps": 1335, "total_steps": 38160, "loss": 0.6065, "lr": 1.7479035639413e-05, "epoch": 0.699685534591195, "percentage": 3.5, "elapsed_time": "0:03:34", "remaining_time": "1:38:35", "throughput": 4020.5, "total_tokens": 862144}
+{"current_steps": 1340, "total_steps": 38160, "loss": 0.6612, "lr": 1.754454926624738e-05, "epoch": 0.7023060796645703, "percentage": 3.51, "elapsed_time": "0:03:35", "remaining_time": "1:38:34", "throughput": 4020.96, "total_tokens": 865440}
+{"current_steps": 1345, "total_steps": 38160, "loss": 0.6392, "lr": 1.761006289308176e-05, "epoch": 0.7049266247379455, "percentage": 3.52, "elapsed_time": "0:03:36", "remaining_time": "1:38:32", "throughput": 4022.64, "total_tokens": 868992}
+{"current_steps": 1350, "total_steps": 38160, "loss": 0.5838, "lr": 1.7675576519916143e-05, "epoch": 0.7075471698113207, "percentage": 3.54, "elapsed_time": "0:03:36", "remaining_time": "1:38:30", "throughput": 4022.55, "total_tokens": 871968}
+{"current_steps": 1355, "total_steps": 38160, "loss": 0.5977, "lr": 1.7741090146750526e-05, "epoch": 0.710167714884696, "percentage": 3.55, "elapsed_time": "0:03:37", "remaining_time": "1:38:27", "throughput": 4022.52, "total_tokens": 874912}
+{"current_steps": 1360, "total_steps": 38160, "loss": 0.6099, "lr": 1.7806603773584906e-05, "epoch": 0.7127882599580713, "percentage": 3.56, "elapsed_time": "0:03:38", "remaining_time": "1:38:27", "throughput": 4022.01, "total_tokens": 878112}
+{"current_steps": 1365, "total_steps": 38160, "loss": 0.5992, "lr": 1.787211740041929e-05, "epoch": 0.7154088050314465, "percentage": 3.58, "elapsed_time": "0:03:39", "remaining_time": "1:38:25", "throughput": 4021.11, "total_tokens": 880928}
+{"current_steps": 1370, "total_steps": 38160, "loss": 0.594, "lr": 1.7937631027253672e-05, "epoch": 0.7180293501048218, "percentage": 3.59, "elapsed_time": "0:03:39", "remaining_time": "1:38:25", "throughput": 4023.55, "total_tokens": 884832}
+{"current_steps": 1375, "total_steps": 38160, "loss": 0.5634, "lr": 1.8003144654088048e-05, "epoch": 0.7206498951781971, "percentage": 3.6, "elapsed_time": "0:03:40", "remaining_time": "1:38:27", "throughput": 4024.8, "total_tokens": 888800}
+{"current_steps": 1380, "total_steps": 38160, "loss": 0.587, "lr": 1.806865828092243e-05, "epoch": 0.7232704402515723, "percentage": 3.62, "elapsed_time": "0:03:41", "remaining_time": "1:38:27", "throughput": 4026.15, "total_tokens": 892416}
+{"current_steps": 1385, "total_steps": 38160, "loss": 0.6955, "lr": 1.8134171907756814e-05, "epoch": 0.7258909853249476, "percentage": 3.63, "elapsed_time": "0:03:42", "remaining_time": "1:38:26", "throughput": 4027.14, "total_tokens": 895840}
+{"current_steps": 1390, "total_steps": 38160, "loss": 0.6141, "lr": 1.8199685534591197e-05, "epoch": 0.7285115303983228, "percentage": 3.64, "elapsed_time": "0:03:43", "remaining_time": "1:38:25", "throughput": 4028.74, "total_tokens": 899392}
+{"current_steps": 1395, "total_steps": 38160, "loss": 0.5578, "lr": 1.8265199161425577e-05, "epoch": 0.7311320754716981, "percentage": 3.66, "elapsed_time": "0:03:44", "remaining_time": "1:38:25", "throughput": 4030.76, "total_tokens": 903264}
+{"current_steps": 1400, "total_steps": 38160, "loss": 0.6471, "lr": 1.833071278825996e-05, "epoch": 0.7337526205450734, "percentage": 3.67, "elapsed_time": "0:03:44", "remaining_time": "1:38:22", "throughput": 4030.55, "total_tokens": 906016}
+{"current_steps": 1405, "total_steps": 38160, "loss": 0.6917, "lr": 1.839622641509434e-05, "epoch": 0.7363731656184487, "percentage": 3.68, "elapsed_time": "0:03:45", "remaining_time": "1:38:23", "throughput": 4028.68, "total_tokens": 909152}
+{"current_steps": 1410, "total_steps": 38160, "loss": 0.6649, "lr": 1.846174004192872e-05, "epoch": 0.7389937106918238, "percentage": 3.69, "elapsed_time": "0:03:46", "remaining_time": "1:38:23", "throughput": 4028.55, "total_tokens": 912416}
+{"current_steps": 1415, "total_steps": 38160, "loss": 0.7838, "lr": 1.8527253668763103e-05, "epoch": 0.7416142557651991, "percentage": 3.71, "elapsed_time": "0:03:47", "remaining_time": "1:38:23", "throughput": 4027.54, "total_tokens": 915616}
+{"current_steps": 1420, "total_steps": 38160, "loss": 0.6336, "lr": 1.8592767295597486e-05, "epoch": 0.7442348008385744, "percentage": 3.72, "elapsed_time": "0:03:48", "remaining_time": "1:38:37", "throughput": 4029.31, "total_tokens": 921568}
+{"current_steps": 1425, "total_steps": 38160, "loss": 0.5353, "lr": 1.865828092243187e-05, "epoch": 0.7468553459119497, "percentage": 3.73, "elapsed_time": "0:03:49", "remaining_time": "1:38:36", "throughput": 4030.03, "total_tokens": 924864}
+{"current_steps": 1430, "total_steps": 38160, "loss": 0.5813, "lr": 1.872379454926625e-05, "epoch": 0.749475890985325, "percentage": 3.75, "elapsed_time": "0:03:50", "remaining_time": "1:38:35", "throughput": 4029.2, "total_tokens": 927968}
+{"current_steps": 1435, "total_steps": 38160, "loss": 0.5945, "lr": 1.878930817610063e-05, "epoch": 0.7520964360587002, "percentage": 3.76, "elapsed_time": "0:03:51", "remaining_time": "1:38:34", "throughput": 4030.69, "total_tokens": 931584}
+{"current_steps": 1440, "total_steps": 38160, "loss": 0.585, "lr": 1.885482180293501e-05, "epoch": 0.7547169811320755, "percentage": 3.77, "elapsed_time": "0:03:51", "remaining_time": "1:38:35", "throughput": 4029.93, "total_tokens": 934816}
+{"current_steps": 1445, "total_steps": 38160, "loss": 0.6382, "lr": 1.892033542976939e-05, "epoch": 0.7573375262054507, "percentage": 3.79, "elapsed_time": "0:03:52", "remaining_time": "1:38:33", "throughput": 4029.8, "total_tokens": 937856}
+{"current_steps": 1450, "total_steps": 38160, "loss": 0.62, "lr": 1.8985849056603774e-05, "epoch": 0.759958071278826, "percentage": 3.8, "elapsed_time": "0:03:54", "remaining_time": "1:38:55", "throughput": 4033.79, "total_tokens": 945632}
+{"current_steps": 1455, "total_steps": 38160, "loss": 0.7001, "lr": 1.9051362683438157e-05, "epoch": 0.7625786163522013, "percentage": 3.81, "elapsed_time": "0:03:55", "remaining_time": "1:38:53", "throughput": 4033.06, "total_tokens": 948608}
+{"current_steps": 1460, "total_steps": 38160, "loss": 0.609, "lr": 1.9116876310272537e-05, "epoch": 0.7651991614255765, "percentage": 3.83, "elapsed_time": "0:03:55", "remaining_time": "1:38:51", "throughput": 4032.46, "total_tokens": 951488}
+{"current_steps": 1465, "total_steps": 38160, "loss": 0.631, "lr": 1.918238993710692e-05, "epoch": 0.7678197064989518, "percentage": 3.84, "elapsed_time": "0:03:56", "remaining_time": "1:38:51", "throughput": 4033.37, "total_tokens": 955072}
+{"current_steps": 1470, "total_steps": 38160, "loss": 0.6494, "lr": 1.92479035639413e-05, "epoch": 0.7704402515723271, "percentage": 3.85, "elapsed_time": "0:03:57", "remaining_time": "1:38:49", "throughput": 4033.6, "total_tokens": 958272}
+{"current_steps": 1475, "total_steps": 38160, "loss": 0.7301, "lr": 1.9313417190775683e-05, "epoch": 0.7730607966457023, "percentage": 3.87, "elapsed_time": "0:03:58", "remaining_time": "1:38:49", "throughput": 4034.53, "total_tokens": 961888}
+{"current_steps": 1480, "total_steps": 38160, "loss": 0.8154, "lr": 1.9378930817610062e-05, "epoch": 0.7756813417190775, "percentage": 3.88, "elapsed_time": "0:03:59", "remaining_time": "1:38:46", "throughput": 4032.28, "total_tokens": 964256}
+{"current_steps": 1485, "total_steps": 38160, "loss": 0.5963, "lr": 1.9444444444444445e-05, "epoch": 0.7783018867924528, "percentage": 3.89, "elapsed_time": "0:04:00", "remaining_time": "1:38:47", "throughput": 4034.13, "total_tokens": 968224}
+{"current_steps": 1490, "total_steps": 38160, "loss": 0.575, "lr": 1.950995807127883e-05, "epoch": 0.7809224318658281, "percentage": 3.9, "elapsed_time": "0:04:00", "remaining_time": "1:38:45", "throughput": 4033.54, "total_tokens": 971136}
+{"current_steps": 1495, "total_steps": 38160, "loss": 0.593, "lr": 1.9575471698113208e-05, "epoch": 0.7835429769392034, "percentage": 3.92, "elapsed_time": "0:04:01", "remaining_time": "1:38:50", "throughput": 4036.14, "total_tokens": 975936}
+{"current_steps": 1500, "total_steps": 38160, "loss": 0.631, "lr": 1.9640985324947588e-05, "epoch": 0.7861635220125787, "percentage": 3.93, "elapsed_time": "0:04:02", "remaining_time": "1:38:49", "throughput": 4036.61, "total_tokens": 979360}
+{"current_steps": 1505, "total_steps": 38160, "loss": 0.5511, "lr": 1.970649895178197e-05, "epoch": 0.7887840670859538, "percentage": 3.94, "elapsed_time": "0:04:03", "remaining_time": "1:38:53", "throughput": 4037.23, "total_tokens": 983616}
+{"current_steps": 1510, "total_steps": 38160, "loss": 0.5986, "lr": 1.9772012578616354e-05, "epoch": 0.7914046121593291, "percentage": 3.96, "elapsed_time": "0:04:04", "remaining_time": "1:38:52", "throughput": 4037.54, "total_tokens": 986944}
+{"current_steps": 1515, "total_steps": 38160, "loss": 0.6223, "lr": 1.9837526205450734e-05, "epoch": 0.7940251572327044, "percentage": 3.97, "elapsed_time": "0:04:05", "remaining_time": "1:38:52", "throughput": 4035.86, "total_tokens": 989888}
+{"current_steps": 1520, "total_steps": 38160, "loss": 0.6239, "lr": 1.9903039832285117e-05, "epoch": 0.7966457023060797, "percentage": 3.98, "elapsed_time": "0:04:06", "remaining_time": "1:38:50", "throughput": 4035.46, "total_tokens": 992768}
+{"current_steps": 1525, "total_steps": 38160, "loss": 0.5101, "lr": 1.99685534591195e-05, "epoch": 0.799266247379455, "percentage": 4.0, "elapsed_time": "0:04:06", "remaining_time": "1:38:53", "throughput": 4036.29, "total_tokens": 996896}
+{"current_steps": 1530, "total_steps": 38160, "loss": 0.6323, "lr": 2.003406708595388e-05, "epoch": 0.8018867924528302, "percentage": 4.01, "elapsed_time": "0:04:07", "remaining_time": "1:38:49", "throughput": 4036.73, "total_tokens": 999840}
+{"current_steps": 1535, "total_steps": 38160, "loss": 0.5301, "lr": 2.009958071278826e-05, "epoch": 0.8045073375262054, "percentage": 4.02, "elapsed_time": "0:04:08", "remaining_time": "1:38:51", "throughput": 4037.42, "total_tokens": 1003680}
+{"current_steps": 1540, "total_steps": 38160, "loss": 0.5482, "lr": 2.0165094339622642e-05, "epoch": 0.8071278825995807, "percentage": 4.04, "elapsed_time": "0:04:09", "remaining_time": "1:38:48", "throughput": 4037.44, "total_tokens": 1006656}
+{"current_steps": 1545, "total_steps": 38160, "loss": 0.5032, "lr": 2.0230607966457025e-05, "epoch": 0.809748427672956, "percentage": 4.05, "elapsed_time": "0:04:10", "remaining_time": "1:38:45", "throughput": 4034.26, "total_tokens": 1008768}
+{"current_steps": 1550, "total_steps": 38160, "loss": 0.5808, "lr": 2.0296121593291405e-05, "epoch": 0.8123689727463312, "percentage": 4.06, "elapsed_time": "0:04:10", "remaining_time": "1:38:44", "throughput": 4034.26, "total_tokens": 1011968}
+{"current_steps": 1555, "total_steps": 38160, "loss": 0.6436, "lr": 2.0361635220125788e-05, "epoch": 0.8149895178197065, "percentage": 4.07, "elapsed_time": "0:04:12", "remaining_time": "1:39:02", "throughput": 4035.51, "total_tokens": 1018784}
+{"current_steps": 1560, "total_steps": 38160, "loss": 0.4018, "lr": 2.0427148846960168e-05, "epoch": 0.8176100628930818, "percentage": 4.09, "elapsed_time": "0:04:13", "remaining_time": "1:39:07", "throughput": 4037.23, "total_tokens": 1023520}
+{"current_steps": 1565, "total_steps": 38160, "loss": 0.5324, "lr": 2.049266247379455e-05, "epoch": 0.820230607966457, "percentage": 4.1, "elapsed_time": "0:04:14", "remaining_time": "1:39:07", "throughput": 4038.71, "total_tokens": 1027200}
+{"current_steps": 1570, "total_steps": 38160, "loss": 0.8955, "lr": 2.055817610062893e-05, "epoch": 0.8228511530398323, "percentage": 4.11, "elapsed_time": "0:04:15", "remaining_time": "1:39:07", "throughput": 4039.22, "total_tokens": 1030848}
+{"current_steps": 1575, "total_steps": 38160, "loss": 0.6222, "lr": 2.0623689727463314e-05, "epoch": 0.8254716981132075, "percentage": 4.13, "elapsed_time": "0:04:16", "remaining_time": "1:39:09", "throughput": 4041.65, "total_tokens": 1035232}
+{"current_steps": 1580, "total_steps": 38160, "loss": 0.5941, "lr": 2.0689203354297697e-05, "epoch": 0.8280922431865828, "percentage": 4.14, "elapsed_time": "0:04:16", "remaining_time": "1:39:06", "throughput": 4042.0, "total_tokens": 1038176}
+{"current_steps": 1585, "total_steps": 38160, "loss": 0.5626, "lr": 2.0754716981132076e-05, "epoch": 0.8307127882599581, "percentage": 4.15, "elapsed_time": "0:04:17", "remaining_time": "1:39:05", "throughput": 4041.33, "total_tokens": 1041216}
+{"current_steps": 1590, "total_steps": 38160, "loss": 0.6039, "lr": 2.0820230607966456e-05, "epoch": 0.8333333333333334, "percentage": 4.17, "elapsed_time": "0:04:18", "remaining_time": "1:39:02", "throughput": 4039.98, "total_tokens": 1043840}
+{"current_steps": 1595, "total_steps": 38160, "loss": 0.4898, "lr": 2.088574423480084e-05, "epoch": 0.8359538784067087, "percentage": 4.18, "elapsed_time": "0:04:19", "remaining_time": "1:38:59", "throughput": 4039.81, "total_tokens": 1046656}
+{"current_steps": 1600, "total_steps": 38160, "loss": 0.5031, "lr": 2.0951257861635222e-05, "epoch": 0.8385744234800838, "percentage": 4.19, "elapsed_time": "0:04:20", "remaining_time": "1:39:05", "throughput": 4041.46, "total_tokens": 1051616}
+{"current_steps": 1605, "total_steps": 38160, "loss": 0.739, "lr": 2.1016771488469602e-05, "epoch": 0.8411949685534591, "percentage": 4.21, "elapsed_time": "0:04:20", "remaining_time": "1:39:02", "throughput": 4038.98, "total_tokens": 1053824}
+{"current_steps": 1610, "total_steps": 38160, "loss": 0.5799, "lr": 2.1082285115303985e-05, "epoch": 0.8438155136268344, "percentage": 4.22, "elapsed_time": "0:04:21", "remaining_time": "1:38:58", "throughput": 4038.64, "total_tokens": 1056512}
+{"current_steps": 1615, "total_steps": 38160, "loss": 0.5544, "lr": 2.1147798742138368e-05, "epoch": 0.8464360587002097, "percentage": 4.23, "elapsed_time": "0:04:22", "remaining_time": "1:38:59", "throughput": 4039.24, "total_tokens": 1060160}
+{"current_steps": 1620, "total_steps": 38160, "loss": 0.738, "lr": 2.1213312368972748e-05, "epoch": 0.8490566037735849, "percentage": 4.25, "elapsed_time": "0:04:23", "remaining_time": "1:38:56", "throughput": 4038.4, "total_tokens": 1062880}
+{"current_steps": 1625, "total_steps": 38160, "loss": 0.6779, "lr": 2.1278825995807127e-05, "epoch": 0.8516771488469602, "percentage": 4.26, "elapsed_time": "0:04:24", "remaining_time": "1:38:56", "throughput": 4039.02, "total_tokens": 1066528}
+{"current_steps": 1630, "total_steps": 38160, "loss": 0.4861, "lr": 2.134433962264151e-05, "epoch": 0.8542976939203354, "percentage": 4.27, "elapsed_time": "0:04:24", "remaining_time": "1:38:56", "throughput": 4037.69, "total_tokens": 1069472}
+{"current_steps": 1635, "total_steps": 38160, "loss": 0.5739, "lr": 2.1409853249475894e-05, "epoch": 0.8569182389937107, "percentage": 4.28, "elapsed_time": "0:04:25", "remaining_time": "1:39:00", "throughput": 4039.63, "total_tokens": 1074304}
+{"current_steps": 1640, "total_steps": 38160, "loss": 0.539, "lr": 2.1475366876310273e-05, "epoch": 0.859538784067086, "percentage": 4.3, "elapsed_time": "0:04:26", "remaining_time": "1:38:59", "throughput": 4039.14, "total_tokens": 1077376}
+{"current_steps": 1645, "total_steps": 38160, "loss": 0.5773, "lr": 2.1540880503144656e-05, "epoch": 0.8621593291404612, "percentage": 4.31, "elapsed_time": "0:04:27", "remaining_time": "1:38:58", "throughput": 4039.73, "total_tokens": 1080672}
+{"current_steps": 1650, "total_steps": 38160, "loss": 0.6302, "lr": 2.160639412997904e-05, "epoch": 0.8647798742138365, "percentage": 4.32, "elapsed_time": "0:04:28", "remaining_time": "1:38:58", "throughput": 4040.16, "total_tokens": 1084224}
+{"current_steps": 1655, "total_steps": 38160, "loss": 0.7895, "lr": 2.1671907756813416e-05, "epoch": 0.8674004192872118, "percentage": 4.34, "elapsed_time": "0:04:29", "remaining_time": "1:38:55", "throughput": 4039.05, "total_tokens": 1086912}
+{"current_steps": 1660, "total_steps": 38160, "loss": 0.681, "lr": 2.17374213836478e-05, "epoch": 0.870020964360587, "percentage": 4.35, "elapsed_time": "0:04:29", "remaining_time": "1:38:54", "throughput": 4037.76, "total_tokens": 1089696}
+{"current_steps": 1665, "total_steps": 38160, "loss": 0.5289, "lr": 2.1802935010482182e-05, "epoch": 0.8726415094339622, "percentage": 4.36, "elapsed_time": "0:04:30", "remaining_time": "1:38:51", "throughput": 4037.53, "total_tokens": 1092608}
+{"current_steps": 1670, "total_steps": 38160, "loss": 0.7744, "lr": 2.186844863731656e-05, "epoch": 0.8752620545073375, "percentage": 4.38, "elapsed_time": "0:04:31", "remaining_time": "1:38:49", "throughput": 4038.01, "total_tokens": 1095808}
+{"current_steps": 1675, "total_steps": 38160, "loss": 0.6631, "lr": 2.1933962264150945e-05, "epoch": 0.8778825995807128, "percentage": 4.39, "elapsed_time": "0:04:32", "remaining_time": "1:38:47", "throughput": 4037.81, "total_tokens": 1098816}
+{"current_steps": 1680, "total_steps": 38160, "loss": 0.6448, "lr": 2.1999475890985328e-05, "epoch": 0.8805031446540881, "percentage": 4.4, "elapsed_time": "0:04:33", "remaining_time": "1:38:49", "throughput": 4039.69, "total_tokens": 1103136}
+{"current_steps": 1685, "total_steps": 38160, "loss": 0.5341, "lr": 2.2064989517819707e-05, "epoch": 0.8831236897274634, "percentage": 4.42, "elapsed_time": "0:04:34", "remaining_time": "1:38:52", "throughput": 4042.3, "total_tokens": 1107872}
+{"current_steps": 1690, "total_steps": 38160, "loss": 0.6872, "lr": 2.2130503144654087e-05, "epoch": 0.8857442348008385, "percentage": 4.43, "elapsed_time": "0:04:34", "remaining_time": "1:38:50", "throughput": 4042.65, "total_tokens": 1111072}
+{"current_steps": 1695, "total_steps": 38160, "loss": 0.6674, "lr": 2.219601677148847e-05, "epoch": 0.8883647798742138, "percentage": 4.44, "elapsed_time": "0:04:35", "remaining_time": "1:38:48", "throughput": 4041.98, "total_tokens": 1113792}
+{"current_steps": 1700, "total_steps": 38160, "loss": 0.6124, "lr": 2.2261530398322853e-05, "epoch": 0.8909853249475891, "percentage": 4.45, "elapsed_time": "0:04:36", "remaining_time": "1:38:45", "throughput": 4040.79, "total_tokens": 1116384}
+{"current_steps": 1705, "total_steps": 38160, "loss": 0.5576, "lr": 2.2327044025157233e-05, "epoch": 0.8936058700209644, "percentage": 4.47, "elapsed_time": "0:04:37", "remaining_time": "1:38:43", "throughput": 4040.34, "total_tokens": 1119296}
+{"current_steps": 1710, "total_steps": 38160, "loss": 0.6771, "lr": 2.2392557651991616e-05, "epoch": 0.8962264150943396, "percentage": 4.48, "elapsed_time": "0:04:37", "remaining_time": "1:38:41", "throughput": 4041.74, "total_tokens": 1122720}
+{"current_steps": 1715, "total_steps": 38160, "loss": 0.7852, "lr": 2.2458071278825996e-05, "epoch": 0.8988469601677149, "percentage": 4.49, "elapsed_time": "0:04:38", "remaining_time": "1:38:38", "throughput": 4039.78, "total_tokens": 1125088}
+{"current_steps": 1720, "total_steps": 38160, "loss": 0.7204, "lr": 2.252358490566038e-05, "epoch": 0.9014675052410901, "percentage": 4.51, "elapsed_time": "0:04:39", "remaining_time": "1:38:35", "throughput": 4040.66, "total_tokens": 1128288}
+{"current_steps": 1725, "total_steps": 38160, "loss": 0.7069, "lr": 2.258909853249476e-05, "epoch": 0.9040880503144654, "percentage": 4.52, "elapsed_time": "0:04:39", "remaining_time": "1:38:32", "throughput": 4040.53, "total_tokens": 1131040}
+{"current_steps": 1730, "total_steps": 38160, "loss": 0.5968, "lr": 2.265461215932914e-05, "epoch": 0.9067085953878407, "percentage": 4.53, "elapsed_time": "0:04:40", "remaining_time": "1:38:34", "throughput": 4041.68, "total_tokens": 1135136}
+{"current_steps": 1735, "total_steps": 38160, "loss": 0.7413, "lr": 2.2720125786163524e-05, "epoch": 0.9093291404612159, "percentage": 4.55, "elapsed_time": "0:04:41", "remaining_time": "1:38:35", "throughput": 4042.61, "total_tokens": 1139008}
+{"current_steps": 1740, "total_steps": 38160, "loss": 0.5417, "lr": 2.2785639412997904e-05, "epoch": 0.9119496855345912, "percentage": 4.56, "elapsed_time": "0:04:42", "remaining_time": "1:38:32", "throughput": 4042.67, "total_tokens": 1141920}
+{"current_steps": 1745, "total_steps": 38160, "loss": 0.6172, "lr": 2.2851153039832284e-05, "epoch": 0.9145702306079665, "percentage": 4.57, "elapsed_time": "0:04:43", "remaining_time": "1:38:30", "throughput": 4042.66, "total_tokens": 1144960}
+{"current_steps": 1750, "total_steps": 38160, "loss": 0.6513, "lr": 2.2916666666666667e-05, "epoch": 0.9171907756813418, "percentage": 4.59, "elapsed_time": "0:04:43", "remaining_time": "1:38:27", "throughput": 4042.24, "total_tokens": 1147712}
+{"current_steps": 1755, "total_steps": 38160, "loss": 0.6143, "lr": 2.298218029350105e-05, "epoch": 0.9198113207547169, "percentage": 4.6, "elapsed_time": "0:04:44", "remaining_time": "1:38:24", "throughput": 4041.85, "total_tokens": 1150528}
+{"current_steps": 1760, "total_steps": 38160, "loss": 0.6818, "lr": 2.304769392033543e-05, "epoch": 0.9224318658280922, "percentage": 4.61, "elapsed_time": "0:04:45", "remaining_time": "1:38:23", "throughput": 4043.47, "total_tokens": 1154272}
+{"current_steps": 1765, "total_steps": 38160, "loss": 0.596, "lr": 2.3113207547169813e-05, "epoch": 0.9250524109014675, "percentage": 4.63, "elapsed_time": "0:04:46", "remaining_time": "1:38:21", "throughput": 4043.23, "total_tokens": 1157088}
+{"current_steps": 1770, "total_steps": 38160, "loss": 0.7229, "lr": 2.3178721174004196e-05, "epoch": 0.9276729559748428, "percentage": 4.64, "elapsed_time": "0:04:46", "remaining_time": "1:38:18", "throughput": 4043.75, "total_tokens": 1160192}
+{"current_steps": 1775, "total_steps": 38160, "loss": 0.6043, "lr": 2.3244234800838576e-05, "epoch": 0.9302935010482181, "percentage": 4.65, "elapsed_time": "0:04:47", "remaining_time": "1:38:17", "throughput": 4044.17, "total_tokens": 1163424}
+{"current_steps": 1780, "total_steps": 38160, "loss": 0.653, "lr": 2.3309748427672955e-05, "epoch": 0.9329140461215933, "percentage": 4.66, "elapsed_time": "0:04:48", "remaining_time": "1:38:15", "throughput": 4045.78, "total_tokens": 1167104}
+{"current_steps": 1785, "total_steps": 38160, "loss": 0.604, "lr": 2.3375262054507338e-05, "epoch": 0.9355345911949685, "percentage": 4.68, "elapsed_time": "0:04:49", "remaining_time": "1:38:15", "throughput": 4045.19, "total_tokens": 1170368}
+{"current_steps": 1790, "total_steps": 38160, "loss": 0.7018, "lr": 2.344077568134172e-05, "epoch": 0.9381551362683438, "percentage": 4.69, "elapsed_time": "0:04:50", "remaining_time": "1:38:15", "throughput": 4045.96, "total_tokens": 1173888}
+{"current_steps": 1795, "total_steps": 38160, "loss": 0.542, "lr": 2.35062893081761e-05, "epoch": 0.9407756813417191, "percentage": 4.7, "elapsed_time": "0:04:50", "remaining_time": "1:38:14", "throughput": 4046.62, "total_tokens": 1177408}
+{"current_steps": 1800, "total_steps": 38160, "loss": 0.5519, "lr": 2.3571802935010484e-05, "epoch": 0.9433962264150944, "percentage": 4.72, "elapsed_time": "0:04:51", "remaining_time": "1:38:14", "throughput": 4047.06, "total_tokens": 1181056}
+{"current_steps": 1805, "total_steps": 38160, "loss": 0.7689, "lr": 2.3637316561844867e-05, "epoch": 0.9460167714884696, "percentage": 4.73, "elapsed_time": "0:04:52", "remaining_time": "1:38:14", "throughput": 4046.67, "total_tokens": 1184256}
+{"current_steps": 1810, "total_steps": 38160, "loss": 0.5198, "lr": 2.3702830188679247e-05, "epoch": 0.9486373165618449, "percentage": 4.74, "elapsed_time": "0:04:53", "remaining_time": "1:38:12", "throughput": 4046.41, "total_tokens": 1187168}
+{"current_steps": 1815, "total_steps": 38160, "loss": 0.5154, "lr": 2.3768343815513627e-05, "epoch": 0.9512578616352201, "percentage": 4.76, "elapsed_time": "0:04:54", "remaining_time": "1:38:10", "throughput": 4047.78, "total_tokens": 1190784}
+{"current_steps": 1820, "total_steps": 38160, "loss": 0.6839, "lr": 2.383385744234801e-05, "epoch": 0.9538784067085954, "percentage": 4.77, "elapsed_time": "0:04:54", "remaining_time": "1:38:08", "throughput": 4046.58, "total_tokens": 1193312}
+{"current_steps": 1825, "total_steps": 38160, "loss": 0.6059, "lr": 2.3899371069182393e-05, "epoch": 0.9564989517819706, "percentage": 4.78, "elapsed_time": "0:04:55", "remaining_time": "1:38:06", "throughput": 4047.32, "total_tokens": 1196608}
+{"current_steps": 1830, "total_steps": 38160, "loss": 0.6567, "lr": 2.3964884696016772e-05, "epoch": 0.9591194968553459, "percentage": 4.8, "elapsed_time": "0:04:56", "remaining_time": "1:38:05", "throughput": 4048.52, "total_tokens": 1200160}
+{"current_steps": 1835, "total_steps": 38160, "loss": 0.5262, "lr": 2.4030398322851155e-05, "epoch": 0.9617400419287212, "percentage": 4.81, "elapsed_time": "0:04:57", "remaining_time": "1:38:02", "throughput": 4048.65, "total_tokens": 1203040}
+{"current_steps": 1840, "total_steps": 38160, "loss": 0.4398, "lr": 2.4095911949685535e-05, "epoch": 0.9643605870020965, "percentage": 4.82, "elapsed_time": "0:04:57", "remaining_time": "1:37:59", "throughput": 4047.2, "total_tokens": 1205408}
+{"current_steps": 1845, "total_steps": 38160, "loss": 0.4673, "lr": 2.4161425576519918e-05, "epoch": 0.9669811320754716, "percentage": 4.83, "elapsed_time": "0:04:58", "remaining_time": "1:37:58", "throughput": 4047.42, "total_tokens": 1208896}
+{"current_steps": 1850, "total_steps": 38160, "loss": 0.5623, "lr": 2.4226939203354298e-05, "epoch": 0.9696016771488469, "percentage": 4.85, "elapsed_time": "0:04:59", "remaining_time": "1:37:58", "throughput": 4048.79, "total_tokens": 1212576}
+{"current_steps": 1855, "total_steps": 38160, "loss": 0.5551, "lr": 2.429245283018868e-05, "epoch": 0.9722222222222222, "percentage": 4.86, "elapsed_time": "0:05:00", "remaining_time": "1:37:56", "throughput": 4049.22, "total_tokens": 1215872}
+{"current_steps": 1860, "total_steps": 38160, "loss": 0.5463, "lr": 2.4357966457023064e-05, "epoch": 0.9748427672955975, "percentage": 4.87, "elapsed_time": "0:05:01", "remaining_time": "1:37:57", "throughput": 4050.91, "total_tokens": 1219968}
+{"current_steps": 1865, "total_steps": 38160, "loss": 0.6082, "lr": 2.4423480083857444e-05, "epoch": 0.9774633123689728, "percentage": 4.89, "elapsed_time": "0:05:01", "remaining_time": "1:37:54", "throughput": 4048.74, "total_tokens": 1222048}
+{"current_steps": 1870, "total_steps": 38160, "loss": 0.6314, "lr": 2.4488993710691823e-05, "epoch": 0.980083857442348, "percentage": 4.9, "elapsed_time": "0:05:02", "remaining_time": "1:37:51", "throughput": 4049.0, "total_tokens": 1225120}
+{"current_steps": 1875, "total_steps": 38160, "loss": 0.5823, "lr": 2.4554507337526206e-05, "epoch": 0.9827044025157232, "percentage": 4.91, "elapsed_time": "0:05:03", "remaining_time": "1:37:52", "throughput": 4050.42, "total_tokens": 1229056}
+{"current_steps": 1880, "total_steps": 38160, "loss": 0.4814, "lr": 2.4620020964360586e-05, "epoch": 0.9853249475890985, "percentage": 4.93, "elapsed_time": "0:05:04", "remaining_time": "1:37:49", "throughput": 4049.82, "total_tokens": 1231776}
+{"current_steps": 1885, "total_steps": 38160, "loss": 0.5674, "lr": 2.468553459119497e-05, "epoch": 0.9879454926624738, "percentage": 4.94, "elapsed_time": "0:05:04", "remaining_time": "1:37:48", "throughput": 4050.1, "total_tokens": 1235040}
+{"current_steps": 1890, "total_steps": 38160, "loss": 0.5268, "lr": 2.4751048218029352e-05, "epoch": 0.9905660377358491, "percentage": 4.95, "elapsed_time": "0:05:05", "remaining_time": "1:37:47", "throughput": 4049.33, "total_tokens": 1238048}
+{"current_steps": 1895, "total_steps": 38160, "loss": 0.6059, "lr": 2.4816561844863735e-05, "epoch": 0.9931865828092243, "percentage": 4.97, "elapsed_time": "0:05:06", "remaining_time": "1:37:45", "throughput": 4050.2, "total_tokens": 1241344}
+{"current_steps": 1900, "total_steps": 38160, "loss": 0.4554, "lr": 2.488207547169811e-05, "epoch": 0.9958071278825996, "percentage": 4.98, "elapsed_time": "0:05:07", "remaining_time": "1:37:43", "throughput": 4048.71, "total_tokens": 1243840}
+{"current_steps": 1905, "total_steps": 38160, "loss": 0.6523, "lr": 2.4947589098532495e-05, "epoch": 0.9984276729559748, "percentage": 4.99, "elapsed_time": "0:05:07", "remaining_time": "1:37:40", "throughput": 4048.78, "total_tokens": 1246816}
+{"current_steps": 1908, "total_steps": 38160, "eval_loss": 0.5840584635734558, "epoch": 1.0, "percentage": 5.0, "elapsed_time": "0:05:18", "remaining_time": "1:40:47", "throughput": 3922.19, "total_tokens": 1248304}
+{"current_steps": 1910, "total_steps": 38160, "loss": 0.5836, "lr": 2.5013102725366878e-05, "epoch": 1.00104821802935, "percentage": 5.01, "elapsed_time": "0:05:19", "remaining_time": "1:41:11", "throughput": 3904.75, "total_tokens": 1249200}
+{"current_steps": 1915, "total_steps": 38160, "loss": 0.6168, "lr": 2.507861635220126e-05, "epoch": 1.0036687631027255, "percentage": 5.02, "elapsed_time": "0:05:20", "remaining_time": "1:41:10", "throughput": 3902.55, "total_tokens": 1251664}
+{"current_steps": 1920, "total_steps": 38160, "loss": 0.502, "lr": 2.514412997903564e-05, "epoch": 1.0062893081761006, "percentage": 5.03, "elapsed_time": "0:05:21", "remaining_time": "1:41:09", "throughput": 3902.59, "total_tokens": 1254928}
+{"current_steps": 1925, "total_steps": 38160, "loss": 0.474, "lr": 2.520964360587002e-05, "epoch": 1.0089098532494758, "percentage": 5.04, "elapsed_time": "0:05:22", "remaining_time": "1:41:07", "throughput": 3901.96, "total_tokens": 1257744}
+{"current_steps": 1930, "total_steps": 38160, "loss": 0.5192, "lr": 2.5275157232704407e-05, "epoch": 1.0115303983228512, "percentage": 5.06, "elapsed_time": "0:05:23", "remaining_time": "1:41:07", "throughput": 3903.95, "total_tokens": 1261904}
+{"current_steps": 1935, "total_steps": 38160, "loss": 0.4531, "lr": 2.5340670859538783e-05, "epoch": 1.0141509433962264, "percentage": 5.07, "elapsed_time": "0:05:23", "remaining_time": "1:41:04", "throughput": 3902.96, "total_tokens": 1264432}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11d79b32e5608200dd1b720b6eb94e1f89e8db6cf7027330292b5c4fa1278214
+size 6353