Training in progress, step 1813

.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 +1,35 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "k_proj",
+    "v_proj",
+    "down_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be941a3544503396cf280290ed78c73b84594786e84b1e6c25d2c9c64958a0d5
+size 58745928

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 +1,58 @@
+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: lora
+lora_rank: 16
+lora_alpha: 16
+lora_dropout: 0.05
+lora_target: k_proj,v_proj,down_proj
+
+### dataset
+dataset: piqa
+template: llama3
+cutoff_len: 2048
+overwrite_cache: true
+preprocessing_num_workers: 16
+dataloader_num_workers: 4
+packing: false
+
+### output
+output_dir: saves/lora/llama-3-8b-instruct/train_piqa_1754507484
+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: 10.0
+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
+
+### eval
+per_device_eval_batch_size: 4
+eval_strategy: steps
+eval_steps: 0.05
+val_size: 0.1

trainer_log.jsonl

@@ -0,0 +1,368 @@
+{"current_steps": 5, "total_steps": 36260, "loss": 1.0028, "lr": 5.5157198014340876e-08, "epoch": 0.0013789299503585218, "percentage": 0.01, "elapsed_time": "0:00:01", "remaining_time": "3:27:49", "throughput": 1656.09, "total_tokens": 2848}
+{"current_steps": 10, "total_steps": 36260, "loss": 0.9575, "lr": 1.2410369553226697e-07, "epoch": 0.0027578599007170436, "percentage": 0.03, "elapsed_time": "0:00:02", "remaining_time": "2:57:29", "throughput": 1960.39, "total_tokens": 5760}
+{"current_steps": 15, "total_steps": 36260, "loss": 1.0904, "lr": 1.9305019305019306e-07, "epoch": 0.004136789851075565, "percentage": 0.04, "elapsed_time": "0:00:03", "remaining_time": "2:40:32", "throughput": 2014.9, "total_tokens": 8032}
+{"current_steps": 20, "total_steps": 36260, "loss": 0.9729, "lr": 2.619966905681192e-07, "epoch": 0.005515719801434087, "percentage": 0.06, "elapsed_time": "0:00:05", "remaining_time": "2:44:18", "throughput": 2152.59, "total_tokens": 11712}
+{"current_steps": 25, "total_steps": 36260, "loss": 0.985, "lr": 3.309431880860452e-07, "epoch": 0.006894649751792609, "percentage": 0.07, "elapsed_time": "0:00:06", "remaining_time": "2:36:32", "throughput": 2177.57, "total_tokens": 14112}
+{"current_steps": 30, "total_steps": 36260, "loss": 1.0233, "lr": 3.998896856039713e-07, "epoch": 0.00827357970215113, "percentage": 0.08, "elapsed_time": "0:00:07", "remaining_time": "2:40:56", "throughput": 2253.08, "total_tokens": 18016}
+{"current_steps": 35, "total_steps": 36260, "loss": 1.0284, "lr": 4.6883618312189744e-07, "epoch": 0.009652509652509652, "percentage": 0.1, "elapsed_time": "0:00:09", "remaining_time": "2:41:41", "throughput": 2317.96, "total_tokens": 21728}
+{"current_steps": 40, "total_steps": 36260, "loss": 0.9388, "lr": 5.377826806398235e-07, "epoch": 0.011031439602868174, "percentage": 0.11, "elapsed_time": "0:00:10", "remaining_time": "2:38:09", "throughput": 2311.49, "total_tokens": 24224}
+{"current_steps": 45, "total_steps": 36260, "loss": 1.0727, "lr": 6.067291781577496e-07, "epoch": 0.012410369553226696, "percentage": 0.12, "elapsed_time": "0:00:11", "remaining_time": "2:35:30", "throughput": 2318.34, "total_tokens": 26880}
+{"current_steps": 50, "total_steps": 36260, "loss": 0.8292, "lr": 6.756756756756758e-07, "epoch": 0.013789299503585218, "percentage": 0.14, "elapsed_time": "0:00:12", "remaining_time": "2:32:07", "throughput": 2305.36, "total_tokens": 29056}
+{"current_steps": 55, "total_steps": 36260, "loss": 1.0232, "lr": 7.446221731936018e-07, "epoch": 0.01516822945394374, "percentage": 0.15, "elapsed_time": "0:00:13", "remaining_time": "2:29:08", "throughput": 2297.46, "total_tokens": 31232}
+{"current_steps": 60, "total_steps": 36260, "loss": 0.6936, "lr": 8.13568670711528e-07, "epoch": 0.01654715940430226, "percentage": 0.17, "elapsed_time": "0:00:14", "remaining_time": "2:28:10", "throughput": 2312.76, "total_tokens": 34080}
+{"current_steps": 65, "total_steps": 36260, "loss": 0.8574, "lr": 8.82515168229454e-07, "epoch": 0.017926089354660783, "percentage": 0.18, "elapsed_time": "0:00:16", "remaining_time": "2:28:39", "throughput": 2331.49, "total_tokens": 37344}
+{"current_steps": 70, "total_steps": 36260, "loss": 0.709, "lr": 9.514616657473801e-07, "epoch": 0.019305019305019305, "percentage": 0.19, "elapsed_time": "0:00:17", "remaining_time": "2:26:51", "throughput": 2330.15, "total_tokens": 39712}
+{"current_steps": 75, "total_steps": 36260, "loss": 1.0253, "lr": 1.020408163265306e-06, "epoch": 0.020683949255377827, "percentage": 0.21, "elapsed_time": "0:00:18", "remaining_time": "2:26:04", "throughput": 2334.03, "total_tokens": 42400}
+{"current_steps": 80, "total_steps": 36260, "loss": 0.8172, "lr": 1.0893546607832323e-06, "epoch": 0.02206287920573635, "percentage": 0.22, "elapsed_time": "0:00:19", "remaining_time": "2:26:01", "throughput": 2340.63, "total_tokens": 45344}
+{"current_steps": 85, "total_steps": 36260, "loss": 1.0014, "lr": 1.1583011583011583e-06, "epoch": 0.02344180915609487, "percentage": 0.23, "elapsed_time": "0:00:20", "remaining_time": "2:25:20", "throughput": 2334.84, "total_tokens": 47840}
+{"current_steps": 90, "total_steps": 36260, "loss": 0.4216, "lr": 1.2272476558190843e-06, "epoch": 0.024820739106453393, "percentage": 0.25, "elapsed_time": "0:00:22", "remaining_time": "2:28:15", "throughput": 2365.15, "total_tokens": 52352}
+{"current_steps": 95, "total_steps": 36260, "loss": 0.878, "lr": 1.2961941533370105e-06, "epoch": 0.026199669056811915, "percentage": 0.26, "elapsed_time": "0:00:23", "remaining_time": "2:27:48", "throughput": 2372.34, "total_tokens": 55264}
+{"current_steps": 100, "total_steps": 36260, "loss": 0.6017, "lr": 1.3651406508549365e-06, "epoch": 0.027578599007170437, "percentage": 0.28, "elapsed_time": "0:00:24", "remaining_time": "2:30:15", "throughput": 2382.22, "total_tokens": 59392}
+{"current_steps": 105, "total_steps": 36260, "loss": 0.8679, "lr": 1.4340871483728628e-06, "epoch": 0.02895752895752896, "percentage": 0.29, "elapsed_time": "0:00:26", "remaining_time": "2:30:10", "throughput": 2390.77, "total_tokens": 62560}
+{"current_steps": 110, "total_steps": 36260, "loss": 0.5288, "lr": 1.5030336458907888e-06, "epoch": 0.03033645890788748, "percentage": 0.3, "elapsed_time": "0:00:27", "remaining_time": "2:29:30", "throughput": 2391.42, "total_tokens": 65280}
+{"current_steps": 115, "total_steps": 36260, "loss": 0.4933, "lr": 1.5719801434087148e-06, "epoch": 0.031715388858246, "percentage": 0.32, "elapsed_time": "0:00:28", "remaining_time": "2:28:26", "throughput": 2390.65, "total_tokens": 67744}
+{"current_steps": 120, "total_steps": 36260, "loss": 0.6139, "lr": 1.6409266409266408e-06, "epoch": 0.03309431880860452, "percentage": 0.33, "elapsed_time": "0:00:29", "remaining_time": "2:29:50", "throughput": 2404.3, "total_tokens": 71776}
+{"current_steps": 125, "total_steps": 36260, "loss": 0.2704, "lr": 1.709873138444567e-06, "epoch": 0.03447324875896304, "percentage": 0.34, "elapsed_time": "0:00:31", "remaining_time": "2:29:53", "throughput": 2408.82, "total_tokens": 74944}
+{"current_steps": 130, "total_steps": 36260, "loss": 0.447, "lr": 1.778819635962493e-06, "epoch": 0.035852178709321565, "percentage": 0.36, "elapsed_time": "0:00:32", "remaining_time": "2:29:40", "throughput": 2409.33, "total_tokens": 77856}
+{"current_steps": 135, "total_steps": 36260, "loss": 0.1624, "lr": 1.847766133480419e-06, "epoch": 0.03723110865968009, "percentage": 0.37, "elapsed_time": "0:00:33", "remaining_time": "2:30:20", "throughput": 2416.91, "total_tokens": 81472}
+{"current_steps": 140, "total_steps": 36260, "loss": 0.2461, "lr": 1.9167126309983453e-06, "epoch": 0.03861003861003861, "percentage": 0.39, "elapsed_time": "0:00:34", "remaining_time": "2:29:30", "throughput": 2414.92, "total_tokens": 83968}
+{"current_steps": 145, "total_steps": 36260, "loss": 0.1736, "lr": 1.9856591285162715e-06, "epoch": 0.03998896856039713, "percentage": 0.4, "elapsed_time": "0:00:35", "remaining_time": "2:28:46", "throughput": 2414.34, "total_tokens": 86528}
+{"current_steps": 150, "total_steps": 36260, "loss": 0.3462, "lr": 2.0546056260341973e-06, "epoch": 0.04136789851075565, "percentage": 0.41, "elapsed_time": "0:00:37", "remaining_time": "2:30:09", "throughput": 2428.25, "total_tokens": 90880}
+{"current_steps": 155, "total_steps": 36260, "loss": 0.2281, "lr": 2.1235521235521236e-06, "epoch": 0.042746828461114175, "percentage": 0.43, "elapsed_time": "0:00:38", "remaining_time": "2:29:38", "throughput": 2426.74, "total_tokens": 93536}
+{"current_steps": 160, "total_steps": 36260, "loss": 0.414, "lr": 2.1924986210700498e-06, "epoch": 0.0441257584114727, "percentage": 0.44, "elapsed_time": "0:00:39", "remaining_time": "2:30:17", "throughput": 2437.19, "total_tokens": 97408}
+{"current_steps": 165, "total_steps": 36260, "loss": 0.1749, "lr": 2.2614451185879756e-06, "epoch": 0.04550468836183122, "percentage": 0.46, "elapsed_time": "0:00:41", "remaining_time": "2:30:16", "throughput": 2437.74, "total_tokens": 100480}
+{"current_steps": 170, "total_steps": 36260, "loss": 0.1878, "lr": 2.330391616105902e-06, "epoch": 0.04688361831218974, "percentage": 0.47, "elapsed_time": "0:00:42", "remaining_time": "2:31:12", "throughput": 2450.03, "total_tokens": 104704}
+{"current_steps": 175, "total_steps": 36260, "loss": 0.025, "lr": 2.399338113623828e-06, "epoch": 0.04826254826254826, "percentage": 0.48, "elapsed_time": "0:00:43", "remaining_time": "2:30:28", "throughput": 2443.29, "total_tokens": 106976}
+{"current_steps": 180, "total_steps": 36260, "loss": 0.2351, "lr": 2.4682846111417543e-06, "epoch": 0.049641478212906785, "percentage": 0.5, "elapsed_time": "0:00:44", "remaining_time": "2:29:45", "throughput": 2439.28, "total_tokens": 109344}
+{"current_steps": 185, "total_steps": 36260, "loss": 0.3986, "lr": 2.53723110865968e-06, "epoch": 0.05102040816326531, "percentage": 0.51, "elapsed_time": "0:00:46", "remaining_time": "2:29:30", "throughput": 2436.12, "total_tokens": 112064}
+{"current_steps": 190, "total_steps": 36260, "loss": 0.231, "lr": 2.6061776061776063e-06, "epoch": 0.05239933811362383, "percentage": 0.52, "elapsed_time": "0:00:47", "remaining_time": "2:28:52", "throughput": 2433.99, "total_tokens": 114528}
+{"current_steps": 195, "total_steps": 36260, "loss": 0.3256, "lr": 2.6751241036955325e-06, "epoch": 0.05377826806398235, "percentage": 0.54, "elapsed_time": "0:00:48", "remaining_time": "2:29:17", "throughput": 2438.14, "total_tokens": 118080}
+{"current_steps": 200, "total_steps": 36260, "loss": 0.357, "lr": 2.7440706012134583e-06, "epoch": 0.05515719801434087, "percentage": 0.55, "elapsed_time": "0:00:49", "remaining_time": "2:28:31", "throughput": 2434.34, "total_tokens": 120320}
+{"current_steps": 205, "total_steps": 36260, "loss": 0.2071, "lr": 2.8130170987313846e-06, "epoch": 0.056536127964699395, "percentage": 0.57, "elapsed_time": "0:00:50", "remaining_time": "2:28:05", "throughput": 2434.04, "total_tokens": 122976}
+{"current_steps": 210, "total_steps": 36260, "loss": 0.2674, "lr": 2.8819635962493108e-06, "epoch": 0.05791505791505792, "percentage": 0.58, "elapsed_time": "0:00:51", "remaining_time": "2:27:28", "throughput": 2431.12, "total_tokens": 125312}
+{"current_steps": 215, "total_steps": 36260, "loss": 0.0931, "lr": 2.950910093767237e-06, "epoch": 0.05929398786541644, "percentage": 0.59, "elapsed_time": "0:00:52", "remaining_time": "2:27:51", "throughput": 2438.26, "total_tokens": 129024}
+{"current_steps": 220, "total_steps": 36260, "loss": 0.3677, "lr": 3.019856591285163e-06, "epoch": 0.06067291781577496, "percentage": 0.61, "elapsed_time": "0:00:54", "remaining_time": "2:27:48", "throughput": 2440.56, "total_tokens": 132128}
+{"current_steps": 225, "total_steps": 36260, "loss": 0.1927, "lr": 3.088803088803089e-06, "epoch": 0.06205184776613348, "percentage": 0.62, "elapsed_time": "0:00:55", "remaining_time": "2:27:29", "throughput": 2442.21, "total_tokens": 134944}
+{"current_steps": 230, "total_steps": 36260, "loss": 0.2736, "lr": 3.157749586321015e-06, "epoch": 0.063430777716492, "percentage": 0.63, "elapsed_time": "0:00:56", "remaining_time": "2:27:49", "throughput": 2444.49, "total_tokens": 138400}
+{"current_steps": 235, "total_steps": 36260, "loss": 0.1733, "lr": 3.226696083838941e-06, "epoch": 0.06480970766685053, "percentage": 0.65, "elapsed_time": "0:00:57", "remaining_time": "2:27:26", "throughput": 2443.73, "total_tokens": 141024}
+{"current_steps": 240, "total_steps": 36260, "loss": 0.1692, "lr": 3.295642581356867e-06, "epoch": 0.06618863761720904, "percentage": 0.66, "elapsed_time": "0:00:58", "remaining_time": "2:27:05", "throughput": 2442.33, "total_tokens": 143616}
+{"current_steps": 245, "total_steps": 36260, "loss": 0.2171, "lr": 3.3645890788747935e-06, "epoch": 0.06756756756756757, "percentage": 0.68, "elapsed_time": "0:01:00", "remaining_time": "2:27:39", "throughput": 2445.5, "total_tokens": 147392}
+{"current_steps": 250, "total_steps": 36260, "loss": 0.1329, "lr": 3.4335355763927193e-06, "epoch": 0.06894649751792609, "percentage": 0.69, "elapsed_time": "0:01:01", "remaining_time": "2:27:37", "throughput": 2446.35, "total_tokens": 150432}
+{"current_steps": 255, "total_steps": 36260, "loss": 0.2505, "lr": 3.5024820739106456e-06, "epoch": 0.07032542746828462, "percentage": 0.7, "elapsed_time": "0:01:02", "remaining_time": "2:27:44", "throughput": 2447.2, "total_tokens": 153632}
+{"current_steps": 260, "total_steps": 36260, "loss": 0.1999, "lr": 3.5714285714285714e-06, "epoch": 0.07170435741864313, "percentage": 0.72, "elapsed_time": "0:01:03", "remaining_time": "2:27:21", "throughput": 2446.0, "total_tokens": 156192}
+{"current_steps": 265, "total_steps": 36260, "loss": 0.2638, "lr": 3.640375068946498e-06, "epoch": 0.07308328736900166, "percentage": 0.73, "elapsed_time": "0:01:05", "remaining_time": "2:28:01", "throughput": 2452.48, "total_tokens": 160352}
+{"current_steps": 270, "total_steps": 36260, "loss": 0.0676, "lr": 3.709321566464424e-06, "epoch": 0.07446221731936017, "percentage": 0.74, "elapsed_time": "0:01:06", "remaining_time": "2:28:12", "throughput": 2455.09, "total_tokens": 163776}
+{"current_steps": 275, "total_steps": 36260, "loss": 0.1888, "lr": 3.77826806398235e-06, "epoch": 0.0758411472697187, "percentage": 0.76, "elapsed_time": "0:01:08", "remaining_time": "2:28:36", "throughput": 2455.59, "total_tokens": 167328}
+{"current_steps": 280, "total_steps": 36260, "loss": 0.3264, "lr": 3.847214561500275e-06, "epoch": 0.07722007722007722, "percentage": 0.77, "elapsed_time": "0:01:09", "remaining_time": "2:28:30", "throughput": 2456.54, "total_tokens": 170336}
+{"current_steps": 285, "total_steps": 36260, "loss": 0.1758, "lr": 3.916161059018202e-06, "epoch": 0.07859900717043575, "percentage": 0.79, "elapsed_time": "0:01:10", "remaining_time": "2:28:11", "throughput": 2457.33, "total_tokens": 173088}
+{"current_steps": 290, "total_steps": 36260, "loss": 0.3236, "lr": 3.985107556536128e-06, "epoch": 0.07997793712079426, "percentage": 0.8, "elapsed_time": "0:01:11", "remaining_time": "2:28:39", "throughput": 2461.21, "total_tokens": 176992}
+{"current_steps": 295, "total_steps": 36260, "loss": 0.0426, "lr": 4.0540540540540545e-06, "epoch": 0.08135686707115279, "percentage": 0.81, "elapsed_time": "0:01:13", "remaining_time": "2:28:27", "throughput": 2463.3, "total_tokens": 179968}
+{"current_steps": 300, "total_steps": 36260, "loss": 0.3515, "lr": 4.12300055157198e-06, "epoch": 0.0827357970215113, "percentage": 0.83, "elapsed_time": "0:01:14", "remaining_time": "2:28:44", "throughput": 2466.2, "total_tokens": 183616}
+{"current_steps": 305, "total_steps": 36260, "loss": 0.2225, "lr": 4.191947049089907e-06, "epoch": 0.08411472697186984, "percentage": 0.84, "elapsed_time": "0:01:15", "remaining_time": "2:28:35", "throughput": 2463.82, "total_tokens": 186336}
+{"current_steps": 310, "total_steps": 36260, "loss": 0.0901, "lr": 4.260893546607833e-06, "epoch": 0.08549365692222835, "percentage": 0.85, "elapsed_time": "0:01:16", "remaining_time": "2:28:34", "throughput": 2465.16, "total_tokens": 189504}
+{"current_steps": 315, "total_steps": 36260, "loss": 0.1338, "lr": 4.329840044125759e-06, "epoch": 0.08687258687258688, "percentage": 0.87, "elapsed_time": "0:01:17", "remaining_time": "2:28:14", "throughput": 2462.42, "total_tokens": 191936}
+{"current_steps": 320, "total_steps": 36260, "loss": 0.3522, "lr": 4.398786541643684e-06, "epoch": 0.0882515168229454, "percentage": 0.88, "elapsed_time": "0:01:19", "remaining_time": "2:28:06", "throughput": 2463.45, "total_tokens": 194912}
+{"current_steps": 325, "total_steps": 36260, "loss": 0.1393, "lr": 4.467733039161611e-06, "epoch": 0.08963044677330391, "percentage": 0.9, "elapsed_time": "0:01:20", "remaining_time": "2:27:58", "throughput": 2463.74, "total_tokens": 197824}
+{"current_steps": 330, "total_steps": 36260, "loss": 0.1981, "lr": 4.536679536679537e-06, "epoch": 0.09100937672366244, "percentage": 0.91, "elapsed_time": "0:01:21", "remaining_time": "2:28:08", "throughput": 2466.26, "total_tokens": 201344}
+{"current_steps": 335, "total_steps": 36260, "loss": 0.1377, "lr": 4.6056260341974635e-06, "epoch": 0.09238830667402095, "percentage": 0.92, "elapsed_time": "0:01:22", "remaining_time": "2:28:13", "throughput": 2466.87, "total_tokens": 204576}
+{"current_steps": 340, "total_steps": 36260, "loss": 0.1207, "lr": 4.674572531715389e-06, "epoch": 0.09376723662437948, "percentage": 0.94, "elapsed_time": "0:01:24", "remaining_time": "2:27:58", "throughput": 2465.91, "total_tokens": 207232}
+{"current_steps": 345, "total_steps": 36260, "loss": 0.2269, "lr": 4.743519029233315e-06, "epoch": 0.095146166574738, "percentage": 0.95, "elapsed_time": "0:01:25", "remaining_time": "2:28:02", "throughput": 2467.45, "total_tokens": 210528}
+{"current_steps": 350, "total_steps": 36260, "loss": 0.0605, "lr": 4.812465526751241e-06, "epoch": 0.09652509652509653, "percentage": 0.97, "elapsed_time": "0:01:26", "remaining_time": "2:27:35", "throughput": 2464.49, "total_tokens": 212704}
+{"current_steps": 355, "total_steps": 36260, "loss": 0.0962, "lr": 4.881412024269167e-06, "epoch": 0.09790402647545504, "percentage": 0.98, "elapsed_time": "0:01:27", "remaining_time": "2:27:32", "throughput": 2465.18, "total_tokens": 215776}
+{"current_steps": 360, "total_steps": 36260, "loss": 0.088, "lr": 4.950358521787093e-06, "epoch": 0.09928295642581357, "percentage": 0.99, "elapsed_time": "0:01:28", "remaining_time": "2:27:18", "throughput": 2465.24, "total_tokens": 218496}
+{"current_steps": 365, "total_steps": 36260, "loss": 0.1856, "lr": 5.019305019305019e-06, "epoch": 0.10066188637617209, "percentage": 1.01, "elapsed_time": "0:01:30", "remaining_time": "2:27:37", "throughput": 2467.37, "total_tokens": 222240}
+{"current_steps": 370, "total_steps": 36260, "loss": 0.0668, "lr": 5.088251516822946e-06, "epoch": 0.10204081632653061, "percentage": 1.02, "elapsed_time": "0:01:31", "remaining_time": "2:27:32", "throughput": 2468.05, "total_tokens": 225248}
+{"current_steps": 375, "total_steps": 36260, "loss": 0.0359, "lr": 5.157198014340872e-06, "epoch": 0.10341974627688913, "percentage": 1.03, "elapsed_time": "0:01:32", "remaining_time": "2:27:18", "throughput": 2467.17, "total_tokens": 227872}
+{"current_steps": 380, "total_steps": 36260, "loss": 0.1299, "lr": 5.226144511858798e-06, "epoch": 0.10479867622724766, "percentage": 1.05, "elapsed_time": "0:01:33", "remaining_time": "2:27:06", "throughput": 2466.03, "total_tokens": 230528}
+{"current_steps": 385, "total_steps": 36260, "loss": 0.2488, "lr": 5.295091009376723e-06, "epoch": 0.10617760617760617, "percentage": 1.06, "elapsed_time": "0:01:34", "remaining_time": "2:27:01", "throughput": 2465.84, "total_tokens": 233440}
+{"current_steps": 390, "total_steps": 36260, "loss": 0.2388, "lr": 5.36403750689465e-06, "epoch": 0.1075565361279647, "percentage": 1.08, "elapsed_time": "0:01:36", "remaining_time": "2:27:22", "throughput": 2467.49, "total_tokens": 237216}
+{"current_steps": 395, "total_steps": 36260, "loss": 0.1085, "lr": 5.432984004412576e-06, "epoch": 0.10893546607832322, "percentage": 1.09, "elapsed_time": "0:01:37", "remaining_time": "2:27:00", "throughput": 2463.65, "total_tokens": 239328}
+{"current_steps": 400, "total_steps": 36260, "loss": 0.0505, "lr": 5.501930501930502e-06, "epoch": 0.11031439602868175, "percentage": 1.1, "elapsed_time": "0:01:38", "remaining_time": "2:26:43", "throughput": 2462.85, "total_tokens": 241856}
+{"current_steps": 405, "total_steps": 36260, "loss": 0.0662, "lr": 5.570876999448428e-06, "epoch": 0.11169332597904026, "percentage": 1.12, "elapsed_time": "0:01:39", "remaining_time": "2:26:30", "throughput": 2461.7, "total_tokens": 244416}
+{"current_steps": 410, "total_steps": 36260, "loss": 0.0659, "lr": 5.639823496966355e-06, "epoch": 0.11307225592939879, "percentage": 1.13, "elapsed_time": "0:01:40", "remaining_time": "2:26:22", "throughput": 2460.95, "total_tokens": 247168}
+{"current_steps": 415, "total_steps": 36260, "loss": 0.2146, "lr": 5.708769994484281e-06, "epoch": 0.1144511858797573, "percentage": 1.14, "elapsed_time": "0:01:41", "remaining_time": "2:26:27", "throughput": 2462.87, "total_tokens": 250560}
+{"current_steps": 420, "total_steps": 36260, "loss": 0.1969, "lr": 5.777716492002206e-06, "epoch": 0.11583011583011583, "percentage": 1.16, "elapsed_time": "0:01:42", "remaining_time": "2:26:24", "throughput": 2462.01, "total_tokens": 253440}
+{"current_steps": 425, "total_steps": 36260, "loss": 0.079, "lr": 5.846662989520132e-06, "epoch": 0.11720904578047435, "percentage": 1.17, "elapsed_time": "0:01:44", "remaining_time": "2:26:13", "throughput": 2460.58, "total_tokens": 256032}
+{"current_steps": 430, "total_steps": 36260, "loss": 0.1566, "lr": 5.915609487038059e-06, "epoch": 0.11858797573083288, "percentage": 1.19, "elapsed_time": "0:01:45", "remaining_time": "2:26:19", "throughput": 2462.1, "total_tokens": 259424}
+{"current_steps": 435, "total_steps": 36260, "loss": 0.1678, "lr": 5.984555984555985e-06, "epoch": 0.1199669056811914, "percentage": 1.2, "elapsed_time": "0:01:46", "remaining_time": "2:26:27", "throughput": 2462.24, "total_tokens": 262720}
+{"current_steps": 440, "total_steps": 36260, "loss": 0.1697, "lr": 6.053502482073911e-06, "epoch": 0.12134583563154992, "percentage": 1.21, "elapsed_time": "0:01:47", "remaining_time": "2:26:16", "throughput": 2460.05, "total_tokens": 265216}
+{"current_steps": 445, "total_steps": 36260, "loss": 0.0977, "lr": 6.122448979591837e-06, "epoch": 0.12272476558190844, "percentage": 1.23, "elapsed_time": "0:01:48", "remaining_time": "2:26:07", "throughput": 2458.48, "total_tokens": 267808}
+{"current_steps": 450, "total_steps": 36260, "loss": 0.2573, "lr": 6.191395477109764e-06, "epoch": 0.12410369553226697, "percentage": 1.24, "elapsed_time": "0:01:50", "remaining_time": "2:26:09", "throughput": 2457.71, "total_tokens": 270848}
+{"current_steps": 455, "total_steps": 36260, "loss": 0.1549, "lr": 6.2603419746276896e-06, "epoch": 0.12548262548262548, "percentage": 1.25, "elapsed_time": "0:01:51", "remaining_time": "2:26:05", "throughput": 2457.66, "total_tokens": 273760}
+{"current_steps": 460, "total_steps": 36260, "loss": 0.3127, "lr": 6.329288472145615e-06, "epoch": 0.126861555432984, "percentage": 1.27, "elapsed_time": "0:01:52", "remaining_time": "2:25:59", "throughput": 2456.89, "total_tokens": 276544}
+{"current_steps": 465, "total_steps": 36260, "loss": 0.1809, "lr": 6.398234969663541e-06, "epoch": 0.12824048538334254, "percentage": 1.28, "elapsed_time": "0:01:53", "remaining_time": "2:25:47", "throughput": 2455.98, "total_tokens": 279072}
+{"current_steps": 470, "total_steps": 36260, "loss": 0.1863, "lr": 6.467181467181467e-06, "epoch": 0.12961941533370105, "percentage": 1.3, "elapsed_time": "0:01:55", "remaining_time": "2:26:26", "throughput": 2458.32, "total_tokens": 283648}
+{"current_steps": 475, "total_steps": 36260, "loss": 0.0864, "lr": 6.5361279646993945e-06, "epoch": 0.13099834528405957, "percentage": 1.31, "elapsed_time": "0:01:56", "remaining_time": "2:26:22", "throughput": 2458.98, "total_tokens": 286656}
+{"current_steps": 480, "total_steps": 36260, "loss": 0.1479, "lr": 6.60507446221732e-06, "epoch": 0.13237727523441808, "percentage": 1.32, "elapsed_time": "0:01:57", "remaining_time": "2:26:14", "throughput": 2457.46, "total_tokens": 289280}
+{"current_steps": 485, "total_steps": 36260, "loss": 0.2347, "lr": 6.674020959735246e-06, "epoch": 0.1337562051847766, "percentage": 1.34, "elapsed_time": "0:01:58", "remaining_time": "2:25:59", "throughput": 2456.91, "total_tokens": 291776}
+{"current_steps": 490, "total_steps": 36260, "loss": 0.0784, "lr": 6.742967457253172e-06, "epoch": 0.13513513513513514, "percentage": 1.35, "elapsed_time": "0:01:59", "remaining_time": "2:25:42", "throughput": 2455.14, "total_tokens": 294016}
+{"current_steps": 495, "total_steps": 36260, "loss": 0.1207, "lr": 6.8119139547710985e-06, "epoch": 0.13651406508549366, "percentage": 1.37, "elapsed_time": "0:02:00", "remaining_time": "2:25:28", "throughput": 2453.63, "total_tokens": 296416}
+{"current_steps": 500, "total_steps": 36260, "loss": 0.1375, "lr": 6.880860452289024e-06, "epoch": 0.13789299503585217, "percentage": 1.38, "elapsed_time": "0:02:02", "remaining_time": "2:25:36", "throughput": 2456.55, "total_tokens": 300096}
+{"current_steps": 505, "total_steps": 36260, "loss": 0.1348, "lr": 6.94980694980695e-06, "epoch": 0.1392719249862107, "percentage": 1.39, "elapsed_time": "0:02:03", "remaining_time": "2:25:36", "throughput": 2455.91, "total_tokens": 303040}
+{"current_steps": 510, "total_steps": 36260, "loss": 0.3134, "lr": 7.018753447324876e-06, "epoch": 0.14065085493656923, "percentage": 1.41, "elapsed_time": "0:02:04", "remaining_time": "2:25:45", "throughput": 2456.47, "total_tokens": 306464}
+{"current_steps": 515, "total_steps": 36260, "loss": 0.1024, "lr": 7.087699944842802e-06, "epoch": 0.14202978488692775, "percentage": 1.42, "elapsed_time": "0:02:06", "remaining_time": "2:25:46", "throughput": 2455.94, "total_tokens": 309472}
+{"current_steps": 520, "total_steps": 36260, "loss": 0.0612, "lr": 7.156646442360728e-06, "epoch": 0.14340871483728626, "percentage": 1.43, "elapsed_time": "0:02:07", "remaining_time": "2:25:55", "throughput": 2457.28, "total_tokens": 313024}
+{"current_steps": 525, "total_steps": 36260, "loss": 0.1524, "lr": 7.225592939878654e-06, "epoch": 0.14478764478764478, "percentage": 1.45, "elapsed_time": "0:02:08", "remaining_time": "2:25:52", "throughput": 2457.13, "total_tokens": 315968}
+{"current_steps": 530, "total_steps": 36260, "loss": 0.0898, "lr": 7.29453943739658e-06, "epoch": 0.14616657473800332, "percentage": 1.46, "elapsed_time": "0:02:09", "remaining_time": "2:25:53", "throughput": 2457.76, "total_tokens": 319136}
+{"current_steps": 535, "total_steps": 36260, "loss": 0.3482, "lr": 7.363485934914506e-06, "epoch": 0.14754550468836183, "percentage": 1.48, "elapsed_time": "0:02:11", "remaining_time": "2:26:02", "throughput": 2459.92, "total_tokens": 322816}
+{"current_steps": 540, "total_steps": 36260, "loss": 0.1018, "lr": 7.432432432432433e-06, "epoch": 0.14892443463872035, "percentage": 1.49, "elapsed_time": "0:02:12", "remaining_time": "2:25:56", "throughput": 2459.06, "total_tokens": 325536}
+{"current_steps": 545, "total_steps": 36260, "loss": 0.1982, "lr": 7.501378929950359e-06, "epoch": 0.15030336458907886, "percentage": 1.5, "elapsed_time": "0:02:13", "remaining_time": "2:25:53", "throughput": 2458.32, "total_tokens": 328384}
+{"current_steps": 550, "total_steps": 36260, "loss": 0.0413, "lr": 7.570325427468285e-06, "epoch": 0.1516822945394374, "percentage": 1.52, "elapsed_time": "0:02:14", "remaining_time": "2:26:01", "throughput": 2460.2, "total_tokens": 332000}
+{"current_steps": 555, "total_steps": 36260, "loss": 0.0576, "lr": 7.63927192498621e-06, "epoch": 0.15306122448979592, "percentage": 1.53, "elapsed_time": "0:02:16", "remaining_time": "2:25:54", "throughput": 2460.1, "total_tokens": 334752}
+{"current_steps": 560, "total_steps": 36260, "loss": 0.2015, "lr": 7.708218422504138e-06, "epoch": 0.15444015444015444, "percentage": 1.54, "elapsed_time": "0:02:17", "remaining_time": "2:25:51", "throughput": 2460.73, "total_tokens": 337792}
+{"current_steps": 565, "total_steps": 36260, "loss": 0.1955, "lr": 7.777164920022063e-06, "epoch": 0.15581908439051295, "percentage": 1.56, "elapsed_time": "0:02:18", "remaining_time": "2:25:38", "throughput": 2459.14, "total_tokens": 340128}
+{"current_steps": 570, "total_steps": 36260, "loss": 0.1704, "lr": 7.84611141753999e-06, "epoch": 0.1571980143408715, "percentage": 1.57, "elapsed_time": "0:02:19", "remaining_time": "2:25:31", "throughput": 2458.69, "total_tokens": 342880}
+{"current_steps": 575, "total_steps": 36260, "loss": 0.1103, "lr": 7.915057915057915e-06, "epoch": 0.15857694429123, "percentage": 1.59, "elapsed_time": "0:02:20", "remaining_time": "2:25:30", "throughput": 2458.33, "total_tokens": 345824}
+{"current_steps": 580, "total_steps": 36260, "loss": 0.1035, "lr": 7.984004412575841e-06, "epoch": 0.15995587424158852, "percentage": 1.6, "elapsed_time": "0:02:21", "remaining_time": "2:25:16", "throughput": 2456.9, "total_tokens": 348128}
+{"current_steps": 585, "total_steps": 36260, "loss": 0.1242, "lr": 8.052950910093768e-06, "epoch": 0.16133480419194704, "percentage": 1.61, "elapsed_time": "0:02:22", "remaining_time": "2:25:12", "throughput": 2456.72, "total_tokens": 350976}
+{"current_steps": 590, "total_steps": 36260, "loss": 0.0733, "lr": 8.121897407611693e-06, "epoch": 0.16271373414230558, "percentage": 1.63, "elapsed_time": "0:02:24", "remaining_time": "2:25:12", "throughput": 2456.47, "total_tokens": 353984}
+{"current_steps": 595, "total_steps": 36260, "loss": 0.2745, "lr": 8.19084390512962e-06, "epoch": 0.1640926640926641, "percentage": 1.64, "elapsed_time": "0:02:25", "remaining_time": "2:25:02", "throughput": 2456.26, "total_tokens": 356608}
+{"current_steps": 600, "total_steps": 36260, "loss": 0.1379, "lr": 8.259790402647546e-06, "epoch": 0.1654715940430226, "percentage": 1.65, "elapsed_time": "0:02:26", "remaining_time": "2:24:55", "throughput": 2456.19, "total_tokens": 359360}
+{"current_steps": 605, "total_steps": 36260, "loss": 0.0612, "lr": 8.328736900165473e-06, "epoch": 0.16685052399338113, "percentage": 1.67, "elapsed_time": "0:02:27", "remaining_time": "2:24:46", "throughput": 2453.95, "total_tokens": 361696}
+{"current_steps": 610, "total_steps": 36260, "loss": 0.3031, "lr": 8.397683397683398e-06, "epoch": 0.16822945394373967, "percentage": 1.68, "elapsed_time": "0:02:28", "remaining_time": "2:24:45", "throughput": 2453.48, "total_tokens": 364640}
+{"current_steps": 615, "total_steps": 36260, "loss": 0.2168, "lr": 8.466629895201323e-06, "epoch": 0.16960838389409819, "percentage": 1.7, "elapsed_time": "0:02:29", "remaining_time": "2:24:43", "throughput": 2452.52, "total_tokens": 367424}
+{"current_steps": 620, "total_steps": 36260, "loss": 0.1513, "lr": 8.535576392719251e-06, "epoch": 0.1709873138444567, "percentage": 1.71, "elapsed_time": "0:02:30", "remaining_time": "2:24:37", "throughput": 2451.96, "total_tokens": 370144}
+{"current_steps": 625, "total_steps": 36260, "loss": 0.1476, "lr": 8.604522890237176e-06, "epoch": 0.17236624379481522, "percentage": 1.72, "elapsed_time": "0:02:32", "remaining_time": "2:24:31", "throughput": 2452.26, "total_tokens": 372960}
+{"current_steps": 630, "total_steps": 36260, "loss": 0.0805, "lr": 8.673469387755103e-06, "epoch": 0.17374517374517376, "percentage": 1.74, "elapsed_time": "0:02:33", "remaining_time": "2:24:23", "throughput": 2451.84, "total_tokens": 375584}
+{"current_steps": 635, "total_steps": 36260, "loss": 0.0878, "lr": 8.742415885273028e-06, "epoch": 0.17512410369553227, "percentage": 1.75, "elapsed_time": "0:02:34", "remaining_time": "2:24:17", "throughput": 2451.76, "total_tokens": 378368}
+{"current_steps": 640, "total_steps": 36260, "loss": 0.2009, "lr": 8.811362382790954e-06, "epoch": 0.1765030336458908, "percentage": 1.77, "elapsed_time": "0:02:35", "remaining_time": "2:24:17", "throughput": 2451.82, "total_tokens": 381408}
+{"current_steps": 645, "total_steps": 36260, "loss": 0.2805, "lr": 8.880308880308881e-06, "epoch": 0.1778819635962493, "percentage": 1.78, "elapsed_time": "0:02:36", "remaining_time": "2:24:25", "throughput": 2454.0, "total_tokens": 385120}
+{"current_steps": 650, "total_steps": 36260, "loss": 0.077, "lr": 8.949255377826806e-06, "epoch": 0.17926089354660782, "percentage": 1.79, "elapsed_time": "0:02:38", "remaining_time": "2:24:35", "throughput": 2454.69, "total_tokens": 388736}
+{"current_steps": 655, "total_steps": 36260, "loss": 0.0423, "lr": 9.018201875344733e-06, "epoch": 0.18063982349696636, "percentage": 1.81, "elapsed_time": "0:02:39", "remaining_time": "2:24:28", "throughput": 2453.66, "total_tokens": 391296}
+{"current_steps": 660, "total_steps": 36260, "loss": 0.1452, "lr": 9.08714837286266e-06, "epoch": 0.18201875344732488, "percentage": 1.82, "elapsed_time": "0:02:40", "remaining_time": "2:24:31", "throughput": 2453.69, "total_tokens": 394464}
+{"current_steps": 665, "total_steps": 36260, "loss": 0.1126, "lr": 9.156094870380586e-06, "epoch": 0.1833976833976834, "percentage": 1.83, "elapsed_time": "0:02:42", "remaining_time": "2:24:34", "throughput": 2454.3, "total_tokens": 397728}
+{"current_steps": 670, "total_steps": 36260, "loss": 0.1552, "lr": 9.225041367898511e-06, "epoch": 0.1847766133480419, "percentage": 1.85, "elapsed_time": "0:02:43", "remaining_time": "2:24:26", "throughput": 2453.58, "total_tokens": 400288}
+{"current_steps": 675, "total_steps": 36260, "loss": 0.1061, "lr": 9.293987865416438e-06, "epoch": 0.18615554329840045, "percentage": 1.86, "elapsed_time": "0:02:44", "remaining_time": "2:24:28", "throughput": 2454.32, "total_tokens": 403552}
+{"current_steps": 680, "total_steps": 36260, "loss": 0.1482, "lr": 9.362934362934363e-06, "epoch": 0.18753447324875896, "percentage": 1.88, "elapsed_time": "0:02:45", "remaining_time": "2:24:22", "throughput": 2454.4, "total_tokens": 406336}
+{"current_steps": 685, "total_steps": 36260, "loss": 0.1531, "lr": 9.43188086045229e-06, "epoch": 0.18891340319911748, "percentage": 1.89, "elapsed_time": "0:02:46", "remaining_time": "2:24:22", "throughput": 2454.19, "total_tokens": 409376}
+{"current_steps": 690, "total_steps": 36260, "loss": 0.0885, "lr": 9.500827357970216e-06, "epoch": 0.190292333149476, "percentage": 1.9, "elapsed_time": "0:02:47", "remaining_time": "2:24:19", "throughput": 2454.47, "total_tokens": 412288}
+{"current_steps": 695, "total_steps": 36260, "loss": 0.0576, "lr": 9.56977385548814e-06, "epoch": 0.19167126309983454, "percentage": 1.92, "elapsed_time": "0:02:49", "remaining_time": "2:24:34", "throughput": 2455.64, "total_tokens": 416256}
+{"current_steps": 700, "total_steps": 36260, "loss": 0.2956, "lr": 9.638720353006067e-06, "epoch": 0.19305019305019305, "percentage": 1.93, "elapsed_time": "0:02:50", "remaining_time": "2:24:26", "throughput": 2455.56, "total_tokens": 418912}
+{"current_steps": 705, "total_steps": 36260, "loss": 0.1216, "lr": 9.707666850523994e-06, "epoch": 0.19442912300055157, "percentage": 1.94, "elapsed_time": "0:02:51", "remaining_time": "2:24:21", "throughput": 2456.58, "total_tokens": 421920}
+{"current_steps": 710, "total_steps": 36260, "loss": 0.0975, "lr": 9.77661334804192e-06, "epoch": 0.19580805295091008, "percentage": 1.96, "elapsed_time": "0:02:52", "remaining_time": "2:24:08", "throughput": 2454.72, "total_tokens": 424000}
+{"current_steps": 715, "total_steps": 36260, "loss": 0.2162, "lr": 9.845559845559846e-06, "epoch": 0.19718698290126863, "percentage": 1.97, "elapsed_time": "0:02:53", "remaining_time": "2:24:03", "throughput": 2454.25, "total_tokens": 426720}
+{"current_steps": 720, "total_steps": 36260, "loss": 0.0845, "lr": 9.914506343077772e-06, "epoch": 0.19856591285162714, "percentage": 1.99, "elapsed_time": "0:02:55", "remaining_time": "2:24:03", "throughput": 2454.94, "total_tokens": 429856}
+{"current_steps": 725, "total_steps": 36260, "loss": 0.1461, "lr": 9.983452840595699e-06, "epoch": 0.19994484280198566, "percentage": 2.0, "elapsed_time": "0:02:56", "remaining_time": "2:24:06", "throughput": 2455.49, "total_tokens": 433184}
+{"current_steps": 730, "total_steps": 36260, "loss": 0.2545, "lr": 1.0052399338113624e-05, "epoch": 0.20132377275234417, "percentage": 2.01, "elapsed_time": "0:02:57", "remaining_time": "2:24:12", "throughput": 2456.28, "total_tokens": 436640}
+{"current_steps": 735, "total_steps": 36260, "loss": 0.1301, "lr": 1.012134583563155e-05, "epoch": 0.20270270270270271, "percentage": 2.03, "elapsed_time": "0:02:59", "remaining_time": "2:24:21", "throughput": 2458.05, "total_tokens": 440512}
+{"current_steps": 740, "total_steps": 36260, "loss": 0.1534, "lr": 1.0190292333149476e-05, "epoch": 0.20408163265306123, "percentage": 2.04, "elapsed_time": "0:03:00", "remaining_time": "2:24:27", "throughput": 2459.14, "total_tokens": 444032}
+{"current_steps": 745, "total_steps": 36260, "loss": 0.1086, "lr": 1.0259238830667404e-05, "epoch": 0.20546056260341974, "percentage": 2.05, "elapsed_time": "0:03:01", "remaining_time": "2:24:28", "throughput": 2459.07, "total_tokens": 447136}
+{"current_steps": 750, "total_steps": 36260, "loss": 0.1734, "lr": 1.0328185328185329e-05, "epoch": 0.20683949255377826, "percentage": 2.07, "elapsed_time": "0:03:02", "remaining_time": "2:24:21", "throughput": 2458.24, "total_tokens": 449728}
+{"current_steps": 755, "total_steps": 36260, "loss": 0.1162, "lr": 1.0397131825703254e-05, "epoch": 0.2082184225041368, "percentage": 2.08, "elapsed_time": "0:03:04", "remaining_time": "2:24:25", "throughput": 2459.57, "total_tokens": 453216}
+{"current_steps": 760, "total_steps": 36260, "loss": 0.1742, "lr": 1.046607832322118e-05, "epoch": 0.20959735245449532, "percentage": 2.1, "elapsed_time": "0:03:05", "remaining_time": "2:24:18", "throughput": 2459.72, "total_tokens": 455968}
+{"current_steps": 765, "total_steps": 36260, "loss": 0.0803, "lr": 1.0535024820739107e-05, "epoch": 0.21097628240485383, "percentage": 2.11, "elapsed_time": "0:03:06", "remaining_time": "2:24:08", "throughput": 2458.14, "total_tokens": 458176}
+{"current_steps": 770, "total_steps": 36260, "loss": 0.11, "lr": 1.0603971318257034e-05, "epoch": 0.21235521235521235, "percentage": 2.12, "elapsed_time": "0:03:07", "remaining_time": "2:24:05", "throughput": 2459.26, "total_tokens": 461280}
+{"current_steps": 775, "total_steps": 36260, "loss": 0.1382, "lr": 1.0672917815774959e-05, "epoch": 0.2137341423055709, "percentage": 2.14, "elapsed_time": "0:03:09", "remaining_time": "2:24:20", "throughput": 2459.98, "total_tokens": 465312}
+{"current_steps": 780, "total_steps": 36260, "loss": 0.2158, "lr": 1.0741864313292885e-05, "epoch": 0.2151130722559294, "percentage": 2.15, "elapsed_time": "0:03:10", "remaining_time": "2:24:29", "throughput": 2461.35, "total_tokens": 469088}
+{"current_steps": 785, "total_steps": 36260, "loss": 0.0473, "lr": 1.0810810810810812e-05, "epoch": 0.21649200220628792, "percentage": 2.16, "elapsed_time": "0:03:11", "remaining_time": "2:24:30", "throughput": 2461.75, "total_tokens": 472320}
+{"current_steps": 790, "total_steps": 36260, "loss": 0.0865, "lr": 1.0879757308328737e-05, "epoch": 0.21787093215664644, "percentage": 2.18, "elapsed_time": "0:03:12", "remaining_time": "2:24:24", "throughput": 2461.55, "total_tokens": 475040}
+{"current_steps": 795, "total_steps": 36260, "loss": 0.0691, "lr": 1.0948703805846664e-05, "epoch": 0.21924986210700498, "percentage": 2.19, "elapsed_time": "0:03:14", "remaining_time": "2:24:16", "throughput": 2460.51, "total_tokens": 477440}
+{"current_steps": 800, "total_steps": 36260, "loss": 0.1587, "lr": 1.1017650303364589e-05, "epoch": 0.2206287920573635, "percentage": 2.21, "elapsed_time": "0:03:15", "remaining_time": "2:24:25", "throughput": 2461.42, "total_tokens": 481216}
+{"current_steps": 805, "total_steps": 36260, "loss": 0.0477, "lr": 1.1086596800882517e-05, "epoch": 0.222007722007722, "percentage": 2.22, "elapsed_time": "0:03:16", "remaining_time": "2:24:30", "throughput": 2461.64, "total_tokens": 484608}
+{"current_steps": 810, "total_steps": 36260, "loss": 0.0163, "lr": 1.1155543298400442e-05, "epoch": 0.22338665195808052, "percentage": 2.23, "elapsed_time": "0:03:18", "remaining_time": "2:24:28", "throughput": 2462.48, "total_tokens": 487712}
+{"current_steps": 815, "total_steps": 36260, "loss": 0.1218, "lr": 1.1224489795918369e-05, "epoch": 0.22476558190843904, "percentage": 2.25, "elapsed_time": "0:03:19", "remaining_time": "2:24:29", "throughput": 2462.03, "total_tokens": 490784}
+{"current_steps": 820, "total_steps": 36260, "loss": 0.2492, "lr": 1.1293436293436294e-05, "epoch": 0.22614451185879758, "percentage": 2.26, "elapsed_time": "0:03:20", "remaining_time": "2:24:22", "throughput": 2461.93, "total_tokens": 493472}
+{"current_steps": 825, "total_steps": 36260, "loss": 0.0771, "lr": 1.136238279095422e-05, "epoch": 0.2275234418091561, "percentage": 2.28, "elapsed_time": "0:03:21", "remaining_time": "2:24:26", "throughput": 2461.92, "total_tokens": 496768}
+{"current_steps": 830, "total_steps": 36260, "loss": 0.2474, "lr": 1.1431329288472147e-05, "epoch": 0.2289023717595146, "percentage": 2.29, "elapsed_time": "0:03:22", "remaining_time": "2:24:23", "throughput": 2461.67, "total_tokens": 499616}
+{"current_steps": 835, "total_steps": 36260, "loss": 0.1169, "lr": 1.1500275785990072e-05, "epoch": 0.23028130170987313, "percentage": 2.3, "elapsed_time": "0:03:24", "remaining_time": "2:24:16", "throughput": 2460.91, "total_tokens": 502144}
+{"current_steps": 840, "total_steps": 36260, "loss": 0.0383, "lr": 1.1569222283507998e-05, "epoch": 0.23166023166023167, "percentage": 2.32, "elapsed_time": "0:03:25", "remaining_time": "2:24:14", "throughput": 2461.46, "total_tokens": 505216}
+{"current_steps": 845, "total_steps": 36260, "loss": 0.0655, "lr": 1.1638168781025923e-05, "epoch": 0.23303916161059018, "percentage": 2.33, "elapsed_time": "0:03:26", "remaining_time": "2:24:12", "throughput": 2460.86, "total_tokens": 508032}
+{"current_steps": 850, "total_steps": 36260, "loss": 0.0876, "lr": 1.1707115278543852e-05, "epoch": 0.2344180915609487, "percentage": 2.34, "elapsed_time": "0:03:27", "remaining_time": "2:24:13", "throughput": 2461.42, "total_tokens": 511264}
+{"current_steps": 855, "total_steps": 36260, "loss": 0.0729, "lr": 1.1776061776061777e-05, "epoch": 0.23579702151130721, "percentage": 2.36, "elapsed_time": "0:03:29", "remaining_time": "2:24:38", "throughput": 2463.99, "total_tokens": 516384}
+{"current_steps": 860, "total_steps": 36260, "loss": 0.037, "lr": 1.1845008273579702e-05, "epoch": 0.23717595146166576, "percentage": 2.37, "elapsed_time": "0:03:30", "remaining_time": "2:24:42", "throughput": 2464.72, "total_tokens": 519872}
+{"current_steps": 865, "total_steps": 36260, "loss": 0.0903, "lr": 1.1913954771097628e-05, "epoch": 0.23855488141202427, "percentage": 2.39, "elapsed_time": "0:03:32", "remaining_time": "2:24:41", "throughput": 2464.11, "total_tokens": 522816}
+{"current_steps": 870, "total_steps": 36260, "loss": 0.2302, "lr": 1.1982901268615555e-05, "epoch": 0.2399338113623828, "percentage": 2.4, "elapsed_time": "0:03:33", "remaining_time": "2:24:45", "throughput": 2464.59, "total_tokens": 526240}
+{"current_steps": 875, "total_steps": 36260, "loss": 0.3389, "lr": 1.2051847766133482e-05, "epoch": 0.2413127413127413, "percentage": 2.41, "elapsed_time": "0:03:34", "remaining_time": "2:24:39", "throughput": 2464.59, "total_tokens": 528960}
+{"current_steps": 880, "total_steps": 36260, "loss": 0.0851, "lr": 1.2120794263651407e-05, "epoch": 0.24269167126309985, "percentage": 2.43, "elapsed_time": "0:03:35", "remaining_time": "2:24:44", "throughput": 2465.52, "total_tokens": 532544}
+{"current_steps": 885, "total_steps": 36260, "loss": 0.0875, "lr": 1.2189740761169333e-05, "epoch": 0.24407060121345836, "percentage": 2.44, "elapsed_time": "0:03:37", "remaining_time": "2:24:40", "throughput": 2465.13, "total_tokens": 535328}
+{"current_steps": 890, "total_steps": 36260, "loss": 0.1118, "lr": 1.225868725868726e-05, "epoch": 0.24544953116381688, "percentage": 2.45, "elapsed_time": "0:03:38", "remaining_time": "2:24:36", "throughput": 2464.59, "total_tokens": 538048}
+{"current_steps": 895, "total_steps": 36260, "loss": 0.0619, "lr": 1.2327633756205185e-05, "epoch": 0.2468284611141754, "percentage": 2.47, "elapsed_time": "0:03:39", "remaining_time": "2:24:31", "throughput": 2464.32, "total_tokens": 540832}
+{"current_steps": 900, "total_steps": 36260, "loss": 0.0613, "lr": 1.2396580253723111e-05, "epoch": 0.24820739106453393, "percentage": 2.48, "elapsed_time": "0:03:41", "remaining_time": "2:24:50", "throughput": 2466.77, "total_tokens": 545632}
+{"current_steps": 905, "total_steps": 36260, "loss": 0.1849, "lr": 1.2465526751241036e-05, "epoch": 0.24958632101489245, "percentage": 2.5, "elapsed_time": "0:03:42", "remaining_time": "2:24:41", "throughput": 2465.28, "total_tokens": 547840}
+{"current_steps": 910, "total_steps": 36260, "loss": 0.2089, "lr": 1.2534473248758963e-05, "epoch": 0.25096525096525096, "percentage": 2.51, "elapsed_time": "0:03:43", "remaining_time": "2:24:30", "throughput": 2463.94, "total_tokens": 549952}
+{"current_steps": 915, "total_steps": 36260, "loss": 0.1107, "lr": 1.260341974627689e-05, "epoch": 0.2523441809156095, "percentage": 2.52, "elapsed_time": "0:03:44", "remaining_time": "2:24:25", "throughput": 2464.21, "total_tokens": 552800}
+{"current_steps": 920, "total_steps": 36260, "loss": 0.148, "lr": 1.2672366243794816e-05, "epoch": 0.253723110865968, "percentage": 2.54, "elapsed_time": "0:03:45", "remaining_time": "2:24:37", "throughput": 2465.24, "total_tokens": 556928}
+{"current_steps": 925, "total_steps": 36260, "loss": 0.0527, "lr": 1.2741312741312741e-05, "epoch": 0.25510204081632654, "percentage": 2.55, "elapsed_time": "0:03:47", "remaining_time": "2:24:34", "throughput": 2464.97, "total_tokens": 559744}
+{"current_steps": 930, "total_steps": 36260, "loss": 0.1174, "lr": 1.2810259238830668e-05, "epoch": 0.2564809707666851, "percentage": 2.56, "elapsed_time": "0:03:48", "remaining_time": "2:24:28", "throughput": 2464.06, "total_tokens": 562272}
+{"current_steps": 935, "total_steps": 36260, "loss": 0.0961, "lr": 1.2879205736348593e-05, "epoch": 0.25785990071704357, "percentage": 2.58, "elapsed_time": "0:03:49", "remaining_time": "2:24:18", "throughput": 2463.17, "total_tokens": 564512}
+{"current_steps": 940, "total_steps": 36260, "loss": 0.0335, "lr": 1.294815223386652e-05, "epoch": 0.2592388306674021, "percentage": 2.59, "elapsed_time": "0:03:50", "remaining_time": "2:24:21", "throughput": 2463.91, "total_tokens": 568000}
+{"current_steps": 945, "total_steps": 36260, "loss": 0.0943, "lr": 1.3017098731384448e-05, "epoch": 0.2606177606177606, "percentage": 2.61, "elapsed_time": "0:03:53", "remaining_time": "2:25:11", "throughput": 2468.76, "total_tokens": 575520}
+{"current_steps": 950, "total_steps": 36260, "loss": 0.0553, "lr": 1.3086045228902371e-05, "epoch": 0.26199669056811914, "percentage": 2.62, "elapsed_time": "0:03:54", "remaining_time": "2:25:13", "throughput": 2469.5, "total_tokens": 578944}
+{"current_steps": 955, "total_steps": 36260, "loss": 0.0884, "lr": 1.31549917264203e-05, "epoch": 0.2633756205184777, "percentage": 2.63, "elapsed_time": "0:03:55", "remaining_time": "2:25:10", "throughput": 2469.64, "total_tokens": 581920}
+{"current_steps": 960, "total_steps": 36260, "loss": 0.3079, "lr": 1.3223938223938226e-05, "epoch": 0.26475455046883617, "percentage": 2.65, "elapsed_time": "0:03:56", "remaining_time": "2:25:03", "throughput": 2468.66, "total_tokens": 584288}
+{"current_steps": 965, "total_steps": 36260, "loss": 0.2113, "lr": 1.329288472145615e-05, "epoch": 0.2661334804191947, "percentage": 2.66, "elapsed_time": "0:03:58", "remaining_time": "2:25:11", "throughput": 2469.77, "total_tokens": 588256}
+{"current_steps": 970, "total_steps": 36260, "loss": 0.0899, "lr": 1.3361831218974078e-05, "epoch": 0.2675124103695532, "percentage": 2.68, "elapsed_time": "0:03:59", "remaining_time": "2:25:07", "throughput": 2468.96, "total_tokens": 590912}
+{"current_steps": 975, "total_steps": 36260, "loss": 0.0264, "lr": 1.3430777716492001e-05, "epoch": 0.26889134031991174, "percentage": 2.69, "elapsed_time": "0:04:00", "remaining_time": "2:25:05", "throughput": 2469.29, "total_tokens": 594016}
+{"current_steps": 980, "total_steps": 36260, "loss": 0.0369, "lr": 1.349972421400993e-05, "epoch": 0.2702702702702703, "percentage": 2.7, "elapsed_time": "0:04:02", "remaining_time": "2:25:20", "throughput": 2471.02, "total_tokens": 598592}
+{"current_steps": 985, "total_steps": 36260, "loss": 0.0794, "lr": 1.3568670711527856e-05, "epoch": 0.2716492002206288, "percentage": 2.72, "elapsed_time": "0:04:03", "remaining_time": "2:25:16", "throughput": 2470.46, "total_tokens": 601280}
+{"current_steps": 990, "total_steps": 36260, "loss": 0.2106, "lr": 1.3637617209045781e-05, "epoch": 0.2730281301709873, "percentage": 2.73, "elapsed_time": "0:04:04", "remaining_time": "2:25:08", "throughput": 2469.84, "total_tokens": 603712}
+{"current_steps": 995, "total_steps": 36260, "loss": 0.3632, "lr": 1.3706563706563708e-05, "epoch": 0.27440706012134586, "percentage": 2.74, "elapsed_time": "0:04:05", "remaining_time": "2:25:08", "throughput": 2470.93, "total_tokens": 607136}
+{"current_steps": 1000, "total_steps": 36260, "loss": 0.104, "lr": 1.3775510204081633e-05, "epoch": 0.27578599007170435, "percentage": 2.76, "elapsed_time": "0:04:07", "remaining_time": "2:25:23", "throughput": 2472.22, "total_tokens": 611648}
+{"current_steps": 1005, "total_steps": 36260, "loss": 0.1017, "lr": 1.384445670159956e-05, "epoch": 0.2771649200220629, "percentage": 2.77, "elapsed_time": "0:04:09", "remaining_time": "2:25:34", "throughput": 2473.37, "total_tokens": 615872}
+{"current_steps": 1010, "total_steps": 36260, "loss": 0.0645, "lr": 1.3913403199117486e-05, "epoch": 0.2785438499724214, "percentage": 2.79, "elapsed_time": "0:04:10", "remaining_time": "2:25:31", "throughput": 2473.76, "total_tokens": 618880}
+{"current_steps": 1015, "total_steps": 36260, "loss": 0.0939, "lr": 1.3982349696635411e-05, "epoch": 0.2799227799227799, "percentage": 2.8, "elapsed_time": "0:04:11", "remaining_time": "2:25:26", "throughput": 2473.37, "total_tokens": 621600}
+{"current_steps": 1020, "total_steps": 36260, "loss": 0.0655, "lr": 1.4051296194153338e-05, "epoch": 0.28130170987313846, "percentage": 2.81, "elapsed_time": "0:04:12", "remaining_time": "2:25:24", "throughput": 2473.55, "total_tokens": 624640}
+{"current_steps": 1025, "total_steps": 36260, "loss": 0.0496, "lr": 1.4120242691671264e-05, "epoch": 0.28268063982349695, "percentage": 2.83, "elapsed_time": "0:04:13", "remaining_time": "2:25:25", "throughput": 2473.52, "total_tokens": 627872}
+{"current_steps": 1030, "total_steps": 36260, "loss": 0.1605, "lr": 1.4189189189189189e-05, "epoch": 0.2840595697738555, "percentage": 2.84, "elapsed_time": "0:04:14", "remaining_time": "2:25:21", "throughput": 2473.76, "total_tokens": 630752}
+{"current_steps": 1035, "total_steps": 36260, "loss": 0.0605, "lr": 1.4258135686707116e-05, "epoch": 0.28543849972421403, "percentage": 2.85, "elapsed_time": "0:04:16", "remaining_time": "2:25:22", "throughput": 2474.41, "total_tokens": 634176}
+{"current_steps": 1040, "total_steps": 36260, "loss": 0.3454, "lr": 1.432708218422504e-05, "epoch": 0.2868174296745725, "percentage": 2.87, "elapsed_time": "0:04:17", "remaining_time": "2:25:19", "throughput": 2474.26, "total_tokens": 637056}
+{"current_steps": 1045, "total_steps": 36260, "loss": 0.1984, "lr": 1.4396028681742967e-05, "epoch": 0.28819635962493106, "percentage": 2.88, "elapsed_time": "0:04:18", "remaining_time": "2:25:15", "throughput": 2474.42, "total_tokens": 639936}
+{"current_steps": 1050, "total_steps": 36260, "loss": 0.1364, "lr": 1.4464975179260896e-05, "epoch": 0.28957528957528955, "percentage": 2.9, "elapsed_time": "0:04:19", "remaining_time": "2:25:07", "throughput": 2474.04, "total_tokens": 642464}
+{"current_steps": 1055, "total_steps": 36260, "loss": 0.1217, "lr": 1.4533921676778819e-05, "epoch": 0.2909542195256481, "percentage": 2.91, "elapsed_time": "0:04:20", "remaining_time": "2:25:04", "throughput": 2473.38, "total_tokens": 645184}
+{"current_steps": 1060, "total_steps": 36260, "loss": 0.0617, "lr": 1.4602868174296747e-05, "epoch": 0.29233314947600664, "percentage": 2.92, "elapsed_time": "0:04:22", "remaining_time": "2:25:06", "throughput": 2473.4, "total_tokens": 648480}
+{"current_steps": 1065, "total_steps": 36260, "loss": 0.1155, "lr": 1.4671814671814674e-05, "epoch": 0.2937120794263651, "percentage": 2.94, "elapsed_time": "0:04:23", "remaining_time": "2:25:06", "throughput": 2473.55, "total_tokens": 651680}
+{"current_steps": 1070, "total_steps": 36260, "loss": 0.0829, "lr": 1.4740761169332597e-05, "epoch": 0.29509100937672367, "percentage": 2.95, "elapsed_time": "0:04:24", "remaining_time": "2:25:04", "throughput": 2473.8, "total_tokens": 654752}
+{"current_steps": 1075, "total_steps": 36260, "loss": 0.1554, "lr": 1.4809707666850526e-05, "epoch": 0.2964699393270822, "percentage": 2.96, "elapsed_time": "0:04:25", "remaining_time": "2:25:03", "throughput": 2474.23, "total_tokens": 657952}
+{"current_steps": 1080, "total_steps": 36260, "loss": 0.0438, "lr": 1.4878654164368449e-05, "epoch": 0.2978488692774407, "percentage": 2.98, "elapsed_time": "0:04:27", "remaining_time": "2:24:59", "throughput": 2474.2, "total_tokens": 660800}
+{"current_steps": 1085, "total_steps": 36260, "loss": 0.1605, "lr": 1.4947600661886377e-05, "epoch": 0.29922779922779924, "percentage": 2.99, "elapsed_time": "0:04:28", "remaining_time": "2:25:08", "throughput": 2475.12, "total_tokens": 664896}
+{"current_steps": 1090, "total_steps": 36260, "loss": 0.0205, "lr": 1.5016547159404304e-05, "epoch": 0.3006067291781577, "percentage": 3.01, "elapsed_time": "0:04:29", "remaining_time": "2:25:09", "throughput": 2474.9, "total_tokens": 668064}
+{"current_steps": 1095, "total_steps": 36260, "loss": 0.0436, "lr": 1.5085493656922229e-05, "epoch": 0.30198565912851627, "percentage": 3.02, "elapsed_time": "0:04:31", "remaining_time": "2:25:12", "throughput": 2475.45, "total_tokens": 671552}
+{"current_steps": 1100, "total_steps": 36260, "loss": 0.0727, "lr": 1.5154440154440155e-05, "epoch": 0.3033645890788748, "percentage": 3.03, "elapsed_time": "0:04:32", "remaining_time": "2:25:05", "throughput": 2474.68, "total_tokens": 674016}
+{"current_steps": 1105, "total_steps": 36260, "loss": 0.0926, "lr": 1.5223386651958082e-05, "epoch": 0.3047435190292333, "percentage": 3.05, "elapsed_time": "0:04:33", "remaining_time": "2:25:01", "throughput": 2473.82, "total_tokens": 676576}
+{"current_steps": 1110, "total_steps": 36260, "loss": 0.1329, "lr": 1.5292333149476005e-05, "epoch": 0.30612244897959184, "percentage": 3.06, "elapsed_time": "0:04:34", "remaining_time": "2:24:54", "throughput": 2473.64, "total_tokens": 679168}
+{"current_steps": 1115, "total_steps": 36260, "loss": 0.3203, "lr": 1.5361279646993934e-05, "epoch": 0.30750137892995033, "percentage": 3.08, "elapsed_time": "0:04:35", "remaining_time": "2:24:58", "throughput": 2474.4, "total_tokens": 682816}
+{"current_steps": 1120, "total_steps": 36260, "loss": 0.111, "lr": 1.543022614451186e-05, "epoch": 0.3088803088803089, "percentage": 3.09, "elapsed_time": "0:04:37", "remaining_time": "2:24:57", "throughput": 2474.81, "total_tokens": 686016}
+{"current_steps": 1125, "total_steps": 36260, "loss": 0.049, "lr": 1.5499172642029787e-05, "epoch": 0.3102592388306674, "percentage": 3.1, "elapsed_time": "0:04:38", "remaining_time": "2:24:50", "throughput": 2474.34, "total_tokens": 688480}
+{"current_steps": 1130, "total_steps": 36260, "loss": 0.0164, "lr": 1.5568119139547712e-05, "epoch": 0.3116381687810259, "percentage": 3.12, "elapsed_time": "0:04:39", "remaining_time": "2:24:48", "throughput": 2474.1, "total_tokens": 691488}
+{"current_steps": 1135, "total_steps": 36260, "loss": 0.0789, "lr": 1.5637065637065637e-05, "epoch": 0.31301709873138445, "percentage": 3.13, "elapsed_time": "0:04:40", "remaining_time": "2:24:42", "throughput": 2473.64, "total_tokens": 693984}
+{"current_steps": 1140, "total_steps": 36260, "loss": 0.212, "lr": 1.5706012134583565e-05, "epoch": 0.314396028681743, "percentage": 3.14, "elapsed_time": "0:04:41", "remaining_time": "2:24:39", "throughput": 2473.76, "total_tokens": 696960}
+{"current_steps": 1145, "total_steps": 36260, "loss": 0.0622, "lr": 1.577495863210149e-05, "epoch": 0.3157749586321015, "percentage": 3.16, "elapsed_time": "0:04:42", "remaining_time": "2:24:35", "throughput": 2473.84, "total_tokens": 699776}
+{"current_steps": 1150, "total_steps": 36260, "loss": 0.1457, "lr": 1.5843905129619415e-05, "epoch": 0.31715388858246, "percentage": 3.17, "elapsed_time": "0:04:43", "remaining_time": "2:24:28", "throughput": 2473.68, "total_tokens": 702368}
+{"current_steps": 1155, "total_steps": 36260, "loss": 0.1343, "lr": 1.5912851627137344e-05, "epoch": 0.3185328185328185, "percentage": 3.19, "elapsed_time": "0:04:45", "remaining_time": "2:24:24", "throughput": 2473.4, "total_tokens": 705120}
+{"current_steps": 1160, "total_steps": 36260, "loss": 0.1003, "lr": 1.598179812465527e-05, "epoch": 0.31991174848317705, "percentage": 3.2, "elapsed_time": "0:04:46", "remaining_time": "2:24:22", "throughput": 2473.13, "total_tokens": 708032}
+{"current_steps": 1165, "total_steps": 36260, "loss": 0.034, "lr": 1.6050744622173193e-05, "epoch": 0.3212906784335356, "percentage": 3.21, "elapsed_time": "0:04:47", "remaining_time": "2:24:19", "throughput": 2473.35, "total_tokens": 710976}
+{"current_steps": 1170, "total_steps": 36260, "loss": 0.086, "lr": 1.6119691119691122e-05, "epoch": 0.3226696083838941, "percentage": 3.23, "elapsed_time": "0:04:48", "remaining_time": "2:24:19", "throughput": 2473.69, "total_tokens": 714240}
+{"current_steps": 1175, "total_steps": 36260, "loss": 0.17, "lr": 1.6188637617209047e-05, "epoch": 0.3240485383342526, "percentage": 3.24, "elapsed_time": "0:04:49", "remaining_time": "2:24:13", "throughput": 2473.2, "total_tokens": 716736}
+{"current_steps": 1180, "total_steps": 36260, "loss": 0.14, "lr": 1.6257584114726972e-05, "epoch": 0.32542746828461117, "percentage": 3.25, "elapsed_time": "0:04:51", "remaining_time": "2:24:13", "throughput": 2473.09, "total_tokens": 719904}
+{"current_steps": 1185, "total_steps": 36260, "loss": 0.0547, "lr": 1.6326530612244897e-05, "epoch": 0.32680639823496965, "percentage": 3.27, "elapsed_time": "0:04:52", "remaining_time": "2:24:18", "throughput": 2473.66, "total_tokens": 723584}
+{"current_steps": 1190, "total_steps": 36260, "loss": 0.0943, "lr": 1.6395477109762825e-05, "epoch": 0.3281853281853282, "percentage": 3.28, "elapsed_time": "0:04:53", "remaining_time": "2:24:15", "throughput": 2473.84, "total_tokens": 726528}
+{"current_steps": 1195, "total_steps": 36260, "loss": 0.1174, "lr": 1.6464423607280753e-05, "epoch": 0.3295642581356867, "percentage": 3.3, "elapsed_time": "0:04:54", "remaining_time": "2:24:08", "throughput": 2473.02, "total_tokens": 728864}
+{"current_steps": 1200, "total_steps": 36260, "loss": 0.169, "lr": 1.6533370104798675e-05, "epoch": 0.3309431880860452, "percentage": 3.31, "elapsed_time": "0:04:55", "remaining_time": "2:24:00", "throughput": 2472.42, "total_tokens": 731232}
+{"current_steps": 1205, "total_steps": 36260, "loss": 0.089, "lr": 1.6602316602316603e-05, "epoch": 0.33232211803640377, "percentage": 3.32, "elapsed_time": "0:04:57", "remaining_time": "2:24:04", "throughput": 2473.0, "total_tokens": 734816}
+{"current_steps": 1210, "total_steps": 36260, "loss": 0.2132, "lr": 1.667126309983453e-05, "epoch": 0.33370104798676226, "percentage": 3.34, "elapsed_time": "0:04:58", "remaining_time": "2:24:03", "throughput": 2472.89, "total_tokens": 737920}
+{"current_steps": 1215, "total_steps": 36260, "loss": 0.0774, "lr": 1.6740209597352453e-05, "epoch": 0.3350799779371208, "percentage": 3.35, "elapsed_time": "0:04:59", "remaining_time": "2:23:56", "throughput": 2472.17, "total_tokens": 740224}
+{"current_steps": 1220, "total_steps": 36260, "loss": 0.1372, "lr": 1.680915609487038e-05, "epoch": 0.33645890788747934, "percentage": 3.36, "elapsed_time": "0:05:00", "remaining_time": "2:23:50", "throughput": 2471.82, "total_tokens": 742752}
+{"current_steps": 1225, "total_steps": 36260, "loss": 0.2211, "lr": 1.6878102592388306e-05, "epoch": 0.33783783783783783, "percentage": 3.38, "elapsed_time": "0:05:01", "remaining_time": "2:23:45", "throughput": 2471.75, "total_tokens": 745472}
+{"current_steps": 1230, "total_steps": 36260, "loss": 0.1056, "lr": 1.6947049089906235e-05, "epoch": 0.33921676778819637, "percentage": 3.39, "elapsed_time": "0:05:02", "remaining_time": "2:23:46", "throughput": 2471.63, "total_tokens": 748640}
+{"current_steps": 1235, "total_steps": 36260, "loss": 0.0477, "lr": 1.701599558742416e-05, "epoch": 0.34059569773855486, "percentage": 3.41, "elapsed_time": "0:05:03", "remaining_time": "2:23:39", "throughput": 2471.0, "total_tokens": 751040}
+{"current_steps": 1240, "total_steps": 36260, "loss": 0.2482, "lr": 1.7084942084942085e-05, "epoch": 0.3419746276889134, "percentage": 3.42, "elapsed_time": "0:05:05", "remaining_time": "2:23:44", "throughput": 2471.74, "total_tokens": 754784}
+{"current_steps": 1245, "total_steps": 36260, "loss": 0.048, "lr": 1.7153888582460013e-05, "epoch": 0.34335355763927194, "percentage": 3.43, "elapsed_time": "0:05:06", "remaining_time": "2:23:39", "throughput": 2471.28, "total_tokens": 757408}
+{"current_steps": 1250, "total_steps": 36260, "loss": 0.0652, "lr": 1.7222835079977938e-05, "epoch": 0.34473248758963043, "percentage": 3.45, "elapsed_time": "0:05:07", "remaining_time": "2:23:45", "throughput": 2471.51, "total_tokens": 761152}
+{"current_steps": 1255, "total_steps": 36260, "loss": 0.0411, "lr": 1.7291781577495863e-05, "epoch": 0.346111417539989, "percentage": 3.46, "elapsed_time": "0:05:09", "remaining_time": "2:23:44", "throughput": 2471.64, "total_tokens": 764288}
+{"current_steps": 1260, "total_steps": 36260, "loss": 0.1879, "lr": 1.736072807501379e-05, "epoch": 0.3474903474903475, "percentage": 3.47, "elapsed_time": "0:05:10", "remaining_time": "2:23:42", "throughput": 2471.34, "total_tokens": 767104}
+{"current_steps": 1265, "total_steps": 36260, "loss": 0.0523, "lr": 1.7429674572531716e-05, "epoch": 0.348869277440706, "percentage": 3.49, "elapsed_time": "0:05:11", "remaining_time": "2:23:46", "throughput": 2471.79, "total_tokens": 770752}
+{"current_steps": 1270, "total_steps": 36260, "loss": 0.099, "lr": 1.749862107004964e-05, "epoch": 0.35024820739106455, "percentage": 3.5, "elapsed_time": "0:05:13", "remaining_time": "2:23:55", "throughput": 2472.88, "total_tokens": 775072}
+{"current_steps": 1275, "total_steps": 36260, "loss": 0.0634, "lr": 1.756756756756757e-05, "epoch": 0.35162713734142304, "percentage": 3.52, "elapsed_time": "0:05:16", "remaining_time": "2:24:38", "throughput": 2474.29, "total_tokens": 782592}
+{"current_steps": 1280, "total_steps": 36260, "loss": 0.0453, "lr": 1.7636514065085495e-05, "epoch": 0.3530060672917816, "percentage": 3.53, "elapsed_time": "0:05:17", "remaining_time": "2:24:32", "throughput": 2474.19, "total_tokens": 785216}
+{"current_steps": 1285, "total_steps": 36260, "loss": 0.0389, "lr": 1.770546056260342e-05, "epoch": 0.3543849972421401, "percentage": 3.54, "elapsed_time": "0:05:18", "remaining_time": "2:24:33", "throughput": 2474.5, "total_tokens": 788512}
+{"current_steps": 1290, "total_steps": 36260, "loss": 0.2205, "lr": 1.7774407060121348e-05, "epoch": 0.3557639271924986, "percentage": 3.56, "elapsed_time": "0:05:19", "remaining_time": "2:24:33", "throughput": 2474.57, "total_tokens": 791712}
+{"current_steps": 1295, "total_steps": 36260, "loss": 0.0313, "lr": 1.7843353557639273e-05, "epoch": 0.35714285714285715, "percentage": 3.57, "elapsed_time": "0:05:20", "remaining_time": "2:24:26", "throughput": 2473.67, "total_tokens": 794048}
+{"current_steps": 1300, "total_steps": 36260, "loss": 0.0477, "lr": 1.79123000551572e-05, "epoch": 0.35852178709321564, "percentage": 3.59, "elapsed_time": "0:05:22", "remaining_time": "2:24:25", "throughput": 2473.5, "total_tokens": 796992}
+{"current_steps": 1305, "total_steps": 36260, "loss": 0.3161, "lr": 1.7981246552675123e-05, "epoch": 0.3599007170435742, "percentage": 3.6, "elapsed_time": "0:05:23", "remaining_time": "2:24:30", "throughput": 2473.45, "total_tokens": 800672}
+{"current_steps": 1310, "total_steps": 36260, "loss": 0.1256, "lr": 1.805019305019305e-05, "epoch": 0.3612796469939327, "percentage": 3.61, "elapsed_time": "0:05:25", "remaining_time": "2:24:31", "throughput": 2473.66, "total_tokens": 804032}
+{"current_steps": 1315, "total_steps": 36260, "loss": 0.1027, "lr": 1.811913954771098e-05, "epoch": 0.3626585769442912, "percentage": 3.63, "elapsed_time": "0:05:26", "remaining_time": "2:24:33", "throughput": 2474.15, "total_tokens": 807552}
+{"current_steps": 1320, "total_steps": 36260, "loss": 0.0386, "lr": 1.81880860452289e-05, "epoch": 0.36403750689464975, "percentage": 3.64, "elapsed_time": "0:05:27", "remaining_time": "2:24:35", "throughput": 2474.32, "total_tokens": 810976}
+{"current_steps": 1325, "total_steps": 36260, "loss": 0.2094, "lr": 1.825703254274683e-05, "epoch": 0.3654164368450083, "percentage": 3.65, "elapsed_time": "0:05:28", "remaining_time": "2:24:29", "throughput": 2473.35, "total_tokens": 813248}
+{"current_steps": 1330, "total_steps": 36260, "loss": 0.0563, "lr": 1.8325979040264754e-05, "epoch": 0.3667953667953668, "percentage": 3.67, "elapsed_time": "0:05:30", "remaining_time": "2:24:42", "throughput": 2474.88, "total_tokens": 818144}
+{"current_steps": 1335, "total_steps": 36260, "loss": 0.0854, "lr": 1.8394925537782683e-05, "epoch": 0.3681742967457253, "percentage": 3.68, "elapsed_time": "0:05:31", "remaining_time": "2:24:41", "throughput": 2474.98, "total_tokens": 821280}
+{"current_steps": 1340, "total_steps": 36260, "loss": 0.1671, "lr": 1.8463872035300608e-05, "epoch": 0.3695532266960838, "percentage": 3.7, "elapsed_time": "0:05:33", "remaining_time": "2:24:47", "throughput": 2475.78, "total_tokens": 825312}
+{"current_steps": 1345, "total_steps": 36260, "loss": 0.0915, "lr": 1.8532818532818533e-05, "epoch": 0.37093215664644236, "percentage": 3.71, "elapsed_time": "0:05:35", "remaining_time": "2:24:57", "throughput": 2476.84, "total_tokens": 829856}
+{"current_steps": 1350, "total_steps": 36260, "loss": 0.1025, "lr": 1.860176503033646e-05, "epoch": 0.3723110865968009, "percentage": 3.72, "elapsed_time": "0:05:36", "remaining_time": "2:24:54", "throughput": 2476.69, "total_tokens": 832736}
+{"current_steps": 1355, "total_steps": 36260, "loss": 0.127, "lr": 1.8670711527854386e-05, "epoch": 0.3736900165471594, "percentage": 3.74, "elapsed_time": "0:05:37", "remaining_time": "2:24:54", "throughput": 2476.63, "total_tokens": 835904}
+{"current_steps": 1360, "total_steps": 36260, "loss": 0.068, "lr": 1.873965802537231e-05, "epoch": 0.37506894649751793, "percentage": 3.75, "elapsed_time": "0:05:39", "remaining_time": "2:25:06", "throughput": 2477.45, "total_tokens": 840512}
+{"current_steps": 1365, "total_steps": 36260, "loss": 0.1928, "lr": 1.880860452289024e-05, "epoch": 0.3764478764478765, "percentage": 3.76, "elapsed_time": "0:05:40", "remaining_time": "2:25:03", "throughput": 2477.36, "total_tokens": 843456}
+{"current_steps": 1370, "total_steps": 36260, "loss": 0.0686, "lr": 1.8877551020408164e-05, "epoch": 0.37782680639823496, "percentage": 3.78, "elapsed_time": "0:05:41", "remaining_time": "2:25:01", "throughput": 2477.4, "total_tokens": 846464}
+{"current_steps": 1375, "total_steps": 36260, "loss": 0.1055, "lr": 1.894649751792609e-05, "epoch": 0.3792057363485935, "percentage": 3.79, "elapsed_time": "0:05:43", "remaining_time": "2:25:08", "throughput": 2478.09, "total_tokens": 850592}
+{"current_steps": 1380, "total_steps": 36260, "loss": 0.1617, "lr": 1.9015444015444017e-05, "epoch": 0.380584666298952, "percentage": 3.81, "elapsed_time": "0:05:44", "remaining_time": "2:25:01", "throughput": 2477.14, "total_tokens": 852768}
+{"current_steps": 1385, "total_steps": 36260, "loss": 0.1458, "lr": 1.9084390512961942e-05, "epoch": 0.38196359624931053, "percentage": 3.82, "elapsed_time": "0:05:45", "remaining_time": "2:25:03", "throughput": 2477.31, "total_tokens": 856224}
+{"current_steps": 1390, "total_steps": 36260, "loss": 0.1465, "lr": 1.9153337010479867e-05, "epoch": 0.3833425261996691, "percentage": 3.83, "elapsed_time": "0:05:46", "remaining_time": "2:24:57", "throughput": 2476.8, "total_tokens": 858688}
+{"current_steps": 1395, "total_steps": 36260, "loss": 0.0437, "lr": 1.9222283507997796e-05, "epoch": 0.38472145615002756, "percentage": 3.85, "elapsed_time": "0:05:47", "remaining_time": "2:24:54", "throughput": 2476.28, "total_tokens": 861440}
+{"current_steps": 1400, "total_steps": 36260, "loss": 0.0893, "lr": 1.929123000551572e-05, "epoch": 0.3861003861003861, "percentage": 3.86, "elapsed_time": "0:05:48", "remaining_time": "2:24:49", "throughput": 2475.72, "total_tokens": 863936}
+{"current_steps": 1405, "total_steps": 36260, "loss": 0.2189, "lr": 1.936017650303365e-05, "epoch": 0.38747931605074465, "percentage": 3.87, "elapsed_time": "0:05:50", "remaining_time": "2:24:56", "throughput": 2476.18, "total_tokens": 868032}
+{"current_steps": 1410, "total_steps": 36260, "loss": 0.1524, "lr": 1.942912300055157e-05, "epoch": 0.38885824600110314, "percentage": 3.89, "elapsed_time": "0:05:52", "remaining_time": "2:25:02", "throughput": 2476.7, "total_tokens": 872000}
+{"current_steps": 1415, "total_steps": 36260, "loss": 0.1867, "lr": 1.94980694980695e-05, "epoch": 0.3902371759514617, "percentage": 3.9, "elapsed_time": "0:05:53", "remaining_time": "2:25:04", "throughput": 2477.21, "total_tokens": 875616}
+{"current_steps": 1420, "total_steps": 36260, "loss": 0.0809, "lr": 1.9567015995587427e-05, "epoch": 0.39161610590182017, "percentage": 3.92, "elapsed_time": "0:05:54", "remaining_time": "2:25:05", "throughput": 2477.52, "total_tokens": 879072}
+{"current_steps": 1425, "total_steps": 36260, "loss": 0.3356, "lr": 1.963596249310535e-05, "epoch": 0.3929950358521787, "percentage": 3.93, "elapsed_time": "0:05:56", "remaining_time": "2:25:05", "throughput": 2477.78, "total_tokens": 882368}
+{"current_steps": 1430, "total_steps": 36260, "loss": 0.0261, "lr": 1.9704908990623277e-05, "epoch": 0.39437396580253725, "percentage": 3.94, "elapsed_time": "0:05:57", "remaining_time": "2:25:04", "throughput": 2477.85, "total_tokens": 885536}
+{"current_steps": 1435, "total_steps": 36260, "loss": 0.0582, "lr": 1.9773855488141205e-05, "epoch": 0.39575289575289574, "percentage": 3.96, "elapsed_time": "0:05:59", "remaining_time": "2:25:16", "throughput": 2478.63, "total_tokens": 890272}
+{"current_steps": 1440, "total_steps": 36260, "loss": 0.0781, "lr": 1.984280198565913e-05, "epoch": 0.3971318257032543, "percentage": 3.97, "elapsed_time": "0:06:00", "remaining_time": "2:25:15", "throughput": 2478.55, "total_tokens": 893376}
+{"current_steps": 1445, "total_steps": 36260, "loss": 0.0845, "lr": 1.9911748483177055e-05, "epoch": 0.39851075565361277, "percentage": 3.99, "elapsed_time": "0:06:02", "remaining_time": "2:25:30", "throughput": 2479.74, "total_tokens": 898528}
+{"current_steps": 1450, "total_steps": 36260, "loss": 0.2178, "lr": 1.998069498069498e-05, "epoch": 0.3998896856039713, "percentage": 4.0, "elapsed_time": "0:06:03", "remaining_time": "2:25:33", "throughput": 2480.15, "total_tokens": 902272}
+{"current_steps": 1455, "total_steps": 36260, "loss": 0.0957, "lr": 2.004964147821291e-05, "epoch": 0.40126861555432985, "percentage": 4.01, "elapsed_time": "0:06:05", "remaining_time": "2:25:37", "throughput": 2480.68, "total_tokens": 906112}
+{"current_steps": 1460, "total_steps": 36260, "loss": 0.0838, "lr": 2.0118587975730834e-05, "epoch": 0.40264754550468834, "percentage": 4.03, "elapsed_time": "0:06:06", "remaining_time": "2:25:30", "throughput": 2479.98, "total_tokens": 908384}
+{"current_steps": 1465, "total_steps": 36260, "loss": 0.1298, "lr": 2.018753447324876e-05, "epoch": 0.4040264754550469, "percentage": 4.04, "elapsed_time": "0:06:07", "remaining_time": "2:25:32", "throughput": 2480.24, "total_tokens": 911936}
+{"current_steps": 1470, "total_steps": 36260, "loss": 0.1378, "lr": 2.0256480970766687e-05, "epoch": 0.40540540540540543, "percentage": 4.05, "elapsed_time": "0:06:08", "remaining_time": "2:25:29", "throughput": 2480.33, "total_tokens": 914912}
+{"current_steps": 1475, "total_steps": 36260, "loss": 0.0387, "lr": 2.0325427468284612e-05, "epoch": 0.4067843353557639, "percentage": 4.07, "elapsed_time": "0:06:10", "remaining_time": "2:25:29", "throughput": 2480.26, "total_tokens": 918112}
+{"current_steps": 1480, "total_steps": 36260, "loss": 0.1245, "lr": 2.0394373965802537e-05, "epoch": 0.40816326530612246, "percentage": 4.08, "elapsed_time": "0:06:11", "remaining_time": "2:25:29", "throughput": 2480.43, "total_tokens": 921440}
+{"current_steps": 1485, "total_steps": 36260, "loss": 0.0727, "lr": 2.0463320463320465e-05, "epoch": 0.40954219525648095, "percentage": 4.1, "elapsed_time": "0:06:12", "remaining_time": "2:25:25", "throughput": 2479.87, "total_tokens": 923968}
+{"current_steps": 1490, "total_steps": 36260, "loss": 0.1256, "lr": 2.053226696083839e-05, "epoch": 0.4109211252068395, "percentage": 4.11, "elapsed_time": "0:06:13", "remaining_time": "2:25:22", "throughput": 2479.79, "total_tokens": 926880}
+{"current_steps": 1495, "total_steps": 36260, "loss": 0.074, "lr": 2.0601213458356315e-05, "epoch": 0.41230005515719803, "percentage": 4.12, "elapsed_time": "0:06:14", "remaining_time": "2:25:16", "throughput": 2479.57, "total_tokens": 929408}
+{"current_steps": 1500, "total_steps": 36260, "loss": 0.1072, "lr": 2.0670159955874243e-05, "epoch": 0.4136789851075565, "percentage": 4.14, "elapsed_time": "0:06:15", "remaining_time": "2:25:11", "throughput": 2479.49, "total_tokens": 932160}
+{"current_steps": 1505, "total_steps": 36260, "loss": 0.1157, "lr": 2.073910645339217e-05, "epoch": 0.41505791505791506, "percentage": 4.15, "elapsed_time": "0:06:17", "remaining_time": "2:25:07", "throughput": 2479.16, "total_tokens": 934816}
+{"current_steps": 1510, "total_steps": 36260, "loss": 0.0287, "lr": 2.0808052950910097e-05, "epoch": 0.4164368450082736, "percentage": 4.16, "elapsed_time": "0:06:18", "remaining_time": "2:25:08", "throughput": 2479.81, "total_tokens": 938368}
+{"current_steps": 1515, "total_steps": 36260, "loss": 0.0598, "lr": 2.087699944842802e-05, "epoch": 0.4178157749586321, "percentage": 4.18, "elapsed_time": "0:06:19", "remaining_time": "2:25:02", "throughput": 2479.1, "total_tokens": 940704}
+{"current_steps": 1520, "total_steps": 36260, "loss": 0.2306, "lr": 2.0945945945945947e-05, "epoch": 0.41919470490899063, "percentage": 4.19, "elapsed_time": "0:06:20", "remaining_time": "2:24:57", "throughput": 2478.73, "total_tokens": 943264}
+{"current_steps": 1525, "total_steps": 36260, "loss": 0.0371, "lr": 2.1014892443463875e-05, "epoch": 0.4205736348593491, "percentage": 4.21, "elapsed_time": "0:06:21", "remaining_time": "2:24:55", "throughput": 2478.47, "total_tokens": 946144}
+{"current_steps": 1530, "total_steps": 36260, "loss": 0.2095, "lr": 2.1083838940981797e-05, "epoch": 0.42195256480970766, "percentage": 4.22, "elapsed_time": "0:06:23", "remaining_time": "2:24:58", "throughput": 2479.16, "total_tokens": 950016}
+{"current_steps": 1535, "total_steps": 36260, "loss": 0.1734, "lr": 2.1152785438499725e-05, "epoch": 0.4233314947600662, "percentage": 4.23, "elapsed_time": "0:06:24", "remaining_time": "2:25:00", "throughput": 2479.65, "total_tokens": 953728}
+{"current_steps": 1540, "total_steps": 36260, "loss": 0.1515, "lr": 2.1221731936017653e-05, "epoch": 0.4247104247104247, "percentage": 4.25, "elapsed_time": "0:06:25", "remaining_time": "2:24:56", "throughput": 2479.2, "total_tokens": 956288}
+{"current_steps": 1545, "total_steps": 36260, "loss": 0.0931, "lr": 2.1290678433535578e-05, "epoch": 0.42608935466078324, "percentage": 4.26, "elapsed_time": "0:06:26", "remaining_time": "2:24:51", "throughput": 2478.65, "total_tokens": 958752}
+{"current_steps": 1550, "total_steps": 36260, "loss": 0.1013, "lr": 2.1359624931053503e-05, "epoch": 0.4274682846111418, "percentage": 4.27, "elapsed_time": "0:06:27", "remaining_time": "2:24:44", "throughput": 2477.9, "total_tokens": 960960}
+{"current_steps": 1555, "total_steps": 36260, "loss": 0.1027, "lr": 2.1428571428571428e-05, "epoch": 0.42884721456150027, "percentage": 4.29, "elapsed_time": "0:06:29", "remaining_time": "2:24:45", "throughput": 2478.21, "total_tokens": 964480}
+{"current_steps": 1560, "total_steps": 36260, "loss": 0.0457, "lr": 2.1497517926089356e-05, "epoch": 0.4302261445118588, "percentage": 4.3, "elapsed_time": "0:06:30", "remaining_time": "2:24:49", "throughput": 2478.74, "total_tokens": 968320}
+{"current_steps": 1565, "total_steps": 36260, "loss": 0.0381, "lr": 2.156646442360728e-05, "epoch": 0.4316050744622173, "percentage": 4.32, "elapsed_time": "0:06:32", "remaining_time": "2:24:50", "throughput": 2479.05, "total_tokens": 971808}
+{"current_steps": 1570, "total_steps": 36260, "loss": 0.1548, "lr": 2.1635410921125206e-05, "epoch": 0.43298400441257584, "percentage": 4.33, "elapsed_time": "0:06:33", "remaining_time": "2:24:48", "throughput": 2479.23, "total_tokens": 974944}
+{"current_steps": 1575, "total_steps": 36260, "loss": 0.1647, "lr": 2.1704357418643135e-05, "epoch": 0.4343629343629344, "percentage": 4.34, "elapsed_time": "0:06:34", "remaining_time": "2:24:54", "throughput": 2479.69, "total_tokens": 979040}
+{"current_steps": 1580, "total_steps": 36260, "loss": 0.1813, "lr": 2.177330391616106e-05, "epoch": 0.43574186431329287, "percentage": 4.36, "elapsed_time": "0:06:35", "remaining_time": "2:24:50", "throughput": 2479.13, "total_tokens": 981536}
+{"current_steps": 1585, "total_steps": 36260, "loss": 0.0513, "lr": 2.1842250413678985e-05, "epoch": 0.4371207942636514, "percentage": 4.37, "elapsed_time": "0:06:37", "remaining_time": "2:24:47", "throughput": 2479.11, "total_tokens": 984448}
+{"current_steps": 1590, "total_steps": 36260, "loss": 0.0778, "lr": 2.1911196911196913e-05, "epoch": 0.43849972421400996, "percentage": 4.38, "elapsed_time": "0:06:38", "remaining_time": "2:24:43", "throughput": 2479.09, "total_tokens": 987264}
+{"current_steps": 1595, "total_steps": 36260, "loss": 0.1787, "lr": 2.1980143408714838e-05, "epoch": 0.43987865416436844, "percentage": 4.4, "elapsed_time": "0:06:39", "remaining_time": "2:24:44", "throughput": 2479.34, "total_tokens": 990752}
+{"current_steps": 1600, "total_steps": 36260, "loss": 0.1242, "lr": 2.2049089906232763e-05, "epoch": 0.441257584114727, "percentage": 4.41, "elapsed_time": "0:06:40", "remaining_time": "2:24:39", "throughput": 2478.95, "total_tokens": 993184}
+{"current_steps": 1605, "total_steps": 36260, "loss": 0.0137, "lr": 2.211803640375069e-05, "epoch": 0.4426365140650855, "percentage": 4.43, "elapsed_time": "0:06:42", "remaining_time": "2:24:40", "throughput": 2479.13, "total_tokens": 996640}
+{"current_steps": 1610, "total_steps": 36260, "loss": 0.0736, "lr": 2.2186982901268616e-05, "epoch": 0.444015444015444, "percentage": 4.44, "elapsed_time": "0:06:42", "remaining_time": "2:24:32", "throughput": 2478.09, "total_tokens": 998624}
+{"current_steps": 1615, "total_steps": 36260, "loss": 0.1704, "lr": 2.2255929398786545e-05, "epoch": 0.44539437396580256, "percentage": 4.45, "elapsed_time": "0:06:44", "remaining_time": "2:24:34", "throughput": 2478.3, "total_tokens": 1002176}
+{"current_steps": 1620, "total_steps": 36260, "loss": 0.249, "lr": 2.232487589630447e-05, "epoch": 0.44677330391616105, "percentage": 4.47, "elapsed_time": "0:06:45", "remaining_time": "2:24:31", "throughput": 2478.15, "total_tokens": 1005024}
+{"current_steps": 1625, "total_steps": 36260, "loss": 0.1029, "lr": 2.2393822393822394e-05, "epoch": 0.4481522338665196, "percentage": 4.48, "elapsed_time": "0:06:46", "remaining_time": "2:24:29", "throughput": 2477.95, "total_tokens": 1007904}
+{"current_steps": 1630, "total_steps": 36260, "loss": 0.1166, "lr": 2.2462768891340323e-05, "epoch": 0.4495311638168781, "percentage": 4.5, "elapsed_time": "0:06:47", "remaining_time": "2:24:23", "throughput": 2477.63, "total_tokens": 1010336}
+{"current_steps": 1635, "total_steps": 36260, "loss": 0.0395, "lr": 2.2531715388858244e-05, "epoch": 0.4509100937672366, "percentage": 4.51, "elapsed_time": "0:06:49", "remaining_time": "2:24:21", "throughput": 2477.5, "total_tokens": 1013312}
+{"current_steps": 1640, "total_steps": 36260, "loss": 0.1431, "lr": 2.2600661886376173e-05, "epoch": 0.45228902371759516, "percentage": 4.52, "elapsed_time": "0:06:50", "remaining_time": "2:24:21", "throughput": 2477.66, "total_tokens": 1016640}
+{"current_steps": 1645, "total_steps": 36260, "loss": 0.0781, "lr": 2.26696083838941e-05, "epoch": 0.45366795366795365, "percentage": 4.54, "elapsed_time": "0:06:51", "remaining_time": "2:24:16", "throughput": 2477.27, "total_tokens": 1019136}
+{"current_steps": 1650, "total_steps": 36260, "loss": 0.1515, "lr": 2.2738554881412026e-05, "epoch": 0.4550468836183122, "percentage": 4.55, "elapsed_time": "0:06:52", "remaining_time": "2:24:10", "throughput": 2476.87, "total_tokens": 1021504}
+{"current_steps": 1655, "total_steps": 36260, "loss": 0.006, "lr": 2.280750137892995e-05, "epoch": 0.45642581356867074, "percentage": 4.56, "elapsed_time": "0:06:53", "remaining_time": "2:24:08", "throughput": 2476.92, "total_tokens": 1024544}
+{"current_steps": 1660, "total_steps": 36260, "loss": 0.1076, "lr": 2.2876447876447876e-05, "epoch": 0.4578047435190292, "percentage": 4.58, "elapsed_time": "0:06:54", "remaining_time": "2:24:05", "throughput": 2476.73, "total_tokens": 1027360}
+{"current_steps": 1665, "total_steps": 36260, "loss": 0.1448, "lr": 2.2945394373965804e-05, "epoch": 0.45918367346938777, "percentage": 4.59, "elapsed_time": "0:06:56", "remaining_time": "2:24:06", "throughput": 2476.77, "total_tokens": 1030656}
+{"current_steps": 1670, "total_steps": 36260, "loss": 0.2095, "lr": 2.301434087148373e-05, "epoch": 0.46056260341974625, "percentage": 4.61, "elapsed_time": "0:06:57", "remaining_time": "2:24:00", "throughput": 2476.21, "total_tokens": 1033024}
+{"current_steps": 1675, "total_steps": 36260, "loss": 0.2896, "lr": 2.3083287369001654e-05, "epoch": 0.4619415333701048, "percentage": 4.62, "elapsed_time": "0:06:58", "remaining_time": "2:23:55", "throughput": 2475.73, "total_tokens": 1035424}
+{"current_steps": 1680, "total_steps": 36260, "loss": 0.0414, "lr": 2.3152233866519583e-05, "epoch": 0.46332046332046334, "percentage": 4.63, "elapsed_time": "0:06:59", "remaining_time": "2:23:51", "throughput": 2475.54, "total_tokens": 1038112}
+{"current_steps": 1685, "total_steps": 36260, "loss": 0.1063, "lr": 2.3221180364037508e-05, "epoch": 0.4646993932708218, "percentage": 4.65, "elapsed_time": "0:07:00", "remaining_time": "2:23:53", "throughput": 2476.09, "total_tokens": 1041856}
+{"current_steps": 1690, "total_steps": 36260, "loss": 0.0757, "lr": 2.3290126861555432e-05, "epoch": 0.46607832322118037, "percentage": 4.66, "elapsed_time": "0:07:01", "remaining_time": "2:23:47", "throughput": 2475.6, "total_tokens": 1044128}
+{"current_steps": 1695, "total_steps": 36260, "loss": 0.0896, "lr": 2.335907335907336e-05, "epoch": 0.4674572531715389, "percentage": 4.67, "elapsed_time": "0:07:03", "remaining_time": "2:23:46", "throughput": 2475.6, "total_tokens": 1047264}
+{"current_steps": 1700, "total_steps": 36260, "loss": 0.071, "lr": 2.3428019856591286e-05, "epoch": 0.4688361831218974, "percentage": 4.69, "elapsed_time": "0:07:04", "remaining_time": "2:23:49", "throughput": 2476.12, "total_tokens": 1051072}
+{"current_steps": 1705, "total_steps": 36260, "loss": 0.0073, "lr": 2.349696635410921e-05, "epoch": 0.47021511307225594, "percentage": 4.7, "elapsed_time": "0:07:05", "remaining_time": "2:23:46", "throughput": 2476.03, "total_tokens": 1053920}
+{"current_steps": 1710, "total_steps": 36260, "loss": 0.2374, "lr": 2.356591285162714e-05, "epoch": 0.47159404302261443, "percentage": 4.72, "elapsed_time": "0:07:06", "remaining_time": "2:23:46", "throughput": 2475.91, "total_tokens": 1057056}
+{"current_steps": 1715, "total_steps": 36260, "loss": 0.1277, "lr": 2.3634859349145064e-05, "epoch": 0.47297297297297297, "percentage": 4.73, "elapsed_time": "0:07:08", "remaining_time": "2:23:44", "throughput": 2476.11, "total_tokens": 1060128}
+{"current_steps": 1720, "total_steps": 36260, "loss": 0.083, "lr": 2.3703805846662992e-05, "epoch": 0.4743519029233315, "percentage": 4.74, "elapsed_time": "0:07:09", "remaining_time": "2:23:38", "throughput": 2475.76, "total_tokens": 1062528}
+{"current_steps": 1725, "total_steps": 36260, "loss": 0.07, "lr": 2.3772752344180917e-05, "epoch": 0.47573083287369, "percentage": 4.76, "elapsed_time": "0:07:10", "remaining_time": "2:23:35", "throughput": 2475.63, "total_tokens": 1065344}
+{"current_steps": 1730, "total_steps": 36260, "loss": 0.0535, "lr": 2.3841698841698842e-05, "epoch": 0.47710976282404854, "percentage": 4.77, "elapsed_time": "0:07:11", "remaining_time": "2:23:34", "throughput": 2475.48, "total_tokens": 1068352}
+{"current_steps": 1735, "total_steps": 36260, "loss": 0.012, "lr": 2.391064533921677e-05, "epoch": 0.4784886927744071, "percentage": 4.78, "elapsed_time": "0:07:12", "remaining_time": "2:23:36", "throughput": 2475.54, "total_tokens": 1071872}
+{"current_steps": 1740, "total_steps": 36260, "loss": 0.1598, "lr": 2.3979591836734696e-05, "epoch": 0.4798676227247656, "percentage": 4.8, "elapsed_time": "0:07:14", "remaining_time": "2:23:34", "throughput": 2475.5, "total_tokens": 1074848}
+{"current_steps": 1745, "total_steps": 36260, "loss": 0.1561, "lr": 2.404853833425262e-05, "epoch": 0.4812465526751241, "percentage": 4.81, "elapsed_time": "0:07:15", "remaining_time": "2:23:30", "throughput": 2475.45, "total_tokens": 1077600}
+{"current_steps": 1750, "total_steps": 36260, "loss": 0.1743, "lr": 2.411748483177055e-05, "epoch": 0.4826254826254826, "percentage": 4.83, "elapsed_time": "0:07:16", "remaining_time": "2:23:29", "throughput": 2475.5, "total_tokens": 1080768}
+{"current_steps": 1755, "total_steps": 36260, "loss": 0.1345, "lr": 2.4186431329288474e-05, "epoch": 0.48400441257584115, "percentage": 4.84, "elapsed_time": "0:07:17", "remaining_time": "2:23:26", "throughput": 2475.57, "total_tokens": 1083680}
+{"current_steps": 1760, "total_steps": 36260, "loss": 0.1239, "lr": 2.42553778268064e-05, "epoch": 0.4853833425261997, "percentage": 4.85, "elapsed_time": "0:07:18", "remaining_time": "2:23:22", "throughput": 2475.07, "total_tokens": 1086240}
+{"current_steps": 1765, "total_steps": 36260, "loss": 0.192, "lr": 2.4324324324324327e-05, "epoch": 0.4867622724765582, "percentage": 4.87, "elapsed_time": "0:07:19", "remaining_time": "2:23:19", "throughput": 2474.75, "total_tokens": 1088864}
+{"current_steps": 1770, "total_steps": 36260, "loss": 0.1495, "lr": 2.4393270821842252e-05, "epoch": 0.4881412024269167, "percentage": 4.88, "elapsed_time": "0:07:21", "remaining_time": "2:23:17", "throughput": 2474.81, "total_tokens": 1091968}
+{"current_steps": 1775, "total_steps": 36260, "loss": 0.136, "lr": 2.4462217319360177e-05, "epoch": 0.4895201323772752, "percentage": 4.9, "elapsed_time": "0:07:22", "remaining_time": "2:23:15", "throughput": 2474.96, "total_tokens": 1095040}
+{"current_steps": 1780, "total_steps": 36260, "loss": 0.1546, "lr": 2.4531163816878102e-05, "epoch": 0.49089906232763375, "percentage": 4.91, "elapsed_time": "0:07:23", "remaining_time": "2:23:15", "throughput": 2475.07, "total_tokens": 1098240}
+{"current_steps": 1785, "total_steps": 36260, "loss": 0.2136, "lr": 2.460011031439603e-05, "epoch": 0.4922779922779923, "percentage": 4.92, "elapsed_time": "0:07:24", "remaining_time": "2:23:11", "throughput": 2474.54, "total_tokens": 1100768}
+{"current_steps": 1790, "total_steps": 36260, "loss": 0.0961, "lr": 2.466905681191396e-05, "epoch": 0.4936569222283508, "percentage": 4.94, "elapsed_time": "0:07:25", "remaining_time": "2:23:06", "throughput": 2474.31, "total_tokens": 1103264}
+{"current_steps": 1795, "total_steps": 36260, "loss": 0.054, "lr": 2.473800330943188e-05, "epoch": 0.4950358521787093, "percentage": 4.95, "elapsed_time": "0:07:27", "remaining_time": "2:23:09", "throughput": 2474.55, "total_tokens": 1107008}
+{"current_steps": 1800, "total_steps": 36260, "loss": 0.1891, "lr": 2.480694980694981e-05, "epoch": 0.49641478212906787, "percentage": 4.96, "elapsed_time": "0:07:28", "remaining_time": "2:23:05", "throughput": 2474.49, "total_tokens": 1109696}
+{"current_steps": 1805, "total_steps": 36260, "loss": 0.0502, "lr": 2.4875896304467734e-05, "epoch": 0.49779371207942635, "percentage": 4.98, "elapsed_time": "0:07:29", "remaining_time": "2:23:05", "throughput": 2474.74, "total_tokens": 1113120}
+{"current_steps": 1810, "total_steps": 36260, "loss": 0.1198, "lr": 2.494484280198566e-05, "epoch": 0.4991726420297849, "percentage": 4.99, "elapsed_time": "0:07:31", "remaining_time": "2:23:06", "throughput": 2475.34, "total_tokens": 1116768}
+{"current_steps": 1813, "total_steps": 36260, "eval_loss": 0.10147319734096527, "epoch": 0.5, "percentage": 5.0, "elapsed_time": "0:08:00", "remaining_time": "2:32:06", "throughput": 2328.26, "total_tokens": 1118368}
+{"current_steps": 1815, "total_steps": 36260, "loss": 0.009, "lr": 2.5013789299503587e-05, "epoch": 0.5005515719801434, "percentage": 5.01, "elapsed_time": "0:08:02", "remaining_time": "2:32:29", "throughput": 2322.77, "total_tokens": 1119808}
+{"current_steps": 1820, "total_steps": 36260, "loss": 0.1332, "lr": 2.5082735797021512e-05, "epoch": 0.5019305019305019, "percentage": 5.02, "elapsed_time": "0:08:03", "remaining_time": "2:32:27", "throughput": 2323.5, "total_tokens": 1123232}
+{"current_steps": 1825, "total_steps": 36260, "loss": 0.0622, "lr": 2.5151682294539437e-05, "epoch": 0.5033094318808604, "percentage": 5.03, "elapsed_time": "0:08:04", "remaining_time": "2:32:25", "throughput": 2324.32, "total_tokens": 1126560}
+{"current_steps": 1830, "total_steps": 36260, "loss": 0.0559, "lr": 2.5220628792057365e-05, "epoch": 0.504688361831219, "percentage": 5.05, "elapsed_time": "0:08:05", "remaining_time": "2:32:20", "throughput": 2324.51, "total_tokens": 1129280}
+{"current_steps": 1835, "total_steps": 36260, "loss": 0.0881, "lr": 2.528957528957529e-05, "epoch": 0.5060672917815775, "percentage": 5.06, "elapsed_time": "0:08:06", "remaining_time": "2:32:15", "throughput": 2324.7, "total_tokens": 1132000}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ebfd06295ea5a7d48aaa7186ec8728b6d3bfa76d39a8ad4d9a5e66dd98dff98
+size 6225