trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9859154929577465,
  "eval_steps": 500,
  "global_step": 159,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.18779342723004694,
      "grad_norm": 1.2853285209837433,
      "learning_rate": 5e-06,
      "loss": 0.7828,
      "step": 10
    },
    {
      "epoch": 0.3755868544600939,
      "grad_norm": 1.5703679315137147,
      "learning_rate": 5e-06,
      "loss": 0.6839,
      "step": 20
    },
    {
      "epoch": 0.5633802816901409,
      "grad_norm": 0.8050661506952912,
      "learning_rate": 5e-06,
      "loss": 0.6552,
      "step": 30
    },
    {
      "epoch": 0.7511737089201878,
      "grad_norm": 0.7119286963130127,
      "learning_rate": 5e-06,
      "loss": 0.6355,
      "step": 40
    },
    {
      "epoch": 0.9389671361502347,
      "grad_norm": 0.7729364900144268,
      "learning_rate": 5e-06,
      "loss": 0.6241,
      "step": 50
    },
    {
      "epoch": 0.9953051643192489,
      "eval_loss": 0.6091234683990479,
      "eval_runtime": 58.3471,
      "eval_samples_per_second": 24.577,
      "eval_steps_per_second": 0.394,
      "step": 53
    },
    {
      "epoch": 1.1267605633802817,
      "grad_norm": 0.8140362088385064,
      "learning_rate": 5e-06,
      "loss": 0.6425,
      "step": 60
    },
    {
      "epoch": 1.3145539906103285,
      "grad_norm": 0.8133655444184705,
      "learning_rate": 5e-06,
      "loss": 0.5674,
      "step": 70
    },
    {
      "epoch": 1.5023474178403755,
      "grad_norm": 0.6492306561049654,
      "learning_rate": 5e-06,
      "loss": 0.5654,
      "step": 80
    },
    {
      "epoch": 1.6901408450704225,
      "grad_norm": 0.7508356214374524,
      "learning_rate": 5e-06,
      "loss": 0.5662,
      "step": 90
    },
    {
      "epoch": 1.8779342723004695,
      "grad_norm": 1.39871277191206,
      "learning_rate": 5e-06,
      "loss": 0.5597,
      "step": 100
    },
    {
      "epoch": 1.9906103286384975,
      "eval_loss": 0.5878855586051941,
      "eval_runtime": 58.1368,
      "eval_samples_per_second": 24.666,
      "eval_steps_per_second": 0.396,
      "step": 106
    },
    {
      "epoch": 2.0657276995305165,
      "grad_norm": 1.0731318648601569,
      "learning_rate": 5e-06,
      "loss": 0.5894,
      "step": 110
    },
    {
      "epoch": 2.2535211267605635,
      "grad_norm": 0.7948806860624125,
      "learning_rate": 5e-06,
      "loss": 0.5112,
      "step": 120
    },
    {
      "epoch": 2.4413145539906105,
      "grad_norm": 0.7896383757774875,
      "learning_rate": 5e-06,
      "loss": 0.5115,
      "step": 130
    },
    {
      "epoch": 2.629107981220657,
      "grad_norm": 1.107201714560591,
      "learning_rate": 5e-06,
      "loss": 0.5121,
      "step": 140
    },
    {
      "epoch": 2.816901408450704,
      "grad_norm": 0.8878715217566517,
      "learning_rate": 5e-06,
      "loss": 0.513,
      "step": 150
    },
    {
      "epoch": 2.9859154929577465,
      "eval_loss": 0.5900523066520691,
      "eval_runtime": 57.5361,
      "eval_samples_per_second": 24.923,
      "eval_steps_per_second": 0.4,
      "step": 159
    },
    {
      "epoch": 2.9859154929577465,
      "step": 159,
      "total_flos": 266121542369280.0,
      "train_loss": 0.5896026683303545,
      "train_runtime": 9662.9595,
      "train_samples_per_second": 8.459,
      "train_steps_per_second": 0.016
    }
  ],
  "logging_steps": 10,
  "max_steps": 159,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 266121542369280.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}