last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0001894692965005021,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.894692965005021e-05,
      "grad_norm": 0.5058638453483582,
      "learning_rate": 2e-05,
      "loss": 0.2283,
      "step": 1
    },
    {
      "epoch": 1.894692965005021e-05,
      "eval_loss": 0.1918485462665558,
      "eval_runtime": 767.0599,
      "eval_samples_per_second": 28.972,
      "eval_steps_per_second": 14.486,
      "step": 1
    },
    {
      "epoch": 3.789385930010042e-05,
      "grad_norm": 0.24720197916030884,
      "learning_rate": 4e-05,
      "loss": 0.132,
      "step": 2
    },
    {
      "epoch": 5.684078895015063e-05,
      "grad_norm": 0.3961828052997589,
      "learning_rate": 6e-05,
      "loss": 0.1826,
      "step": 3
    },
    {
      "epoch": 5.684078895015063e-05,
      "eval_loss": 0.1860792487859726,
      "eval_runtime": 767.3047,
      "eval_samples_per_second": 28.962,
      "eval_steps_per_second": 14.482,
      "step": 3
    },
    {
      "epoch": 7.578771860020084e-05,
      "grad_norm": 0.36349087953567505,
      "learning_rate": 8e-05,
      "loss": 0.1655,
      "step": 4
    },
    {
      "epoch": 9.473464825025105e-05,
      "grad_norm": 0.37379804253578186,
      "learning_rate": 0.0001,
      "loss": 0.1599,
      "step": 5
    },
    {
      "epoch": 0.00011368157790030126,
      "grad_norm": 0.37054234743118286,
      "learning_rate": 0.00012,
      "loss": 0.1353,
      "step": 6
    },
    {
      "epoch": 0.00011368157790030126,
      "eval_loss": 0.11403009295463562,
      "eval_runtime": 767.1933,
      "eval_samples_per_second": 28.967,
      "eval_steps_per_second": 14.484,
      "step": 6
    },
    {
      "epoch": 0.00013262850755035147,
      "grad_norm": 0.33067765831947327,
      "learning_rate": 0.00014,
      "loss": 0.0919,
      "step": 7
    },
    {
      "epoch": 0.00015157543720040168,
      "grad_norm": 0.26524150371551514,
      "learning_rate": 0.00016,
      "loss": 0.0832,
      "step": 8
    },
    {
      "epoch": 0.0001705223668504519,
      "grad_norm": 0.21210353076457977,
      "learning_rate": 0.00018,
      "loss": 0.0591,
      "step": 9
    },
    {
      "epoch": 0.0001705223668504519,
      "eval_loss": 0.05260061100125313,
      "eval_runtime": 766.1969,
      "eval_samples_per_second": 29.004,
      "eval_steps_per_second": 14.503,
      "step": 9
    },
    {
      "epoch": 0.0001894692965005021,
      "grad_norm": 0.2930048108100891,
      "learning_rate": 0.0002,
      "loss": 0.0468,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 862667895472128.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}