last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.007172084042113006,
  "eval_steps": 10,
  "global_step": 39,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00018389959082341041,
      "eval_loss": 1.2216368913650513,
      "eval_runtime": 139.2415,
      "eval_samples_per_second": 16.446,
      "eval_steps_per_second": 8.223,
      "step": 1
    },
    {
      "epoch": 0.0009194979541170521,
      "grad_norm": 0.311225950717926,
      "learning_rate": 5e-05,
      "loss": 1.2753,
      "step": 5
    },
    {
      "epoch": 0.0018389959082341042,
      "grad_norm": 0.36874568462371826,
      "learning_rate": 0.0001,
      "loss": 1.1951,
      "step": 10
    },
    {
      "epoch": 0.0018389959082341042,
      "eval_loss": 1.1231805086135864,
      "eval_runtime": 135.6701,
      "eval_samples_per_second": 16.879,
      "eval_steps_per_second": 8.44,
      "step": 10
    },
    {
      "epoch": 0.002758493862351156,
      "grad_norm": 0.3702358901500702,
      "learning_rate": 9.619397662556435e-05,
      "loss": 1.005,
      "step": 15
    },
    {
      "epoch": 0.0036779918164682084,
      "grad_norm": 0.4076727330684662,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.8239,
      "step": 20
    },
    {
      "epoch": 0.0036779918164682084,
      "eval_loss": 0.709395706653595,
      "eval_runtime": 135.516,
      "eval_samples_per_second": 16.898,
      "eval_steps_per_second": 8.449,
      "step": 20
    },
    {
      "epoch": 0.00459748977058526,
      "grad_norm": 1.03678560256958,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.6475,
      "step": 25
    },
    {
      "epoch": 0.005516987724702312,
      "grad_norm": 0.556086540222168,
      "learning_rate": 5e-05,
      "loss": 0.4812,
      "step": 30
    },
    {
      "epoch": 0.005516987724702312,
      "eval_loss": 0.4218650162220001,
      "eval_runtime": 135.24,
      "eval_samples_per_second": 16.933,
      "eval_steps_per_second": 8.466,
      "step": 30
    },
    {
      "epoch": 0.006436485678819365,
      "grad_norm": 0.5369921326637268,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.403,
      "step": 35
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 13,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5412321188904960.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}