last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.014776505356483192,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0014776505356483192,
      "grad_norm": 1.2122458219528198,
      "learning_rate": 2e-05,
      "loss": 2.4179,
      "step": 1
    },
    {
      "epoch": 0.0014776505356483192,
      "eval_loss": 2.4440231323242188,
      "eval_runtime": 53.951,
      "eval_samples_per_second": 5.283,
      "eval_steps_per_second": 2.651,
      "step": 1
    },
    {
      "epoch": 0.0029553010712966383,
      "grad_norm": 0.8069053292274475,
      "learning_rate": 4e-05,
      "loss": 2.2724,
      "step": 2
    },
    {
      "epoch": 0.004432951606944958,
      "grad_norm": 1.7609020471572876,
      "learning_rate": 6e-05,
      "loss": 2.5391,
      "step": 3
    },
    {
      "epoch": 0.004432951606944958,
      "eval_loss": 2.4400010108947754,
      "eval_runtime": 54.3025,
      "eval_samples_per_second": 5.248,
      "eval_steps_per_second": 2.633,
      "step": 3
    },
    {
      "epoch": 0.005910602142593277,
      "grad_norm": 0.7547000646591187,
      "learning_rate": 8e-05,
      "loss": 2.3439,
      "step": 4
    },
    {
      "epoch": 0.007388252678241596,
      "grad_norm": 0.8642814755439758,
      "learning_rate": 0.0001,
      "loss": 2.4535,
      "step": 5
    },
    {
      "epoch": 0.008865903213889916,
      "grad_norm": 0.7741430997848511,
      "learning_rate": 0.00012,
      "loss": 2.2524,
      "step": 6
    },
    {
      "epoch": 0.008865903213889916,
      "eval_loss": 2.4153623580932617,
      "eval_runtime": 54.5172,
      "eval_samples_per_second": 5.228,
      "eval_steps_per_second": 2.623,
      "step": 6
    },
    {
      "epoch": 0.010343553749538234,
      "grad_norm": 0.8248753547668457,
      "learning_rate": 0.00014,
      "loss": 2.2024,
      "step": 7
    },
    {
      "epoch": 0.011821204285186553,
      "grad_norm": 0.8654381036758423,
      "learning_rate": 0.00016,
      "loss": 2.4941,
      "step": 8
    },
    {
      "epoch": 0.013298854820834873,
      "grad_norm": 0.7903196811676025,
      "learning_rate": 0.00018,
      "loss": 2.3585,
      "step": 9
    },
    {
      "epoch": 0.013298854820834873,
      "eval_loss": 2.3736534118652344,
      "eval_runtime": 54.3524,
      "eval_samples_per_second": 5.244,
      "eval_steps_per_second": 2.631,
      "step": 9
    },
    {
      "epoch": 0.014776505356483192,
      "grad_norm": 0.747567355632782,
      "learning_rate": 0.0002,
      "loss": 2.3932,
      "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": 2545835153817600.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}