last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.005128205128205128,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005128205128205128,
      "grad_norm": 28.866498947143555,
      "learning_rate": 2e-05,
      "loss": 22.3345,
      "step": 1
    },
    {
      "epoch": 0.0005128205128205128,
      "eval_loss": 4.872819423675537,
      "eval_runtime": 12.6874,
      "eval_samples_per_second": 64.788,
      "eval_steps_per_second": 32.394,
      "step": 1
    },
    {
      "epoch": 0.0010256410256410256,
      "grad_norm": 26.673473358154297,
      "learning_rate": 4e-05,
      "loss": 20.8146,
      "step": 2
    },
    {
      "epoch": 0.0015384615384615385,
      "grad_norm": 24.699934005737305,
      "learning_rate": 6e-05,
      "loss": 19.8149,
      "step": 3
    },
    {
      "epoch": 0.0015384615384615385,
      "eval_loss": 4.819361686706543,
      "eval_runtime": 12.0615,
      "eval_samples_per_second": 68.151,
      "eval_steps_per_second": 34.075,
      "step": 3
    },
    {
      "epoch": 0.0020512820512820513,
      "grad_norm": 19.96166229248047,
      "learning_rate": 8e-05,
      "loss": 19.845,
      "step": 4
    },
    {
      "epoch": 0.002564102564102564,
      "grad_norm": 14.74158763885498,
      "learning_rate": 0.0001,
      "loss": 15.1953,
      "step": 5
    },
    {
      "epoch": 0.003076923076923077,
      "grad_norm": 23.97884178161621,
      "learning_rate": 0.00012,
      "loss": 18.7039,
      "step": 6
    },
    {
      "epoch": 0.003076923076923077,
      "eval_loss": 4.507215976715088,
      "eval_runtime": 12.1994,
      "eval_samples_per_second": 67.38,
      "eval_steps_per_second": 33.69,
      "step": 6
    },
    {
      "epoch": 0.0035897435897435897,
      "grad_norm": 28.07773780822754,
      "learning_rate": 0.00014,
      "loss": 20.8542,
      "step": 7
    },
    {
      "epoch": 0.0041025641025641026,
      "grad_norm": 24.7640323638916,
      "learning_rate": 0.00016,
      "loss": 17.5828,
      "step": 8
    },
    {
      "epoch": 0.004615384615384616,
      "grad_norm": 29.232324600219727,
      "learning_rate": 0.00018,
      "loss": 18.9759,
      "step": 9
    },
    {
      "epoch": 0.004615384615384616,
      "eval_loss": 3.9376187324523926,
      "eval_runtime": 12.1491,
      "eval_samples_per_second": 67.659,
      "eval_steps_per_second": 33.83,
      "step": 9
    },
    {
      "epoch": 0.005128205128205128,
      "grad_norm": 43.22773361206055,
      "learning_rate": 0.0002,
      "loss": 18.0903,
      "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": 298765642629120.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}