last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.47619047619047616,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.047619047619047616,
      "grad_norm": 3.609731912612915,
      "learning_rate": 2e-05,
      "loss": 3.6344,
      "step": 1
    },
    {
      "epoch": 0.047619047619047616,
      "eval_loss": 4.260468006134033,
      "eval_runtime": 1.7036,
      "eval_samples_per_second": 5.283,
      "eval_steps_per_second": 2.935,
      "step": 1
    },
    {
      "epoch": 0.09523809523809523,
      "grad_norm": 3.572953462600708,
      "learning_rate": 4e-05,
      "loss": 3.1556,
      "step": 2
    },
    {
      "epoch": 0.14285714285714285,
      "grad_norm": 3.6638119220733643,
      "learning_rate": 6e-05,
      "loss": 4.9191,
      "step": 3
    },
    {
      "epoch": 0.14285714285714285,
      "eval_loss": 4.2523651123046875,
      "eval_runtime": 0.3317,
      "eval_samples_per_second": 27.13,
      "eval_steps_per_second": 15.072,
      "step": 3
    },
    {
      "epoch": 0.19047619047619047,
      "grad_norm": 4.614552974700928,
      "learning_rate": 8e-05,
      "loss": 4.0998,
      "step": 4
    },
    {
      "epoch": 0.23809523809523808,
      "grad_norm": 3.882472276687622,
      "learning_rate": 0.0001,
      "loss": 4.1582,
      "step": 5
    },
    {
      "epoch": 0.2857142857142857,
      "grad_norm": 4.788486957550049,
      "learning_rate": 0.00012,
      "loss": 4.7095,
      "step": 6
    },
    {
      "epoch": 0.2857142857142857,
      "eval_loss": 3.967926025390625,
      "eval_runtime": 0.3377,
      "eval_samples_per_second": 26.647,
      "eval_steps_per_second": 14.804,
      "step": 6
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 5.178266525268555,
      "learning_rate": 0.00014,
      "loss": 4.2387,
      "step": 7
    },
    {
      "epoch": 0.38095238095238093,
      "grad_norm": 6.597425937652588,
      "learning_rate": 0.00016,
      "loss": 4.3673,
      "step": 8
    },
    {
      "epoch": 0.42857142857142855,
      "grad_norm": 4.210269451141357,
      "learning_rate": 0.00018,
      "loss": 3.3903,
      "step": 9
    },
    {
      "epoch": 0.42857142857142855,
      "eval_loss": 3.175935983657837,
      "eval_runtime": 0.341,
      "eval_samples_per_second": 26.394,
      "eval_steps_per_second": 14.663,
      "step": 9
    },
    {
      "epoch": 0.47619047619047616,
      "grad_norm": 4.450435638427734,
      "learning_rate": 0.0002,
      "loss": 3.3034,
      "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": 1849564248145920.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}