last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.021727322107550243,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0021727322107550242,
      "grad_norm": 38.3082275390625,
      "learning_rate": 2e-05,
      "loss": 26.2925,
      "step": 1
    },
    {
      "epoch": 0.0021727322107550242,
      "eval_loss": 6.362123489379883,
      "eval_runtime": 69.9182,
      "eval_samples_per_second": 2.775,
      "eval_steps_per_second": 1.387,
      "step": 1
    },
    {
      "epoch": 0.0043454644215100485,
      "grad_norm": 34.35226821899414,
      "learning_rate": 4e-05,
      "loss": 23.3629,
      "step": 2
    },
    {
      "epoch": 0.006518196632265073,
      "grad_norm": 46.58251953125,
      "learning_rate": 6e-05,
      "loss": 31.458,
      "step": 3
    },
    {
      "epoch": 0.006518196632265073,
      "eval_loss": 6.052692413330078,
      "eval_runtime": 69.8071,
      "eval_samples_per_second": 2.779,
      "eval_steps_per_second": 1.39,
      "step": 3
    },
    {
      "epoch": 0.008690928843020097,
      "grad_norm": 34.892906188964844,
      "learning_rate": 8e-05,
      "loss": 22.7088,
      "step": 4
    },
    {
      "epoch": 0.010863661053775122,
      "grad_norm": 27.14177703857422,
      "learning_rate": 0.0001,
      "loss": 19.6406,
      "step": 5
    },
    {
      "epoch": 0.013036393264530146,
      "grad_norm": 33.119747161865234,
      "learning_rate": 0.00012,
      "loss": 24.4518,
      "step": 6
    },
    {
      "epoch": 0.013036393264530146,
      "eval_loss": 4.772887706756592,
      "eval_runtime": 70.422,
      "eval_samples_per_second": 2.755,
      "eval_steps_per_second": 1.377,
      "step": 6
    },
    {
      "epoch": 0.015209125475285171,
      "grad_norm": 29.63689422607422,
      "learning_rate": 0.00014,
      "loss": 19.6883,
      "step": 7
    },
    {
      "epoch": 0.017381857686040194,
      "grad_norm": 27.822065353393555,
      "learning_rate": 0.00016,
      "loss": 18.6915,
      "step": 8
    },
    {
      "epoch": 0.01955458989679522,
      "grad_norm": 29.270885467529297,
      "learning_rate": 0.00018,
      "loss": 17.4177,
      "step": 9
    },
    {
      "epoch": 0.01955458989679522,
      "eval_loss": 3.6328675746917725,
      "eval_runtime": 69.2865,
      "eval_samples_per_second": 2.8,
      "eval_steps_per_second": 1.4,
      "step": 9
    },
    {
      "epoch": 0.021727322107550243,
      "grad_norm": 35.51152038574219,
      "learning_rate": 0.0002,
      "loss": 15.9869,
      "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": 1803185647779840.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}