last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00023673121537805974,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.3673121537805975e-05,
      "grad_norm": 4.070494651794434,
      "learning_rate": 2e-05,
      "loss": 4.5492,
      "step": 1
    },
    {
      "epoch": 2.3673121537805975e-05,
      "eval_loss": 4.50961446762085,
      "eval_runtime": 341.9104,
      "eval_samples_per_second": 52.022,
      "eval_steps_per_second": 26.013,
      "step": 1
    },
    {
      "epoch": 4.734624307561195e-05,
      "grad_norm": 4.397290229797363,
      "learning_rate": 4e-05,
      "loss": 4.7238,
      "step": 2
    },
    {
      "epoch": 7.101936461341793e-05,
      "grad_norm": 4.730429172515869,
      "learning_rate": 6e-05,
      "loss": 4.4912,
      "step": 3
    },
    {
      "epoch": 7.101936461341793e-05,
      "eval_loss": 4.477484226226807,
      "eval_runtime": 335.0046,
      "eval_samples_per_second": 53.095,
      "eval_steps_per_second": 26.549,
      "step": 3
    },
    {
      "epoch": 9.46924861512239e-05,
      "grad_norm": 4.465548038482666,
      "learning_rate": 8e-05,
      "loss": 4.8216,
      "step": 4
    },
    {
      "epoch": 0.00011836560768902987,
      "grad_norm": 4.55939245223999,
      "learning_rate": 0.0001,
      "loss": 4.3881,
      "step": 5
    },
    {
      "epoch": 0.00014203872922683586,
      "grad_norm": 5.568608283996582,
      "learning_rate": 0.00012,
      "loss": 4.6893,
      "step": 6
    },
    {
      "epoch": 0.00014203872922683586,
      "eval_loss": 3.9816391468048096,
      "eval_runtime": 333.254,
      "eval_samples_per_second": 53.374,
      "eval_steps_per_second": 26.688,
      "step": 6
    },
    {
      "epoch": 0.0001657118507646418,
      "grad_norm": 5.301882266998291,
      "learning_rate": 0.00014,
      "loss": 4.0845,
      "step": 7
    },
    {
      "epoch": 0.0001893849723024478,
      "grad_norm": 5.137835502624512,
      "learning_rate": 0.00016,
      "loss": 4.0146,
      "step": 8
    },
    {
      "epoch": 0.00021305809384025378,
      "grad_norm": 6.15079402923584,
      "learning_rate": 0.00018,
      "loss": 3.206,
      "step": 9
    },
    {
      "epoch": 0.00021305809384025378,
      "eval_loss": 3.147989511489868,
      "eval_runtime": 332.2808,
      "eval_samples_per_second": 53.53,
      "eval_steps_per_second": 26.767,
      "step": 9
    },
    {
      "epoch": 0.00023673121537805974,
      "grad_norm": 6.127230167388916,
      "learning_rate": 0.0002,
      "loss": 3.2116,
      "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": 489477793382400.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}