last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.016842105263157894,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0016842105263157896,
      "grad_norm": 23.902877807617188,
      "learning_rate": 2e-05,
      "loss": 11.0946,
      "step": 1
    },
    {
      "epoch": 0.0016842105263157896,
      "eval_loss": 2.922970771789551,
      "eval_runtime": 2.918,
      "eval_samples_per_second": 85.675,
      "eval_steps_per_second": 42.837,
      "step": 1
    },
    {
      "epoch": 0.003368421052631579,
      "grad_norm": 37.24565505981445,
      "learning_rate": 4e-05,
      "loss": 9.4751,
      "step": 2
    },
    {
      "epoch": 0.0050526315789473685,
      "grad_norm": 27.14742660522461,
      "learning_rate": 6e-05,
      "loss": 11.7116,
      "step": 3
    },
    {
      "epoch": 0.0050526315789473685,
      "eval_loss": 2.919999599456787,
      "eval_runtime": 2.5673,
      "eval_samples_per_second": 97.38,
      "eval_steps_per_second": 48.69,
      "step": 3
    },
    {
      "epoch": 0.006736842105263158,
      "grad_norm": 36.44790267944336,
      "learning_rate": 8e-05,
      "loss": 10.008,
      "step": 4
    },
    {
      "epoch": 0.008421052631578947,
      "grad_norm": 28.15861701965332,
      "learning_rate": 0.0001,
      "loss": 10.1501,
      "step": 5
    },
    {
      "epoch": 0.010105263157894737,
      "grad_norm": 33.711490631103516,
      "learning_rate": 0.00012,
      "loss": 10.5486,
      "step": 6
    },
    {
      "epoch": 0.010105263157894737,
      "eval_loss": 2.9084646701812744,
      "eval_runtime": 2.5795,
      "eval_samples_per_second": 96.916,
      "eval_steps_per_second": 48.458,
      "step": 6
    },
    {
      "epoch": 0.011789473684210527,
      "grad_norm": 24.055999755859375,
      "learning_rate": 0.00014,
      "loss": 9.1365,
      "step": 7
    },
    {
      "epoch": 0.013473684210526317,
      "grad_norm": 63.981239318847656,
      "learning_rate": 0.00016,
      "loss": 12.1075,
      "step": 8
    },
    {
      "epoch": 0.015157894736842105,
      "grad_norm": 32.312252044677734,
      "learning_rate": 0.00018,
      "loss": 13.2968,
      "step": 9
    },
    {
      "epoch": 0.015157894736842105,
      "eval_loss": 2.880253791809082,
      "eval_runtime": 2.5756,
      "eval_samples_per_second": 97.064,
      "eval_steps_per_second": 48.532,
      "step": 9
    },
    {
      "epoch": 0.016842105263157894,
      "grad_norm": 35.027122497558594,
      "learning_rate": 0.0002,
      "loss": 10.6329,
      "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": 87728565780480.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}