last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.46511627906976744,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.046511627906976744,
      "grad_norm": 16.168792724609375,
      "learning_rate": 2e-05,
      "loss": 12.5927,
      "step": 1
    },
    {
      "epoch": 0.046511627906976744,
      "eval_loss": 3.677544593811035,
      "eval_runtime": 0.1596,
      "eval_samples_per_second": 56.401,
      "eval_steps_per_second": 31.334,
      "step": 1
    },
    {
      "epoch": 0.09302325581395349,
      "grad_norm": 14.232671737670898,
      "learning_rate": 4e-05,
      "loss": 13.2225,
      "step": 2
    },
    {
      "epoch": 0.13953488372093023,
      "grad_norm": 16.136791229248047,
      "learning_rate": 6e-05,
      "loss": 12.742,
      "step": 3
    },
    {
      "epoch": 0.13953488372093023,
      "eval_loss": 3.6357693672180176,
      "eval_runtime": 0.1605,
      "eval_samples_per_second": 56.06,
      "eval_steps_per_second": 31.144,
      "step": 3
    },
    {
      "epoch": 0.18604651162790697,
      "grad_norm": 11.944231986999512,
      "learning_rate": 8e-05,
      "loss": 9.897,
      "step": 4
    },
    {
      "epoch": 0.23255813953488372,
      "grad_norm": 17.337947845458984,
      "learning_rate": 0.0001,
      "loss": 12.6181,
      "step": 5
    },
    {
      "epoch": 0.27906976744186046,
      "grad_norm": 17.71935272216797,
      "learning_rate": 0.00012,
      "loss": 11.1099,
      "step": 6
    },
    {
      "epoch": 0.27906976744186046,
      "eval_loss": 2.780703067779541,
      "eval_runtime": 0.1757,
      "eval_samples_per_second": 51.211,
      "eval_steps_per_second": 28.45,
      "step": 6
    },
    {
      "epoch": 0.32558139534883723,
      "grad_norm": 18.756610870361328,
      "learning_rate": 0.00014,
      "loss": 9.2466,
      "step": 7
    },
    {
      "epoch": 0.37209302325581395,
      "grad_norm": 18.909032821655273,
      "learning_rate": 0.00016,
      "loss": 8.2619,
      "step": 8
    },
    {
      "epoch": 0.4186046511627907,
      "grad_norm": 22.742977142333984,
      "learning_rate": 0.00018,
      "loss": 5.61,
      "step": 9
    },
    {
      "epoch": 0.4186046511627907,
      "eval_loss": 0.8675842881202698,
      "eval_runtime": 0.1575,
      "eval_samples_per_second": 57.131,
      "eval_steps_per_second": 31.74,
      "step": 9
    },
    {
      "epoch": 0.46511627906976744,
      "grad_norm": 17.746034622192383,
      "learning_rate": 0.0002,
      "loss": 3.3071,
      "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": 298572369100800.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}