last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.03007518796992481,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0030075187969924814,
      "grad_norm": 0.6942528486251831,
      "learning_rate": 2e-05,
      "loss": 2.2174,
      "step": 1
    },
    {
      "epoch": 0.0030075187969924814,
      "eval_loss": 2.6961584091186523,
      "eval_runtime": 22.5368,
      "eval_samples_per_second": 6.212,
      "eval_steps_per_second": 3.106,
      "step": 1
    },
    {
      "epoch": 0.006015037593984963,
      "grad_norm": 0.5156584978103638,
      "learning_rate": 4e-05,
      "loss": 2.869,
      "step": 2
    },
    {
      "epoch": 0.009022556390977444,
      "grad_norm": 0.8239279389381409,
      "learning_rate": 6e-05,
      "loss": 2.1524,
      "step": 3
    },
    {
      "epoch": 0.009022556390977444,
      "eval_loss": 2.6957452297210693,
      "eval_runtime": 22.5539,
      "eval_samples_per_second": 6.207,
      "eval_steps_per_second": 3.104,
      "step": 3
    },
    {
      "epoch": 0.012030075187969926,
      "grad_norm": 0.4614211618900299,
      "learning_rate": 8e-05,
      "loss": 2.7253,
      "step": 4
    },
    {
      "epoch": 0.015037593984962405,
      "grad_norm": 0.5375364422798157,
      "learning_rate": 0.0001,
      "loss": 2.3831,
      "step": 5
    },
    {
      "epoch": 0.01804511278195489,
      "grad_norm": 0.6730604767799377,
      "learning_rate": 0.00012,
      "loss": 2.72,
      "step": 6
    },
    {
      "epoch": 0.01804511278195489,
      "eval_loss": 2.6837499141693115,
      "eval_runtime": 22.4492,
      "eval_samples_per_second": 6.236,
      "eval_steps_per_second": 3.118,
      "step": 6
    },
    {
      "epoch": 0.021052631578947368,
      "grad_norm": 0.4505942165851593,
      "learning_rate": 0.00014,
      "loss": 2.9096,
      "step": 7
    },
    {
      "epoch": 0.02406015037593985,
      "grad_norm": 0.46443304419517517,
      "learning_rate": 0.00016,
      "loss": 2.6855,
      "step": 8
    },
    {
      "epoch": 0.02706766917293233,
      "grad_norm": 0.34115874767303467,
      "learning_rate": 0.00018,
      "loss": 2.8562,
      "step": 9
    },
    {
      "epoch": 0.02706766917293233,
      "eval_loss": 2.641460418701172,
      "eval_runtime": 22.479,
      "eval_samples_per_second": 6.228,
      "eval_steps_per_second": 3.114,
      "step": 9
    },
    {
      "epoch": 0.03007518796992481,
      "grad_norm": 0.42337003350257874,
      "learning_rate": 0.0002,
      "loss": 3.1771,
      "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": 543199583010816.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}