last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.6153846153846154,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06153846153846154,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 4.3026,
      "step": 1
    },
    {
      "epoch": 0.06153846153846154,
      "eval_loss": 4.9456562995910645,
      "eval_runtime": 1.9932,
      "eval_samples_per_second": 13.546,
      "eval_steps_per_second": 2.007,
      "step": 1
    },
    {
      "epoch": 0.12307692307692308,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 4.3929,
      "step": 2
    },
    {
      "epoch": 0.18461538461538463,
      "grad_norm": 2.3017876148223877,
      "learning_rate": 2e-05,
      "loss": 4.491,
      "step": 3
    },
    {
      "epoch": 0.18461538461538463,
      "eval_loss": 4.9456562995910645,
      "eval_runtime": 1.2397,
      "eval_samples_per_second": 21.78,
      "eval_steps_per_second": 3.227,
      "step": 3
    },
    {
      "epoch": 0.24615384615384617,
      "grad_norm": 2.2400166988372803,
      "learning_rate": 4e-05,
      "loss": 4.5093,
      "step": 4
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 2.200295925140381,
      "learning_rate": 6e-05,
      "loss": 4.4495,
      "step": 5
    },
    {
      "epoch": 0.36923076923076925,
      "grad_norm": 2.5379796028137207,
      "learning_rate": 8e-05,
      "loss": 4.2715,
      "step": 6
    },
    {
      "epoch": 0.36923076923076925,
      "eval_loss": 4.600525856018066,
      "eval_runtime": 1.2603,
      "eval_samples_per_second": 21.423,
      "eval_steps_per_second": 3.174,
      "step": 6
    },
    {
      "epoch": 0.4307692307692308,
      "grad_norm": 2.464343547821045,
      "learning_rate": 0.0001,
      "loss": 4.1546,
      "step": 7
    },
    {
      "epoch": 0.49230769230769234,
      "grad_norm": 2.381273031234741,
      "learning_rate": 0.00012,
      "loss": 3.6829,
      "step": 8
    },
    {
      "epoch": 0.5538461538461539,
      "grad_norm": 3.1305465698242188,
      "learning_rate": 0.00014,
      "loss": 3.8566,
      "step": 9
    },
    {
      "epoch": 0.5538461538461539,
      "eval_loss": 3.2923731803894043,
      "eval_runtime": 1.2727,
      "eval_samples_per_second": 21.214,
      "eval_steps_per_second": 3.143,
      "step": 9
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 3.802546739578247,
      "learning_rate": 0.00016,
      "loss": 3.4508,
      "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": 5.592309412397056e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}