last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.1201923076923077,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002403846153846154,
      "eval_loss": NaN,
      "eval_runtime": 1.0006,
      "eval_samples_per_second": 175.886,
      "eval_steps_per_second": 87.943,
      "step": 1
    },
    {
      "epoch": 0.01201923076923077,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.02403846153846154,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.02403846153846154,
      "eval_loss": NaN,
      "eval_runtime": 0.6587,
      "eval_samples_per_second": 267.18,
      "eval_steps_per_second": 133.59,
      "step": 10
    },
    {
      "epoch": 0.036057692307692304,
      "grad_norm": NaN,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.04807692307692308,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.04807692307692308,
      "eval_loss": NaN,
      "eval_runtime": 0.6914,
      "eval_samples_per_second": 254.565,
      "eval_steps_per_second": 127.283,
      "step": 20
    },
    {
      "epoch": 0.06009615384615385,
      "grad_norm": NaN,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.07211538461538461,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.07211538461538461,
      "eval_loss": NaN,
      "eval_runtime": 0.6965,
      "eval_samples_per_second": 252.674,
      "eval_steps_per_second": 126.337,
      "step": 30
    },
    {
      "epoch": 0.08413461538461539,
      "grad_norm": NaN,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.0,
      "step": 35
    },
    {
      "epoch": 0.09615384615384616,
      "grad_norm": NaN,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.0,
      "step": 40
    },
    {
      "epoch": 0.09615384615384616,
      "eval_loss": NaN,
      "eval_runtime": 0.6803,
      "eval_samples_per_second": 258.719,
      "eval_steps_per_second": 129.359,
      "step": 40
    },
    {
      "epoch": 0.10817307692307693,
      "grad_norm": NaN,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.0,
      "step": 45
    },
    {
      "epoch": 0.1201923076923077,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 50
    },
    {
      "epoch": 0.1201923076923077,
      "eval_loss": NaN,
      "eval_runtime": 0.6888,
      "eval_samples_per_second": 255.511,
      "eval_steps_per_second": 127.756,
      "step": 50
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 13,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1294978252800.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}