last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.020931449502878074,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0020931449502878076,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.0020931449502878076,
      "eval_loss": NaN,
      "eval_runtime": 3.4488,
      "eval_samples_per_second": 58.571,
      "eval_steps_per_second": 29.286,
      "step": 1
    },
    {
      "epoch": 0.004186289900575615,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.006279434850863423,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.006279434850863423,
      "eval_loss": NaN,
      "eval_runtime": 3.0344,
      "eval_samples_per_second": 66.57,
      "eval_steps_per_second": 33.285,
      "step": 3
    },
    {
      "epoch": 0.00837257980115123,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.010465724751439037,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.012558869701726845,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.012558869701726845,
      "eval_loss": NaN,
      "eval_runtime": 2.9848,
      "eval_samples_per_second": 67.677,
      "eval_steps_per_second": 33.838,
      "step": 6
    },
    {
      "epoch": 0.014652014652014652,
      "grad_norm": NaN,
      "learning_rate": 0.00014,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.01674515960230246,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.018838304552590265,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.018838304552590265,
      "eval_loss": NaN,
      "eval_runtime": 3.0711,
      "eval_samples_per_second": 65.774,
      "eval_steps_per_second": 32.887,
      "step": 9
    },
    {
      "epoch": 0.020931449502878074,
      "grad_norm": NaN,
      "learning_rate": 0.0002,
      "loss": 0.0,
      "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": 255791978250240.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}