last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.005692329585883023,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005692329585883023,
      "grad_norm": 0.8005009889602661,
      "learning_rate": 2e-05,
      "loss": 2.519,
      "step": 1
    },
    {
      "epoch": 0.0005692329585883023,
      "eval_loss": 2.5949466228485107,
      "eval_runtime": 23.1251,
      "eval_samples_per_second": 32.0,
      "eval_steps_per_second": 16.0,
      "step": 1
    },
    {
      "epoch": 0.0011384659171766046,
      "grad_norm": 0.879878044128418,
      "learning_rate": 4e-05,
      "loss": 2.5677,
      "step": 2
    },
    {
      "epoch": 0.0017076988757649069,
      "grad_norm": 0.9343793988227844,
      "learning_rate": 6e-05,
      "loss": 2.541,
      "step": 3
    },
    {
      "epoch": 0.0017076988757649069,
      "eval_loss": 2.590181350708008,
      "eval_runtime": 22.4639,
      "eval_samples_per_second": 32.942,
      "eval_steps_per_second": 16.471,
      "step": 3
    },
    {
      "epoch": 0.002276931834353209,
      "grad_norm": 0.927487313747406,
      "learning_rate": 8e-05,
      "loss": 2.6392,
      "step": 4
    },
    {
      "epoch": 0.0028461647929415115,
      "grad_norm": 0.9168305397033691,
      "learning_rate": 0.0001,
      "loss": 2.4596,
      "step": 5
    },
    {
      "epoch": 0.0034153977515298138,
      "grad_norm": 0.9559779763221741,
      "learning_rate": 0.00012,
      "loss": 2.4561,
      "step": 6
    },
    {
      "epoch": 0.0034153977515298138,
      "eval_loss": 2.5247347354888916,
      "eval_runtime": 22.5955,
      "eval_samples_per_second": 32.75,
      "eval_steps_per_second": 16.375,
      "step": 6
    },
    {
      "epoch": 0.003984630710118116,
      "grad_norm": 1.058866024017334,
      "learning_rate": 0.00014,
      "loss": 2.4268,
      "step": 7
    },
    {
      "epoch": 0.004553863668706418,
      "grad_norm": 0.981652557849884,
      "learning_rate": 0.00016,
      "loss": 2.5081,
      "step": 8
    },
    {
      "epoch": 0.00512309662729472,
      "grad_norm": 0.8454458713531494,
      "learning_rate": 0.00018,
      "loss": 2.392,
      "step": 9
    },
    {
      "epoch": 0.00512309662729472,
      "eval_loss": 2.4337289333343506,
      "eval_runtime": 22.5925,
      "eval_samples_per_second": 32.754,
      "eval_steps_per_second": 16.377,
      "step": 9
    },
    {
      "epoch": 0.005692329585883023,
      "grad_norm": 0.8924505114555359,
      "learning_rate": 0.0002,
      "loss": 2.4098,
      "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": 1849564248145920.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}