last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.000364549578742709,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 4.050550874918989e-05,
      "grad_norm": 3.6397786140441895,
      "learning_rate": 2e-05,
      "loss": 2.5374,
      "step": 1
    },
    {
      "epoch": 4.050550874918989e-05,
      "eval_loss": 2.4582314491271973,
      "eval_runtime": 1080.2619,
      "eval_samples_per_second": 4.812,
      "eval_steps_per_second": 4.812,
      "step": 1
    },
    {
      "epoch": 8.101101749837978e-05,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 2.321,
      "step": 2
    },
    {
      "epoch": 0.00012151652624756967,
      "grad_norm": 4.1562933921813965,
      "learning_rate": 4e-05,
      "loss": 2.3433,
      "step": 3
    },
    {
      "epoch": 0.00012151652624756967,
      "eval_loss": 2.450849771499634,
      "eval_runtime": 1078.6512,
      "eval_samples_per_second": 4.819,
      "eval_steps_per_second": 4.819,
      "step": 3
    },
    {
      "epoch": 0.00016202203499675956,
      "grad_norm": 3.986837148666382,
      "learning_rate": 6e-05,
      "loss": 2.5492,
      "step": 4
    },
    {
      "epoch": 0.00020252754374594945,
      "grad_norm": 2.91995906829834,
      "learning_rate": 8e-05,
      "loss": 1.8304,
      "step": 5
    },
    {
      "epoch": 0.00024303305249513934,
      "grad_norm": 3.264693260192871,
      "learning_rate": 0.0001,
      "loss": 2.1765,
      "step": 6
    },
    {
      "epoch": 0.00024303305249513934,
      "eval_loss": 2.262515068054199,
      "eval_runtime": 1079.0465,
      "eval_samples_per_second": 4.817,
      "eval_steps_per_second": 4.817,
      "step": 6
    },
    {
      "epoch": 0.00028353856124432923,
      "grad_norm": 5.041287899017334,
      "learning_rate": 0.00012,
      "loss": 2.4483,
      "step": 7
    },
    {
      "epoch": 0.0003240440699935191,
      "grad_norm": 4.383685111999512,
      "learning_rate": 0.00014,
      "loss": 2.01,
      "step": 8
    },
    {
      "epoch": 0.000364549578742709,
      "grad_norm": 5.683202266693115,
      "learning_rate": 0.00016,
      "loss": 2.0384,
      "step": 9
    },
    {
      "epoch": 0.000364549578742709,
      "eval_loss": 1.791993260383606,
      "eval_runtime": 1079.1282,
      "eval_samples_per_second": 4.817,
      "eval_steps_per_second": 4.817,
      "step": 9
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6676985552044032.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}