last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.003604902667627974,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0003604902667627974,
      "grad_norm": 0.5396327972412109,
      "learning_rate": 2e-05,
      "loss": 3.1034,
      "step": 1
    },
    {
      "epoch": 0.0003604902667627974,
      "eval_loss": 0.8934842944145203,
      "eval_runtime": 73.7074,
      "eval_samples_per_second": 15.86,
      "eval_steps_per_second": 7.937,
      "step": 1
    },
    {
      "epoch": 0.0007209805335255948,
      "grad_norm": 0.3926020860671997,
      "learning_rate": 4e-05,
      "loss": 3.1027,
      "step": 2
    },
    {
      "epoch": 0.0010814708002883922,
      "grad_norm": 0.33328357338905334,
      "learning_rate": 6e-05,
      "loss": 4.0248,
      "step": 3
    },
    {
      "epoch": 0.0010814708002883922,
      "eval_loss": 0.89348304271698,
      "eval_runtime": 73.7703,
      "eval_samples_per_second": 15.846,
      "eval_steps_per_second": 7.93,
      "step": 3
    },
    {
      "epoch": 0.0014419610670511895,
      "grad_norm": 0.4036581218242645,
      "learning_rate": 8e-05,
      "loss": 3.5279,
      "step": 4
    },
    {
      "epoch": 0.001802451333813987,
      "grad_norm": 0.4844096601009369,
      "learning_rate": 0.0001,
      "loss": 3.4007,
      "step": 5
    },
    {
      "epoch": 0.0021629416005767843,
      "grad_norm": 0.3808601498603821,
      "learning_rate": 0.00012,
      "loss": 3.1719,
      "step": 6
    },
    {
      "epoch": 0.0021629416005767843,
      "eval_loss": 0.8924697637557983,
      "eval_runtime": 73.8054,
      "eval_samples_per_second": 15.839,
      "eval_steps_per_second": 7.926,
      "step": 6
    },
    {
      "epoch": 0.0025234318673395817,
      "grad_norm": 0.3815302848815918,
      "learning_rate": 0.00014,
      "loss": 2.9936,
      "step": 7
    },
    {
      "epoch": 0.002883922134102379,
      "grad_norm": 0.4225100874900818,
      "learning_rate": 0.00016,
      "loss": 3.8196,
      "step": 8
    },
    {
      "epoch": 0.0032444124008651765,
      "grad_norm": 0.41114550828933716,
      "learning_rate": 0.00018,
      "loss": 3.8682,
      "step": 9
    },
    {
      "epoch": 0.0032444124008651765,
      "eval_loss": 0.8888127207756042,
      "eval_runtime": 73.8329,
      "eval_samples_per_second": 15.833,
      "eval_steps_per_second": 7.923,
      "step": 9
    },
    {
      "epoch": 0.003604902667627974,
      "grad_norm": 0.3832067549228668,
      "learning_rate": 0.0002,
      "loss": 3.2433,
      "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": 3257498484080640.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}