last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0014334605529574083,
  "eval_steps": 5,
  "global_step": 5,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0002866921105914817,
      "grad_norm": 2.3434934616088867,
      "learning_rate": 1e-05,
      "loss": 2.3633,
      "step": 1
    },
    {
      "epoch": 0.0002866921105914817,
      "eval_loss": 2.5438878536224365,
      "eval_runtime": 193.7172,
      "eval_samples_per_second": 30.328,
      "eval_steps_per_second": 15.166,
      "step": 1
    },
    {
      "epoch": 0.0005733842211829634,
      "grad_norm": 1.747134804725647,
      "learning_rate": 2e-05,
      "loss": 2.4938,
      "step": 2
    },
    {
      "epoch": 0.000860076331774445,
      "grad_norm": 1.7026760578155518,
      "learning_rate": 3e-05,
      "loss": 2.338,
      "step": 3
    },
    {
      "epoch": 0.0011467684423659267,
      "grad_norm": 1.9191794395446777,
      "learning_rate": 4e-05,
      "loss": 2.6796,
      "step": 4
    },
    {
      "epoch": 0.0014334605529574083,
      "grad_norm": 1.7047806978225708,
      "learning_rate": 5e-05,
      "loss": 2.4081,
      "step": 5
    },
    {
      "epoch": 0.0014334605529574083,
      "eval_loss": 2.463175058364868,
      "eval_runtime": 48.5493,
      "eval_samples_per_second": 121.011,
      "eval_steps_per_second": 60.516,
      "step": 5
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 511583956500480.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}