last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.013323464100666173,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0014803849000740192,
      "grad_norm": 0.380435049533844,
      "learning_rate": 2e-05,
      "loss": 0.7388,
      "step": 1
    },
    {
      "epoch": 0.0014803849000740192,
      "eval_loss": 0.7957301735877991,
      "eval_runtime": 17.7627,
      "eval_samples_per_second": 16.045,
      "eval_steps_per_second": 8.051,
      "step": 1
    },
    {
      "epoch": 0.0029607698001480384,
      "grad_norm": 0.437558650970459,
      "learning_rate": 4e-05,
      "loss": 0.7884,
      "step": 2
    },
    {
      "epoch": 0.0044411547002220575,
      "grad_norm": 0.47545790672302246,
      "learning_rate": 6e-05,
      "loss": 0.7885,
      "step": 3
    },
    {
      "epoch": 0.0044411547002220575,
      "eval_loss": 0.7941223978996277,
      "eval_runtime": 15.7296,
      "eval_samples_per_second": 18.119,
      "eval_steps_per_second": 9.091,
      "step": 3
    },
    {
      "epoch": 0.005921539600296077,
      "grad_norm": 0.46282222867012024,
      "learning_rate": 8e-05,
      "loss": 0.7392,
      "step": 4
    },
    {
      "epoch": 0.007401924500370096,
      "grad_norm": 0.47253677248954773,
      "learning_rate": 0.0001,
      "loss": 0.8891,
      "step": 5
    },
    {
      "epoch": 0.008882309400444115,
      "grad_norm": 0.4877263903617859,
      "learning_rate": 0.00012,
      "loss": 0.8255,
      "step": 6
    },
    {
      "epoch": 0.008882309400444115,
      "eval_loss": 0.7716551423072815,
      "eval_runtime": 15.7283,
      "eval_samples_per_second": 18.12,
      "eval_steps_per_second": 9.092,
      "step": 6
    },
    {
      "epoch": 0.010362694300518135,
      "grad_norm": 0.3506518006324768,
      "learning_rate": 0.00014,
      "loss": 0.4081,
      "step": 7
    },
    {
      "epoch": 0.011843079200592153,
      "grad_norm": 0.5045745968818665,
      "learning_rate": 0.00016,
      "loss": 0.9159,
      "step": 8
    },
    {
      "epoch": 0.013323464100666173,
      "grad_norm": 0.4290393888950348,
      "learning_rate": 0.00018,
      "loss": 0.7865,
      "step": 9
    },
    {
      "epoch": 0.013323464100666173,
      "eval_loss": 0.7267910242080688,
      "eval_runtime": 15.7422,
      "eval_samples_per_second": 18.104,
      "eval_steps_per_second": 9.084,
      "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": 3329215646662656.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}