last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.007297938332421091,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0007297938332421091,
      "grad_norm": 18.8606014251709,
      "learning_rate": 2e-05,
      "loss": 37.0272,
      "step": 1
    },
    {
      "epoch": 0.0007297938332421091,
      "eval_loss": 9.656542778015137,
      "eval_runtime": 176.6537,
      "eval_samples_per_second": 3.266,
      "eval_steps_per_second": 1.636,
      "step": 1
    },
    {
      "epoch": 0.0014595876664842181,
      "grad_norm": 20.115055084228516,
      "learning_rate": 4e-05,
      "loss": 37.8305,
      "step": 2
    },
    {
      "epoch": 0.0021893814997263274,
      "grad_norm": 21.362010955810547,
      "learning_rate": 6e-05,
      "loss": 39.1861,
      "step": 3
    },
    {
      "epoch": 0.0021893814997263274,
      "eval_loss": 9.560768127441406,
      "eval_runtime": 177.637,
      "eval_samples_per_second": 3.248,
      "eval_steps_per_second": 1.627,
      "step": 3
    },
    {
      "epoch": 0.0029191753329684362,
      "grad_norm": 21.96184730529785,
      "learning_rate": 8e-05,
      "loss": 36.8291,
      "step": 4
    },
    {
      "epoch": 0.0036489691662105455,
      "grad_norm": 23.605018615722656,
      "learning_rate": 0.0001,
      "loss": 35.7411,
      "step": 5
    },
    {
      "epoch": 0.004378762999452655,
      "grad_norm": 26.890573501586914,
      "learning_rate": 0.00012,
      "loss": 38.8873,
      "step": 6
    },
    {
      "epoch": 0.004378762999452655,
      "eval_loss": 8.529736518859863,
      "eval_runtime": 177.7105,
      "eval_samples_per_second": 3.247,
      "eval_steps_per_second": 1.626,
      "step": 6
    },
    {
      "epoch": 0.005108556832694764,
      "grad_norm": 32.43141174316406,
      "learning_rate": 0.00014,
      "loss": 33.8516,
      "step": 7
    },
    {
      "epoch": 0.0058383506659368724,
      "grad_norm": 41.67256546020508,
      "learning_rate": 0.00016,
      "loss": 30.5001,
      "step": 8
    },
    {
      "epoch": 0.006568144499178982,
      "grad_norm": 44.105892181396484,
      "learning_rate": 0.00018,
      "loss": 22.1419,
      "step": 9
    },
    {
      "epoch": 0.006568144499178982,
      "eval_loss": 4.630993843078613,
      "eval_runtime": 177.6493,
      "eval_samples_per_second": 3.248,
      "eval_steps_per_second": 1.627,
      "step": 9
    },
    {
      "epoch": 0.007297938332421091,
      "grad_norm": 50.32173156738281,
      "learning_rate": 0.0002,
      "loss": 19.0138,
      "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": 7368366033469440.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}