last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.011217049915872126,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0011217049915872126,
      "grad_norm": 0.010207759216427803,
      "learning_rate": 2e-05,
      "loss": 11.9305,
      "step": 1
    },
    {
      "epoch": 0.0011217049915872126,
      "eval_loss": 11.931302070617676,
      "eval_runtime": 1.5864,
      "eval_samples_per_second": 473.392,
      "eval_steps_per_second": 118.506,
      "step": 1
    },
    {
      "epoch": 0.002243409983174425,
      "grad_norm": 0.014218014664947987,
      "learning_rate": 4e-05,
      "loss": 11.9324,
      "step": 2
    },
    {
      "epoch": 0.0033651149747616375,
      "grad_norm": 0.01008047629147768,
      "learning_rate": 6e-05,
      "loss": 11.9253,
      "step": 3
    },
    {
      "epoch": 0.0033651149747616375,
      "eval_loss": 11.931297302246094,
      "eval_runtime": 1.6298,
      "eval_samples_per_second": 460.789,
      "eval_steps_per_second": 115.351,
      "step": 3
    },
    {
      "epoch": 0.00448681996634885,
      "grad_norm": 0.005606319755315781,
      "learning_rate": 8e-05,
      "loss": 11.9338,
      "step": 4
    },
    {
      "epoch": 0.005608524957936063,
      "grad_norm": 0.009023796766996384,
      "learning_rate": 0.0001,
      "loss": 11.9314,
      "step": 5
    },
    {
      "epoch": 0.006730229949523275,
      "grad_norm": 0.01568673551082611,
      "learning_rate": 0.00012,
      "loss": 11.9317,
      "step": 6
    },
    {
      "epoch": 0.006730229949523275,
      "eval_loss": 11.931279182434082,
      "eval_runtime": 1.6377,
      "eval_samples_per_second": 458.556,
      "eval_steps_per_second": 114.792,
      "step": 6
    },
    {
      "epoch": 0.007851934941110488,
      "grad_norm": 0.016139866784214973,
      "learning_rate": 0.00014,
      "loss": 11.927,
      "step": 7
    },
    {
      "epoch": 0.0089736399326977,
      "grad_norm": 0.016835955902934074,
      "learning_rate": 0.00016,
      "loss": 11.9367,
      "step": 8
    },
    {
      "epoch": 0.010095344924284913,
      "grad_norm": 0.01244762446731329,
      "learning_rate": 0.00018,
      "loss": 11.9306,
      "step": 9
    },
    {
      "epoch": 0.010095344924284913,
      "eval_loss": 11.931243896484375,
      "eval_runtime": 1.6456,
      "eval_samples_per_second": 456.359,
      "eval_steps_per_second": 114.242,
      "step": 9
    },
    {
      "epoch": 0.011217049915872126,
      "grad_norm": 0.010091469623148441,
      "learning_rate": 0.0002,
      "loss": 11.9328,
      "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": 2392768512.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}