last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.007136485280999108,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0007136485280999108,
      "grad_norm": 2.884699583053589,
      "learning_rate": 2e-05,
      "loss": 7.3707,
      "step": 1
    },
    {
      "epoch": 0.0007136485280999108,
      "eval_loss": 1.9733723402023315,
      "eval_runtime": 40.6713,
      "eval_samples_per_second": 14.507,
      "eval_steps_per_second": 7.253,
      "step": 1
    },
    {
      "epoch": 0.0014272970561998216,
      "grad_norm": 3.539494276046753,
      "learning_rate": 4e-05,
      "loss": 6.7833,
      "step": 2
    },
    {
      "epoch": 0.0021409455842997323,
      "grad_norm": 1.5795257091522217,
      "learning_rate": 6e-05,
      "loss": 7.0235,
      "step": 3
    },
    {
      "epoch": 0.0021409455842997323,
      "eval_loss": 1.9712296724319458,
      "eval_runtime": 40.7455,
      "eval_samples_per_second": 14.48,
      "eval_steps_per_second": 7.24,
      "step": 3
    },
    {
      "epoch": 0.002854594112399643,
      "grad_norm": 1.5734566450119019,
      "learning_rate": 8e-05,
      "loss": 7.9809,
      "step": 4
    },
    {
      "epoch": 0.003568242640499554,
      "grad_norm": 3.3591525554656982,
      "learning_rate": 0.0001,
      "loss": 6.837,
      "step": 5
    },
    {
      "epoch": 0.004281891168599465,
      "grad_norm": 1.540778398513794,
      "learning_rate": 0.00012,
      "loss": 8.0313,
      "step": 6
    },
    {
      "epoch": 0.004281891168599465,
      "eval_loss": 1.9519037008285522,
      "eval_runtime": 40.7799,
      "eval_samples_per_second": 14.468,
      "eval_steps_per_second": 7.234,
      "step": 6
    },
    {
      "epoch": 0.0049955396966993755,
      "grad_norm": 1.8274019956588745,
      "learning_rate": 0.00014,
      "loss": 8.3603,
      "step": 7
    },
    {
      "epoch": 0.005709188224799286,
      "grad_norm": 1.92953360080719,
      "learning_rate": 0.00016,
      "loss": 7.3223,
      "step": 8
    },
    {
      "epoch": 0.006422836752899197,
      "grad_norm": 2.5140371322631836,
      "learning_rate": 0.00018,
      "loss": 7.4628,
      "step": 9
    },
    {
      "epoch": 0.006422836752899197,
      "eval_loss": 1.858712911605835,
      "eval_runtime": 40.7818,
      "eval_samples_per_second": 14.467,
      "eval_steps_per_second": 7.234,
      "step": 9
    },
    {
      "epoch": 0.007136485280999108,
      "grad_norm": 2.170633316040039,
      "learning_rate": 0.0002,
      "loss": 6.9033,
      "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": 3176061021978624.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}