last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.33613445378151263,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03361344537815126,
      "grad_norm": 21.176664352416992,
      "learning_rate": 2e-05,
      "loss": 3.109,
      "step": 1
    },
    {
      "epoch": 0.03361344537815126,
      "eval_loss": 0.8350883722305298,
      "eval_runtime": 1.1526,
      "eval_samples_per_second": 11.279,
      "eval_steps_per_second": 6.073,
      "step": 1
    },
    {
      "epoch": 0.06722689075630252,
      "grad_norm": 29.099716186523438,
      "learning_rate": 4e-05,
      "loss": 3.4425,
      "step": 2
    },
    {
      "epoch": 0.10084033613445378,
      "grad_norm": 22.868017196655273,
      "learning_rate": 6e-05,
      "loss": 3.1076,
      "step": 3
    },
    {
      "epoch": 0.10084033613445378,
      "eval_loss": 0.8264645934104919,
      "eval_runtime": 0.2029,
      "eval_samples_per_second": 64.057,
      "eval_steps_per_second": 34.492,
      "step": 3
    },
    {
      "epoch": 0.13445378151260504,
      "grad_norm": 19.981307983398438,
      "learning_rate": 8e-05,
      "loss": 3.1769,
      "step": 4
    },
    {
      "epoch": 0.16806722689075632,
      "grad_norm": 27.104169845581055,
      "learning_rate": 0.0001,
      "loss": 2.9308,
      "step": 5
    },
    {
      "epoch": 0.20168067226890757,
      "grad_norm": 14.100279808044434,
      "learning_rate": 0.00012,
      "loss": 2.733,
      "step": 6
    },
    {
      "epoch": 0.20168067226890757,
      "eval_loss": 0.7169986963272095,
      "eval_runtime": 0.2067,
      "eval_samples_per_second": 62.906,
      "eval_steps_per_second": 33.872,
      "step": 6
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 23.43442153930664,
      "learning_rate": 0.00014,
      "loss": 2.9591,
      "step": 7
    },
    {
      "epoch": 0.2689075630252101,
      "grad_norm": 19.78339195251465,
      "learning_rate": 0.00016,
      "loss": 2.6446,
      "step": 8
    },
    {
      "epoch": 0.3025210084033613,
      "grad_norm": 21.895431518554688,
      "learning_rate": 0.00018,
      "loss": 2.0876,
      "step": 9
    },
    {
      "epoch": 0.3025210084033613,
      "eval_loss": 0.5609821081161499,
      "eval_runtime": 0.2125,
      "eval_samples_per_second": 61.172,
      "eval_steps_per_second": 32.939,
      "step": 9
    },
    {
      "epoch": 0.33613445378151263,
      "grad_norm": 38.3336296081543,
      "learning_rate": 0.0002,
      "loss": 2.2363,
      "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": 87728565780480.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}