last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.06191950464396285,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.001238390092879257,
      "eval_loss": NaN,
      "eval_runtime": 24.0496,
      "eval_samples_per_second": 14.137,
      "eval_steps_per_second": 7.069,
      "step": 1
    },
    {
      "epoch": 0.006191950464396285,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.01238390092879257,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.01238390092879257,
      "eval_loss": NaN,
      "eval_runtime": 22.9903,
      "eval_samples_per_second": 14.789,
      "eval_steps_per_second": 7.394,
      "step": 10
    },
    {
      "epoch": 0.018575851393188854,
      "grad_norm": NaN,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.02476780185758514,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.4791,
      "step": 20
    },
    {
      "epoch": 0.02476780185758514,
      "eval_loss": NaN,
      "eval_runtime": 23.0187,
      "eval_samples_per_second": 14.771,
      "eval_steps_per_second": 7.385,
      "step": 20
    },
    {
      "epoch": 0.030959752321981424,
      "grad_norm": NaN,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.03715170278637771,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.03715170278637771,
      "eval_loss": NaN,
      "eval_runtime": 23.1016,
      "eval_samples_per_second": 14.718,
      "eval_steps_per_second": 7.359,
      "step": 30
    },
    {
      "epoch": 0.043343653250773995,
      "grad_norm": NaN,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.0,
      "step": 35
    },
    {
      "epoch": 0.04953560371517028,
      "grad_norm": NaN,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.0,
      "step": 40
    },
    {
      "epoch": 0.04953560371517028,
      "eval_loss": NaN,
      "eval_runtime": 23.0164,
      "eval_samples_per_second": 14.772,
      "eval_steps_per_second": 7.386,
      "step": 40
    },
    {
      "epoch": 0.05572755417956656,
      "grad_norm": NaN,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 2.7407,
      "step": 45
    },
    {
      "epoch": 0.06191950464396285,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 50
    },
    {
      "epoch": 0.06191950464396285,
      "eval_loss": NaN,
      "eval_runtime": 23.0344,
      "eval_samples_per_second": 14.761,
      "eval_steps_per_second": 7.38,
      "step": 50
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 13,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 9001625343492096.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}