last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.14450867052023122,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002890173410404624,
      "eval_loss": NaN,
      "eval_runtime": 21.3252,
      "eval_samples_per_second": 6.846,
      "eval_steps_per_second": 3.423,
      "step": 1
    },
    {
      "epoch": 0.014450867052023121,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.028901734104046242,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.028901734104046242,
      "eval_loss": NaN,
      "eval_runtime": 19.2525,
      "eval_samples_per_second": 7.583,
      "eval_steps_per_second": 3.792,
      "step": 10
    },
    {
      "epoch": 0.04335260115606936,
      "grad_norm": NaN,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.057803468208092484,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.057803468208092484,
      "eval_loss": NaN,
      "eval_runtime": 25.0682,
      "eval_samples_per_second": 5.824,
      "eval_steps_per_second": 2.912,
      "step": 20
    },
    {
      "epoch": 0.07225433526011561,
      "grad_norm": NaN,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.08670520231213873,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.08670520231213873,
      "eval_loss": NaN,
      "eval_runtime": 19.2346,
      "eval_samples_per_second": 7.59,
      "eval_steps_per_second": 3.795,
      "step": 30
    },
    {
      "epoch": 0.10115606936416185,
      "grad_norm": NaN,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.0,
      "step": 35
    },
    {
      "epoch": 0.11560693641618497,
      "grad_norm": NaN,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.0,
      "step": 40
    },
    {
      "epoch": 0.11560693641618497,
      "eval_loss": NaN,
      "eval_runtime": 19.2386,
      "eval_samples_per_second": 7.589,
      "eval_steps_per_second": 3.794,
      "step": 40
    },
    {
      "epoch": 0.13005780346820808,
      "grad_norm": NaN,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.0,
      "step": 45
    },
    {
      "epoch": 0.14450867052023122,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 50
    },
    {
      "epoch": 0.14450867052023122,
      "eval_loss": NaN,
      "eval_runtime": 19.3451,
      "eval_samples_per_second": 7.547,
      "eval_steps_per_second": 3.774,
      "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": 3234875128676352.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}