last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.003915963424901611,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 7.831926849803223e-05,
      "eval_loss": NaN,
      "eval_runtime": 244.0622,
      "eval_samples_per_second": 22.031,
      "eval_steps_per_second": 11.018,
      "step": 1
    },
    {
      "epoch": 0.00039159634249016115,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.0007831926849803223,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.0007831926849803223,
      "eval_loss": NaN,
      "eval_runtime": 245.6983,
      "eval_samples_per_second": 21.885,
      "eval_steps_per_second": 10.944,
      "step": 10
    },
    {
      "epoch": 0.0011747890274704834,
      "grad_norm": NaN,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.0015663853699606446,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.0015663853699606446,
      "eval_loss": NaN,
      "eval_runtime": 243.8672,
      "eval_samples_per_second": 22.049,
      "eval_steps_per_second": 11.026,
      "step": 20
    },
    {
      "epoch": 0.0019579817124508055,
      "grad_norm": NaN,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.002349578054940967,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.002349578054940967,
      "eval_loss": NaN,
      "eval_runtime": 245.3077,
      "eval_samples_per_second": 21.919,
      "eval_steps_per_second": 10.962,
      "step": 30
    },
    {
      "epoch": 0.002741174397431128,
      "grad_norm": NaN,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.0,
      "step": 35
    },
    {
      "epoch": 0.003132770739921289,
      "grad_norm": NaN,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.0,
      "step": 40
    },
    {
      "epoch": 0.003132770739921289,
      "eval_loss": NaN,
      "eval_runtime": 243.2584,
      "eval_samples_per_second": 22.104,
      "eval_steps_per_second": 11.054,
      "step": 40
    },
    {
      "epoch": 0.00352436708241145,
      "grad_norm": NaN,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.0,
      "step": 45
    },
    {
      "epoch": 0.003915963424901611,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 50
    },
    {
      "epoch": 0.003915963424901611,
      "eval_loss": NaN,
      "eval_runtime": 243.9558,
      "eval_samples_per_second": 22.041,
      "eval_steps_per_second": 11.022,
      "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": 3226767672213504.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}