last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.09022556390977443,
  "eval_steps": 5,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0030075187969924814,
      "eval_loss": NaN,
      "eval_runtime": 27.481,
      "eval_samples_per_second": 5.094,
      "eval_steps_per_second": 2.547,
      "step": 1
    },
    {
      "epoch": 0.009022556390977444,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.015037593984962405,
      "eval_loss": NaN,
      "eval_runtime": 27.5842,
      "eval_samples_per_second": 5.075,
      "eval_steps_per_second": 2.538,
      "step": 5
    },
    {
      "epoch": 0.01804511278195489,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.02706766917293233,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.03007518796992481,
      "eval_loss": NaN,
      "eval_runtime": 27.6578,
      "eval_samples_per_second": 5.062,
      "eval_steps_per_second": 2.531,
      "step": 10
    },
    {
      "epoch": 0.03609022556390978,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.045112781954887216,
      "grad_norm": NaN,
      "learning_rate": 0.0002,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.045112781954887216,
      "eval_loss": NaN,
      "eval_runtime": 26.6388,
      "eval_samples_per_second": 5.255,
      "eval_steps_per_second": 2.628,
      "step": 15
    },
    {
      "epoch": 0.05413533834586466,
      "grad_norm": NaN,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.06015037593984962,
      "eval_loss": NaN,
      "eval_runtime": 27.7237,
      "eval_samples_per_second": 5.05,
      "eval_steps_per_second": 2.525,
      "step": 20
    },
    {
      "epoch": 0.06315789473684211,
      "grad_norm": NaN,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.07218045112781955,
      "grad_norm": NaN,
      "learning_rate": 6.909830056250527e-05,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.07518796992481203,
      "eval_loss": NaN,
      "eval_runtime": 27.6693,
      "eval_samples_per_second": 5.06,
      "eval_steps_per_second": 2.53,
      "step": 25
    },
    {
      "epoch": 0.081203007518797,
      "grad_norm": NaN,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 0.0,
      "step": 27
    },
    {
      "epoch": 0.09022556390977443,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.09022556390977443,
      "eval_loss": NaN,
      "eval_runtime": 24.5914,
      "eval_samples_per_second": 5.693,
      "eval_steps_per_second": 2.847,
      "step": 30
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 15,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.2170659010445312e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}