last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.003585964534810751,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 7.171929069621502e-05,
      "eval_loss": NaN,
      "eval_runtime": 56.2144,
      "eval_samples_per_second": 104.439,
      "eval_steps_per_second": 52.229,
      "step": 1
    },
    {
      "epoch": 0.00035859645348107505,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.0007171929069621501,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.0007171929069621501,
      "eval_loss": NaN,
      "eval_runtime": 55.1052,
      "eval_samples_per_second": 106.542,
      "eval_steps_per_second": 53.28,
      "step": 10
    },
    {
      "epoch": 0.0010757893604432251,
      "grad_norm": NaN,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.0014343858139243002,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.0014343858139243002,
      "eval_loss": NaN,
      "eval_runtime": 55.5858,
      "eval_samples_per_second": 105.621,
      "eval_steps_per_second": 52.819,
      "step": 20
    },
    {
      "epoch": 0.0017929822674053754,
      "grad_norm": NaN,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.0021515787208864503,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.0021515787208864503,
      "eval_loss": NaN,
      "eval_runtime": 54.9022,
      "eval_samples_per_second": 106.936,
      "eval_steps_per_second": 53.477,
      "step": 30
    },
    {
      "epoch": 0.0025101751743675255,
      "grad_norm": NaN,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.0,
      "step": 35
    },
    {
      "epoch": 0.0028687716278486004,
      "grad_norm": NaN,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.0,
      "step": 40
    },
    {
      "epoch": 0.0028687716278486004,
      "eval_loss": NaN,
      "eval_runtime": 54.7168,
      "eval_samples_per_second": 107.298,
      "eval_steps_per_second": 53.658,
      "step": 40
    },
    {
      "epoch": 0.0032273680813296756,
      "grad_norm": NaN,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.0,
      "step": 45
    },
    {
      "epoch": 0.003585964534810751,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 50
    },
    {
      "epoch": 0.003585964534810751,
      "eval_loss": NaN,
      "eval_runtime": 54.7114,
      "eval_samples_per_second": 107.309,
      "eval_steps_per_second": 53.663,
      "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": 270836545093632.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}