last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.05263157894736842,
  "eval_steps": 5,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0017543859649122807,
      "eval_loss": NaN,
      "eval_runtime": 66.4871,
      "eval_samples_per_second": 3.61,
      "eval_steps_per_second": 1.805,
      "step": 1
    },
    {
      "epoch": 0.005263157894736842,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.008771929824561403,
      "eval_loss": NaN,
      "eval_runtime": 66.447,
      "eval_samples_per_second": 3.612,
      "eval_steps_per_second": 1.806,
      "step": 5
    },
    {
      "epoch": 0.010526315789473684,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.015789473684210527,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.017543859649122806,
      "eval_loss": NaN,
      "eval_runtime": 66.4259,
      "eval_samples_per_second": 3.613,
      "eval_steps_per_second": 1.807,
      "step": 10
    },
    {
      "epoch": 0.021052631578947368,
      "grad_norm": NaN,
      "learning_rate": 0.00019510565162951537,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.02631578947368421,
      "grad_norm": NaN,
      "learning_rate": 0.00017071067811865476,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.02631578947368421,
      "eval_loss": NaN,
      "eval_runtime": 66.4783,
      "eval_samples_per_second": 3.61,
      "eval_steps_per_second": 1.805,
      "step": 15
    },
    {
      "epoch": 0.031578947368421054,
      "grad_norm": NaN,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.03508771929824561,
      "eval_loss": NaN,
      "eval_runtime": 66.4897,
      "eval_samples_per_second": 3.61,
      "eval_steps_per_second": 1.805,
      "step": 20
    },
    {
      "epoch": 0.03684210526315789,
      "grad_norm": NaN,
      "learning_rate": 8.435655349597689e-05,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.042105263157894736,
      "grad_norm": NaN,
      "learning_rate": 4.12214747707527e-05,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.043859649122807015,
      "eval_loss": NaN,
      "eval_runtime": 66.6293,
      "eval_samples_per_second": 3.602,
      "eval_steps_per_second": 1.801,
      "step": 25
    },
    {
      "epoch": 0.04736842105263158,
      "grad_norm": NaN,
      "learning_rate": 1.0899347581163221e-05,
      "loss": 0.0,
      "step": 27
    },
    {
      "epoch": 0.05263157894736842,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.05263157894736842,
      "eval_loss": NaN,
      "eval_runtime": 66.8071,
      "eval_samples_per_second": 3.592,
      "eval_steps_per_second": 1.796,
      "step": 30
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.983522292314931e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}