last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.15037593984962405,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0030075187969924814,
      "eval_loss": NaN,
      "eval_runtime": 14.9633,
      "eval_samples_per_second": 9.356,
      "eval_steps_per_second": 4.678,
      "step": 1
    },
    {
      "epoch": 0.015037593984962405,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 1.4893,
      "step": 5
    },
    {
      "epoch": 0.03007518796992481,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.03007518796992481,
      "eval_loss": NaN,
      "eval_runtime": 13.9467,
      "eval_samples_per_second": 10.038,
      "eval_steps_per_second": 5.019,
      "step": 10
    },
    {
      "epoch": 0.045112781954887216,
      "grad_norm": NaN,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.06015037593984962,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.06015037593984962,
      "eval_loss": NaN,
      "eval_runtime": 13.9547,
      "eval_samples_per_second": 10.032,
      "eval_steps_per_second": 5.016,
      "step": 20
    },
    {
      "epoch": 0.07518796992481203,
      "grad_norm": NaN,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.09022556390977443,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.09022556390977443,
      "eval_loss": NaN,
      "eval_runtime": 13.9623,
      "eval_samples_per_second": 10.027,
      "eval_steps_per_second": 5.014,
      "step": 30
    },
    {
      "epoch": 0.10526315789473684,
      "grad_norm": NaN,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.0,
      "step": 35
    },
    {
      "epoch": 0.12030075187969924,
      "grad_norm": NaN,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.0,
      "step": 40
    },
    {
      "epoch": 0.12030075187969924,
      "eval_loss": NaN,
      "eval_runtime": 14.0349,
      "eval_samples_per_second": 9.975,
      "eval_steps_per_second": 4.988,
      "step": 40
    },
    {
      "epoch": 0.13533834586466165,
      "grad_norm": NaN,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.0,
      "step": 45
    },
    {
      "epoch": 0.15037593984962405,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 50
    },
    {
      "epoch": 0.15037593984962405,
      "eval_loss": NaN,
      "eval_runtime": 13.9926,
      "eval_samples_per_second": 10.005,
      "eval_steps_per_second": 5.003,
      "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": 1.4638142787158016e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}