last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0035382574082264487,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00035382574082264487,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 14.8697,
      "step": 1
    },
    {
      "epoch": 0.00035382574082264487,
      "eval_loss": 3.9243927001953125,
      "eval_runtime": 117.1959,
      "eval_samples_per_second": 5.077,
      "eval_steps_per_second": 5.077,
      "step": 1
    },
    {
      "epoch": 0.0007076514816452897,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 17.5113,
      "step": 2
    },
    {
      "epoch": 0.0010614772224679346,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 18.748,
      "step": 3
    },
    {
      "epoch": 0.0010614772224679346,
      "eval_loss": 3.9243927001953125,
      "eval_runtime": 117.0949,
      "eval_samples_per_second": 5.081,
      "eval_steps_per_second": 5.081,
      "step": 3
    },
    {
      "epoch": 0.0014153029632905795,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 16.9349,
      "step": 4
    },
    {
      "epoch": 0.0017691287041132243,
      "grad_norm": 95.18614196777344,
      "learning_rate": 2e-05,
      "loss": 15.3692,
      "step": 5
    },
    {
      "epoch": 0.002122954444935869,
      "grad_norm": 110.96968841552734,
      "learning_rate": 4e-05,
      "loss": 15.6412,
      "step": 6
    },
    {
      "epoch": 0.002122954444935869,
      "eval_loss": 3.736077308654785,
      "eval_runtime": 116.5658,
      "eval_samples_per_second": 5.104,
      "eval_steps_per_second": 5.104,
      "step": 6
    },
    {
      "epoch": 0.002476780185758514,
      "grad_norm": 87.67485046386719,
      "learning_rate": 6e-05,
      "loss": 12.3813,
      "step": 7
    },
    {
      "epoch": 0.002830605926581159,
      "grad_norm": 123.6387939453125,
      "learning_rate": 8e-05,
      "loss": 17.1364,
      "step": 8
    },
    {
      "epoch": 0.003184431667403804,
      "grad_norm": 77.38605499267578,
      "learning_rate": 0.0001,
      "loss": 11.6124,
      "step": 9
    },
    {
      "epoch": 0.003184431667403804,
      "eval_loss": 2.0625381469726562,
      "eval_runtime": 116.5884,
      "eval_samples_per_second": 5.103,
      "eval_steps_per_second": 5.103,
      "step": 9
    },
    {
      "epoch": 0.0035382574082264487,
      "grad_norm": 60.3310546875,
      "learning_rate": 0.00012,
      "loss": 7.6766,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7034387363266560.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}