last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.012391573729863693,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0012391573729863693,
      "grad_norm": 1.9326956272125244,
      "learning_rate": 2e-05,
      "loss": 1.2225,
      "step": 1
    },
    {
      "epoch": 0.0012391573729863693,
      "eval_loss": 1.1873403787612915,
      "eval_runtime": 14.7896,
      "eval_samples_per_second": 22.989,
      "eval_steps_per_second": 11.495,
      "step": 1
    },
    {
      "epoch": 0.0024783147459727386,
      "grad_norm": 1.8224650621414185,
      "learning_rate": 4e-05,
      "loss": 1.1186,
      "step": 2
    },
    {
      "epoch": 0.0037174721189591076,
      "grad_norm": 1.713597297668457,
      "learning_rate": 6e-05,
      "loss": 1.095,
      "step": 3
    },
    {
      "epoch": 0.0037174721189591076,
      "eval_loss": 1.1725218296051025,
      "eval_runtime": 13.5731,
      "eval_samples_per_second": 25.05,
      "eval_steps_per_second": 12.525,
      "step": 3
    },
    {
      "epoch": 0.004956629491945477,
      "grad_norm": 1.9229203462600708,
      "learning_rate": 8e-05,
      "loss": 1.1895,
      "step": 4
    },
    {
      "epoch": 0.006195786864931847,
      "grad_norm": 1.4946612119674683,
      "learning_rate": 0.0001,
      "loss": 1.1122,
      "step": 5
    },
    {
      "epoch": 0.007434944237918215,
      "grad_norm": 1.8240691423416138,
      "learning_rate": 0.00012,
      "loss": 1.0115,
      "step": 6
    },
    {
      "epoch": 0.007434944237918215,
      "eval_loss": 0.9989386796951294,
      "eval_runtime": 13.6382,
      "eval_samples_per_second": 24.93,
      "eval_steps_per_second": 12.465,
      "step": 6
    },
    {
      "epoch": 0.008674101610904586,
      "grad_norm": 1.7445862293243408,
      "learning_rate": 0.00014,
      "loss": 0.9263,
      "step": 7
    },
    {
      "epoch": 0.009913258983890954,
      "grad_norm": 1.9537849426269531,
      "learning_rate": 0.00016,
      "loss": 0.8324,
      "step": 8
    },
    {
      "epoch": 0.011152416356877323,
      "grad_norm": 1.4942197799682617,
      "learning_rate": 0.00018,
      "loss": 1.057,
      "step": 9
    },
    {
      "epoch": 0.011152416356877323,
      "eval_loss": 0.8678604960441589,
      "eval_runtime": 13.7065,
      "eval_samples_per_second": 24.806,
      "eval_steps_per_second": 12.403,
      "step": 9
    },
    {
      "epoch": 0.012391573729863693,
      "grad_norm": 1.5920920372009277,
      "learning_rate": 0.0002,
      "loss": 0.865,
      "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": 2112182707814400.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}