last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004968326915911067,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0004968326915911067,
      "grad_norm": 9.575961112976074,
      "learning_rate": 2e-05,
      "loss": 13.4967,
      "step": 1
    },
    {
      "epoch": 0.0004968326915911067,
      "eval_loss": 4.478808879852295,
      "eval_runtime": 9.6736,
      "eval_samples_per_second": 87.661,
      "eval_steps_per_second": 43.831,
      "step": 1
    },
    {
      "epoch": 0.0009936653831822133,
      "grad_norm": 14.79227066040039,
      "learning_rate": 4e-05,
      "loss": 18.8801,
      "step": 2
    },
    {
      "epoch": 0.0014904980747733202,
      "grad_norm": 11.625609397888184,
      "learning_rate": 6e-05,
      "loss": 16.0218,
      "step": 3
    },
    {
      "epoch": 0.0014904980747733202,
      "eval_loss": 4.443297863006592,
      "eval_runtime": 9.5987,
      "eval_samples_per_second": 88.345,
      "eval_steps_per_second": 44.173,
      "step": 3
    },
    {
      "epoch": 0.0019873307663644266,
      "grad_norm": 15.475156784057617,
      "learning_rate": 8e-05,
      "loss": 19.5836,
      "step": 4
    },
    {
      "epoch": 0.0024841634579555335,
      "grad_norm": 11.649445533752441,
      "learning_rate": 0.0001,
      "loss": 14.2257,
      "step": 5
    },
    {
      "epoch": 0.0029809961495466403,
      "grad_norm": 13.620771408081055,
      "learning_rate": 0.00012,
      "loss": 18.1914,
      "step": 6
    },
    {
      "epoch": 0.0029809961495466403,
      "eval_loss": 4.233336925506592,
      "eval_runtime": 9.9171,
      "eval_samples_per_second": 85.509,
      "eval_steps_per_second": 42.754,
      "step": 6
    },
    {
      "epoch": 0.0034778288411377468,
      "grad_norm": 10.729366302490234,
      "learning_rate": 0.00014,
      "loss": 16.1147,
      "step": 7
    },
    {
      "epoch": 0.003974661532728853,
      "grad_norm": 18.95943832397461,
      "learning_rate": 0.00016,
      "loss": 20.5381,
      "step": 8
    },
    {
      "epoch": 0.0044714942243199605,
      "grad_norm": 19.25127601623535,
      "learning_rate": 0.00018,
      "loss": 18.0612,
      "step": 9
    },
    {
      "epoch": 0.0044714942243199605,
      "eval_loss": 3.78324556350708,
      "eval_runtime": 9.6298,
      "eval_samples_per_second": 88.06,
      "eval_steps_per_second": 44.03,
      "step": 9
    },
    {
      "epoch": 0.004968326915911067,
      "grad_norm": 13.589359283447266,
      "learning_rate": 0.0002,
      "loss": 15.8584,
      "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": 75069652992000.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}