last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.08488964346349745,
  "eval_steps": 5,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.003395585738539898,
      "grad_norm": NaN,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.003395585738539898,
      "eval_loss": NaN,
      "eval_runtime": 4.1118,
      "eval_samples_per_second": 60.314,
      "eval_steps_per_second": 7.539,
      "step": 1
    },
    {
      "epoch": 0.006791171477079796,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.010186757215619695,
      "grad_norm": NaN,
      "learning_rate": 3e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.013582342954159592,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.01697792869269949,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.01697792869269949,
      "eval_loss": NaN,
      "eval_runtime": 3.5099,
      "eval_samples_per_second": 70.658,
      "eval_steps_per_second": 8.832,
      "step": 5
    },
    {
      "epoch": 0.02037351443123939,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.023769100169779286,
      "grad_norm": NaN,
      "learning_rate": 7e-05,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.027164685908319185,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.030560271646859084,
      "grad_norm": NaN,
      "learning_rate": 9e-05,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.03395585738539898,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.03395585738539898,
      "eval_loss": NaN,
      "eval_runtime": 3.4967,
      "eval_samples_per_second": 70.924,
      "eval_steps_per_second": 8.866,
      "step": 10
    },
    {
      "epoch": 0.03735144312393888,
      "grad_norm": NaN,
      "learning_rate": 9.98458666866564e-05,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.04074702886247878,
      "grad_norm": NaN,
      "learning_rate": 9.938441702975689e-05,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.044142614601018676,
      "grad_norm": NaN,
      "learning_rate": 9.861849601988383e-05,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.04753820033955857,
      "grad_norm": NaN,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 0.050933786078098474,
      "grad_norm": NaN,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.050933786078098474,
      "eval_loss": NaN,
      "eval_runtime": 3.4974,
      "eval_samples_per_second": 70.911,
      "eval_steps_per_second": 8.864,
      "step": 15
    },
    {
      "epoch": 0.05432937181663837,
      "grad_norm": NaN,
      "learning_rate": 9.45503262094184e-05,
      "loss": 0.0,
      "step": 16
    },
    {
      "epoch": 0.057724957555178265,
      "grad_norm": NaN,
      "learning_rate": 9.263200821770461e-05,
      "loss": 0.0,
      "step": 17
    },
    {
      "epoch": 0.06112054329371817,
      "grad_norm": NaN,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.06451612903225806,
      "grad_norm": NaN,
      "learning_rate": 8.802029828000156e-05,
      "loss": 0.0,
      "step": 19
    },
    {
      "epoch": 0.06791171477079797,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.06791171477079797,
      "eval_loss": NaN,
      "eval_runtime": 3.5133,
      "eval_samples_per_second": 70.588,
      "eval_steps_per_second": 8.824,
      "step": 20
    },
    {
      "epoch": 0.07130730050933787,
      "grad_norm": NaN,
      "learning_rate": 8.247240241650918e-05,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.07470288624787776,
      "grad_norm": NaN,
      "learning_rate": 7.938926261462366e-05,
      "loss": 0.0,
      "step": 22
    },
    {
      "epoch": 0.07809847198641766,
      "grad_norm": NaN,
      "learning_rate": 7.612492823579745e-05,
      "loss": 0.0,
      "step": 23
    },
    {
      "epoch": 0.08149405772495756,
      "grad_norm": NaN,
      "learning_rate": 7.269952498697734e-05,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.08488964346349745,
      "grad_norm": NaN,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.08488964346349745,
      "eval_loss": NaN,
      "eval_runtime": 3.5006,
      "eval_samples_per_second": 70.845,
      "eval_steps_per_second": 8.856,
      "step": 25
    }
  ],
  "logging_steps": 1,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 294850488434688.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}