last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.012394645513138325,
  "eval_steps": 4,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0004957858205255329,
      "grad_norm": NaN,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.0004957858205255329,
      "eval_loss": NaN,
      "eval_runtime": 1218.3774,
      "eval_samples_per_second": 2.091,
      "eval_steps_per_second": 0.523,
      "step": 1
    },
    {
      "epoch": 0.0009915716410510659,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.001487357461576599,
      "grad_norm": NaN,
      "learning_rate": 3e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.0019831432821021317,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.0019831432821021317,
      "eval_loss": NaN,
      "eval_runtime": 1218.7463,
      "eval_samples_per_second": 2.091,
      "eval_steps_per_second": 0.523,
      "step": 4
    },
    {
      "epoch": 0.002478929102627665,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.002974714923153198,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.0034705007436787306,
      "grad_norm": NaN,
      "learning_rate": 7e-05,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.0039662865642042635,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.0039662865642042635,
      "eval_loss": NaN,
      "eval_runtime": 1219.5471,
      "eval_samples_per_second": 2.089,
      "eval_steps_per_second": 0.522,
      "step": 8
    },
    {
      "epoch": 0.004462072384729797,
      "grad_norm": NaN,
      "learning_rate": 9e-05,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.00495785820525533,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.005453644025780862,
      "grad_norm": NaN,
      "learning_rate": 9.98458666866564e-05,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.005949429846306396,
      "grad_norm": NaN,
      "learning_rate": 9.938441702975689e-05,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.005949429846306396,
      "eval_loss": NaN,
      "eval_runtime": 1219.0556,
      "eval_samples_per_second": 2.09,
      "eval_steps_per_second": 0.523,
      "step": 12
    },
    {
      "epoch": 0.006445215666831929,
      "grad_norm": NaN,
      "learning_rate": 9.861849601988383e-05,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.006941001487357461,
      "grad_norm": NaN,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 0.0074367873078829945,
      "grad_norm": NaN,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.007932573128408527,
      "grad_norm": NaN,
      "learning_rate": 9.45503262094184e-05,
      "loss": 0.0,
      "step": 16
    },
    {
      "epoch": 0.007932573128408527,
      "eval_loss": NaN,
      "eval_runtime": 1219.4314,
      "eval_samples_per_second": 2.089,
      "eval_steps_per_second": 0.522,
      "step": 16
    },
    {
      "epoch": 0.00842835894893406,
      "grad_norm": NaN,
      "learning_rate": 9.263200821770461e-05,
      "loss": 0.0,
      "step": 17
    },
    {
      "epoch": 0.008924144769459593,
      "grad_norm": NaN,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.009419930589985127,
      "grad_norm": NaN,
      "learning_rate": 8.802029828000156e-05,
      "loss": 0.0,
      "step": 19
    },
    {
      "epoch": 0.00991571641051066,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.00991571641051066,
      "eval_loss": NaN,
      "eval_runtime": 1219.0367,
      "eval_samples_per_second": 2.09,
      "eval_steps_per_second": 0.523,
      "step": 20
    },
    {
      "epoch": 0.010411502231036193,
      "grad_norm": NaN,
      "learning_rate": 8.247240241650918e-05,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.010907288051561725,
      "grad_norm": NaN,
      "learning_rate": 7.938926261462366e-05,
      "loss": 0.0,
      "step": 22
    },
    {
      "epoch": 0.011403073872087258,
      "grad_norm": NaN,
      "learning_rate": 7.612492823579745e-05,
      "loss": 0.0,
      "step": 23
    },
    {
      "epoch": 0.011898859692612791,
      "grad_norm": NaN,
      "learning_rate": 7.269952498697734e-05,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.011898859692612791,
      "eval_loss": NaN,
      "eval_runtime": 1219.4405,
      "eval_samples_per_second": 2.089,
      "eval_steps_per_second": 0.522,
      "step": 24
    },
    {
      "epoch": 0.012394645513138325,
      "grad_norm": NaN,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.0,
      "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": 1.119015678246912e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}