last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.014342058085335245,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0014342058085335247,
      "grad_norm": 0.10692407935857773,
      "learning_rate": 2e-05,
      "loss": 11.9329,
      "step": 1
    },
    {
      "epoch": 0.0014342058085335247,
      "eval_loss": 11.946812629699707,
      "eval_runtime": 4.3539,
      "eval_samples_per_second": 67.526,
      "eval_steps_per_second": 33.763,
      "step": 1
    },
    {
      "epoch": 0.0028684116170670493,
      "grad_norm": 0.1532188206911087,
      "learning_rate": 4e-05,
      "loss": 11.9316,
      "step": 2
    },
    {
      "epoch": 0.004302617425600574,
      "grad_norm": 0.1192634180188179,
      "learning_rate": 6e-05,
      "loss": 11.9346,
      "step": 3
    },
    {
      "epoch": 0.004302617425600574,
      "eval_loss": 11.946681022644043,
      "eval_runtime": 3.9584,
      "eval_samples_per_second": 74.273,
      "eval_steps_per_second": 37.136,
      "step": 3
    },
    {
      "epoch": 0.005736823234134099,
      "grad_norm": 0.13755089044570923,
      "learning_rate": 8e-05,
      "loss": 11.9327,
      "step": 4
    },
    {
      "epoch": 0.007171029042667623,
      "grad_norm": 0.10198943316936493,
      "learning_rate": 0.0001,
      "loss": 11.9306,
      "step": 5
    },
    {
      "epoch": 0.008605234851201148,
      "grad_norm": 0.1001814678311348,
      "learning_rate": 0.00012,
      "loss": 11.9502,
      "step": 6
    },
    {
      "epoch": 0.008605234851201148,
      "eval_loss": 11.94594669342041,
      "eval_runtime": 4.0437,
      "eval_samples_per_second": 72.706,
      "eval_steps_per_second": 36.353,
      "step": 6
    },
    {
      "epoch": 0.010039440659734672,
      "grad_norm": 0.1335136592388153,
      "learning_rate": 0.00014,
      "loss": 11.9258,
      "step": 7
    },
    {
      "epoch": 0.011473646468268197,
      "grad_norm": 0.11691147834062576,
      "learning_rate": 0.00016,
      "loss": 11.9326,
      "step": 8
    },
    {
      "epoch": 0.012907852276801721,
      "grad_norm": 0.16463994979858398,
      "learning_rate": 0.00018,
      "loss": 11.9372,
      "step": 9
    },
    {
      "epoch": 0.012907852276801721,
      "eval_loss": 11.944936752319336,
      "eval_runtime": 3.9428,
      "eval_samples_per_second": 74.567,
      "eval_steps_per_second": 37.283,
      "step": 9
    },
    {
      "epoch": 0.014342058085335245,
      "grad_norm": 0.12192214280366898,
      "learning_rate": 0.0002,
      "loss": 11.9502,
      "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": 2471829700608.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}