trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 196,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05118362124120281,
      "grad_norm": 4.597589015960693,
      "learning_rate": 4.5e-06,
      "loss": 0.7947,
      "step": 10
    },
    {
      "epoch": 0.10236724248240563,
      "grad_norm": 1.5900923013687134,
      "learning_rate": 9.5e-06,
      "loss": 0.6757,
      "step": 20
    },
    {
      "epoch": 0.15355086372360843,
      "grad_norm": 1.4815870523452759,
      "learning_rate": 9.935617890443557e-06,
      "loss": 0.5963,
      "step": 30
    },
    {
      "epoch": 0.20473448496481125,
      "grad_norm": 1.2894426584243774,
      "learning_rate": 9.715190263989562e-06,
      "loss": 0.5643,
      "step": 40
    },
    {
      "epoch": 0.2559181062060141,
      "grad_norm": 1.3262934684753418,
      "learning_rate": 9.344925248293837e-06,
      "loss": 0.5508,
      "step": 50
    },
    {
      "epoch": 0.30710172744721687,
      "grad_norm": 1.3410526514053345,
      "learning_rate": 8.836588973708129e-06,
      "loss": 0.5369,
      "step": 60
    },
    {
      "epoch": 0.3582853486884197,
      "grad_norm": 1.025560975074768,
      "learning_rate": 8.206335142623305e-06,
      "loss": 0.5291,
      "step": 70
    },
    {
      "epoch": 0.4094689699296225,
      "grad_norm": 1.1452240943908691,
      "learning_rate": 7.474191703716339e-06,
      "loss": 0.5237,
      "step": 80
    },
    {
      "epoch": 0.46065259117082535,
      "grad_norm": 0.9583773612976074,
      "learning_rate": 6.663424411982121e-06,
      "loss": 0.5133,
      "step": 90
    },
    {
      "epoch": 0.5118362124120281,
      "grad_norm": 0.8556775450706482,
      "learning_rate": 5.799797499079301e-06,
      "loss": 0.5135,
      "step": 100
    },
    {
      "epoch": 0.5630198336532309,
      "grad_norm": 0.7606714367866516,
      "learning_rate": 4.91075494810577e-06,
      "loss": 0.505,
      "step": 110
    },
    {
      "epoch": 0.6142034548944337,
      "grad_norm": 0.7366161346435547,
      "learning_rate": 4.02454838991936e-06,
      "loss": 0.5029,
      "step": 120
    },
    {
      "epoch": 0.6653870761356366,
      "grad_norm": 0.7205525040626526,
      "learning_rate": 3.169339334358105e-06,
      "loss": 0.4997,
      "step": 130
    },
    {
      "epoch": 0.7165706973768394,
      "grad_norm": 0.6518215537071228,
      "learning_rate": 2.372304265289436e-06,
      "loss": 0.4949,
      "step": 140
    },
    {
      "epoch": 0.7677543186180422,
      "grad_norm": 0.6340447664260864,
      "learning_rate": 1.6587710374121203e-06,
      "loss": 0.4961,
      "step": 150
    },
    {
      "epoch": 0.818937939859245,
      "grad_norm": 0.6365537047386169,
      "learning_rate": 1.0514140180404202e-06,
      "loss": 0.4935,
      "step": 160
    },
    {
      "epoch": 0.8701215611004478,
      "grad_norm": 0.5110655426979065,
      "learning_rate": 5.69533550325988e-07,
      "loss": 0.4949,
      "step": 170
    },
    {
      "epoch": 0.9213051823416507,
      "grad_norm": 0.530006468296051,
      "learning_rate": 2.2844263484068097e-07,
      "loss": 0.4971,
      "step": 180
    },
    {
      "epoch": 0.9724888035828535,
      "grad_norm": 0.5080718398094177,
      "learning_rate": 3.8980319302407976e-08,
      "loss": 0.4934,
      "step": 190
    },
    {
      "epoch": 1.0,
      "step": 196,
      "total_flos": 1.3677572943251505e+19,
      "train_loss": 0.5380685463243601,
      "train_runtime": 10684.4182,
      "train_samples_per_second": 2.34,
      "train_steps_per_second": 0.018
    }
  ],
  "logging_steps": 10,
  "max_steps": 196,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 196,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.3677572943251505e+19,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}