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": 4.1302571296691895,
      "learning_rate": 4.5e-06,
      "loss": 1.1016,
      "step": 10
    },
    {
      "epoch": 0.10236724248240563,
      "grad_norm": 1.156183123588562,
      "learning_rate": 9.5e-06,
      "loss": 0.9459,
      "step": 20
    },
    {
      "epoch": 0.15355086372360843,
      "grad_norm": 0.9259429574012756,
      "learning_rate": 9.935617890443557e-06,
      "loss": 0.8445,
      "step": 30
    },
    {
      "epoch": 0.20473448496481125,
      "grad_norm": 0.8677472472190857,
      "learning_rate": 9.715190263989562e-06,
      "loss": 0.8,
      "step": 40
    },
    {
      "epoch": 0.2559181062060141,
      "grad_norm": 1.0556738376617432,
      "learning_rate": 9.344925248293837e-06,
      "loss": 0.7673,
      "step": 50
    },
    {
      "epoch": 0.30710172744721687,
      "grad_norm": 0.9036213755607605,
      "learning_rate": 8.836588973708129e-06,
      "loss": 0.7517,
      "step": 60
    },
    {
      "epoch": 0.3582853486884197,
      "grad_norm": 0.8972534537315369,
      "learning_rate": 8.206335142623305e-06,
      "loss": 0.7542,
      "step": 70
    },
    {
      "epoch": 0.4094689699296225,
      "grad_norm": 0.7269328236579895,
      "learning_rate": 7.474191703716339e-06,
      "loss": 0.738,
      "step": 80
    },
    {
      "epoch": 0.46065259117082535,
      "grad_norm": 0.6029865145683289,
      "learning_rate": 6.663424411982121e-06,
      "loss": 0.7288,
      "step": 90
    },
    {
      "epoch": 0.5118362124120281,
      "grad_norm": 0.7237837314605713,
      "learning_rate": 5.799797499079301e-06,
      "loss": 0.7279,
      "step": 100
    },
    {
      "epoch": 0.5630198336532309,
      "grad_norm": 0.6429140567779541,
      "learning_rate": 4.91075494810577e-06,
      "loss": 0.7382,
      "step": 110
    },
    {
      "epoch": 0.6142034548944337,
      "grad_norm": 0.647485613822937,
      "learning_rate": 4.02454838991936e-06,
      "loss": 0.7134,
      "step": 120
    },
    {
      "epoch": 0.6653870761356366,
      "grad_norm": 0.5843088030815125,
      "learning_rate": 3.169339334358105e-06,
      "loss": 0.7241,
      "step": 130
    },
    {
      "epoch": 0.7165706973768394,
      "grad_norm": 0.5550548434257507,
      "learning_rate": 2.372304265289436e-06,
      "loss": 0.7201,
      "step": 140
    },
    {
      "epoch": 0.7677543186180422,
      "grad_norm": 0.5591825246810913,
      "learning_rate": 1.6587710374121203e-06,
      "loss": 0.7056,
      "step": 150
    },
    {
      "epoch": 0.818937939859245,
      "grad_norm": 0.5044201612472534,
      "learning_rate": 1.0514140180404202e-06,
      "loss": 0.707,
      "step": 160
    },
    {
      "epoch": 0.8701215611004478,
      "grad_norm": 0.49126315116882324,
      "learning_rate": 5.69533550325988e-07,
      "loss": 0.7143,
      "step": 170
    },
    {
      "epoch": 0.9213051823416507,
      "grad_norm": 0.4847228229045868,
      "learning_rate": 2.2844263484068097e-07,
      "loss": 0.7036,
      "step": 180
    },
    {
      "epoch": 0.9724888035828535,
      "grad_norm": 0.4863457977771759,
      "learning_rate": 3.8980319302407976e-08,
      "loss": 0.702,
      "step": 190
    },
    {
      "epoch": 1.0,
      "step": 196,
      "total_flos": 8.284210212756259e+18,
      "train_loss": 0.7670857638728862,
      "train_runtime": 7137.9383,
      "train_samples_per_second": 3.502,
      "train_steps_per_second": 0.027
    }
  ],
  "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": 8.284210212756259e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}