trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.886075949367089,
  "eval_steps": 100,
  "global_step": 195,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.25316455696202533,
      "grad_norm": 14.92142105102539,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 2.4108,
      "step": 10
    },
    {
      "epoch": 0.5063291139240507,
      "grad_norm": 10.319029808044434,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 2.3752,
      "step": 20
    },
    {
      "epoch": 0.759493670886076,
      "grad_norm": 11.658527374267578,
      "learning_rate": 3e-06,
      "loss": 2.2968,
      "step": 30
    },
    {
      "epoch": 1.0,
      "grad_norm": 11.538076400756836,
      "learning_rate": 4.000000000000001e-06,
      "loss": 2.1359,
      "step": 40
    },
    {
      "epoch": 1.2531645569620253,
      "grad_norm": 15.206498146057129,
      "learning_rate": 5e-06,
      "loss": 2.1211,
      "step": 50
    },
    {
      "epoch": 1.5063291139240507,
      "grad_norm": 14.679222106933594,
      "learning_rate": 6e-06,
      "loss": 1.9345,
      "step": 60
    },
    {
      "epoch": 1.759493670886076,
      "grad_norm": 14.933147430419922,
      "learning_rate": 7.000000000000001e-06,
      "loss": 1.715,
      "step": 70
    },
    {
      "epoch": 2.0,
      "grad_norm": 9.749479293823242,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.5099,
      "step": 80
    },
    {
      "epoch": 2.2531645569620253,
      "grad_norm": 16.478574752807617,
      "learning_rate": 9e-06,
      "loss": 1.3413,
      "step": 90
    },
    {
      "epoch": 2.5063291139240507,
      "grad_norm": 11.706594467163086,
      "learning_rate": 1e-05,
      "loss": 1.0757,
      "step": 100
    },
    {
      "epoch": 2.5063291139240507,
      "eval_accuracy": 0.5857142857142857,
      "eval_f1": 0.5055477855477856,
      "eval_loss": 1.2395234107971191,
      "eval_runtime": 5.2819,
      "eval_samples_per_second": 13.253,
      "eval_steps_per_second": 1.704,
      "step": 100
    },
    {
      "epoch": 2.759493670886076,
      "grad_norm": 11.618650436401367,
      "learning_rate": 1.1000000000000001e-05,
      "loss": 0.9368,
      "step": 110
    },
    {
      "epoch": 3.0,
      "grad_norm": 14.1014986038208,
      "learning_rate": 1.2e-05,
      "loss": 0.7523,
      "step": 120
    },
    {
      "epoch": 3.2531645569620253,
      "grad_norm": 14.018274307250977,
      "learning_rate": 1.3000000000000001e-05,
      "loss": 0.5237,
      "step": 130
    },
    {
      "epoch": 3.5063291139240507,
      "grad_norm": 14.574962615966797,
      "learning_rate": 1.4000000000000001e-05,
      "loss": 0.5173,
      "step": 140
    },
    {
      "epoch": 3.759493670886076,
      "grad_norm": 29.751333236694336,
      "learning_rate": 1.5e-05,
      "loss": 0.577,
      "step": 150
    },
    {
      "epoch": 4.0,
      "grad_norm": 13.803457260131836,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.4925,
      "step": 160
    },
    {
      "epoch": 4.253164556962025,
      "grad_norm": 11.551584243774414,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 0.2708,
      "step": 170
    },
    {
      "epoch": 4.506329113924051,
      "grad_norm": 15.792594909667969,
      "learning_rate": 1.8e-05,
      "loss": 0.318,
      "step": 180
    },
    {
      "epoch": 4.759493670886076,
      "grad_norm": 7.049367904663086,
      "learning_rate": 1.9e-05,
      "loss": 0.3567,
      "step": 190
    }
  ],
  "logging_steps": 10,
  "max_steps": 195,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1673689415946240.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}