trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 42.10526315789474,
  "eval_steps": 500,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.1052631578947367,
      "grad_norm": 0.32421875,
      "learning_rate": 0.00019876883405951377,
      "loss": 0.9472,
      "step": 10
    },
    {
      "epoch": 4.2105263157894735,
      "grad_norm": 0.298828125,
      "learning_rate": 0.00019510565162951537,
      "loss": 0.744,
      "step": 20
    },
    {
      "epoch": 6.315789473684211,
      "grad_norm": 0.1695556640625,
      "learning_rate": 0.0001891006524188368,
      "loss": 0.6417,
      "step": 30
    },
    {
      "epoch": 8.421052631578947,
      "grad_norm": 0.128173828125,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.5931,
      "step": 40
    },
    {
      "epoch": 10.526315789473685,
      "grad_norm": 0.1119384765625,
      "learning_rate": 0.00017071067811865476,
      "loss": 0.5781,
      "step": 50
    },
    {
      "epoch": 12.631578947368421,
      "grad_norm": 0.12359619140625,
      "learning_rate": 0.00015877852522924732,
      "loss": 0.5593,
      "step": 60
    },
    {
      "epoch": 14.736842105263158,
      "grad_norm": 0.1312255859375,
      "learning_rate": 0.00014539904997395468,
      "loss": 0.5535,
      "step": 70
    },
    {
      "epoch": 16.842105263157894,
      "grad_norm": 0.1319580078125,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.5372,
      "step": 80
    },
    {
      "epoch": 18.94736842105263,
      "grad_norm": 0.1417236328125,
      "learning_rate": 0.0001156434465040231,
      "loss": 0.5316,
      "step": 90
    },
    {
      "epoch": 21.05263157894737,
      "grad_norm": 0.1668701171875,
      "learning_rate": 0.0001,
      "loss": 0.5192,
      "step": 100
    },
    {
      "epoch": 23.157894736842106,
      "grad_norm": 0.19091796875,
      "learning_rate": 8.435655349597689e-05,
      "loss": 0.5119,
      "step": 110
    },
    {
      "epoch": 25.263157894736842,
      "grad_norm": 0.1781005859375,
      "learning_rate": 6.909830056250527e-05,
      "loss": 0.5092,
      "step": 120
    },
    {
      "epoch": 27.36842105263158,
      "grad_norm": 0.174560546875,
      "learning_rate": 5.4600950026045326e-05,
      "loss": 0.4983,
      "step": 130
    },
    {
      "epoch": 29.473684210526315,
      "grad_norm": 0.186767578125,
      "learning_rate": 4.12214747707527e-05,
      "loss": 0.4911,
      "step": 140
    },
    {
      "epoch": 31.57894736842105,
      "grad_norm": 0.185302734375,
      "learning_rate": 2.9289321881345254e-05,
      "loss": 0.4875,
      "step": 150
    },
    {
      "epoch": 33.68421052631579,
      "grad_norm": 0.19384765625,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 0.4846,
      "step": 160
    },
    {
      "epoch": 35.78947368421053,
      "grad_norm": 0.1925048828125,
      "learning_rate": 1.0899347581163221e-05,
      "loss": 0.4835,
      "step": 170
    },
    {
      "epoch": 37.89473684210526,
      "grad_norm": 0.1820068359375,
      "learning_rate": 4.8943483704846475e-06,
      "loss": 0.4818,
      "step": 180
    },
    {
      "epoch": 40.0,
      "grad_norm": 0.1949462890625,
      "learning_rate": 1.231165940486234e-06,
      "loss": 0.481,
      "step": 190
    },
    {
      "epoch": 42.10526315789474,
      "grad_norm": 0.1903076171875,
      "learning_rate": 0.0,
      "loss": 0.4806,
      "step": 200
    },
    {
      "epoch": 42.10526315789474,
      "step": 200,
      "total_flos": 4.049973067697357e+16,
      "train_loss": 0.5557145524024963,
      "train_runtime": 584.5958,
      "train_samples_per_second": 10.948,
      "train_steps_per_second": 0.342
    }
  ],
  "logging_steps": 10,
  "max_steps": 200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "total_flos": 4.049973067697357e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}