trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.975609756097561,
  "eval_steps": 500,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04878048780487805,
      "grad_norm": 0.16131359338760376,
      "learning_rate": 5e-06,
      "loss": 0.3192,
      "step": 1
    },
    {
      "epoch": 0.0975609756097561,
      "grad_norm": 0.15827403962612152,
      "learning_rate": 1e-05,
      "loss": 0.2608,
      "step": 2
    },
    {
      "epoch": 0.14634146341463414,
      "grad_norm": 0.17321467399597168,
      "learning_rate": 9.924038765061042e-06,
      "loss": 0.2871,
      "step": 3
    },
    {
      "epoch": 0.1951219512195122,
      "grad_norm": 0.16689953207969666,
      "learning_rate": 9.698463103929542e-06,
      "loss": 0.3039,
      "step": 4
    },
    {
      "epoch": 0.24390243902439024,
      "grad_norm": 0.17536021769046783,
      "learning_rate": 9.330127018922195e-06,
      "loss": 0.3278,
      "step": 5
    },
    {
      "epoch": 0.2926829268292683,
      "grad_norm": 0.2078198343515396,
      "learning_rate": 8.83022221559489e-06,
      "loss": 0.3152,
      "step": 6
    },
    {
      "epoch": 0.34146341463414637,
      "grad_norm": 0.188552588224411,
      "learning_rate": 8.213938048432697e-06,
      "loss": 0.2712,
      "step": 7
    },
    {
      "epoch": 0.3902439024390244,
      "grad_norm": 0.19117948412895203,
      "learning_rate": 7.500000000000001e-06,
      "loss": 0.3114,
      "step": 8
    },
    {
      "epoch": 0.43902439024390244,
      "grad_norm": 0.20401953160762787,
      "learning_rate": 6.710100716628345e-06,
      "loss": 0.3357,
      "step": 9
    },
    {
      "epoch": 0.4878048780487805,
      "grad_norm": 0.19046278297901154,
      "learning_rate": 5.8682408883346535e-06,
      "loss": 0.2572,
      "step": 10
    },
    {
      "epoch": 0.5365853658536586,
      "grad_norm": 0.19299212098121643,
      "learning_rate": 5e-06,
      "loss": 0.27,
      "step": 11
    },
    {
      "epoch": 0.5853658536585366,
      "grad_norm": 0.18482521176338196,
      "learning_rate": 4.131759111665349e-06,
      "loss": 0.2629,
      "step": 12
    },
    {
      "epoch": 0.6341463414634146,
      "grad_norm": 0.15758441388607025,
      "learning_rate": 3.289899283371657e-06,
      "loss": 0.315,
      "step": 13
    },
    {
      "epoch": 0.6829268292682927,
      "grad_norm": 0.1637982428073883,
      "learning_rate": 2.5000000000000015e-06,
      "loss": 0.3473,
      "step": 14
    },
    {
      "epoch": 0.7317073170731707,
      "grad_norm": 0.17859426140785217,
      "learning_rate": 1.7860619515673034e-06,
      "loss": 0.3389,
      "step": 15
    },
    {
      "epoch": 0.7804878048780488,
      "grad_norm": 0.16861049830913544,
      "learning_rate": 1.1697777844051105e-06,
      "loss": 0.2522,
      "step": 16
    },
    {
      "epoch": 0.8292682926829268,
      "grad_norm": 0.17381973564624786,
      "learning_rate": 6.698729810778065e-07,
      "loss": 0.2758,
      "step": 17
    },
    {
      "epoch": 0.8780487804878049,
      "grad_norm": 0.16618315875530243,
      "learning_rate": 3.015368960704584e-07,
      "loss": 0.2548,
      "step": 18
    },
    {
      "epoch": 0.926829268292683,
      "grad_norm": 0.16824068129062653,
      "learning_rate": 7.59612349389599e-08,
      "loss": 0.3216,
      "step": 19
    },
    {
      "epoch": 0.975609756097561,
      "grad_norm": 0.1719992458820343,
      "learning_rate": 0.0,
      "loss": 0.3547,
      "step": 20
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.002307552949043e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}