trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 11.940298507462687,
  "eval_steps": 500,
  "global_step": 1600,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.746268656716418,
      "grad_norm": 0.2015659064054489,
      "learning_rate": 3.73134328358209e-06,
      "loss": 0.1198,
      "step": 100
    },
    {
      "epoch": 1.4925373134328357,
      "grad_norm": 0.16025546193122864,
      "learning_rate": 7.46268656716418e-06,
      "loss": 0.0693,
      "step": 200
    },
    {
      "epoch": 2.2388059701492535,
      "grad_norm": 0.1018073558807373,
      "learning_rate": 9.867330016583748e-06,
      "loss": 0.0496,
      "step": 300
    },
    {
      "epoch": 2.9850746268656714,
      "grad_norm": 0.100601427257061,
      "learning_rate": 9.45273631840796e-06,
      "loss": 0.0341,
      "step": 400
    },
    {
      "epoch": 3.7313432835820897,
      "grad_norm": 0.0992753654718399,
      "learning_rate": 9.038142620232173e-06,
      "loss": 0.0216,
      "step": 500
    },
    {
      "epoch": 4.477611940298507,
      "grad_norm": 0.08199736475944519,
      "learning_rate": 8.623548922056384e-06,
      "loss": 0.0187,
      "step": 600
    },
    {
      "epoch": 5.223880597014926,
      "grad_norm": 0.08397415280342102,
      "learning_rate": 8.208955223880599e-06,
      "loss": 0.0166,
      "step": 700
    },
    {
      "epoch": 5.970149253731344,
      "grad_norm": 0.13193649053573608,
      "learning_rate": 7.79436152570481e-06,
      "loss": 0.013,
      "step": 800
    },
    {
      "epoch": 6.7164179104477615,
      "grad_norm": 0.06592541933059692,
      "learning_rate": 7.3797678275290215e-06,
      "loss": 0.0109,
      "step": 900
    },
    {
      "epoch": 7.462686567164179,
      "grad_norm": 0.0223555825650692,
      "learning_rate": 6.965174129353235e-06,
      "loss": 0.0105,
      "step": 1000
    },
    {
      "epoch": 8.208955223880597,
      "grad_norm": 0.021410465240478516,
      "learning_rate": 6.550580431177447e-06,
      "loss": 0.0098,
      "step": 1100
    },
    {
      "epoch": 8.955223880597014,
      "grad_norm": 0.09453420341014862,
      "learning_rate": 6.135986733001659e-06,
      "loss": 0.0091,
      "step": 1200
    },
    {
      "epoch": 9.701492537313433,
      "grad_norm": 0.06435668468475342,
      "learning_rate": 5.721393034825871e-06,
      "loss": 0.0084,
      "step": 1300
    },
    {
      "epoch": 10.447761194029852,
      "grad_norm": 0.09910629689693451,
      "learning_rate": 5.306799336650083e-06,
      "loss": 0.0088,
      "step": 1400
    },
    {
      "epoch": 11.194029850746269,
      "grad_norm": 0.03590040281414986,
      "learning_rate": 4.892205638474296e-06,
      "loss": 0.0071,
      "step": 1500
    },
    {
      "epoch": 11.940298507462687,
      "grad_norm": 0.04040609300136566,
      "learning_rate": 4.477611940298508e-06,
      "loss": 0.0073,
      "step": 1600
    }
  ],
  "logging_steps": 100,
  "max_steps": 2680,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 800,
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}