checkpoint-10000/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.29411764705882354,
  "eval_steps": 500,
  "global_step": 10000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.014705882352941176,
      "grad_norm": 0.35731062293052673,
      "learning_rate": 4.975490196078432e-05,
      "loss": 0.5354,
      "step": 500
    },
    {
      "epoch": 0.029411764705882353,
      "grad_norm": 2.1595866680145264,
      "learning_rate": 4.9509803921568634e-05,
      "loss": 0.1627,
      "step": 1000
    },
    {
      "epoch": 0.04411764705882353,
      "grad_norm": 0.14825384318828583,
      "learning_rate": 4.9264705882352944e-05,
      "loss": 0.101,
      "step": 1500
    },
    {
      "epoch": 0.058823529411764705,
      "grad_norm": 0.07121703773736954,
      "learning_rate": 4.901960784313725e-05,
      "loss": 0.0952,
      "step": 2000
    },
    {
      "epoch": 0.07352941176470588,
      "grad_norm": 0.06089532747864723,
      "learning_rate": 4.877450980392157e-05,
      "loss": 0.0881,
      "step": 2500
    },
    {
      "epoch": 0.08823529411764706,
      "grad_norm": 0.037034619599580765,
      "learning_rate": 4.8529411764705885e-05,
      "loss": 0.0717,
      "step": 3000
    },
    {
      "epoch": 0.10294117647058823,
      "grad_norm": 0.01999847963452339,
      "learning_rate": 4.82843137254902e-05,
      "loss": 0.0848,
      "step": 3500
    },
    {
      "epoch": 0.11764705882352941,
      "grad_norm": 0.018650399520993233,
      "learning_rate": 4.803921568627452e-05,
      "loss": 0.068,
      "step": 4000
    },
    {
      "epoch": 0.1323529411764706,
      "grad_norm": 0.023023229092359543,
      "learning_rate": 4.7794117647058826e-05,
      "loss": 0.0626,
      "step": 4500
    },
    {
      "epoch": 0.14705882352941177,
      "grad_norm": 0.018715515732765198,
      "learning_rate": 4.7549019607843135e-05,
      "loss": 0.0555,
      "step": 5000
    },
    {
      "epoch": 0.16176470588235295,
      "grad_norm": 0.11842140555381775,
      "learning_rate": 4.730392156862745e-05,
      "loss": 0.0658,
      "step": 5500
    },
    {
      "epoch": 0.17647058823529413,
      "grad_norm": 0.04816881939768791,
      "learning_rate": 4.705882352941177e-05,
      "loss": 0.0648,
      "step": 6000
    },
    {
      "epoch": 0.19117647058823528,
      "grad_norm": 0.025240018963813782,
      "learning_rate": 4.681372549019608e-05,
      "loss": 0.0632,
      "step": 6500
    },
    {
      "epoch": 0.20588235294117646,
      "grad_norm": 0.007390766404569149,
      "learning_rate": 4.656862745098039e-05,
      "loss": 0.0705,
      "step": 7000
    },
    {
      "epoch": 0.22058823529411764,
      "grad_norm": 0.011664963327348232,
      "learning_rate": 4.632352941176471e-05,
      "loss": 0.0552,
      "step": 7500
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 39.69175720214844,
      "learning_rate": 4.607843137254902e-05,
      "loss": 0.0686,
      "step": 8000
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.007366931065917015,
      "learning_rate": 4.5833333333333334e-05,
      "loss": 0.061,
      "step": 8500
    },
    {
      "epoch": 0.2647058823529412,
      "grad_norm": 0.003396671498194337,
      "learning_rate": 4.558823529411765e-05,
      "loss": 0.0464,
      "step": 9000
    },
    {
      "epoch": 0.27941176470588236,
      "grad_norm": 0.045366521924734116,
      "learning_rate": 4.5343137254901966e-05,
      "loss": 0.053,
      "step": 9500
    },
    {
      "epoch": 0.29411764705882354,
      "grad_norm": 1.137495756149292,
      "learning_rate": 4.5098039215686275e-05,
      "loss": 0.0484,
      "step": 10000
    }
  ],
  "logging_steps": 500,
  "max_steps": 102000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 6.19997036101632e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}