trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1407,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.806774509884417,
      "epoch": 0.21344717182497333,
      "grad_norm": 0.26953125,
      "learning_rate": 0.0001859275053304904,
      "loss": 0.8233,
      "mean_token_accuracy": 0.7962292967736722,
      "num_tokens": 3815906.0,
      "step": 100
    },
    {
      "entropy": 0.7081028638407588,
      "epoch": 0.42689434364994666,
      "grad_norm": 0.2333984375,
      "learning_rate": 0.0001717128642501777,
      "loss": 0.7291,
      "mean_token_accuracy": 0.8129953134059906,
      "num_tokens": 7626566.0,
      "step": 200
    },
    {
      "entropy": 0.6894140053912997,
      "epoch": 0.6403415154749199,
      "grad_norm": 0.25390625,
      "learning_rate": 0.00015749822316986497,
      "loss": 0.7077,
      "mean_token_accuracy": 0.8166575956344605,
      "num_tokens": 11456784.0,
      "step": 300
    },
    {
      "entropy": 0.6752224664017558,
      "epoch": 0.8537886872998933,
      "grad_norm": 0.248046875,
      "learning_rate": 0.00014328358208955225,
      "loss": 0.6934,
      "mean_token_accuracy": 0.819763490408659,
      "num_tokens": 15297842.0,
      "step": 400
    },
    {
      "entropy": 0.5943935247671664,
      "epoch": 1.0661686232657417,
      "grad_norm": 0.212890625,
      "learning_rate": 0.00012906894100923953,
      "loss": 0.5923,
      "mean_token_accuracy": 0.8431494048791914,
      "num_tokens": 19106846.0,
      "step": 500
    },
    {
      "entropy": 0.4187217493355274,
      "epoch": 1.279615795090715,
      "grad_norm": 0.2265625,
      "learning_rate": 0.0001148542999289268,
      "loss": 0.3844,
      "mean_token_accuracy": 0.892728195488453,
      "num_tokens": 22944206.0,
      "step": 600
    },
    {
      "entropy": 0.411714296489954,
      "epoch": 1.4930629669156883,
      "grad_norm": 0.2236328125,
      "learning_rate": 0.00010063965884861408,
      "loss": 0.3803,
      "mean_token_accuracy": 0.8937244528532028,
      "num_tokens": 26787706.0,
      "step": 700
    },
    {
      "entropy": 0.4113427203521132,
      "epoch": 1.7065101387406618,
      "grad_norm": 0.2353515625,
      "learning_rate": 8.642501776830136e-05,
      "loss": 0.3803,
      "mean_token_accuracy": 0.8936202809214592,
      "num_tokens": 30652821.0,
      "step": 800
    },
    {
      "entropy": 0.40327337332069874,
      "epoch": 1.9199573105656351,
      "grad_norm": 0.2470703125,
      "learning_rate": 7.221037668798864e-05,
      "loss": 0.3717,
      "mean_token_accuracy": 0.896033379137516,
      "num_tokens": 34453009.0,
      "step": 900
    },
    {
      "entropy": 0.3362068770518854,
      "epoch": 2.1323372465314834,
      "grad_norm": 0.2001953125,
      "learning_rate": 5.7995735607675904e-05,
      "loss": 0.2826,
      "mean_token_accuracy": 0.9232019716171763,
      "num_tokens": 38246767.0,
      "step": 1000
    },
    {
      "entropy": 0.282884752638638,
      "epoch": 2.3457844183564567,
      "grad_norm": 0.2109375,
      "learning_rate": 4.3781094527363184e-05,
      "loss": 0.2226,
      "mean_token_accuracy": 0.9403098250925541,
      "num_tokens": 42074506.0,
      "step": 1100
    },
    {
      "entropy": 0.27763547843322156,
      "epoch": 2.55923159018143,
      "grad_norm": 0.2138671875,
      "learning_rate": 2.9566453447050464e-05,
      "loss": 0.2188,
      "mean_token_accuracy": 0.9416492365300655,
      "num_tokens": 45883112.0,
      "step": 1200
    },
    {
      "entropy": 0.2768232163786888,
      "epoch": 2.7726787620064033,
      "grad_norm": 0.23828125,
      "learning_rate": 1.535181236673774e-05,
      "loss": 0.2191,
      "mean_token_accuracy": 0.9414364543557167,
      "num_tokens": 49691137.0,
      "step": 1300
    },
    {
      "entropy": 0.28273184813559055,
      "epoch": 2.9861259338313766,
      "grad_norm": 0.1962890625,
      "learning_rate": 1.1371712864250178e-06,
      "loss": 0.2225,
      "mean_token_accuracy": 0.9403267921507359,
      "num_tokens": 53554833.0,
      "step": 1400
    }
  ],
  "logging_steps": 100,
  "max_steps": 1407,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 1.0385484190580736e+18,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}