{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 416,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.24096385542168675,
      "grad_norm": 7.01993989944458,
      "learning_rate": 0.0005109898816807756,
      "loss": 900.5307,
      "mean_token_accuracy": 0.03312907457351685,
      "num_tokens": 1635565.0,
      "step": 25
    },
    {
      "epoch": 0.4819277108433735,
      "grad_norm": 0.3315601050853729,
      "learning_rate": 0.0005091581413578166,
      "loss": 0.516,
      "mean_token_accuracy": 0.02481155812740326,
      "num_tokens": 3269577.0,
      "step": 50
    },
    {
      "epoch": 0.7228915662650602,
      "grad_norm": 0.34700050950050354,
      "learning_rate": 0.0005041904358813742,
      "loss": 0.8675,
      "mean_token_accuracy": 0.0355426150560379,
      "num_tokens": 4905219.0,
      "step": 75
    },
    {
      "epoch": 0.963855421686747,
      "grad_norm": 1.0790128707885742,
      "learning_rate": 0.000496148180789635,
      "loss": 0.3346,
      "mean_token_accuracy": 0.04646581411361694,
      "num_tokens": 6540612.0,
      "step": 100
    },
    {
      "epoch": 1.0,
      "eval_runtime": 8.662,
      "eval_samples_per_second": 42.6,
      "eval_steps_per_second": 5.426,
      "step": 104
    },
    {
      "epoch": 1.202409638554217,
      "grad_norm": 0.0,
      "learning_rate": 0.0004851308021510392,
      "loss": 0.7956,
      "mean_token_accuracy": 0.04010541932751434,
      "num_tokens": 8123852.0,
      "step": 125
    },
    {
      "epoch": 1.4433734939759035,
      "grad_norm": 0.0,
      "learning_rate": 0.00047127450736389253,
      "loss": 16.2755,
      "mean_token_accuracy": 0.036344459056854246,
      "num_tokens": 9757781.0,
      "step": 150
    },
    {
      "epoch": 1.6843373493975904,
      "grad_norm": 0.0,
      "learning_rate": 0.0004547506012299107,
      "loss": 5.4428,
      "mean_token_accuracy": 0.0777336436510086,
      "num_tokens": 11393027.0,
      "step": 175
    },
    {
      "epoch": 1.9253012048192772,
      "grad_norm": 0.0,
      "learning_rate": 0.00043576336812001216,
      "loss": 3.1535,
      "mean_token_accuracy": 0.058714132905006405,
      "num_tokens": 13028615.0,
      "step": 200
    },
    {
      "epoch": 2.0,
      "eval_runtime": 8.6797,
      "eval_samples_per_second": 42.513,
      "eval_steps_per_second": 5.415,
      "step": 208
    },
    {
      "epoch": 2.163855421686747,
      "grad_norm": 0.0,
      "learning_rate": 0.0004145475464150544,
      "loss": 4.6324,
      "mean_token_accuracy": 0.034899900055894946,
      "num_tokens": 14611187.0,
      "step": 225
    },
    {
      "epoch": 2.404819277108434,
      "grad_norm": 0.0,
      "learning_rate": 0.00039136542644490095,
      "loss": 9.1752,
      "mean_token_accuracy": 0.02672989547252655,
      "num_tokens": 16247622.0,
      "step": 250
    },
    {
      "epoch": 2.6457831325301204,
      "grad_norm": 0.0,
      "learning_rate": 0.00036650360780387636,
      "loss": 0.3624,
      "mean_token_accuracy": 0.03860188961029053,
      "num_tokens": 17882680.0,
      "step": 275
    },
    {
      "epoch": 2.886746987951807,
      "grad_norm": 0.0,
      "learning_rate": 0.0003402694561317866,
      "loss": 0.5308,
      "mean_token_accuracy": 0.03807927131652832,
      "num_tokens": 19516398.0,
      "step": 300
    },
    {
      "epoch": 3.0,
      "eval_runtime": 8.6633,
      "eval_samples_per_second": 42.593,
      "eval_steps_per_second": 5.425,
      "step": 312
    },
    {
      "epoch": 3.125301204819277,
      "grad_norm": 0.6956835985183716,
      "learning_rate": 0.00031298730316517496,
      "loss": 0.4736,
      "mean_token_accuracy": 0.048145625928435666,
      "num_tokens": 21098092.0,
      "step": 325
    },
    {
      "epoch": 3.3662650602409636,
      "grad_norm": 0.1215895265340805,
      "learning_rate": 0.00028499443703742166,
      "loss": 0.329,
      "mean_token_accuracy": 0.02645561397075653,
      "num_tokens": 22732961.0,
      "step": 350
    },
    {
      "epoch": 3.6072289156626507,
      "grad_norm": 0.059805743396282196,
      "learning_rate": 0.0002566369323994442,
      "loss": 0.4127,
      "mean_token_accuracy": 0.03806757390499115,
      "num_tokens": 24367482.0,
      "step": 375
    },
    {
      "epoch": 3.8481927710843373,
      "grad_norm": 0.04674985259771347,
      "learning_rate": 0.0002282653719130422,
      "loss": 0.0992,
      "mean_token_accuracy": 0.03934468150138855,
      "num_tokens": 26003224.0,
      "step": 400
    },
    {
      "epoch": 4.0,
      "eval_runtime": 8.6842,
      "eval_samples_per_second": 42.491,
      "eval_steps_per_second": 5.412,
      "step": 416
    }
  ],
  "logging_steps": 25,
  "max_steps": 728,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.1783526738338775e+18,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}