{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.05491848038068492,
  "eval_steps": 8,
  "global_step": 88,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0006240736406896014,
      "eval_loss": 2.868605375289917,
      "eval_runtime": 574.6067,
      "eval_samples_per_second": 9.394,
      "eval_steps_per_second": 4.697,
      "step": 1
    },
    {
      "epoch": 0.001872220922068804,
      "grad_norm": 3.5454490184783936,
      "learning_rate": 3e-05,
      "loss": 2.8225,
      "step": 3
    },
    {
      "epoch": 0.003744441844137608,
      "grad_norm": 3.0127978324890137,
      "learning_rate": 6e-05,
      "loss": 2.5466,
      "step": 6
    },
    {
      "epoch": 0.004992589125516811,
      "eval_loss": 1.3012374639511108,
      "eval_runtime": 574.4376,
      "eval_samples_per_second": 9.397,
      "eval_steps_per_second": 4.699,
      "step": 8
    },
    {
      "epoch": 0.005616662766206412,
      "grad_norm": 2.2846736907958984,
      "learning_rate": 9e-05,
      "loss": 1.6608,
      "step": 9
    },
    {
      "epoch": 0.007488883688275216,
      "grad_norm": 1.0156153440475464,
      "learning_rate": 0.00012,
      "loss": 1.1555,
      "step": 12
    },
    {
      "epoch": 0.00936110461034402,
      "grad_norm": 0.878858208656311,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.9501,
      "step": 15
    },
    {
      "epoch": 0.009985178251033622,
      "eval_loss": 0.856473982334137,
      "eval_runtime": 574.4848,
      "eval_samples_per_second": 9.396,
      "eval_steps_per_second": 4.698,
      "step": 16
    },
    {
      "epoch": 0.011233325532412824,
      "grad_norm": 0.6313814520835876,
      "learning_rate": 0.00018,
      "loss": 0.9535,
      "step": 18
    },
    {
      "epoch": 0.013105546454481629,
      "grad_norm": 12.354997634887695,
      "learning_rate": 0.00019989930665413147,
      "loss": 0.998,
      "step": 21
    },
    {
      "epoch": 0.014977767376550433,
      "grad_norm": 0.4437302052974701,
      "learning_rate": 0.00019839295885986296,
      "loss": 0.8152,
      "step": 24
    },
    {
      "epoch": 0.014977767376550433,
      "eval_loss": 0.8276342153549194,
      "eval_runtime": 574.4332,
      "eval_samples_per_second": 9.397,
      "eval_steps_per_second": 4.699,
      "step": 24
    },
    {
      "epoch": 0.016849988298619237,
      "grad_norm": 0.40756919980049133,
      "learning_rate": 0.00019510565162951537,
      "loss": 0.8186,
      "step": 27
    },
    {
      "epoch": 0.01872220922068804,
      "grad_norm": 0.5308417081832886,
      "learning_rate": 0.0001900968867902419,
      "loss": 0.8643,
      "step": 30
    },
    {
      "epoch": 0.019970356502067244,
      "eval_loss": 0.7987203001976013,
      "eval_runtime": 575.0353,
      "eval_samples_per_second": 9.387,
      "eval_steps_per_second": 4.694,
      "step": 32
    },
    {
      "epoch": 0.020594430142756845,
      "grad_norm": 0.4313737452030182,
      "learning_rate": 0.00018345732537213027,
      "loss": 0.8467,
      "step": 33
    },
    {
      "epoch": 0.02246665106482565,
      "grad_norm": 0.43028154969215393,
      "learning_rate": 0.00017530714660036112,
      "loss": 0.8378,
      "step": 36
    },
    {
      "epoch": 0.024338871986894453,
      "grad_norm": 0.392749547958374,
      "learning_rate": 0.00016579387259397127,
      "loss": 0.8232,
      "step": 39
    },
    {
      "epoch": 0.024962945627584054,
      "eval_loss": 0.7788376212120056,
      "eval_runtime": 574.4954,
      "eval_samples_per_second": 9.396,
      "eval_steps_per_second": 4.698,
      "step": 40
    },
    {
      "epoch": 0.026211092908963257,
      "grad_norm": 0.6331055164337158,
      "learning_rate": 0.00015508969814521025,
      "loss": 0.8609,
      "step": 42
    },
    {
      "epoch": 0.02808331383103206,
      "grad_norm": 0.3308071494102478,
      "learning_rate": 0.00014338837391175582,
      "loss": 0.7784,
      "step": 45
    },
    {
      "epoch": 0.029955534753100865,
      "grad_norm": 0.3475176692008972,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.7766,
      "step": 48
    },
    {
      "epoch": 0.029955534753100865,
      "eval_loss": 0.7678617835044861,
      "eval_runtime": 574.3416,
      "eval_samples_per_second": 9.399,
      "eval_steps_per_second": 4.699,
      "step": 48
    },
    {
      "epoch": 0.03182775567516967,
      "grad_norm": 0.3206414580345154,
      "learning_rate": 0.00011785568947986367,
      "loss": 0.8059,
      "step": 51
    },
    {
      "epoch": 0.033699976597238473,
      "grad_norm": 0.330831378698349,
      "learning_rate": 0.00010448648303505151,
      "loss": 0.8291,
      "step": 54
    },
    {
      "epoch": 0.034948123878617676,
      "eval_loss": 0.7632117867469788,
      "eval_runtime": 574.4268,
      "eval_samples_per_second": 9.397,
      "eval_steps_per_second": 4.699,
      "step": 56
    },
    {
      "epoch": 0.03557219751930728,
      "grad_norm": 0.3289456069469452,
      "learning_rate": 9.103606910965666e-05,
      "loss": 0.7862,
      "step": 57
    },
    {
      "epoch": 0.03744441844137608,
      "grad_norm": 0.33111053705215454,
      "learning_rate": 7.774790660436858e-05,
      "loss": 0.8385,
      "step": 60
    },
    {
      "epoch": 0.039316639363444886,
      "grad_norm": 0.34499919414520264,
      "learning_rate": 6.486251759186572e-05,
      "loss": 0.8008,
      "step": 63
    },
    {
      "epoch": 0.03994071300413449,
      "eval_loss": 0.7575622797012329,
      "eval_runtime": 574.6227,
      "eval_samples_per_second": 9.394,
      "eval_steps_per_second": 4.697,
      "step": 64
    },
    {
      "epoch": 0.04118886028551369,
      "grad_norm": 0.35130053758621216,
      "learning_rate": 5.261313375270014e-05,
      "loss": 0.8259,
      "step": 66
    },
    {
      "epoch": 0.043061081207582494,
      "grad_norm": 0.32489344477653503,
      "learning_rate": 4.12214747707527e-05,
      "loss": 0.765,
      "step": 69
    },
    {
      "epoch": 0.0449333021296513,
      "grad_norm": 0.3194122910499573,
      "learning_rate": 3.089373510131354e-05,
      "loss": 0.7716,
      "step": 72
    },
    {
      "epoch": 0.0449333021296513,
      "eval_loss": 0.7537363767623901,
      "eval_runtime": 574.4019,
      "eval_samples_per_second": 9.398,
      "eval_steps_per_second": 4.699,
      "step": 72
    },
    {
      "epoch": 0.0468055230517201,
      "grad_norm": 0.39400896430015564,
      "learning_rate": 2.181685175319702e-05,
      "loss": 0.8015,
      "step": 75
    },
    {
      "epoch": 0.048677743973788906,
      "grad_norm": 0.382411390542984,
      "learning_rate": 1.415512063981339e-05,
      "loss": 0.8092,
      "step": 78
    },
    {
      "epoch": 0.04992589125516811,
      "eval_loss": 0.752012312412262,
      "eval_runtime": 574.415,
      "eval_samples_per_second": 9.397,
      "eval_steps_per_second": 4.699,
      "step": 80
    },
    {
      "epoch": 0.05054996489585771,
      "grad_norm": 0.31652024388313293,
      "learning_rate": 8.047222744854943e-06,
      "loss": 0.7595,
      "step": 81
    },
    {
      "epoch": 0.052422185817926514,
      "grad_norm": 0.3957296907901764,
      "learning_rate": 3.6037139304146762e-06,
      "loss": 0.802,
      "step": 84
    },
    {
      "epoch": 0.05429440673999532,
      "grad_norm": 0.32933101058006287,
      "learning_rate": 9.0502382320653e-07,
      "loss": 0.802,
      "step": 87
    },
    {
      "epoch": 0.05491848038068492,
      "eval_loss": 0.7515419125556946,
      "eval_runtime": 574.5854,
      "eval_samples_per_second": 9.395,
      "eval_steps_per_second": 4.697,
      "step": 88
    }
  ],
  "logging_steps": 3,
  "max_steps": 90,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 8,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.681769044200653e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}