{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.5384615384615383,
  "eval_steps": 9,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03076923076923077,
      "grad_norm": 0.1967356950044632,
      "learning_rate": 1e-05,
      "loss": 10.3769,
      "step": 1
    },
    {
      "epoch": 0.03076923076923077,
      "eval_loss": 10.386632919311523,
      "eval_runtime": 0.079,
      "eval_samples_per_second": 1380.359,
      "eval_steps_per_second": 50.655,
      "step": 1
    },
    {
      "epoch": 0.06153846153846154,
      "grad_norm": 0.2022944986820221,
      "learning_rate": 2e-05,
      "loss": 10.3764,
      "step": 2
    },
    {
      "epoch": 0.09230769230769231,
      "grad_norm": 0.18788686394691467,
      "learning_rate": 3e-05,
      "loss": 10.382,
      "step": 3
    },
    {
      "epoch": 0.12307692307692308,
      "grad_norm": 0.19782385230064392,
      "learning_rate": 4e-05,
      "loss": 10.3854,
      "step": 4
    },
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 0.2000645250082016,
      "learning_rate": 5e-05,
      "loss": 10.3846,
      "step": 5
    },
    {
      "epoch": 0.18461538461538463,
      "grad_norm": 0.21505430340766907,
      "learning_rate": 6e-05,
      "loss": 10.3951,
      "step": 6
    },
    {
      "epoch": 0.2153846153846154,
      "grad_norm": 0.19195610284805298,
      "learning_rate": 7e-05,
      "loss": 10.4099,
      "step": 7
    },
    {
      "epoch": 0.24615384615384617,
      "grad_norm": 0.20345796644687653,
      "learning_rate": 8e-05,
      "loss": 10.4002,
      "step": 8
    },
    {
      "epoch": 0.27692307692307694,
      "grad_norm": 0.2024480700492859,
      "learning_rate": 9e-05,
      "loss": 10.4227,
      "step": 9
    },
    {
      "epoch": 0.27692307692307694,
      "eval_loss": 10.384278297424316,
      "eval_runtime": 0.0734,
      "eval_samples_per_second": 1484.068,
      "eval_steps_per_second": 54.461,
      "step": 9
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 0.20689363777637482,
      "learning_rate": 0.0001,
      "loss": 10.3974,
      "step": 10
    },
    {
      "epoch": 0.3384615384615385,
      "grad_norm": 0.2552158832550049,
      "learning_rate": 9.996740476948385e-05,
      "loss": 10.3596,
      "step": 11
    },
    {
      "epoch": 0.36923076923076925,
      "grad_norm": 0.2340978980064392,
      "learning_rate": 9.98696615758975e-05,
      "loss": 10.383,
      "step": 12
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.2439499795436859,
      "learning_rate": 9.970689785771798e-05,
      "loss": 10.3855,
      "step": 13
    },
    {
      "epoch": 0.4307692307692308,
      "grad_norm": 0.21027112007141113,
      "learning_rate": 9.947932582778188e-05,
      "loss": 10.3948,
      "step": 14
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 0.2078929841518402,
      "learning_rate": 9.918724219660013e-05,
      "loss": 10.3575,
      "step": 15
    },
    {
      "epoch": 0.49230769230769234,
      "grad_norm": 0.20251798629760742,
      "learning_rate": 9.883102778550434e-05,
      "loss": 10.3808,
      "step": 16
    },
    {
      "epoch": 0.5230769230769231,
      "grad_norm": 0.19867950677871704,
      "learning_rate": 9.841114703012817e-05,
      "loss": 10.3645,
      "step": 17
    },
    {
      "epoch": 0.5538461538461539,
      "grad_norm": 0.22018198668956757,
      "learning_rate": 9.792814737487207e-05,
      "loss": 10.4185,
      "step": 18
    },
    {
      "epoch": 0.5538461538461539,
      "eval_loss": 10.378647804260254,
      "eval_runtime": 0.0748,
      "eval_samples_per_second": 1456.458,
      "eval_steps_per_second": 53.448,
      "step": 18
    },
    {
      "epoch": 0.5846153846153846,
      "grad_norm": 0.24947059154510498,
      "learning_rate": 9.738265855914013e-05,
      "loss": 10.3531,
      "step": 19
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 0.2289813756942749,
      "learning_rate": 9.677539179628005e-05,
      "loss": 10.3866,
      "step": 20
    },
    {
      "epoch": 0.6461538461538462,
      "grad_norm": 0.22504808008670807,
      "learning_rate": 9.610713884629666e-05,
      "loss": 10.3638,
      "step": 21
    },
    {
      "epoch": 0.676923076923077,
      "grad_norm": 0.24337053298950195,
      "learning_rate": 9.537877098354786e-05,
      "loss": 10.3807,
      "step": 22
    },
    {
      "epoch": 0.7076923076923077,
      "grad_norm": 0.22690436244010925,
      "learning_rate": 9.459123786076912e-05,
      "loss": 10.3491,
      "step": 23
    },
    {
      "epoch": 0.7384615384615385,
      "grad_norm": 0.2087947279214859,
      "learning_rate": 9.374556627090749e-05,
      "loss": 10.3742,
      "step": 24
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 0.21224865317344666,
      "learning_rate": 9.284285880837946e-05,
      "loss": 10.3702,
      "step": 25
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.2412348985671997,
      "learning_rate": 9.188429243149824e-05,
      "loss": 10.3665,
      "step": 26
    },
    {
      "epoch": 0.8307692307692308,
      "grad_norm": 0.2303854376077652,
      "learning_rate": 9.087111692794459e-05,
      "loss": 10.3636,
      "step": 27
    },
    {
      "epoch": 0.8307692307692308,
      "eval_loss": 10.372750282287598,
      "eval_runtime": 0.0826,
      "eval_samples_per_second": 1319.489,
      "eval_steps_per_second": 48.422,
      "step": 27
    },
    {
      "epoch": 0.8615384615384616,
      "grad_norm": 0.2295921891927719,
      "learning_rate": 8.980465328528219e-05,
      "loss": 10.3577,
      "step": 28
    },
    {
      "epoch": 0.8923076923076924,
      "grad_norm": 0.2032311111688614,
      "learning_rate": 8.868629196864182e-05,
      "loss": 10.3765,
      "step": 29
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 0.22687426209449768,
      "learning_rate": 8.751749110782012e-05,
      "loss": 10.3815,
      "step": 30
    },
    {
      "epoch": 0.9538461538461539,
      "grad_norm": 0.22927479445934296,
      "learning_rate": 8.629977459615655e-05,
      "loss": 10.3654,
      "step": 31
    },
    {
      "epoch": 0.9846153846153847,
      "grad_norm": 0.23933324217796326,
      "learning_rate": 8.503473010366713e-05,
      "loss": 10.3814,
      "step": 32
    },
    {
      "epoch": 1.0153846153846153,
      "grad_norm": 0.32213953137397766,
      "learning_rate": 8.37240070070257e-05,
      "loss": 14.7436,
      "step": 33
    },
    {
      "epoch": 1.0461538461538462,
      "grad_norm": 0.2868359386920929,
      "learning_rate": 8.236931423909138e-05,
      "loss": 11.6626,
      "step": 34
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 0.21313560009002686,
      "learning_rate": 8.097241806078615e-05,
      "loss": 9.8745,
      "step": 35
    },
    {
      "epoch": 1.1076923076923078,
      "grad_norm": 0.2324608415365219,
      "learning_rate": 7.953513975822755e-05,
      "loss": 9.8428,
      "step": 36
    },
    {
      "epoch": 1.1076923076923078,
      "eval_loss": 10.366357803344727,
      "eval_runtime": 0.0742,
      "eval_samples_per_second": 1469.155,
      "eval_steps_per_second": 53.914,
      "step": 36
    },
    {
      "epoch": 1.1384615384615384,
      "grad_norm": 0.27083510160446167,
      "learning_rate": 7.805935326811912e-05,
      "loss": 9.634,
      "step": 37
    },
    {
      "epoch": 1.1692307692307693,
      "grad_norm": 0.30902621150016785,
      "learning_rate": 7.654698273449435e-05,
      "loss": 11.505,
      "step": 38
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.35695773363113403,
      "learning_rate": 7.500000000000001e-05,
      "loss": 11.8729,
      "step": 39
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 0.23437827825546265,
      "learning_rate": 7.342042203498951e-05,
      "loss": 9.4989,
      "step": 40
    },
    {
      "epoch": 1.2615384615384615,
      "grad_norm": 0.2547559440135956,
      "learning_rate": 7.181030830777837e-05,
      "loss": 8.6384,
      "step": 41
    },
    {
      "epoch": 1.2923076923076924,
      "grad_norm": 0.26535457372665405,
      "learning_rate": 7.017175809949044e-05,
      "loss": 9.577,
      "step": 42
    },
    {
      "epoch": 1.323076923076923,
      "grad_norm": 0.34555530548095703,
      "learning_rate": 6.850690776699573e-05,
      "loss": 13.4974,
      "step": 43
    },
    {
      "epoch": 1.353846153846154,
      "grad_norm": 0.2884383499622345,
      "learning_rate": 6.681792795750875e-05,
      "loss": 10.2425,
      "step": 44
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.26960813999176025,
      "learning_rate": 6.510702077847863e-05,
      "loss": 8.4271,
      "step": 45
    },
    {
      "epoch": 1.3846153846153846,
      "eval_loss": 10.359579086303711,
      "eval_runtime": 0.0835,
      "eval_samples_per_second": 1305.361,
      "eval_steps_per_second": 47.903,
      "step": 45
    },
    {
      "epoch": 1.4153846153846155,
      "grad_norm": 0.37105077505111694,
      "learning_rate": 6.337641692646106e-05,
      "loss": 10.7346,
      "step": 46
    },
    {
      "epoch": 1.4461538461538461,
      "grad_norm": 0.31993967294692993,
      "learning_rate": 6.162837277871553e-05,
      "loss": 10.932,
      "step": 47
    },
    {
      "epoch": 1.476923076923077,
      "grad_norm": 0.31407782435417175,
      "learning_rate": 5.9865167451320005e-05,
      "loss": 10.4683,
      "step": 48
    },
    {
      "epoch": 1.5076923076923077,
      "grad_norm": 0.29343822598457336,
      "learning_rate": 5.808909982763825e-05,
      "loss": 9.5898,
      "step": 49
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.40266671776771545,
      "learning_rate": 5.6302485561014475e-05,
      "loss": 12.5307,
      "step": 50
    }
  ],
  "logging_steps": 1,
  "max_steps": 97,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 10460489318400.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}