{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.721170395869191,
  "eval_steps": 2000,
  "global_step": 8000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "grad_norm": 2.5756406784057617,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.5053,
      "step": 100
    },
    {
      "epoch": 0.04,
      "grad_norm": 2.708136796951294,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.4834,
      "step": 200
    },
    {
      "epoch": 0.06,
      "grad_norm": 2.605242967605591,
      "learning_rate": 1.2e-05,
      "loss": 0.4346,
      "step": 300
    },
    {
      "epoch": 0.09,
      "grad_norm": 3.3453047275543213,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.4465,
      "step": 400
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.8422973155975342,
      "learning_rate": 2e-05,
      "loss": 0.4542,
      "step": 500
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.822622299194336,
      "learning_rate": 1.98974358974359e-05,
      "loss": 0.4507,
      "step": 600
    },
    {
      "epoch": 0.15,
      "grad_norm": 2.805553674697876,
      "learning_rate": 1.9794871794871798e-05,
      "loss": 0.4556,
      "step": 700
    },
    {
      "epoch": 0.17,
      "grad_norm": 2.9561212062835693,
      "learning_rate": 1.9692307692307696e-05,
      "loss": 0.452,
      "step": 800
    },
    {
      "epoch": 0.19,
      "grad_norm": 4.11840295791626,
      "learning_rate": 1.958974358974359e-05,
      "loss": 0.4535,
      "step": 900
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.7729780673980713,
      "learning_rate": 1.9487179487179488e-05,
      "loss": 0.4735,
      "step": 1000
    },
    {
      "epoch": 0.24,
      "grad_norm": 1.458228588104248,
      "learning_rate": 1.9384615384615386e-05,
      "loss": 0.3946,
      "step": 1100
    },
    {
      "epoch": 0.26,
      "grad_norm": 2.6751554012298584,
      "learning_rate": 1.9282051282051284e-05,
      "loss": 0.4231,
      "step": 1200
    },
    {
      "epoch": 0.28,
      "grad_norm": 2.862396240234375,
      "learning_rate": 1.9179487179487182e-05,
      "loss": 0.3933,
      "step": 1300
    },
    {
      "epoch": 0.3,
      "grad_norm": 2.2546944618225098,
      "learning_rate": 1.907692307692308e-05,
      "loss": 0.4084,
      "step": 1400
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.8640025854110718,
      "learning_rate": 1.8974358974358975e-05,
      "loss": 0.4353,
      "step": 1500
    },
    {
      "epoch": 0.34,
      "grad_norm": 2.273890972137451,
      "learning_rate": 1.8871794871794873e-05,
      "loss": 0.4283,
      "step": 1600
    },
    {
      "epoch": 0.37,
      "grad_norm": 2.463329315185547,
      "learning_rate": 1.876923076923077e-05,
      "loss": 0.4219,
      "step": 1700
    },
    {
      "epoch": 0.39,
      "grad_norm": 2.782228946685791,
      "learning_rate": 1.866666666666667e-05,
      "loss": 0.4186,
      "step": 1800
    },
    {
      "epoch": 0.41,
      "grad_norm": 2.084019422531128,
      "learning_rate": 1.8564102564102567e-05,
      "loss": 0.4552,
      "step": 1900
    },
    {
      "epoch": 0.43,
      "grad_norm": 2.8400542736053467,
      "learning_rate": 1.8461538461538465e-05,
      "loss": 0.3935,
      "step": 2000
    },
    {
      "epoch": 0.43,
      "eval_loss": 0.3142565190792084,
      "eval_runtime": 65.2006,
      "eval_samples_per_second": 15.337,
      "eval_steps_per_second": 3.834,
      "step": 2000
    },
    {
      "epoch": 0.45,
      "grad_norm": 2.180964946746826,
      "learning_rate": 1.835897435897436e-05,
      "loss": 0.3924,
      "step": 2100
    },
    {
      "epoch": 0.47,
      "grad_norm": 2.9051389694213867,
      "learning_rate": 1.8256410256410257e-05,
      "loss": 0.3901,
      "step": 2200
    },
    {
      "epoch": 0.49,
      "grad_norm": 1.5432188510894775,
      "learning_rate": 1.8153846153846155e-05,
      "loss": 0.4155,
      "step": 2300
    },
    {
      "epoch": 0.52,
      "grad_norm": 2.0124869346618652,
      "learning_rate": 1.8051282051282053e-05,
      "loss": 0.4078,
      "step": 2400
    },
    {
      "epoch": 0.54,
      "grad_norm": 1.4548277854919434,
      "learning_rate": 1.794871794871795e-05,
      "loss": 0.4033,
      "step": 2500
    },
    {
      "epoch": 0.56,
      "grad_norm": 1.5465322732925415,
      "learning_rate": 1.784615384615385e-05,
      "loss": 0.4038,
      "step": 2600
    },
    {
      "epoch": 0.58,
      "grad_norm": 2.648297071456909,
      "learning_rate": 1.7743589743589744e-05,
      "loss": 0.3933,
      "step": 2700
    },
    {
      "epoch": 0.6,
      "grad_norm": 4.687922477722168,
      "learning_rate": 1.7641025641025642e-05,
      "loss": 0.4,
      "step": 2800
    },
    {
      "epoch": 0.62,
      "grad_norm": 1.872753620147705,
      "learning_rate": 1.753846153846154e-05,
      "loss": 0.3588,
      "step": 2900
    },
    {
      "epoch": 0.65,
      "grad_norm": 1.814994215965271,
      "learning_rate": 1.7435897435897438e-05,
      "loss": 0.4426,
      "step": 3000
    },
    {
      "epoch": 0.67,
      "grad_norm": 3.421119451522827,
      "learning_rate": 1.7333333333333336e-05,
      "loss": 0.3985,
      "step": 3100
    },
    {
      "epoch": 0.69,
      "grad_norm": 1.6136844158172607,
      "learning_rate": 1.7230769230769234e-05,
      "loss": 0.3522,
      "step": 3200
    },
    {
      "epoch": 0.71,
      "grad_norm": 2.4265120029449463,
      "learning_rate": 1.7128205128205128e-05,
      "loss": 0.3473,
      "step": 3300
    },
    {
      "epoch": 0.73,
      "grad_norm": 2.5966241359710693,
      "learning_rate": 1.7025641025641026e-05,
      "loss": 0.3807,
      "step": 3400
    },
    {
      "epoch": 0.75,
      "grad_norm": 1.95503568649292,
      "learning_rate": 1.6923076923076924e-05,
      "loss": 0.3731,
      "step": 3500
    },
    {
      "epoch": 0.77,
      "grad_norm": 1.2861559391021729,
      "learning_rate": 1.6820512820512822e-05,
      "loss": 0.3952,
      "step": 3600
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.7286114692687988,
      "learning_rate": 1.671794871794872e-05,
      "loss": 0.376,
      "step": 3700
    },
    {
      "epoch": 0.82,
      "grad_norm": 1.7622095346450806,
      "learning_rate": 1.6615384615384618e-05,
      "loss": 0.3517,
      "step": 3800
    },
    {
      "epoch": 0.84,
      "grad_norm": 1.7638356685638428,
      "learning_rate": 1.6512820512820513e-05,
      "loss": 0.3529,
      "step": 3900
    },
    {
      "epoch": 0.86,
      "grad_norm": 2.3966591358184814,
      "learning_rate": 1.641025641025641e-05,
      "loss": 0.3614,
      "step": 4000
    },
    {
      "epoch": 0.86,
      "eval_loss": 0.2630089223384857,
      "eval_runtime": 65.7983,
      "eval_samples_per_second": 15.198,
      "eval_steps_per_second": 3.799,
      "step": 4000
    },
    {
      "epoch": 0.88,
      "grad_norm": 2.331758499145508,
      "learning_rate": 1.630769230769231e-05,
      "loss": 0.3815,
      "step": 4100
    },
    {
      "epoch": 0.9,
      "grad_norm": 1.6653661727905273,
      "learning_rate": 1.6205128205128207e-05,
      "loss": 0.3701,
      "step": 4200
    },
    {
      "epoch": 0.93,
      "grad_norm": 1.6307231187820435,
      "learning_rate": 1.6102564102564105e-05,
      "loss": 0.3811,
      "step": 4300
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.9386046528816223,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.3408,
      "step": 4400
    },
    {
      "epoch": 0.97,
      "grad_norm": 1.8476628065109253,
      "learning_rate": 1.5897435897435897e-05,
      "loss": 0.3515,
      "step": 4500
    },
    {
      "epoch": 0.99,
      "grad_norm": 1.7403932809829712,
      "learning_rate": 1.5794871794871795e-05,
      "loss": 0.366,
      "step": 4600
    },
    {
      "epoch": 1.01,
      "grad_norm": 1.0889652967453003,
      "learning_rate": 1.5692307692307693e-05,
      "loss": 0.3001,
      "step": 4700
    },
    {
      "epoch": 1.03,
      "grad_norm": 1.335119605064392,
      "learning_rate": 1.558974358974359e-05,
      "loss": 0.218,
      "step": 4800
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.9958767890930176,
      "learning_rate": 1.548717948717949e-05,
      "loss": 0.2403,
      "step": 4900
    },
    {
      "epoch": 1.08,
      "grad_norm": 2.5801336765289307,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.228,
      "step": 5000
    },
    {
      "epoch": 1.1,
      "grad_norm": 2.4592084884643555,
      "learning_rate": 1.5282051282051282e-05,
      "loss": 0.2946,
      "step": 5100
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.4749181568622589,
      "learning_rate": 1.517948717948718e-05,
      "loss": 0.2307,
      "step": 5200
    },
    {
      "epoch": 1.14,
      "grad_norm": 2.7117793560028076,
      "learning_rate": 1.5076923076923078e-05,
      "loss": 0.2213,
      "step": 5300
    },
    {
      "epoch": 1.16,
      "grad_norm": 1.6409398317337036,
      "learning_rate": 1.4974358974358976e-05,
      "loss": 0.2448,
      "step": 5400
    },
    {
      "epoch": 1.18,
      "grad_norm": 2.0051097869873047,
      "learning_rate": 1.4871794871794874e-05,
      "loss": 0.23,
      "step": 5500
    },
    {
      "epoch": 1.2,
      "grad_norm": 1.724003791809082,
      "learning_rate": 1.4769230769230772e-05,
      "loss": 0.2273,
      "step": 5600
    },
    {
      "epoch": 1.23,
      "grad_norm": 1.6625137329101562,
      "learning_rate": 1.4666666666666666e-05,
      "loss": 0.231,
      "step": 5700
    },
    {
      "epoch": 1.25,
      "grad_norm": 2.87290358543396,
      "learning_rate": 1.4564102564102564e-05,
      "loss": 0.2426,
      "step": 5800
    },
    {
      "epoch": 1.27,
      "grad_norm": 1.323164701461792,
      "learning_rate": 1.4461538461538462e-05,
      "loss": 0.2498,
      "step": 5900
    },
    {
      "epoch": 1.29,
      "grad_norm": 1.9449268579483032,
      "learning_rate": 1.435897435897436e-05,
      "loss": 0.2241,
      "step": 6000
    },
    {
      "epoch": 1.29,
      "eval_loss": 0.24455103278160095,
      "eval_runtime": 65.1007,
      "eval_samples_per_second": 15.361,
      "eval_steps_per_second": 3.84,
      "step": 6000
    },
    {
      "epoch": 1.31,
      "grad_norm": 3.3016514778137207,
      "learning_rate": 1.4256410256410258e-05,
      "loss": 0.2341,
      "step": 6100
    },
    {
      "epoch": 1.33,
      "grad_norm": 1.8060470819473267,
      "learning_rate": 1.4153846153846156e-05,
      "loss": 0.2389,
      "step": 6200
    },
    {
      "epoch": 1.36,
      "grad_norm": 1.6190078258514404,
      "learning_rate": 1.405128205128205e-05,
      "loss": 0.2288,
      "step": 6300
    },
    {
      "epoch": 1.38,
      "grad_norm": 2.1221323013305664,
      "learning_rate": 1.3948717948717949e-05,
      "loss": 0.2344,
      "step": 6400
    },
    {
      "epoch": 1.4,
      "grad_norm": 1.716636300086975,
      "learning_rate": 1.3846153846153847e-05,
      "loss": 0.2553,
      "step": 6500
    },
    {
      "epoch": 1.42,
      "grad_norm": 2.039508581161499,
      "learning_rate": 1.3743589743589745e-05,
      "loss": 0.2504,
      "step": 6600
    },
    {
      "epoch": 1.44,
      "grad_norm": 2.224332809448242,
      "learning_rate": 1.3641025641025643e-05,
      "loss": 0.2327,
      "step": 6700
    },
    {
      "epoch": 1.46,
      "grad_norm": 2.1847877502441406,
      "learning_rate": 1.353846153846154e-05,
      "loss": 0.2363,
      "step": 6800
    },
    {
      "epoch": 1.48,
      "grad_norm": 1.154945731163025,
      "learning_rate": 1.3435897435897435e-05,
      "loss": 0.2528,
      "step": 6900
    },
    {
      "epoch": 1.51,
      "grad_norm": 2.06807279586792,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.2465,
      "step": 7000
    },
    {
      "epoch": 1.53,
      "grad_norm": 3.2610838413238525,
      "learning_rate": 1.3231794871794872e-05,
      "loss": 0.248,
      "step": 7100
    },
    {
      "epoch": 1.55,
      "grad_norm": 1.7421317100524902,
      "learning_rate": 1.312923076923077e-05,
      "loss": 0.2221,
      "step": 7200
    },
    {
      "epoch": 1.57,
      "grad_norm": 1.0780401229858398,
      "learning_rate": 1.3026666666666667e-05,
      "loss": 0.2245,
      "step": 7300
    },
    {
      "epoch": 1.59,
      "grad_norm": 2.143017292022705,
      "learning_rate": 1.2924102564102565e-05,
      "loss": 0.218,
      "step": 7400
    },
    {
      "epoch": 1.61,
      "grad_norm": 1.5114802122116089,
      "learning_rate": 1.2821538461538463e-05,
      "loss": 0.2222,
      "step": 7500
    },
    {
      "epoch": 1.64,
      "grad_norm": 1.4918652772903442,
      "learning_rate": 1.271897435897436e-05,
      "loss": 0.1958,
      "step": 7600
    },
    {
      "epoch": 1.66,
      "grad_norm": 1.8960070610046387,
      "learning_rate": 1.2616410256410257e-05,
      "loss": 0.2351,
      "step": 7700
    },
    {
      "epoch": 1.68,
      "grad_norm": 1.8487619161605835,
      "learning_rate": 1.2513846153846155e-05,
      "loss": 0.2293,
      "step": 7800
    },
    {
      "epoch": 1.7,
      "grad_norm": 2.432880163192749,
      "learning_rate": 1.2411282051282051e-05,
      "loss": 0.2344,
      "step": 7900
    },
    {
      "epoch": 1.72,
      "grad_norm": 2.5734400749206543,
      "learning_rate": 1.230871794871795e-05,
      "loss": 0.245,
      "step": 8000
    },
    {
      "epoch": 1.72,
      "eval_loss": 0.22669802606105804,
      "eval_runtime": 65.0878,
      "eval_samples_per_second": 15.364,
      "eval_steps_per_second": 3.841,
      "step": 8000
    }
  ],
  "logging_steps": 100,
  "max_steps": 20000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 2000,
  "total_flos": 2.5172343959165338e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}