{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 50,
  "global_step": 2073,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0964785335262904,
      "grad_norm": 0.7586275935173035,
      "learning_rate": 2.364864864864865e-06,
      "loss": 0.8531,
      "step": 50
    },
    {
      "epoch": 0.1929570670525808,
      "grad_norm": 1.5941067934036255,
      "learning_rate": 4.7779922779922784e-06,
      "loss": 0.5954,
      "step": 100
    },
    {
      "epoch": 0.2894356005788712,
      "grad_norm": 0.40901508927345276,
      "learning_rate": 7.191119691119691e-06,
      "loss": 0.3425,
      "step": 150
    },
    {
      "epoch": 0.3859141341051616,
      "grad_norm": 0.567613959312439,
      "learning_rate": 9.604247104247105e-06,
      "loss": 0.3213,
      "step": 200
    },
    {
      "epoch": 0.482392667631452,
      "grad_norm": 0.7472495436668396,
      "learning_rate": 1.2017374517374517e-05,
      "loss": 0.3009,
      "step": 250
    },
    {
      "epoch": 0.5788712011577424,
      "grad_norm": 0.8184729218482971,
      "learning_rate": 1.443050193050193e-05,
      "loss": 0.2567,
      "step": 300
    },
    {
      "epoch": 0.6753497346840328,
      "grad_norm": 0.6335634589195251,
      "learning_rate": 1.6843629343629346e-05,
      "loss": 0.2685,
      "step": 350
    },
    {
      "epoch": 0.7718282682103232,
      "grad_norm": 0.4045080840587616,
      "learning_rate": 1.916023166023166e-05,
      "loss": 0.289,
      "step": 400
    },
    {
      "epoch": 0.8683068017366136,
      "grad_norm": 0.5611971020698547,
      "learning_rate": 2.1573359073359074e-05,
      "loss": 0.2925,
      "step": 450
    },
    {
      "epoch": 0.964785335262904,
      "grad_norm": 0.7453213930130005,
      "learning_rate": 2.398648648648649e-05,
      "loss": 0.263,
      "step": 500
    },
    {
      "epoch": 0.9995176073323685,
      "eval_loss": 0.3949769139289856,
      "eval_runtime": 193.7983,
      "eval_samples_per_second": 1.192,
      "eval_steps_per_second": 0.15,
      "step": 518
    },
    {
      "epoch": 1.0612638687891944,
      "grad_norm": 0.6185714602470398,
      "learning_rate": 2.6399613899613903e-05,
      "loss": 0.2808,
      "step": 550
    },
    {
      "epoch": 1.1577424023154848,
      "grad_norm": 0.7272781729698181,
      "learning_rate": 2.8812741312741313e-05,
      "loss": 0.222,
      "step": 600
    },
    {
      "epoch": 1.2542209358417753,
      "grad_norm": 0.48480555415153503,
      "learning_rate": 3.122586872586873e-05,
      "loss": 0.252,
      "step": 650
    },
    {
      "epoch": 1.3506994693680656,
      "grad_norm": 0.6064177751541138,
      "learning_rate": 3.3638996138996145e-05,
      "loss": 0.2599,
      "step": 700
    },
    {
      "epoch": 1.447178002894356,
      "grad_norm": 0.5045720934867859,
      "learning_rate": 3.605212355212355e-05,
      "loss": 0.2588,
      "step": 750
    },
    {
      "epoch": 1.5436565364206465,
      "grad_norm": 0.4221721291542053,
      "learning_rate": 3.8465250965250966e-05,
      "loss": 0.2612,
      "step": 800
    },
    {
      "epoch": 1.6401350699469368,
      "grad_norm": 0.4660094976425171,
      "learning_rate": 4.087837837837838e-05,
      "loss": 0.2209,
      "step": 850
    },
    {
      "epoch": 1.7366136034732271,
      "grad_norm": 0.6750354766845703,
      "learning_rate": 4.3291505791505795e-05,
      "loss": 0.2152,
      "step": 900
    },
    {
      "epoch": 1.8330921369995177,
      "grad_norm": 0.522652268409729,
      "learning_rate": 4.5704633204633205e-05,
      "loss": 0.2418,
      "step": 950
    },
    {
      "epoch": 1.929570670525808,
      "grad_norm": 1.7187763452529907,
      "learning_rate": 4.811776061776062e-05,
      "loss": 0.2496,
      "step": 1000
    },
    {
      "epoch": 1.9990352146647372,
      "eval_loss": 0.36865848302841187,
      "eval_runtime": 192.2269,
      "eval_samples_per_second": 1.202,
      "eval_steps_per_second": 0.151,
      "step": 1036
    },
    {
      "epoch": 2.0260492040520983,
      "grad_norm": 0.980817973613739,
      "learning_rate": 4.9867277992277996e-05,
      "loss": 0.2725,
      "step": 1050
    },
    {
      "epoch": 2.122527737578389,
      "grad_norm": 0.4250291585922241,
      "learning_rate": 4.9263996138996145e-05,
      "loss": 0.2423,
      "step": 1100
    },
    {
      "epoch": 2.2190062711046794,
      "grad_norm": 0.4263651669025421,
      "learning_rate": 4.866071428571429e-05,
      "loss": 0.2347,
      "step": 1150
    },
    {
      "epoch": 2.3154848046309695,
      "grad_norm": 0.43781450390815735,
      "learning_rate": 4.8057432432432437e-05,
      "loss": 0.2025,
      "step": 1200
    },
    {
      "epoch": 2.41196333815726,
      "grad_norm": 0.49764326214790344,
      "learning_rate": 4.745415057915058e-05,
      "loss": 0.2336,
      "step": 1250
    },
    {
      "epoch": 2.5084418716835506,
      "grad_norm": 0.5718632340431213,
      "learning_rate": 4.685086872586873e-05,
      "loss": 0.2025,
      "step": 1300
    },
    {
      "epoch": 2.6049204052098407,
      "grad_norm": 0.5540716648101807,
      "learning_rate": 4.624758687258687e-05,
      "loss": 0.2312,
      "step": 1350
    },
    {
      "epoch": 2.7013989387361312,
      "grad_norm": 0.5893882513046265,
      "learning_rate": 4.5644305019305026e-05,
      "loss": 0.1978,
      "step": 1400
    },
    {
      "epoch": 2.797877472262422,
      "grad_norm": 0.5099292397499084,
      "learning_rate": 4.504102316602317e-05,
      "loss": 0.204,
      "step": 1450
    },
    {
      "epoch": 2.894356005788712,
      "grad_norm": 0.5618692636489868,
      "learning_rate": 4.443774131274132e-05,
      "loss": 0.225,
      "step": 1500
    },
    {
      "epoch": 2.9908345393150024,
      "grad_norm": 0.41460365056991577,
      "learning_rate": 4.383445945945946e-05,
      "loss": 0.2233,
      "step": 1550
    },
    {
      "epoch": 2.9985528219971056,
      "eval_loss": 0.356653094291687,
      "eval_runtime": 192.3283,
      "eval_samples_per_second": 1.201,
      "eval_steps_per_second": 0.151,
      "step": 1554
    },
    {
      "epoch": 3.087313072841293,
      "grad_norm": 0.5067611336708069,
      "learning_rate": 4.323117760617761e-05,
      "loss": 0.1889,
      "step": 1600
    },
    {
      "epoch": 3.183791606367583,
      "grad_norm": 0.4298361539840698,
      "learning_rate": 4.262789575289575e-05,
      "loss": 0.1774,
      "step": 1650
    },
    {
      "epoch": 3.2802701398938736,
      "grad_norm": 0.2998252213001251,
      "learning_rate": 4.20246138996139e-05,
      "loss": 0.1961,
      "step": 1700
    },
    {
      "epoch": 3.376748673420164,
      "grad_norm": 0.3195660412311554,
      "learning_rate": 4.142133204633205e-05,
      "loss": 0.2116,
      "step": 1750
    },
    {
      "epoch": 3.4732272069464543,
      "grad_norm": 0.39473968744277954,
      "learning_rate": 4.08180501930502e-05,
      "loss": 0.1959,
      "step": 1800
    },
    {
      "epoch": 3.569705740472745,
      "grad_norm": 0.3995039165019989,
      "learning_rate": 4.0214768339768347e-05,
      "loss": 0.202,
      "step": 1850
    },
    {
      "epoch": 3.6661842739990353,
      "grad_norm": 0.8156320452690125,
      "learning_rate": 3.961148648648649e-05,
      "loss": 0.2374,
      "step": 1900
    },
    {
      "epoch": 3.7626628075253254,
      "grad_norm": 0.3621225655078888,
      "learning_rate": 3.900820463320464e-05,
      "loss": 0.2155,
      "step": 1950
    },
    {
      "epoch": 3.859141341051616,
      "grad_norm": 0.6666299700737,
      "learning_rate": 3.840492277992278e-05,
      "loss": 0.2013,
      "step": 2000
    },
    {
      "epoch": 3.9556198745779065,
      "grad_norm": 0.5147814154624939,
      "learning_rate": 3.780164092664093e-05,
      "loss": 0.1901,
      "step": 2050
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.35413017868995667,
      "eval_runtime": 192.2471,
      "eval_samples_per_second": 1.202,
      "eval_steps_per_second": 0.151,
      "step": 2073
    }
  ],
  "logging_steps": 50,
  "max_steps": 5180,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 8.872900451228713e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}