{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 3748,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.026689797824781477,
      "grad_norm": 2.147552967071533,
      "learning_rate": 2.6133333333333334e-06,
      "loss": 0.7325,
      "step": 50
    },
    {
      "epoch": 0.053379595649562954,
      "grad_norm": 1.0603501796722412,
      "learning_rate": 5.28e-06,
      "loss": 0.3367,
      "step": 100
    },
    {
      "epoch": 0.08006939347434443,
      "grad_norm": 0.6418544054031372,
      "learning_rate": 7.946666666666666e-06,
      "loss": 0.3204,
      "step": 150
    },
    {
      "epoch": 0.10675919129912591,
      "grad_norm": 0.5255997776985168,
      "learning_rate": 1.0613333333333334e-05,
      "loss": 0.3304,
      "step": 200
    },
    {
      "epoch": 0.13344898912390737,
      "grad_norm": 0.3151794970035553,
      "learning_rate": 1.3280000000000002e-05,
      "loss": 0.2894,
      "step": 250
    },
    {
      "epoch": 0.16013878694868885,
      "grad_norm": 0.8053786754608154,
      "learning_rate": 1.5946666666666668e-05,
      "loss": 0.2991,
      "step": 300
    },
    {
      "epoch": 0.18682858477347034,
      "grad_norm": 0.02903924137353897,
      "learning_rate": 1.8613333333333334e-05,
      "loss": 0.3086,
      "step": 350
    },
    {
      "epoch": 0.21351838259825182,
      "grad_norm": 0.7904114723205566,
      "learning_rate": 1.9997501717954778e-05,
      "loss": 0.3291,
      "step": 400
    },
    {
      "epoch": 0.2402081804230333,
      "grad_norm": 0.41448771953582764,
      "learning_rate": 1.9976257383527736e-05,
      "loss": 0.3051,
      "step": 450
    },
    {
      "epoch": 0.26689797824781475,
      "grad_norm": 0.3972539007663727,
      "learning_rate": 1.9933381071415265e-05,
      "loss": 0.3149,
      "step": 500
    },
    {
      "epoch": 0.26689797824781475,
      "eval_loss": 0.4319009780883789,
      "eval_runtime": 118.7587,
      "eval_samples_per_second": 29.185,
      "eval_steps_per_second": 29.185,
      "step": 500
    },
    {
      "epoch": 0.29358777607259623,
      "grad_norm": 0.9960196018218994,
      "learning_rate": 1.9868965752296757e-05,
      "loss": 0.2993,
      "step": 550
    },
    {
      "epoch": 0.3202775738973777,
      "grad_norm": 0.4068482518196106,
      "learning_rate": 1.978315110087108e-05,
      "loss": 0.289,
      "step": 600
    },
    {
      "epoch": 0.3469673717221592,
      "grad_norm": 0.16865970194339752,
      "learning_rate": 1.967612319299347e-05,
      "loss": 0.3031,
      "step": 650
    },
    {
      "epoch": 0.37365716954694067,
      "grad_norm": 0.5249218344688416,
      "learning_rate": 1.954811410219871e-05,
      "loss": 0.296,
      "step": 700
    },
    {
      "epoch": 0.40034696737172215,
      "grad_norm": 0.2836828827857971,
      "learning_rate": 1.9399401396485418e-05,
      "loss": 0.2986,
      "step": 750
    },
    {
      "epoch": 0.42703676519650363,
      "grad_norm": 0.024336909875273705,
      "learning_rate": 1.923030753645258e-05,
      "loss": 0.2963,
      "step": 800
    },
    {
      "epoch": 0.4537265630212851,
      "grad_norm": 0.15610457956790924,
      "learning_rate": 1.9041199176093403e-05,
      "loss": 0.2939,
      "step": 850
    },
    {
      "epoch": 0.4804163608460666,
      "grad_norm": 0.3119260370731354,
      "learning_rate": 1.8832486367762608e-05,
      "loss": 0.3027,
      "step": 900
    },
    {
      "epoch": 0.5071061586708481,
      "grad_norm": 1.0931050777435303,
      "learning_rate": 1.8604621673041056e-05,
      "loss": 0.3102,
      "step": 950
    },
    {
      "epoch": 0.5337959564956295,
      "grad_norm": 0.16979913413524628,
      "learning_rate": 1.8358099181425628e-05,
      "loss": 0.2789,
      "step": 1000
    },
    {
      "epoch": 0.5337959564956295,
      "eval_loss": 0.42964276671409607,
      "eval_runtime": 118.8013,
      "eval_samples_per_second": 29.175,
      "eval_steps_per_second": 29.175,
      "step": 1000
    },
    {
      "epoch": 0.560485754320411,
      "grad_norm": 0.5423979163169861,
      "learning_rate": 1.809345343897229e-05,
      "loss": 0.2991,
      "step": 1050
    },
    {
      "epoch": 0.5871755521451925,
      "grad_norm": 0.9803158640861511,
      "learning_rate": 1.7811258289215265e-05,
      "loss": 0.2987,
      "step": 1100
    },
    {
      "epoch": 0.613865349969974,
      "grad_norm": 0.25261151790618896,
      "learning_rate": 1.7512125628875722e-05,
      "loss": 0.3019,
      "step": 1150
    },
    {
      "epoch": 0.6405551477947554,
      "grad_norm": 0.34850814938545227,
      "learning_rate": 1.7196704081057955e-05,
      "loss": 0.2888,
      "step": 1200
    },
    {
      "epoch": 0.667244945619537,
      "grad_norm": 0.32236334681510925,
      "learning_rate": 1.6865677588810112e-05,
      "loss": 0.2895,
      "step": 1250
    },
    {
      "epoch": 0.6939347434443184,
      "grad_norm": 0.32494327425956726,
      "learning_rate": 1.6519763932099e-05,
      "loss": 0.2983,
      "step": 1300
    },
    {
      "epoch": 0.7206245412690999,
      "grad_norm": 0.022694284096360207,
      "learning_rate": 1.615971317141477e-05,
      "loss": 0.2883,
      "step": 1350
    },
    {
      "epoch": 0.7473143390938813,
      "grad_norm": 0.02763156034052372,
      "learning_rate": 1.578630602138029e-05,
      "loss": 0.3111,
      "step": 1400
    },
    {
      "epoch": 0.7740041369186629,
      "grad_norm": 0.4600750207901001,
      "learning_rate": 1.54003521578917e-05,
      "loss": 0.3019,
      "step": 1450
    },
    {
      "epoch": 0.8006939347434443,
      "grad_norm": 0.024699728935956955,
      "learning_rate": 1.5002688462460931e-05,
      "loss": 0.3074,
      "step": 1500
    },
    {
      "epoch": 0.8006939347434443,
      "eval_loss": 0.42814743518829346,
      "eval_runtime": 118.8349,
      "eval_samples_per_second": 29.167,
      "eval_steps_per_second": 29.167,
      "step": 1500
    },
    {
      "epoch": 0.8273837325682258,
      "grad_norm": 0.43051406741142273,
      "learning_rate": 1.459417720756705e-05,
      "loss": 0.2929,
      "step": 1550
    },
    {
      "epoch": 0.8540735303930073,
      "grad_norm": 0.03149699047207832,
      "learning_rate": 1.4175704186951178e-05,
      "loss": 0.2858,
      "step": 1600
    },
    {
      "epoch": 0.8807633282177888,
      "grad_norm": 0.30078360438346863,
      "learning_rate": 1.3748176794909173e-05,
      "loss": 0.296,
      "step": 1650
    },
    {
      "epoch": 0.9074531260425702,
      "grad_norm": 0.2556113302707672,
      "learning_rate": 1.3312522058746883e-05,
      "loss": 0.2844,
      "step": 1700
    },
    {
      "epoch": 0.9341429238673518,
      "grad_norm": 0.09224811941385269,
      "learning_rate": 1.2869684628664158e-05,
      "loss": 0.2997,
      "step": 1750
    },
    {
      "epoch": 0.9608327216921332,
      "grad_norm": 0.4570249319076538,
      "learning_rate": 1.2420624729426419e-05,
      "loss": 0.286,
      "step": 1800
    },
    {
      "epoch": 0.9875225195169146,
      "grad_norm": 0.013208149001002312,
      "learning_rate": 1.1966316078265114e-05,
      "loss": 0.2889,
      "step": 1850
    },
    {
      "epoch": 1.0138786948688863,
      "grad_norm": 0.049777038395404816,
      "learning_rate": 1.150774377352188e-05,
      "loss": 0.2915,
      "step": 1900
    },
    {
      "epoch": 1.0405684926936678,
      "grad_norm": 0.019075842574238777,
      "learning_rate": 1.1045902158614493e-05,
      "loss": 0.2997,
      "step": 1950
    },
    {
      "epoch": 1.0672582905184493,
      "grad_norm": 0.3245074152946472,
      "learning_rate": 1.0581792665956311e-05,
      "loss": 0.2819,
      "step": 2000
    },
    {
      "epoch": 1.0672582905184493,
      "eval_loss": 0.4269392490386963,
      "eval_runtime": 118.8174,
      "eval_samples_per_second": 29.171,
      "eval_steps_per_second": 29.171,
      "step": 2000
    },
    {
      "epoch": 1.0939480883432309,
      "grad_norm": 0.1012788936495781,
      "learning_rate": 1.0116421645504322e-05,
      "loss": 0.3026,
      "step": 2050
    },
    {
      "epoch": 1.1206378861680122,
      "grad_norm": 0.005864244420081377,
      "learning_rate": 9.650798182644238e-06,
      "loss": 0.2996,
      "step": 2100
    },
    {
      "epoch": 1.1473276839927937,
      "grad_norm": 0.20553366839885712,
      "learning_rate": 9.185931910144259e-06,
      "loss": 0.2947,
      "step": 2150
    },
    {
      "epoch": 1.1740174818175753,
      "grad_norm": 0.1598707139492035,
      "learning_rate": 8.722830818921908e-06,
      "loss": 0.3082,
      "step": 2200
    },
    {
      "epoch": 1.2007072796423568,
      "grad_norm": 0.013660128228366375,
      "learning_rate": 8.262499072370962e-06,
      "loss": 0.2808,
      "step": 2250
    },
    {
      "epoch": 1.227397077467138,
      "grad_norm": 0.2775779068470001,
      "learning_rate": 7.805934828987778e-06,
      "loss": 0.2956,
      "step": 2300
    },
    {
      "epoch": 1.2540868752919196,
      "grad_norm": 0.11179913580417633,
      "learning_rate": 7.354128078018343e-06,
      "loss": 0.2766,
      "step": 2350
    },
    {
      "epoch": 1.2807766731167012,
      "grad_norm": 0.013183626346290112,
      "learning_rate": 6.908058492819043e-06,
      "loss": 0.3051,
      "step": 2400
    },
    {
      "epoch": 1.3074664709414825,
      "grad_norm": 0.030972259119153023,
      "learning_rate": 6.468693306585873e-06,
      "loss": 0.2966,
      "step": 2450
    },
    {
      "epoch": 1.334156268766264,
      "grad_norm": 0.004325371701270342,
      "learning_rate": 6.036985215058232e-06,
      "loss": 0.2993,
      "step": 2500
    },
    {
      "epoch": 1.334156268766264,
      "eval_loss": 0.4272440969944,
      "eval_runtime": 118.9153,
      "eval_samples_per_second": 29.147,
      "eval_steps_per_second": 29.147,
      "step": 2500
    },
    {
      "epoch": 1.3608460665910456,
      "grad_norm": 0.005641893949359655,
      "learning_rate": 5.613870310744911e-06,
      "loss": 0.2827,
      "step": 2550
    },
    {
      "epoch": 1.387535864415827,
      "grad_norm": 0.06420216709375381,
      "learning_rate": 5.20026605315167e-06,
      "loss": 0.2675,
      "step": 2600
    },
    {
      "epoch": 1.4142256622406086,
      "grad_norm": 0.7545241117477417,
      "learning_rate": 4.797069279411617e-06,
      "loss": 0.3023,
      "step": 2650
    },
    {
      "epoch": 1.44091546006539,
      "grad_norm": 0.01520050223916769,
      "learning_rate": 4.405154259631967e-06,
      "loss": 0.2743,
      "step": 2700
    },
    {
      "epoch": 1.4676052578901715,
      "grad_norm": 0.11380592733621597,
      "learning_rate": 4.0253708011739915e-06,
      "loss": 0.2878,
      "step": 2750
    },
    {
      "epoch": 1.494295055714953,
      "grad_norm": 0.013577022589743137,
      "learning_rate": 3.6585424059766296e-06,
      "loss": 0.2902,
      "step": 2800
    },
    {
      "epoch": 1.5209848535397343,
      "grad_norm": 0.020401863381266594,
      "learning_rate": 3.3054644849193495e-06,
      "loss": 0.2884,
      "step": 2850
    },
    {
      "epoch": 1.5476746513645159,
      "grad_norm": 0.006767068989574909,
      "learning_rate": 2.966902633096178e-06,
      "loss": 0.2952,
      "step": 2900
    },
    {
      "epoch": 1.5743644491892974,
      "grad_norm": 0.01109298225492239,
      "learning_rate": 2.643590969740637e-06,
      "loss": 0.2839,
      "step": 2950
    },
    {
      "epoch": 1.601054247014079,
      "grad_norm": 0.014958917163312435,
      "learning_rate": 2.33623054640124e-06,
      "loss": 0.2909,
      "step": 3000
    },
    {
      "epoch": 1.601054247014079,
      "eval_loss": 0.42682579159736633,
      "eval_runtime": 118.7962,
      "eval_samples_per_second": 29.176,
      "eval_steps_per_second": 29.176,
      "step": 3000
    },
    {
      "epoch": 1.6277440448388605,
      "grad_norm": 0.26821330189704895,
      "learning_rate": 2.0454878268191925e-06,
      "loss": 0.2818,
      "step": 3050
    },
    {
      "epoch": 1.6544338426636418,
      "grad_norm": 0.21184305846691132,
      "learning_rate": 1.7719932418044105e-06,
      "loss": 0.2839,
      "step": 3100
    },
    {
      "epoch": 1.6811236404884233,
      "grad_norm": 0.1029989942908287,
      "learning_rate": 1.516339822243398e-06,
      "loss": 0.2828,
      "step": 3150
    },
    {
      "epoch": 1.7078134383132046,
      "grad_norm": 0.03526095300912857,
      "learning_rate": 1.2790819132030974e-06,
      "loss": 0.2851,
      "step": 3200
    },
    {
      "epoch": 1.7345032361379862,
      "grad_norm": 0.004737787880003452,
      "learning_rate": 1.0607339719190002e-06,
      "loss": 0.2797,
      "step": 3250
    },
    {
      "epoch": 1.7611930339627677,
      "grad_norm": 0.010323897004127502,
      "learning_rate": 8.617694522738518e-07,
      "loss": 0.274,
      "step": 3300
    },
    {
      "epoch": 1.7878828317875493,
      "grad_norm": 0.6822851896286011,
      "learning_rate": 6.826197781858324e-07,
      "loss": 0.2727,
      "step": 3350
    },
    {
      "epoch": 1.8145726296123308,
      "grad_norm": 0.007676210254430771,
      "learning_rate": 5.236734081322281e-07,
      "loss": 0.2932,
      "step": 3400
    },
    {
      "epoch": 1.8412624274371123,
      "grad_norm": 0.45130208134651184,
      "learning_rate": 3.852749928370536e-07,
      "loss": 0.2885,
      "step": 3450
    },
    {
      "epoch": 1.8679522252618936,
      "grad_norm": 0.007974912412464619,
      "learning_rate": 2.677246279490309e-07,
      "loss": 0.2996,
      "step": 3500
    },
    {
      "epoch": 1.8679522252618936,
      "eval_loss": 0.42648929357528687,
      "eval_runtime": 118.9628,
      "eval_samples_per_second": 29.135,
      "eval_steps_per_second": 29.135,
      "step": 3500
    },
    {
      "epoch": 1.894642023086675,
      "grad_norm": 0.20535898208618164,
      "learning_rate": 1.7127720333040442e-07,
      "loss": 0.2885,
      "step": 3550
    },
    {
      "epoch": 1.9213318209114565,
      "grad_norm": 0.005906387697905302,
      "learning_rate": 9.614185036752155e-08,
      "loss": 0.2867,
      "step": 3600
    },
    {
      "epoch": 1.948021618736238,
      "grad_norm": 0.1817421317100525,
      "learning_rate": 4.248148850162892e-08,
      "loss": 0.273,
      "step": 3650
    },
    {
      "epoch": 1.9747114165610196,
      "grad_norm": 0.4754043221473694,
      "learning_rate": 1.041247196316264e-08,
      "loss": 0.3001,
      "step": 3700
    },
    {
      "epoch": 2.0,
      "step": 3748,
      "total_flos": 3.7463946988098355e+17,
      "train_loss": 0.3005621947244748,
      "train_runtime": 4284.8367,
      "train_samples_per_second": 6.995,
      "train_steps_per_second": 0.875
    }
  ],
  "logging_steps": 50,
  "max_steps": 3748,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.7463946988098355e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}