{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 24,
  "global_step": 288,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.010416666666666666,
      "eval_loss": 1.9108808040618896,
      "eval_runtime": 3.3051,
      "eval_samples_per_second": 49.016,
      "eval_steps_per_second": 6.354,
      "step": 1
    },
    {
      "epoch": 0.03125,
      "grad_norm": 0.41242459416389465,
      "learning_rate": 1.5e-05,
      "loss": 1.8948,
      "step": 3
    },
    {
      "epoch": 0.0625,
      "grad_norm": 0.5248602032661438,
      "learning_rate": 3e-05,
      "loss": 1.9034,
      "step": 6
    },
    {
      "epoch": 0.09375,
      "grad_norm": 0.4381963312625885,
      "learning_rate": 4.5e-05,
      "loss": 1.8402,
      "step": 9
    },
    {
      "epoch": 0.125,
      "grad_norm": 0.3652794063091278,
      "learning_rate": 4.99936149886953e-05,
      "loss": 1.7817,
      "step": 12
    },
    {
      "epoch": 0.15625,
      "grad_norm": 0.30795106291770935,
      "learning_rate": 4.996010259708474e-05,
      "loss": 1.7055,
      "step": 15
    },
    {
      "epoch": 0.1875,
      "grad_norm": 0.35701417922973633,
      "learning_rate": 4.989790503518888e-05,
      "loss": 1.7031,
      "step": 18
    },
    {
      "epoch": 0.21875,
      "grad_norm": 0.3083578944206238,
      "learning_rate": 4.980709378298851e-05,
      "loss": 1.6675,
      "step": 21
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.2675102651119232,
      "learning_rate": 4.9687773204487064e-05,
      "loss": 1.6823,
      "step": 24
    },
    {
      "epoch": 0.25,
      "eval_loss": 1.6583374738693237,
      "eval_runtime": 3.3211,
      "eval_samples_per_second": 48.78,
      "eval_steps_per_second": 6.323,
      "step": 24
    },
    {
      "epoch": 0.28125,
      "grad_norm": 0.2701338827610016,
      "learning_rate": 4.954008042777125e-05,
      "loss": 1.666,
      "step": 27
    },
    {
      "epoch": 0.3125,
      "grad_norm": 0.35075831413269043,
      "learning_rate": 4.93641851874178e-05,
      "loss": 1.6302,
      "step": 30
    },
    {
      "epoch": 0.34375,
      "grad_norm": 0.31512945890426636,
      "learning_rate": 4.916028962942762e-05,
      "loss": 1.6324,
      "step": 33
    },
    {
      "epoch": 0.375,
      "grad_norm": 0.2673735022544861,
      "learning_rate": 4.892862807891131e-05,
      "loss": 1.5973,
      "step": 36
    },
    {
      "epoch": 0.40625,
      "grad_norm": 0.27960431575775146,
      "learning_rate": 4.866946677079314e-05,
      "loss": 1.5657,
      "step": 39
    },
    {
      "epoch": 0.4375,
      "grad_norm": 0.3442665934562683,
      "learning_rate": 4.8383103543843036e-05,
      "loss": 1.564,
      "step": 42
    },
    {
      "epoch": 0.46875,
      "grad_norm": 0.3022296130657196,
      "learning_rate": 4.806986749838807e-05,
      "loss": 1.557,
      "step": 45
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.30464908480644226,
      "learning_rate": 4.773011861809693e-05,
      "loss": 1.5419,
      "step": 48
    },
    {
      "epoch": 0.5,
      "eval_loss": 1.552250862121582,
      "eval_runtime": 3.3169,
      "eval_samples_per_second": 48.84,
      "eval_steps_per_second": 6.331,
      "step": 48
    },
    {
      "epoch": 0.53125,
      "grad_norm": 0.284551739692688,
      "learning_rate": 4.736424735627193e-05,
      "loss": 1.5701,
      "step": 51
    },
    {
      "epoch": 0.5625,
      "grad_norm": 0.3086010813713074,
      "learning_rate": 4.697267418712415e-05,
      "loss": 1.5359,
      "step": 54
    },
    {
      "epoch": 0.59375,
      "grad_norm": 0.31514284014701843,
      "learning_rate": 4.655584912254727e-05,
      "loss": 1.5334,
      "step": 57
    },
    {
      "epoch": 0.625,
      "grad_norm": 0.31487005949020386,
      "learning_rate": 4.611425119494551e-05,
      "loss": 1.5224,
      "step": 60
    },
    {
      "epoch": 0.65625,
      "grad_norm": 0.36550524830818176,
      "learning_rate": 4.564838790671e-05,
      "loss": 1.5222,
      "step": 63
    },
    {
      "epoch": 0.6875,
      "grad_norm": 0.29623621702194214,
      "learning_rate": 4.515879464697629e-05,
      "loss": 1.5206,
      "step": 66
    },
    {
      "epoch": 0.71875,
      "grad_norm": 0.3229241967201233,
      "learning_rate": 4.464603407633326e-05,
      "loss": 1.5232,
      "step": 69
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.2879750728607178,
      "learning_rate": 4.41106954801906e-05,
      "loss": 1.5028,
      "step": 72
    },
    {
      "epoch": 0.75,
      "eval_loss": 1.493626594543457,
      "eval_runtime": 3.3364,
      "eval_samples_per_second": 48.555,
      "eval_steps_per_second": 6.294,
      "step": 72
    },
    {
      "epoch": 0.78125,
      "grad_norm": 0.3321720063686371,
      "learning_rate": 4.355339409154788e-05,
      "loss": 1.4901,
      "step": 75
    },
    {
      "epoch": 0.8125,
      "grad_norm": 0.349568635225296,
      "learning_rate": 4.297477038394368e-05,
      "loss": 1.5304,
      "step": 78
    },
    {
      "epoch": 0.84375,
      "grad_norm": 0.38185468316078186,
      "learning_rate": 4.237548933539718e-05,
      "loss": 1.502,
      "step": 81
    },
    {
      "epoch": 0.875,
      "grad_norm": 0.37284818291664124,
      "learning_rate": 4.1756239664188273e-05,
      "loss": 1.4548,
      "step": 84
    },
    {
      "epoch": 0.90625,
      "grad_norm": 0.3401886522769928,
      "learning_rate": 4.1117733037354314e-05,
      "loss": 1.49,
      "step": 87
    },
    {
      "epoch": 0.9375,
      "grad_norm": 0.31730660796165466,
      "learning_rate": 4.0460703252813326e-05,
      "loss": 1.485,
      "step": 90
    },
    {
      "epoch": 0.96875,
      "grad_norm": 0.34687182307243347,
      "learning_rate": 3.978590539605338e-05,
      "loss": 1.5021,
      "step": 93
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.4444548487663269,
      "learning_rate": 3.9094114972357524e-05,
      "loss": 1.4711,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.452917218208313,
      "eval_runtime": 3.3544,
      "eval_samples_per_second": 48.294,
      "eval_steps_per_second": 6.26,
      "step": 96
    },
    {
      "epoch": 1.03125,
      "grad_norm": 0.38303127884864807,
      "learning_rate": 3.838612701556138e-05,
      "loss": 1.3992,
      "step": 99
    },
    {
      "epoch": 1.0625,
      "grad_norm": 0.34194791316986084,
      "learning_rate": 3.7662755174367795e-05,
      "loss": 1.397,
      "step": 102
    },
    {
      "epoch": 1.09375,
      "grad_norm": 0.3771461248397827,
      "learning_rate": 3.6924830777268425e-05,
      "loss": 1.4063,
      "step": 105
    },
    {
      "epoch": 1.125,
      "grad_norm": 0.36158788204193115,
      "learning_rate": 3.6173201877147134e-05,
      "loss": 1.4279,
      "step": 108
    },
    {
      "epoch": 1.15625,
      "grad_norm": 0.3729671835899353,
      "learning_rate": 3.540873227666288e-05,
      "loss": 1.4246,
      "step": 111
    },
    {
      "epoch": 1.1875,
      "grad_norm": 0.4584656357765198,
      "learning_rate": 3.463230053553241e-05,
      "loss": 1.3891,
      "step": 114
    },
    {
      "epoch": 1.21875,
      "grad_norm": 0.44859328866004944,
      "learning_rate": 3.384479896085353e-05,
      "loss": 1.4136,
      "step": 117
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.42308565974235535,
      "learning_rate": 3.30471325816293e-05,
      "loss": 1.4095,
      "step": 120
    },
    {
      "epoch": 1.25,
      "eval_loss": 1.425024390220642,
      "eval_runtime": 3.3273,
      "eval_samples_per_second": 48.688,
      "eval_steps_per_second": 6.311,
      "step": 120
    },
    {
      "epoch": 1.28125,
      "grad_norm": 0.3992334008216858,
      "learning_rate": 3.2240218108671684e-05,
      "loss": 1.4394,
      "step": 123
    },
    {
      "epoch": 1.3125,
      "grad_norm": 0.4292496144771576,
      "learning_rate": 3.142498288108007e-05,
      "loss": 1.4048,
      "step": 126
    },
    {
      "epoch": 1.34375,
      "grad_norm": 0.42058807611465454,
      "learning_rate": 3.060236380050519e-05,
      "loss": 1.3989,
      "step": 129
    },
    {
      "epoch": 1.375,
      "grad_norm": 0.49857133626937866,
      "learning_rate": 2.9773306254423515e-05,
      "loss": 1.3915,
      "step": 132
    },
    {
      "epoch": 1.40625,
      "grad_norm": 0.3670888841152191,
      "learning_rate": 2.893876302965925e-05,
      "loss": 1.3718,
      "step": 135
    },
    {
      "epoch": 1.4375,
      "grad_norm": 0.36883431673049927,
      "learning_rate": 2.8099693217402806e-05,
      "loss": 1.3771,
      "step": 138
    },
    {
      "epoch": 1.46875,
      "grad_norm": 0.4503380060195923,
      "learning_rate": 2.7257061110984007e-05,
      "loss": 1.4497,
      "step": 141
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.4777998924255371,
      "learning_rate": 2.6411835097666747e-05,
      "loss": 1.4025,
      "step": 144
    },
    {
      "epoch": 1.5,
      "eval_loss": 1.4068902730941772,
      "eval_runtime": 3.3295,
      "eval_samples_per_second": 48.656,
      "eval_steps_per_second": 6.307,
      "step": 144
    },
    {
      "epoch": 1.53125,
      "grad_norm": 0.4338071346282959,
      "learning_rate": 2.5564986545738766e-05,
      "loss": 1.4139,
      "step": 147
    },
    {
      "epoch": 1.5625,
      "grad_norm": 0.47734135389328003,
      "learning_rate": 2.4717488688175512e-05,
      "loss": 1.3902,
      "step": 150
    },
    {
      "epoch": 1.59375,
      "grad_norm": 0.468674898147583,
      "learning_rate": 2.3870315504160995e-05,
      "loss": 1.3863,
      "step": 153
    },
    {
      "epoch": 1.625,
      "grad_norm": 0.45810505747795105,
      "learning_rate": 2.3024440599751133e-05,
      "loss": 1.3942,
      "step": 156
    },
    {
      "epoch": 1.65625,
      "grad_norm": 0.4274010956287384,
      "learning_rate": 2.218083608896583e-05,
      "loss": 1.3502,
      "step": 159
    },
    {
      "epoch": 1.6875,
      "grad_norm": 0.43965771794319153,
      "learning_rate": 2.1340471476595835e-05,
      "loss": 1.3646,
      "step": 162
    },
    {
      "epoch": 1.71875,
      "grad_norm": 0.5207608938217163,
      "learning_rate": 2.0504312544008193e-05,
      "loss": 1.3609,
      "step": 165
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.4664800465106964,
      "learning_rate": 1.967332023923078e-05,
      "loss": 1.3641,
      "step": 168
    },
    {
      "epoch": 1.75,
      "eval_loss": 1.3919411897659302,
      "eval_runtime": 3.3286,
      "eval_samples_per_second": 48.67,
      "eval_steps_per_second": 6.309,
      "step": 168
    },
    {
      "epoch": 1.78125,
      "grad_norm": 0.43235448002815247,
      "learning_rate": 1.8848449572591628e-05,
      "loss": 1.388,
      "step": 171
    },
    {
      "epoch": 1.8125,
      "grad_norm": 0.47917473316192627,
      "learning_rate": 1.8030648519181924e-05,
      "loss": 1.3944,
      "step": 174
    },
    {
      "epoch": 1.84375,
      "grad_norm": 0.4614221453666687,
      "learning_rate": 1.7220856929404343e-05,
      "loss": 1.3947,
      "step": 177
    },
    {
      "epoch": 1.875,
      "grad_norm": 0.5674882531166077,
      "learning_rate": 1.6420005448858523e-05,
      "loss": 1.328,
      "step": 180
    },
    {
      "epoch": 1.90625,
      "grad_norm": 0.5045132637023926,
      "learning_rate": 1.562901444880508e-05,
      "loss": 1.3731,
      "step": 183
    },
    {
      "epoch": 1.9375,
      "grad_norm": 0.43909093737602234,
      "learning_rate": 1.4848792968437376e-05,
      "loss": 1.377,
      "step": 186
    },
    {
      "epoch": 1.96875,
      "grad_norm": 0.462323397397995,
      "learning_rate": 1.4080237670176455e-05,
      "loss": 1.3566,
      "step": 189
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.5580944418907166,
      "learning_rate": 1.3324231809189985e-05,
      "loss": 1.3483,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.3825615644454956,
      "eval_runtime": 3.3381,
      "eval_samples_per_second": 48.531,
      "eval_steps_per_second": 6.291,
      "step": 192
    },
    {
      "epoch": 2.03125,
      "grad_norm": 0.4587874710559845,
      "learning_rate": 1.25816442183193e-05,
      "loss": 1.3393,
      "step": 195
    },
    {
      "epoch": 2.0625,
      "grad_norm": 0.41174420714378357,
      "learning_rate": 1.1853328309581139e-05,
      "loss": 1.3425,
      "step": 198
    },
    {
      "epoch": 2.09375,
      "grad_norm": 0.4823381304740906,
      "learning_rate": 1.1140121093391736e-05,
      "loss": 1.3173,
      "step": 201
    },
    {
      "epoch": 2.125,
      "grad_norm": 0.6109301447868347,
      "learning_rate": 1.0442842216640166e-05,
      "loss": 1.3513,
      "step": 204
    },
    {
      "epoch": 2.15625,
      "grad_norm": 0.49659690260887146,
      "learning_rate": 9.762293020716695e-06,
      "loss": 1.3281,
      "step": 207
    },
    {
      "epoch": 2.1875,
      "grad_norm": 0.46417319774627686,
      "learning_rate": 9.099255620578451e-06,
      "loss": 1.3498,
      "step": 210
    },
    {
      "epoch": 2.21875,
      "grad_norm": 0.5012263059616089,
      "learning_rate": 8.454492005910941e-06,
      "loss": 1.341,
      "step": 213
    },
    {
      "epoch": 2.25,
      "grad_norm": 0.530414879322052,
      "learning_rate": 7.828743165418392e-06,
      "loss": 1.3427,
      "step": 216
    },
    {
      "epoch": 2.25,
      "eval_loss": 1.377941370010376,
      "eval_runtime": 3.3435,
      "eval_samples_per_second": 48.453,
      "eval_steps_per_second": 6.281,
      "step": 216
    },
    {
      "epoch": 2.28125,
      "grad_norm": 0.45930707454681396,
      "learning_rate": 7.222728235249196e-06,
      "loss": 1.3354,
      "step": 219
    },
    {
      "epoch": 2.3125,
      "grad_norm": 0.505244255065918,
      "learning_rate": 6.637143672535281e-06,
      "loss": 1.2967,
      "step": 222
    },
    {
      "epoch": 2.34375,
      "grad_norm": 0.4718848764896393,
      "learning_rate": 6.072662454995101e-06,
      "loss": 1.345,
      "step": 225
    },
    {
      "epoch": 2.375,
      "grad_norm": 0.4575720727443695,
      "learning_rate": 5.529933307520102e-06,
      "loss": 1.3259,
      "step": 228
    },
    {
      "epoch": 2.40625,
      "grad_norm": 0.4919467866420746,
      "learning_rate": 5.009579956633578e-06,
      "loss": 1.3407,
      "step": 231
    },
    {
      "epoch": 2.4375,
      "grad_norm": 0.6009489297866821,
      "learning_rate": 4.512200413678672e-06,
      "loss": 1.3246,
      "step": 234
    },
    {
      "epoch": 2.46875,
      "grad_norm": 0.49300843477249146,
      "learning_rate": 4.0383662875592446e-06,
      "loss": 1.3679,
      "step": 237
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.5008527040481567,
      "learning_rate": 3.588622127823604e-06,
      "loss": 1.3168,
      "step": 240
    },
    {
      "epoch": 2.5,
      "eval_loss": 1.3748859167099,
      "eval_runtime": 3.3426,
      "eval_samples_per_second": 48.466,
      "eval_steps_per_second": 6.283,
      "step": 240
    },
    {
      "epoch": 2.53125,
      "grad_norm": 0.46080684661865234,
      "learning_rate": 3.1634847988458615e-06,
      "loss": 1.3173,
      "step": 243
    },
    {
      "epoch": 2.5625,
      "grad_norm": 0.46182775497436523,
      "learning_rate": 2.7634428858242995e-06,
      "loss": 1.3378,
      "step": 246
    },
    {
      "epoch": 2.59375,
      "grad_norm": 0.48507875204086304,
      "learning_rate": 2.388956133279266e-06,
      "loss": 1.339,
      "step": 249
    },
    {
      "epoch": 2.625,
      "grad_norm": 0.5142520666122437,
      "learning_rate": 2.040454916695972e-06,
      "loss": 1.3046,
      "step": 252
    },
    {
      "epoch": 2.65625,
      "grad_norm": 0.5162761807441711,
      "learning_rate": 1.7183397479194174e-06,
      "loss": 1.3473,
      "step": 255
    },
    {
      "epoch": 2.6875,
      "grad_norm": 0.5032109022140503,
      "learning_rate": 1.4229808148697733e-06,
      "loss": 1.3258,
      "step": 258
    },
    {
      "epoch": 2.71875,
      "grad_norm": 0.5153303146362305,
      "learning_rate": 1.1547175561073154e-06,
      "loss": 1.2873,
      "step": 261
    },
    {
      "epoch": 2.75,
      "grad_norm": 0.5076745748519897,
      "learning_rate": 9.13858270735743e-07,
      "loss": 1.3198,
      "step": 264
    },
    {
      "epoch": 2.75,
      "eval_loss": 1.3732330799102783,
      "eval_runtime": 3.334,
      "eval_samples_per_second": 48.591,
      "eval_steps_per_second": 6.299,
      "step": 264
    },
    {
      "epoch": 2.78125,
      "grad_norm": 0.48225662112236023,
      "learning_rate": 7.006797640922436e-07,
      "loss": 1.3077,
      "step": 267
    },
    {
      "epoch": 2.8125,
      "grad_norm": 0.4646640419960022,
      "learning_rate": 5.154270296314878e-07,
      "loss": 1.2931,
      "step": 270
    },
    {
      "epoch": 2.84375,
      "grad_norm": 0.5893747210502625,
      "learning_rate": 3.583129673691427e-07,
      "loss": 1.3327,
      "step": 273
    },
    {
      "epoch": 2.875,
      "grad_norm": 0.601911723613739,
      "learning_rate": 2.295181392084511e-07,
      "loss": 1.3169,
      "step": 276
    },
    {
      "epoch": 2.90625,
      "grad_norm": 0.5594666600227356,
      "learning_rate": 1.2919056143113062e-07,
      "loss": 1.3379,
      "step": 279
    },
    {
      "epoch": 2.9375,
      "grad_norm": 0.5165249705314636,
      "learning_rate": 5.7445534591002436e-08,
      "loss": 1.3493,
      "step": 282
    },
    {
      "epoch": 2.96875,
      "grad_norm": 0.48071709275245667,
      "learning_rate": 1.4365511005878795e-08,
      "loss": 1.3008,
      "step": 285
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.5523079037666321,
      "learning_rate": 0.0,
      "loss": 1.3255,
      "step": 288
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.3729829788208008,
      "eval_runtime": 3.3405,
      "eval_samples_per_second": 48.496,
      "eval_steps_per_second": 6.287,
      "step": 288
    }
  ],
  "logging_steps": 3,
  "max_steps": 288,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 24,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.472770514845696e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}